GitHub/LineageOS/G12/android_kernel_amlogic_linux-4.9.git
16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/bart/ide-2.6
Linus Torvalds [Tue, 15 Jul 2008 18:15:36 +0000 (11:15 -0700)]
Merge git://git./linux/kernel/git/bart/ide-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/bart/ide-2.6: (80 commits)
  ide-floppy: fix unfortunate function naming
  ide-tape: unify idetape_create_read/write_cmd
  ide: add ide_pc_intr() helper
  ide-{floppy,scsi}: read Status Register before stopping DMA engine
  ide-scsi: add more debugging to idescsi_pc_intr()
  ide-scsi: use pc->callback
  ide-floppy: add more debugging to idefloppy_pc_intr()
  ide-tape: always log debug info in idetape_pc_intr() if debugging is enabled
  ide-tape: add ide_tape_io_buffers() helper
  ide-tape: factor out DSC handling from idetape_pc_intr()
  ide-{floppy,tape}: move checking of ->failed_pc to ->callback
  ide: add ide_issue_pc() helper
  ide: add PC_FLAG_DRQ_INTERRUPT pc flag
  ide-scsi: move idescsi_map_sg() call out from idescsi_issue_pc()
  ide: add ide_transfer_pc() helper
  ide-scsi: set drive->scsi flag for devices handled by the driver
  ide-{cd,floppy,tape}: remove checking for drive->scsi
  ide: add PC_FLAG_ZIP_DRIVE pc flag
  ide-tape: factor out waiting for good ireason from idetape_transfer_pc()
  ide-tape: set PC_FLAG_DMA_IN_PROGRESS flag in idetape_transfer_pc()
  ...

16 years agoide-floppy: fix unfortunate function naming
Borislav Petkov [Tue, 15 Jul 2008 19:22:03 +0000 (21:22 +0200)]
ide-floppy: fix unfortunate function naming

mv idefloppy_transfer_pc1 idefloppy_start_pc_transfer
mv idefloppy_transfer_pc2 idefloppy_transfer_pc

which describes their functionality and disambiguates them. There should be no
functionality change introduced by this patch.

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-tape: unify idetape_create_read/write_cmd
Borislav Petkov [Tue, 15 Jul 2008 19:22:03 +0000 (21:22 +0200)]
ide-tape: unify idetape_create_read/write_cmd

A straightforward one. There should be no functional change resulting from this
change.

[bart: minor fixups]

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: add ide_pc_intr() helper
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:22:03 +0000 (21:22 +0200)]
ide: add ide_pc_intr() helper

* ide-tape.c: add 'drive' argument to idetape_update_buffers().

* Add generic ide_pc_intr() helper to ide-atapi.c and then
  convert ide-{floppy,tape,scsi} device drivers to use it.

* ide-tape.c: remove no longer needed DBG_PC_INTR.

There should be no functional changes caused by this patch
(unless the debugging is explicitely compiled in).

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-{floppy,scsi}: read Status Register before stopping DMA engine
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:22:03 +0000 (21:22 +0200)]
ide-{floppy,scsi}: read Status Register before stopping DMA engine

Read Status Register before stopping DMA engine to match ide-tape
device driver - it should be safe and shouldn't affect anything.

This is a preparation for adding generic ide_pc_intr() helper.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-scsi: add more debugging to idescsi_pc_intr()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:22:02 +0000 (21:22 +0200)]
ide-scsi: add more debugging to idescsi_pc_intr()

Add more debugging to idescsi_pc_intr() to match ide-tape's
idetape_pc_intr().

While at it:

* Correct the first debug message.

This is a preparation for adding generic ide_pc_intr() helper.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-scsi: use pc->callback
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:22:02 +0000 (21:22 +0200)]
ide-scsi: use pc->callback

* Add ide_scsi_callback() pc->callback implementation, then update
  idescsi_check_condition() and idescsi_queue() to setup ->callback.

* Convert idescsi_pc_intr() to use pc->callback.

This is a preparation for adding generic ide_pc_intr() helper.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-floppy: add more debugging to idefloppy_pc_intr()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:22:02 +0000 (21:22 +0200)]
ide-floppy: add more debugging to idefloppy_pc_intr()

Add more debugging to idefloppy_pc_intr() to match ide-tape's
idetape_pc_intr().

