UbiquitousOS
/
OpenCloudOS-Kernel
mirror of https://gitee.com/OpenCloudOS/OpenCloudOS-Kernel.git
11
0
Fork 0
Code Issues Projects Releases Wiki Activity
OpenCloudOS-Kernel/include/asm-ia64/machvec.h

421 lines
15 KiB
C
Raw Normal View History

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
/*
* Machine vector for IA-64.
*
* Copyright (C) 1999 Silicon Graphics, Inc.
* Copyright (C) Srinivasa Thirumalachar <sprasad@engr.sgi.com>
* Copyright (C) Vijay Chander <vijay@engr.sgi.com>
* Copyright (C) 1999-2001, 2003-2004 Hewlett-Packard Co.
* David Mosberger-Tang <davidm@hpl.hp.com>
*/
#ifndef _ASM_IA64_MACHVEC_H
#define _ASM_IA64_MACHVEC_H
#include <linux/types.h>
/* forward declarations: */
struct device;
struct pt_regs;
struct scatterlist;
struct page;
struct mm_struct;
struct pci_bus;
[IA64] hooks to wait for mmio writes to drain when migrating processes On SN2, MMIO writes which are issued from separate processors are not guaranteed to arrive in any particular order at the IO hardware. When performing such writes from the kernel this is not a problem, as a kernel thread will not migrate to another CPU during execution, and mmiowb() calls can guarantee write ordering when control of the IO resource is allowed to move between threads. However, when MMIO writes can be performed from user space (e.g. DRM) there are no such guarantees and mechanisms, as the process may context-switch at any time, and may migrate to a different CPU as part of the switch. For such programs/hardware to operate correctly, it is required that the MMIO writes from the old CPU be accepted by the IO hardware before subsequent writes from the new CPU can be issued. The following patch implements this behavior on SN2 by waiting for a Shub register to indicate that these writes have been accepted. This is placed in the context switch-in path, and only performs the wait when the newly scheduled task changes CPUs. Signed-off-by: Prarit Bhargava <prarit@sgi.com> Signed-off-by: Brent Casavant <bcasavan@sgi.com>
2006-01-27 07:55:52 +08:00
struct task_struct;
[PATCH] msi: refactor and move the msi irq_chip into the arch code It turns out msi_ops was simply not enough to abstract the architecture specific details of msi. So I have moved the resposibility of constructing the struct irq_chip to the architectures, and have two architecture specific functions arch_setup_msi_irq, and arch_teardown_msi_irq. For simple architectures those functions can do all of the work. For architectures with platform dependencies they can call into the appropriate platform code. With this msi.c is finally free of assuming you have an apic, and this actually takes less code. The helpers for the architecture specific code are declared in the linux/msi.h to keep them separate from the msi functions used by drivers in linux/pci.h Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Tony Luck <tony.luck@intel.com> Cc: Andi Kleen <ak@suse.de> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Greg KH <greg@kroah.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 17:16:59 +08:00
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
typedef void ia64_mv_setup_t (char **);
typedef void ia64_mv_cpu_init_t (void);
typedef void ia64_mv_irq_init_t (void);
typedef void ia64_mv_send_ipi_t (int, int, int, int);
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
typedef void ia64_mv_timer_interrupt_t (int, void *);
[IA64] - Avoid slow TLB purges on SGI Altix systems flush_tlb_all() can be a scaling issue on large SGI Altix systems since it uses the global call_lock and always executes on all cpus. When a process enters flush_tlb_range() to purge TLBs for another process, it is possible to avoid flush_tlb_all() and instead allow sn2_global_tlb_purge() to purge TLBs only where necessary. This patch modifies flush_tlb_range() so that this case can be handled by platform TLB purge functions and updates ia64_global_tlb_purge() accordingly. sn2_global_tlb_purge() now calculates the region register value from the mm argument introduced with this patch. Signed-off-by: Dean Roe <roe@sgi.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-10-28 04:41:04 +08:00
