UbiquitousOS
/
OpenCloudOS-Kernel
mirror of https://gitee.com/OpenCloudOS/OpenCloudOS-Kernel.git
11
0
Fork 0
Code Issues Projects Releases Wiki Activity
OpenCloudOS-Kernel/drivers/scsi/megaraid/megaraid_mbox.c

4073 lines
98 KiB
C
Raw Normal View History

treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 152 Based on 1 normalized pattern(s): this program is free software you can redistribute it and or modify it under the terms of the gnu general public license as published by the free software foundation either version 2 of the license or at your option any later version extracted by the scancode license scanner the SPDX license identifier GPL-2.0-or-later has been chosen to replace the boilerplate/reference in 3029 file(s). Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Allison Randal <allison@lohutok.net> Cc: linux-spdx@vger.kernel.org Link: https://lkml.kernel.org/r/20190527070032.746973796@linutronix.de Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2019-05-27 14:55:01 +08:00
// SPDX-License-Identifier: GPL-2.0-or-later
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/*
*
* Linux MegaRAID device driver
*
* Copyright (c) 2003-2004 LSI Logic Corporation.
*
* FILE : megaraid_mbox.c
[SCSI] megaraid_{mm,mbox}: init fix for kdump 1. Changes in Initialization to fix kdump failure. Send SYNC command on loading. This command clears the pending commands in the adapter and re-initialize its internal RAID structure. Without this change, megaraid driver either panics or fails to initialize the adapter during kdump's second kernel boot if there are pending commands or interrupts from other devices sharing the same IRQ. 2. Authors email-id domain name changed from lsil.com to lsi.com. Also modified the MODULE_AUTHOR to megaraidlinux@lsi.com Signed-off-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-05 23:10:09 +08:00
* Version : v2.20.5.1 (Nov 16 2006)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* Authors:
[SCSI] megaraid_{mm,mbox}: init fix for kdump 1. Changes in Initialization to fix kdump failure. Send SYNC command on loading. This command clears the pending commands in the adapter and re-initialize its internal RAID structure. Without this change, megaraid driver either panics or fails to initialize the adapter during kdump's second kernel boot if there are pending commands or interrupts from other devices sharing the same IRQ. 2. Authors email-id domain name changed from lsil.com to lsi.com. Also modified the MODULE_AUTHOR to megaraidlinux@lsi.com Signed-off-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-05 23:10:09 +08:00
* Atul Mukker <Atul.Mukker@lsi.com>
* Sreenivas Bagalkote <Sreenivas.Bagalkote@lsi.com>
* Manoj Jose <Manoj.Jose@lsi.com>
* Seokmann Ju
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* List of supported controllers
*
* OEM Product Name VID DID SSVID SSID
* --- ------------ --- --- ---- ----
* Dell PERC3/QC 101E 1960 1028 0471
* Dell PERC3/DC 101E 1960 1028 0493
* Dell PERC3/SC 101E 1960 1028 0475
* Dell PERC3/Di 1028 1960 1028 0123
* Dell PERC4/SC 1000 1960 1028 0520
* Dell PERC4/DC 1000 1960 1028 0518
* Dell PERC4/QC 1000 0407 1028 0531
* Dell PERC4/Di 1028 000F 1028 014A
* Dell PERC 4e/Si 1028 0013 1028 016c
* Dell PERC 4e/Di 1028 0013 1028 016d
* Dell PERC 4e/Di 1028 0013 1028 016e
* Dell PERC 4e/Di 1028 0013 1028 016f
* Dell PERC 4e/Di 1028 0013 1028 0170
* Dell PERC 4e/DC 1000 0408 1028 0002
* Dell PERC 4e/SC 1000 0408 1028 0001
*
* LSI MegaRAID SCSI 320-0 1000 1960 1000 A520
* LSI MegaRAID SCSI 320-1 1000 1960 1000 0520
* LSI MegaRAID SCSI 320-2 1000 1960 1000 0518
* LSI MegaRAID SCSI 320-0X 1000 0407 1000 0530
* LSI MegaRAID SCSI 320-2X 1000 0407 1000 0532
* LSI MegaRAID SCSI 320-4X 1000 0407 1000 0531
* LSI MegaRAID SCSI 320-1E 1000 0408 1000 0001
* LSI MegaRAID SCSI 320-2E 1000 0408 1000 0002
* LSI MegaRAID SATA 150-4 1000 1960 1000 4523
* LSI MegaRAID SATA 150-6 1000 1960 1000 0523
* LSI MegaRAID SATA 300-4X 1000 0409 1000 3004
* LSI MegaRAID SATA 300-8X 1000 0409 1000 3008
*
* INTEL RAID Controller SRCU42X 1000 0407 8086 0532
* INTEL RAID Controller SRCS16 1000 1960 8086 0523
* INTEL RAID Controller SRCU42E 1000 0408 8086 0002
* INTEL RAID Controller SRCZCRX 1000 0407 8086 0530
* INTEL RAID Controller SRCS28X 1000 0409 8086 3008
* INTEL RAID Controller SROMBU42E 1000 0408 8086 3431
* INTEL RAID Controller SROMBU42E 1000 0408 8086 3499
* INTEL RAID Controller SRCU51L 1000 1960 8086 0520
*
* FSC MegaRAID PCI Express ROMB 1000 0408 1734 1065
*
* ACER MegaRAID ROMB-2E 1000 0408 1025 004D
*
* NEC MegaRAID PCI Express ROMB 1000 0408 1033 8287
*
doc: fix broken references There are numerous broken references to Documentation files (in other Documentation files, in comments, etc.). These broken references are caused by typo's in the references, and by renames or removals of the Documentation files. Some broken references are simply odd. Fix these broken references, sometimes by dropping the irrelevant text they were part of. Signed-off-by: Paul Bolle <pebolle@tiscali.nl> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2011-08-15 08:02:26 +08:00
* For history of changes, see Documentation/scsi/ChangeLog.megaraid
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#include <linux/slab.h>
scsi: Fix up files implicitly depending on module.h inclusion The module.h header was implicitly present everywhere, so files with no explicit include of the module infrastructure would build anyway. We are now removing the implicit include, and so we need to call out the module.h file that we need explicitly. Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-05-27 21:47:43 +08:00
#include <linux/module.h>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
#include "megaraid_mbox.h"
static int megaraid_init(void);
static void megaraid_exit(void);
static int megaraid_probe_one(struct pci_dev*, const struct pci_device_id *);
static void megaraid_detach_one(struct pci_dev *);
[PATCH] PCI: Convert megaraid to use pci_driver shutdown method Convert megaraid to use pci_driver's shutdown method rather than the generic device_driver shutdown method. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2005-10-17 04:32:46 +08:00
static void megaraid_mbox_shutdown(struct pci_dev *);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
static int megaraid_io_attach(adapter_t *);
static void megaraid_io_detach(adapter_t *);
static int megaraid_init_mbox(adapter_t *);
static void megaraid_fini_mbox(adapter_t *);
static int megaraid_alloc_cmd_packets(adapter_t *);
static void megaraid_free_cmd_packets(adapter_t *);
static int megaraid_mbox_setup_dma_pools(adapter_t *);
static void megaraid_mbox_teardown_dma_pools(adapter_t *);
static int megaraid_sysfs_alloc_resources(adapter_t *);
static void megaraid_sysfs_free_resources(adapter_t *);
static int megaraid_abort_handler(struct scsi_cmnd *);
static int megaraid_reset_handler(struct scsi_cmnd *);
static int mbox_post_sync_cmd(adapter_t *, uint8_t []);
static int mbox_post_sync_cmd_fast(adapter_t *, uint8_t []);
static int megaraid_busywait_mbox(mraid_device_t *);
static int megaraid_mbox_product_info(adapter_t *);
static int megaraid_mbox_extended_cdb(adapter_t *);
static int megaraid_mbox_support_ha(adapter_t *, uint16_t *);
static int megaraid_mbox_support_random_del(adapter_t *);
static int megaraid_mbox_get_max_sg(adapter_t *);
static void megaraid_mbox_enum_raid_scsi(adapter_t *);
static void megaraid_mbox_flush_cache(adapter_t *);
[SCSI] megaraid_{mm,mbox}: init fix for kdump 1. Changes in Initialization to fix kdump failure. Send SYNC command on loading. This command clears the pending commands in the adapter and re-initialize its internal RAID structure. Without this change, megaraid driver either panics or fails to initialize the adapter during kdump's second kernel boot if there are pending commands or interrupts from other devices sharing the same IRQ. 2. Authors email-id domain name changed from lsil.com to lsi.com. Also modified the MODULE_AUTHOR to megaraidlinux@lsi.com Signed-off-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-05 23:10:09 +08:00
static int megaraid_mbox_fire_sync_cmd(adapter_t *);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
static void megaraid_mbox_display_scb(adapter_t *, scb_t *);
static void megaraid_mbox_setup_device_map(adapter_t *);
SCSI host lock push-down Move the mid-layer's ->queuecommand() invocation from being locked with the host lock to being unlocked to facilitate speeding up the critical path for drivers who don't need this lock taken anyway. The patch below presents a simple SCSI host lock push-down as an equivalent transformation. No locking or other behavior should change with this patch. All existing bugs and locking orders are preserved. Additionally, add one parameter to queuecommand, struct Scsi_Host * and remove one parameter from queuecommand, void (*done)(struct scsi_cmnd *) Scsi_Host* is a convenient pointer that most host drivers need anyway, and 'done' is redundant to struct scsi_cmnd->scsi_done. Minimal code disturbance was attempted with this change. Most drivers needed only two one-line modifications for their host lock push-down. Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Acked-by: James Bottomley <James.Bottomley@suse.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-11-16 15:10:29 +08:00
static int megaraid_queue_command(struct Scsi_Host *, struct scsi_cmnd *);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
static scb_t *megaraid_mbox_build_cmd(adapter_t *, struct scsi_cmnd *, int *);
static void megaraid_mbox_runpendq(adapter_t *, scb_t *);
static void megaraid_mbox_prepare_pthru(adapter_t *, scb_t *,
struct scsi_cmnd *);
static void megaraid_mbox_prepare_epthru(adapter_t *, scb_t *,
struct scsi_cmnd *);
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 21:55:46 +08:00
static irqreturn_t megaraid_isr(int, void *);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
static void megaraid_mbox_dpc(unsigned long);
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-22 07:13:36 +08:00
static ssize_t megaraid_sysfs_show_app_hndl(struct device *, struct device_attribute *attr, char *);
[PATCH] Driver Core: drivers/s390/net/qeth_sys.c - drivers/usb/gadget/pxa2xx_udc.c: update device attribute callbacks Signed-off-by: Yani Ioannou <yani.ioannou@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2005-05-17 18:43:37 +08:00
static ssize_t megaraid_sysfs_show_ldnum(struct device *, struct device_attribute *attr, char *);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
static int megaraid_cmm_register(adapter_t *);
static int megaraid_cmm_unregister(adapter_t *);
static int megaraid_mbox_mm_handler(unsigned long, uioc_t *, uint32_t);
static int megaraid_mbox_mm_command(adapter_t *, uioc_t *);
static void megaraid_mbox_mm_done(adapter_t *, scb_t *);
static int gather_hbainfo(adapter_t *, mraid_hba_info_t *);
static int wait_till_fw_empty(adapter_t *);
[SCSI] megaraid_{mm,mbox}: init fix for kdump 1. Changes in Initialization to fix kdump failure. Send SYNC command on loading. This command clears the pending commands in the adapter and re-initialize its internal RAID structure. Without this change, megaraid driver either panics or fails to initialize the adapter during kdump's second kernel boot if there are pending commands or interrupts from other devices sharing the same IRQ. 2. Authors email-id domain name changed from lsil.com to lsi.com. Also modified the MODULE_AUTHOR to megaraidlinux@lsi.com Signed-off-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-05 23:10:09 +08:00
MODULE_AUTHOR("megaraidlinux@lsi.com");
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
MODULE_DESCRIPTION("LSI Logic MegaRAID Mailbox Driver");
MODULE_LICENSE("GPL");
MODULE_VERSION(MEGARAID_VERSION);
/*
* ### modules parameters for driver ###
*/
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
/*
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* Set to enable driver to expose unconfigured disk to kernel
*/
static int megaraid_expose_unconf_disks = 0;
module_param_named(unconf_disks, megaraid_expose_unconf_disks, int, 0);
MODULE_PARM_DESC(unconf_disks,
"Set to expose unconfigured disks to kernel (default=0)");
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
/*
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* driver wait time if the adapter's mailbox is busy
*/
static unsigned int max_mbox_busy_wait = MBOX_BUSY_WAIT;
module_param_named(busy_wait, max_mbox_busy_wait, int, 0);
MODULE_PARM_DESC(busy_wait,
"Max wait for mailbox in microseconds if busy (default=10)");
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
/*
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* number of sectors per IO command
*/
static unsigned int megaraid_max_sectors = MBOX_MAX_SECTORS;
module_param_named(max_sectors, megaraid_max_sectors, int, 0);
MODULE_PARM_DESC(max_sectors,
"Maximum number of sectors per IO command (default=128)");
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
/*
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* number of commands per logical unit
*/
static unsigned int megaraid_cmd_per_lun = MBOX_DEF_CMD_PER_LUN;
module_param_named(cmd_per_lun, megaraid_cmd_per_lun, int, 0);
MODULE_PARM_DESC(cmd_per_lun,
"Maximum number of commands per logical unit (default=64)");
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
/*
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* Fast driver load option, skip scanning for physical devices during load.
* This would result in non-disk devices being skipped during driver load
* time. These can be later added though, using /proc/scsi/scsi
*/
static unsigned int megaraid_fast_load = 0;
module_param_named(fast_load, megaraid_fast_load, int, 0);
MODULE_PARM_DESC(fast_load,
"Faster loading of the driver, skips physical devices! (default=0)");
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
/*
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* mraid_debug level - threshold for amount of information to be displayed by
* the driver. This level can be changed through modules parameters, ioctl or
* sysfs/proc interface. By default, print the announcement messages only.
*/
int mraid_debug_level = CL_ANN;
module_param_named(debug_level, mraid_debug_level, int, 0);
MODULE_PARM_DESC(debug_level, "Debug level for driver (default=0)");
/*
* PCI table for all supported controllers.
*/
static struct pci_device_id pci_id_table_g[] = {
{
PCI_VENDOR_ID_DELL,
PCI_DEVICE_ID_PERC4_DI_DISCOVERY,
PCI_VENDOR_ID_DELL,
PCI_SUBSYS_ID_PERC4_DI_DISCOVERY,
},
{
PCI_VENDOR_ID_LSI_LOGIC,
PCI_DEVICE_ID_PERC4_SC,
PCI_VENDOR_ID_DELL,
PCI_SUBSYS_ID_PERC4_SC,
},
{
PCI_VENDOR_ID_LSI_LOGIC,
PCI_DEVICE_ID_PERC4_DC,
PCI_VENDOR_ID_DELL,
PCI_SUBSYS_ID_PERC4_DC,
},
{
PCI_VENDOR_ID_LSI_LOGIC,
[SCSI] megaraid version 2.20.4.6 Signed-off by: Seokmann Ju <sju@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-05-17 06:32:17 +08:00
PCI_DEVICE_ID_VERDE,
PCI_ANY_ID,
PCI_ANY_ID,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
},
{
PCI_VENDOR_ID_DELL,
PCI_DEVICE_ID_PERC4_DI_EVERGLADES,
PCI_VENDOR_ID_DELL,
PCI_SUBSYS_ID_PERC4_DI_EVERGLADES,
},
{
PCI_VENDOR_ID_DELL,
PCI_DEVICE_ID_PERC4E_SI_BIGBEND,
PCI_VENDOR_ID_DELL,
PCI_SUBSYS_ID_PERC4E_SI_BIGBEND,
},
{
PCI_VENDOR_ID_DELL,
PCI_DEVICE_ID_PERC4E_DI_KOBUK,
PCI_VENDOR_ID_DELL,
PCI_SUBSYS_ID_PERC4E_DI_KOBUK,
},
{
PCI_VENDOR_ID_DELL,
PCI_DEVICE_ID_PERC4E_DI_CORVETTE,
PCI_VENDOR_ID_DELL,
PCI_SUBSYS_ID_PERC4E_DI_CORVETTE,
},
{
PCI_VENDOR_ID_DELL,
PCI_DEVICE_ID_PERC4E_DI_EXPEDITION,
PCI_VENDOR_ID_DELL,
PCI_SUBSYS_ID_PERC4E_DI_EXPEDITION,
},
{
PCI_VENDOR_ID_DELL,
PCI_DEVICE_ID_PERC4E_DI_GUADALUPE,
PCI_VENDOR_ID_DELL,
PCI_SUBSYS_ID_PERC4E_DI_GUADALUPE,
},
{
PCI_VENDOR_ID_LSI_LOGIC,
[SCSI] megaraid version 2.20.4.6 Signed-off by: Seokmann Ju <sju@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-05-17 06:32:17 +08:00
PCI_DEVICE_ID_DOBSON,
PCI_ANY_ID,
PCI_ANY_ID,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
},
{
PCI_VENDOR_ID_AMI,
PCI_DEVICE_ID_AMI_MEGARAID3,
[SCSI] megaraid_{mbox,mm} : remove PCI Id overlaping between megaraid_legacy and megaraid_{mbox,mm} Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-18 02:17:25 +08:00
PCI_ANY_ID,
PCI_ANY_ID,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
},
{
PCI_VENDOR_ID_LSI_LOGIC,
[SCSI] megaraid_{mbox,mm} : remove PCI Id overlaping between megaraid_legacy and megaraid_{mbox,mm} Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-18 02:17:25 +08:00
PCI_DEVICE_ID_AMI_MEGARAID3,
PCI_ANY_ID,
PCI_ANY_ID,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
},
{
PCI_VENDOR_ID_LSI_LOGIC,
[SCSI] megaraid version 2.20.4.6 Signed-off by: Seokmann Ju <sju@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-05-17 06:32:17 +08:00
PCI_DEVICE_ID_LINDSAY,
PCI_ANY_ID,
PCI_ANY_ID,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
},
{0} /* Terminating entry */
};
MODULE_DEVICE_TABLE(pci, pci_id_table_g);
[SCSI] megaraid: fix section mismatch Change megaraid_pci_driver_g variable name so that it matches the modpost whitelist that allows pointers to init text/data. WARNING: vmlinux.o(.data+0x1a8e30): Section mismatch: reference to .init.text:megaraid_probe_one (between 'megaraid_pci_driver_g' and 'class_device_attr_megaraid_mbox_app_hndl') Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-01-11 06:33:16 +08:00
static struct pci_driver megaraid_pci_driver = {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
.name = "megaraid",
.id_table = pci_id_table_g,
.probe = megaraid_probe_one,
Drivers: scsi: remove __dev* attributes. CONFIG_HOTPLUG is going away as an option. As a result, the __dev* markings need to be removed. This change removes the use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit from these drivers. Based on patches originally written by Bill Pemberton, but redone by me in order to handle some of the coding style issues better, by hand. Cc: Bill Pemberton <wfp5p@virginia.edu> Cc: Adam Radford <linuxraid@lsi.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-22 05:08:55 +08:00
.remove = megaraid_detach_one,
[PATCH] PCI: Convert megaraid to use pci_driver shutdown method Convert megaraid to use pci_driver's shutdown method rather than the generic device_driver shutdown method. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2005-10-17 04:32:46 +08:00
.shutdown = megaraid_mbox_shutdown,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
};
// definitions for the device attributes for exporting logical drive number
// for a scsi address (Host, Channel, Id, Lun)
scsi: megaraid: make two symbols static in megaraid_mbox.c Fix the following sparse warning: drivers/scsi/megaraid/megaraid_mbox.c:305:5: warning: symbol 'dev_attr_megaraid_mbox_app_hndl' was not declared. Should it be static? drivers/scsi/megaraid/megaraid_mbox.c:315:5: warning: symbol 'dev_attr_megaraid_mbox_ld' was not declared. Should it be static? Link: https://lore.kernel.org/r/20200407092827.18074-2-yanaijie@huawei.com Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Jason Yan <yanaijie@huawei.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-04-07 17:28:24 +08:00
static DEVICE_ATTR(megaraid_mbox_app_hndl, S_IRUSR, megaraid_sysfs_show_app_hndl,
NULL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
// Host template initializer for megaraid mbox sysfs device attributes
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-22 07:13:36 +08:00
static struct device_attribute *megaraid_shost_attrs[] = {
&dev_attr_megaraid_mbox_app_hndl,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
NULL,
};
scsi: megaraid: make two symbols static in megaraid_mbox.c Fix the following sparse warning: drivers/scsi/megaraid/megaraid_mbox.c:305:5: warning: symbol 'dev_attr_megaraid_mbox_app_hndl' was not declared. Should it be static? drivers/scsi/megaraid/megaraid_mbox.c:315:5: warning: symbol 'dev_attr_megaraid_mbox_ld' was not declared. Should it be static? Link: https://lore.kernel.org/r/20200407092827.18074-2-yanaijie@huawei.com Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Jason Yan <yanaijie@huawei.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2020-04-07 17:28:24 +08:00
static DEVICE_ATTR(megaraid_mbox_ld, S_IRUSR, megaraid_sysfs_show_ldnum, NULL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
// Host template initializer for megaraid mbox sysfs device attributes
static struct device_attribute *megaraid_sdev_attrs[] = {
&dev_attr_megaraid_mbox_ld,
NULL,
};
/*
* Scsi host template for megaraid unified driver
*/
static struct scsi_host_template megaraid_template_g = {
.module = THIS_MODULE,
.name = "LSI Logic MegaRAID driver",
.proc_name = "megaraid",
.queuecommand = megaraid_queue_command,
.eh_abort_handler = megaraid_abort_handler,
.eh_host_reset_handler = megaraid_reset_handler,
scsi: drop reason argument from ->change_queue_depth Drop the now unused reason argument from the ->change_queue_depth method. Also add a return value to scsi_adjust_queue_depth, and rename it to scsi_change_queue_depth now that it can be used as the default ->change_queue_depth implementation. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Mike Christie <michaelc@cs.wisc.edu> Reviewed-by: Hannes Reinecke <hare@suse.de>
2014-11-13 22:08:42 +08:00
.change_queue_depth = scsi_change_queue_depth,
[SCSI] Disable WRITE SAME for RAID and virtual host adapter drivers Some host adapters do not pass commands through to the target disk directly. Instead they provide an emulated target which may or may not accurately report its capabilities. In some cases the physical device characteristics are reported even when the host adapter is processing commands on the device's behalf. This can lead to adapter firmware hangs or excessive I/O errors. This patch disables WRITE SAME for devices connected to host adapters that provide an emulated target. Driver writers can disable WRITE SAME by setting the no_write_same flag in the host adapter template. [jejb: fix up rejections due to eh_deadline patch] Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Cc: stable@kernel.org Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-10-23 18:25:40 +08:00
.no_write_same = 1,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
.sdev_attrs = megaraid_sdev_attrs,
.shost_attrs = megaraid_shost_attrs,
};
/**
* megaraid_init - module load hook
*
* We register ourselves as hotplug enabled module and let PCI subsystem
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* discover our adapters.