While at it:

* Correct the first debug message.

* Log device name with "DMA finished" message.

This is a preparation for adding generic ide_pc_intr() helper.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-tape: always log debug info in idetape_pc_intr() if debugging is enabled
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:22:01 +0000 (21:22 +0200)]
ide-tape: always log debug info in idetape_pc_intr() if debugging is enabled

Add DBG_PC_INTR debug level and use it to always log debug info
in idetape_pc_intr() if debugging is enabled.

While at it:

* Use drive->name instead of tape->name.

* Log device name with "DMA finished" message.

This is a preparation for adding generic ide_pc_intr() helper.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-tape: add ide_tape_io_buffers() helper
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:22:01 +0000 (21:22 +0200)]
ide-tape: add ide_tape_io_buffers() helper

* Add ide_tape_io_buffers() helper which is a wrapper for
  idetape_{in,out}put_buffers() and convert idetape_pc_intr()
  to use it.

* Remove no longer used idetape_io_buf typedef.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>---
16 years agoide-tape: factor out DSC handling from idetape_pc_intr()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:22:01 +0000 (21:22 +0200)]
ide-tape: factor out DSC handling from idetape_pc_intr()

Factor out DSC handling from idetape_pc_intr() to ide_tape_handle_dsc()
helper as a preparation for adding generic ide_pc_intr() helper.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-{floppy,tape}: move checking of ->failed_pc to ->callback
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:22:01 +0000 (21:22 +0200)]
ide-{floppy,tape}: move checking of ->failed_pc to ->callback

Move checking/resetting of ->failed_pc from ide*_pc_intr() to ->callback
as a preparation for adding generic ide_pc_intr() helper.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>---
16 years agoide: add ide_issue_pc() helper
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:22:00 +0000 (21:22 +0200)]
ide: add ide_issue_pc() helper

Add generic ide_issue_pc() helper to ide-atapi.c and then
convert ide-{floppy,tape,scsi} device drivers to use it.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: add PC_FLAG_DRQ_INTERRUPT pc flag
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:59 +0000 (21:21 +0200)]
ide: add PC_FLAG_DRQ_INTERRUPT pc flag

Add PC_FLAG_DRQ_INTERRUPT pc flag, set it in ide*_do_request()
and check for it (instead of checking for IDE*_FLAG_DRQ_INTERRUPT)
in ide*_issue_pc().  This is a preparation for adding generic
ide_issue_pc() helper.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-scsi: move idescsi_map_sg() call out from idescsi_issue_pc()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:58 +0000 (21:21 +0200)]
ide-scsi: move idescsi_map_sg() call out from idescsi_issue_pc()

Move idescsi_map_sg() call out from idescsi_issue_pc()
to idescsi_do_request() as a preparation to adding generic
ide_issue_pc() helper.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: add ide_transfer_pc() helper
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:58 +0000 (21:21 +0200)]
ide: add ide_transfer_pc() helper

* Add ide-atapi.c file for generic ATAPI support together with
  CONFIG_IDE_ATAPI config option.

* Add generic ide_transfer_pc() helper to ide-atapi.c and then
  convert ide-{floppy,tape,scsi} device drivers to use it.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-scsi: set drive->scsi flag for devices handled by the driver
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:58 +0000 (21:21 +0200)]
ide-scsi: set drive->scsi flag for devices handled by the driver

This is a preparation for adding generic ide_transfer_pc() helper.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-{cd,floppy,tape}: remove checking for drive->scsi
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:57 +0000 (21:21 +0200)]
ide-{cd,floppy,tape}: remove checking for drive->scsi

Remove checking for drive->scsi which is no longer set by IDE core code
(leave the flag since it will be re-used for generic ATAPI support).