typedef void ia64_mv_global_tlb_purge_t (struct mm_struct *, unsigned long, unsigned long, unsigned long);
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
typedef void ia64_mv_tlb_migrate_finish_t (struct mm_struct *);
typedef unsigned int ia64_mv_local_vector_to_irq (u8);
typedef char *ia64_mv_pci_get_legacy_mem_t (struct pci_bus *);
typedef int ia64_mv_pci_legacy_read_t (struct pci_bus *, u16 port, u32 *val,
u8 size);
typedef int ia64_mv_pci_legacy_write_t (struct pci_bus *, u16 port, u32 val,
u8 size);
[IA64] hooks to wait for mmio writes to drain when migrating processes On SN2, MMIO writes which are issued from separate processors are not guaranteed to arrive in any particular order at the IO hardware. When performing such writes from the kernel this is not a problem, as a kernel thread will not migrate to another CPU during execution, and mmiowb() calls can guarantee write ordering when control of the IO resource is allowed to move between threads. However, when MMIO writes can be performed from user space (e.g. DRM) there are no such guarantees and mechanisms, as the process may context-switch at any time, and may migrate to a different CPU as part of the switch. For such programs/hardware to operate correctly, it is required that the MMIO writes from the old CPU be accepted by the IO hardware before subsequent writes from the new CPU can be issued. The following patch implements this behavior on SN2 by waiting for a Shub register to indicate that these writes have been accepted. This is placed in the context switch-in path, and only performs the wait when the newly scheduled task changes CPUs. Signed-off-by: Prarit Bhargava <prarit@sgi.com> Signed-off-by: Brent Casavant <bcasavan@sgi.com>
2006-01-27 07:55:52 +08:00
typedef void ia64_mv_migrate_t(struct task_struct * task);
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
/* DMA-mapping interface: */
typedef void ia64_mv_dma_init (void);
[PATCH] gfp_t: dma-mapping (ia64) ... and related annotations for amd64 - swiotlb code is shared, but prototypes are not. Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-10-21 15:21:03 +08:00
typedef void *ia64_mv_dma_alloc_coherent (struct device *, size_t, dma_addr_t *, gfp_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
typedef void ia64_mv_dma_free_coherent (struct device *, size_t, void *, dma_addr_t);
typedef dma_addr_t ia64_mv_dma_map_single (struct device *, void *, size_t, int);
typedef void ia64_mv_dma_unmap_single (struct device *, dma_addr_t, size_t, int);
typedef int ia64_mv_dma_map_sg (struct device *, struct scatterlist *, int, int);
typedef void ia64_mv_dma_unmap_sg (struct device *, struct scatterlist *, int, int);
typedef void ia64_mv_dma_sync_single_for_cpu (struct device *, dma_addr_t, size_t, int);
typedef void ia64_mv_dma_sync_sg_for_cpu (struct device *, struct scatterlist *, int, int);
typedef void ia64_mv_dma_sync_single_for_device (struct device *, dma_addr_t, size_t, int);
typedef void ia64_mv_dma_sync_sg_for_device (struct device *, struct scatterlist *, int, int);
typedef int ia64_mv_dma_mapping_error (dma_addr_t dma_addr);
typedef int ia64_mv_dma_supported (struct device *, u64);
/*
* WARNING: The legacy I/O space is _architected_. Platforms are
* expected to follow this architected model (see Section 10.7 in the
* IA-64 Architecture Software Developer's Manual). Unfortunately,
* some broken machines do not follow that model, which is why we have
* to make the inX/outX operations part of the machine vector.
* Platform designers should follow the architected model whenever
* possible.
