2005-04-17 06:20:36 +08:00
|
|
|
/*
|
|
|
|
* Copyright (C) 1999-2003 Andre Hedrick <andre@linux-ide.org>
|
|
|
|
* Portions Copyright (C) 2001 Sun Microsystems, Inc.
|
|
|
|
* Portions Copyright (C) 2003 Red Hat Inc
|
2007-10-19 06:30:09 +08:00
|
|
|
* Portions Copyright (C) 2007 Bartlomiej Zolnierkiewicz
|
2009-04-18 23:42:19 +08:00
|
|
|
* Portions Copyright (C) 2005-2009 MontaVista Software, Inc.
|
2005-04-17 06:20:36 +08:00
|
|
|
*
|
|
|
|
* Thanks to HighPoint Technologies for their assistance, and hardware.
|
|
|
|
* Special Thanks to Jon Burchmore in SanDiego for the deep pockets, his
|
|
|
|
* donation of an ABit BP6 mainboard, processor, and memory acellerated
|
|
|
|
* development and support.
|
|
|
|
*
|
2005-06-28 06:24:27 +08:00
|
|
|
*
|
2006-12-13 16:35:47 +08:00
|
|
|
* HighPoint has its own drivers (open source except for the RAID part)
|
2010-10-18 17:03:14 +08:00
|
|
|
* available from http://www.highpoint-tech.com/USA_new/service_support.htm
|
2006-12-13 16:35:47 +08:00
|
|
|
* This may be useful to anyone wanting to work on this driver, however do not
|
|
|
|
* trust them too much since the code tends to become less and less meaningful
|
|
|
|
* as the time passes... :-/
|
2005-06-28 06:24:27 +08:00
|
|
|
*
|
2005-04-17 06:20:36 +08:00
|
|
|
* Note that final HPT370 support was done by force extraction of GPL.
|
|
|
|
*
|
|
|
|
* - add function for getting/setting power status of drive
|
|
|
|
* - the HPT370's state machine can get confused. reset it before each dma
|
|
|
|
* xfer to prevent that from happening.
|
|
|
|
* - reset state engine whenever we get an error.
|
|
|
|
* - check for busmaster state at end of dma.
|
|
|
|
* - use new highpoint timings.
|
|
|
|
* - detect bus speed using highpoint register.
|
|
|
|
* - use pll if we don't have a clock table. added a 66MHz table that's
|
|
|
|
* just 2x the 33MHz table.
|
|
|
|
* - removed turnaround. NOTE: we never want to switch between pll and
|
|
|
|
* pci clocks as the chip can glitch in those cases. the highpoint
|
|
|
|
* approved workaround slows everything down too much to be useful. in
|
|
|
|
* addition, we would have to serialize access to each chip.
|
|
|
|
* Adrian Sun <a.sun@sun.com>
|
|
|
|
*
|
|
|
|
* add drive timings for 66MHz PCI bus,
|
|
|
|
* fix ATA Cable signal detection, fix incorrect /proc info
|
|
|
|
* add /proc display for per-drive PIO/DMA/UDMA mode and
|
|
|
|
* per-channel ATA-33/66 Cable detect.
|
|
|
|
* Duncan Laurie <void@sun.com>
|
|
|
|
*
|
|
|
|
* fixup /proc output for multiple controllers
|
|
|
|
* Tim Hockin <thockin@sun.com>
|
|
|
|
*
|
|
|
|
* On hpt366:
|
|
|
|
* Reset the hpt366 on error, reset on dma
|
|
|
|
* Fix disabling Fast Interrupt hpt366.
|
|
|
|
* Mike Waychison <crlf@sun.com>
|
|
|
|
*
|
|
|
|
* Added support for 372N clocking and clock switching. The 372N needs
|
|
|
|
* different clocks on read/write. This requires overloading rw_disk and
|
|
|
|
* other deeply crazy things. Thanks to <http://www.hoerstreich.de> for
|
|
|
|
* keeping me sane.
|
2008-11-03 04:40:08 +08:00
|
|
|
* Alan Cox <alan@lxorguk.ukuu.org.uk>
|
2005-04-17 06:20:36 +08:00
|
|
|
*
|
2006-12-13 16:35:47 +08:00
|
|
|
* - fix the clock turnaround code: it was writing to the wrong ports when
|
|
|
|
* called for the secondary channel, caching the current clock mode per-
|
|
|
|
* channel caused the cached register value to get out of sync with the
|
|
|
|
* actual one, the channels weren't serialized, the turnaround shouldn't
|
|
|
|
* be done on 66 MHz PCI bus
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
* - disable UltraATA/100 for HPT370 by default as the 33 MHz clock being used
|
|
|
|
* does not allow for this speed anyway
|
|
|
|
* - avoid touching disabled channels (e.g. HPT371/N are single channel chips,
|
|
|
|
* their primary channel is kind of virtual, it isn't tied to any pins)
|
2006-12-13 16:35:49 +08:00
|
|
|
* - fix/remove bad/unused timing tables and use one set of tables for the whole
|
|
|
|
* HPT37x chip family; save space by introducing the separate transfer mode
|
|
|
|
* table in which the mode lookup is done
|
2006-12-13 16:35:52 +08:00
|
|
|
* - use f_CNT value saved by the HighPoint BIOS as reading it directly gives
|
2007-09-12 04:28:35 +08:00
|
|
|
* the wrong PCI frequency since DPLL has already been calibrated by BIOS;
|
|
|
|
* read it only from the function 0 of HPT374 chips
|
2006-12-13 16:35:50 +08:00
|
|
|
* - fix the hotswap code: it caused RESET- to glitch when tristating the bus,
|
|
|
|
* and for HPT36x the obsolete HDIO_TRISTATE_HWIF handler was called instead
|
2006-12-13 16:35:51 +08:00
|
|
|
* - pass to init_chipset() handlers a copy of the IDE PCI device structure as
|
|
|
|
* they tamper with its fields
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
* - pass to the init_setup handlers a copy of the ide_pci_device_t structure
|
|
|
|
* since they may tamper with its fields
|
2007-02-08 01:17:51 +08:00
|
|
|
* - prefix the driver startup messages with the real chip name
|
|
|
|
* - claim the extra 240 bytes of I/O space for all chips
|
2007-07-10 05:17:55 +08:00
|
|
|
* - optimize the UltraDMA filtering and the drive list lookup code
|
2007-02-08 01:17:54 +08:00
|
|
|
* - use pci_get_slot() to get to the function 1 of HPT36x/374
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
* - cache offset of the channel's misc. control registers (MCRs) being used
|
|
|
|
* throughout the driver
|
|
|
|
* - only touch the relevant MCR when detecting the cable type on HPT374's
|
|
|
|
* function 1
|
2007-02-08 01:18:05 +08:00
|
|
|
* - rename all the register related variables consistently
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
* - move all the interrupt twiddling code from the speedproc handlers into
|
|
|
|
* init_hwif_hpt366(), also grouping all the DMA related code together there
|
2008-01-26 05:17:05 +08:00
|
|
|
* - merge HPT36x/HPT37x speedproc handlers, fix PIO timing register mask and
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
* separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
|
|
|
|
* when setting an UltraDMA mode
|
|
|
|
* - fix hpt3xx_tune_drive() to set the PIO mode requested, not always select
|
|
|
|
* the best possible one
|
2007-02-08 01:18:13 +08:00
|
|
|
* - clean up DMA timeout handling for HPT370
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
* - switch to using the enumeration type to differ between the numerous chip
|
|
|
|
* variants, matching PCI device/revision ID with the chip type early, at the
|
|
|
|
* init_setup stage
|
|
|
|
* - extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
|
|
|
|
* stop duplicating it for each channel by storing the pointer in the pci_dev
|
|
|
|
* structure: first, at the init_setup stage, point it to a static "template"
|
|
|
|
* with only the chip type and its specific base DPLL frequency, the highest
|
2007-07-10 05:17:55 +08:00
|
|
|
* UltraDMA mode, and the chip settings table pointer filled, then, at the
|
|
|
|
* init_chipset stage, allocate per-chip instance and fill it with the rest
|
|
|
|
* of the necessary information
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
* - get rid of the constant thresholds in the HPT37x PCI clock detection code,
|
|
|
|
* switch to calculating PCI clock frequency based on the chip's base DPLL
|
|
|
|
* frequency
|
|
|
|
* - switch to using the DPLL clock and enable UltraATA/133 mode by default on
|
2007-06-08 21:14:32 +08:00
|
|
|
* anything newer than HPT370/A (except HPT374 that is not capable of this
|
|
|
|
* mode according to the manual)
|
2007-02-08 01:18:20 +08:00
|
|
|
* - fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
|
|
|
|
* also fixing the interchanged 25/40 MHz PCI clock cases for HPT36x chips;
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
* unify HPT36x/37x timing setup code and the speedproc handlers by joining
|
|
|
|
* the register setting lists into the table indexed by the clock selected
|
2007-07-10 05:17:55 +08:00
|
|
|
* - set the correct hwif->ultra_mask for each individual chip
|
2007-10-12 05:53:58 +08:00
|
|
|
* - add Ultra and MW DMA mode filtering for the HPT37[24] based SATA cards
|
2009-04-18 23:42:19 +08:00
|
|
|
* - stop resetting HPT370's state machine before each DMA transfer as that has
|
|
|
|
* caused more harm than good
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
* Sergei Shtylyov, <sshtylyov@ru.mvista.com> or <source@mvista.com>
|
2005-04-17 06:20:36 +08:00
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/types.h>
|
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/delay.h>
|
|
|
|
#include <linux/blkdev.h>
|
|
|
|
#include <linux/interrupt.h>
|
|
|
|
#include <linux/pci.h>
|
|
|
|
#include <linux/init.h>
|
|
|
|
#include <linux/ide.h>
|
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>
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2016-12-25 03:46:01 +08:00
|
|
|
#include <linux/uaccess.h>
|
2005-04-17 06:20:36 +08:00
|
|
|
#include <asm/io.h>
|
|
|
|
|
2008-07-25 04:53:32 +08:00
|
|
|
#define DRV_NAME "hpt366"
|
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
/* various tuning parameters */
|
2009-04-18 23:42:19 +08:00
|
|
|
#undef HPT_RESET_STATE_ENGINE
|
2006-12-13 16:35:47 +08:00
|
|
|
#undef HPT_DELAY_INTERRUPT
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
static const char *bad_ata100_5[] = {
|
|
|
|
"IBM-DTLA-307075",
|
|
|
|
"IBM-DTLA-307060",
|
|
|
|
"IBM-DTLA-307045",
|
|
|
|
"IBM-DTLA-307030",
|
|
|
|
"IBM-DTLA-307020",
|
|
|
|
"IBM-DTLA-307015",
|
|
|
|
"IBM-DTLA-305040",
|
|
|
|
"IBM-DTLA-305030",
|
|
|
|
"IBM-DTLA-305020",
|
|
|
|
"IC35L010AVER07-0",
|
|
|
|
"IC35L020AVER07-0",
|
|
|
|
"IC35L030AVER07-0",
|
|
|
|
"IC35L040AVER07-0",
|
|
|
|
"IC35L060AVER07-0",
|
|
|
|
"WDC AC310200R",
|
|
|
|
NULL
|
|
|
|
};
|
|
|
|
|
|
|
|
static const char *bad_ata66_4[] = {
|
|
|
|
"IBM-DTLA-307075",
|
|
|
|
"IBM-DTLA-307060",
|
|
|
|
"IBM-DTLA-307045",
|
|
|
|
"IBM-DTLA-307030",
|
|
|
|
"IBM-DTLA-307020",
|
|
|
|
"IBM-DTLA-307015",
|
|
|
|
"IBM-DTLA-305040",
|
|
|
|
"IBM-DTLA-305030",
|
|
|
|
"IBM-DTLA-305020",
|
|
|
|
"IC35L010AVER07-0",
|
|
|
|
"IC35L020AVER07-0",
|
|
|
|
"IC35L030AVER07-0",
|
|
|
|
"IC35L040AVER07-0",
|
|
|
|
"IC35L060AVER07-0",
|
|
|
|
"WDC AC310200R",
|
2007-07-04 04:28:35 +08:00
|
|
|
"MAXTOR STM3320620A",
|
2005-04-17 06:20:36 +08:00
|
|
|
NULL
|
|
|
|
};
|
|
|
|
|
|
|
|
static const char *bad_ata66_3[] = {
|
|
|
|
"WDC AC310200R",
|
|
|
|
NULL
|
|
|
|
};
|
|
|
|
|
|
|
|
static const char *bad_ata33[] = {
|
|
|
|
"Maxtor 92720U8", "Maxtor 92040U6", "Maxtor 91360U4", "Maxtor 91020U3", "Maxtor 90845U3", "Maxtor 90650U2",
|
|
|
|
"Maxtor 91360D8", "Maxtor 91190D7", "Maxtor 91020D6", "Maxtor 90845D5", "Maxtor 90680D4", "Maxtor 90510D3", "Maxtor 90340D2",
|
|
|
|
"Maxtor 91152D8", "Maxtor 91008D7", "Maxtor 90845D6", "Maxtor 90840D6", "Maxtor 90720D5", "Maxtor 90648D5", "Maxtor 90576D4",
|
|
|
|
"Maxtor 90510D4",
|
|
|
|
"Maxtor 90432D3", "Maxtor 90288D2", "Maxtor 90256D2",
|
|
|
|
"Maxtor 91000D8", "Maxtor 90910D8", "Maxtor 90875D7", "Maxtor 90840D7", "Maxtor 90750D6", "Maxtor 90625D5", "Maxtor 90500D4",
|
|
|
|
"Maxtor 91728D8", "Maxtor 91512D7", "Maxtor 91303D6", "Maxtor 91080D5", "Maxtor 90845D4", "Maxtor 90680D4", "Maxtor 90648D3", "Maxtor 90432D2",
|
|
|
|
NULL
|
|
|
|
};
|
|
|
|
|
2006-12-13 16:35:49 +08:00
|
|
|
static u8 xfer_speeds[] = {
|
|
|
|
XFER_UDMA_6,
|
|
|
|
XFER_UDMA_5,
|
|
|
|
XFER_UDMA_4,
|
|
|
|
XFER_UDMA_3,
|
|
|
|
XFER_UDMA_2,
|
|
|
|
XFER_UDMA_1,
|
|
|
|
XFER_UDMA_0,
|
|
|
|
|
|
|
|
XFER_MW_DMA_2,
|
|
|
|
XFER_MW_DMA_1,
|
|
|
|
XFER_MW_DMA_0,
|
|
|
|
|
|
|
|
XFER_PIO_4,
|
|
|
|
XFER_PIO_3,
|
|
|
|
XFER_PIO_2,
|
|
|
|
XFER_PIO_1,
|
|
|
|
XFER_PIO_0
|
2005-04-17 06:20:36 +08:00
|
|
|
};
|
|
|
|
|
2006-12-13 16:35:49 +08:00
|
|
|
/* Key for bus clock timings
|
|
|
|
* 36x 37x
|
|
|
|
* bits bits
|
|
|
|
* 0:3 0:3 data_high_time. Inactive time of DIOW_/DIOR_ for PIO and MW DMA.
|
|
|
|
* cycles = value + 1
|
|
|
|
* 4:7 4:8 data_low_time. Active time of DIOW_/DIOR_ for PIO and MW DMA.
|
|
|
|
* cycles = value + 1
|
|
|
|
* 8:11 9:12 cmd_high_time. Inactive time of DIOW_/DIOR_ during task file
|
|
|
|
* register access.