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: add PC_FLAG_ZIP_DRIVE pc flag
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:57 +0000 (21:21 +0200)]
ide: add PC_FLAG_ZIP_DRIVE pc flag

Add PC_FLAG_ZIP_DRIVE pc flag, set it in idefloppy_do_request()
and check for it (instead of checking for IDEFLOPPY_FLAG_ZIP_DRIVE)
in idefloppy_transfer_pc().  This is a preparation for adding
generic ide_transfer_pc() helper.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-tape: factor out waiting for good ireason from idetape_transfer_pc()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:57 +0000 (21:21 +0200)]
ide-tape: factor out waiting for good ireason from idetape_transfer_pc()

Factor out waiting for good ireason from idetape_transfer_pc()
to ide_tape_wait_ireason() as a preparation for adding generic
ide_transfer_pc() helper.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-tape: set PC_FLAG_DMA_IN_PROGRESS flag in idetape_transfer_pc()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:57 +0000 (21:21 +0200)]
ide-tape: set PC_FLAG_DMA_IN_PROGRESS flag in idetape_transfer_pc()

Set PC_FLAG_DMA_IN_PROGRESS flag in idetape_transfer_pc() instead of
idetape_issue_pc() to match the other ATAPI device drivers.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-floppy: start DMA engine in idefloppy_transfer_pc1()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:56 +0000 (21:21 +0200)]
ide-floppy: start DMA engine in idefloppy_transfer_pc1()

Start DMA engine and set PC_FLAG_DMA_IN_PROGRESS flag in
idefloppy_transfer_pc1() instead of idefloppy_issue_pc()
so the Status Register and the Interrupt Reason Register
are checked first.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-{floppy,tape}: PC_FLAG_DMA_RECOMMENDED -> PC_FLAG_DMA_OK
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:56 +0000 (21:21 +0200)]
ide-{floppy,tape}: PC_FLAG_DMA_RECOMMENDED -> PC_FLAG_DMA_OK

* Use PC_FLAG_DMA_OK flag instead of PC_FLAG_DMA_RECOMMENDED one.

* Remove no longer used PC_FLAG_DMA_RECOMMENDED flag.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-{floppy,tape}: merge pc->idefloppy_callback and pc->idetape_callback
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:56 +0000 (21:21 +0200)]
ide-{floppy,tape}: merge pc->idefloppy_callback and pc->idetape_callback

Merge pc->idefloppy_callback and pc->idetape_callback into pc->callback.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-floppy: merge callbacks
Borislav Petkov [Tue, 15 Jul 2008 19:21:56 +0000 (21:21 +0200)]
ide-floppy: merge callbacks

The appropriate functionality of the callback is established through querying
the ATAPI packet command in pc->c[0].

While at it, simplify if (floppy->failed_pc)-branch to be found in the original
idefloppy_request_sense_callback().

Bart:
- keep handling for blk_pc_request() requests unchanged + add FIXME
- add uptodate variable + leave just one idefloppy_end_request() call
- add newline to the debug message

Signed-off-by: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-tape: make pc->idetape_callback void
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:55 +0000 (21:21 +0200)]
ide-tape: make pc->idetape_callback void

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-tape: merge callbacks
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:55 +0000 (21:21 +0200)]
ide-tape: merge callbacks

The appropriate functionality of the callback is established through querying
the ATAPI packet command in pc->c[0].

While at it:
- add uptodate variable + leave just one idetape_end_request() call
- don't use HWGROUP() macro

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-tape: make idetape_retry_pc() void
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:55 +0000 (21:21 +0200)]
ide-tape: make idetape_retry_pc() void

idetape_retry_pc() always returns ide_stopped so make it void.

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-{floppy,tape,scsi}: log device name instead of driver name
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:54 +0000 (21:21 +0200)]
ide-{floppy,tape,scsi}: log device name instead of driver name

Log device name instead of driver name in *_pc_intr() and *_transfer_pc*().

While at it:

* Merge two consecutive printk()-s in *_pc_intr() together.

* Replace "floppy"/"tape"/"scsi" references in printk()-s by "device".

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-floppy: merge idefloppy_transfer_pc() and idefloppy_transfer_pc1()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:54 +0000 (21:21 +0200)]
ide-floppy: merge idefloppy_transfer_pc() and idefloppy_transfer_pc1()

* Check IDEFLOPPY_FLAG_ZIP_DRIVE flag in idefloppy_transfer_pc1()
  and skip idefloppy_transfer_pc2()-phase if the flag is not set.

* Always use idefloppy_transfer_pc1() in idefloppy_issue_pc()
  and remove no longer needed idefloppy_transfer_pc().