*/
typedef unsigned int ia64_mv_inb_t (unsigned long);
typedef unsigned int ia64_mv_inw_t (unsigned long);
typedef unsigned int ia64_mv_inl_t (unsigned long);
typedef void ia64_mv_outb_t (unsigned char, unsigned long);
typedef void ia64_mv_outw_t (unsigned short, unsigned long);
typedef void ia64_mv_outl_t (unsigned int, unsigned long);
typedef void ia64_mv_mmiowb_t (void);
typedef unsigned char ia64_mv_readb_t (const volatile void __iomem *);
typedef unsigned short ia64_mv_readw_t (const volatile void __iomem *);
typedef unsigned int ia64_mv_readl_t (const volatile void __iomem *);
typedef unsigned long ia64_mv_readq_t (const volatile void __iomem *);
typedef unsigned char ia64_mv_readb_relaxed_t (const volatile void __iomem *);
typedef unsigned short ia64_mv_readw_relaxed_t (const volatile void __iomem *);
typedef unsigned int ia64_mv_readl_relaxed_t (const volatile void __iomem *);
typedef unsigned long ia64_mv_readq_relaxed_t (const volatile void __iomem *);
[PATCH] msi: refactor and move the msi irq_chip into the arch code It turns out msi_ops was simply not enough to abstract the architecture specific details of msi. So I have moved the resposibility of constructing the struct irq_chip to the architectures, and have two architecture specific functions arch_setup_msi_irq, and arch_teardown_msi_irq. For simple architectures those functions can do all of the work. For architectures with platform dependencies they can call into the appropriate platform code. With this msi.c is finally free of assuming you have an apic, and this actually takes less code. The helpers for the architecture specific code are declared in the linux/msi.h to keep them separate from the msi functions used by drivers in linux/pci.h Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Tony Luck <tony.luck@intel.com> Cc: Andi Kleen <ak@suse.de> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Greg KH <greg@kroah.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 17:16:59 +08:00
typedef int ia64_mv_setup_msi_irq_t (unsigned int irq, struct pci_dev *pdev);
typedef void ia64_mv_teardown_msi_irq_t (unsigned int 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
static inline void
machvec_noop (void)
{
}
static inline void
machvec_noop_mm (struct mm_struct *mm)
{
}
[IA64] hooks to wait for mmio writes to drain when migrating processes On SN2, MMIO writes which are issued from separate processors are not guaranteed to arrive in any particular order at the IO hardware. When performing such writes from the kernel this is not a problem, as a kernel thread will not migrate to another CPU during execution, and mmiowb() calls can guarantee write ordering when control of the IO resource is allowed to move between threads. However, when MMIO writes can be performed from user space (e.g. DRM) there are no such guarantees and mechanisms, as the process may context-switch at any time, and may migrate to a different CPU as part of the switch. For such programs/hardware to operate correctly, it is required that the MMIO writes from the old CPU be accepted by the IO hardware before subsequent writes from the new CPU can be issued. The following patch implements this behavior on SN2 by waiting for a Shub register to indicate that these writes have been accepted. This is placed in the context switch-in path, and only performs the wait when the newly scheduled task changes CPUs. Signed-off-by: Prarit Bhargava <prarit@sgi.com> Signed-off-by: Brent Casavant <bcasavan@sgi.com>
2006-01-27 07:55:52 +08:00
static inline void
machvec_noop_task (struct task_struct *task)
{
}
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
extern void machvec_setup (char **);
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
extern void machvec_timer_interrupt (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
extern void machvec_dma_sync_single (struct device *, dma_addr_t, size_t, int);
extern void machvec_dma_sync_sg (struct device *, struct scatterlist *, int, int);
extern void machvec_tlb_migrate_finish (struct mm_struct *);