*/
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
static int __init
megaraid_init(void)
{
int rval;
// Announce the driver version
con_log(CL_ANN, (KERN_INFO "megaraid: %s %s\n", MEGARAID_VERSION,
MEGARAID_EXT_VERSION));
// check validity of module parameters
if (megaraid_cmd_per_lun > MBOX_MAX_SCSI_CMDS) {
con_log(CL_ANN, (KERN_WARNING
"megaraid mailbox: max commands per lun reset to %d\n",
MBOX_MAX_SCSI_CMDS));
megaraid_cmd_per_lun = MBOX_MAX_SCSI_CMDS;
}
// register as a PCI hot-plug driver module
[SCSI] megaraid: fix section mismatch Change megaraid_pci_driver_g variable name so that it matches the modpost whitelist that allows pointers to init text/data. WARNING: vmlinux.o(.data+0x1a8e30): Section mismatch: reference to .init.text:megaraid_probe_one (between 'megaraid_pci_driver_g' and 'class_device_attr_megaraid_mbox_app_hndl') Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-01-11 06:33:16 +08:00
rval = pci_register_driver(&megaraid_pci_driver);
[SCSI] megaraid version 2.20.4.6 Signed-off by: Seokmann Ju <sju@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-05-17 06:32:17 +08:00
if (rval < 0) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
con_log(CL_ANN, (KERN_WARNING
"megaraid: could not register hotplug support.\n"));
}
return rval;
}
/**
* megaraid_exit - driver unload entry point
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* We simply unwrap the megaraid_init routine here.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static void __exit
megaraid_exit(void)
{
con_log(CL_DLEVEL1, (KERN_NOTICE "megaraid: unloading framework\n"));
// unregister as PCI hotplug driver
[SCSI] megaraid: fix section mismatch Change megaraid_pci_driver_g variable name so that it matches the modpost whitelist that allows pointers to init text/data. WARNING: vmlinux.o(.data+0x1a8e30): Section mismatch: reference to .init.text:megaraid_probe_one (between 'megaraid_pci_driver_g' and 'class_device_attr_megaraid_mbox_app_hndl') Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-01-11 06:33:16 +08:00
pci_unregister_driver(&megaraid_pci_driver);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return;
}
/**
* megaraid_probe_one - PCI hotplug entry point
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @pdev : handle to this controller's PCI configuration space
* @id : pci device id of the class of controllers
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* This routine should be called whenever a new adapter is detected by the
Fix misspellings of "system", "controller", "interrupt" and "necessary". Fix the various misspellings of "system", controller", "interrupt" and "[un]necessary". Signed-off-by: Robert P. J. Day <rpjday@mindspring.com> Signed-off-by: Adrian Bunk <bunk@kernel.org>
2007-10-20 05:10:43 +08:00
* PCI hotplug susbsystem.
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
*/
Drivers: scsi: remove __dev* attributes. CONFIG_HOTPLUG is going away as an option. As a result, the __dev* markings need to be removed. This change removes the use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit from these drivers. Based on patches originally written by Bill Pemberton, but redone by me in order to handle some of the coding style issues better, by hand. Cc: Bill Pemberton <wfp5p@virginia.edu> Cc: Adam Radford <linuxraid@lsi.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-22 05:08:55 +08:00
static int
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
megaraid_probe_one(struct pci_dev *pdev, const struct pci_device_id *id)
{
adapter_t *adapter;
// detected a new controller
con_log(CL_ANN, (KERN_INFO
"megaraid: probe new device %#4.04x:%#4.04x:%#4.04x:%#4.04x: ",
pdev->vendor, pdev->device, pdev->subsystem_vendor,
pdev->subsystem_device));
con_log(CL_ANN, ("bus %d:slot %d:func %d\n", pdev->bus->number,
PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn)));
if (pci_enable_device(pdev)) {
con_log(CL_ANN, (KERN_WARNING
"megaraid: pci_enable_device failed\n"));
return -ENODEV;
}
// Enable bus-mastering on this controller
pci_set_master(pdev);
// Allocate the per driver initialization structure
some kmalloc/memset ->kzalloc (tree wide) Transform some calls to kmalloc/memset to a single kzalloc (or kcalloc). Here is a short excerpt of the semantic patch performing this transformation: @@ type T2; expression x; identifier f,fld; expression E; expression E1,E2; expression e1,e2,e3,y; statement S; @@ x = - kmalloc + kzalloc (E1,E2) ... when != \(x->fld=E;\|y=f(...,x,...);\|f(...,x,...);\|x=E;\|while(...) S\|for(e1;e2;e3) S\) - memset((T2)x,0,E1); @@ expression E1,E2,E3; @@ - kzalloc(E1 * E2,E3) + kcalloc(E1,E2,E3) [akpm@linux-foundation.org: get kcalloc args the right way around] Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Acked-by: Russell King <rmk@arm.linux.org.uk> Cc: Bryan Wu <bryan.wu@analog.com> Acked-by: Jiri Slaby <jirislaby@gmail.com> Cc: Dave Airlie <airlied@linux.ie> Acked-by: Roland Dreier <rolandd@cisco.com> Cc: Jiri Kosina <jkosina@suse.cz> Acked-by: Dmitry Torokhov <dtor@mail.ru> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Acked-by: Mauro Carvalho Chehab <mchehab@infradead.org> Acked-by: Pierre Ossman <drzeus-list@drzeus.cx> Cc: Jeff Garzik <jeff@garzik.org> Cc: "David S. Miller" <davem@davemloft.net> Acked-by: Greg KH <greg@kroah.com> Cc: James Bottomley <James.Bottomley@steeleye.com> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-19 16:49:03 +08:00
adapter = kzalloc(sizeof(adapter_t), GFP_KERNEL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (adapter == NULL) {
con_log(CL_ANN, (KERN_WARNING
[SCSI] replace __FUNCTION__ with __func__ [jejb: fixed up a ton of missed conversions. All of you are on notice this has happened, driver trees will now need to be rebased] Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Cc: SCSI List <linux-scsi@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-07-04 14:47:27 +08:00
"megaraid: out of memory, %s %d.\n", __func__, __LINE__));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
goto out_probe_one;
}
// set up PCI related soft state and other pre-known parameters
adapter->unique_id = pdev->bus->number << 8 | pdev->devfn;
adapter->irq = pdev->irq;
adapter->pdev = pdev;
atomic_set(&adapter->being_detached, 0);
// Setup the default DMA mask. This would be changed later on
// depending on hardware capabilities
scsi: megaraid_mbox: switch to generic DMA API Switch from the legacy PCI DMA API to the generic DMA API. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-10-11 15:34:24 +08:00
if (dma_set_mask(&adapter->pdev->dev, DMA_BIT_MASK(32))) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
con_log(CL_ANN, (KERN_WARNING
scsi: megaraid_mbox: switch to generic DMA API Switch from the legacy PCI DMA API to the generic DMA API. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-10-11 15:34:24 +08:00
"megaraid: dma_set_mask failed:%d\n", __LINE__));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
goto out_free_adapter;
}
// Initialize the synchronization lock for kernel and LLD
spin_lock_init(&adapter->lock);
// Initialize the command queues: the list of free SCBs and the list
// of pending SCBs.
INIT_LIST_HEAD(&adapter->kscb_pool);
spin_lock_init(SCSI_FREE_LIST_LOCK(adapter));
INIT_LIST_HEAD(&adapter->pend_list);
spin_lock_init(PENDING_LIST_LOCK(adapter));
INIT_LIST_HEAD(&adapter->completed_list);
spin_lock_init(COMPLETED_LIST_LOCK(adapter));
// Start the mailbox based controller
if (megaraid_init_mbox(adapter) != 0) {
con_log(CL_ANN, (KERN_WARNING
scsi: megaraid: fix spelling mistake "maibox" -> "mailbox" Trivial fix to spelling mistake in warning message and comments Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-09-24 06:25:48 +08:00
"megaraid: mailbox adapter did not initialize\n"));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
goto out_free_adapter;
}
// Register with LSI Common Management Module
if (megaraid_cmm_register(adapter) != 0) {
con_log(CL_ANN, (KERN_WARNING
"megaraid: could not register with management module\n"));
goto out_fini_mbox;
}
// setup adapter handle in PCI soft state
pci_set_drvdata(pdev, adapter);
// attach with scsi mid-layer
if (megaraid_io_attach(adapter) != 0) {
con_log(CL_ANN, (KERN_WARNING "megaraid: io attach failed\n"));
goto out_cmm_unreg;
}
return 0;
out_cmm_unreg:
megaraid_cmm_unregister(adapter);
out_fini_mbox:
megaraid_fini_mbox(adapter);
out_free_adapter:
kfree(adapter);
out_probe_one:
pci_disable_device(pdev);
return -ENODEV;
}
/**
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* megaraid_detach_one - release framework resources and call LLD release routine
scsi/megaraid fixed several typos in comments Fixed several typos in comments in megaraid_mbox.c, megaraid_mm.c and megaraid_sas_fusion.h. Signed-off-by: Matthias Schid <aircrach115@gmail.com> Signed-off-by: Stefan Huber <steffhip@gmail.com> Signed-off-by: Simon Puels <simon.puels@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2013-06-26 22:25:53 +08:00
* @pdev : handle for our PCI configuration space
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* This routine is called during driver unload. We free all the allocated
* resources and call the corresponding LLD so that it can also release all
* its resources.
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* This routine is also called from the PCI hotplug system.
*/
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
static void
megaraid_detach_one(struct pci_dev *pdev)
{
adapter_t *adapter;
struct Scsi_Host *host;
// Start a rollback on this adapter
adapter = pci_get_drvdata(pdev);
if (!adapter) {
con_log(CL_ANN, (KERN_CRIT
"megaraid: Invalid detach on %#4.04x:%#4.04x:%#4.04x:%#4.04x\n",
pdev->vendor, pdev->device, pdev->subsystem_vendor,
pdev->subsystem_device));
return;
}
else {
con_log(CL_ANN, (KERN_NOTICE
"megaraid: detaching device %#4.04x:%#4.04x:%#4.04x:%#4.04x\n",
pdev->vendor, pdev->device, pdev->subsystem_vendor,
pdev->subsystem_device));
}
host = adapter->host;
// do not allow any more requests from the management module for this
// adapter.
// FIXME: How do we account for the request which might still be
// pending with us?
atomic_set(&adapter->being_detached, 1);
// detach from the IO sub-system
megaraid_io_detach(adapter);
// Unregister from common management module
//
// FIXME: this must return success or failure for conditions if there
// is a command pending with LLD or not.
megaraid_cmm_unregister(adapter);
// finalize the mailbox based controller and release all resources
megaraid_fini_mbox(adapter);
kfree(adapter);
scsi_host_put(host);
pci_disable_device(pdev);
return;
}
/**
* megaraid_mbox_shutdown - PCI shutdown for megaraid HBA
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @pdev : generic driver model device
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Shutdown notification, perform flush cache.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static void
[PATCH] PCI: Convert megaraid to use pci_driver shutdown method Convert megaraid to use pci_driver's shutdown method rather than the generic device_driver shutdown method. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2005-10-17 04:32:46 +08:00
megaraid_mbox_shutdown(struct pci_dev *pdev)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
[PATCH] PCI: Convert megaraid to use pci_driver shutdown method Convert megaraid to use pci_driver's shutdown method rather than the generic device_driver shutdown method. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2005-10-17 04:32:46 +08:00
adapter_t *adapter = pci_get_drvdata(pdev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
static int counter;
if (!adapter) {
con_log(CL_ANN, (KERN_WARNING
"megaraid: null device in shutdown\n"));
return;
}
// flush caches now
con_log(CL_ANN, (KERN_INFO "megaraid: flushing adapter %d...",
counter++));
megaraid_mbox_flush_cache(adapter);
con_log(CL_ANN, ("done\n"));
}
/**
* megaraid_io_attach - attach a device with the IO subsystem
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : controller's soft state
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Attach this device with the IO subsystem.
*/
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
static int
megaraid_io_attach(adapter_t *adapter)
{
struct Scsi_Host *host;
// Initialize SCSI Host structure
host = scsi_host_alloc(&megaraid_template_g, 8);
if (!host) {
con_log(CL_ANN, (KERN_WARNING
"megaraid mbox: scsi_register failed\n"));
return -1;
}
SCSIHOST2ADAP(host) = (caddr_t)adapter;
adapter->host = host;
host->irq = adapter->irq;
host->unique_id = adapter->unique_id;
host->can_queue = adapter->max_cmds;
host->this_id = adapter->init_id;
host->sg_tablesize = adapter->sglen;
host->max_sectors = adapter->max_sectors;
host->cmd_per_lun = adapter->cmd_per_lun;
host->max_channel = adapter->max_channel;
host->max_id = adapter->max_target;
host->max_lun = adapter->max_lun;
// notify mid-layer about the new controller
if (scsi_add_host(host, &adapter->pdev->dev)) {
con_log(CL_ANN, (KERN_WARNING
"megaraid mbox: scsi_add_host failed\n"));
scsi_host_put(host);
return -1;
}
scsi_scan_host(host);
return 0;
}
/**
* megaraid_io_detach - detach a device from the IO subsystem
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : controller's soft state
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Detach this device from the IO subsystem.
*/
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
static void
megaraid_io_detach(adapter_t *adapter)
{
struct Scsi_Host *host;
con_log(CL_DLEVEL1, (KERN_INFO "megaraid: io detach\n"));
host = adapter->host;
scsi_remove_host(host);
return;
}
/*
* START: Mailbox Low Level Driver
*
* This is section specific to the single mailbox based controllers
*/
/**
* megaraid_init_mbox - initialize controller
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : our soft state
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* - Allocate 16-byte aligned mailbox memory for firmware handshake
* - Allocate controller's memory resources
* - Find out all initialization data
* - Allocate memory required for all the commands
* - Use internal library of FW routines, build up complete soft state
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
Drivers: scsi: remove __dev* attributes. CONFIG_HOTPLUG is going away as an option. As a result, the __dev* markings need to be removed. This change removes the use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit from these drivers. Based on patches originally written by Bill Pemberton, but redone by me in order to handle some of the coding style issues better, by hand. Cc: Bill Pemberton <wfp5p@virginia.edu> Cc: Adam Radford <linuxraid@lsi.com> Cc: "James E.J. Bottomley" <JBottomley@parallels.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-22 05:08:55 +08:00
static int
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
megaraid_init_mbox(adapter_t *adapter)
{
struct pci_dev *pdev;
mraid_device_t *raid_dev;
int i;
[SCSI] megaraid_{mm,mbox}: 64-bit DMA capability checker This patch contains - a fix for 64-bit DMA capability check in megaraid_{mm,mbox} driver. - includes changes (going back to 32-bit DMA mask if 64-bit DMA mask failes) suggested by James with previous patch. - addition of SATA 150-4/6 as commented by Vasily Averin. With patch, the driver access PCIconfiguration space with dedicated offset to read a signature. If the signature read, it means that the controller has capability to handle 64-bit DMA. Without this patch, the driver used to blindly claim 64-bit DMA capability. The issue has been reported by Vasily Averin [vvs@sw.ru]. Thank you Vasily for the reporting. Signed-Off By: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-07-25 22:44:48 +08:00
uint32_t magic64;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
adapter->ito = MBOX_TIMEOUT;
pdev = adapter->pdev;
/*
* Allocate and initialize the init data structure for mailbox
* controllers
*/
some kmalloc/memset ->kzalloc (tree wide) Transform some calls to kmalloc/memset to a single kzalloc (or kcalloc). Here is a short excerpt of the semantic patch performing this transformation: @@ type T2; expression x; identifier f,fld; expression E; expression E1,E2; expression e1,e2,e3,y; statement S; @@ x = - kmalloc + kzalloc (E1,E2) ... when != \(x->fld=E;\|y=f(...,x,...);\|f(...,x,...);\|x=E;\|while(...) S\|for(e1;e2;e3) S\) - memset((T2)x,0,E1); @@ expression E1,E2,E3; @@ - kzalloc(E1 * E2,E3) + kcalloc(E1,E2,E3) [akpm@linux-foundation.org: get kcalloc args the right way around] Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Acked-by: Russell King <rmk@arm.linux.org.uk> Cc: Bryan Wu <bryan.wu@analog.com> Acked-by: Jiri Slaby <jirislaby@gmail.com> Cc: Dave Airlie <airlied@linux.ie> Acked-by: Roland Dreier <rolandd@cisco.com> Cc: Jiri Kosina <jkosina@suse.cz> Acked-by: Dmitry Torokhov <dtor@mail.ru> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Acked-by: Mauro Carvalho Chehab <mchehab@infradead.org> Acked-by: Pierre Ossman <drzeus-list@drzeus.cx> Cc: Jeff Garzik <jeff@garzik.org> Cc: "David S. Miller" <davem@davemloft.net> Acked-by: Greg KH <greg@kroah.com> Cc: James Bottomley <James.Bottomley@steeleye.com> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-19 16:49:03 +08:00
raid_dev = kzalloc(sizeof(mraid_device_t), GFP_KERNEL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (raid_dev == NULL) return -1;
/*
* Attach the adapter soft state to raid device soft state
*/
adapter->raid_device = (caddr_t)raid_dev;
raid_dev->fast_load = megaraid_fast_load;
// our baseport
raid_dev->baseport = pci_resource_start(pdev, 0);
if (pci_request_regions(pdev, "MegaRAID: LSI Logic Corporation") != 0) {
con_log(CL_ANN, (KERN_WARNING
"megaraid: mem region busy\n"));
goto out_free_raid_dev;
}
remove ioremap_nocache and devm_ioremap_nocache ioremap has provided non-cached semantics by default since the Linux 2.6 days, so remove the additional ioremap_nocache interface. Signed-off-by: Christoph Hellwig <hch@lst.de> Acked-by: Arnd Bergmann <arnd@arndb.de>
2020-01-06 16:43:50 +08:00
raid_dev->baseaddr = ioremap(raid_dev->baseport, 128);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (!raid_dev->baseaddr) {
con_log(CL_ANN, (KERN_WARNING
"megaraid: could not map hba memory\n") );
goto out_release_regions;
}
[SCSI] megaraid_{mm,mbox}: init fix for kdump 1. Changes in Initialization to fix kdump failure. Send SYNC command on loading. This command clears the pending commands in the adapter and re-initialize its internal RAID structure. Without this change, megaraid driver either panics or fails to initialize the adapter during kdump's second kernel boot if there are pending commands or interrupts from other devices sharing the same IRQ. 2. Authors email-id domain name changed from lsil.com to lsi.com. Also modified the MODULE_AUTHOR to megaraidlinux@lsi.com Signed-off-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-05 23:10:09 +08:00
/* initialize the mutual exclusion lock for the mailbox */
spin_lock_init(&raid_dev->mailbox_lock);
/* allocate memory required for commands */
if (megaraid_alloc_cmd_packets(adapter) != 0)
goto out_iounmap;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
[SCSI] megaraid_{mm,mbox}: init fix for kdump 1. Changes in Initialization to fix kdump failure. Send SYNC command on loading. This command clears the pending commands in the adapter and re-initialize its internal RAID structure. Without this change, megaraid driver either panics or fails to initialize the adapter during kdump's second kernel boot if there are pending commands or interrupts from other devices sharing the same IRQ. 2. Authors email-id domain name changed from lsil.com to lsi.com. Also modified the MODULE_AUTHOR to megaraidlinux@lsi.com Signed-off-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-05 23:10:09 +08:00
/*
* Issue SYNC cmd to flush the pending cmds in the adapter
* and initialize its internal state
*/
if (megaraid_mbox_fire_sync_cmd(adapter))
con_log(CL_ANN, ("megaraid: sync cmd failed\n"));
/*
* Setup the rest of the soft state using the library of
* FW routines
*/
/* request IRQ and register the interrupt service routine */
[PATCH] irq-flags: scsi: Use the new IRQF_ constants Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@elte.hu> Cc: "David S. Miller" <davem@davemloft.net> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: James Bottomley <James.Bottomley@steeleye.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-07-02 10:29:42 +08:00
if (request_irq(adapter->irq, megaraid_isr, IRQF_SHARED, "megaraid",
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
adapter)) {
con_log(CL_ANN, (KERN_WARNING
"megaraid: Couldn't register IRQ %d!\n", adapter->irq));
[SCSI] megaraid_{mm,mbox}: init fix for kdump 1. Changes in Initialization to fix kdump failure. Send SYNC command on loading. This command clears the pending commands in the adapter and re-initialize its internal RAID structure. Without this change, megaraid driver either panics or fails to initialize the adapter during kdump's second kernel boot if there are pending commands or interrupts from other devices sharing the same IRQ. 2. Authors email-id domain name changed from lsil.com to lsi.com. Also modified the MODULE_AUTHOR to megaraidlinux@lsi.com Signed-off-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-05 23:10:09 +08:00
goto out_alloc_cmds;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
// Product info
[SCSI] megaraid_{mm,mbox}: init fix for kdump 1. Changes in Initialization to fix kdump failure. Send SYNC command on loading. This command clears the pending commands in the adapter and re-initialize its internal RAID structure. Without this change, megaraid driver either panics or fails to initialize the adapter during kdump's second kernel boot if there are pending commands or interrupts from other devices sharing the same IRQ. 2. Authors email-id domain name changed from lsil.com to lsi.com. Also modified the MODULE_AUTHOR to megaraidlinux@lsi.com Signed-off-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-05 23:10:09 +08:00
if (megaraid_mbox_product_info(adapter) != 0)
goto out_free_irq;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
// Do we support extended CDBs
adapter->max_cdb_sz = 10;
if (megaraid_mbox_extended_cdb(adapter) == 0) {
adapter->max_cdb_sz = 16;
}
/*
* Do we support cluster environment, if we do, what is the initiator
* id.
* NOTE: In a non-cluster aware firmware environment, the LLD should
* return 7 as initiator id.
*/
adapter->ha = 0;
adapter->init_id = -1;
if (megaraid_mbox_support_ha(adapter, &adapter->init_id) == 0) {
adapter->ha = 1;
}
/*
* Prepare the device ids array to have the mapping between the kernel
* device address and megaraid device address.
* We export the physical devices on their actual addresses. The
* logical drives are exported on a virtual SCSI channel
*/
megaraid_mbox_setup_device_map(adapter);
// If the firmware supports random deletion, update the device id map
if (megaraid_mbox_support_random_del(adapter)) {
// Change the logical drives numbers in device_ids array one
// slot in device_ids is reserved for target id, that's why
// "<=" below
for (i = 0; i <= MAX_LOGICAL_DRIVES_40LD; i++) {
adapter->device_ids[adapter->max_channel][i] += 0x80;
}
adapter->device_ids[adapter->max_channel][adapter->init_id] =
0xFF;
raid_dev->random_del_supported = 1;
}
/*
* find out the maximum number of scatter-gather elements supported by
* this firmware
*/
adapter->sglen = megaraid_mbox_get_max_sg(adapter);
// enumerate RAID and SCSI channels so that all devices on SCSI
// channels can later be exported, including disk devices
megaraid_mbox_enum_raid_scsi(adapter);
/*
* Other parameters required by upper layer
*
* maximum number of sectors per IO command
*/
adapter->max_sectors = megaraid_max_sectors;
/*
* number of queued commands per LUN.