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-tape: remove SIMULATE_ERRORS debug code
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:54 +0000 (21:21 +0200)]
ide-tape: remove SIMULATE_ERRORS debug code

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-tape: remove stale comments from idetape_pc_intr()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:54 +0000 (21:21 +0200)]
ide-tape: remove stale comments from idetape_pc_intr()

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-tape: remove unneeded CONFIG_BLK_DEV_IDEDMA ifdef
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:54 +0000 (21:21 +0200)]
ide-tape: remove unneeded CONFIG_BLK_DEV_IDEDMA ifdef

PC_FLAG_DMA_IN_PROGRESS flag is never set if DMA support is disabled.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-tape: remove superfluous warning message from idetape_issue_pc()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:53 +0000 (21:21 +0200)]
ide-tape: remove superfluous warning message from idetape_issue_pc()

ide_dma_off() prints info about DMA being disabled.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-tape: remove superfluous error message from idetape_pc_intr()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:53 +0000 (21:21 +0200)]
ide-tape: remove superfluous error message from idetape_pc_intr()

ide_dma_off() prints info about DMA being disabled.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-tape: idetape_pc_intr() should use local_irq_enable_in_hardirq()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:53 +0000 (21:21 +0200)]
ide-tape: idetape_pc_intr() should use local_irq_enable_in_hardirq()

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-scsi: add debug_log() macro
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:52 +0000 (21:21 +0200)]
ide-scsi: add debug_log() macro

Add debug_log() macro and convert the driver to use it.

[ This makes debug messages to be always prefixed with "ide-scsi: "
  and use KERN_INFO level. ]

While at it:

* Change "DMA complete" debug message to "DMA finished" to match
  other ATAPI device drivers.

* Use __func__.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-scsi: remove superfluous BUG_ON() from idescsi_transfer_pc()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:52 +0000 (21:21 +0200)]
ide-scsi: remove superfluous BUG_ON() from idescsi_transfer_pc()

ide_set_handler() bugs on ->handler == NULL so no need to do it
in idescsi_transfer_pc().

There should be no functional changes caused by this patch.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-scsi: merge idescsi_input_buffers() and idescsi_output_buffers()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:52 +0000 (21:21 +0200)]
ide-scsi: merge idescsi_input_buffers() and idescsi_output_buffers()

* Merge idescsi_input_buffers() and idescsi_output_buffers()
  into ide_scsi_io_buffers() helper.

While at it:

* Log device name instead of driver name on error.

* Use xfer_func_t.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-scsi: fix Interrupt Reason checking in idescsi_pc_intr()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:52 +0000 (21:21 +0200)]
ide-scsi: fix Interrupt Reason checking in idescsi_pc_intr()

Set PC_FLAG_WRITING pc flag in idescsi_queue() (if needed)
and then fix Interrupt Reason checking in idescsi_pc_intr().

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-scsi: fix handling of DMA errors in idescsi_pc_intr()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:51 +0000 (21:21 +0200)]
ide-scsi: fix handling of DMA errors in idescsi_pc_intr()

Check return value of ->dma_end method and if there was a DMA error
handle it accordingly (set PC_FLAG_DMA_ERROR pc flag, don't update
pc->xferred and increase rq->errors).

Also move debug message in the right place while at it.

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-scsi: fix DRQ checking for DMA transfers in idescsi_pc_intr()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:51 +0000 (21:21 +0200)]
ide-scsi: fix DRQ checking for DMA transfers in idescsi_pc_intr()

If DRQ bit of Status Register is not cleared it is an error condition
and should be handled accordingly (disable DMA + reset the device).

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-scsi: fix race in idescsi_transfer_pc()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:51 +0000 (21:21 +0200)]
ide-scsi: fix race in idescsi_transfer_pc()

Start DMA engine before sending content of packet command (otherwise
it is possible that IRQ will happen before DMA engine is started).

Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: remove action argument in ide_do_drive_cmd
FUJITA Tomonori [Tue, 15 Jul 2008 19:21:51 +0000 (21:21 +0200)]
ide: remove action argument in ide_do_drive_cmd

ide_do_drive_cmd is called only with ide_preempt action argument. So
we can remove the action argument in ide_do_drive_cmd and ide_action_t
typedef.