# if defined (CONFIG_IA64_HP_SIM)
# include <asm/machvec_hpsim.h>
# elif defined (CONFIG_IA64_DIG)
# include <asm/machvec_dig.h>
# elif defined (CONFIG_IA64_HP_ZX1)
# include <asm/machvec_hpzx1.h>
# elif defined (CONFIG_IA64_HP_ZX1_SWIOTLB)
# include <asm/machvec_hpzx1_swiotlb.h>
# elif defined (CONFIG_IA64_SGI_SN2)
# include <asm/machvec_sn2.h>
# elif defined (CONFIG_IA64_GENERIC)
# ifdef MACHVEC_PLATFORM_HEADER
# include MACHVEC_PLATFORM_HEADER
# else
# define platform_name ia64_mv.name
# define platform_setup ia64_mv.setup
# define platform_cpu_init ia64_mv.cpu_init
# define platform_irq_init ia64_mv.irq_init
# define platform_send_ipi ia64_mv.send_ipi
# define platform_timer_interrupt ia64_mv.timer_interrupt
# define platform_global_tlb_purge ia64_mv.global_tlb_purge
# define platform_tlb_migrate_finish ia64_mv.tlb_migrate_finish
# define platform_dma_init ia64_mv.dma_init
# define platform_dma_alloc_coherent ia64_mv.dma_alloc_coherent
# define platform_dma_free_coherent ia64_mv.dma_free_coherent
# define platform_dma_map_single ia64_mv.dma_map_single
# define platform_dma_unmap_single ia64_mv.dma_unmap_single
# define platform_dma_map_sg ia64_mv.dma_map_sg
# define platform_dma_unmap_sg ia64_mv.dma_unmap_sg
# define platform_dma_sync_single_for_cpu ia64_mv.dma_sync_single_for_cpu
# define platform_dma_sync_sg_for_cpu ia64_mv.dma_sync_sg_for_cpu
# define platform_dma_sync_single_for_device ia64_mv.dma_sync_single_for_device
# define platform_dma_sync_sg_for_device ia64_mv.dma_sync_sg_for_device
# define platform_dma_mapping_error ia64_mv.dma_mapping_error
# define platform_dma_supported ia64_mv.dma_supported
# define platform_local_vector_to_irq ia64_mv.local_vector_to_irq
# define platform_pci_get_legacy_mem ia64_mv.pci_get_legacy_mem
# define platform_pci_legacy_read ia64_mv.pci_legacy_read
# define platform_pci_legacy_write ia64_mv.pci_legacy_write
# define platform_inb ia64_mv.inb
# define platform_inw ia64_mv.inw
# define platform_inl ia64_mv.inl
# define platform_outb ia64_mv.outb
# define platform_outw ia64_mv.outw
# define platform_outl ia64_mv.outl
# define platform_mmiowb ia64_mv.mmiowb
# define platform_readb ia64_mv.readb
# define platform_readw ia64_mv.readw
# define platform_readl ia64_mv.readl
# define platform_readq ia64_mv.readq
# define platform_readb_relaxed ia64_mv.readb_relaxed
# define platform_readw_relaxed ia64_mv.readw_relaxed
# define platform_readl_relaxed ia64_mv.readl_relaxed
# define platform_readq_relaxed ia64_mv.readq_relaxed
[IA64] hooks to wait for mmio writes to drain when migrating processes On SN2, MMIO writes which are issued from separate processors are not guaranteed to arrive in any particular order at the IO hardware. When performing such writes from the kernel this is not a problem, as a kernel thread will not migrate to another CPU during execution, and mmiowb() calls can guarantee write ordering when control of the IO resource is allowed to move between threads. However, when MMIO writes can be performed from user space (e.g. DRM) there are no such guarantees and mechanisms, as the process may context-switch at any time, and may migrate to a different CPU as part of the switch. For such programs/hardware to operate correctly, it is required that the MMIO writes from the old CPU be accepted by the IO hardware before subsequent writes from the new CPU can be issued. The following patch implements this behavior on SN2 by waiting for a Shub register to indicate that these writes have been accepted. This is placed in the context switch-in path, and only performs the wait when the newly scheduled task changes CPUs. Signed-off-by: Prarit Bhargava <prarit@sgi.com> Signed-off-by: Brent Casavant <bcasavan@sgi.com>
2006-01-27 07:55:52 +08:00
# define platform_migrate ia64_mv.migrate
[PATCH] msi: refactor and move the msi irq_chip into the arch code It turns out msi_ops was simply not enough to abstract the architecture specific details of msi. So I have moved the resposibility of constructing the struct irq_chip to the architectures, and have two architecture specific functions arch_setup_msi_irq, and arch_teardown_msi_irq. For simple architectures those functions can do all of the work. For architectures with platform dependencies they can call into the appropriate platform code. With this msi.c is finally free of assuming you have an apic, and this actually takes less code. The helpers for the architecture specific code are declared in the linux/msi.h to keep them separate from the msi functions used by drivers in linux/pci.h Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Tony Luck <tony.luck@intel.com> Cc: Andi Kleen <ak@suse.de> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Greg KH <greg@kroah.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 17:16:59 +08:00
# define platform_setup_msi_irq ia64_mv.setup_msi_irq
# define platform_teardown_msi_irq ia64_mv.teardown_msi_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
# endif
/* __attribute__((__aligned__(16))) is required to make size of the
* structure multiple of 16 bytes.
* This will fillup the holes created because of section 3.3.1 in
* Software Conventions guide.