*/
adapter->cmd_per_lun = megaraid_cmd_per_lun;
/*
* Allocate resources required to issue FW calls, when sysfs is
* accessed
*/
[SCSI] megaraid_{mm,mbox}: init fix for kdump 1. Changes in Initialization to fix kdump failure. Send SYNC command on loading. This command clears the pending commands in the adapter and re-initialize its internal RAID structure. Without this change, megaraid driver either panics or fails to initialize the adapter during kdump's second kernel boot if there are pending commands or interrupts from other devices sharing the same IRQ. 2. Authors email-id domain name changed from lsil.com to lsi.com. Also modified the MODULE_AUTHOR to megaraidlinux@lsi.com Signed-off-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-05 23:10:09 +08:00
if (megaraid_sysfs_alloc_resources(adapter) != 0)
goto out_free_irq;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
// Set the DMA mask to 64-bit. All supported controllers as capable of
// DMA in this range
[SCSI] megaraid_{mm,mbox}: 64-bit DMA capability checker This patch contains - a fix for 64-bit DMA capability check in megaraid_{mm,mbox} driver. - includes changes (going back to 32-bit DMA mask if 64-bit DMA mask failes) suggested by James with previous patch. - addition of SATA 150-4/6 as commented by Vasily Averin. With patch, the driver access PCIconfiguration space with dedicated offset to read a signature. If the signature read, it means that the controller has capability to handle 64-bit DMA. Without this patch, the driver used to blindly claim 64-bit DMA capability. The issue has been reported by Vasily Averin [vvs@sw.ru]. Thank you Vasily for the reporting. Signed-Off By: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-07-25 22:44:48 +08:00
pci_read_config_dword(adapter->pdev, PCI_CONF_AMISIG64, &magic64);
if (((magic64 == HBA_SIGNATURE_64_BIT) &&
((adapter->pdev->subsystem_device !=
[PATCH] scsi: megaraid_{mm,mbox}: 64-bit DMA capability fix It is known that 2 LSI Logic MegaRAID SATA RAID Controllers (150-4 and 150-6) don't support 64-bit DMA. Unfortunately currently this check is wrong and driver sets 64-bit DMA mode for these devices. Signed-off-by: Andrey Mirkin <amirkin@sw.ru> Acked-by: Vasily Averin <vvs@sw.ru> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-16 16:08:43 +08:00
PCI_SUBSYS_ID_MEGARAID_SATA_150_6) &&
[SCSI] megaraid_{mm,mbox}: 64-bit DMA capability checker This patch contains - a fix for 64-bit DMA capability check in megaraid_{mm,mbox} driver. - includes changes (going back to 32-bit DMA mask if 64-bit DMA mask failes) suggested by James with previous patch. - addition of SATA 150-4/6 as commented by Vasily Averin. With patch, the driver access PCIconfiguration space with dedicated offset to read a signature. If the signature read, it means that the controller has capability to handle 64-bit DMA. Without this patch, the driver used to blindly claim 64-bit DMA capability. The issue has been reported by Vasily Averin [vvs@sw.ru]. Thank you Vasily for the reporting. Signed-Off By: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-07-25 22:44:48 +08:00
(adapter->pdev->subsystem_device !=
PCI_SUBSYS_ID_MEGARAID_SATA_150_4))) ||
(adapter->pdev->vendor == PCI_VENDOR_ID_LSI_LOGIC &&
adapter->pdev->device == PCI_DEVICE_ID_VERDE) ||
(adapter->pdev->vendor == PCI_VENDOR_ID_LSI_LOGIC &&
adapter->pdev->device == PCI_DEVICE_ID_DOBSON) ||
(adapter->pdev->vendor == PCI_VENDOR_ID_LSI_LOGIC &&
adapter->pdev->device == PCI_DEVICE_ID_LINDSAY) ||
(adapter->pdev->vendor == PCI_VENDOR_ID_DELL &&
adapter->pdev->device == PCI_DEVICE_ID_PERC4_DI_EVERGLADES) ||
(adapter->pdev->vendor == PCI_VENDOR_ID_DELL &&
adapter->pdev->device == PCI_DEVICE_ID_PERC4E_DI_KOBUK)) {
scsi: megaraid_mbox: switch to generic DMA API Switch from the legacy PCI DMA API to the generic DMA API. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-10-11 15:34:24 +08:00
if (dma_set_mask(&adapter->pdev->dev, DMA_BIT_MASK(64))) {
[SCSI] megaraid_{mm,mbox}: 64-bit DMA capability checker This patch contains - a fix for 64-bit DMA capability check in megaraid_{mm,mbox} driver. - includes changes (going back to 32-bit DMA mask if 64-bit DMA mask failes) suggested by James with previous patch. - addition of SATA 150-4/6 as commented by Vasily Averin. With patch, the driver access PCIconfiguration space with dedicated offset to read a signature. If the signature read, it means that the controller has capability to handle 64-bit DMA. Without this patch, the driver used to blindly claim 64-bit DMA capability. The issue has been reported by Vasily Averin [vvs@sw.ru]. Thank you Vasily for the reporting. Signed-Off By: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-07-25 22:44:48 +08:00
con_log(CL_ANN, (KERN_WARNING
"megaraid: DMA mask for 64-bit failed\n"));
scsi: megaraid_mbox: switch to generic DMA API Switch from the legacy PCI DMA API to the generic DMA API. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-10-11 15:34:24 +08:00
if (dma_set_mask(&adapter->pdev->dev,
DMA_BIT_MASK(32))) {
[SCSI] megaraid_{mm,mbox}: 64-bit DMA capability checker This patch contains - a fix for 64-bit DMA capability check in megaraid_{mm,mbox} driver. - includes changes (going back to 32-bit DMA mask if 64-bit DMA mask failes) suggested by James with previous patch. - addition of SATA 150-4/6 as commented by Vasily Averin. With patch, the driver access PCIconfiguration space with dedicated offset to read a signature. If the signature read, it means that the controller has capability to handle 64-bit DMA. Without this patch, the driver used to blindly claim 64-bit DMA capability. The issue has been reported by Vasily Averin [vvs@sw.ru]. Thank you Vasily for the reporting. Signed-Off By: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-07-25 22:44:48 +08:00
con_log(CL_ANN, (KERN_WARNING
"megaraid: 32-bit DMA mask failed\n"));
goto out_free_sysfs_res;
}
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
// setup tasklet for DPC
tasklet_init(&adapter->dpc_h, megaraid_mbox_dpc,
(unsigned long)adapter);
con_log(CL_DLEVEL1, (KERN_INFO
"megaraid mbox hba successfully initialized\n"));
return 0;
out_free_sysfs_res:
megaraid_sysfs_free_resources(adapter);
out_free_irq:
free_irq(adapter->irq, adapter);
[SCSI] megaraid_{mm,mbox}: init fix for kdump 1. Changes in Initialization to fix kdump failure. Send SYNC command on loading. This command clears the pending commands in the adapter and re-initialize its internal RAID structure. Without this change, megaraid driver either panics or fails to initialize the adapter during kdump's second kernel boot if there are pending commands or interrupts from other devices sharing the same IRQ. 2. Authors email-id domain name changed from lsil.com to lsi.com. Also modified the MODULE_AUTHOR to megaraidlinux@lsi.com Signed-off-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-05 23:10:09 +08:00
out_alloc_cmds:
megaraid_free_cmd_packets(adapter);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
out_iounmap:
iounmap(raid_dev->baseaddr);
out_release_regions:
pci_release_regions(pdev);
out_free_raid_dev:
kfree(raid_dev);
return -1;
}
/**
* megaraid_fini_mbox - undo controller initialization
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : our soft state
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static void
megaraid_fini_mbox(adapter_t *adapter)
{
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
// flush all caches
megaraid_mbox_flush_cache(adapter);
tasklet_kill(&adapter->dpc_h);
megaraid_sysfs_free_resources(adapter);
megaraid_free_cmd_packets(adapter);
free_irq(adapter->irq, adapter);
iounmap(raid_dev->baseaddr);
pci_release_regions(adapter->pdev);
kfree(raid_dev);
return;
}
/**
* megaraid_alloc_cmd_packets - allocate shared mailbox
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : soft state of the raid controller
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
scsi: megaraid: fix spelling mistake "maibox" -> "mailbox" Trivial fix to spelling mistake in warning message and comments Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-09-24 06:25:48 +08:00
* Allocate and align the shared mailbox. This mailbox is used to issue
scsi/megaraid fixed several typos in comments Fixed several typos in comments in megaraid_mbox.c, megaraid_mm.c and megaraid_sas_fusion.h. Signed-off-by: Matthias Schid <aircrach115@gmail.com> Signed-off-by: Stefan Huber <steffhip@gmail.com> Signed-off-by: Simon Puels <simon.puels@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2013-06-26 22:25:53 +08:00
* all the commands. For IO based controllers, the mailbox is also registered
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* with the FW. Allocate memory for all commands as well.
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* This is our big allocator.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static int
megaraid_alloc_cmd_packets(adapter_t *adapter)
{
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
struct pci_dev *pdev;
unsigned long align;
scb_t *scb;
mbox_ccb_t *ccb;
struct mraid_pci_blk *epthru_pci_blk;
struct mraid_pci_blk *sg_pci_blk;
struct mraid_pci_blk *mbox_pci_blk;
int i;
pdev = adapter->pdev;
/*
* Setup the mailbox
* Allocate the common 16-byte aligned memory for the handshake
* mailbox.
*/
cross-tree: phase out dma_zalloc_coherent() We already need to zero out memory for dma_alloc_coherent(), as such using dma_zalloc_coherent() is superflous. Phase it out. This change was generated with the following Coccinelle SmPL patch: @ replace_dma_zalloc_coherent @ expression dev, size, data, handle, flags; @@ -dma_zalloc_coherent(dev, size, handle, flags) +dma_alloc_coherent(dev, size, handle, flags) Suggested-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Luis Chamberlain <mcgrof@kernel.org> [hch: re-ran the script on the latest tree] Signed-off-by: Christoph Hellwig <hch@lst.de>
2019-01-04 16:23:09 +08:00
raid_dev->una_mbox64 = dma_alloc_coherent(&adapter->pdev->dev,
sizeof(mbox64_t),
&raid_dev->una_mbox64_dma,
GFP_KERNEL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (!raid_dev->una_mbox64) {
con_log(CL_ANN, (KERN_WARNING
[SCSI] replace __FUNCTION__ with __func__ [jejb: fixed up a ton of missed conversions. All of you are on notice this has happened, driver trees will now need to be rebased] Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Cc: SCSI List <linux-scsi@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-07-04 14:47:27 +08:00
"megaraid: out of memory, %s %d\n", __func__,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
__LINE__));
return -1;
}
/*
* Align the mailbox at 16-byte boundary
*/
raid_dev->mbox = &raid_dev->una_mbox64->mbox32;
raid_dev->mbox = (mbox_t *)((((unsigned long)raid_dev->mbox) + 15) &
(~0UL ^ 0xFUL));
raid_dev->mbox64 = (mbox64_t *)(((unsigned long)raid_dev->mbox) - 8);
align = ((void *)raid_dev->mbox -
((void *)&raid_dev->una_mbox64->mbox32));
raid_dev->mbox_dma = (unsigned long)raid_dev->una_mbox64_dma + 8 +
align;
// Allocate memory for commands issued internally
cross-tree: phase out dma_zalloc_coherent() We already need to zero out memory for dma_alloc_coherent(), as such using dma_zalloc_coherent() is superflous. Phase it out. This change was generated with the following Coccinelle SmPL patch: @ replace_dma_zalloc_coherent @ expression dev, size, data, handle, flags; @@ -dma_zalloc_coherent(dev, size, handle, flags) +dma_alloc_coherent(dev, size, handle, flags) Suggested-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Luis Chamberlain <mcgrof@kernel.org> [hch: re-ran the script on the latest tree] Signed-off-by: Christoph Hellwig <hch@lst.de>
2019-01-04 16:23:09 +08:00
adapter->ibuf = dma_alloc_coherent(&pdev->dev, MBOX_IBUF_SIZE,
&adapter->ibuf_dma_h, GFP_KERNEL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (!adapter->ibuf) {
con_log(CL_ANN, (KERN_WARNING
[SCSI] replace __FUNCTION__ with __func__ [jejb: fixed up a ton of missed conversions. All of you are on notice this has happened, driver trees will now need to be rebased] Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Cc: SCSI List <linux-scsi@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-07-04 14:47:27 +08:00
"megaraid: out of memory, %s %d\n", __func__,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
__LINE__));
goto out_free_common_mbox;
}
// Allocate memory for our SCSI Command Blocks and their associated
// memory
/*
* Allocate memory for the base list of scb. Later allocate memory for
* CCBs and embedded components of each CCB and point the pointers in
* scb to the allocated components
* NOTE: The code to allocate SCB will be duplicated in all the LLD
* since the calling routine does not yet know the number of available
* commands.
*/
some kmalloc/memset ->kzalloc (tree wide) Transform some calls to kmalloc/memset to a single kzalloc (or kcalloc). Here is a short excerpt of the semantic patch performing this transformation: @@ type T2; expression x; identifier f,fld; expression E; expression E1,E2; expression e1,e2,e3,y; statement S; @@ x = - kmalloc + kzalloc (E1,E2) ... when != \(x->fld=E;\|y=f(...,x,...);\|f(...,x,...);\|x=E;\|while(...) S\|for(e1;e2;e3) S\) - memset((T2)x,0,E1); @@ expression E1,E2,E3; @@ - kzalloc(E1 * E2,E3) + kcalloc(E1,E2,E3) [akpm@linux-foundation.org: get kcalloc args the right way around] Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Acked-by: Russell King <rmk@arm.linux.org.uk> Cc: Bryan Wu <bryan.wu@analog.com> Acked-by: Jiri Slaby <jirislaby@gmail.com> Cc: Dave Airlie <airlied@linux.ie> Acked-by: Roland Dreier <rolandd@cisco.com> Cc: Jiri Kosina <jkosina@suse.cz> Acked-by: Dmitry Torokhov <dtor@mail.ru> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Acked-by: Mauro Carvalho Chehab <mchehab@infradead.org> Acked-by: Pierre Ossman <drzeus-list@drzeus.cx> Cc: Jeff Garzik <jeff@garzik.org> Cc: "David S. Miller" <davem@davemloft.net> Acked-by: Greg KH <greg@kroah.com> Cc: James Bottomley <James.Bottomley@steeleye.com> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-19 16:49:03 +08:00
adapter->kscb_list = kcalloc(MBOX_MAX_SCSI_CMDS, sizeof(scb_t), GFP_KERNEL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (adapter->kscb_list == NULL) {
con_log(CL_ANN, (KERN_WARNING
[SCSI] replace __FUNCTION__ with __func__ [jejb: fixed up a ton of missed conversions. All of you are on notice this has happened, driver trees will now need to be rebased] Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Cc: SCSI List <linux-scsi@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-07-04 14:47:27 +08:00
"megaraid: out of memory, %s %d\n", __func__,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
__LINE__));
goto out_free_ibuf;
}
// memory allocation for our command packets
if (megaraid_mbox_setup_dma_pools(adapter) != 0) {
con_log(CL_ANN, (KERN_WARNING
[SCSI] replace __FUNCTION__ with __func__ [jejb: fixed up a ton of missed conversions. All of you are on notice this has happened, driver trees will now need to be rebased] Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Cc: SCSI List <linux-scsi@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-07-04 14:47:27 +08:00
"megaraid: out of memory, %s %d\n", __func__,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
__LINE__));
goto out_free_scb_list;
}
// Adjust the scb pointers and link in the free pool
epthru_pci_blk = raid_dev->epthru_pool;
sg_pci_blk = raid_dev->sg_pool;
mbox_pci_blk = raid_dev->mbox_pool;
for (i = 0; i < MBOX_MAX_SCSI_CMDS; i++) {
scb = adapter->kscb_list + i;
ccb = raid_dev->ccb_list + i;
ccb->mbox = (mbox_t *)(mbox_pci_blk[i].vaddr + 16);
ccb->raw_mbox = (uint8_t *)ccb->mbox;
ccb->mbox64 = (mbox64_t *)(mbox_pci_blk[i].vaddr + 8);
ccb->mbox_dma_h = (unsigned long)mbox_pci_blk[i].dma_addr + 16;
// make sure the mailbox is aligned properly
if (ccb->mbox_dma_h & 0x0F) {
con_log(CL_ANN, (KERN_CRIT
"megaraid mbox: not aligned on 16-bytes\n"));
goto out_teardown_dma_pools;
}
ccb->epthru = (mraid_epassthru_t *)
epthru_pci_blk[i].vaddr;
ccb->epthru_dma_h = epthru_pci_blk[i].dma_addr;
ccb->pthru = (mraid_passthru_t *)ccb->epthru;
ccb->pthru_dma_h = ccb->epthru_dma_h;
ccb->sgl64 = (mbox_sgl64 *)sg_pci_blk[i].vaddr;
ccb->sgl_dma_h = sg_pci_blk[i].dma_addr;
ccb->sgl32 = (mbox_sgl32 *)ccb->sgl64;
scb->ccb = (caddr_t)ccb;
scb->gp = 0;
scb->sno = i; // command index
scb->scp = NULL;
scb->state = SCB_FREE;
scsi: megaraid_mbox: switch to generic DMA API Switch from the legacy PCI DMA API to the generic DMA API. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-10-11 15:34:24 +08:00
scb->dma_direction = DMA_NONE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
scb->dma_type = MRAID_DMA_NONE;
scb->dev_channel = -1;
scb->dev_target = -1;
// put scb in the free pool
list_add_tail(&scb->list, &adapter->kscb_pool);
}
return 0;
out_teardown_dma_pools:
megaraid_mbox_teardown_dma_pools(adapter);
out_free_scb_list:
kfree(adapter->kscb_list);
out_free_ibuf:
scsi: megaraid_mbox: switch to generic DMA API Switch from the legacy PCI DMA API to the generic DMA API. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-10-11 15:34:24 +08:00
dma_free_coherent(&pdev->dev, MBOX_IBUF_SIZE, (void *)adapter->ibuf,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
adapter->ibuf_dma_h);
out_free_common_mbox:
scsi: megaraid_mbox: switch to generic DMA API Switch from the legacy PCI DMA API to the generic DMA API. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-10-11 15:34:24 +08:00
dma_free_coherent(&adapter->pdev->dev, sizeof(mbox64_t),
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
(caddr_t)raid_dev->una_mbox64, raid_dev->una_mbox64_dma);
return -1;
}
/**
* megaraid_free_cmd_packets - free memory
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : soft state of the raid controller
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Release memory resources allocated for commands.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static void
megaraid_free_cmd_packets(adapter_t *adapter)
{
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
megaraid_mbox_teardown_dma_pools(adapter);
kfree(adapter->kscb_list);
scsi: megaraid_mbox: switch to generic DMA API Switch from the legacy PCI DMA API to the generic DMA API. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-10-11 15:34:24 +08:00
dma_free_coherent(&adapter->pdev->dev, MBOX_IBUF_SIZE,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
(void *)adapter->ibuf, adapter->ibuf_dma_h);
scsi: megaraid_mbox: switch to generic DMA API Switch from the legacy PCI DMA API to the generic DMA API. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-10-11 15:34:24 +08:00
dma_free_coherent(&adapter->pdev->dev, sizeof(mbox64_t),
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
(caddr_t)raid_dev->una_mbox64, raid_dev->una_mbox64_dma);
return;
}
/**
* megaraid_mbox_setup_dma_pools - setup dma pool for command packets
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : HBA soft state
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Setup the dma pools for mailbox, passthru and extended passthru structures,
* and scatter-gather lists.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static int
megaraid_mbox_setup_dma_pools(adapter_t *adapter)
{
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
struct mraid_pci_blk *epthru_pci_blk;
struct mraid_pci_blk *sg_pci_blk;
struct mraid_pci_blk *mbox_pci_blk;
int i;
// Allocate memory for 16-bytes aligned mailboxes
scsi: megaraid: Replace PCI pool old API The PCI pool API is deprecated. This commit replaces the PCI pool old API by the appropriate function with the DMA pool API. Signed-off-by: Romain Perier <romain.perier@collabora.com> Reviewed-by: Peter Senna Tschudin <peter.senna@collabora.com> Acked-by: Sumit Saxena <sumit.saxena@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-07-06 16:13:06 +08:00
raid_dev->mbox_pool_handle = dma_pool_create("megaraid mbox pool",
&adapter->pdev->dev,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
sizeof(mbox64_t) + 16,
16, 0);
if (raid_dev->mbox_pool_handle == NULL) {
goto fail_setup_dma_pool;
}
mbox_pci_blk = raid_dev->mbox_pool;
for (i = 0; i < MBOX_MAX_SCSI_CMDS; i++) {
scsi: megaraid: Replace PCI pool old API The PCI pool API is deprecated. This commit replaces the PCI pool old API by the appropriate function with the DMA pool API. Signed-off-by: Romain Perier <romain.perier@collabora.com> Reviewed-by: Peter Senna Tschudin <peter.senna@collabora.com> Acked-by: Sumit Saxena <sumit.saxena@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-07-06 16:13:06 +08:00
mbox_pci_blk[i].vaddr = dma_pool_alloc(
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
raid_dev->mbox_pool_handle,
GFP_KERNEL,
&mbox_pci_blk[i].dma_addr);
if (!mbox_pci_blk[i].vaddr) {
goto fail_setup_dma_pool;
}
}
/*
* Allocate memory for each embedded passthru strucuture pointer
* Request for a 128 bytes aligned structure for each passthru command
* structure
* Since passthru and extended passthru commands are exclusive, they
* share common memory pool. Passthru structures piggyback on memory
* allocted to extended passthru since passthru is smaller of the two
*/
scsi: megaraid: Replace PCI pool old API The PCI pool API is deprecated. This commit replaces the PCI pool old API by the appropriate function with the DMA pool API. Signed-off-by: Romain Perier <romain.perier@collabora.com> Reviewed-by: Peter Senna Tschudin <peter.senna@collabora.com> Acked-by: Sumit Saxena <sumit.saxena@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-07-06 16:13:06 +08:00
raid_dev->epthru_pool_handle = dma_pool_create("megaraid mbox pthru",
&adapter->pdev->dev, sizeof(mraid_epassthru_t), 128, 0);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (raid_dev->epthru_pool_handle == NULL) {
goto fail_setup_dma_pool;
}
epthru_pci_blk = raid_dev->epthru_pool;
for (i = 0; i < MBOX_MAX_SCSI_CMDS; i++) {
scsi: megaraid: Replace PCI pool old API The PCI pool API is deprecated. This commit replaces the PCI pool old API by the appropriate function with the DMA pool API. Signed-off-by: Romain Perier <romain.perier@collabora.com> Reviewed-by: Peter Senna Tschudin <peter.senna@collabora.com> Acked-by: Sumit Saxena <sumit.saxena@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-07-06 16:13:06 +08:00
epthru_pci_blk[i].vaddr = dma_pool_alloc(
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
raid_dev->epthru_pool_handle,
GFP_KERNEL,
&epthru_pci_blk[i].dma_addr);
if (!epthru_pci_blk[i].vaddr) {
goto fail_setup_dma_pool;
}
}
// Allocate memory for each scatter-gather list. Request for 512 bytes
// alignment for each sg list
scsi: megaraid: Replace PCI pool old API The PCI pool API is deprecated. This commit replaces the PCI pool old API by the appropriate function with the DMA pool API. Signed-off-by: Romain Perier <romain.perier@collabora.com> Reviewed-by: Peter Senna Tschudin <peter.senna@collabora.com> Acked-by: Sumit Saxena <sumit.saxena@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-07-06 16:13:06 +08:00
raid_dev->sg_pool_handle = dma_pool_create("megaraid mbox sg",
&adapter->pdev->dev,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
sizeof(mbox_sgl64) * MBOX_MAX_SG_SIZE,
512, 0);
if (raid_dev->sg_pool_handle == NULL) {
goto fail_setup_dma_pool;
}
sg_pci_blk = raid_dev->sg_pool;
for (i = 0; i < MBOX_MAX_SCSI_CMDS; i++) {
scsi: megaraid: Replace PCI pool old API The PCI pool API is deprecated. This commit replaces the PCI pool old API by the appropriate function with the DMA pool API. Signed-off-by: Romain Perier <romain.perier@collabora.com> Reviewed-by: Peter Senna Tschudin <peter.senna@collabora.com> Acked-by: Sumit Saxena <sumit.saxena@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-07-06 16:13:06 +08:00
sg_pci_blk[i].vaddr = dma_pool_alloc(
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
raid_dev->sg_pool_handle,
GFP_KERNEL,
&sg_pci_blk[i].dma_addr);
if (!sg_pci_blk[i].vaddr) {
goto fail_setup_dma_pool;
}
}
return 0;
fail_setup_dma_pool:
megaraid_mbox_teardown_dma_pools(adapter);
return -1;
}
/**
* megaraid_mbox_teardown_dma_pools - teardown dma pools for command packets
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : HBA soft state
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Teardown the dma pool for mailbox, passthru and extended passthru
* structures, and scatter-gather lists.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static void
megaraid_mbox_teardown_dma_pools(adapter_t *adapter)
{
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
struct mraid_pci_blk *epthru_pci_blk;
struct mraid_pci_blk *sg_pci_blk;
struct mraid_pci_blk *mbox_pci_blk;
int i;
sg_pci_blk = raid_dev->sg_pool;
for (i = 0; i < MBOX_MAX_SCSI_CMDS && sg_pci_blk[i].vaddr; i++) {
scsi: megaraid: Replace PCI pool old API The PCI pool API is deprecated. This commit replaces the PCI pool old API by the appropriate function with the DMA pool API. Signed-off-by: Romain Perier <romain.perier@collabora.com> Reviewed-by: Peter Senna Tschudin <peter.senna@collabora.com> Acked-by: Sumit Saxena <sumit.saxena@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-07-06 16:13:06 +08:00
dma_pool_free(raid_dev->sg_pool_handle, sg_pci_blk[i].vaddr,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
sg_pci_blk[i].dma_addr);
}
scsi: megaraid_sas: NULL check before some freeing functions is not needed NULL check before some freeing functions is not needed. Signed-off-by: Thomas Meyer <thomas@m3y3r.de> Acked-by: Sumit Saxena <sumit.saxena@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-12-03 04:52:11 +08:00
dma_pool_destroy(raid_dev->sg_pool_handle);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
epthru_pci_blk = raid_dev->epthru_pool;
for (i = 0; i < MBOX_MAX_SCSI_CMDS && epthru_pci_blk[i].vaddr; i++) {
scsi: megaraid: Replace PCI pool old API The PCI pool API is deprecated. This commit replaces the PCI pool old API by the appropriate function with the DMA pool API. Signed-off-by: Romain Perier <romain.perier@collabora.com> Reviewed-by: Peter Senna Tschudin <peter.senna@collabora.com> Acked-by: Sumit Saxena <sumit.saxena@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-07-06 16:13:06 +08:00
dma_pool_free(raid_dev->epthru_pool_handle,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
epthru_pci_blk[i].vaddr, epthru_pci_blk[i].dma_addr);
}
scsi: megaraid_sas: NULL check before some freeing functions is not needed NULL check before some freeing functions is not needed. Signed-off-by: Thomas Meyer <thomas@m3y3r.de> Acked-by: Sumit Saxena <sumit.saxena@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-12-03 04:52:11 +08:00
dma_pool_destroy(raid_dev->epthru_pool_handle);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
mbox_pci_blk = raid_dev->mbox_pool;
for (i = 0; i < MBOX_MAX_SCSI_CMDS && mbox_pci_blk[i].vaddr; i++) {
scsi: megaraid: Replace PCI pool old API The PCI pool API is deprecated. This commit replaces the PCI pool old API by the appropriate function with the DMA pool API. Signed-off-by: Romain Perier <romain.perier@collabora.com> Reviewed-by: Peter Senna Tschudin <peter.senna@collabora.com> Acked-by: Sumit Saxena <sumit.saxena@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-07-06 16:13:06 +08:00
dma_pool_free(raid_dev->mbox_pool_handle,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
mbox_pci_blk[i].vaddr, mbox_pci_blk[i].dma_addr);
}
scsi: megaraid_sas: NULL check before some freeing functions is not needed NULL check before some freeing functions is not needed. Signed-off-by: Thomas Meyer <thomas@m3y3r.de> Acked-by: Sumit Saxena <sumit.saxena@broadcom.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-12-03 04:52:11 +08:00
dma_pool_destroy(raid_dev->mbox_pool_handle);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return;
}
/**
* megaraid_alloc_scb - detach and return a scb from the free list
* @adapter : controller's soft state
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @scp : pointer to the scsi command to be executed
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Return the scb from the head of the free list. %NULL if there are none
* available.