This patch also includes two minor cleanups: 1) ide_do_drive_cmd
always succeeds so we don't need the return value; 2) the callers use
blk_rq_init before ide_do_drive_cmd so there is no need to initialize
rq->errors.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-scsi: replace ide_do_drive_cmd with blk_execute_rq_nowait
FUJITA Tomonori [Tue, 15 Jul 2008 19:21:50 +0000 (21:21 +0200)]
ide-scsi: replace ide_do_drive_cmd with blk_execute_rq_nowait

All the callers of ide_do_drive_cmd() except for ide-scsi use
ide_preempt action argument. This converts ide-scsi to use
blk_execute_rq_nowait instead of ide_do_drive_cmd so that we can
remove the action argument in ide_do_drive_cmd and ide_action_t
typedef.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: remove drive->ctl
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:50 +0000 (21:21 +0200)]
ide: remove drive->ctl

Remove drive->ctl (it is always equal to 0x08 after init time).

While at it:

* Use ATA_DEVCTL_OBS define.

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: use ->OUTBSYNC in ide_set_irq()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:50 +0000 (21:21 +0200)]
ide: use ->OUTBSYNC in ide_set_irq()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: use ->OUTBSYNC in init_irq()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:50 +0000 (21:21 +0200)]
ide: use ->OUTBSYNC in init_irq()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: pass 'hwif *' instead of 'drive *' to ->OUTBSYNC method
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:49 +0000 (21:21 +0200)]
ide: pass 'hwif *' instead of 'drive *' to ->OUTBSYNC method

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: remove ->mmio flag from ide_hwif_t
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:49 +0000 (21:21 +0200)]
ide: remove ->mmio flag from ide_hwif_t

Since scc_pata host driver no longer uses IDE PCI layer / ide_dma_setup()
and all other ->mmio users set also IDE_HFLAG_MMIO host flag we can safely
remove ->mmio flag.

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: move IRQ unmasking out from ->tf_load method
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:48 +0000 (21:21 +0200)]
ide: move IRQ unmasking out from ->tf_load method

Move IRQ unmasking out from ->tf_load method to its users.

There should be no functional changes caused by this patch
(SELECT_MASK() is NOP except for hpt366, icside and sgiioc4).

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: remove IDE_TFLAG_NO_SELECT_MASK taskfile flag
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:48 +0000 (21:21 +0200)]
ide: remove IDE_TFLAG_NO_SELECT_MASK taskfile flag

Always call SELECT_MASK(..., 0) in ide_tf_load() (needs to be done
to match ide_set_irq(..., 1)) and then remove IDE_TFLAG_NO_SELECT_MASK
taskfile flag.

This change should only affect hpt366 and icside host drivers since
->maskproc(..., 0) for sgiioc4 is equivalent to ide_set_irq(..., 1).

Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: fix do_probe() to use SELECT_DRIVE()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:48 +0000 (21:21 +0200)]
ide: fix do_probe() to use SELECT_DRIVE()

Fix one place in do_probe() which used ->OUTB directly instead
of calling SELECT_DRIVE() (so ->selectproc method is also used).

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: remove commented out code from ide_config_drive_speed()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:47 +0000 (21:21 +0200)]
ide: remove commented out code from ide_config_drive_speed()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agosgiioc4: use ->extra_base instead of ->dma_status for dma_handle
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:47 +0000 (21:21 +0200)]
sgiioc4: use ->extra_base instead of ->dma_status for dma_handle

This is a preparation for removal of ->dma_status field from ide_hwif_t.

There should be no functional changes caused by this patch.

Acked-by: Jeremy Higdon <jeremy@sgi.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: remove try_to_flush_leftover_data()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:47 +0000 (21:21 +0200)]
ide: remove try_to_flush_leftover_data()

Just use the new & shiny ide_pad_transfer() helper instead.

Also remove the superfluous check for 'drive->media == ide_disk'
while at it (ide_ata_error() is used only for ide_disk devices).

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: remove obsoleted "ide=" kernel parameters
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:47 +0000 (21:21 +0200)]
ide: remove obsoleted "ide=" kernel parameters

* Remove obsoleted "ide=" kernel parameters.