|
|
|
|
* 12:15 13:17 cmd_low_time. Active time of DIOW_/DIOR_ during task file
|
|
|
|
* register access.
|
|
|
|
* 16:18 18:20 udma_cycle_time. Clock cycles for UDMA xfer.
|
|
|
|
* - 21 CLK frequency: 0=ATA clock, 1=dual ATA clock.
|
|
|
|
* 19:21 22:24 pre_high_time. Time to initialize the 1st cycle for PIO and
|
|
|
|
* MW DMA xfer.
|
|
|
|
* 22:24 25:27 cmd_pre_high_time. Time to initialize the 1st PIO cycle for
|
|
|
|
* task file register access.
|
|
|
|
* 28 28 UDMA enable.
|
|
|
|
* 29 29 DMA enable.
|
|
|
|
* 30 30 PIO MST enable. If set, the chip is in bus master mode during
|
|
|
|
* PIO xfer.
|
|
|
|
* 31 31 FIFO enable.
|
2005-04-17 06:20:36 +08:00
|
|
|
*/
|
|
|
|
|
2006-12-13 16:35:49 +08:00
|
|
|
static u32 forty_base_hpt36x[] = {
|
|
|
|
/* XFER_UDMA_6 */ 0x900fd943,
|
|
|
|
/* XFER_UDMA_5 */ 0x900fd943,
|
|
|
|
/* XFER_UDMA_4 */ 0x900fd943,
|
|
|
|
/* XFER_UDMA_3 */ 0x900ad943,
|
|
|
|
/* XFER_UDMA_2 */ 0x900bd943,
|
|
|
|
/* XFER_UDMA_1 */ 0x9008d943,
|
|
|
|
/* XFER_UDMA_0 */ 0x9008d943,
|
|
|
|
|
|
|
|
/* XFER_MW_DMA_2 */ 0xa008d943,
|
|
|
|
/* XFER_MW_DMA_1 */ 0xa010d955,
|
|
|
|
/* XFER_MW_DMA_0 */ 0xa010d9fc,
|
|
|
|
|
|
|
|
/* XFER_PIO_4 */ 0xc008d963,
|
|
|
|
/* XFER_PIO_3 */ 0xc010d974,
|
|
|
|
/* XFER_PIO_2 */ 0xc010d997,
|
|
|
|
/* XFER_PIO_1 */ 0xc010d9c7,
|
|
|
|
/* XFER_PIO_0 */ 0xc018d9d9
|
2005-04-17 06:20:36 +08:00
|
|
|
};
|
|
|
|
|
2006-12-13 16:35:49 +08:00
|
|
|
static u32 thirty_three_base_hpt36x[] = {
|
|
|
|
/* XFER_UDMA_6 */ 0x90c9a731,
|
|
|
|
/* XFER_UDMA_5 */ 0x90c9a731,
|
|
|
|
/* XFER_UDMA_4 */ 0x90c9a731,
|
|
|
|
/* XFER_UDMA_3 */ 0x90cfa731,
|
|
|
|
/* XFER_UDMA_2 */ 0x90caa731,
|
|
|
|
/* XFER_UDMA_1 */ 0x90cba731,
|
|
|
|
/* XFER_UDMA_0 */ 0x90c8a731,
|
|
|
|
|
|
|
|
/* XFER_MW_DMA_2 */ 0xa0c8a731,
|
|
|
|
/* XFER_MW_DMA_1 */ 0xa0c8a732, /* 0xa0c8a733 */
|
|
|
|
/* XFER_MW_DMA_0 */ 0xa0c8a797,
|
|
|
|
|
|
|
|
/* XFER_PIO_4 */ 0xc0c8a731,
|
|
|
|
/* XFER_PIO_3 */ 0xc0c8a742,
|
|
|
|
/* XFER_PIO_2 */ 0xc0d0a753,
|
|
|
|
/* XFER_PIO_1 */ 0xc0d0a7a3, /* 0xc0d0a793 */
|
|
|
|
/* XFER_PIO_0 */ 0xc0d0a7aa /* 0xc0d0a7a7 */
|
2005-04-17 06:20:36 +08:00
|
|
|
};
|
|
|
|
|
2006-12-13 16:35:49 +08:00
|
|
|
static u32 twenty_five_base_hpt36x[] = {
|
|
|
|
/* XFER_UDMA_6 */ 0x90c98521,
|
|
|
|
/* XFER_UDMA_5 */ 0x90c98521,
|
|
|
|
/* XFER_UDMA_4 */ 0x90c98521,
|
|
|
|
/* XFER_UDMA_3 */ 0x90cf8521,
|
|
|
|
/* XFER_UDMA_2 */ 0x90cf8521,
|
|
|
|
/* XFER_UDMA_1 */ 0x90cb8521,
|
|
|
|
/* XFER_UDMA_0 */ 0x90cb8521,
|
|
|
|
|
|
|
|
/* XFER_MW_DMA_2 */ 0xa0ca8521,
|
|
|
|
/* XFER_MW_DMA_1 */ 0xa0ca8532,
|
|
|
|
/* XFER_MW_DMA_0 */ 0xa0ca8575,
|
|
|
|
|
|
|
|
/* XFER_PIO_4 */ 0xc0ca8521,
|
|
|
|
/* XFER_PIO_3 */ 0xc0ca8532,
|
|
|
|
/* XFER_PIO_2 */ 0xc0ca8542,
|
|
|
|
/* XFER_PIO_1 */ 0xc0d08572,
|
|
|
|
/* XFER_PIO_0 */ 0xc0d08585
|
2005-04-17 06:20:36 +08:00
|
|
|
};
|
|
|
|
|
2007-12-13 06:31:58 +08:00
|
|
|
/*
|
|
|
|
* The following are the new timing tables with PIO mode data/taskfile transfer
|
|
|
|
* overclocking fixed...
|
|
|
|
*/
|
|
|
|
|
|
|
|
/* This table is taken from the HPT370 data manual rev. 1.02 */
|
|
|
|
static u32 thirty_three_base_hpt37x[] = {
|
|
|
|
/* XFER_UDMA_6 */ 0x16455031, /* 0x16655031 ?? */
|
|
|
|
/* XFER_UDMA_5 */ 0x16455031,
|
|
|
|
/* XFER_UDMA_4 */ 0x16455031,
|
|
|
|
/* XFER_UDMA_3 */ 0x166d5031,
|
|
|
|
/* XFER_UDMA_2 */ 0x16495031,
|
|
|
|
/* XFER_UDMA_1 */ 0x164d5033,
|
|
|
|
/* XFER_UDMA_0 */ 0x16515097,
|
|
|
|
|
|
|
|
/* XFER_MW_DMA_2 */ 0x26515031,
|
|
|
|
/* XFER_MW_DMA_1 */ 0x26515033,
|
|
|
|
/* XFER_MW_DMA_0 */ 0x26515097,
|
|
|
|
|
|
|
|
/* XFER_PIO_4 */ 0x06515021,
|
|
|
|
/* XFER_PIO_3 */ 0x06515022,
|
|
|
|
/* XFER_PIO_2 */ 0x06515033,
|
|
|
|
/* XFER_PIO_1 */ 0x06915065,
|
|
|
|
/* XFER_PIO_0 */ 0x06d1508a
|
|
|
|
};
|
|
|
|
|
|
|
|
static u32 fifty_base_hpt37x[] = {
|
|
|
|
/* XFER_UDMA_6 */ 0x1a861842,
|
|
|
|
/* XFER_UDMA_5 */ 0x1a861842,
|
|
|
|
/* XFER_UDMA_4 */ 0x1aae1842,
|
|
|
|
/* XFER_UDMA_3 */ 0x1a8e1842,
|
|
|
|
/* XFER_UDMA_2 */ 0x1a0e1842,
|
|
|
|
/* XFER_UDMA_1 */ 0x1a161854,
|
|
|
|
/* XFER_UDMA_0 */ 0x1a1a18ea,
|
|
|
|
|
|
|
|
/* XFER_MW_DMA_2 */ 0x2a821842,
|
|
|
|
/* XFER_MW_DMA_1 */ 0x2a821854,
|
|
|
|
/* XFER_MW_DMA_0 */ 0x2a8218ea,
|
|
|
|
|
|
|
|
/* XFER_PIO_4 */ 0x0a821842,
|
|
|
|
/* XFER_PIO_3 */ 0x0a821843,
|
|
|
|
/* XFER_PIO_2 */ 0x0a821855,
|
|
|
|
/* XFER_PIO_1 */ 0x0ac218a8,
|
|
|
|
/* XFER_PIO_0 */ 0x0b02190c
|
|
|
|
};
|
|
|
|
|
|
|
|
static u32 sixty_six_base_hpt37x[] = {
|
|
|
|
/* XFER_UDMA_6 */ 0x1c86fe62,
|
|
|
|
/* XFER_UDMA_5 */ 0x1caefe62, /* 0x1c8afe62 */
|
|
|
|
/* XFER_UDMA_4 */ 0x1c8afe62,
|
|
|
|
/* XFER_UDMA_3 */ 0x1c8efe62,
|
|
|
|
/* XFER_UDMA_2 */ 0x1c92fe62,
|
|
|
|
/* XFER_UDMA_1 */ 0x1c9afe62,
|
|
|
|
/* XFER_UDMA_0 */ 0x1c82fe62,
|
|
|
|
|
|
|
|
/* XFER_MW_DMA_2 */ 0x2c82fe62,
|
|
|
|
/* XFER_MW_DMA_1 */ 0x2c82fe66,
|
|
|
|
/* XFER_MW_DMA_0 */ 0x2c82ff2e,
|
|
|
|
|
|
|
|
/* XFER_PIO_4 */ 0x0c82fe62,
|
|
|
|
/* XFER_PIO_3 */ 0x0c82fe84,
|
|
|
|
/* XFER_PIO_2 */ 0x0c82fea6,
|
|
|
|
/* XFER_PIO_1 */ 0x0d02ff26,
|
|
|
|
/* XFER_PIO_0 */ 0x0d42ff7f
|
|
|
|
};
|
2005-04-17 06:20:36 +08:00
|
|
|
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
#define HPT371_ALLOW_ATA133_6 1
|
|
|
|
#define HPT302_ALLOW_ATA133_6 1
|
|
|
|
#define HPT372_ALLOW_ATA133_6 1
|
2007-02-08 01:17:37 +08:00
|
|
|
#define HPT370_ALLOW_ATA100_5 0
|
2005-04-17 06:20:36 +08:00
|
|
|
#define HPT366_ALLOW_ATA66_4 1
|
|
|
|
#define HPT366_ALLOW_ATA66_3 1
|
|
|
|
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
/* Supported ATA clock frequencies */
|
|
|
|
enum ata_clock {
|
|
|
|
ATA_CLOCK_25MHZ,
|
|
|
|
ATA_CLOCK_33MHZ,
|
|
|
|
ATA_CLOCK_40MHZ,
|
|
|
|
ATA_CLOCK_50MHZ,
|
|
|
|
ATA_CLOCK_66MHZ,
|
|
|
|
NUM_ATA_CLOCKS
|
|
|
|
};
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2008-01-26 05:17:05 +08:00
|
|
|
struct hpt_timings {
|
|
|
|
u32 pio_mask;
|
|
|
|
u32 dma_mask;
|
|
|
|
u32 ultra_mask;
|
|
|
|
u32 *clock_table[NUM_ATA_CLOCKS];
|
|
|
|
};
|
|
|
|
|
2005-06-28 06:24:27 +08:00
|
|
|
/*
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
* Hold all the HighPoint chip information in one place.