*/
struct ia64_machine_vector {
const char *name;
ia64_mv_setup_t *setup;
ia64_mv_cpu_init_t *cpu_init;
ia64_mv_irq_init_t *irq_init;
ia64_mv_send_ipi_t *send_ipi;
ia64_mv_timer_interrupt_t *timer_interrupt;
ia64_mv_global_tlb_purge_t *global_tlb_purge;
ia64_mv_tlb_migrate_finish_t *tlb_migrate_finish;
ia64_mv_dma_init *dma_init;
ia64_mv_dma_alloc_coherent *dma_alloc_coherent;
ia64_mv_dma_free_coherent *dma_free_coherent;
ia64_mv_dma_map_single *dma_map_single;
ia64_mv_dma_unmap_single *dma_unmap_single;
ia64_mv_dma_map_sg *dma_map_sg;
ia64_mv_dma_unmap_sg *dma_unmap_sg;
ia64_mv_dma_sync_single_for_cpu *dma_sync_single_for_cpu;
ia64_mv_dma_sync_sg_for_cpu *dma_sync_sg_for_cpu;
ia64_mv_dma_sync_single_for_device *dma_sync_single_for_device;
ia64_mv_dma_sync_sg_for_device *dma_sync_sg_for_device;
ia64_mv_dma_mapping_error *dma_mapping_error;
ia64_mv_dma_supported *dma_supported;
ia64_mv_local_vector_to_irq *local_vector_to_irq;
ia64_mv_pci_get_legacy_mem_t *pci_get_legacy_mem;
ia64_mv_pci_legacy_read_t *pci_legacy_read;
ia64_mv_pci_legacy_write_t *pci_legacy_write;
ia64_mv_inb_t *inb;
ia64_mv_inw_t *inw;
ia64_mv_inl_t *inl;
ia64_mv_outb_t *outb;
ia64_mv_outw_t *outw;
ia64_mv_outl_t *outl;
ia64_mv_mmiowb_t *mmiowb;
ia64_mv_readb_t *readb;
ia64_mv_readw_t *readw;
ia64_mv_readl_t *readl;
ia64_mv_readq_t *readq;
ia64_mv_readb_relaxed_t *readb_relaxed;
ia64_mv_readw_relaxed_t *readw_relaxed;
ia64_mv_readl_relaxed_t *readl_relaxed;
ia64_mv_readq_relaxed_t *readq_relaxed;
[IA64] hooks to wait for mmio writes to drain when migrating processes On SN2, MMIO writes which are issued from separate processors are not guaranteed to arrive in any particular order at the IO hardware. When performing such writes from the kernel this is not a problem, as a kernel thread will not migrate to another CPU during execution, and mmiowb() calls can guarantee write ordering when control of the IO resource is allowed to move between threads. However, when MMIO writes can be performed from user space (e.g. DRM) there are no such guarantees and mechanisms, as the process may context-switch at any time, and may migrate to a different CPU as part of the switch. For such programs/hardware to operate correctly, it is required that the MMIO writes from the old CPU be accepted by the IO hardware before subsequent writes from the new CPU can be issued. The following patch implements this behavior on SN2 by waiting for a Shub register to indicate that these writes have been accepted. This is placed in the context switch-in path, and only performs the wait when the newly scheduled task changes CPUs. Signed-off-by: Prarit Bhargava <prarit@sgi.com> Signed-off-by: Brent Casavant <bcasavan@sgi.com>
2006-01-27 07:55:52 +08:00
ia64_mv_migrate_t *migrate;
[PATCH] msi: refactor and move the msi irq_chip into the arch code It turns out msi_ops was simply not enough to abstract the architecture specific details of msi. So I have moved the resposibility of constructing the struct irq_chip to the architectures, and have two architecture specific functions arch_setup_msi_irq, and arch_teardown_msi_irq. For simple architectures those functions can do all of the work. For architectures with platform dependencies they can call into the appropriate platform code. With this msi.c is finally free of assuming you have an apic, and this actually takes less code. The helpers for the architecture specific code are declared in the linux/msi.h to keep them separate from the msi functions used by drivers in linux/pci.h Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Tony Luck <tony.luck@intel.com> Cc: Andi Kleen <ak@suse.de> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Greg KH <greg@kroah.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 17:16:59 +08:00
ia64_mv_setup_msi_irq_t *setup_msi_irq;
ia64_mv_teardown_msi_irq_t *teardown_msi_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
} __attribute__((__aligned__(16))); /* align attrib? see above comment */
#define MACHVEC_INIT(name) \
{ \
#name, \
platform_setup, \
platform_cpu_init, \
platform_irq_init, \
platform_send_ipi, \
platform_timer_interrupt, \
platform_global_tlb_purge, \
platform_tlb_migrate_finish, \
platform_dma_init, \
platform_dma_alloc_coherent, \