*/
[PATCH] Unlinline a bunch of other functions Remove the "inline" keyword from a bunch of big functions in the kernel with the goal of shrinking it by 30kb to 40kb Signed-off-by: Arjan van de Ven <arjan@infradead.org> Signed-off-by: Ingo Molnar <mingo@elte.hu> Acked-by: Jeff Garzik <jgarzik@pobox.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-15 05:20:43 +08:00
static scb_t *
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
megaraid_alloc_scb(adapter_t *adapter, struct scsi_cmnd *scp)
{
struct list_head *head = &adapter->kscb_pool;
scb_t *scb = NULL;
unsigned long flags;
// detach scb from free pool
spin_lock_irqsave(SCSI_FREE_LIST_LOCK(adapter), flags);
if (list_empty(head)) {
spin_unlock_irqrestore(SCSI_FREE_LIST_LOCK(adapter), flags);
return NULL;
}
scb = list_entry(head->next, scb_t, list);
list_del_init(&scb->list);
spin_unlock_irqrestore(SCSI_FREE_LIST_LOCK(adapter), flags);
scb->state = SCB_ACTIVE;
scb->scp = scp;
scb->dma_type = MRAID_DMA_NONE;
return scb;
}
/**
* megaraid_dealloc_scb - return the scb to the free pool
* @adapter : controller's soft state
* @scb : scb to be freed
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Return the scb back to the free list of scbs. The caller must 'flush' the
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* SCB before calling us. E.g., performing pci_unamp and/or pci_sync etc.
* NOTE NOTE: Make sure the scb is not on any list before calling this
* routine.
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
*/
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
static inline void
megaraid_dealloc_scb(adapter_t *adapter, scb_t *scb)
{
unsigned long flags;
// put scb in the free pool
scb->state = SCB_FREE;
scb->scp = NULL;
spin_lock_irqsave(SCSI_FREE_LIST_LOCK(adapter), flags);
list_add(&scb->list, &adapter->kscb_pool);
spin_unlock_irqrestore(SCSI_FREE_LIST_LOCK(adapter), flags);
return;
}
/**
* megaraid_mbox_mksgl - make the scatter-gather list
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : controller's soft state
* @scb : scsi control block
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Prepare the scatter-gather list.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
[PATCH] Unlinline a bunch of other functions Remove the "inline" keyword from a bunch of big functions in the kernel with the goal of shrinking it by 30kb to 40kb Signed-off-by: Arjan van de Ven <arjan@infradead.org> Signed-off-by: Ingo Molnar <mingo@elte.hu> Acked-by: Jeff Garzik <jgarzik@pobox.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-15 05:20:43 +08:00
static int
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
megaraid_mbox_mksgl(adapter_t *adapter, scb_t *scb)
{
struct scatterlist *sgl;
mbox_ccb_t *ccb;
struct scsi_cmnd *scp;
int sgcnt;
int i;
scp = scb->scp;
ccb = (mbox_ccb_t *)scb->ccb;
[SCSI] megaraid: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:13:44 +08:00
sgcnt = scsi_dma_map(scp);
BUG_ON(sgcnt < 0 || sgcnt > adapter->sglen);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
// no mapping required if no data to be transferred
[SCSI] megaraid: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:13:44 +08:00
if (!sgcnt)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return 0;
scb->dma_type = MRAID_DMA_WSG;
[SCSI] megaraid: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:13:44 +08:00
scsi_for_each_sg(scp, sgl, sgcnt, i) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
ccb->sgl64[i].address = sg_dma_address(sgl);
ccb->sgl64[i].length = sg_dma_len(sgl);
}
// Return count of SG nodes
return sgcnt;
}
/**
* mbox_post_cmd - issue a mailbox command
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : controller's soft state
* @scb : command to be issued
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Post the command to the controller if mailbox is available.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
[PATCH] Unlinline a bunch of other functions Remove the "inline" keyword from a bunch of big functions in the kernel with the goal of shrinking it by 30kb to 40kb Signed-off-by: Arjan van de Ven <arjan@infradead.org> Signed-off-by: Ingo Molnar <mingo@elte.hu> Acked-by: Jeff Garzik <jgarzik@pobox.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-15 05:20:43 +08:00
static int
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
mbox_post_cmd(adapter_t *adapter, scb_t *scb)
{
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
mbox64_t *mbox64;
mbox_t *mbox;
mbox_ccb_t *ccb;
unsigned long flags;
unsigned int i = 0;
ccb = (mbox_ccb_t *)scb->ccb;
mbox = raid_dev->mbox;
mbox64 = raid_dev->mbox64;
/*
* Check for busy mailbox. If it is, return failure - the caller
* should retry later.
*/
spin_lock_irqsave(MAILBOX_LOCK(raid_dev), flags);
if (unlikely(mbox->busy)) {
do {
udelay(1);
i++;
rmb();
} while(mbox->busy && (i < max_mbox_busy_wait));
if (mbox->busy) {
spin_unlock_irqrestore(MAILBOX_LOCK(raid_dev), flags);
return -1;
}
}
// Copy this command's mailbox data into "adapter's" mailbox
memcpy((caddr_t)mbox64, (caddr_t)ccb->mbox64, 22);
mbox->cmdid = scb->sno;
adapter->outstanding_cmds++;
mbox->busy = 1; // Set busy
mbox->poll = 0;
mbox->ack = 0;
wmb();
WRINDOOR(raid_dev, raid_dev->mbox_dma | 0x1);
spin_unlock_irqrestore(MAILBOX_LOCK(raid_dev), flags);
return 0;
}
/**
* megaraid_queue_command - generic queue entry point for all LLDs
* @scp : pointer to the scsi command to be executed
* @done : callback routine to be called after the cmd has be completed
*
* Queue entry point for mailbox based controllers.
*/
static int
SCSI host lock push-down Move the mid-layer's ->queuecommand() invocation from being locked with the host lock to being unlocked to facilitate speeding up the critical path for drivers who don't need this lock taken anyway. The patch below presents a simple SCSI host lock push-down as an equivalent transformation. No locking or other behavior should change with this patch. All existing bugs and locking orders are preserved. Additionally, add one parameter to queuecommand, struct Scsi_Host * and remove one parameter from queuecommand, void (*done)(struct scsi_cmnd *) Scsi_Host* is a convenient pointer that most host drivers need anyway, and 'done' is redundant to struct scsi_cmnd->scsi_done. Minimal code disturbance was attempted with this change. Most drivers needed only two one-line modifications for their host lock push-down. Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Acked-by: James Bottomley <James.Bottomley@suse.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-11-16 15:10:29 +08:00
megaraid_queue_command_lck(struct scsi_cmnd *scp, void (*done)(struct scsi_cmnd *))
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
adapter_t *adapter;
scb_t *scb;
int if_busy;
adapter = SCP2ADAPTER(scp);
scp->scsi_done = done;
scp->result = 0;
/*
* Allocate and build a SCB request
* if_busy flag will be set if megaraid_mbox_build_cmd() command could
* not allocate scb. We will return non-zero status in that case.
* NOTE: scb can be null even though certain commands completed
* successfully, e.g., MODE_SENSE and TEST_UNIT_READY, it would
* return 0 in that case, and we would do the callback right away.
*/
if_busy = 0;
[SCSI] megaraid_mbox: remove scsi_assign_lock usage also remove the adapter->host_lock alias for adapter->lock and remove some superflous locking aswell as removing the tiny locking wrappers for the EH routines. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 04:06:02 +08:00
scb = megaraid_mbox_build_cmd(adapter, scp, &if_busy);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (!scb) { // command already completed
done(scp);
return 0;
}
[SCSI] megaraid_mbox: remove scsi_assign_lock usage also remove the adapter->host_lock alias for adapter->lock and remove some superflous locking aswell as removing the tiny locking wrappers for the EH routines. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 04:06:02 +08:00
megaraid_mbox_runpendq(adapter, scb);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return if_busy;
}
SCSI host lock push-down Move the mid-layer's ->queuecommand() invocation from being locked with the host lock to being unlocked to facilitate speeding up the critical path for drivers who don't need this lock taken anyway. The patch below presents a simple SCSI host lock push-down as an equivalent transformation. No locking or other behavior should change with this patch. All existing bugs and locking orders are preserved. Additionally, add one parameter to queuecommand, struct Scsi_Host * and remove one parameter from queuecommand, void (*done)(struct scsi_cmnd *) Scsi_Host* is a convenient pointer that most host drivers need anyway, and 'done' is redundant to struct scsi_cmnd->scsi_done. Minimal code disturbance was attempted with this change. Most drivers needed only two one-line modifications for their host lock push-down. Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Acked-by: James Bottomley <James.Bottomley@suse.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-11-16 15:10:29 +08:00
static DEF_SCSI_QCMD(megaraid_queue_command)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/**
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* megaraid_mbox_build_cmd - transform the mid-layer scsi commands
* @adapter : controller's soft state
* @scp : mid-layer scsi command pointer
* @busy : set if request could not be completed because of lack of
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* resources
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Transform the mid-layer scsi command to megaraid firmware lingua.
* Convert the command issued by mid-layer to format understood by megaraid
* firmware. We also complete certain commands without sending them to firmware.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static scb_t *
megaraid_mbox_build_cmd(adapter_t *adapter, struct scsi_cmnd *scp, int *busy)
{
mraid_device_t *rdev = ADAP2RAIDDEV(adapter);
int channel;
int target;
int islogical;
mbox_ccb_t *ccb;
mraid_passthru_t *pthru;
mbox64_t *mbox64;
mbox_t *mbox;
scb_t *scb;
char skip[] = "skipping";
char scan[] = "scanning";
char *ss;
/*
* Get the appropriate device map for the device this command is
* intended for
*/
MRAID_GET_DEVICE_MAP(adapter, scp, channel, target, islogical);
/*
* Logical drive commands
*/
if (islogical) {
switch (scp->cmnd[0]) {
case TEST_UNIT_READY:
/*
* Do we support clustering and is the support enabled
* If no, return success always
*/
if (!adapter->ha) {
scp->result = (DID_OK << 16);
return NULL;
}
if (!(scb = megaraid_alloc_scb(adapter, scp))) {
scp->result = (DID_ERROR << 16);
*busy = 1;
return NULL;
}
scb->dma_direction = scp->sc_data_direction;
scb->dev_channel = 0xFF;
scb->dev_target = target;
ccb = (mbox_ccb_t *)scb->ccb;
/*
* The command id will be provided by the command
* issuance routine
*/
ccb->raw_mbox[0] = CLUSTER_CMD;
ccb->raw_mbox[2] = RESERVATION_STATUS;
ccb->raw_mbox[3] = target;
return scb;
case MODE_SENSE:
[SCSI] megaraid: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:13:44 +08:00
{
struct scatterlist *sgl;
caddr_t vaddr;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
[SCSI] megaraid: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:13:44 +08:00
sgl = scsi_sglist(scp);
[SG] Update drivers to use sg helpers Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2007-10-23 03:19:53 +08:00
if (sg_page(sgl)) {
vaddr = (caddr_t) sg_virt(&sgl[0]);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
[SCSI] megaraid: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:13:44 +08:00
memset(vaddr, 0, scp->cmnd[4]);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
else {
[SCSI] megaraid: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:13:44 +08:00
con_log(CL_ANN, (KERN_WARNING
"megaraid mailbox: invalid sg:%d\n",
__LINE__));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
[SCSI] megaraid: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:13:44 +08:00
}
scp->result = (DID_OK << 16);
return NULL;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
case INQUIRY:
/*
* Display the channel scan for logical drives
* Do not display scan for a channel if already done.
*/
if (!(rdev->last_disp & (1L << SCP2CHANNEL(scp)))) {
con_log(CL_ANN, (KERN_INFO
"scsi[%d]: scanning scsi channel %d",
adapter->host->host_no,
SCP2CHANNEL(scp)));
con_log(CL_ANN, (
" [virtual] for logical drives\n"));
rdev->last_disp |= (1L << SCP2CHANNEL(scp));
}
[SCSI] megaraid_{mm,mbox}: a fix on INQUIRY with EVPD With this patch, driver will protect data corruption created by INQUIRY with EVPD request to megaraid controllers. As specified in the changelog, megaraid F/W already has fixed the issue and being under process of release. Meanwhile, driver will protect the system with this patch. Signed-Off By: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-07-25 22:44:58 +08:00
if (scp->cmnd[1] & MEGA_SCSI_INQ_EVPD) {
scp->sense_buffer[0] = 0x70;
scp->sense_buffer[2] = ILLEGAL_REQUEST;
scp->sense_buffer[12] = MEGA_INVALID_FIELD_IN_CDB;
scp->result = CHECK_CONDITION << 1;
return NULL;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/* Fall through */
case READ_CAPACITY:
/*
* Do not allow LUN > 0 for logical drives and
* requests for more than 40 logical drives
*/
if (SCP2LUN(scp)) {
scp->result = (DID_BAD_TARGET << 16);
return NULL;
}
if ((target % 0x80) >= MAX_LOGICAL_DRIVES_40LD) {
scp->result = (DID_BAD_TARGET << 16);
return NULL;
}
/* Allocate a SCB and initialize passthru */
if (!(scb = megaraid_alloc_scb(adapter, scp))) {
scp->result = (DID_ERROR << 16);
*busy = 1;
return NULL;
}
ccb = (mbox_ccb_t *)scb->ccb;
scb->dev_channel = 0xFF;
scb->dev_target = target;
pthru = ccb->pthru;
mbox = ccb->mbox;
mbox64 = ccb->mbox64;
pthru->timeout = 0;
pthru->ars = 1;
pthru->reqsenselen = 14;
pthru->islogical = 1;
pthru->logdrv = target;
pthru->cdblen = scp->cmd_len;
memcpy(pthru->cdb, scp->cmnd, scp->cmd_len);
mbox->cmd = MBOXCMD_PASSTHRU64;
scb->dma_direction = scp->sc_data_direction;
[SCSI] megaraid: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:13:44 +08:00
pthru->dataxferlen = scsi_bufflen(scp);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
pthru->dataxferaddr = ccb->sgl_dma_h;
pthru->numsge = megaraid_mbox_mksgl(adapter,
scb);
mbox->xferaddr = 0xFFFFFFFF;
mbox64->xferaddr_lo = (uint32_t )ccb->pthru_dma_h;
mbox64->xferaddr_hi = 0;
return scb;
case READ_6:
case WRITE_6:
case READ_10:
case WRITE_10:
case READ_12:
case WRITE_12:
/*
* Allocate a SCB and initialize mailbox
*/
if (!(scb = megaraid_alloc_scb(adapter, scp))) {
scp->result = (DID_ERROR << 16);
*busy = 1;
return NULL;
}
ccb = (mbox_ccb_t *)scb->ccb;
scb->dev_channel = 0xFF;
scb->dev_target = target;
mbox = ccb->mbox;
mbox64 = ccb->mbox64;
mbox->logdrv = target;
/*
* A little HACK: 2nd bit is zero for all scsi read
* commands and is set for all scsi write commands
*/
mbox->cmd = (scp->cmnd[0] & 0x02) ? MBOXCMD_LWRITE64:
MBOXCMD_LREAD64 ;
/*
* 6-byte READ(0x08) or WRITE(0x0A) cdb
*/
if (scp->cmd_len == 6) {
mbox->numsectors = (uint32_t)scp->cmnd[4];
mbox->lba =
((uint32_t)scp->cmnd[1] << 16) |
((uint32_t)scp->cmnd[2] << 8) |
(uint32_t)scp->cmnd[3];
mbox->lba &= 0x1FFFFF;
}
/*
* 10-byte READ(0x28) or WRITE(0x2A) cdb
*/
else if (scp->cmd_len == 10) {
mbox->numsectors =
(uint32_t)scp->cmnd[8] |
((uint32_t)scp->cmnd[7] << 8);
mbox->lba =
((uint32_t)scp->cmnd[2] << 24) |
((uint32_t)scp->cmnd[3] << 16) |
((uint32_t)scp->cmnd[4] << 8) |
(uint32_t)scp->cmnd[5];
}
/*
* 12-byte READ(0xA8) or WRITE(0xAA) cdb
*/
else if (scp->cmd_len == 12) {
mbox->lba =
((uint32_t)scp->cmnd[2] << 24) |
((uint32_t)scp->cmnd[3] << 16) |
((uint32_t)scp->cmnd[4] << 8) |
(uint32_t)scp->cmnd[5];
mbox->numsectors =
((uint32_t)scp->cmnd[6] << 24) |
((uint32_t)scp->cmnd[7] << 16) |
((uint32_t)scp->cmnd[8] << 8) |
(uint32_t)scp->cmnd[9];
}
else {
con_log(CL_ANN, (KERN_WARNING
"megaraid: unsupported CDB length\n"));
megaraid_dealloc_scb(adapter, scb);
scp->result = (DID_ERROR << 16);
return NULL;
}
scb->dma_direction = scp->sc_data_direction;
// Calculate Scatter-Gather info
mbox64->xferaddr_lo = (uint32_t )ccb->sgl_dma_h;
mbox->numsge = megaraid_mbox_mksgl(adapter,
scb);
mbox->xferaddr = 0xFFFFFFFF;
mbox64->xferaddr_hi = 0;
return scb;
case RESERVE:
case RELEASE:
/*
* Do we support clustering and is the support enabled
*/
if (!adapter->ha) {
scp->result = (DID_BAD_TARGET << 16);
return NULL;
}
/*
* Allocate a SCB and initialize mailbox
*/
if (!(scb = megaraid_alloc_scb(adapter, scp))) {
scp->result = (DID_ERROR << 16);
*busy = 1;
return NULL;
}
ccb = (mbox_ccb_t *)scb->ccb;
scb->dev_channel = 0xFF;
scb->dev_target = target;
ccb->raw_mbox[0] = CLUSTER_CMD;
ccb->raw_mbox[2] = (scp->cmnd[0] == RESERVE) ?
RESERVE_LD : RELEASE_LD;
ccb->raw_mbox[3] = target;
scb->dma_direction = scp->sc_data_direction;
return scb;
default:
scp->result = (DID_BAD_TARGET << 16);
return NULL;
}
}
else { // Passthru device commands
// Do not allow access to target id > 15 or LUN > 7
if (target > 15 || SCP2LUN(scp) > 7) {
scp->result = (DID_BAD_TARGET << 16);
return NULL;
}
// if fast load option was set and scan for last device is
// over, reset the fast_load flag so that during a possible
// next scan, devices can be made available
if (rdev->fast_load && (target == 15) &&
(SCP2CHANNEL(scp) == adapter->max_channel -1)) {
con_log(CL_ANN, (KERN_INFO
"megaraid[%d]: physical device scan re-enabled\n",
adapter->host->host_no));
rdev->fast_load = 0;
}
/*
* Display the channel scan for physical devices
*/
if (!(rdev->last_disp & (1L << SCP2CHANNEL(scp)))) {
ss = rdev->fast_load ? skip : scan;
con_log(CL_ANN, (KERN_INFO
"scsi[%d]: %s scsi channel %d [Phy %d]",
adapter->host->host_no, ss, SCP2CHANNEL(scp),
channel));
con_log(CL_ANN, (
" for non-raid devices\n"));
rdev->last_disp |= (1L << SCP2CHANNEL(scp));
}
// disable channel sweep if fast load option given
if (rdev->fast_load) {
scp->result = (DID_BAD_TARGET << 16);
return NULL;
}
// Allocate a SCB and initialize passthru
if (!(scb = megaraid_alloc_scb(adapter, scp))) {
scp->result = (DID_ERROR << 16);
*busy = 1;
return NULL;
}
ccb = (mbox_ccb_t *)scb->ccb;
scb->dev_channel = channel;
scb->dev_target = target;
scb->dma_direction = scp->sc_data_direction;
mbox = ccb->mbox;
mbox64 = ccb->mbox64;
// Does this firmware support extended CDBs
if (adapter->max_cdb_sz == 16) {
mbox->cmd = MBOXCMD_EXTPTHRU;
megaraid_mbox_prepare_epthru(adapter, scb, scp);
mbox64->xferaddr_lo = (uint32_t)ccb->epthru_dma_h;
mbox64->xferaddr_hi = 0;
mbox->xferaddr = 0xFFFFFFFF;
}
else {
mbox->cmd = MBOXCMD_PASSTHRU64;
megaraid_mbox_prepare_pthru(adapter, scb, scp);
mbox64->xferaddr_lo = (uint32_t)ccb->pthru_dma_h;
mbox64->xferaddr_hi = 0;
mbox->xferaddr = 0xFFFFFFFF;
}
return scb;
}
// NOT REACHED
}
/**
* megaraid_mbox_runpendq - execute commands queued in the pending queue
* @adapter : controller's soft state
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @scb_q : SCB to be queued in the pending list
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Scan the pending list for commands which are not yet issued and try to
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* post to the controller. The SCB can be a null pointer, which would indicate
* no SCB to be queue, just try to execute the ones in the pending list.