|
2005-06-28 06:24:27 +08:00
|
|
|
*/
|
2005-04-17 06:20:36 +08:00
|
|
|
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
struct hpt_info {
|
2007-10-19 06:30:09 +08:00
|
|
|
char *chip_name; /* Chip name */
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
u8 chip_type; /* Chip type */
|
2007-10-19 06:30:09 +08:00
|
|
|
u8 udma_mask; /* Allowed UltraDMA modes mask. */
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
u8 dpll_clk; /* DPLL clock in MHz */
|
|
|
|
u8 pci_clk; /* PCI clock in MHz */
|
2008-01-26 05:17:05 +08:00
|
|
|
struct hpt_timings *timings; /* Chipset timing data */
|
|
|
|
u8 clock; /* ATA clock selected */
|
2005-06-28 06:24:27 +08:00
|
|
|
};
|
|
|
|
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
/* Supported HighPoint chips */
|
|
|
|
enum {
|
|
|
|
HPT36x,
|
|
|
|
HPT370,
|
|
|
|
HPT370A,
|
|
|
|
HPT374,
|
|
|
|
HPT372,
|
|
|
|
HPT372A,
|
|
|
|
HPT302,
|
|
|
|
HPT371,
|
|
|
|
HPT372N,
|
|
|
|
HPT302N,
|
|
|
|
HPT371N
|
|
|
|
};
|
2005-06-28 06:24:27 +08:00
|
|
|
|
2008-01-26 05:17:05 +08:00
|
|
|
static struct hpt_timings hpt36x_timings = {
|
|
|
|
.pio_mask = 0xc1f8ffff,
|
|
|
|
.dma_mask = 0x303800ff,
|
|
|
|
.ultra_mask = 0x30070000,
|
|
|
|
.clock_table = {
|
|
|
|
[ATA_CLOCK_25MHZ] = twenty_five_base_hpt36x,
|
|
|
|
[ATA_CLOCK_33MHZ] = thirty_three_base_hpt36x,
|
|
|
|
[ATA_CLOCK_40MHZ] = forty_base_hpt36x,
|
|
|
|
[ATA_CLOCK_50MHZ] = NULL,
|
|
|
|
[ATA_CLOCK_66MHZ] = NULL
|
|
|
|
}
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
};
|
2007-02-08 01:17:37 +08:00
|
|
|
|
2008-01-26 05:17:05 +08:00
|
|
|
static struct hpt_timings hpt37x_timings = {
|
|
|
|
.pio_mask = 0xcfc3ffff,
|
|
|
|
.dma_mask = 0x31c001ff,
|
|
|
|
.ultra_mask = 0x303c0000,
|
|
|
|
.clock_table = {
|
|
|
|
[ATA_CLOCK_25MHZ] = NULL,
|
|
|
|
[ATA_CLOCK_33MHZ] = thirty_three_base_hpt37x,
|
|
|
|
[ATA_CLOCK_40MHZ] = NULL,
|
|
|
|
[ATA_CLOCK_50MHZ] = fifty_base_hpt37x,
|
|
|
|
[ATA_CLOCK_66MHZ] = sixty_six_base_hpt37x
|
|
|
|
}
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
};
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static const struct hpt_info hpt36x = {
|
2007-10-19 06:30:09 +08:00
|
|
|
.chip_name = "HPT36x",
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.chip_type = HPT36x,
|
2007-10-19 06:30:09 +08:00
|
|
|
.udma_mask = HPT366_ALLOW_ATA66_3 ? (HPT366_ALLOW_ATA66_4 ? ATA_UDMA4 : ATA_UDMA3) : ATA_UDMA2,
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.dpll_clk = 0, /* no DPLL */
|
2008-01-26 05:17:05 +08:00
|
|
|
.timings = &hpt36x_timings
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
};
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static const struct hpt_info hpt370 = {
|
2007-10-19 06:30:09 +08:00
|
|
|
.chip_name = "HPT370",
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.chip_type = HPT370,
|
2007-10-19 06:30:09 +08:00
|
|
|
.udma_mask = HPT370_ALLOW_ATA100_5 ? ATA_UDMA5 : ATA_UDMA4,
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.dpll_clk = 48,
|
2008-01-26 05:17:05 +08:00
|
|
|
.timings = &hpt37x_timings
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
};
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static const struct hpt_info hpt370a = {
|
2007-10-19 06:30:09 +08:00
|
|
|
.chip_name = "HPT370A",
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.chip_type = HPT370A,
|
2007-10-19 06:30:09 +08:00
|
|
|
.udma_mask = HPT370_ALLOW_ATA100_5 ? ATA_UDMA5 : ATA_UDMA4,
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.dpll_clk = 48,
|
2008-01-26 05:17:05 +08:00
|
|
|
.timings = &hpt37x_timings
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
};
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static const struct hpt_info hpt374 = {
|
2007-10-19 06:30:09 +08:00
|
|
|
.chip_name = "HPT374",
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.chip_type = HPT374,
|
2007-10-19 06:30:09 +08:00
|
|
|
.udma_mask = ATA_UDMA5,
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.dpll_clk = 48,
|
2008-01-26 05:17:05 +08:00
|
|
|
.timings = &hpt37x_timings
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
};
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static const struct hpt_info hpt372 = {
|
2007-10-19 06:30:09 +08:00
|
|
|
.chip_name = "HPT372",
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.chip_type = HPT372,
|
2007-10-19 06:30:09 +08:00
|
|
|
.udma_mask = HPT372_ALLOW_ATA133_6 ? ATA_UDMA6 : ATA_UDMA5,
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.dpll_clk = 55,
|
2008-01-26 05:17:05 +08:00
|
|
|
.timings = &hpt37x_timings
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
};
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static const struct hpt_info hpt372a = {
|
2007-10-19 06:30:09 +08:00
|
|
|
.chip_name = "HPT372A",
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.chip_type = HPT372A,
|
2007-10-19 06:30:09 +08:00
|
|
|
.udma_mask = HPT372_ALLOW_ATA133_6 ? ATA_UDMA6 : ATA_UDMA5,
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.dpll_clk = 66,
|
2008-01-26 05:17:05 +08:00
|
|
|
.timings = &hpt37x_timings
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
};
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static const struct hpt_info hpt302 = {
|
2007-10-19 06:30:09 +08:00
|
|
|
.chip_name = "HPT302",
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.chip_type = HPT302,
|
2007-10-19 06:30:09 +08:00
|
|
|
.udma_mask = HPT302_ALLOW_ATA133_6 ? ATA_UDMA6 : ATA_UDMA5,
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.dpll_clk = 66,
|
2008-01-26 05:17:05 +08:00
|
|
|
.timings = &hpt37x_timings
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
};
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static const struct hpt_info hpt371 = {
|
2007-10-19 06:30:09 +08:00
|
|
|
.chip_name = "HPT371",
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.chip_type = HPT371,
|
2007-10-19 06:30:09 +08:00
|
|
|
.udma_mask = HPT371_ALLOW_ATA133_6 ? ATA_UDMA6 : ATA_UDMA5,
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.dpll_clk = 66,
|
2008-01-26 05:17:05 +08:00
|
|
|
.timings = &hpt37x_timings
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
};
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static const struct hpt_info hpt372n = {
|
2007-10-19 06:30:09 +08:00
|
|
|
.chip_name = "HPT372N",
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.chip_type = HPT372N,
|
2007-10-19 06:30:09 +08:00
|
|
|
.udma_mask = HPT372_ALLOW_ATA133_6 ? ATA_UDMA6 : ATA_UDMA5,
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.dpll_clk = 77,
|
2008-01-26 05:17:05 +08:00
|
|
|
.timings = &hpt37x_timings
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
};
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static const struct hpt_info hpt302n = {
|
2007-10-19 06:30:09 +08:00
|
|
|
.chip_name = "HPT302N",
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.chip_type = HPT302N,
|
2007-10-19 06:30:09 +08:00
|
|
|
.udma_mask = HPT302_ALLOW_ATA133_6 ? ATA_UDMA6 : ATA_UDMA5,
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.dpll_clk = 77,
|
2008-01-26 05:17:05 +08:00
|
|
|
.timings = &hpt37x_timings
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
};
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static const struct hpt_info hpt371n = {
|
2007-10-19 06:30:09 +08:00
|
|
|
.chip_name = "HPT371N",
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.chip_type = HPT371N,
|
2007-10-19 06:30:09 +08:00
|
|
|
.udma_mask = HPT371_ALLOW_ATA133_6 ? ATA_UDMA6 : ATA_UDMA5,
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.dpll_clk = 77,
|
2008-01-26 05:17:05 +08:00
|
|
|
.timings = &hpt37x_timings
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
};
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2016-03-18 05:22:35 +08:00
|
|
|
static bool check_in_drive_list(ide_drive_t *drive, const char **list)
|
2007-02-08 01:17:37 +08:00
|
|
|
{
|
2016-03-18 05:22:35 +08:00
|
|
|
return match_string(list, -1, (char *)&drive->id[ATA_ID_PROD]) >= 0;
|
2007-02-08 01:17:37 +08:00
|
|
|
}
|
2005-04-17 06:20:36 +08:00
|
|
|
|
ide: Fix pointer arithmetic in hpt3xx driver code (3rd try)
git commit 74811f355f4f69a187fa74892dcf2a684b84ce99 causes crash at
module load (or boot) time on my machine with a hpt374 controller.
The reason for this is that for initializing second controller which sets
(hwif->dev == host->dev[1]) to true (1), adds 1 to a void ptr, which
advances it by one byte instead of advancing it by sizeof(hpt_info) bytes.
Because of this, all initialization functions get corrupted data in info
variable which causes a crash at boot time.
This patch fixes that and makes my machine boot again.
The card itself is a HPT374 raid conroller: Here is the lspci -v output:
03:06.0 RAID bus controller: HighPoint Technologies, Inc. HPT374 (rev
07)
Subsystem: HighPoint Technologies, Inc. Unknown device 0001
Flags: bus master, 66MHz, medium devsel, latency 120, IRQ 28
I/O ports at 8000 [size=8]
I/O ports at 7800 [size=4]
I/O ports at 7400 [size=8]
I/O ports at 7000 [size=4]
I/O ports at 6800 [size=256]
Expansion ROM at fe8e0000 [disabled] [size=128K]
Capabilities: [60] Power Management version 2
03:06.1 RAID bus controller: HighPoint Technologies, Inc. HPT374 (rev
07)
Subsystem: HighPoint Technologies, Inc. Unknown device 0001
Flags: bus master, 66MHz, medium devsel, latency 120, IRQ 28
I/O ports at 9800 [size=8]
I/O ports at 9400 [size=4]
I/O ports at 9000 [size=8]
I/O ports at 8800 [size=4]
I/O ports at 8400 [size=256]
Capabilities: [60] Power Management version 2
Signed-off-by: Masoud Sharbiani <masouds@google.com>
Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
[bart: use dev_get_drvdata() per Sergei's suggestion]
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2008-09-11 04:22:34 +08:00
|
|
|
static struct hpt_info *hpt3xx_get_info(struct device *dev)
|
|
|
|
{
|
|
|
|
struct ide_host *host = dev_get_drvdata(dev);
|
|
|
|
struct hpt_info *info = (struct hpt_info *)host->host_priv;
|
|
|
|
|
|
|
|
return dev == host->dev[1] ? info + 1 : info;
|
|
|
|
}
|
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
/*
|
2007-09-12 04:28:36 +08:00
|
|
|
* The Marvell bridge chips used on the HighPoint SATA cards do not seem
|
|
|
|
* to support the UltraDMA modes 1, 2, and 3 as well as any MWDMA modes...
|
2005-04-17 06:20:36 +08:00
|
|
|
*/
|
2007-05-10 06:01:08 +08:00
|
|
|
|
|
|
|
static u8 hpt3xx_udma_filter(ide_drive_t *drive)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2009-01-07 00:20:52 +08:00
|
|
|
ide_hwif_t *hwif = drive->hwif;
|
ide: Fix pointer arithmetic in hpt3xx driver code (3rd try)
git commit 74811f355f4f69a187fa74892dcf2a684b84ce99 causes crash at
module load (or boot) time on my machine with a hpt374 controller.
The reason for this is that for initializing second controller which sets
(hwif->dev == host->dev[1]) to true (1), adds 1 to a void ptr, which
advances it by one byte instead of advancing it by sizeof(hpt_info) bytes.
Because of this, all initialization functions get corrupted data in info
variable which causes a crash at boot time.
This patch fixes that and makes my machine boot again.
The card itself is a HPT374 raid conroller: Here is the lspci -v output:
03:06.0 RAID bus controller: HighPoint Technologies, Inc. HPT374 (rev
07)
Subsystem: HighPoint Technologies, Inc. Unknown device 0001
Flags: bus master, 66MHz, medium devsel, latency 120, IRQ 28
I/O ports at 8000 [size=8]
I/O ports at 7800 [size=4]
I/O ports at 7400 [size=8]
I/O ports at 7000 [size=4]
I/O ports at 6800 [size=256]
Expansion ROM at fe8e0000 [disabled] [size=128K]
Capabilities: [60] Power Management version 2
03:06.1 RAID bus controller: HighPoint Technologies, Inc. HPT374 (rev
07)
Subsystem: HighPoint Technologies, Inc. Unknown device 0001
Flags: bus master, 66MHz, medium devsel, latency 120, IRQ 28
I/O ports at 9800 [size=8]
I/O ports at 9400 [size=4]
I/O ports at 9000 [size=8]
I/O ports at 8800 [size=4]
I/O ports at 8400 [size=256]
Capabilities: [60] Power Management version 2
Signed-off-by: Masoud Sharbiani <masouds@google.com>
Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
[bart: use dev_get_drvdata() per Sergei's suggestion]
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2008-09-11 04:22:34 +08:00
|
|
|
struct hpt_info *info = hpt3xx_get_info(hwif->dev);
|
2007-09-12 04:28:36 +08:00
|
|
|
u8 mask = hwif->ultra_mask;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2007-07-10 05:17:55 +08:00
|
|
|
switch (info->chip_type) {
|
|
|
|
case HPT36x:
|
|
|
|
if (!HPT366_ALLOW_ATA66_4 ||
|
|
|
|
check_in_drive_list(drive, bad_ata66_4))
|
2007-09-12 04:28:36 +08:00
|
|
|
mask = ATA_UDMA3;
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
|
2007-07-10 05:17:55 +08:00
|
|
|
if (!HPT366_ALLOW_ATA66_3 ||
|
|
|
|
check_in_drive_list(drive, bad_ata66_3))
|
2007-09-12 04:28:36 +08:00
|
|
|
mask = ATA_UDMA2;
|
2007-07-10 05:17:55 +08:00
|
|
|
break;
|
2007-09-12 04:28:36 +08:00
|
|
|
case HPT370:
|
|
|
|
if (!HPT370_ALLOW_ATA100_5 ||
|
|
|
|
check_in_drive_list(drive, bad_ata100_5))
|
|
|
|
mask = ATA_UDMA4;
|
|
|
|
break;
|
|
|
|
case HPT370A:
|
|
|
|
if (!HPT370_ALLOW_ATA100_5 ||
|
|
|
|
check_in_drive_list(drive, bad_ata100_5))
|
|
|
|
return ATA_UDMA4;
|
|
|
|
case HPT372 :
|
|
|
|
case HPT372A:
|
|
|
|
case HPT372N:
|
|
|
|
case HPT374 :
|
2008-10-11 04:39:30 +08:00
|
|
|
if (ata_id_is_sata(drive->id))
|
2007-09-12 04:28:36 +08:00
|
|
|
mask &= ~0x0e;
|
|
|
|
/* Fall thru */
|
2007-07-10 05:17:55 +08:00
|
|
|
default:
|
2007-09-12 04:28:36 +08:00
|
|
|
return mask;
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
2007-07-10 05:17:55 +08:00
|
|
|
|
|
|
|
return check_in_drive_list(drive, bad_ata33) ? 0x00 : mask;
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2007-10-12 05:53:58 +08:00
|
|
|
static u8 hpt3xx_mdma_filter(ide_drive_t *drive)
|
|
|
|
{
|
2009-01-07 00:20:52 +08:00
|
|
|
ide_hwif_t *hwif = drive->hwif;
|
ide: Fix pointer arithmetic in hpt3xx driver code (3rd try)
git commit 74811f355f4f69a187fa74892dcf2a684b84ce99 causes crash at
module load (or boot) time on my machine with a hpt374 controller.
The reason for this is that for initializing second controller which sets
(hwif->dev == host->dev[1]) to true (1), adds 1 to a void ptr, which
advances it by one byte instead of advancing it by sizeof(hpt_info) bytes.
Because of this, all initialization functions get corrupted data in info
variable which causes a crash at boot time.
This patch fixes that and makes my machine boot again.
The card itself is a HPT374 raid conroller: Here is the lspci -v output:
03:06.0 RAID bus controller: HighPoint Technologies, Inc. HPT374 (rev
07)
Subsystem: HighPoint Technologies, Inc. Unknown device 0001
Flags: bus master, 66MHz, medium devsel, latency 120, IRQ 28
I/O ports at 8000 [size=8]
I/O ports at 7800 [size=4]
I/O ports at 7400 [size=8]
I/O ports at 7000 [size=4]
I/O ports at 6800 [size=256]
Expansion ROM at fe8e0000 [disabled] [size=128K]
Capabilities: [60] Power Management version 2
03:06.1 RAID bus controller: HighPoint Technologies, Inc. HPT374 (rev
07)
Subsystem: HighPoint Technologies, Inc. Unknown device 0001
Flags: bus master, 66MHz, medium devsel, latency 120, IRQ 28
I/O ports at 9800 [size=8]
I/O ports at 9400 [size=4]
I/O ports at 9000 [size=8]
I/O ports at 8800 [size=4]
I/O ports at 8400 [size=256]
Capabilities: [60] Power Management version 2
Signed-off-by: Masoud Sharbiani <masouds@google.com>
Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
[bart: use dev_get_drvdata() per Sergei's suggestion]
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2008-09-11 04:22:34 +08:00
|
|
|
struct hpt_info *info = hpt3xx_get_info(hwif->dev);
|
2007-10-12 05:53:58 +08:00
|
|
|
|
|
|
|
switch (info->chip_type) {
|
|
|
|
case HPT372 :
|
|
|
|
case HPT372A:
|
|
|
|
case HPT372N:
|
|
|
|
case HPT374 :
|
2008-10-11 04:39:30 +08:00
|
|
|
if (ata_id_is_sata(drive->id))
|
2007-10-12 05:53:58 +08:00
|
|
|
return 0x00;
|
|
|
|
/* Fall thru */
|
|
|
|
default:
|
|
|
|
return 0x07;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
static u32 get_speed_setting(u8 speed, struct hpt_info *info)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2006-12-13 16:35:49 +08:00
|
|
|
int i;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Lookup the transfer mode table to get the index into
|
|
|
|
* the timing table.
|
|
|
|
*
|
|
|
|
* NOTE: For XFER_PIO_SLOW, PIO mode 0 timings will be used.
|
|
|
|
*/
|
|
|
|
for (i = 0; i < ARRAY_SIZE(xfer_speeds) - 1; i++)
|
|
|
|
if (xfer_speeds[i] == speed)
|
|
|
|
break;
|
2008-01-26 05:17:05 +08:00
|
|
|
|
|
|
|
return info->timings->clock_table[info->clock][i];
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2010-01-19 17:45:29 +08:00
|
|
|
static void hpt3xx_set_mode(ide_hwif_t *hwif, ide_drive_t *drive)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2008-07-25 04:53:15 +08:00
|
|
|
struct pci_dev *dev = to_pci_dev(hwif->dev);
|
ide: Fix pointer arithmetic in hpt3xx driver code (3rd try)
git commit 74811f355f4f69a187fa74892dcf2a684b84ce99 causes crash at
module load (or boot) time on my machine with a hpt374 controller.