platform_dma_free_coherent, \
platform_dma_map_single, \
platform_dma_unmap_single, \
platform_dma_map_sg, \
platform_dma_unmap_sg, \
platform_dma_sync_single_for_cpu, \
platform_dma_sync_sg_for_cpu, \
platform_dma_sync_single_for_device, \
platform_dma_sync_sg_for_device, \
platform_dma_mapping_error, \
platform_dma_supported, \
platform_local_vector_to_irq, \
platform_pci_get_legacy_mem, \
platform_pci_legacy_read, \
platform_pci_legacy_write, \
platform_inb, \
platform_inw, \
platform_inl, \
platform_outb, \
platform_outw, \
platform_outl, \
platform_mmiowb, \
platform_readb, \
platform_readw, \
platform_readl, \
platform_readq, \
platform_readb_relaxed, \
platform_readw_relaxed, \
platform_readl_relaxed, \
platform_readq_relaxed, \
[IA64] hooks to wait for mmio writes to drain when migrating processes On SN2, MMIO writes which are issued from separate processors are not guaranteed to arrive in any particular order at the IO hardware. When performing such writes from the kernel this is not a problem, as a kernel thread will not migrate to another CPU during execution, and mmiowb() calls can guarantee write ordering when control of the IO resource is allowed to move between threads. However, when MMIO writes can be performed from user space (e.g. DRM) there are no such guarantees and mechanisms, as the process may context-switch at any time, and may migrate to a different CPU as part of the switch. For such programs/hardware to operate correctly, it is required that the MMIO writes from the old CPU be accepted by the IO hardware before subsequent writes from the new CPU can be issued. The following patch implements this behavior on SN2 by waiting for a Shub register to indicate that these writes have been accepted. This is placed in the context switch-in path, and only performs the wait when the newly scheduled task changes CPUs. Signed-off-by: Prarit Bhargava <prarit@sgi.com> Signed-off-by: Brent Casavant <bcasavan@sgi.com>
2006-01-27 07:55:52 +08:00
platform_migrate, \
[PATCH] msi: refactor and move the msi irq_chip into the arch code It turns out msi_ops was simply not enough to abstract the architecture specific details of msi. So I have moved the resposibility of constructing the struct irq_chip to the architectures, and have two architecture specific functions arch_setup_msi_irq, and arch_teardown_msi_irq. For simple architectures those functions can do all of the work. For architectures with platform dependencies they can call into the appropriate platform code. With this msi.c is finally free of assuming you have an apic, and this actually takes less code. The helpers for the architecture specific code are declared in the linux/msi.h to keep them separate from the msi functions used by drivers in linux/pci.h Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Tony Luck <tony.luck@intel.com> Cc: Andi Kleen <ak@suse.de> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Greg KH <greg@kroah.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 17:16:59 +08:00
platform_setup_msi_irq, \
platform_teardown_msi_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
}
extern struct ia64_machine_vector ia64_mv;
extern void machvec_init (const char *name);
# else
# error Unknown configuration. Update asm-ia64/machvec.h.
# endif /* CONFIG_IA64_GENERIC */
/*
* Declare default routines which aren't declared anywhere else:
*/
extern ia64_mv_dma_init swiotlb_init;
extern ia64_mv_dma_alloc_coherent swiotlb_alloc_coherent;
extern ia64_mv_dma_free_coherent swiotlb_free_coherent;
extern ia64_mv_dma_map_single swiotlb_map_single;
extern ia64_mv_dma_unmap_single swiotlb_unmap_single;
extern ia64_mv_dma_map_sg swiotlb_map_sg;
extern ia64_mv_dma_unmap_sg swiotlb_unmap_sg;
extern ia64_mv_dma_sync_single_for_cpu swiotlb_sync_single_for_cpu;
extern ia64_mv_dma_sync_sg_for_cpu swiotlb_sync_sg_for_cpu;
extern ia64_mv_dma_sync_single_for_device swiotlb_sync_single_for_device;
extern ia64_mv_dma_sync_sg_for_device swiotlb_sync_sg_for_device;
extern ia64_mv_dma_mapping_error swiotlb_dma_mapping_error;
extern ia64_mv_dma_supported swiotlb_dma_supported;
/*
* Define default versions so we can extend machvec for new platforms without having
* to update the machvec files for all existing platforms.