*
* NOTE: We do not actually traverse the pending list. The SCBs are plucked
* out from the head of the pending list. If it is successfully issued, the
* next SCB is at the head now.
*/
static void
megaraid_mbox_runpendq(adapter_t *adapter, scb_t *scb_q)
{
scb_t *scb;
unsigned long flags;
spin_lock_irqsave(PENDING_LIST_LOCK(adapter), flags);
if (scb_q) {
scb_q->state = SCB_PENDQ;
list_add_tail(&scb_q->list, &adapter->pend_list);
}
// if the adapter in not in quiescent mode, post the commands to FW
if (adapter->quiescent) {
spin_unlock_irqrestore(PENDING_LIST_LOCK(adapter), flags);
return;
}
while (!list_empty(&adapter->pend_list)) {
assert_spin_locked(PENDING_LIST_LOCK(adapter));
scb = list_entry(adapter->pend_list.next, scb_t, list);
// remove the scb from the pending list and try to
// issue. If we are unable to issue it, put back in
// the pending list and return
list_del_init(&scb->list);
spin_unlock_irqrestore(PENDING_LIST_LOCK(adapter), flags);
// if mailbox was busy, return SCB back to pending
// list. Make sure to add at the head, since that's
// where it would have been removed from
scb->state = SCB_ISSUED;
if (mbox_post_cmd(adapter, scb) != 0) {
spin_lock_irqsave(PENDING_LIST_LOCK(adapter), flags);
scb->state = SCB_PENDQ;
list_add(&scb->list, &adapter->pend_list);
spin_unlock_irqrestore(PENDING_LIST_LOCK(adapter),
flags);
return;
}
spin_lock_irqsave(PENDING_LIST_LOCK(adapter), flags);
}
spin_unlock_irqrestore(PENDING_LIST_LOCK(adapter), flags);
return;
}
/**
* megaraid_mbox_prepare_pthru - prepare a command for physical devices
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : pointer to controller's soft state
* @scb : scsi control block
* @scp : scsi command from the mid-layer
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Prepare a command for the scsi physical devices.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static void
megaraid_mbox_prepare_pthru(adapter_t *adapter, scb_t *scb,
struct scsi_cmnd *scp)
{
mbox_ccb_t *ccb;
mraid_passthru_t *pthru;
uint8_t channel;
uint8_t target;
ccb = (mbox_ccb_t *)scb->ccb;
pthru = ccb->pthru;
channel = scb->dev_channel;
target = scb->dev_target;
// 0=6sec, 1=60sec, 2=10min, 3=3hrs, 4=NO timeout
pthru->timeout = 4;
pthru->ars = 1;
pthru->islogical = 0;
pthru->channel = 0;
pthru->target = (channel << 4) | target;
pthru->logdrv = SCP2LUN(scp);
pthru->reqsenselen = 14;
pthru->cdblen = scp->cmd_len;
memcpy(pthru->cdb, scp->cmnd, scp->cmd_len);
[SCSI] megaraid: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:13:44 +08:00
if (scsi_bufflen(scp)) {
pthru->dataxferlen = scsi_bufflen(scp);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
pthru->dataxferaddr = ccb->sgl_dma_h;
pthru->numsge = megaraid_mbox_mksgl(adapter, scb);
}
else {
pthru->dataxferaddr = 0;
pthru->dataxferlen = 0;
pthru->numsge = 0;
}
return;
}
/**
* megaraid_mbox_prepare_epthru - prepare a command for physical devices
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : pointer to controller's soft state
* @scb : scsi control block
* @scp : scsi command from the mid-layer
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
scsi/megaraid fixed several typos in comments Fixed several typos in comments in megaraid_mbox.c, megaraid_mm.c and megaraid_sas_fusion.h. Signed-off-by: Matthias Schid <aircrach115@gmail.com> Signed-off-by: Stefan Huber <steffhip@gmail.com> Signed-off-by: Simon Puels <simon.puels@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2013-06-26 22:25:53 +08:00
* Prepare a command for the scsi physical devices. This routine prepares
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* commands for devices which can take extended CDBs (>10 bytes).
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static void
megaraid_mbox_prepare_epthru(adapter_t *adapter, scb_t *scb,
struct scsi_cmnd *scp)
{
mbox_ccb_t *ccb;
mraid_epassthru_t *epthru;
uint8_t channel;
uint8_t target;
ccb = (mbox_ccb_t *)scb->ccb;
epthru = ccb->epthru;
channel = scb->dev_channel;
target = scb->dev_target;
// 0=6sec, 1=60sec, 2=10min, 3=3hrs, 4=NO timeout
epthru->timeout = 4;
epthru->ars = 1;
epthru->islogical = 0;
epthru->channel = 0;
epthru->target = (channel << 4) | target;
epthru->logdrv = SCP2LUN(scp);
epthru->reqsenselen = 14;
epthru->cdblen = scp->cmd_len;
memcpy(epthru->cdb, scp->cmnd, scp->cmd_len);
[SCSI] megaraid: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:13:44 +08:00
if (scsi_bufflen(scp)) {
epthru->dataxferlen = scsi_bufflen(scp);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
epthru->dataxferaddr = ccb->sgl_dma_h;
epthru->numsge = megaraid_mbox_mksgl(adapter, scb);
}
else {
epthru->dataxferaddr = 0;
epthru->dataxferlen = 0;
epthru->numsge = 0;
}
return;
}
/**
* megaraid_ack_sequence - interrupt ack sequence for memory mapped HBAs
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : controller's soft state
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Interrupt acknowledgement sequence for memory mapped HBAs. Find out the
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* completed command and put them on the completed list for later processing.
*
* Returns: 1 if the interrupt is valid, 0 otherwise
*/
[PATCH] Unlinline a bunch of other functions Remove the "inline" keyword from a bunch of big functions in the kernel with the goal of shrinking it by 30kb to 40kb Signed-off-by: Arjan van de Ven <arjan@infradead.org> Signed-off-by: Ingo Molnar <mingo@elte.hu> Acked-by: Jeff Garzik <jgarzik@pobox.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-15 05:20:43 +08:00
static int
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
megaraid_ack_sequence(adapter_t *adapter)
{
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
mbox_t *mbox;
scb_t *scb;
uint8_t nstatus;
uint8_t completed[MBOX_MAX_FIRMWARE_STATUS];
struct list_head clist;
int handled;
uint32_t dword;
unsigned long flags;
int i, j;
mbox = raid_dev->mbox;
// move the SCBs from the firmware completed array to our local list
INIT_LIST_HEAD(&clist);
// loop till F/W has more commands for us to complete
handled = 0;
spin_lock_irqsave(MAILBOX_LOCK(raid_dev), flags);
do {
/*
* Check if a valid interrupt is pending. If found, force the
* interrupt line low.
*/
dword = RDOUTDOOR(raid_dev);
if (dword != 0x10001234) break;
handled = 1;
WROUTDOOR(raid_dev, 0x10001234);
nstatus = 0;
// wait for valid numstatus to post
for (i = 0; i < 0xFFFFF; i++) {
if (mbox->numstatus != 0xFF) {
nstatus = mbox->numstatus;
break;
}
rmb();
}
mbox->numstatus = 0xFF;
adapter->outstanding_cmds -= nstatus;
for (i = 0; i < nstatus; i++) {
// wait for valid command index to post
for (j = 0; j < 0xFFFFF; j++) {
if (mbox->completed[i] != 0xFF) break;
rmb();
}
completed[i] = mbox->completed[i];
mbox->completed[i] = 0xFF;
if (completed[i] == 0xFF) {
con_log(CL_ANN, (KERN_CRIT
"megaraid: command posting timed out\n"));
BUG();
continue;
}
// Get SCB associated with this command id
if (completed[i] >= MBOX_MAX_SCSI_CMDS) {
// a cmm command
scb = adapter->uscb_list + (completed[i] -
MBOX_MAX_SCSI_CMDS);
}
else {
// an os command
scb = adapter->kscb_list + completed[i];
}
scb->status = mbox->status;
list_add_tail(&scb->list, &clist);
}
// Acknowledge interrupt
WRINDOOR(raid_dev, 0x02);
} while(1);
spin_unlock_irqrestore(MAILBOX_LOCK(raid_dev), flags);
// put the completed commands in the completed list. DPC would
// complete these commands later
spin_lock_irqsave(COMPLETED_LIST_LOCK(adapter), flags);
list_splice(&clist, &adapter->completed_list);
spin_unlock_irqrestore(COMPLETED_LIST_LOCK(adapter), flags);
// schedule the DPC if there is some work for it
if (handled)
tasklet_schedule(&adapter->dpc_h);
return handled;
}
/**
* megaraid_isr - isr for memory based mailbox based controllers
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @irq : irq
* @devp : pointer to our soft state
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* Interrupt service routine for memory-mapped mailbox controllers.
*/
static irqreturn_t
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 21:55:46 +08:00
megaraid_isr(int irq, void *devp)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
adapter_t *adapter = devp;
int handled;
handled = megaraid_ack_sequence(adapter);
/* Loop through any pending requests */
if (!adapter->quiescent) {
megaraid_mbox_runpendq(adapter, NULL);
}
return IRQ_RETVAL(handled);
}
/**
* megaraid_mbox_dpc - the tasklet to complete the commands from completed list
* @devp : pointer to HBA soft state
*
* Pick up the commands from the completed list and send back to the owners.
* This is a reentrant function and does not assume any locks are held while
* it is being called.
*/
static void
megaraid_mbox_dpc(unsigned long devp)
{
adapter_t *adapter = (adapter_t *)devp;
mraid_device_t *raid_dev;
struct list_head clist;
struct scatterlist *sgl;
scb_t *scb;
scb_t *tmp;
struct scsi_cmnd *scp;
mraid_passthru_t *pthru;
mraid_epassthru_t *epthru;
mbox_ccb_t *ccb;
int islogical;
int pdev_index;
int pdev_state;
mbox_t *mbox;
unsigned long flags;
uint8_t c;
int status;
[SCSI] megaraid_{mm,mbox}: fix a bug in reset handler When abort failed, the driver gets reset handleer called. In the reset handler, driver calls 'scsi_done()' callback for same SCSI command packet (struct scsi_cmnd) multiple times if there are multiple SCSI command packet in the pend_list. More over, if there are entry in the pend_lsit with IOCTL packet associated, the driver returns it to wrong free_list so that, in turn, the driver could end up with 'NULL pointer dereference..' during I/O command building with incorrect resource. Also, the patch contains several minor/cosmetic changes besides this. Signed-off-by: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-04-27 17:33:06 +08:00
uioc_t *kioc;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (!adapter) return;
raid_dev = ADAP2RAIDDEV(adapter);
// move the SCBs from the completed list to our local list
INIT_LIST_HEAD(&clist);
spin_lock_irqsave(COMPLETED_LIST_LOCK(adapter), flags);
list_splice_init(&adapter->completed_list, &clist);
spin_unlock_irqrestore(COMPLETED_LIST_LOCK(adapter), flags);
list_for_each_entry_safe(scb, tmp, &clist, list) {
status = scb->status;
scp = scb->scp;
ccb = (mbox_ccb_t *)scb->ccb;
pthru = ccb->pthru;
epthru = ccb->epthru;
mbox = ccb->mbox;
// Make sure f/w has completed a valid command
if (scb->state != SCB_ISSUED) {
con_log(CL_ANN, (KERN_CRIT
"megaraid critical err: invalid command %d:%d:%p\n",
scb->sno, scb->state, scp));
BUG();
continue; // Must never happen!
}
// check for the management command and complete it right away
if (scb->sno >= MBOX_MAX_SCSI_CMDS) {
scb->state = SCB_FREE;
scb->status = status;
// remove from local clist
list_del_init(&scb->list);
[SCSI] megaraid_{mm,mbox}: fix a bug in reset handler When abort failed, the driver gets reset handleer called. In the reset handler, driver calls 'scsi_done()' callback for same SCSI command packet (struct scsi_cmnd) multiple times if there are multiple SCSI command packet in the pend_list. More over, if there are entry in the pend_lsit with IOCTL packet associated, the driver returns it to wrong free_list so that, in turn, the driver could end up with 'NULL pointer dereference..' during I/O command building with incorrect resource. Also, the patch contains several minor/cosmetic changes besides this. Signed-off-by: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-04-27 17:33:06 +08:00
kioc = (uioc_t *)scb->gp;
kioc->status = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
megaraid_mbox_mm_done(adapter, scb);
continue;
}
// Was an abort issued for this command earlier
if (scb->state & SCB_ABORT) {
con_log(CL_ANN, (KERN_NOTICE
[SCSI] remove cmd->serial_number litter Stop using cmd->serial_number in printks. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-04 21:42:14 +08:00
"megaraid: aborted cmd [%x] completed\n",
scb->sno));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
/*
* If the inquiry came of a disk drive which is not part of
* any RAID array, expose it to the kernel. For this to be
* enabled, user must set the "megaraid_expose_unconf_disks"
* flag to 1 by specifying it on module parameter list.
* This would enable data migration off drives from other
* configurations.
*/
islogical = MRAID_IS_LOGICAL(adapter, scp);
if (scp->cmnd[0] == INQUIRY && status == 0 && islogical == 0
&& IS_RAID_CH(raid_dev, scb->dev_channel)) {
[SCSI] megaraid: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:13:44 +08:00
sgl = scsi_sglist(scp);
[SG] Update drivers to use sg helpers Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2007-10-23 03:19:53 +08:00
if (sg_page(sgl)) {
c = *(unsigned char *) sg_virt(&sgl[0]);
[SCSI] megaraid: convert to use the data buffer accessors - remove the unnecessary map_single path. - convert to use the new accessors for the sg lists and the parameters. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-05-14 19:13:44 +08:00
} else {
con_log(CL_ANN, (KERN_WARNING
"megaraid mailbox: invalid sg:%d\n",
__LINE__));
c = 0;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
if ((c & 0x1F ) == TYPE_DISK) {
pdev_index = (scb->dev_channel * 16) +
scb->dev_target;
pdev_state =
raid_dev->pdrv_state[pdev_index] & 0x0F;
if (pdev_state == PDRV_ONLINE ||
pdev_state == PDRV_FAILED ||
pdev_state == PDRV_RBLD ||
pdev_state == PDRV_HOTSPARE ||
megaraid_expose_unconf_disks == 0) {
status = 0xF0;
}
}
}
// Convert MegaRAID status to Linux error code
switch (status) {
case 0x00:
scp->result = (DID_OK << 16);
break;
case 0x02:
/* set sense_buffer and result fields */
if (mbox->cmd == MBOXCMD_PASSTHRU ||
mbox->cmd == MBOXCMD_PASSTHRU64) {
memcpy(scp->sense_buffer, pthru->reqsensearea,
14);
scp->result = DRIVER_SENSE << 24 |
DID_OK << 16 | CHECK_CONDITION << 1;
}
else {
if (mbox->cmd == MBOXCMD_EXTPTHRU) {
memcpy(scp->sense_buffer,
epthru->reqsensearea, 14);
scp->result = DRIVER_SENSE << 24 |
DID_OK << 16 |
CHECK_CONDITION << 1;
} else {
scp->sense_buffer[0] = 0x70;
scp->sense_buffer[2] = ABORTED_COMMAND;
scp->result = CHECK_CONDITION << 1;
}
}
break;
case 0x08:
scp->result = DID_BUS_BUSY << 16 | status;
break;
default:
/*
* If TEST_UNIT_READY fails, we know RESERVATION_STATUS
* failed
*/
if (scp->cmnd[0] == TEST_UNIT_READY) {
scp->result = DID_ERROR << 16 |
RESERVATION_CONFLICT << 1;
}
else
/*
* Error code returned is 1 if Reserve or Release
* failed or the input parameter is invalid
*/
if (status == 1 && (scp->cmnd[0] == RESERVE ||
scp->cmnd[0] == RELEASE)) {
scp->result = DID_ERROR << 16 |
RESERVATION_CONFLICT << 1;
}
else {
scp->result = DID_BAD_TARGET << 16 | status;
}
}
// print a debug message for all failed commands
if (status) {
megaraid_mbox_display_scb(adapter, scb);
}
scsi: megaraid_mbox: remove bogus use of pci_dma_sync_sg_* APIs The dma_map_sg / dma_unmap_sg APIs called from scsi_dma_map / scsi_dma_unmap already transfer memory ownership to the device or cpu respectively. Adding additional calls to pci_dma_sync_sg_* will in fact lead to data corruption if we end up using swiotlb for some reason. Also remove the now pointless megaraid_mbox_sync_scb function. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-10-11 15:15:35 +08:00
scsi_dma_unmap(scp);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
// remove from local clist
list_del_init(&scb->list);
// put back in free list
megaraid_dealloc_scb(adapter, scb);
// send the scsi packet back to kernel
scp->scsi_done(scp);
}
return;
}
/**
* megaraid_abort_handler - abort the scsi command
* @scp : command to be aborted
*
* Abort a previous SCSI request. Only commands on the pending list can be
* aborted. All the commands issued to the F/W must complete.
**/
static int
[SCSI] megaraid_mbox: remove scsi_assign_lock usage also remove the adapter->host_lock alias for adapter->lock and remove some superflous locking aswell as removing the tiny locking wrappers for the EH routines. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 04:06:02 +08:00
megaraid_abort_handler(struct scsi_cmnd *scp)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
adapter_t *adapter;
mraid_device_t *raid_dev;
scb_t *scb;
scb_t *tmp;
int found;
unsigned long flags;
int i;
adapter = SCP2ADAPTER(scp);
raid_dev = ADAP2RAIDDEV(adapter);
con_log(CL_ANN, (KERN_WARNING
[SCSI] remove cmd->serial_number litter Stop using cmd->serial_number in printks. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-04 21:42:14 +08:00
"megaraid: aborting cmd=%x <c=%d t=%d l=%d>\n",
scp->cmnd[0], SCP2CHANNEL(scp),
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
SCP2TARGET(scp), SCP2LUN(scp)));
// If FW has stopped responding, simply return failure
if (raid_dev->hw_error) {
con_log(CL_ANN, (KERN_NOTICE
"megaraid: hw error, not aborting\n"));
return FAILED;
}
// There might a race here, where the command was completed by the
// firmware and now it is on the completed list. Before we could
// complete the command to the kernel in dpc, the abort came.
// Find out if this is the case to avoid the race.
scb = NULL;
spin_lock_irqsave(COMPLETED_LIST_LOCK(adapter), flags);
list_for_each_entry_safe(scb, tmp, &adapter->completed_list, list) {
if (scb->scp == scp) { // Found command
list_del_init(&scb->list); // from completed list
con_log(CL_ANN, (KERN_WARNING
[SCSI] remove cmd->serial_number litter Stop using cmd->serial_number in printks. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-04 21:42:14 +08:00
"megaraid: %d[%d:%d], abort from completed list\n",
scb->sno, scb->dev_channel, scb->dev_target));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
scp->result = (DID_ABORT << 16);
scp->scsi_done(scp);
megaraid_dealloc_scb(adapter, scb);
spin_unlock_irqrestore(COMPLETED_LIST_LOCK(adapter),
flags);
return SUCCESS;
}
}
spin_unlock_irqrestore(COMPLETED_LIST_LOCK(adapter), flags);
// Find out if this command is still on the pending list. If it is and
// was never issued, abort and return success. If the command is owned
// by the firmware, we must wait for it to complete by the FW.
spin_lock_irqsave(PENDING_LIST_LOCK(adapter), flags);
list_for_each_entry_safe(scb, tmp, &adapter->pend_list, list) {
if (scb->scp == scp) { // Found command
list_del_init(&scb->list); // from pending list
ASSERT(!(scb->state & SCB_ISSUED));
con_log(CL_ANN, (KERN_WARNING
[SCSI] remove cmd->serial_number litter Stop using cmd->serial_number in printks. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-04 21:42:14 +08:00
"megaraid abort: [%d:%d], driver owner\n",
scb->dev_channel, scb->dev_target));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
scp->result = (DID_ABORT << 16);
scp->scsi_done(scp);
megaraid_dealloc_scb(adapter, scb);
spin_unlock_irqrestore(PENDING_LIST_LOCK(adapter),
flags);
return SUCCESS;
}
}
spin_unlock_irqrestore(PENDING_LIST_LOCK(adapter), flags);
// Check do we even own this command, in which case this would be
// owned by the firmware. The only way to locate the FW scb is to
// traverse through the list of all SCB, since driver does not
// maintain these SCBs on any list
found = 0;
[SCSI] megaraid_mbox: remove scsi_assign_lock usage also remove the adapter->host_lock alias for adapter->lock and remove some superflous locking aswell as removing the tiny locking wrappers for the EH routines. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 04:06:02 +08:00
spin_lock_irq(&adapter->lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
for (i = 0; i < MBOX_MAX_SCSI_CMDS; i++) {
scb = adapter->kscb_list + i;
if (scb->scp == scp) {
found = 1;
if (!(scb->state & SCB_ISSUED)) {
con_log(CL_ANN, (KERN_WARNING
[SCSI] remove cmd->serial_number litter Stop using cmd->serial_number in printks. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-04 21:42:14 +08:00
"megaraid abort: %d[%d:%d], invalid state\n",
scb->sno, scb->dev_channel, scb->dev_target));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
BUG();
}
else {
con_log(CL_ANN, (KERN_WARNING
[SCSI] remove cmd->serial_number litter Stop using cmd->serial_number in printks. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-04 21:42:14 +08:00
"megaraid abort: %d[%d:%d], fw owner\n",
scb->sno, scb->dev_channel, scb->dev_target));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
}
}
[SCSI] megaraid_mbox: remove scsi_assign_lock usage also remove the adapter->host_lock alias for adapter->lock and remove some superflous locking aswell as removing the tiny locking wrappers for the EH routines. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 04:06:02 +08:00
spin_unlock_irq(&adapter->lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (!found) {
[SCSI] remove cmd->serial_number litter Stop using cmd->serial_number in printks. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-04 21:42:14 +08:00
con_log(CL_ANN, (KERN_WARNING "megaraid abort: do now own\n"));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
// FIXME: Should there be a callback for this command?