* Remove no longer needed:
  - ide_setup()
  - parse_options()
  - __setup("", ...)
  - module_param(options, ...)

* Use module_{init,exit}() for MODULE=y case and remove MODULE ifdef.

* Make ide_*acpi* and ide_doubler variables static.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: cleanup init_ide_data()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:46 +0000 (21:21 +0200)]
ide: cleanup init_ide_data()

* Remove no longer need init_ide_data() call from ide_setup().

* Cleanup init_ide_data().

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: remove obsoleted "hdx=" kernel parameters
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:46 +0000 (21:21 +0200)]
ide: remove obsoleted "hdx=" kernel parameters

* Remove obsoleted "hdx=" kernel parameters.

* Remove no longer used stridx() and match_parm().

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: remove obsoleted "idebus=" kernel parameter
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:46 +0000 (21:21 +0200)]
ide: remove obsoleted "idebus=" kernel parameter

* Remove obsoleted "idebus=" kernel parameter.

* Remove no longer needed ide_system_bus_speed() and system_bus_clock()
  (together with idebus_parameter and system_bus_speed variables).

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoblock: unexport blk_end_sync_rq
FUJITA Tomonori [Tue, 15 Jul 2008 19:21:45 +0000 (21:21 +0200)]
block: unexport blk_end_sync_rq

All the users of blk_end_sync_rq has gone (they are converted to use
blk_execute_rq). This unexports blk_end_sync_rq.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoblock: remove the checking for NULL queue in blk_put_request
FUJITA Tomonori [Tue, 15 Jul 2008 19:21:45 +0000 (21:21 +0200)]
block: remove the checking for NULL queue in blk_put_request

Some uses blk_put_request asymmetrically, that is, they uses it with
requests that not allocated by blk_get_request. As a result,
blk_put_request has a hack to catch a NULL request_queue. Now such
callers are fixed (they use blk_get_request properly). So we can
safely remove the hack in blk_put_request.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoblock: convert pd_special_command to use blk_execute_rq
FUJITA Tomonori [Tue, 15 Jul 2008 19:21:45 +0000 (21:21 +0200)]
block: convert pd_special_command to use blk_execute_rq

pd_special_command uses blk_put_request with struct request on the
stack. As a result, blk_put_request needs a hack to catch a NULL
request_queue.  This converts pd_special_command to use
blk_execute_rq.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-cd: remove ide_cd_init_rq
FUJITA Tomonori [Tue, 15 Jul 2008 19:21:45 +0000 (21:21 +0200)]
ide-cd: remove ide_cd_init_rq

ide_cd_init_rq is not used by ide-cd_ioctl any more. Only
cdrom_queue_request_sense use it. This converts
cdrom_queue_request_sense to use blk_rq_init directly and removes
ide_cd_init_rq.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: remove ide_init_drive_cmd
FUJITA Tomonori [Tue, 15 Jul 2008 19:21:44 +0000 (21:21 +0200)]
ide: remove ide_init_drive_cmd

ide_init_drive_cmd just calls blk_rq_init. This converts the users of
ide_init_drive_cmd to use blk_rq_init directly and removes
ide_init_drive_cmd.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: remove ide_wait/head_wait path in ide_do_drive_cmd
FUJITA Tomonori [Tue, 15 Jul 2008 19:21:44 +0000 (21:21 +0200)]
ide: remove ide_wait/head_wait path in ide_do_drive_cmd

Now all the users of ide_do_drive_cmd using ide_wait/head_wait are
converted to use blk_execute_rq this removes the ide_wait/head_wait
path in ide_do_drive_cmd.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: convert ide_do_drive_cmd path to use blk_execute_rq
FUJITA Tomonori [Tue, 15 Jul 2008 19:21:44 +0000 (21:21 +0200)]
ide: convert ide_do_drive_cmd path to use blk_execute_rq

This converts the ide_do_drive_cmd path using ide_[head_]wait to use
blk_execute_rq.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoblock: handle blk_pm_resume_request() requests in blk_execute_rq_nowait()
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:43 +0000 (21:21 +0200)]
block: handle blk_pm_resume_request() requests in blk_execute_rq_nowait()

For blk_pm_resume_request() requests (which are used only by IDE subsystem
currently) the queue is stopped so we need to call ->request_fn explicitly.