*/
#ifndef platform_setup
# define platform_setup machvec_setup
#endif
#ifndef platform_cpu_init
# define platform_cpu_init machvec_noop
#endif
#ifndef platform_irq_init
# define platform_irq_init machvec_noop
#endif
#ifndef platform_send_ipi
# define platform_send_ipi ia64_send_ipi /* default to architected version */
#endif
#ifndef platform_timer_interrupt
# define platform_timer_interrupt machvec_timer_interrupt
#endif
#ifndef platform_global_tlb_purge
# define platform_global_tlb_purge ia64_global_tlb_purge /* default to architected version */
#endif
#ifndef platform_tlb_migrate_finish
# define platform_tlb_migrate_finish machvec_noop_mm
#endif
#ifndef platform_dma_init
# define platform_dma_init swiotlb_init
#endif
#ifndef platform_dma_alloc_coherent
# define platform_dma_alloc_coherent swiotlb_alloc_coherent
#endif
#ifndef platform_dma_free_coherent
# define platform_dma_free_coherent swiotlb_free_coherent
#endif
#ifndef platform_dma_map_single
# define platform_dma_map_single swiotlb_map_single
#endif
#ifndef platform_dma_unmap_single
# define platform_dma_unmap_single swiotlb_unmap_single
#endif
#ifndef platform_dma_map_sg
# define platform_dma_map_sg swiotlb_map_sg
#endif
#ifndef platform_dma_unmap_sg
# define platform_dma_unmap_sg swiotlb_unmap_sg
#endif
#ifndef platform_dma_sync_single_for_cpu
# define platform_dma_sync_single_for_cpu swiotlb_sync_single_for_cpu
#endif
#ifndef platform_dma_sync_sg_for_cpu
# define platform_dma_sync_sg_for_cpu swiotlb_sync_sg_for_cpu
#endif
#ifndef platform_dma_sync_single_for_device
# define platform_dma_sync_single_for_device swiotlb_sync_single_for_device
#endif
#ifndef platform_dma_sync_sg_for_device
# define platform_dma_sync_sg_for_device swiotlb_sync_sg_for_device
#endif
#ifndef platform_dma_mapping_error
# define platform_dma_mapping_error swiotlb_dma_mapping_error
#endif
#ifndef platform_dma_supported
# define platform_dma_supported swiotlb_dma_supported
#endif
#ifndef platform_local_vector_to_irq
# define platform_local_vector_to_irq __ia64_local_vector_to_irq
#endif
#ifndef platform_pci_get_legacy_mem
# define platform_pci_get_legacy_mem ia64_pci_get_legacy_mem
#endif
#ifndef platform_pci_legacy_read
# define platform_pci_legacy_read ia64_pci_legacy_read
[IA64] eliminate compile time warnings This patch removes following compile time warnings: drivers/pci/pci-sysfs.c: In function `pci_read_legacy_io': drivers/pci/pci-sysfs.c:257: warning: implicit declaration of function `ia64_pci_legacy_read' drivers/pci/pci-sysfs.c: In function `pci_write_legacy_io': drivers/pci/pci-sysfs.c:280: warning: implicit declaration of function `ia64_pci_legacy_write' It also fixes wrong definition of ia64_pci_legacy_write (type of `bus' is not `pci_dev', but `pci_bus'). Signed-Off-By: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2006-04-20 17:49:48 +08:00
extern int ia64_pci_legacy_read(struct pci_bus *bus, u16 port, u32 *val, u8 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
#endif
#ifndef platform_pci_legacy_write
# define platform_pci_legacy_write ia64_pci_legacy_write
[IA64] eliminate compile time warnings This patch removes following compile time warnings: drivers/pci/pci-sysfs.c: In function `pci_read_legacy_io': drivers/pci/pci-sysfs.c:257: warning: implicit declaration of function `ia64_pci_legacy_read' drivers/pci/pci-sysfs.c: In function `pci_write_legacy_io': drivers/pci/pci-sysfs.c:280: warning: implicit declaration of function `ia64_pci_legacy_write' It also fixes wrong definition of ia64_pci_legacy_write (type of `bus' is not `pci_dev', but `pci_bus'). Signed-Off-By: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com> Signed-off-by: Tony Luck <tony.luck@intel.com>
2006-04-20 17:49:48 +08:00
extern int ia64_pci_legacy_write(struct pci_bus *bus, u16 port, u32 val, u8 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
#endif
#ifndef platform_inb
# define platform_inb __ia64_inb
#endif
#ifndef platform_inw
# define platform_inw __ia64_inw
#endif