The reason for this is that for initializing second controller which sets
(hwif->dev == host->dev[1]) to true (1), adds 1 to a void ptr, which
advances it by one byte instead of advancing it by sizeof(hpt_info) bytes.
Because of this, all initialization functions get corrupted data in info
variable which causes a crash at boot time.
This patch fixes that and makes my machine boot again.
The card itself is a HPT374 raid conroller: Here is the lspci -v output:
03:06.0 RAID bus controller: HighPoint Technologies, Inc. HPT374 (rev
07)
Subsystem: HighPoint Technologies, Inc. Unknown device 0001
Flags: bus master, 66MHz, medium devsel, latency 120, IRQ 28
I/O ports at 8000 [size=8]
I/O ports at 7800 [size=4]
I/O ports at 7400 [size=8]
I/O ports at 7000 [size=4]
I/O ports at 6800 [size=256]
Expansion ROM at fe8e0000 [disabled] [size=128K]
Capabilities: [60] Power Management version 2
03:06.1 RAID bus controller: HighPoint Technologies, Inc. HPT374 (rev
07)
Subsystem: HighPoint Technologies, Inc. Unknown device 0001
Flags: bus master, 66MHz, medium devsel, latency 120, IRQ 28
I/O ports at 9800 [size=8]
I/O ports at 9400 [size=4]
I/O ports at 9000 [size=8]
I/O ports at 8800 [size=4]
I/O ports at 8400 [size=256]
Capabilities: [60] Power Management version 2
Signed-off-by: Masoud Sharbiani <masouds@google.com>
Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
[bart: use dev_get_drvdata() per Sergei's suggestion]
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2008-09-11 04:22:34 +08:00
|
|
|
struct hpt_info *info = hpt3xx_get_info(hwif->dev);
|
2008-01-26 05:17:05 +08:00
|
|
|
struct hpt_timings *t = info->timings;
|
|
|
|
u8 itr_addr = 0x40 + (drive->dn * 4);
|
2007-02-08 01:18:11 +08:00
|
|
|
u32 old_itr = 0;
|
2010-01-19 17:45:29 +08:00
|
|
|
const u8 speed = drive->dma_mode;
|
2008-01-26 05:17:04 +08:00
|
|
|
u32 new_itr = get_speed_setting(speed, info);
|
2008-01-26 05:17:05 +08:00
|
|
|
u32 itr_mask = speed < XFER_MW_DMA_0 ? t->pio_mask :
|
|
|
|
(speed < XFER_UDMA_0 ? t->dma_mask :
|
|
|
|
t->ultra_mask);
|
2005-06-28 06:24:27 +08:00
|
|
|
|
2008-01-26 05:17:04 +08:00
|
|
|
pci_read_config_dword(dev, itr_addr, &old_itr);
|
|
|
|
new_itr = (old_itr & ~itr_mask) | (new_itr & itr_mask);
|
2005-04-17 06:20:36 +08:00
|
|
|
/*
|
2007-02-08 01:18:05 +08:00
|
|
|
* Disable on-chip PIO FIFO/buffer (and PIO MST mode as well)
|
|
|
|
* to avoid problems handling I/O errors later
|
2005-04-17 06:20:36 +08:00
|
|
|
*/
|
2007-02-08 01:18:05 +08:00
|
|
|
new_itr &= ~0xc0000000;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2007-02-08 01:18:05 +08:00
|
|
|
pci_write_config_dword(dev, itr_addr, new_itr);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2010-01-19 17:44:41 +08:00
|
|
|
static void hpt3xx_set_pio_mode(ide_hwif_t *hwif, ide_drive_t *drive)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2010-01-19 17:45:29 +08:00
|
|
|
drive->dma_mode = drive->pio_mode;
|
|
|
|
hpt3xx_set_mode(hwif, drive);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2007-02-08 01:18:11 +08:00
|
|
|
static void hpt3xx_maskproc(ide_drive_t *drive, int mask)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2009-01-07 00:20:52 +08:00
|
|
|
ide_hwif_t *hwif = drive->hwif;
|
2008-02-02 06:09:31 +08:00
|
|
|
struct pci_dev *dev = to_pci_dev(hwif->dev);
|
ide: Fix pointer arithmetic in hpt3xx driver code (3rd try)
git commit 74811f355f4f69a187fa74892dcf2a684b84ce99 causes crash at
module load (or boot) time on my machine with a hpt374 controller.
The reason for this is that for initializing second controller which sets
(hwif->dev == host->dev[1]) to true (1), adds 1 to a void ptr, which
advances it by one byte instead of advancing it by sizeof(hpt_info) bytes.
Because of this, all initialization functions get corrupted data in info
variable which causes a crash at boot time.
This patch fixes that and makes my machine boot again.
The card itself is a HPT374 raid conroller: Here is the lspci -v output:
03:06.0 RAID bus controller: HighPoint Technologies, Inc. HPT374 (rev
07)
Subsystem: HighPoint Technologies, Inc. Unknown device 0001
Flags: bus master, 66MHz, medium devsel, latency 120, IRQ 28
I/O ports at 8000 [size=8]
I/O ports at 7800 [size=4]
I/O ports at 7400 [size=8]
I/O ports at 7000 [size=4]
I/O ports at 6800 [size=256]
Expansion ROM at fe8e0000 [disabled] [size=128K]
Capabilities: [60] Power Management version 2
03:06.1 RAID bus controller: HighPoint Technologies, Inc. HPT374 (rev
07)
Subsystem: HighPoint Technologies, Inc. Unknown device 0001
Flags: bus master, 66MHz, medium devsel, latency 120, IRQ 28
I/O ports at 9800 [size=8]
I/O ports at 9400 [size=4]
I/O ports at 9000 [size=8]
I/O ports at 8800 [size=4]
I/O ports at 8400 [size=256]
Capabilities: [60] Power Management version 2
Signed-off-by: Masoud Sharbiani <masouds@google.com>
Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
[bart: use dev_get_drvdata() per Sergei's suggestion]
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2008-09-11 04:22:34 +08:00
|
|
|
struct hpt_info *info = hpt3xx_get_info(hwif->dev);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2009-06-07 21:37:10 +08:00
|
|
|
if ((drive->dev_flags & IDE_DFLAG_NIEN_QUIRK) == 0)
|
2008-10-18 00:09:15 +08:00
|
|
|
return;
|
|
|
|
|
|
|
|
if (info->chip_type >= HPT370) {
|
|
|
|
u8 scr1 = 0;
|
|
|
|
|
|
|
|
pci_read_config_byte(dev, 0x5a, &scr1);
|
|
|
|
if (((scr1 & 0x10) >> 4) != mask) {
|
2007-02-08 01:18:05 +08:00
|
|
|
if (mask)
|
2008-10-18 00:09:15 +08:00
|
|
|
scr1 |= 0x10;
|
2007-02-08 01:18:05 +08:00
|
|
|
else
|
2008-10-18 00:09:15 +08:00
|
|
|
scr1 &= ~0x10;
|
|
|
|
pci_write_config_byte(dev, 0x5a, scr1);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
2008-10-18 00:09:15 +08:00
|
|
|
} else if (mask)
|
|
|
|
disable_irq(hwif->irq);
|
|
|
|
else
|
|
|
|
enable_irq(hwif->irq);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2007-02-08 01:18:05 +08:00
|
|
|
* This is specific to the HPT366 UDMA chipset
|
2005-04-17 06:20:36 +08:00
|
|
|
* by HighPoint|Triones Technologies, Inc.
|
|
|
|
*/
|
2007-07-10 05:17:54 +08:00
|
|
|
static void hpt366_dma_lost_irq(ide_drive_t *drive)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2008-02-02 06:09:31 +08:00
|
|
|
struct pci_dev *dev = to_pci_dev(drive->hwif->dev);
|
2007-02-08 01:18:05 +08:00
|
|
|
u8 mcr1 = 0, mcr3 = 0, scr1 = 0;
|
|
|
|
|
|
|
|
pci_read_config_byte(dev, 0x50, &mcr1);
|
|
|
|
pci_read_config_byte(dev, 0x52, &mcr3);
|
|
|
|
pci_read_config_byte(dev, 0x5a, &scr1);
|
|
|
|
printk("%s: (%s) mcr1=0x%02x, mcr3=0x%02x, scr1=0x%02x\n",
|
2008-04-27 04:25:20 +08:00
|
|
|
drive->name, __func__, mcr1, mcr3, scr1);
|
2007-02-08 01:18:05 +08:00
|
|
|
if (scr1 & 0x10)
|
|
|
|
pci_write_config_byte(dev, 0x5a, scr1 & ~0x10);
|
2007-07-10 05:17:54 +08:00
|
|
|
ide_dma_lost_irq(drive);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2007-02-08 01:18:13 +08:00
|
|
|
static void hpt370_clear_engine(ide_drive_t *drive)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2009-01-07 00:20:52 +08:00
|
|
|
ide_hwif_t *hwif = drive->hwif;
|
2008-02-02 06:09:31 +08:00
|
|
|
struct pci_dev *dev = to_pci_dev(hwif->dev);
|
2007-02-08 01:18:05 +08:00
|
|
|
|
2008-02-02 06:09:31 +08:00
|
|
|
pci_write_config_byte(dev, hwif->select_data, 0x37);
|
2005-04-17 06:20:36 +08:00
|
|
|
udelay(10);
|
|
|
|
}
|
|
|
|
|
2007-02-08 01:18:13 +08:00
|
|
|
static void hpt370_irq_timeout(ide_drive_t *drive)
|
|
|
|
{
|
2009-01-07 00:20:52 +08:00
|
|
|
ide_hwif_t *hwif = drive->hwif;
|
2008-02-02 06:09:31 +08:00
|
|
|
struct pci_dev *dev = to_pci_dev(hwif->dev);
|
2007-02-08 01:18:13 +08:00
|
|
|
u16 bfifo = 0;
|
|
|
|
u8 dma_cmd;
|
|
|
|
|
2008-02-02 06:09:31 +08:00
|
|
|
pci_read_config_word(dev, hwif->select_data + 2, &bfifo);
|
2007-02-08 01:18:13 +08:00
|
|
|
printk(KERN_DEBUG "%s: %d bytes in FIFO\n", drive->name, bfifo & 0x1ff);
|
|
|
|
|
|
|
|
/* get DMA command mode */
|
2008-07-24 01:55:51 +08:00
|
|
|
dma_cmd = inb(hwif->dma_base + ATA_DMA_CMD);
|
2007-02-08 01:18:13 +08:00
|
|
|
/* stop DMA */
|
2009-04-18 23:42:19 +08:00
|
|
|
outb(dma_cmd & ~ATA_DMA_START, hwif->dma_base + ATA_DMA_CMD);
|
2007-02-08 01:18:13 +08:00
|
|
|
hpt370_clear_engine(drive);
|
|
|
|
}
|
|
|
|
|
2008-04-27 04:25:24 +08:00
|
|
|
static void hpt370_dma_start(ide_drive_t *drive)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
|
|
|
#ifdef HPT_RESET_STATE_ENGINE
|
|
|
|
hpt370_clear_engine(drive);
|
|
|
|
#endif
|
|
|
|
ide_dma_start(drive);
|
|
|
|
}
|
|
|
|
|
2008-04-27 04:25:24 +08:00
|
|
|
static int hpt370_dma_end(ide_drive_t *drive)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2009-01-07 00:20:52 +08:00
|
|
|
ide_hwif_t *hwif = drive->hwif;
|
2008-07-24 01:55:51 +08:00
|
|
|
u8 dma_stat = inb(hwif->dma_base + ATA_DMA_STATUS);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2009-04-18 23:42:19 +08:00
|
|
|
if (dma_stat & ATA_DMA_ACTIVE) {
|
2005-04-17 06:20:36 +08:00
|
|
|
/* wait a little */
|
|
|
|
udelay(20);
|
2008-07-24 01:55:51 +08:00
|
|
|
dma_stat = inb(hwif->dma_base + ATA_DMA_STATUS);
|
2009-04-18 23:42:19 +08:00
|
|
|
if (dma_stat & ATA_DMA_ACTIVE)
|
2007-02-08 01:18:13 +08:00
|
|
|
hpt370_irq_timeout(drive);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
2008-10-14 03:39:46 +08:00
|
|
|
return ide_dma_end(drive);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/* returns 1 if DMA IRQ issued, 0 otherwise */
|
2008-04-27 04:25:24 +08:00
|
|
|
static int hpt374_dma_test_irq(ide_drive_t *drive)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2009-01-07 00:20:52 +08:00
|
|
|
ide_hwif_t *hwif = drive->hwif;
|
2008-02-02 06:09:31 +08:00
|
|
|
struct pci_dev *dev = to_pci_dev(hwif->dev);
|
2005-04-17 06:20:36 +08:00
|
|
|
u16 bfifo = 0;
|
2007-02-08 01:18:05 +08:00
|
|
|
u8 dma_stat;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2008-02-02 06:09:31 +08:00
|
|
|
pci_read_config_word(dev, hwif->select_data + 2, &bfifo);
|
2005-04-17 06:20:36 +08:00
|
|
|
if (bfifo & 0x1FF) {
|
|
|
|
// printk("%s: %d bytes in FIFO\n", drive->name, bfifo);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2008-07-24 01:55:51 +08:00
|
|
|
dma_stat = inb(hwif->dma_base + ATA_DMA_STATUS);
|
2005-04-17 06:20:36 +08:00
|
|
|
/* return 1 if INTR asserted */
|
2009-04-18 23:42:19 +08:00
|
|
|
if (dma_stat & ATA_DMA_INTR)
|
2005-04-17 06:20:36 +08:00
|
|
|
return 1;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2008-04-27 04:25:24 +08:00
|
|
|
static int hpt374_dma_end(ide_drive_t *drive)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2009-01-07 00:20:52 +08:00
|
|
|
ide_hwif_t *hwif = drive->hwif;
|
2008-02-02 06:09:31 +08:00
|
|
|
struct pci_dev *dev = to_pci_dev(hwif->dev);
|
2007-02-08 01:18:05 +08:00
|
|
|
u8 mcr = 0, mcr_addr = hwif->select_data;
|
|
|
|
u8 bwsr = 0, mask = hwif->channel ? 0x02 : 0x01;
|
|
|
|
|
|
|
|
pci_read_config_byte(dev, 0x6a, &bwsr);
|
|
|
|
pci_read_config_byte(dev, mcr_addr, &mcr);
|
|
|
|
if (bwsr & mask)
|
|
|
|
pci_write_config_byte(dev, mcr_addr, mcr | 0x30);
|
2008-10-14 03:39:46 +08:00
|
|
|
return ide_dma_end(drive);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2006-12-13 16:35:47 +08:00
|
|
|
* hpt3xxn_set_clock - perform clock switching dance
|
|
|
|
* @hwif: hwif to switch
|
|
|
|
* @mode: clocking mode (0x21 for write, 0x23 otherwise)
|
2005-04-17 06:20:36 +08:00
|
|
|
*
|
2006-12-13 16:35:47 +08:00
|
|
|
* Switch the DPLL clock on the HPT3xxN devices. This is a right mess.