return SUCCESS;
}
// We cannot actually abort a command owned by firmware, return
// failure and wait for reset. In host reset handler, we will find out
// if the HBA is still live
return FAILED;
}
/**
scsi/megaraid fixed several typos in comments Fixed several typos in comments in megaraid_mbox.c, megaraid_mm.c and megaraid_sas_fusion.h. Signed-off-by: Matthias Schid <aircrach115@gmail.com> Signed-off-by: Stefan Huber <steffhip@gmail.com> Signed-off-by: Simon Puels <simon.puels@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2013-06-26 22:25:53 +08:00
* megaraid_reset_handler - device reset handler for mailbox based driver
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* @scp : reference command
*
* Reset handler for the mailbox based controller. First try to find out if
* the FW is still live, in which case the outstanding commands counter mut go
* down to 0. If that happens, also issue the reservation reset command to
* relinquish (possible) reservations on the logical drives connected to this
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* host.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
**/
static int
[SCSI] megaraid_mbox: remove scsi_assign_lock usage also remove the adapter->host_lock alias for adapter->lock and remove some superflous locking aswell as removing the tiny locking wrappers for the EH routines. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 04:06:02 +08:00
megaraid_reset_handler(struct scsi_cmnd *scp)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
adapter_t *adapter;
scb_t *scb;
scb_t *tmp;
mraid_device_t *raid_dev;
unsigned long flags;
uint8_t raw_mbox[sizeof(mbox_t)];
int rval;
int recovery_window;
int i;
[SCSI] megaraid_{mm,mbox}: fix a bug in reset handler When abort failed, the driver gets reset handleer called. In the reset handler, driver calls 'scsi_done()' callback for same SCSI command packet (struct scsi_cmnd) multiple times if there are multiple SCSI command packet in the pend_list. More over, if there are entry in the pend_lsit with IOCTL packet associated, the driver returns it to wrong free_list so that, in turn, the driver could end up with 'NULL pointer dereference..' during I/O command building with incorrect resource. Also, the patch contains several minor/cosmetic changes besides this. Signed-off-by: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-04-27 17:33:06 +08:00
uioc_t *kioc;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
adapter = SCP2ADAPTER(scp);
raid_dev = ADAP2RAIDDEV(adapter);
// return failure if adapter is not responding
if (raid_dev->hw_error) {
con_log(CL_ANN, (KERN_NOTICE
"megaraid: hw error, cannot reset\n"));
return FAILED;
}
// Under exceptional conditions, FW can take up to 3 minutes to
// complete command processing. Wait for additional 2 minutes for the
// pending commands counter to go down to 0. If it doesn't, let the
// controller be marked offline
// Also, reset all the commands currently owned by the driver
spin_lock_irqsave(PENDING_LIST_LOCK(adapter), flags);
list_for_each_entry_safe(scb, tmp, &adapter->pend_list, list) {
list_del_init(&scb->list); // from pending list
[SCSI] megaraid_{mm,mbox}: fix a bug in reset handler When abort failed, the driver gets reset handleer called. In the reset handler, driver calls 'scsi_done()' callback for same SCSI command packet (struct scsi_cmnd) multiple times if there are multiple SCSI command packet in the pend_list. More over, if there are entry in the pend_lsit with IOCTL packet associated, the driver returns it to wrong free_list so that, in turn, the driver could end up with 'NULL pointer dereference..' during I/O command building with incorrect resource. Also, the patch contains several minor/cosmetic changes besides this. Signed-off-by: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-04-27 17:33:06 +08:00
if (scb->sno >= MBOX_MAX_SCSI_CMDS) {
con_log(CL_ANN, (KERN_WARNING
"megaraid: IOCTL packet with %d[%d:%d] being reset\n",
scb->sno, scb->dev_channel, scb->dev_target));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
[SCSI] megaraid_{mm,mbox}: fix a bug in reset handler When abort failed, the driver gets reset handleer called. In the reset handler, driver calls 'scsi_done()' callback for same SCSI command packet (struct scsi_cmnd) multiple times if there are multiple SCSI command packet in the pend_list. More over, if there are entry in the pend_lsit with IOCTL packet associated, the driver returns it to wrong free_list so that, in turn, the driver could end up with 'NULL pointer dereference..' during I/O command building with incorrect resource. Also, the patch contains several minor/cosmetic changes besides this. Signed-off-by: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-04-27 17:33:06 +08:00
scb->status = -1;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
[SCSI] megaraid_{mm,mbox}: fix a bug in reset handler When abort failed, the driver gets reset handleer called. In the reset handler, driver calls 'scsi_done()' callback for same SCSI command packet (struct scsi_cmnd) multiple times if there are multiple SCSI command packet in the pend_list. More over, if there are entry in the pend_lsit with IOCTL packet associated, the driver returns it to wrong free_list so that, in turn, the driver could end up with 'NULL pointer dereference..' during I/O command building with incorrect resource. Also, the patch contains several minor/cosmetic changes besides this. Signed-off-by: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-04-27 17:33:06 +08:00
kioc = (uioc_t *)scb->gp;
kioc->status = -EFAULT;
megaraid_mbox_mm_done(adapter, scb);
} else {
if (scb->scp == scp) { // Found command
con_log(CL_ANN, (KERN_WARNING
[SCSI] remove cmd->serial_number litter Stop using cmd->serial_number in printks. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
2011-04-04 21:42:14 +08:00
"megaraid: %d[%d:%d], reset from pending list\n",
scb->sno, scb->dev_channel, scb->dev_target));
[SCSI] megaraid_{mm,mbox}: fix a bug in reset handler When abort failed, the driver gets reset handleer called. In the reset handler, driver calls 'scsi_done()' callback for same SCSI command packet (struct scsi_cmnd) multiple times if there are multiple SCSI command packet in the pend_list. More over, if there are entry in the pend_lsit with IOCTL packet associated, the driver returns it to wrong free_list so that, in turn, the driver could end up with 'NULL pointer dereference..' during I/O command building with incorrect resource. Also, the patch contains several minor/cosmetic changes besides this. Signed-off-by: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-04-27 17:33:06 +08:00
} else {
con_log(CL_ANN, (KERN_WARNING
"megaraid: IO packet with %d[%d:%d] being reset\n",
scb->sno, scb->dev_channel, scb->dev_target));
}
scb->scp->result = (DID_RESET << 16);
scb->scp->scsi_done(scb->scp);
megaraid_dealloc_scb(adapter, scb);
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
spin_unlock_irqrestore(PENDING_LIST_LOCK(adapter), flags);
if (adapter->outstanding_cmds) {
con_log(CL_ANN, (KERN_NOTICE
"megaraid: %d outstanding commands. Max wait %d sec\n",
[SCSI] megaraid_{mm,mbox}: fix a bug in reset handler When abort failed, the driver gets reset handleer called. In the reset handler, driver calls 'scsi_done()' callback for same SCSI command packet (struct scsi_cmnd) multiple times if there are multiple SCSI command packet in the pend_list. More over, if there are entry in the pend_lsit with IOCTL packet associated, the driver returns it to wrong free_list so that, in turn, the driver could end up with 'NULL pointer dereference..' during I/O command building with incorrect resource. Also, the patch contains several minor/cosmetic changes besides this. Signed-off-by: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-04-27 17:33:06 +08:00
adapter->outstanding_cmds,
(MBOX_RESET_WAIT + MBOX_RESET_EXT_WAIT)));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
recovery_window = MBOX_RESET_WAIT + MBOX_RESET_EXT_WAIT;
[SCSI] megaraid_{mm,mbox}: fix a bug in reset handler When abort failed, the driver gets reset handleer called. In the reset handler, driver calls 'scsi_done()' callback for same SCSI command packet (struct scsi_cmnd) multiple times if there are multiple SCSI command packet in the pend_list. More over, if there are entry in the pend_lsit with IOCTL packet associated, the driver returns it to wrong free_list so that, in turn, the driver could end up with 'NULL pointer dereference..' during I/O command building with incorrect resource. Also, the patch contains several minor/cosmetic changes besides this. Signed-off-by: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-04-27 17:33:06 +08:00
for (i = 0; i < recovery_window; i++) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
megaraid_ack_sequence(adapter);
// print a message once every 5 seconds only
if (!(i % 5)) {
con_log(CL_ANN, (
"megaraid mbox: Wait for %d commands to complete:%d\n",
adapter->outstanding_cmds,
[SCSI] megaraid_{mm,mbox}: fix a bug in reset handler When abort failed, the driver gets reset handleer called. In the reset handler, driver calls 'scsi_done()' callback for same SCSI command packet (struct scsi_cmnd) multiple times if there are multiple SCSI command packet in the pend_list. More over, if there are entry in the pend_lsit with IOCTL packet associated, the driver returns it to wrong free_list so that, in turn, the driver could end up with 'NULL pointer dereference..' during I/O command building with incorrect resource. Also, the patch contains several minor/cosmetic changes besides this. Signed-off-by: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-04-27 17:33:06 +08:00
(MBOX_RESET_WAIT + MBOX_RESET_EXT_WAIT) - i));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
Fix common misspellings Fixes generated by 'codespell' and manually reviewed. Signed-off-by: Lucas De Marchi <lucas.demarchi@profusion.mobi>
2011-03-31 09:57:33 +08:00
// bailout if no recovery happened in reset time
[SCSI] megaraid_{mm,mbox}: fix a bug in reset handler When abort failed, the driver gets reset handleer called. In the reset handler, driver calls 'scsi_done()' callback for same SCSI command packet (struct scsi_cmnd) multiple times if there are multiple SCSI command packet in the pend_list. More over, if there are entry in the pend_lsit with IOCTL packet associated, the driver returns it to wrong free_list so that, in turn, the driver could end up with 'NULL pointer dereference..' during I/O command building with incorrect resource. Also, the patch contains several minor/cosmetic changes besides this. Signed-off-by: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-04-27 17:33:06 +08:00
if (adapter->outstanding_cmds == 0) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
break;
}
msleep(1000);
}
[SCSI] megaraid_mbox: remove scsi_assign_lock usage also remove the adapter->host_lock alias for adapter->lock and remove some superflous locking aswell as removing the tiny locking wrappers for the EH routines. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 04:06:02 +08:00
spin_lock(&adapter->lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
// If still outstanding commands, bail out
if (adapter->outstanding_cmds) {
con_log(CL_ANN, (KERN_WARNING
"megaraid mbox: critical hardware error!\n"));
raid_dev->hw_error = 1;
[SCSI] megaraid_mbox: remove scsi_assign_lock usage also remove the adapter->host_lock alias for adapter->lock and remove some superflous locking aswell as removing the tiny locking wrappers for the EH routines. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 04:06:02 +08:00
rval = FAILED;
goto out;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
else {
con_log(CL_ANN, (KERN_NOTICE
tree-wide: fix assorted typos all over the place That is "success", "unknown", "through", "performance", "[re|un]mapping" , "access", "default", "reasonable", "[con]currently", "temperature" , "channel", "[un]used", "application", "example","hierarchy", "therefore" , "[over|under]flow", "contiguous", "threshold", "enough" and others. Signed-off-by: André Goddard Rosa <andre.goddard@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2009-11-14 23:09:05 +08:00
"megaraid mbox: reset sequence completed successfully\n"));
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
// If the controller supports clustering, reset reservations
[SCSI] megaraid_mbox: remove scsi_assign_lock usage also remove the adapter->host_lock alias for adapter->lock and remove some superflous locking aswell as removing the tiny locking wrappers for the EH routines. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 04:06:02 +08:00
if (!adapter->ha) {
rval = SUCCESS;
goto out;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
// clear reservations if any
raw_mbox[0] = CLUSTER_CMD;
raw_mbox[2] = RESET_RESERVATIONS;
rval = SUCCESS;
if (mbox_post_sync_cmd_fast(adapter, raw_mbox) == 0) {
con_log(CL_ANN,
(KERN_INFO "megaraid: reservation reset\n"));
}
else {
rval = FAILED;
con_log(CL_ANN, (KERN_WARNING
"megaraid: reservation reset failed\n"));
}
[SCSI] megaraid_mbox: remove scsi_assign_lock usage also remove the adapter->host_lock alias for adapter->lock and remove some superflous locking aswell as removing the tiny locking wrappers for the EH routines. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 04:06:02 +08:00
out:
[SCSI] megaraid: remove a spurious IRQ enable We took this lock with spin_lock() so we should unlock it with spin_unlock() instead of spin_unlock_irq(). This was introduced in f2c8dc402b "[SCSI] megaraid_mbox: remove scsi_assign_lock usage". Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Acked-by: Adam Radford <aradford@gmail.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2012-06-27 17:00:30 +08:00
spin_unlock(&adapter->lock);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return rval;
}
/*
* START: internal commands library
*
* This section of the driver has the common routine used by the driver and
* also has all the FW routines
*/
/**
* mbox_post_sync_cmd() - blocking command to the mailbox based controllers
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : controller's soft state
* @raw_mbox : the mailbox
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* Issue a scb in synchronous and non-interrupt mode for mailbox based
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* controllers.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static int
mbox_post_sync_cmd(adapter_t *adapter, uint8_t raw_mbox[])
{
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
mbox_t *mbox;
uint8_t status;
int i;
mbox = raid_dev->mbox;
/*
* Wait until mailbox is free
*/
if (megaraid_busywait_mbox(raid_dev) != 0)
goto blocked_mailbox;
/*
* Copy mailbox data into host structure
*/
memcpy((caddr_t)mbox, (caddr_t)raw_mbox, 16);
mbox->cmdid = 0xFE;
mbox->busy = 1;
mbox->poll = 0;
mbox->ack = 0;
mbox->numstatus = 0xFF;
mbox->status = 0xFF;
wmb();
WRINDOOR(raid_dev, raid_dev->mbox_dma | 0x1);
// wait for maximum 1 second for status to post. If the status is not
// available within 1 second, assume FW is initializing and wait
// for an extended amount of time
if (mbox->numstatus == 0xFF) { // status not yet available
[PATCH] s/;;/;/g Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-03-24 19:16:13 +08:00
udelay(25);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
for (i = 0; mbox->numstatus == 0xFF && i < 1000; i++) {
rmb();
msleep(1);
}
if (i == 1000) {
con_log(CL_ANN, (KERN_NOTICE
"megaraid mailbox: wait for FW to boot "));
for (i = 0; (mbox->numstatus == 0xFF) &&
(i < MBOX_RESET_WAIT); i++) {
rmb();
con_log(CL_ANN, ("\b\b\b\b\b[%03d]",
MBOX_RESET_WAIT - i));
msleep(1000);
}
if (i == MBOX_RESET_WAIT) {
con_log(CL_ANN, (
"\nmegaraid mailbox: status not available\n"));
return -1;
}
con_log(CL_ANN, ("\b\b\b\b\b[ok] \n"));
}
}
// wait for maximum 1 second for poll semaphore
if (mbox->poll != 0x77) {
udelay(25);
for (i = 0; (mbox->poll != 0x77) && (i < 1000); i++) {
rmb();
msleep(1);
}
if (i == 1000) {
con_log(CL_ANN, (KERN_WARNING
"megaraid mailbox: could not get poll semaphore\n"));
return -1;
}
}
WRINDOOR(raid_dev, raid_dev->mbox_dma | 0x2);
wmb();
// wait for maximum 1 second for acknowledgement
if (RDINDOOR(raid_dev) & 0x2) {
udelay(25);
for (i = 0; (RDINDOOR(raid_dev) & 0x2) && (i < 1000); i++) {
rmb();
msleep(1);
}
if (i == 1000) {
con_log(CL_ANN, (KERN_WARNING
"megaraid mailbox: could not acknowledge\n"));
return -1;
}
}
mbox->poll = 0;
mbox->ack = 0x77;
status = mbox->status;
// invalidate the completed command id array. After command
// completion, firmware would write the valid id.
mbox->numstatus = 0xFF;
mbox->status = 0xFF;
for (i = 0; i < MBOX_MAX_FIRMWARE_STATUS; i++) {
mbox->completed[i] = 0xFF;
}
return status;
blocked_mailbox:
con_log(CL_ANN, (KERN_WARNING "megaraid: blocked mailbox\n") );
return -1;
}
/**
* mbox_post_sync_cmd_fast - blocking command to the mailbox based controllers
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : controller's soft state
* @raw_mbox : the mailbox
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* Issue a scb in synchronous and non-interrupt mode for mailbox based
* controllers. This is a faster version of the synchronous command and
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* therefore can be called in interrupt-context as well.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static int
mbox_post_sync_cmd_fast(adapter_t *adapter, uint8_t raw_mbox[])
{
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
mbox_t *mbox;
long i;
mbox = raid_dev->mbox;
// return immediately if the mailbox is busy
if (mbox->busy) return -1;
// Copy mailbox data into host structure
memcpy((caddr_t)mbox, (caddr_t)raw_mbox, 14);
mbox->cmdid = 0xFE;
mbox->busy = 1;
mbox->poll = 0;
mbox->ack = 0;
mbox->numstatus = 0xFF;
mbox->status = 0xFF;
wmb();
WRINDOOR(raid_dev, raid_dev->mbox_dma | 0x1);
[SCSI] megaraid_{mm,mbox}: fix a bug in reset handler When abort failed, the driver gets reset handleer called. In the reset handler, driver calls 'scsi_done()' callback for same SCSI command packet (struct scsi_cmnd) multiple times if there are multiple SCSI command packet in the pend_list. More over, if there are entry in the pend_lsit with IOCTL packet associated, the driver returns it to wrong free_list so that, in turn, the driver could end up with 'NULL pointer dereference..' during I/O command building with incorrect resource. Also, the patch contains several minor/cosmetic changes besides this. Signed-off-by: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-04-27 17:33:06 +08:00
for (i = 0; i < MBOX_SYNC_WAIT_CNT; i++) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (mbox->numstatus != 0xFF) break;
[SCSI] megaraid_{mbox,mm} : remove PCI Id overlaping between megaraid_legacy and megaraid_{mbox,mm} Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-18 02:17:25 +08:00
rmb();
[SCSI] megaraid_{mm,mbox}: fix a bug in reset handler When abort failed, the driver gets reset handleer called. In the reset handler, driver calls 'scsi_done()' callback for same SCSI command packet (struct scsi_cmnd) multiple times if there are multiple SCSI command packet in the pend_list. More over, if there are entry in the pend_lsit with IOCTL packet associated, the driver returns it to wrong free_list so that, in turn, the driver could end up with 'NULL pointer dereference..' during I/O command building with incorrect resource. Also, the patch contains several minor/cosmetic changes besides this. Signed-off-by: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-04-27 17:33:06 +08:00
udelay(MBOX_SYNC_DELAY_200);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
}
[SCSI] megaraid_{mm,mbox}: fix a bug in reset handler When abort failed, the driver gets reset handleer called. In the reset handler, driver calls 'scsi_done()' callback for same SCSI command packet (struct scsi_cmnd) multiple times if there are multiple SCSI command packet in the pend_list. More over, if there are entry in the pend_lsit with IOCTL packet associated, the driver returns it to wrong free_list so that, in turn, the driver could end up with 'NULL pointer dereference..' during I/O command building with incorrect resource. Also, the patch contains several minor/cosmetic changes besides this. Signed-off-by: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-04-27 17:33:06 +08:00
if (i == MBOX_SYNC_WAIT_CNT) {
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
// We may need to re-calibrate the counter
con_log(CL_ANN, (KERN_CRIT
"megaraid: fast sync command timed out\n"));
}
WRINDOOR(raid_dev, raid_dev->mbox_dma | 0x2);
wmb();
return mbox->status;
}
/**
* megaraid_busywait_mbox() - Wait until the controller's mailbox is available
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @raid_dev : RAID device (HBA) soft state
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Wait until the controller's mailbox is available to accept more commands.
* Wait for at most 1 second.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static int
megaraid_busywait_mbox(mraid_device_t *raid_dev)
{
mbox_t *mbox = raid_dev->mbox;
int i = 0;
if (mbox->busy) {
udelay(25);
for (i = 0; mbox->busy && i < 1000; i++)
msleep(1);
}
if (i < 1000) return 0;
else return -1;
}
/**
* megaraid_mbox_product_info - some static information about the controller
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : our soft state
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Issue commands to the controller to grab some parameters required by our
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* caller.
*/
static int
megaraid_mbox_product_info(adapter_t *adapter)
{
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
mbox_t *mbox;
uint8_t raw_mbox[sizeof(mbox_t)];
mraid_pinfo_t *pinfo;
dma_addr_t pinfo_dma_h;
mraid_inquiry3_t *mraid_inq3;
int i;
memset((caddr_t)raw_mbox, 0, sizeof(raw_mbox));
mbox = (mbox_t *)raw_mbox;
/*
* Issue an ENQUIRY3 command to find out certain adapter parameters,
* e.g., max channels, max commands etc.
*/
cross-tree: phase out dma_zalloc_coherent() We already need to zero out memory for dma_alloc_coherent(), as such using dma_zalloc_coherent() is superflous. Phase it out. This change was generated with the following Coccinelle SmPL patch: @ replace_dma_zalloc_coherent @ expression dev, size, data, handle, flags; @@ -dma_zalloc_coherent(dev, size, handle, flags) +dma_alloc_coherent(dev, size, handle, flags) Suggested-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Luis Chamberlain <mcgrof@kernel.org> [hch: re-ran the script on the latest tree] Signed-off-by: Christoph Hellwig <hch@lst.de>
2019-01-04 16:23:09 +08:00
pinfo = dma_alloc_coherent(&adapter->pdev->dev, sizeof(mraid_pinfo_t),
&pinfo_dma_h, GFP_KERNEL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (pinfo == NULL) {
con_log(CL_ANN, (KERN_WARNING
[SCSI] replace __FUNCTION__ with __func__ [jejb: fixed up a ton of missed conversions. All of you are on notice this has happened, driver trees will now need to be rebased] Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Cc: SCSI List <linux-scsi@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-07-04 14:47:27 +08:00
"megaraid: out of memory, %s %d\n", __func__,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
__LINE__));
return -1;
}
mbox->xferaddr = (uint32_t)adapter->ibuf_dma_h;
memset((void *)adapter->ibuf, 0, MBOX_IBUF_SIZE);
raw_mbox[0] = FC_NEW_CONFIG;
raw_mbox[2] = NC_SUBOP_ENQUIRY3;
raw_mbox[3] = ENQ3_GET_SOLICITED_FULL;
// Issue the command
if (mbox_post_sync_cmd(adapter, raw_mbox) != 0) {
con_log(CL_ANN, (KERN_WARNING "megaraid: Inquiry3 failed\n"));
scsi: megaraid_mbox: switch to generic DMA API Switch from the legacy PCI DMA API to the generic DMA API. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-10-11 15:34:24 +08:00
dma_free_coherent(&adapter->pdev->dev, sizeof(mraid_pinfo_t),
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
pinfo, pinfo_dma_h);
return -1;
}
/*
* Collect information about state of each physical drive
* attached to the controller. We will expose all the disks
* which are not part of RAID
*/
mraid_inq3 = (mraid_inquiry3_t *)adapter->ibuf;
for (i = 0; i < MBOX_MAX_PHYSICAL_DRIVES; i++) {
raid_dev->pdrv_state[i] = mraid_inq3->pdrv_state[i];
}
/*
* Get product info for information like number of channels,
* maximum commands supported.
*/
memset((caddr_t)raw_mbox, 0, sizeof(raw_mbox));
mbox->xferaddr = (uint32_t)pinfo_dma_h;
raw_mbox[0] = FC_NEW_CONFIG;
raw_mbox[2] = NC_SUBOP_PRODUCT_INFO;
if (mbox_post_sync_cmd(adapter, raw_mbox) != 0) {
con_log(CL_ANN, (KERN_WARNING
"megaraid: product info failed\n"));
scsi: megaraid_mbox: switch to generic DMA API Switch from the legacy PCI DMA API to the generic DMA API. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-10-11 15:34:24 +08:00
dma_free_coherent(&adapter->pdev->dev, sizeof(mraid_pinfo_t),
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
pinfo, pinfo_dma_h);
return -1;
}
/*
* Setup some parameters for host, as required by our caller
*/
adapter->max_channel = pinfo->nchannels;
/*
* we will export all the logical drives on a single channel.