Thanks to:
- Rafael for reporting/bisecting the bug
- Borislav/Rafael for testing the fix

This is a preparation for converting IDE to use blk_execute_rq().

Cc: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-tape: convert ide_do_drive_cmd path to use blk_execute_rq
FUJITA Tomonori [Tue, 15 Jul 2008 19:21:43 +0000 (21:21 +0200)]
ide-tape: convert ide_do_drive_cmd path to use blk_execute_rq

This converts the ide_do_drive_cmd path using ide_wait to use
blk_execute_rq.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-taskfile: convert ide_do_drive_cmd path to use blk_execute_rq
FUJITA Tomonori [Tue, 15 Jul 2008 19:21:43 +0000 (21:21 +0200)]
ide-taskfile: convert ide_do_drive_cmd path to use blk_execute_rq

This converts the ide_do_drive_cmd path using ide_wait to use
blk_execute_rq.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-floppy: convert ide_do_drive_cmd path to use blk_execute_rq
FUJITA Tomonori [Tue, 15 Jul 2008 19:21:43 +0000 (21:21 +0200)]
ide-floppy: convert ide_do_drive_cmd path to use blk_execute_rq

This converts the ide_do_drive_cmd path using ide_wait to use
blk_execute_rq.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-disk: convert ide_do_drive_cmd path to use blk_execute_rq
FUJITA Tomonori [Tue, 15 Jul 2008 19:21:43 +0000 (21:21 +0200)]
ide-disk: convert ide_do_drive_cmd path to use blk_execute_rq

This converts the ide_do_drive_cmd path using ide_wait to use
blk_execute_rq.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-cd: convert ide_do_drive_cmd path to use blk_execute_rq
FUJITA Tomonori [Tue, 15 Jul 2008 19:21:42 +0000 (21:21 +0200)]
ide-cd: convert ide_do_drive_cmd path to use blk_execute_rq

This converts the ide_do_drive_cmd path using ide_wait to use
blk_execute_rq.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-cd: convert ide_cd_queue_pc to use blk_execute_rq
FUJITA Tomonori [Tue, 15 Jul 2008 19:21:42 +0000 (21:21 +0200)]
ide-cd: convert ide_cd_queue_pc to use blk_execute_rq

This converts ide_cd_queue_pc to use blk_execute_rq, necessitating
changing the ide_cd_queue_pc prototype into a form that doesn't takes
a pointer to request struct. ide_cd_queue_pc works like scsi_execute.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: use __generic_unplug_device() in ide_do_drive_cmd() (take 2)
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:42 +0000 (21:21 +0200)]
ide: use __generic_unplug_device() in ide_do_drive_cmd() (take 2)

* Call __elv_add_request() with 'plug' == 1 (so the device will be
  plugged) and then use __generic_unplug_device() instead of calling
  ide_do_request() directly.

v2:
* For blk_pm_resume_request() requests the queue is stopped so we
  need to call ->request_fn explicitly.

  Thanks to:
  - Rafael for reporting/bisecting the bug
  - Borislav/Rafael for testing the fix

This is a preparation for converting IDE to use blk_execute_rq().

Cc: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: set REQ_PREEMPT request flag in ide_do_drive_cmd() users
Bartlomiej Zolnierkiewicz [Tue, 15 Jul 2008 19:21:41 +0000 (21:21 +0200)]
ide: set REQ_PREEMPT request flag in ide_do_drive_cmd() users

* Set REQ_PREEMPT request flag in ide_do_drive_cmd() users
  for ide_preempt and ide_head_wait action types.

* Remove setting REQ_PREEMPT from ide_do_drive_cmd().

While at it:

* Set 'where' variable outside ide_lock.

This is a preparation for converting IDE to use blk_execute_rq().

There should be no functional changes caused by this patch.