|
2005-04-17 06:20:36 +08:00
|
|
|
*/
|
2006-12-13 16:35:47 +08:00
|
|
|
|
|
|
|
static void hpt3xxn_set_clock(ide_hwif_t *hwif, u8 mode)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2008-01-26 05:17:05 +08:00
|
|
|
unsigned long base = hwif->extra_base;
|
|
|
|
u8 scr2 = inb(base + 0x6b);
|
2006-12-13 16:35:47 +08:00
|
|
|
|
|
|
|
if ((scr2 & 0x7f) == mode)
|
|
|
|
return;
|
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
/* Tristate the bus */
|
2008-01-26 05:17:05 +08:00
|
|
|
outb(0x80, base + 0x63);
|
|
|
|
outb(0x80, base + 0x67);
|
2006-12-13 16:35:47 +08:00
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
/* Switch clock and reset channels */
|
2008-01-26 05:17:05 +08:00
|
|
|
outb(mode, base + 0x6b);
|
|
|
|
outb(0xc0, base + 0x69);
|
2006-12-13 16:35:47 +08:00
|
|
|
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
/*
|
|
|
|
* Reset the state machines.
|
|
|
|
* NOTE: avoid accidentally enabling the disabled channels.
|
|
|
|
*/
|
2008-01-26 05:17:05 +08:00
|
|
|
outb(inb(base + 0x60) | 0x32, base + 0x60);
|
|
|
|
outb(inb(base + 0x64) | 0x32, base + 0x64);
|
2006-12-13 16:35:47 +08:00
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
/* Complete reset */
|
2008-01-26 05:17:05 +08:00
|
|
|
outb(0x00, base + 0x69);
|
2006-12-13 16:35:47 +08:00
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
/* Reconnect channels to bus */
|
2008-01-26 05:17:05 +08:00
|
|
|
outb(0x00, base + 0x63);
|
|
|
|
outb(0x00, base + 0x67);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2006-12-13 16:35:47 +08:00
|
|
|
* hpt3xxn_rw_disk - prepare for I/O
|
2005-04-17 06:20:36 +08:00
|
|
|
* @drive: drive for command
|
|
|
|
* @rq: block request structure
|
|
|
|
*
|
2006-12-13 16:35:47 +08:00
|
|
|
* This is called when a disk I/O is issued to HPT3xxN.
|
2005-04-17 06:20:36 +08:00
|
|
|
* We need it because of the clock switching.
|
|
|
|
*/
|
|
|
|
|
2006-12-13 16:35:47 +08:00
|
|
|
static void hpt3xxn_rw_disk(ide_drive_t *drive, struct request *rq)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2010-09-28 02:01:32 +08:00
|
|
|
hpt3xxn_set_clock(drive->hwif, rq_data_dir(rq) ? 0x21 : 0x23);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
/**
|
|
|
|
* hpt37x_calibrate_dpll - calibrate the DPLL
|
|
|
|
* @dev: PCI device
|
|
|
|
*
|
|
|
|
* Perform a calibration cycle on the DPLL.
|
|
|
|
* Returns 1 if this succeeds
|
|
|
|
*/
|
2008-10-11 04:39:32 +08:00
|
|
|
static int hpt37x_calibrate_dpll(struct pci_dev *dev, u16 f_low, u16 f_high)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
u32 dpll = (f_high << 16) | f_low | 0x100;
|
|
|
|
u8 scr2;
|
|
|
|
int i;
|
2005-06-28 06:24:27 +08:00
|
|
|
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
pci_write_config_dword(dev, 0x5c, dpll);
|
2005-06-28 06:24:27 +08:00
|
|
|
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
/* Wait for oscillator ready */
|
|
|
|
for(i = 0; i < 0x5000; ++i) {
|
|
|
|
udelay(50);
|
|
|
|
pci_read_config_byte(dev, 0x5b, &scr2);
|
|
|
|
if (scr2 & 0x80)
|
2005-06-28 06:24:27 +08:00
|
|
|
break;
|
|
|
|
}
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
/* See if it stays ready (we'll just bail out if it's not yet) */
|
|
|
|
for(i = 0; i < 0x1000; ++i) {
|
|
|
|
pci_read_config_byte(dev, 0x5b, &scr2);
|
|
|
|
/* DPLL destabilized? */
|
|
|
|
if(!(scr2 & 0x80))
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
/* Turn off tuning, we have the DPLL set */
|
|
|
|
pci_read_config_dword (dev, 0x5c, &dpll);
|
|
|
|
pci_write_config_dword(dev, 0x5c, (dpll & ~0x100));
|
|
|
|
return 1;
|
2005-06-28 06:24:27 +08:00
|
|
|
}
|
|
|
|
|
2008-10-11 04:39:32 +08:00
|
|
|
static void hpt3xx_disable_fast_irq(struct pci_dev *dev, u8 mcr_addr)
|
2008-10-11 04:39:32 +08:00
|
|
|
{
|
|
|
|
struct ide_host *host = pci_get_drvdata(dev);
|
|
|
|
struct hpt_info *info = host->host_priv + (&dev->dev == host->dev[1]);
|
|
|
|
u8 chip_type = info->chip_type;
|
|
|
|
u8 new_mcr, old_mcr = 0;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Disable the "fast interrupt" prediction. Don't hold off
|
|
|
|
* on interrupts. (== 0x01 despite what the docs say)
|
|
|
|
*/
|
|
|
|
pci_read_config_byte(dev, mcr_addr + 1, &old_mcr);
|
|
|
|
|
|
|
|
if (chip_type >= HPT374)
|
|
|
|
new_mcr = old_mcr & ~0x07;
|
|
|
|
else if (chip_type >= HPT370) {
|
|
|
|
new_mcr = old_mcr;
|
|
|
|
new_mcr &= ~0x02;
|
|
|
|
#ifdef HPT_DELAY_INTERRUPT
|
|
|
|
new_mcr &= ~0x01;
|
|
|
|
#else
|
|
|
|
new_mcr |= 0x01;
|
|
|
|
#endif
|
|
|
|
} else /* HPT366 and HPT368 */
|
|
|
|
new_mcr = old_mcr & ~0x80;
|
|
|
|
|
|
|
|
if (new_mcr != old_mcr)
|
|
|
|
pci_write_config_byte(dev, mcr_addr + 1, new_mcr);
|
|
|
|
}
|
|
|
|
|
2009-03-25 06:22:53 +08:00
|
|
|
static int init_chipset_hpt366(struct pci_dev *dev)
|
2005-06-28 06:24:27 +08:00
|
|
|
{
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
unsigned long io_base = pci_resource_start(dev, 4);
|
ide: Fix pointer arithmetic in hpt3xx driver code (3rd try)
git commit 74811f355f4f69a187fa74892dcf2a684b84ce99 causes crash at
module load (or boot) time on my machine with a hpt374 controller.
The reason for this is that for initializing second controller which sets
(hwif->dev == host->dev[1]) to true (1), adds 1 to a void ptr, which
advances it by one byte instead of advancing it by sizeof(hpt_info) bytes.
Because of this, all initialization functions get corrupted data in info
variable which causes a crash at boot time.
This patch fixes that and makes my machine boot again.
The card itself is a HPT374 raid conroller: Here is the lspci -v output:
03:06.0 RAID bus controller: HighPoint Technologies, Inc. HPT374 (rev
07)
Subsystem: HighPoint Technologies, Inc. Unknown device 0001
Flags: bus master, 66MHz, medium devsel, latency 120, IRQ 28
I/O ports at 8000 [size=8]
I/O ports at 7800 [size=4]
I/O ports at 7400 [size=8]
I/O ports at 7000 [size=4]
I/O ports at 6800 [size=256]
Expansion ROM at fe8e0000 [disabled] [size=128K]
Capabilities: [60] Power Management version 2
03:06.1 RAID bus controller: HighPoint Technologies, Inc. HPT374 (rev
07)
Subsystem: HighPoint Technologies, Inc. Unknown device 0001
Flags: bus master, 66MHz, medium devsel, latency 120, IRQ 28
I/O ports at 9800 [size=8]
I/O ports at 9400 [size=4]
I/O ports at 9000 [size=8]
I/O ports at 8800 [size=4]
I/O ports at 8400 [size=256]
Capabilities: [60] Power Management version 2
Signed-off-by: Masoud Sharbiani <masouds@google.com>
Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
[bart: use dev_get_drvdata() per Sergei's suggestion]
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2008-09-11 04:22:34 +08:00
|
|
|
struct hpt_info *info = hpt3xx_get_info(&dev->dev);
|
2008-07-25 04:53:33 +08:00
|
|
|
const char *name = DRV_NAME;
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
u8 pci_clk, dpll_clk = 0; /* PCI and DPLL clock in MHz */
|
2007-09-12 04:28:35 +08:00
|
|
|
u8 chip_type;
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
enum ata_clock clock;
|
|
|
|
|
2007-09-12 04:28:35 +08:00
|
|
|
chip_type = info->chip_type;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
pci_write_config_byte(dev, PCI_CACHE_LINE_SIZE, (L1_CACHE_BYTES / 4));
|
|
|
|
pci_write_config_byte(dev, PCI_LATENCY_TIMER, 0x78);
|
|
|
|
pci_write_config_byte(dev, PCI_MIN_GNT, 0x08);
|
|
|
|
pci_write_config_byte(dev, PCI_MAX_LAT, 0x08);
|
2006-12-13 16:35:52 +08:00
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
/*
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
* First, try to estimate the PCI clock frequency...
|
2005-04-17 06:20:36 +08:00
|
|
|
*/
|
2007-09-12 04:28:35 +08:00
|
|
|
if (chip_type >= HPT370) {
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
u8 scr1 = 0;
|
|
|
|
u16 f_cnt = 0;
|
|
|
|
u32 temp = 0;
|
|
|
|
|
|
|
|
/* Interrupt force enable. */
|
|
|
|
pci_read_config_byte(dev, 0x5a, &scr1);
|
|
|
|
if (scr1 & 0x10)
|
|
|
|
pci_write_config_byte(dev, 0x5a, scr1 & ~0x10);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* HighPoint does this for HPT372A.
|
|
|
|
* NOTE: This register is only writeable via I/O space.
|
|
|
|
*/
|
2007-09-12 04:28:35 +08:00
|
|
|
if (chip_type == HPT372A)
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
outb(0x0e, io_base + 0x9c);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Default to PCI clock. Make sure MA15/16 are set to output
|
|
|
|
* to prevent drives having problems with 40-pin cables.
|
|
|
|
*/
|
|
|
|
pci_write_config_byte(dev, 0x5b, 0x23);
|
2006-12-13 16:35:47 +08:00
|
|
|
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
/*
|
|
|
|
* We'll have to read f_CNT value in order to determine
|
|
|
|
* the PCI clock frequency according to the following ratio:
|
|
|
|
*
|
|
|
|
* f_CNT = Fpci * 192 / Fdpll
|
|
|
|
*
|
|
|
|
* First try reading the register in which the HighPoint BIOS
|
|
|
|
* saves f_CNT value before reprogramming the DPLL from its
|
|
|
|
* default setting (which differs for the various chips).
|
|
|
|
*
|
2007-09-12 04:28:35 +08:00
|
|
|
* NOTE: This register is only accessible via I/O space;
|
|
|
|
* HPT374 BIOS only saves it for the function 0, so we have to
|
|
|
|
* always read it from there -- no need to check the result of
|
|
|
|
* pci_get_slot() for the function 0 as the whole device has
|
|
|
|
* been already "pinned" (via function 1) in init_setup_hpt374()
|
|
|
|
*/
|
|
|
|
if (chip_type == HPT374 && (PCI_FUNC(dev->devfn) & 1)) {
|
|
|
|
struct pci_dev *dev1 = pci_get_slot(dev->bus,
|
|
|
|
dev->devfn - 1);
|
|
|
|
unsigned long io_base = pci_resource_start(dev1, 4);
|
|
|
|
|
|
|
|
temp = inl(io_base + 0x90);
|
|
|
|
pci_dev_put(dev1);
|
|
|
|
} else
|
|
|
|
temp = inl(io_base + 0x90);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* In case the signature check fails, we'll have to
|
|
|
|
* resort to reading the f_CNT register itself in hopes
|
|
|
|
* that nobody has touched the DPLL yet...
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
*/
|
|
|
|
if ((temp & 0xFFFFF000) != 0xABCDE000) {
|
|
|
|
int i;
|
|
|
|
|
2008-07-25 04:53:31 +08:00
|
|
|
printk(KERN_WARNING "%s %s: no clock data saved by "
|
|
|
|
"BIOS\n", name, pci_name(dev));
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
|
|
|
|
/* Calculate the average value of f_CNT. */
|
|
|
|
for (temp = i = 0; i < 128; i++) {
|
|
|
|
pci_read_config_word(dev, 0x78, &f_cnt);
|
|
|
|
temp += f_cnt & 0x1ff;
|
|
|
|
mdelay(1);
|
|
|
|
}
|
|
|
|
f_cnt = temp / 128;
|
|
|
|
} else
|
|
|
|
f_cnt = temp & 0x1ff;
|
|
|
|
|
|
|
|
dpll_clk = info->dpll_clk;
|
|
|
|
pci_clk = (f_cnt * dpll_clk) / 192;
|
|
|
|
|
|
|
|
/* Clamp PCI clock to bands. */
|
|
|
|
if (pci_clk < 40)
|
|
|
|
pci_clk = 33;
|
|
|
|
else if(pci_clk < 45)
|
|
|
|
pci_clk = 40;
|
|
|
|
else if(pci_clk < 55)
|
|
|
|
pci_clk = 50;
|
2005-04-17 06:20:36 +08:00
|
|
|
else
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
pci_clk = 66;
|
2006-12-13 16:35:47 +08:00
|
|
|
|
2008-07-25 04:53:31 +08:00
|
|
|
printk(KERN_INFO "%s %s: DPLL base: %d MHz, f_CNT: %d, "
|
|
|
|
"assuming %d MHz PCI\n", name, pci_name(dev),
|
|
|
|
dpll_clk, f_cnt, pci_clk);
|
2007-02-08 01:17:51 +08:00
|
|
|
} else {
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
u32 itr1 = 0;
|
|
|
|
|
|
|
|
pci_read_config_dword(dev, 0x40, &itr1);
|
|
|
|
|
|
|
|
/* Detect PCI clock by looking at cmd_high_time. */
|
2016-07-12 18:59:39 +08:00
|
|
|
switch ((itr1 >> 8) & 0x0f) {
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
case 0x09:
|
|
|
|
pci_clk = 40;
|
2007-02-08 01:18:20 +08:00
|
|
|
break;
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
case 0x05:
|
|
|
|
pci_clk = 25;
|
2007-02-08 01:18:20 +08:00
|
|
|
break;
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
case 0x07:
|
|
|
|
default:
|
|
|
|
pci_clk = 33;
|
2007-02-08 01:18:20 +08:00
|
|
|
break;
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
}
|
2006-12-13 16:35:47 +08:00
|
|
|
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
/* Let's assume we'll use PCI clock for the ATA clock... */
|
|
|
|
switch (pci_clk) {
|
|
|
|
case 25:
|
|
|
|
clock = ATA_CLOCK_25MHZ;
|
|
|
|
break;
|
|
|
|
case 33:
|
|
|
|
default:
|
|
|
|
clock = ATA_CLOCK_33MHZ;
|
|
|
|
break;
|
|
|
|
case 40:
|
|
|
|
clock = ATA_CLOCK_40MHZ;
|
|
|
|
break;
|
|
|
|
case 50:
|
|
|
|
clock = ATA_CLOCK_50MHZ;
|
|
|
|
break;
|
|
|
|
case 66:
|
|
|
|
clock = ATA_CLOCK_66MHZ;
|
|
|
|
break;
|
|
|
|
}
|
2006-12-13 16:35:47 +08:00
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
/*
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
* Only try the DPLL if we don't have a table for the PCI clock that
|
|
|
|
* we are running at for HPT370/A, always use it for anything newer...