#ifndef platform_inl
# define platform_inl __ia64_inl
#endif
#ifndef platform_outb
# define platform_outb __ia64_outb
#endif
#ifndef platform_outw
# define platform_outw __ia64_outw
#endif
#ifndef platform_outl
# define platform_outl __ia64_outl
#endif
#ifndef platform_mmiowb
# define platform_mmiowb __ia64_mmiowb
#endif
#ifndef platform_readb
# define platform_readb __ia64_readb
#endif
#ifndef platform_readw
# define platform_readw __ia64_readw
#endif
#ifndef platform_readl
# define platform_readl __ia64_readl
#endif
#ifndef platform_readq
# define platform_readq __ia64_readq
#endif
#ifndef platform_readb_relaxed
# define platform_readb_relaxed __ia64_readb_relaxed
#endif
#ifndef platform_readw_relaxed
# define platform_readw_relaxed __ia64_readw_relaxed
#endif
#ifndef platform_readl_relaxed
# define platform_readl_relaxed __ia64_readl_relaxed
#endif
#ifndef platform_readq_relaxed
# define platform_readq_relaxed __ia64_readq_relaxed
#endif
[IA64] hooks to wait for mmio writes to drain when migrating processes On SN2, MMIO writes which are issued from separate processors are not guaranteed to arrive in any particular order at the IO hardware. When performing such writes from the kernel this is not a problem, as a kernel thread will not migrate to another CPU during execution, and mmiowb() calls can guarantee write ordering when control of the IO resource is allowed to move between threads. However, when MMIO writes can be performed from user space (e.g. DRM) there are no such guarantees and mechanisms, as the process may context-switch at any time, and may migrate to a different CPU as part of the switch. For such programs/hardware to operate correctly, it is required that the MMIO writes from the old CPU be accepted by the IO hardware before subsequent writes from the new CPU can be issued. The following patch implements this behavior on SN2 by waiting for a Shub register to indicate that these writes have been accepted. This is placed in the context switch-in path, and only performs the wait when the newly scheduled task changes CPUs. Signed-off-by: Prarit Bhargava <prarit@sgi.com> Signed-off-by: Brent Casavant <bcasavan@sgi.com>
2006-01-27 07:55:52 +08:00
#ifndef platform_migrate
# define platform_migrate machvec_noop_task
#endif
[PATCH] msi: refactor and move the msi irq_chip into the arch code It turns out msi_ops was simply not enough to abstract the architecture specific details of msi. So I have moved the resposibility of constructing the struct irq_chip to the architectures, and have two architecture specific functions arch_setup_msi_irq, and arch_teardown_msi_irq. For simple architectures those functions can do all of the work. For architectures with platform dependencies they can call into the appropriate platform code. With this msi.c is finally free of assuming you have an apic, and this actually takes less code. The helpers for the architecture specific code are declared in the linux/msi.h to keep them separate from the msi functions used by drivers in linux/pci.h Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Tony Luck <tony.luck@intel.com> Cc: Andi Kleen <ak@suse.de> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Greg KH <greg@kroah.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-10-04 17:16:59 +08:00
#ifndef platform_setup_msi_irq
# define platform_setup_msi_irq ((ia64_mv_setup_msi_irq_t*)NULL)
#endif
#ifndef platform_teardown_msi_irq
# define platform_teardown_msi_irq ((ia64_mv_teardown_msi_irq_t*)NULL)
[PATCH] PCI: msi abstractions and support for altix Abstract portions of the MSI core for platforms that do not use standard APIC interrupt controllers. This is implemented through a new arch-specific msi setup routine, and a set of msi ops which can be set on a per platform basis. Signed-off-by: Mark Maule <maule@sgi.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2006-04-11 10:17:48 +08:00
#endif
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
#endif /* _ASM_IA64_MACHVEC_H */