* Add 1 since inquires do not come for inititor ID
*/
adapter->max_target = MAX_LOGICAL_DRIVES_40LD + 1;
adapter->max_lun = 8; // up to 8 LUNs for non-disk devices
/*
* These are the maximum outstanding commands for the scsi-layer
*/
adapter->max_cmds = MBOX_MAX_SCSI_CMDS;
memset(adapter->fw_version, 0, VERSION_SIZE);
memset(adapter->bios_version, 0, VERSION_SIZE);
memcpy(adapter->fw_version, pinfo->fw_version, 4);
adapter->fw_version[4] = 0;
memcpy(adapter->bios_version, pinfo->bios_version, 4);
adapter->bios_version[4] = 0;
con_log(CL_ANN, (KERN_NOTICE
"megaraid: fw version:[%s] bios version:[%s]\n",
adapter->fw_version, adapter->bios_version));
scsi: megaraid_mbox: switch to generic DMA API Switch from the legacy PCI DMA API to the generic DMA API. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-10-11 15:34:24 +08:00
dma_free_coherent(&adapter->pdev->dev, sizeof(mraid_pinfo_t), pinfo,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
pinfo_dma_h);
return 0;
}
/**
* megaraid_mbox_extended_cdb - check for support for extended CDBs
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : soft state for the controller
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* This routine check whether the controller in question supports extended
* ( > 10 bytes ) CDBs.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static int
megaraid_mbox_extended_cdb(adapter_t *adapter)
{
mbox_t *mbox;
uint8_t raw_mbox[sizeof(mbox_t)];
int rval;
mbox = (mbox_t *)raw_mbox;
memset((caddr_t)raw_mbox, 0, sizeof(raw_mbox));
mbox->xferaddr = (uint32_t)adapter->ibuf_dma_h;
memset((void *)adapter->ibuf, 0, MBOX_IBUF_SIZE);
raw_mbox[0] = MAIN_MISC_OPCODE;
raw_mbox[2] = SUPPORT_EXT_CDB;
/*
* Issue the command
*/
rval = 0;
if (mbox_post_sync_cmd(adapter, raw_mbox) != 0) {
rval = -1;
}
return rval;
}
/**
* megaraid_mbox_support_ha - Do we support clustering
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : soft state for the controller
* @init_id : ID of the initiator
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* Determine if the firmware supports clustering and the ID of the initiator.
*/
static int
megaraid_mbox_support_ha(adapter_t *adapter, uint16_t *init_id)
{
mbox_t *mbox;
uint8_t raw_mbox[sizeof(mbox_t)];
int rval;
mbox = (mbox_t *)raw_mbox;
memset((caddr_t)raw_mbox, 0, sizeof(raw_mbox));
mbox->xferaddr = (uint32_t)adapter->ibuf_dma_h;
memset((void *)adapter->ibuf, 0, MBOX_IBUF_SIZE);
raw_mbox[0] = GET_TARGET_ID;
// Issue the command
*init_id = 7;
rval = -1;
if (mbox_post_sync_cmd(adapter, raw_mbox) == 0) {
*init_id = *(uint8_t *)adapter->ibuf;
con_log(CL_ANN, (KERN_INFO
"megaraid: cluster firmware, initiator ID: %d\n",
*init_id));
rval = 0;
}
return rval;
}
/**
* megaraid_mbox_support_random_del - Do we support random deletion
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : soft state for the controller
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Determine if the firmware supports random deletion.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* Return: 1 is operation supported, 0 otherwise
*/
static int
megaraid_mbox_support_random_del(adapter_t *adapter)
{
uint8_t raw_mbox[sizeof(mbox_t)];
int rval;
[SCSI] megaraid_mbox: fix Dell CERC firmware problem Newer Dell CERC firmware (>= 6.62) implement a random deletion handling compatible with the legacy megaraid driver. The legacy handling shifted the target ID by 0x80 only for I/O commands (READ/WRITE/etc), whereas megaraid_mbox shifts the target ID always if random deletion is supported. The resulted in megaraid_mbox sending an INQUIRY to the wrong channel, and not finding any devices, obviously. So we disable the random deletion support if the offending firmware is found. Addresses http://bugzilla.kernel.org/show_bug.cgi?id=6695 Signed-off-by: Hannes Reinecke <hare@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Acked-by: "Yang, Bo" <Bo.Yang@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-04-19 04:57:20 +08:00
/*
* Newer firmware on Dell CERC expect a different
* random deletion handling, so disable it.
*/
if (adapter->pdev->vendor == PCI_VENDOR_ID_AMI &&
adapter->pdev->device == PCI_DEVICE_ID_AMI_MEGARAID3 &&
adapter->pdev->subsystem_vendor == PCI_VENDOR_ID_DELL &&
adapter->pdev->subsystem_device == PCI_SUBSYS_ID_CERC_ATA100_4CH &&
(adapter->fw_version[0] > '6' ||
(adapter->fw_version[0] == '6' &&
adapter->fw_version[2] > '6') ||
(adapter->fw_version[0] == '6'
&& adapter->fw_version[2] == '6'
&& adapter->fw_version[3] > '1'))) {
con_log(CL_DLEVEL1, ("megaraid: disable random deletion\n"));
return 0;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
memset((caddr_t)raw_mbox, 0, sizeof(mbox_t));
raw_mbox[0] = FC_DEL_LOGDRV;
raw_mbox[2] = OP_SUP_DEL_LOGDRV;
// Issue the command
rval = 0;
if (mbox_post_sync_cmd(adapter, raw_mbox) == 0) {
con_log(CL_DLEVEL1, ("megaraid: supports random deletion\n"));
rval = 1;
}
return rval;
}
/**
* megaraid_mbox_get_max_sg - maximum sg elements supported by the firmware
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : soft state for the controller
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* Find out the maximum number of scatter-gather elements supported by the
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* firmware.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static int
megaraid_mbox_get_max_sg(adapter_t *adapter)
{
mbox_t *mbox;
uint8_t raw_mbox[sizeof(mbox_t)];
int nsg;
mbox = (mbox_t *)raw_mbox;
memset((caddr_t)raw_mbox, 0, sizeof(mbox_t));
mbox->xferaddr = (uint32_t)adapter->ibuf_dma_h;
memset((void *)adapter->ibuf, 0, MBOX_IBUF_SIZE);
raw_mbox[0] = MAIN_MISC_OPCODE;
raw_mbox[2] = GET_MAX_SG_SUPPORT;
// Issue the command
if (mbox_post_sync_cmd(adapter, raw_mbox) == 0) {
nsg = *(uint8_t *)adapter->ibuf;
}
else {
nsg = MBOX_DEFAULT_SG_SIZE;
}
if (nsg > MBOX_MAX_SG_SIZE) nsg = MBOX_MAX_SG_SIZE;
return nsg;
}
/**
* megaraid_mbox_enum_raid_scsi - enumerate the RAID and SCSI channels
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : soft state for the controller
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Enumerate the RAID and SCSI channels for ROMB platforms so that channels
* can be exported as regular SCSI channels.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static void
megaraid_mbox_enum_raid_scsi(adapter_t *adapter)
{
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
mbox_t *mbox;
uint8_t raw_mbox[sizeof(mbox_t)];
mbox = (mbox_t *)raw_mbox;
memset((caddr_t)raw_mbox, 0, sizeof(mbox_t));
mbox->xferaddr = (uint32_t)adapter->ibuf_dma_h;
memset((void *)adapter->ibuf, 0, MBOX_IBUF_SIZE);
raw_mbox[0] = CHNL_CLASS;
raw_mbox[2] = GET_CHNL_CLASS;
// Issue the command. If the command fails, all channels are RAID
// channels
raid_dev->channel_class = 0xFF;
if (mbox_post_sync_cmd(adapter, raw_mbox) == 0) {
raid_dev->channel_class = *(uint8_t *)adapter->ibuf;
}
return;
}
/**
* megaraid_mbox_flush_cache - flush adapter and disks cache
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : soft state for the controller
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Flush adapter cache followed by disks cache.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static void
megaraid_mbox_flush_cache(adapter_t *adapter)
{
uint8_t raw_mbox[sizeof(mbox_t)];
memset((caddr_t)raw_mbox, 0, sizeof(mbox_t));
raw_mbox[0] = FLUSH_ADAPTER;
if (mbox_post_sync_cmd(adapter, raw_mbox) != 0) {
con_log(CL_ANN, ("megaraid: flush adapter failed\n"));
}
raw_mbox[0] = FLUSH_SYSTEM;
if (mbox_post_sync_cmd(adapter, raw_mbox) != 0) {
con_log(CL_ANN, ("megaraid: flush disks cache failed\n"));
}
return;
}
[SCSI] megaraid_{mm,mbox}: init fix for kdump 1. Changes in Initialization to fix kdump failure. Send SYNC command on loading. This command clears the pending commands in the adapter and re-initialize its internal RAID structure. Without this change, megaraid driver either panics or fails to initialize the adapter during kdump's second kernel boot if there are pending commands or interrupts from other devices sharing the same IRQ. 2. Authors email-id domain name changed from lsil.com to lsi.com. Also modified the MODULE_AUTHOR to megaraidlinux@lsi.com Signed-off-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-05 23:10:09 +08:00
/**
* megaraid_mbox_fire_sync_cmd - fire the sync cmd
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : soft state for the controller
[SCSI] megaraid_{mm,mbox}: init fix for kdump 1. Changes in Initialization to fix kdump failure. Send SYNC command on loading. This command clears the pending commands in the adapter and re-initialize its internal RAID structure. Without this change, megaraid driver either panics or fails to initialize the adapter during kdump's second kernel boot if there are pending commands or interrupts from other devices sharing the same IRQ. 2. Authors email-id domain name changed from lsil.com to lsi.com. Also modified the MODULE_AUTHOR to megaraidlinux@lsi.com Signed-off-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-05 23:10:09 +08:00
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Clears the pending cmds in FW and reinits its RAID structs.
[SCSI] megaraid_{mm,mbox}: init fix for kdump 1. Changes in Initialization to fix kdump failure. Send SYNC command on loading. This command clears the pending commands in the adapter and re-initialize its internal RAID structure. Without this change, megaraid driver either panics or fails to initialize the adapter during kdump's second kernel boot if there are pending commands or interrupts from other devices sharing the same IRQ. 2. Authors email-id domain name changed from lsil.com to lsi.com. Also modified the MODULE_AUTHOR to megaraidlinux@lsi.com Signed-off-by: Sumant Patro <sumant.patro@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-05 23:10:09 +08:00
*/
static int
megaraid_mbox_fire_sync_cmd(adapter_t *adapter)
{
mbox_t *mbox;
uint8_t raw_mbox[sizeof(mbox_t)];
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
int status = 0;
int i;
uint32_t dword;
mbox = (mbox_t *)raw_mbox;
memset((caddr_t)raw_mbox, 0, sizeof(mbox_t));
raw_mbox[0] = 0xFF;
mbox = raid_dev->mbox;
/* Wait until mailbox is free */
if (megaraid_busywait_mbox(raid_dev) != 0) {
status = 1;
goto blocked_mailbox;
}
/* Copy mailbox data into host structure */
memcpy((caddr_t)mbox, (caddr_t)raw_mbox, 16);
mbox->cmdid = 0xFE;
mbox->busy = 1;
mbox->poll = 0;
mbox->ack = 0;
mbox->numstatus = 0;
mbox->status = 0;
wmb();
WRINDOOR(raid_dev, raid_dev->mbox_dma | 0x1);
/* Wait for maximum 1 min for status to post.
* If the Firmware SUPPORTS the ABOVE COMMAND,
* mbox->cmd will be set to 0
* else
* the firmware will reject the command with
* mbox->numstatus set to 1
*/
i = 0;
status = 0;
while (!mbox->numstatus && mbox->cmd == 0xFF) {
rmb();
msleep(1);
i++;
if (i > 1000 * 60) {
status = 1;
break;
}
}
if (mbox->numstatus == 1)
status = 1; /*cmd not supported*/
/* Check for interrupt line */
dword = RDOUTDOOR(raid_dev);
WROUTDOOR(raid_dev, dword);
WRINDOOR(raid_dev,2);
return status;
blocked_mailbox:
con_log(CL_ANN, (KERN_WARNING "megaraid: blocked mailbox\n"));
return status;
}
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/**
* megaraid_mbox_display_scb - display SCB information, mostly debug purposes
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : controller's soft state
* @scb : SCB to be displayed
* @level : debug level for console print
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* Diplay information about the given SCB iff the current debug level is
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* verbose.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static void
megaraid_mbox_display_scb(adapter_t *adapter, scb_t *scb)
{
mbox_ccb_t *ccb;
struct scsi_cmnd *scp;
mbox_t *mbox;
int level;
int i;
ccb = (mbox_ccb_t *)scb->ccb;
scp = scb->scp;
mbox = ccb->mbox;
level = CL_DLEVEL3;
con_log(level, (KERN_NOTICE
"megaraid mailbox: status:%#x cmd:%#x id:%#x ", scb->status,
mbox->cmd, scb->sno));
con_log(level, ("sec:%#x lba:%#x addr:%#x ld:%d sg:%d\n",
mbox->numsectors, mbox->lba, mbox->xferaddr, mbox->logdrv,
mbox->numsge));
if (!scp) return;
con_log(level, (KERN_NOTICE "scsi cmnd: "));
for (i = 0; i < scp->cmd_len; i++) {
con_log(level, ("%#2.02x ", scp->cmnd[i]));
}
con_log(level, ("\n"));
return;
}
/**
* megaraid_mbox_setup_device_map - manage device ids
* @adapter : Driver's soft state
*
scsi/megaraid fixed several typos in comments Fixed several typos in comments in megaraid_mbox.c, megaraid_mm.c and megaraid_sas_fusion.h. Signed-off-by: Matthias Schid <aircrach115@gmail.com> Signed-off-by: Stefan Huber <steffhip@gmail.com> Signed-off-by: Simon Puels <simon.puels@gmail.com> Signed-off-by: Jiri Kosina <jkosina@suse.cz>
2013-06-26 22:25:53 +08:00
* Manage the device ids to have an appropriate mapping between the kernel
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* scsi addresses and megaraid scsi and logical drive addresses. We export
* scsi devices on their actual addresses, whereas the logical drives are
* exported on a virtual scsi channel.
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
*/
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
static void
megaraid_mbox_setup_device_map(adapter_t *adapter)
{
uint8_t c;
uint8_t t;
/*
* First fill the values on the logical drive channel
*/
for (t = 0; t < LSI_MAX_LOGICAL_DRIVES_64LD; t++)
adapter->device_ids[adapter->max_channel][t] =
(t < adapter->init_id) ? t : t - 1;
adapter->device_ids[adapter->max_channel][adapter->init_id] = 0xFF;
/*
* Fill the values on the physical devices channels
*/
for (c = 0; c < adapter->max_channel; c++)
for (t = 0; t < LSI_MAX_LOGICAL_DRIVES_64LD; t++)
adapter->device_ids[c][t] = (c << 8) | t;
}
/*
* END: internal commands library
*/
/*
* START: Interface for the common management module
*
drivers/scsi/: Spelling fixes Signed-off-by: Joe Perches <joe@perches.com> Acked-by: Andrew Vasquez <andrew.vasquez@qlogic.com> Acked-by: James Smart <james.smart@emulex.com> Acked-by: Darrick J. Wong <djwong@us.ibm.com> Acked-by: David Somayajulu <david.somayajulu@qlogic.com> Acked-by: Mark Salyzyn <mark_salyzyn@adaptec.com> Signed-off-by: Adrian Bunk <bunk@kernel.org>
2008-02-03 23:28:22 +08:00
* This is the module, which interfaces with the common management module to
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* provide support for ioctl and sysfs
*/
/**
drivers/scsi/: Spelling fixes Signed-off-by: Joe Perches <joe@perches.com> Acked-by: Andrew Vasquez <andrew.vasquez@qlogic.com> Acked-by: James Smart <james.smart@emulex.com> Acked-by: Darrick J. Wong <djwong@us.ibm.com> Acked-by: David Somayajulu <david.somayajulu@qlogic.com> Acked-by: Mark Salyzyn <mark_salyzyn@adaptec.com> Signed-off-by: Adrian Bunk <bunk@kernel.org>
2008-02-03 23:28:22 +08:00
* megaraid_cmm_register - register with the management module
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : HBA soft state
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* Register with the management module, which allows applications to issue
* ioctl calls to the drivers. This interface is used by the management module
* to setup sysfs support as well.
*/
static int
megaraid_cmm_register(adapter_t *adapter)
{
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
mraid_mmadp_t adp;
scb_t *scb;
mbox_ccb_t *ccb;
int rval;
int i;
// Allocate memory for the base list of scb for management module.
some kmalloc/memset ->kzalloc (tree wide) Transform some calls to kmalloc/memset to a single kzalloc (or kcalloc). Here is a short excerpt of the semantic patch performing this transformation: @@ type T2; expression x; identifier f,fld; expression E; expression E1,E2; expression e1,e2,e3,y; statement S; @@ x = - kmalloc + kzalloc (E1,E2) ... when != \(x->fld=E;\|y=f(...,x,...);\|f(...,x,...);\|x=E;\|while(...) S\|for(e1;e2;e3) S\) - memset((T2)x,0,E1); @@ expression E1,E2,E3; @@ - kzalloc(E1 * E2,E3) + kcalloc(E1,E2,E3) [akpm@linux-foundation.org: get kcalloc args the right way around] Signed-off-by: Yoann Padioleau <padator@wanadoo.fr> Cc: Richard Henderson <rth@twiddle.net> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Acked-by: Russell King <rmk@arm.linux.org.uk> Cc: Bryan Wu <bryan.wu@analog.com> Acked-by: Jiri Slaby <jirislaby@gmail.com> Cc: Dave Airlie <airlied@linux.ie> Acked-by: Roland Dreier <rolandd@cisco.com> Cc: Jiri Kosina <jkosina@suse.cz> Acked-by: Dmitry Torokhov <dtor@mail.ru> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Acked-by: Mauro Carvalho Chehab <mchehab@infradead.org> Acked-by: Pierre Ossman <drzeus-list@drzeus.cx> Cc: Jeff Garzik <jeff@garzik.org> Cc: "David S. Miller" <davem@davemloft.net> Acked-by: Greg KH <greg@kroah.com> Cc: James Bottomley <James.Bottomley@steeleye.com> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-07-19 16:49:03 +08:00
adapter->uscb_list = kcalloc(MBOX_MAX_USER_CMDS, sizeof(scb_t), GFP_KERNEL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (adapter->uscb_list == NULL) {
con_log(CL_ANN, (KERN_WARNING
[SCSI] replace __FUNCTION__ with __func__ [jejb: fixed up a ton of missed conversions. All of you are on notice this has happened, driver trees will now need to be rebased] Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Cc: SCSI List <linux-scsi@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-07-04 14:47:27 +08:00
"megaraid: out of memory, %s %d\n", __func__,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
__LINE__));
return -1;
}
// Initialize the synchronization parameters for resources for
// commands for management module
INIT_LIST_HEAD(&adapter->uscb_pool);
spin_lock_init(USER_FREE_LIST_LOCK(adapter));
// link all the packets. Note, CCB for commands, coming from the
// commom management module, mailbox physical address are already
// setup by it. We just need placeholder for that in our local command
// control blocks
for (i = 0; i < MBOX_MAX_USER_CMDS; i++) {
scb = adapter->uscb_list + i;
ccb = raid_dev->uccb_list + i;
scb->ccb = (caddr_t)ccb;
ccb->mbox64 = raid_dev->umbox64 + i;
ccb->mbox = &ccb->mbox64->mbox32;
ccb->raw_mbox = (uint8_t *)ccb->mbox;
scb->gp = 0;
// COMMAND ID 0 - (MBOX_MAX_SCSI_CMDS-1) ARE RESERVED FOR
// COMMANDS COMING FROM IO SUBSYSTEM (MID-LAYER)
scb->sno = i + MBOX_MAX_SCSI_CMDS;
scb->scp = NULL;
scb->state = SCB_FREE;
scsi: megaraid_mbox: switch to generic DMA API Switch from the legacy PCI DMA API to the generic DMA API. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-10-11 15:34:24 +08:00
scb->dma_direction = DMA_NONE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
scb->dma_type = MRAID_DMA_NONE;
scb->dev_channel = -1;
scb->dev_target = -1;
// put scb in the free pool
list_add_tail(&scb->list, &adapter->uscb_pool);
}
adp.unique_id = adapter->unique_id;
adp.drvr_type = DRVRTYPE_MBOX;
adp.drvr_data = (unsigned long)adapter;
adp.pdev = adapter->pdev;
adp.issue_uioc = megaraid_mbox_mm_handler;
[SCSI] megaraid_{mm,mbox}: fix a bug in reset handler When abort failed, the driver gets reset handleer called. In the reset handler, driver calls 'scsi_done()' callback for same SCSI command packet (struct scsi_cmnd) multiple times if there are multiple SCSI command packet in the pend_list. More over, if there are entry in the pend_lsit with IOCTL packet associated, the driver returns it to wrong free_list so that, in turn, the driver could end up with 'NULL pointer dereference..' during I/O command building with incorrect resource. Also, the patch contains several minor/cosmetic changes besides this. Signed-off-by: Seokmann Ju <seokmann.ju@lsil.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2006-04-27 17:33:06 +08:00
adp.timeout = MBOX_RESET_WAIT + MBOX_RESET_EXT_WAIT;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
adp.max_kioc = MBOX_MAX_USER_CMDS;
if ((rval = mraid_mm_register_adp(&adp)) != 0) {
con_log(CL_ANN, (KERN_WARNING
"megaraid mbox: did not register with CMM\n"));
kfree(adapter->uscb_list);
}
return rval;
}
/**
drivers/scsi/: Spelling fixes Signed-off-by: Joe Perches <joe@perches.com> Acked-by: Andrew Vasquez <andrew.vasquez@qlogic.com> Acked-by: James Smart <james.smart@emulex.com> Acked-by: Darrick J. Wong <djwong@us.ibm.com> Acked-by: David Somayajulu <david.somayajulu@qlogic.com> Acked-by: Mark Salyzyn <mark_salyzyn@adaptec.com> Signed-off-by: Adrian Bunk <bunk@kernel.org>
2008-02-03 23:28:22 +08:00
* megaraid_cmm_unregister - un-register with the management module
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : HBA soft state
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* Un-register with the management module.
* FIXME: mgmt module must return failure for unregister if it has pending
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* commands in LLD.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static int
megaraid_cmm_unregister(adapter_t *adapter)
{
kfree(adapter->uscb_list);
mraid_mm_unregister_adp(adapter->unique_id);
return 0;
}
/**
* megaraid_mbox_mm_handler - interface for CMM to issue commands to LLD
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @drvr_data : LLD specific data
* @kioc : CMM interface packet
* @action : command action
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
drivers/scsi/: Spelling fixes Signed-off-by: Joe Perches <joe@perches.com> Acked-by: Andrew Vasquez <andrew.vasquez@qlogic.com> Acked-by: James Smart <james.smart@emulex.com> Acked-by: Darrick J. Wong <djwong@us.ibm.com> Acked-by: David Somayajulu <david.somayajulu@qlogic.com> Acked-by: Mark Salyzyn <mark_salyzyn@adaptec.com> Signed-off-by: Adrian Bunk <bunk@kernel.org>
2008-02-03 23:28:22 +08:00
* This routine is invoked whenever the Common Management Module (CMM) has a
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* command for us. The 'action' parameter specifies if this is a new command
* or otherwise.