|
2005-06-28 06:24:27 +08:00
|
|
|
*
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
* NOTE: Using the internal DPLL results in slow reads on 33 MHz PCI.
|
|
|
|
* We also don't like using the DPLL because this causes glitches
|
|
|
|
* on PRST-/SRST- when the state engine gets reset...
|
2005-04-17 06:20:36 +08:00
|
|
|
*/
|
2008-01-26 05:17:05 +08:00
|
|
|
if (chip_type >= HPT374 || info->timings->clock_table[clock] == NULL) {
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
u16 f_low, delta = pci_clk < 50 ? 2 : 4;
|
|
|
|
int adjust;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Select 66 MHz DPLL clock only if UltraATA/133 mode is
|
|
|
|
* supported/enabled, use 50 MHz DPLL clock otherwise...
|
|
|
|
*/
|
2007-10-19 06:30:09 +08:00
|
|
|
if (info->udma_mask == ATA_UDMA6) {
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
dpll_clk = 66;
|
|
|
|
clock = ATA_CLOCK_66MHZ;
|
|
|
|
} else if (dpll_clk) { /* HPT36x chips don't have DPLL */
|
|
|
|
dpll_clk = 50;
|
|
|
|
clock = ATA_CLOCK_50MHZ;
|
|
|
|
}
|
2005-06-28 06:24:27 +08:00
|
|
|
|
2008-01-26 05:17:05 +08:00
|
|
|
if (info->timings->clock_table[clock] == NULL) {
|
2008-07-25 04:53:31 +08:00
|
|
|
printk(KERN_ERR "%s %s: unknown bus timing!\n",
|
|
|
|
name, pci_name(dev));
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
return -EIO;
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
/* Select the DPLL clock. */
|
|
|
|
pci_write_config_byte(dev, 0x5b, 0x21);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Adjust the DPLL based upon PCI clock, enable it,
|
|
|
|
* and wait for stabilization...
|
|
|
|
*/
|
|
|
|
f_low = (pci_clk * 48) / dpll_clk;
|
|
|
|
|
|
|
|
for (adjust = 0; adjust < 8; adjust++) {
|
|
|
|
if(hpt37x_calibrate_dpll(dev, f_low, f_low + delta))
|
|
|
|
break;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* See if it'll settle at a fractionally different clock
|
|
|
|
*/
|
|
|
|
if (adjust & 1)
|
|
|
|
f_low -= adjust >> 1;
|
|
|
|
else
|
|
|
|
f_low += adjust >> 1;
|
|
|
|
}
|
|
|
|
if (adjust == 8) {
|
2008-07-25 04:53:31 +08:00
|
|
|
printk(KERN_ERR "%s %s: DPLL did not stabilize!\n",
|
|
|
|
name, pci_name(dev));
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
return -EIO;
|
|
|
|
}
|
|
|
|
|
2008-07-25 04:53:31 +08:00
|
|
|
printk(KERN_INFO "%s %s: using %d MHz DPLL clock\n",
|
|
|
|
name, pci_name(dev), dpll_clk);
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
} else {
|
|
|
|
/* Mark the fact that we're not using the DPLL. */
|
|
|
|
dpll_clk = 0;
|
|
|
|
|
2008-07-25 04:53:31 +08:00
|
|
|
printk(KERN_INFO "%s %s: using %d MHz PCI clock\n",
|
|
|
|
name, pci_name(dev), pci_clk);
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
}
|
2005-06-28 06:24:27 +08:00
|
|
|
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
/* Store the clock frequencies. */
|
|
|
|
info->dpll_clk = dpll_clk;
|
|
|
|
info->pci_clk = pci_clk;
|
2008-01-26 05:17:05 +08:00
|
|
|
info->clock = clock;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2007-09-12 04:28:35 +08:00
|
|
|
if (chip_type >= HPT370) {
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
u8 mcr1, mcr4;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Reset the state engines.
|
|
|
|
* NOTE: Avoid accidentally enabling the disabled channels.
|
|
|
|
*/
|
|
|
|
pci_read_config_byte (dev, 0x50, &mcr1);
|
|
|
|
pci_read_config_byte (dev, 0x54, &mcr4);
|
|
|
|
pci_write_config_byte(dev, 0x50, (mcr1 | 0x32));
|
|
|
|
pci_write_config_byte(dev, 0x54, (mcr4 | 0x32));
|
|
|
|
udelay(100);
|
2007-02-08 01:18:11 +08:00
|
|
|
}
|
2005-04-17 06:20:36 +08:00
|
|
|
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
/*
|
|
|
|
* On HPT371N, if ATA clock is 66 MHz we must set bit 2 in
|
|
|
|
* the MISC. register to stretch the UltraDMA Tss timing.
|
|
|
|
* NOTE: This register is only writeable via I/O space.
|
|
|
|
*/
|
2007-09-12 04:28:35 +08:00
|
|
|
if (chip_type == HPT371N && clock == ATA_CLOCK_66MHZ)
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
outb(inb(io_base + 0x9c) | 0x04, io_base + 0x9c);
|
|
|
|
|
2008-10-11 04:39:32 +08:00
|
|
|
hpt3xx_disable_fast_irq(dev, 0x50);
|
|
|
|
hpt3xx_disable_fast_irq(dev, 0x54);
|
|
|
|
|
2009-03-25 06:22:53 +08:00
|
|
|
return 0;
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2008-08-06 00:17:04 +08:00
|
|
|
static u8 hpt3xx_cable_detect(ide_hwif_t *hwif)
|
2008-02-03 02:56:31 +08:00
|
|
|
{
|
|
|
|
struct pci_dev *dev = to_pci_dev(hwif->dev);
|
ide: Fix pointer arithmetic in hpt3xx driver code (3rd try)
git commit 74811f355f4f69a187fa74892dcf2a684b84ce99 causes crash at
module load (or boot) time on my machine with a hpt374 controller.
The reason for this is that for initializing second controller which sets
(hwif->dev == host->dev[1]) to true (1), adds 1 to a void ptr, which
advances it by one byte instead of advancing it by sizeof(hpt_info) bytes.
Because of this, all initialization functions get corrupted data in info
variable which causes a crash at boot time.
This patch fixes that and makes my machine boot again.
The card itself is a HPT374 raid conroller: Here is the lspci -v output:
03:06.0 RAID bus controller: HighPoint Technologies, Inc. HPT374 (rev
07)
Subsystem: HighPoint Technologies, Inc. Unknown device 0001
Flags: bus master, 66MHz, medium devsel, latency 120, IRQ 28
I/O ports at 8000 [size=8]
I/O ports at 7800 [size=4]
I/O ports at 7400 [size=8]
I/O ports at 7000 [size=4]
I/O ports at 6800 [size=256]
Expansion ROM at fe8e0000 [disabled] [size=128K]
Capabilities: [60] Power Management version 2
03:06.1 RAID bus controller: HighPoint Technologies, Inc. HPT374 (rev
07)
Subsystem: HighPoint Technologies, Inc. Unknown device 0001
Flags: bus master, 66MHz, medium devsel, latency 120, IRQ 28
I/O ports at 9800 [size=8]
I/O ports at 9400 [size=4]
I/O ports at 9000 [size=8]
I/O ports at 8800 [size=4]
I/O ports at 8400 [size=256]
Capabilities: [60] Power Management version 2
Signed-off-by: Masoud Sharbiani <masouds@google.com>
Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
[bart: use dev_get_drvdata() per Sergei's suggestion]
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2008-09-11 04:22:34 +08:00
|
|
|
struct hpt_info *info = hpt3xx_get_info(hwif->dev);
|
2008-02-03 02:56:31 +08:00
|
|
|
u8 chip_type = info->chip_type;
|
|
|
|
u8 scr1 = 0, ata66 = hwif->channel ? 0x01 : 0x02;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The HPT37x uses the CBLID pins as outputs for MA15/MA16
|
|
|
|
* address lines to access an external EEPROM. To read valid
|
|
|
|
* cable detect state the pins must be enabled as inputs.
|
|
|
|
*/
|
|
|
|
if (chip_type == HPT374 && (PCI_FUNC(dev->devfn) & 1)) {
|
|
|
|
/*
|
|
|
|
* HPT374 PCI function 1
|
|
|
|
* - set bit 15 of reg 0x52 to enable TCBLID as input
|
|
|
|
* - set bit 15 of reg 0x56 to enable FCBLID as input
|
|
|
|
*/
|
|
|
|
u8 mcr_addr = hwif->select_data + 2;
|
|
|
|
u16 mcr;
|
|
|
|
|
|
|
|
pci_read_config_word(dev, mcr_addr, &mcr);
|
2010-09-28 02:00:40 +08:00
|
|
|
pci_write_config_word(dev, mcr_addr, mcr | 0x8000);
|
|
|
|
/* Debounce, then read cable ID register */
|
|
|
|
udelay(10);
|
2008-02-03 02:56:31 +08:00
|
|
|
pci_read_config_byte(dev, 0x5a, &scr1);
|
|
|
|
pci_write_config_word(dev, mcr_addr, mcr);
|
|
|
|
} else if (chip_type >= HPT370) {
|
|
|
|
/*
|
|
|
|
* HPT370/372 and 374 pcifn 0
|
|
|
|
* - clear bit 0 of reg 0x5b to enable P/SCBLID as inputs
|
|
|
|
*/
|
|
|
|
u8 scr2 = 0;
|
|
|
|
|
|
|
|
pci_read_config_byte(dev, 0x5b, &scr2);
|
2010-09-28 02:00:40 +08:00
|
|
|
pci_write_config_byte(dev, 0x5b, scr2 & ~1);
|
|
|
|
/* Debounce, then read cable ID register */
|
|
|
|
udelay(10);
|
2008-02-03 02:56:31 +08:00
|
|
|
pci_read_config_byte(dev, 0x5a, &scr1);
|
2010-09-28 02:00:40 +08:00
|
|
|
pci_write_config_byte(dev, 0x5b, scr2);
|
2008-02-03 02:56:31 +08:00
|
|
|
} else
|
|
|
|
pci_read_config_byte(dev, 0x5a, &scr1);
|
|
|
|
|
|
|
|
return (scr1 & ata66) ? ATA_CBL_PATA40 : ATA_CBL_PATA80;
|
|
|
|
}
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static void init_hwif_hpt366(ide_hwif_t *hwif)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
ide: Fix pointer arithmetic in hpt3xx driver code (3rd try)
git commit 74811f355f4f69a187fa74892dcf2a684b84ce99 causes crash at
module load (or boot) time on my machine with a hpt374 controller.
The reason for this is that for initializing second controller which sets
(hwif->dev == host->dev[1]) to true (1), adds 1 to a void ptr, which
advances it by one byte instead of advancing it by sizeof(hpt_info) bytes.
Because of this, all initialization functions get corrupted data in info
variable which causes a crash at boot time.
This patch fixes that and makes my machine boot again.
The card itself is a HPT374 raid conroller: Here is the lspci -v output:
03:06.0 RAID bus controller: HighPoint Technologies, Inc. HPT374 (rev
07)
Subsystem: HighPoint Technologies, Inc. Unknown device 0001
Flags: bus master, 66MHz, medium devsel, latency 120, IRQ 28
I/O ports at 8000 [size=8]
I/O ports at 7800 [size=4]
I/O ports at 7400 [size=8]
I/O ports at 7000 [size=4]
I/O ports at 6800 [size=256]
Expansion ROM at fe8e0000 [disabled] [size=128K]
Capabilities: [60] Power Management version 2
03:06.1 RAID bus controller: HighPoint Technologies, Inc. HPT374 (rev
07)
Subsystem: HighPoint Technologies, Inc. Unknown device 0001
Flags: bus master, 66MHz, medium devsel, latency 120, IRQ 28
I/O ports at 9800 [size=8]
I/O ports at 9400 [size=4]
I/O ports at 9000 [size=8]
I/O ports at 8800 [size=4]
I/O ports at 8400 [size=256]
Capabilities: [60] Power Management version 2
Signed-off-by: Masoud Sharbiani <masouds@google.com>
Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
[bart: use dev_get_drvdata() per Sergei's suggestion]
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2008-09-11 04:22:34 +08:00
|
|
|
struct hpt_info *info = hpt3xx_get_info(hwif->dev);
|
2007-09-12 04:28:36 +08:00
|
|
|
u8 chip_type = info->chip_type;
|
2007-02-08 01:18:05 +08:00
|
|
|
|
|
|
|
/* Cache the channel's MISC. control registers' offset */
|
2007-09-12 04:28:36 +08:00
|
|
|
hwif->select_data = hwif->channel ? 0x54 : 0x50;
|
2007-02-08 01:18:05 +08:00
|
|
|
|
2006-12-13 16:35:47 +08:00
|
|
|
/*
|
|
|
|
* HPT3xxN chips have some complications:
|
|
|
|
*
|
|
|
|
* - on 33 MHz PCI we must clock switch
|
|
|
|
* - on 66 MHz PCI we must NOT use the PCI clock
|
|
|
|
*/
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
if (chip_type >= HPT372N && info->dpll_clk && info->pci_clk < 66) {
|
2006-12-13 16:35:47 +08:00
|
|
|
/*
|
|
|
|
* Clock is shared between the channels,
|
|
|
|
* so we'll have to serialize them... :-(
|
|
|
|
*/
|
2008-12-30 03:27:36 +08:00
|
|
|
hwif->host->host_flags |= IDE_HFLAG_SERIALIZE;
|
2006-12-13 16:35:47 +08:00
|
|
|
hwif->rw_disk = &hpt3xxn_rw_disk;
|
|
|
|
}
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static int init_dma_hpt366(ide_hwif_t *hwif,
|
2008-04-27 04:25:22 +08:00
|
|
|
const struct ide_port_info *d)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2008-02-02 06:09:31 +08:00
|
|
|
struct pci_dev *dev = to_pci_dev(hwif->dev);
|
2008-04-27 04:25:22 +08:00
|
|
|
unsigned long flags, base = ide_pci_dma_base(hwif, d);
|
|
|
|
u8 dma_old, dma_new, masterdma = 0, slavedma = 0;
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2008-07-24 01:55:51 +08:00
|
|
|
if (base == 0)
|
|
|
|
return -1;
|
|
|
|
|
|
|
|
hwif->dma_base = base;
|
|
|
|
|
|
|
|
if (ide_pci_check_simplex(hwif, d) < 0)
|
|
|
|
return -1;
|
|
|
|
|
|
|
|
if (ide_pci_set_master(dev, d->name) < 0)
|
2008-04-27 04:25:22 +08:00
|
|
|
return -1;
|
|
|
|
|
|
|
|
dma_old = inb(base + 2);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
local_irq_save(flags);
|
|
|
|
|
|
|
|
dma_new = dma_old;
|
2007-02-08 01:18:05 +08:00
|
|
|
pci_read_config_byte(dev, hwif->channel ? 0x4b : 0x43, &masterdma);
|
|
|
|
pci_read_config_byte(dev, hwif->channel ? 0x4f : 0x47, &slavedma);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
if (masterdma & 0x30) dma_new |= 0x20;
|
2007-02-08 01:18:05 +08:00
|
|
|
if ( slavedma & 0x30) dma_new |= 0x40;
|
2005-04-17 06:20:36 +08:00
|
|
|
if (dma_new != dma_old)
|
2008-04-27 04:25:22 +08:00
|
|
|
outb(dma_new, base + 2);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
local_irq_restore(flags);
|
2008-04-27 04:25:22 +08:00
|
|
|
|
|
|
|
printk(KERN_INFO " %s: BM-DMA at 0x%04lx-0x%04lx\n",
|
|
|
|
hwif->name, base, base + 7);
|
|
|
|
|
|
|
|
hwif->extra_base = base + (hwif->channel ? 8 : 16);
|
|
|
|
|
|
|
|
if (ide_allocate_dma_engine(hwif))
|
|
|
|
return -1;
|
|
|
|
|
|
|
|
return 0;
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static void hpt374_init(struct pci_dev *dev, struct pci_dev *dev2)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2007-10-19 06:30:09 +08:00
|
|
|
if (dev2->irq != dev->irq) {
|
|
|
|
/* FIXME: we need a core pci_set_interrupt() */
|
|
|
|
dev2->irq = dev->irq;
|
2008-07-25 04:53:32 +08:00
|
|
|
printk(KERN_INFO DRV_NAME " %s: PCI config space interrupt "
|
2008-07-25 04:53:31 +08:00
|
|
|
"fixed\n", pci_name(dev2));
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static void hpt371_init(struct pci_dev *dev)
|
2006-12-13 16:35:47 +08:00
|
|
|
{
|
2007-06-09 06:46:36 +08:00
|
|
|
u8 mcr1 = 0;
|
2007-02-08 01:17:51 +08:00
|
|
|
|
2006-12-13 16:35:47 +08:00
|
|
|
/*
|
|
|
|
* HPT371 chips physically have only one channel, the secondary one,
|
|
|
|
* but the primary channel registers do exist! Go figure...
|
|
|
|
* So, we manually disable the non-existing channel here
|
|
|
|
* (if the BIOS hasn't done this already).
|
|
|
|
*/
|
|
|
|
pci_read_config_byte(dev, 0x50, &mcr1);
|
|
|
|
if (mcr1 & 0x04)
|
2007-02-08 01:17:51 +08:00
|
|
|
pci_write_config_byte(dev, 0x50, mcr1 & ~0x04);
|
|
|
|
}
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static int hpt36x_init(struct pci_dev *dev, struct pci_dev *dev2)
|
2007-02-08 01:17:51 +08:00
|
|
|
{
|
2007-10-19 06:30:09 +08:00
|
|
|
u8 mcr1 = 0, pin1 = 0, pin2 = 0;
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
|
2007-10-19 06:30:09 +08:00
|
|
|
/*
|
|
|
|
* Now we'll have to force both channels enabled if
|
|
|
|
* at least one of them has been enabled by BIOS...
|
|
|
|
*/
|
|
|
|
pci_read_config_byte(dev, 0x50, &mcr1);
|
|
|
|
if (mcr1 & 0x30)
|
|
|
|
pci_write_config_byte(dev, 0x50, mcr1 | 0x30);
|
2006-12-13 16:35:47 +08:00
|
|
|
|
2007-10-19 06:30:09 +08:00
|
|
|
pci_read_config_byte(dev, PCI_INTERRUPT_PIN, &pin1);
|
|
|
|
pci_read_config_byte(dev2, PCI_INTERRUPT_PIN, &pin2);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2007-10-19 06:30:09 +08:00
|
|
|
if (pin1 != pin2 && dev->irq == dev2->irq) {
|
2008-07-25 04:53:32 +08:00
|
|
|
printk(KERN_INFO DRV_NAME " %s: onboard version of chipset, "
|
2008-07-25 04:53:31 +08:00
|
|
|
"pin1=%d pin2=%d\n", pci_name(dev), pin1, pin2);
|
2007-10-19 06:30:09 +08:00
|
|
|
return 1;
|
2007-07-10 05:17:55 +08:00
|
|
|
}
|
|
|
|
|
2007-10-19 06:30:09 +08:00
|
|
|
return 0;
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2008-01-26 05:17:18 +08:00
|
|
|
#define IDE_HFLAGS_HPT3XX \
|
|
|
|
(IDE_HFLAG_NO_ATAPI_DMA | \
|
|
|
|
IDE_HFLAG_OFF_BOARD)
|
|
|
|
|
2008-04-27 04:25:14 +08:00
|
|
|
static const struct ide_port_ops hpt3xx_port_ops = {
|
|
|
|
.set_pio_mode = hpt3xx_set_pio_mode,
|
|
|
|
.set_dma_mode = hpt3xx_set_mode,
|
|
|
|
.maskproc = hpt3xx_maskproc,
|
|
|
|
.mdma_filter = hpt3xx_mdma_filter,
|
|
|
|
.udma_filter = hpt3xx_udma_filter,
|
|
|
|
.cable_detect = hpt3xx_cable_detect,
|
|
|
|
};
|
|
|
|
|
2008-04-27 04:25:24 +08:00
|
|
|
static const struct ide_dma_ops hpt37x_dma_ops = {
|
|
|
|
.dma_host_set = ide_dma_host_set,
|
|
|
|
.dma_setup = ide_dma_setup,
|
|
|
|
.dma_start = ide_dma_start,
|
2008-04-27 04:25:24 +08:00
|
|
|
.dma_end = hpt374_dma_end,
|
|
|
|
.dma_test_irq = hpt374_dma_test_irq,
|
2008-04-27 04:25:24 +08:00
|
|
|
.dma_lost_irq = ide_dma_lost_irq,
|
2009-03-27 19:46:47 +08:00
|
|
|
.dma_timer_expiry = ide_dma_sff_timer_expiry,
|
2009-01-07 00:21:02 +08:00
|
|
|
.dma_sff_read_status = ide_dma_sff_read_status,
|
2008-04-27 04:25:24 +08:00
|
|
|
};
|
|
|
|
|
2008-04-27 04:25:24 +08:00
|
|
|
static const struct ide_dma_ops hpt370_dma_ops = {
|
|
|
|
.dma_host_set = ide_dma_host_set,
|
|
|
|
.dma_setup = ide_dma_setup,
|
2008-04-27 04:25:24 +08:00
|
|
|
.dma_start = hpt370_dma_start,
|
|
|
|
.dma_end = hpt370_dma_end,
|
2008-04-27 04:25:24 +08:00
|
|
|
.dma_test_irq = ide_dma_test_irq,
|
|
|
|
.dma_lost_irq = ide_dma_lost_irq,
|
2009-03-27 19:46:47 +08:00
|
|
|
.dma_timer_expiry = ide_dma_sff_timer_expiry,
|
2009-04-01 02:15:19 +08:00
|
|
|
.dma_clear = hpt370_irq_timeout,
|
2009-01-07 00:21:02 +08:00
|
|
|
.dma_sff_read_status = ide_dma_sff_read_status,
|
2008-04-27 04:25:24 +08:00
|
|
|
};
|
|
|
|
|
2008-04-27 04:25:24 +08:00
|
|
|
static const struct ide_dma_ops hpt36x_dma_ops = {
|
|
|
|
.dma_host_set = ide_dma_host_set,
|
|
|
|
.dma_setup = ide_dma_setup,
|
|
|
|
.dma_start = ide_dma_start,
|
2008-10-14 03:39:46 +08:00
|
|
|
.dma_end = ide_dma_end,
|
2008-04-27 04:25:24 +08:00
|
|
|
.dma_test_irq = ide_dma_test_irq,
|
2008-04-27 04:25:24 +08:00
|
|
|
.dma_lost_irq = hpt366_dma_lost_irq,
|
2009-03-27 19:46:47 +08:00
|
|
|
.dma_timer_expiry = ide_dma_sff_timer_expiry,
|
2009-01-07 00:21:02 +08:00
|
|
|
.dma_sff_read_status = ide_dma_sff_read_status,
|
2008-04-27 04:25:24 +08:00
|
|
|
};
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static const struct ide_port_info hpt366_chipsets[] = {
|
2008-07-25 04:53:32 +08:00
|
|
|
{ /* 0: HPT36x */
|
|
|
|
.name = DRV_NAME,
|
2005-04-17 06:20:36 +08:00
|
|
|
.init_chipset = init_chipset_hpt366,
|
|
|
|
.init_hwif = init_hwif_hpt366,
|
|
|
|
.init_dma = init_dma_hpt366,
|
2007-10-19 06:30:09 +08:00
|
|
|
/*
|
|
|
|
* HPT36x chips have one channel per function and have
|
|
|
|
* both channel enable bits located differently and visible
|
|
|
|
* to both functions -- really stupid design decision... :-(
|
|
|
|
* Bit 4 is for the primary channel, bit 5 for the secondary.
|
|
|
|
*/
|
|
|
|
.enablebits = {{0x50,0x10,0x10}, {0x54,0x04,0x04}},
|
2008-04-27 04:25:14 +08:00
|
|
|
.port_ops = &hpt3xx_port_ops,
|
2008-04-27 04:25:24 +08:00
|
|
|
.dma_ops = &hpt36x_dma_ops,
|
2008-01-26 05:17:18 +08:00
|
|
|
.host_flags = IDE_HFLAGS_HPT3XX | IDE_HFLAG_SINGLE,
|
2007-07-20 07:11:59 +08:00
|
|
|
.pio_mask = ATA_PIO4,
|
2007-10-19 06:30:07 +08:00
|
|
|
.mwdma_mask = ATA_MWDMA2,
|
2008-07-25 04:53:32 +08:00
|
|
|
},
|
|
|
|
{ /* 1: HPT3xx */
|
|
|
|
.name = DRV_NAME,
|
2005-04-17 06:20:36 +08:00
|
|
|
.init_chipset = init_chipset_hpt366,
|
|
|
|
.init_hwif = init_hwif_hpt366,
|
|
|
|
.init_dma = init_dma_hpt366,
|
hpt366: init code rewrite
Finally, rework the driver init. code to correctly handle all the chip
variants HighPoint has created so far. This should cure the rest of the
timing issues in the driver (especially, on 66 MHz PCI) caused by the
HighPoint's habit of switching the base DPLL clock with every new revision
of the chips...
- switch to using the enumeration type to differ between the numerous chip
variants, matching PCI device/revision ID with the chip type early, at the
init_setup stage;
- extend the hpt_info structure to hold the DPLL and PCI clock frequencies,
stop duplicating it for each channel by storing the pointer in the pci_dev
structure: first, at the init_setup stage, point it to a static "template"
with only the chip type and its specific base DPLL frequency, the highest
supported DMA mode, and the chip settings table pointer filled, then, at
the init_chipset stage, allocate per-chip instance and fill it with the
rest of the necessary information;
- get rid of the constant thresholds in the HPT37x PCI clock detection code,
switch to calculating PCI clock frequency based on the chip's base DPLL
frequency;
- switch to using the DPLL clock and enable UltraATA/133 mode by default on
anything newer than HPT370/A;
- fold PCI clock detection and DPLL setup code into init_chipset_hpt366(),
unify the HPT36x/37x setup code and the speedproc handlers by joining the
register setting lists into the table indexed by the clock selected;
- add enablebits for all the chips to avoid touching disabled channels
(though the HighPoint BIOS seem to only disable the primary one on
HPT371/N);
- separate the UltraDMA and MWDMA masks there to avoid changing PIO timings
when setting an UltraDMA mode in hpt37x_tune_chipset().
This version has been tested on HPT370/302/371N.
Thanks to Alan for the inspiration. Hopefully, his libata driver will also
benefit from the work done on this "obsolete" driver...
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
2007-02-08 01:18:16 +08:00
|
|
|
.enablebits = {{0x50,0x04,0x04}, {0x54,0x04,0x04}},
|
2008-04-27 04:25:14 +08:00
|
|
|
.port_ops = &hpt3xx_port_ops,
|
2008-04-27 04:25:24 +08:00
|
|
|
.dma_ops = &hpt37x_dma_ops,
|
2008-01-26 05:17:18 +08:00
|
|
|
.host_flags = IDE_HFLAGS_HPT3XX,
|
2007-07-20 07:11:59 +08:00
|
|
|
.pio_mask = ATA_PIO4,
|
2007-10-19 06:30:07 +08:00
|
|
|
.mwdma_mask = ATA_MWDMA2,
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
};
|
|
|
|
|
|
|
|
/**
|
|
|
|
* hpt366_init_one - called when an HPT366 is found
|
|
|
|
* @dev: the hpt366 device
|
|
|
|
* @id: the matching pci id
|
|
|
|
*
|
|
|
|
* Called when the PCI registration layer (or the IDE initialization)
|
|
|
|
* finds a device matching our IDE device tables.
|
|
|
|
*/
|
2012-12-22 05:21:03 +08:00
|
|
|
static int hpt366_init_one(struct pci_dev *dev, const struct pci_device_id *id)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2007-10-27 02:31:15 +08:00
|
|
|
const struct hpt_info *info = NULL;
|
2008-07-25 04:53:15 +08:00
|
|
|
struct hpt_info *dyn_info;
|
2007-10-19 06:30:09 +08:00
|
|
|
struct pci_dev *dev2 = NULL;
|
2007-10-20 06:32:34 +08:00
|
|
|
struct ide_port_info d;
|
2007-10-19 06:30:09 +08:00
|
|
|
u8 idx = id->driver_data;
|
|
|
|
u8 rev = dev->revision;
|
2008-07-25 04:53:15 +08:00
|
|
|
int ret;
|
2007-10-19 06:30:09 +08:00
|
|
|
|
|
|
|
if ((idx == 0 || idx == 4) && (PCI_FUNC(dev->devfn) & 1))
|
|
|
|
return -ENODEV;
|
|
|
|
|
|
|
|
switch (idx) {
|
|
|
|
case 0:
|
|
|
|
if (rev < 3)
|
|
|
|
info = &hpt36x;
|
|
|
|
else {
|
2008-02-27 04:50:33 +08:00
|
|
|
switch (min_t(u8, rev, 6)) {
|
|
|
|
case 3: info = &hpt370; break;
|
|
|
|
case 4: info = &hpt370a; break;
|
|
|
|
case 5: info = &hpt372; break;
|
|
|
|
case 6: info = &hpt372n; break;
|
|
|
|
}
|
2007-10-19 06:30:09 +08:00
|
|
|
idx++;
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
case 1:
|
|
|
|
info = (rev > 1) ? &hpt372n : &hpt372a;
|
|
|
|
break;
|
|
|
|
case 2:
|
|
|
|
info = (rev > 1) ? &hpt302n : &hpt302;
|
|
|
|
break;
|
|
|
|
case 3:
|
|
|
|
hpt371_init(dev);
|
|
|
|
info = (rev > 1) ? &hpt371n : &hpt371;
|
|
|
|
break;
|
|
|
|
case 4:
|
|
|
|
info = &hpt374;
|
|
|
|
break;
|
|
|
|
case 5:
|
|
|
|
info = &hpt372n;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
2008-07-25 04:53:32 +08:00
|
|
|
printk(KERN_INFO DRV_NAME ": %s chipset detected\n", info->chip_name);
|
|
|
|
|
|
|
|
d = hpt366_chipsets[min_t(u8, idx, 1)];
|
2007-10-19 06:30:09 +08:00
|
|
|
|
|
|
|
d.udma_mask = info->udma_mask;
|
|
|
|
|
2008-04-27 04:25:24 +08:00
|
|
|
/* fixup ->dma_ops for HPT370/HPT370A */
|
|
|
|
if (info == &hpt370 || info == &hpt370a)
|
|
|
|
d.dma_ops = &hpt370_dma_ops;
|
|
|
|
|
2007-10-19 06:30:09 +08:00
|
|
|
if (info == &hpt36x || info == &hpt374)
|
|
|
|
dev2 = pci_get_slot(dev->bus, dev->devfn + 1);
|
|
|
|
|
treewide: kzalloc() -> kcalloc()
The kzalloc() function has a 2-factor argument form, kcalloc(). This
patch replaces cases of:
kzalloc(a * b, gfp)
with:
kcalloc(a * b, gfp)
as well as handling cases of:
kzalloc(a * b * c, gfp)
with:
kzalloc(array3_size(a, b, c), gfp)
as it's slightly less ugly than:
kzalloc_array(array_size(a, b), c, gfp)
This does, however, attempt to ignore constant size factors like:
kzalloc(4 * 1024, gfp)
though any constants defined via macros get caught up in the conversion.