*/
static int
megaraid_mbox_mm_handler(unsigned long drvr_data, uioc_t *kioc, uint32_t action)
{
adapter_t *adapter;
if (action != IOCTL_ISSUE) {
con_log(CL_ANN, (KERN_WARNING
"megaraid: unsupported management action:%#2x\n",
action));
return (-ENOTSUPP);
}
adapter = (adapter_t *)drvr_data;
// make sure this adapter is not being detached right now.
if (atomic_read(&adapter->being_detached)) {
con_log(CL_ANN, (KERN_WARNING
"megaraid: reject management request, detaching\n"));
return (-ENODEV);
}
switch (kioc->opcode) {
case GET_ADAP_INFO:
kioc->status = gather_hbainfo(adapter, (mraid_hba_info_t *)
(unsigned long)kioc->buf_vaddr);
kioc->done(kioc);
return kioc->status;
case MBOX_CMD:
return megaraid_mbox_mm_command(adapter, kioc);
default:
kioc->status = (-EINVAL);
kioc->done(kioc);
return (-EINVAL);
}
return 0; // not reached
}
/**
* megaraid_mbox_mm_command - issues commands routed through CMM
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : HBA soft state
* @kioc : management command packet
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* Issues commands, which are routed through the management module.
*/
static int
megaraid_mbox_mm_command(adapter_t *adapter, uioc_t *kioc)
{
struct list_head *head = &adapter->uscb_pool;
mbox64_t *mbox64;
uint8_t *raw_mbox;
scb_t *scb;
mbox_ccb_t *ccb;
unsigned long flags;
// detach one scb from free pool
spin_lock_irqsave(USER_FREE_LIST_LOCK(adapter), flags);
if (list_empty(head)) { // should never happen because of CMM
con_log(CL_ANN, (KERN_WARNING
"megaraid mbox: bug in cmm handler, lost resources\n"));
spin_unlock_irqrestore(USER_FREE_LIST_LOCK(adapter), flags);
return (-EINVAL);
}
scb = list_entry(head->next, scb_t, list);
list_del_init(&scb->list);
spin_unlock_irqrestore(USER_FREE_LIST_LOCK(adapter), flags);
scb->state = SCB_ACTIVE;
scb->dma_type = MRAID_DMA_NONE;
scsi: megaraid_mbox: switch to generic DMA API Switch from the legacy PCI DMA API to the generic DMA API. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-10-11 15:34:24 +08:00
scb->dma_direction = DMA_NONE;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
ccb = (mbox_ccb_t *)scb->ccb;
mbox64 = (mbox64_t *)(unsigned long)kioc->cmdbuf;
raw_mbox = (uint8_t *)&mbox64->mbox32;
memcpy(ccb->mbox64, mbox64, sizeof(mbox64_t));
scb->gp = (unsigned long)kioc;
/*
* If it is a logdrv random delete operation, we have to wait till
* there are no outstanding cmds at the fw and then issue it directly
*/
if (raw_mbox[0] == FC_DEL_LOGDRV && raw_mbox[2] == OP_DEL_LOGDRV) {
if (wait_till_fw_empty(adapter)) {
con_log(CL_ANN, (KERN_NOTICE
"megaraid mbox: LD delete, timed out\n"));
kioc->status = -ETIME;
scb->status = -1;
megaraid_mbox_mm_done(adapter, scb);
return (-ETIME);
}
INIT_LIST_HEAD(&scb->list);
scb->state = SCB_ISSUED;
if (mbox_post_cmd(adapter, scb) != 0) {
con_log(CL_ANN, (KERN_NOTICE
"megaraid mbox: LD delete, mailbox busy\n"));
kioc->status = -EBUSY;
scb->status = -1;
megaraid_mbox_mm_done(adapter, scb);
return (-EBUSY);
}
return 0;
}
// put the command on the pending list and execute
megaraid_mbox_runpendq(adapter, scb);
return 0;
}
static int
wait_till_fw_empty(adapter_t *adapter)
{
unsigned long flags = 0;
int i;
/*
* Set the quiescent flag to stop issuing cmds to FW.
*/
[SCSI] megaraid_mbox: remove scsi_assign_lock usage also remove the adapter->host_lock alias for adapter->lock and remove some superflous locking aswell as removing the tiny locking wrappers for the EH routines. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 04:06:02 +08:00
spin_lock_irqsave(&adapter->lock, flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
adapter->quiescent++;
[SCSI] megaraid_mbox: remove scsi_assign_lock usage also remove the adapter->host_lock alias for adapter->lock and remove some superflous locking aswell as removing the tiny locking wrappers for the EH routines. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2005-11-01 04:06:02 +08:00
spin_unlock_irqrestore(&adapter->lock, flags);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/*
* Wait till there are no more cmds outstanding at FW. Try for at most
* 60 seconds
*/
for (i = 0; i < 60 && adapter->outstanding_cmds; i++) {
con_log(CL_DLEVEL1, (KERN_INFO
"megaraid: FW has %d pending commands\n",
adapter->outstanding_cmds));
msleep(1000);
}
return adapter->outstanding_cmds;
}
/**
* megaraid_mbox_mm_done - callback for CMM commands
* @adapter : HBA soft state
* @scb : completed command
*
* Callback routine for internal commands originated from the management
* module.
*/
static void
megaraid_mbox_mm_done(adapter_t *adapter, scb_t *scb)
{
uioc_t *kioc;
mbox64_t *mbox64;
uint8_t *raw_mbox;
unsigned long flags;
kioc = (uioc_t *)scb->gp;
mbox64 = (mbox64_t *)(unsigned long)kioc->cmdbuf;
mbox64->mbox32.status = scb->status;
raw_mbox = (uint8_t *)&mbox64->mbox32;
// put scb in the free pool
scb->state = SCB_FREE;
scb->scp = NULL;
spin_lock_irqsave(USER_FREE_LIST_LOCK(adapter), flags);
list_add(&scb->list, &adapter->uscb_pool);
spin_unlock_irqrestore(USER_FREE_LIST_LOCK(adapter), flags);
// if a delete logical drive operation succeeded, restart the
// controller
if (raw_mbox[0] == FC_DEL_LOGDRV && raw_mbox[2] == OP_DEL_LOGDRV) {
adapter->quiescent--;
megaraid_mbox_runpendq(adapter, NULL);
}
kioc->done(kioc);
return;
}
/**
* gather_hbainfo - HBA characteristics for the applications
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : HBA soft state
* @hinfo : pointer to the caller's host info strucuture
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static int
gather_hbainfo(adapter_t *adapter, mraid_hba_info_t *hinfo)
{
hinfo->pci_vendor_id = adapter->pdev->vendor;
hinfo->pci_device_id = adapter->pdev->device;
hinfo->subsys_vendor_id = adapter->pdev->subsystem_vendor;
hinfo->subsys_device_id = adapter->pdev->subsystem_device;
hinfo->pci_bus = adapter->pdev->bus->number;
hinfo->pci_dev_fn = adapter->pdev->devfn;
hinfo->pci_slot = PCI_SLOT(adapter->pdev->devfn);
hinfo->irq = adapter->host->irq;
hinfo->baseport = ADAP2RAIDDEV(adapter)->baseport;
hinfo->unique_id = (hinfo->pci_bus << 8) | adapter->pdev->devfn;
hinfo->host_no = adapter->host->host_no;
return 0;
}
/*
* END: Interface for the common management module
*/
/**
* megaraid_sysfs_alloc_resources - allocate sysfs related resources
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : controller's soft state
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* Allocate packets required to issue FW calls whenever the sysfs attributes
* are read. These attributes would require up-to-date information from the
* FW. Also set up resources for mutual exclusion to share these resources and
* the wait queue.
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Return 0 on success.
* Return -ERROR_CODE on failure.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static int
megaraid_sysfs_alloc_resources(adapter_t *adapter)
{
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
int rval = 0;
raid_dev->sysfs_uioc = kmalloc(sizeof(uioc_t), GFP_KERNEL);
raid_dev->sysfs_mbox64 = kmalloc(sizeof(mbox64_t), GFP_KERNEL);
scsi: megaraid_mbox: switch to generic DMA API Switch from the legacy PCI DMA API to the generic DMA API. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-10-11 15:34:24 +08:00
raid_dev->sysfs_buffer = dma_alloc_coherent(&adapter->pdev->dev,
PAGE_SIZE, &raid_dev->sysfs_buffer_dma, GFP_KERNEL);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (!raid_dev->sysfs_uioc || !raid_dev->sysfs_mbox64 ||
!raid_dev->sysfs_buffer) {
con_log(CL_ANN, (KERN_WARNING
[SCSI] replace __FUNCTION__ with __func__ [jejb: fixed up a ton of missed conversions. All of you are on notice this has happened, driver trees will now need to be rebased] Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Cc: SCSI List <linux-scsi@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
2008-07-04 14:47:27 +08:00
"megaraid: out of memory, %s %d\n", __func__,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
__LINE__));
rval = -ENOMEM;
megaraid_sysfs_free_resources(adapter);
}
[SCSI] megaraid_mbox: use mutex instead of semaphore The Megaraid Mailbox driver uses a semaphore as mutex. Use the mutex API instead of the (binary) semaphore. Signed-off-by: Matthias Kaehlcke <matthias.kaehlcke@gmail.com> Acked-by: "Patro, Sumant" <Sumant.Patro@lsi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-07-10 03:00:11 +08:00
mutex_init(&raid_dev->sysfs_mtx);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
init_waitqueue_head(&raid_dev->sysfs_wait_q);
return rval;
}
/**
* megaraid_sysfs_free_resources - free sysfs related resources
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : controller's soft state
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* Free packets allocated for sysfs FW commands
*/
static void
megaraid_sysfs_free_resources(adapter_t *adapter)
{
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
[PATCH] kfree cleanup: drivers/scsi This is the drivers/scsi/ part of the big kfree cleanup patch. Remove pointless checks for NULL prior to calling kfree() in drivers/scsi/. Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com> Cc: James Bottomley <James.Bottomley@steeleye.com> Acked-by: Kai Makisara <kai.makisara@kolumbus.fi> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-11-07 17:01:26 +08:00
kfree(raid_dev->sysfs_uioc);
kfree(raid_dev->sysfs_mbox64);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
if (raid_dev->sysfs_buffer) {
scsi: megaraid_mbox: switch to generic DMA API Switch from the legacy PCI DMA API to the generic DMA API. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2018-10-11 15:34:24 +08:00
dma_free_coherent(&adapter->pdev->dev, PAGE_SIZE,
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
raid_dev->sysfs_buffer, raid_dev->sysfs_buffer_dma);
}
}
/**
* megaraid_sysfs_get_ldmap_done - callback for get ldmap
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @uioc : completed packet
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* Callback routine called in the ISR/tasklet context for get ldmap call
*/
static void
megaraid_sysfs_get_ldmap_done(uioc_t *uioc)
{
adapter_t *adapter = (adapter_t *)uioc->buf_vaddr;
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
uioc->status = 0;
wake_up(&raid_dev->sysfs_wait_q);
}
/**
* megaraid_sysfs_get_ldmap_timeout - timeout handling for get ldmap
scsi: megaraid: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Also consolidates the timer setup functions arguments, which are all identical, and corrects on-stack timer usage. Cc: Kashyap Desai <kashyap.desai@broadcom.com> Cc: Sumit Saxena <sumit.saxena@broadcom.com> Cc: Shivasharan S <shivasharan.srikanteshwara@broadcom.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: megaraidlinux.pdl@broadcom.com Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-10-23 06:30:04 +08:00
* @t : timed out timer
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* Timeout routine to recover and return to application, in case the adapter
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* has stopped responding. A timeout of 60 seconds for this command seems like
* a good value.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static void
scsi: megaraid: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Also consolidates the timer setup functions arguments, which are all identical, and corrects on-stack timer usage. Cc: Kashyap Desai <kashyap.desai@broadcom.com> Cc: Sumit Saxena <sumit.saxena@broadcom.com> Cc: Shivasharan S <shivasharan.srikanteshwara@broadcom.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: megaraidlinux.pdl@broadcom.com Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-10-23 06:30:04 +08:00
megaraid_sysfs_get_ldmap_timeout(struct timer_list *t)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
scsi: megaraid: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Also consolidates the timer setup functions arguments, which are all identical, and corrects on-stack timer usage. Cc: Kashyap Desai <kashyap.desai@broadcom.com> Cc: Sumit Saxena <sumit.saxena@broadcom.com> Cc: Shivasharan S <shivasharan.srikanteshwara@broadcom.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: megaraidlinux.pdl@broadcom.com Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-10-23 06:30:04 +08:00
struct uioc_timeout *timeout = from_timer(timeout, t, timer);
uioc_t *uioc = timeout->uioc;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
adapter_t *adapter = (adapter_t *)uioc->buf_vaddr;
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
uioc->status = -ETIME;
wake_up(&raid_dev->sysfs_wait_q);
}
/**
* megaraid_sysfs_get_ldmap - get update logical drive map
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @adapter : controller's soft state
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* This routine will be called whenever user reads the logical drive
* attributes, go get the current logical drive mapping table from the
drivers/scsi/: Spelling fixes Signed-off-by: Joe Perches <joe@perches.com> Acked-by: Andrew Vasquez <andrew.vasquez@qlogic.com> Acked-by: James Smart <james.smart@emulex.com> Acked-by: Darrick J. Wong <djwong@us.ibm.com> Acked-by: David Somayajulu <david.somayajulu@qlogic.com> Acked-by: Mark Salyzyn <mark_salyzyn@adaptec.com> Signed-off-by: Adrian Bunk <bunk@kernel.org>
2008-02-03 23:28:22 +08:00
* firmware. We use the management API's to issue commands to the controller.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* NOTE: The commands issuance functionality is not generalized and
* implemented in context of "get ld map" command only. If required, the
* command issuance logical can be trivially pulled out and implemented as a
Fix common misspellings Fixes generated by 'codespell' and manually reviewed. Signed-off-by: Lucas De Marchi <lucas.demarchi@profusion.mobi>
2011-03-31 09:57:33 +08:00
* standalone library. For now, this should suffice since there is no other
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
* user of this interface.
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* Return 0 on success.
* Return -1 on failure.
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static int
megaraid_sysfs_get_ldmap(adapter_t *adapter)
{
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
uioc_t *uioc;
mbox64_t *mbox64;
mbox_t *mbox;
char *raw_mbox;
scsi: megaraid: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Also consolidates the timer setup functions arguments, which are all identical, and corrects on-stack timer usage. Cc: Kashyap Desai <kashyap.desai@broadcom.com> Cc: Sumit Saxena <sumit.saxena@broadcom.com> Cc: Shivasharan S <shivasharan.srikanteshwara@broadcom.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: megaraidlinux.pdl@broadcom.com Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-10-23 06:30:04 +08:00
struct uioc_timeout timeout;
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
caddr_t ldmap;
int rval = 0;
/*
* Allow only one read at a time to go through the sysfs attributes
*/
[SCSI] megaraid_mbox: use mutex instead of semaphore The Megaraid Mailbox driver uses a semaphore as mutex. Use the mutex API instead of the (binary) semaphore. Signed-off-by: Matthias Kaehlcke <matthias.kaehlcke@gmail.com> Acked-by: "Patro, Sumant" <Sumant.Patro@lsi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-07-10 03:00:11 +08:00
mutex_lock(&raid_dev->sysfs_mtx);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
uioc = raid_dev->sysfs_uioc;
mbox64 = raid_dev->sysfs_mbox64;
ldmap = raid_dev->sysfs_buffer;
memset(uioc, 0, sizeof(uioc_t));
memset(mbox64, 0, sizeof(mbox64_t));
memset(ldmap, 0, sizeof(raid_dev->curr_ldmap));
mbox = &mbox64->mbox32;
raw_mbox = (char *)mbox;
uioc->cmdbuf = (uint64_t)(unsigned long)mbox64;
uioc->buf_vaddr = (caddr_t)adapter;
uioc->status = -ENODATA;
uioc->done = megaraid_sysfs_get_ldmap_done;
/*
* Prepare the mailbox packet to get the current logical drive mapping
* table
*/
mbox->xferaddr = (uint32_t)raid_dev->sysfs_buffer_dma;
raw_mbox[0] = FC_DEL_LOGDRV;
raw_mbox[2] = OP_GET_LDID_MAP;
/*
* Setup a timer to recover from a non-responding controller
*/
scsi: megaraid: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Also consolidates the timer setup functions arguments, which are all identical, and corrects on-stack timer usage. Cc: Kashyap Desai <kashyap.desai@broadcom.com> Cc: Sumit Saxena <sumit.saxena@broadcom.com> Cc: Shivasharan S <shivasharan.srikanteshwara@broadcom.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: megaraidlinux.pdl@broadcom.com Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-10-23 06:30:04 +08:00
timeout.uioc = uioc;
timer_setup_on_stack(&timeout.timer,
megaraid_sysfs_get_ldmap_timeout, 0);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
scsi: megaraid: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Also consolidates the timer setup functions arguments, which are all identical, and corrects on-stack timer usage. Cc: Kashyap Desai <kashyap.desai@broadcom.com> Cc: Sumit Saxena <sumit.saxena@broadcom.com> Cc: Shivasharan S <shivasharan.srikanteshwara@broadcom.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: megaraidlinux.pdl@broadcom.com Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-10-23 06:30:04 +08:00
timeout.timer.expires = jiffies + 60 * HZ;
add_timer(&timeout.timer);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
/*
* Send the command to the firmware
*/
rval = megaraid_mbox_mm_command(adapter, uioc);
if (rval == 0) { // command successfully issued
wait_event(raid_dev->sysfs_wait_q, (uioc->status != -ENODATA));
/*
* Check if the command timed out
*/
if (uioc->status == -ETIME) {
con_log(CL_ANN, (KERN_NOTICE
"megaraid: sysfs get ld map timed out\n"));
rval = -ETIME;
}
else {
rval = mbox->status;
}
if (rval == 0) {
memcpy(raid_dev->curr_ldmap, ldmap,
sizeof(raid_dev->curr_ldmap));
}
else {
con_log(CL_ANN, (KERN_NOTICE
"megaraid: get ld map failed with %x\n", rval));
}
}
else {
con_log(CL_ANN, (KERN_NOTICE
"megaraid: could not issue ldmap command:%x\n", rval));
}
scsi: megaraid: Convert timers to use timer_setup() In preparation for unconditionally passing the struct timer_list pointer to all timer callbacks, switch to using the new timer_setup() and from_timer() to pass the timer pointer explicitly. Also consolidates the timer setup functions arguments, which are all identical, and corrects on-stack timer usage. Cc: Kashyap Desai <kashyap.desai@broadcom.com> Cc: Sumit Saxena <sumit.saxena@broadcom.com> Cc: Shivasharan S <shivasharan.srikanteshwara@broadcom.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: megaraidlinux.pdl@broadcom.com Cc: linux-scsi@vger.kernel.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: Martin K. Petersen <martin.petersen@oracle.com>
2017-10-23 06:30:04 +08:00
del_timer_sync(&timeout.timer);
destroy_timer_on_stack(&timeout.timer);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
[SCSI] megaraid_mbox: use mutex instead of semaphore The Megaraid Mailbox driver uses a semaphore as mutex. Use the mutex API instead of the (binary) semaphore. Signed-off-by: Matthias Kaehlcke <matthias.kaehlcke@gmail.com> Acked-by: "Patro, Sumant" <Sumant.Patro@lsi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-07-10 03:00:11 +08:00
mutex_unlock(&raid_dev->sysfs_mtx);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
return rval;
}
/**
* megaraid_sysfs_show_app_hndl - display application handle for this adapter
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @cdev : class device object representation for the host
* @buf : buffer to send data to
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* Display the handle used by the applications while executing management
* tasks on the adapter. We invoke a management module API to get the adapter
* handle, since we do not interface with applications directly.
*/
static ssize_t
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-22 07:13:36 +08:00
megaraid_sysfs_show_app_hndl(struct device *dev, struct device_attribute *attr,
char *buf)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
SCSI: convert struct class_device to struct device It's big, but there doesn't seem to be a way to split it up smaller... Signed-off-by: Tony Jones <tonyj@suse.de> Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Cc: Roland Dreier <rolandd@cisco.com> Cc: Sean Hefty <sean.hefty@intel.com> Cc: Hal Rosenstock <hal.rosenstock@gmail.com> Cc: James Bottomley <James.Bottomley@HansenPartnership.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2008-02-22 07:13:36 +08:00
struct Scsi_Host *shost = class_to_shost(dev);
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
adapter_t *adapter = (adapter_t *)SCSIHOST2ADAP(shost);
uint32_t app_hndl;
app_hndl = mraid_mm_adapter_app_handle(adapter->unique_id);
return snprintf(buf, 8, "%u\n", app_hndl);
}
/**
* megaraid_sysfs_show_ldnum - display the logical drive number for this device
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* @dev : device object representation for the scsi device
* @attr : device attribute to show
* @buf : buffer to send data to
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* Display the logical drive number for the device in question, if it a valid
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* logical drive. For physical devices, "-1" is returned.
*
* The logical drive number is displayed in following format:
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*
* <SCSI ID> <LD NUM> <LD STICKY ID> <APP ADAPTER HANDLE>
*
[SCSI] megaraid: fix kernel-doc kernel-doc modifications: - change "@param var" notation to @var; - change function/description separator from ':' to '-'; - change var/description separator from '-' to ':'; - fix a few doc. typos; - don't use kernel-doc /** lead-in when the doc. block is not kernel-doc; - use Linux common */ ending comment format instead of **/; - use correct function parameter names; - place function parameters immediately after the function short description; - place kernel-doc immediately before its function or macro; Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com> Acked-by: Sumant Patro <sumantp@lsil.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
2007-01-06 14:41:48 +08:00
* <int> <int> <int> <int>
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
*/
static ssize_t
[PATCH] Driver Core: drivers/s390/net/qeth_sys.c - drivers/usb/gadget/pxa2xx_udc.c: update device attribute callbacks Signed-off-by: Yani Ioannou <yani.ioannou@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2005-05-17 18:43:37 +08:00
megaraid_sysfs_show_ldnum(struct device *dev, struct device_attribute *attr, char *buf)
Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip!
2005-04-17 06:20:36 +08:00
{
struct scsi_device *sdev = to_scsi_device(dev);
adapter_t *adapter = (adapter_t *)SCSIHOST2ADAP(sdev->host);
mraid_device_t *raid_dev = ADAP2RAIDDEV(adapter);
int scsi_id = -1;
int logical_drv = -1;
int ldid_map = -1;
uint32_t app_hndl = 0;
int mapped_sdev_id;
int rval;
int i;
if (raid_dev->random_del_supported &&
MRAID_IS_LOGICAL_SDEV(adapter, sdev)) {
rval = megaraid_sysfs_get_ldmap(adapter);
if (rval == 0) {
for (i = 0; i < MAX_LOGICAL_DRIVES_40LD; i++) {
mapped_sdev_id = sdev->id;
if (sdev->id > adapter->init_id) {
mapped_sdev_id -= 1;
}
if (raid_dev->curr_ldmap[i] == mapped_sdev_id) {
scsi_id = sdev->id;
logical_drv = i;
ldid_map = raid_dev->curr_ldmap[i];
app_hndl = mraid_mm_adapter_app_handle(
adapter->unique_id);
break;
}
}
}
else {
con_log(CL_ANN, (KERN_NOTICE
"megaraid: sysfs get ld map failed: %x\n",
rval));
}
}
return snprintf(buf, 36, "%d %d %d %d\n", scsi_id, logical_drv,
ldid_map, app_hndl);
}
/*
* END: Mailbox Low Level Driver
*/
module_init(megaraid_init);
module_exit(megaraid_exit);
/* vim: set ts=8 sw=8 tw=78 ai si: */