Any factors with a sizeof() of "unsigned char", "char", and "u8" were
dropped, since they're redundant.
The Coccinelle script used for this was:
// Fix redundant parens around sizeof().
@@
type TYPE;
expression THING, E;
@@
(
kzalloc(
- (sizeof(TYPE)) * E
+ sizeof(TYPE) * E
, ...)
|
kzalloc(
- (sizeof(THING)) * E
+ sizeof(THING) * E
, ...)
)
// Drop single-byte sizes and redundant parens.
@@
expression COUNT;
typedef u8;
typedef __u8;
@@
(
kzalloc(
- sizeof(u8) * (COUNT)
+ COUNT
, ...)
|
kzalloc(
- sizeof(__u8) * (COUNT)
+ COUNT
, ...)
|
kzalloc(
- sizeof(char) * (COUNT)
+ COUNT
, ...)
|
kzalloc(
- sizeof(unsigned char) * (COUNT)
+ COUNT
, ...)
|
kzalloc(
- sizeof(u8) * COUNT
+ COUNT
, ...)
|
kzalloc(
- sizeof(__u8) * COUNT
+ COUNT
, ...)
|
kzalloc(
- sizeof(char) * COUNT
+ COUNT
, ...)
|
kzalloc(
- sizeof(unsigned char) * COUNT
+ COUNT
, ...)
)
// 2-factor product with sizeof(type/expression) and identifier or constant.
@@
type TYPE;
expression THING;
identifier COUNT_ID;
constant COUNT_CONST;
@@
(
- kzalloc
+ kcalloc
(
- sizeof(TYPE) * (COUNT_ID)
+ COUNT_ID, sizeof(TYPE)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(TYPE) * COUNT_ID
+ COUNT_ID, sizeof(TYPE)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(TYPE) * (COUNT_CONST)
+ COUNT_CONST, sizeof(TYPE)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(TYPE) * COUNT_CONST
+ COUNT_CONST, sizeof(TYPE)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(THING) * (COUNT_ID)
+ COUNT_ID, sizeof(THING)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(THING) * COUNT_ID
+ COUNT_ID, sizeof(THING)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(THING) * (COUNT_CONST)
+ COUNT_CONST, sizeof(THING)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(THING) * COUNT_CONST
+ COUNT_CONST, sizeof(THING)
, ...)
)
// 2-factor product, only identifiers.
@@
identifier SIZE, COUNT;
@@
- kzalloc
+ kcalloc
(
- SIZE * COUNT
+ COUNT, SIZE
, ...)
// 3-factor product with 1 sizeof(type) or sizeof(expression), with
// redundant parens removed.
@@
expression THING;
identifier STRIDE, COUNT;
type TYPE;
@@
(
kzalloc(
- sizeof(TYPE) * (COUNT) * (STRIDE)
+ array3_size(COUNT, STRIDE, sizeof(TYPE))
, ...)
|
kzalloc(
- sizeof(TYPE) * (COUNT) * STRIDE
+ array3_size(COUNT, STRIDE, sizeof(TYPE))
, ...)
|
kzalloc(
- sizeof(TYPE) * COUNT * (STRIDE)
+ array3_size(COUNT, STRIDE, sizeof(TYPE))
, ...)
|
kzalloc(
- sizeof(TYPE) * COUNT * STRIDE
+ array3_size(COUNT, STRIDE, sizeof(TYPE))
, ...)
|
kzalloc(
- sizeof(THING) * (COUNT) * (STRIDE)
+ array3_size(COUNT, STRIDE, sizeof(THING))
, ...)
|
kzalloc(
- sizeof(THING) * (COUNT) * STRIDE
+ array3_size(COUNT, STRIDE, sizeof(THING))
, ...)
|
kzalloc(
- sizeof(THING) * COUNT * (STRIDE)
+ array3_size(COUNT, STRIDE, sizeof(THING))
, ...)
|
kzalloc(
- sizeof(THING) * COUNT * STRIDE
+ array3_size(COUNT, STRIDE, sizeof(THING))
, ...)
)
// 3-factor product with 2 sizeof(variable), with redundant parens removed.
@@
expression THING1, THING2;
identifier COUNT;
type TYPE1, TYPE2;
@@
(
kzalloc(
- sizeof(TYPE1) * sizeof(TYPE2) * COUNT
+ array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2))
, ...)
|
kzalloc(
- sizeof(TYPE1) * sizeof(THING2) * (COUNT)
+ array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2))
, ...)
|
kzalloc(
- sizeof(THING1) * sizeof(THING2) * COUNT
+ array3_size(COUNT, sizeof(THING1), sizeof(THING2))
, ...)
|
kzalloc(
- sizeof(THING1) * sizeof(THING2) * (COUNT)
+ array3_size(COUNT, sizeof(THING1), sizeof(THING2))
, ...)
|
kzalloc(
- sizeof(TYPE1) * sizeof(THING2) * COUNT
+ array3_size(COUNT, sizeof(TYPE1), sizeof(THING2))
, ...)
|
kzalloc(
- sizeof(TYPE1) * sizeof(THING2) * (COUNT)
+ array3_size(COUNT, sizeof(TYPE1), sizeof(THING2))
, ...)
)
// 3-factor product, only identifiers, with redundant parens removed.
@@
identifier STRIDE, SIZE, COUNT;
@@
(
kzalloc(
- (COUNT) * STRIDE * SIZE
+ array3_size(COUNT, STRIDE, SIZE)
, ...)
|
kzalloc(
- COUNT * (STRIDE) * SIZE
+ array3_size(COUNT, STRIDE, SIZE)
, ...)
|
kzalloc(
- COUNT * STRIDE * (SIZE)
+ array3_size(COUNT, STRIDE, SIZE)
, ...)
|
kzalloc(
- (COUNT) * (STRIDE) * SIZE
+ array3_size(COUNT, STRIDE, SIZE)
, ...)
|
kzalloc(
- COUNT * (STRIDE) * (SIZE)
+ array3_size(COUNT, STRIDE, SIZE)
, ...)
|
kzalloc(
- (COUNT) * STRIDE * (SIZE)
+ array3_size(COUNT, STRIDE, SIZE)
, ...)
|
kzalloc(
- (COUNT) * (STRIDE) * (SIZE)
+ array3_size(COUNT, STRIDE, SIZE)
, ...)
|
kzalloc(
- COUNT * STRIDE * SIZE
+ array3_size(COUNT, STRIDE, SIZE)
, ...)
)
// Any remaining multi-factor products, first at least 3-factor products,
// when they're not all constants...
@@
expression E1, E2, E3;
constant C1, C2, C3;
@@
(
kzalloc(C1 * C2 * C3, ...)
|
kzalloc(
- (E1) * E2 * E3
+ array3_size(E1, E2, E3)
, ...)
|
kzalloc(
- (E1) * (E2) * E3
+ array3_size(E1, E2, E3)
, ...)
|
kzalloc(
- (E1) * (E2) * (E3)
+ array3_size(E1, E2, E3)
, ...)
|
kzalloc(
- E1 * E2 * E3
+ array3_size(E1, E2, E3)
, ...)
)
// And then all remaining 2 factors products when they're not all constants,
// keeping sizeof() as the second factor argument.
@@
expression THING, E1, E2;
type TYPE;
constant C1, C2, C3;
@@
(
kzalloc(sizeof(THING) * C2, ...)
|
kzalloc(sizeof(TYPE) * C2, ...)
|
kzalloc(C1 * C2 * C3, ...)
|
kzalloc(C1 * C2, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(TYPE) * (E2)
+ E2, sizeof(TYPE)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(TYPE) * E2
+ E2, sizeof(TYPE)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(THING) * (E2)
+ E2, sizeof(THING)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(THING) * E2
+ E2, sizeof(THING)
, ...)
|
- kzalloc
+ kcalloc
(
- (E1) * E2
+ E1, E2
, ...)
|
- kzalloc
+ kcalloc
(
- (E1) * (E2)
+ E1, E2
, ...)
|
- kzalloc
+ kcalloc
(
- E1 * E2
+ E1, E2
, ...)
)
Signed-off-by: Kees Cook <keescook@chromium.org>
2018-06-13 05:03:40 +08:00
|
|
|
dyn_info = kcalloc(dev2 ? 2 : 1, sizeof(*dyn_info), GFP_KERNEL);
|
2008-07-25 04:53:15 +08:00
|
|
|
if (dyn_info == NULL) {
|
2008-07-25 04:53:31 +08:00
|
|
|
printk(KERN_ERR "%s %s: out of memory!\n",
|
|
|
|
d.name, pci_name(dev));
|
2008-07-25 04:53:15 +08:00
|
|
|
pci_dev_put(dev2);
|
|
|
|
return -ENOMEM;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Copy everything from a static "template" structure
|
|
|
|
* to just allocated per-chip hpt_info structure.
|
|
|
|
*/
|
|
|
|
memcpy(dyn_info, info, sizeof(*dyn_info));
|
2007-10-19 06:30:09 +08:00
|
|
|
|
2008-07-25 04:53:15 +08:00
|
|
|
if (dev2) {
|
|
|
|
memcpy(dyn_info + 1, info, sizeof(*dyn_info));
|
2007-10-19 06:30:09 +08:00
|
|
|
|
|
|
|
if (info == &hpt374)
|
|
|
|
hpt374_init(dev, dev2);
|
|
|
|
else {
|
|
|
|
if (hpt36x_init(dev, dev2))
|
2008-04-26 23:36:35 +08:00
|
|
|
d.host_flags &= ~IDE_HFLAG_NON_BOOTABLE;
|
2007-10-19 06:30:09 +08:00
|
|
|
}
|
|
|
|
|
2008-07-25 04:53:15 +08:00
|
|
|
ret = ide_pci_init_two(dev, dev2, &d, dyn_info);
|
|
|
|
if (ret < 0) {
|
2007-10-19 06:30:09 +08:00
|
|
|
pci_dev_put(dev2);
|
2008-07-25 04:53:15 +08:00
|
|
|
kfree(dyn_info);
|
|
|
|
}
|
2007-10-19 06:30:09 +08:00
|
|
|
return ret;
|
|
|
|
}
|
2005-04-17 06:20:36 +08:00
|
|
|
|
2008-07-25 04:53:15 +08:00
|
|
|
ret = ide_pci_init_one(dev, &d, dyn_info);
|
|
|
|
if (ret < 0)
|
|
|
|
kfree(dyn_info);
|
|
|
|
|
|
|
|
return ret;
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static void hpt366_remove(struct pci_dev *dev)
|
2008-07-25 04:53:21 +08:00
|
|
|
{
|
|
|
|
struct ide_host *host = pci_get_drvdata(dev);
|
|
|
|
struct ide_info *info = host->host_priv;
|
|
|
|
struct pci_dev *dev2 = host->dev[1] ? to_pci_dev(host->dev[1]) : NULL;
|
|
|
|
|
|
|
|
ide_pci_remove(dev);
|
|
|
|
pci_dev_put(dev2);
|
|
|
|
kfree(info);
|
|
|
|
}
|
|
|
|
|
2012-12-22 05:21:03 +08:00
|
|
|
static const struct pci_device_id hpt366_pci_tbl[] = {
|
2007-10-17 04:29:56 +08:00
|
|
|
{ PCI_VDEVICE(TTI, PCI_DEVICE_ID_TTI_HPT366), 0 },
|
|
|
|
{ PCI_VDEVICE(TTI, PCI_DEVICE_ID_TTI_HPT372), 1 },
|
|
|
|
{ PCI_VDEVICE(TTI, PCI_DEVICE_ID_TTI_HPT302), 2 },
|
|
|
|
{ PCI_VDEVICE(TTI, PCI_DEVICE_ID_TTI_HPT371), 3 },
|
|
|
|
{ PCI_VDEVICE(TTI, PCI_DEVICE_ID_TTI_HPT374), 4 },
|
|
|
|
{ PCI_VDEVICE(TTI, PCI_DEVICE_ID_TTI_HPT372N), 5 },
|
2005-04-17 06:20:36 +08:00
|
|
|
{ 0, },
|
|
|
|
};
|
|
|
|
MODULE_DEVICE_TABLE(pci, hpt366_pci_tbl);
|
|
|
|
|
2008-10-14 03:39:41 +08:00
|
|
|
static struct pci_driver hpt366_pci_driver = {
|
2005-04-17 06:20:36 +08:00
|
|
|
.name = "HPT366_IDE",
|
|
|
|
.id_table = hpt366_pci_tbl,
|
|
|
|
.probe = hpt366_init_one,
|
2012-12-22 05:21:03 +08:00
|
|
|
.remove = hpt366_remove,
|
2008-10-11 04:39:32 +08:00
|
|
|
.suspend = ide_pci_suspend,
|
|
|
|
.resume = ide_pci_resume,
|
2005-04-17 06:20:36 +08:00
|
|
|
};
|
|
|
|
|
2007-01-27 20:46:56 +08:00
|
|
|
static int __init hpt366_ide_init(void)
|
2005-04-17 06:20:36 +08:00
|
|
|
{
|
2008-10-14 03:39:41 +08:00
|
|
|
return ide_pci_register_driver(&hpt366_pci_driver);
|
2005-04-17 06:20:36 +08:00
|
|
|
}
|
|
|
|
|
2008-07-25 04:53:21 +08:00
|
|
|
static void __exit hpt366_ide_exit(void)
|
|
|
|
{
|
2008-10-14 03:39:41 +08:00
|
|
|
pci_unregister_driver(&hpt366_pci_driver);
|
2008-07-25 04:53:21 +08:00
|
|
|
}
|
|
|
|
|
2005-04-17 06:20:36 +08:00
|
|
|
module_init(hpt366_ide_init);
|
2008-07-25 04:53:21 +08:00
|
|
|
module_exit(hpt366_ide_exit);
|
2005-04-17 06:20:36 +08:00
|
|
|
|
|
|
|
MODULE_AUTHOR("Andre Hedrick");
|
|
|
|
MODULE_DESCRIPTION("PCI driver module for Highpoint HPT366 IDE");
|
|
|
|
MODULE_LICENSE("GPL");
|