Cc: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Borislav Petkov <petkovbb@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: use get_unaligned_* helpers
Harvey Harrison [Tue, 15 Jul 2008 19:21:41 +0000 (21:21 +0200)]
ide: use get_unaligned_* helpers

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide-tape: use clamp_t() rather than nested min_t()/max_t()
Harvey Harrison [Tue, 15 Jul 2008 19:21:41 +0000 (21:21 +0200)]
ide-tape: use clamp_t() rather than nested min_t()/max_t()

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: eliminate FIT() macro
Harvey Harrison [Tue, 15 Jul 2008 19:21:41 +0000 (21:21 +0200)]
ide: eliminate FIT() macro

Replace the FIT() macro with the kernel-provided clamp_val() macro.

FIT was always being called with a member of a struct ide_timing,
which are shorts, and constant constraints for the min and max.

Thus we can use clamp_val, rather than clamp_t.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoide: disable drive interrupts in ide_driveid_update()
Mikhail Cherkashin [Tue, 15 Jul 2008 19:21:40 +0000 (21:21 +0200)]
ide: disable drive interrupts in ide_driveid_update()

Since ide_driveid_update() uses polling to execute the IDENTIFY DEVICE command
but clears nIEN bit in the control register and doesn't mask the IDE interrupt,
the latter does happen and lead to the corresponding message to appear:

ide0: unexpected interrupt, status=0x58, count=1

when e.g. running hdparm with option -X with a non-PCI IDE driver...

Signed-off-by: Mikhail Cherkashin <mcherkashin@ru.mvista.com>
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agopalm_bk3710: fix tRP for UDMA mode 4
Mikhail Cherkashin [Tue, 15 Jul 2008 19:21:40 +0000 (21:21 +0200)]
palm_bk3710: fix tRP for UDMA mode 4

Fix tRP timing for UDMA mode 4 according to the ATA specification.

Signed-off-by: Mikhail Cherkashin <mcherkhashin@ru.mvista.com>
Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
16 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/shaggy...
Linus Torvalds [Tue, 15 Jul 2008 18:03:19 +0000 (11:03 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/shaggy/jfs-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/shaggy/jfs-2.6:
  jfs: remove DIRENTSIZ
  JFS: diAlloc() should return -EIO rather than EIO
  JFS: skip bad iput() call in error path
  JFS: switch to seq_files
  JFS: 0 is not valid errno value so return NULL from jfs_lookup

16 years agoFix printk format warning in clocksource/acpi_pm.c
Linus Torvalds [Tue, 15 Jul 2008 18:01:39 +0000 (11:01 -0700)]
Fix printk format warning in clocksource/acpi_pm.c

For real, this time.  The earlier attempt just moved the warning around.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
16 years agoMerge branch 'timers/for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Tue, 15 Jul 2008 17:39:57 +0000 (10:39 -0700)]
Merge branch 'timers/for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'timers/for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  x86: add PCI ID for 6300ESB force hpet
  x86: add another PCI ID for ICH6 force-hpet
  kernel-paramaters: document pmtmr= command line option
  acpi_pm clccksource: fix printk format warning
  nohz: don't stop idle tick if softirqs are pending.
  pmtmr: allow command line override of ioport
  nohz: reduce jiffies polling overhead
  hrtimer: Remove unused variables in ktime_divns()
  hrtimer: remove warning in hres_timers_resume
  posix-timers: print RT watchdog message

16 years agoMerge branch 'genirq' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux...
Linus Torvalds [Tue, 15 Jul 2008 17:39:22 +0000 (10:39 -0700)]
Merge branch 'genirq' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'genirq' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  genirq: remove extraneous checks in manage.c
  genirq: Expose default irq affinity mask (take 3)

16 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-2.6-nmw
Linus Torvalds [Tue, 15 Jul 2008 17:38:46 +0000 (10:38 -0700)]
Merge git://git./linux/kernel/git/steve/gfs2-2.6-nmw

* git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-2.6-nmw:
  [GFS2] Fix GFS2's use of do_div() in its quota calculations
  [GFS2] Remove unused declaration
  [GFS2] Remove support for unused and pointless flag
  [GFS2] Replace rgrp "recent list" with mru list
  [GFS2] Allow local DF locks when holding a cached EX glock
  [GFS2] Fix delayed demote race
  [GFS2] don't call permission()
  [GFS2] Fix module building
  [GFS2] Glock documentation
  [GFS2] Remove all_list from lock_dlm
  [GFS2] Remove obsolete conversion deadlock avoidance code
  [GFS2] Remove remote lock dropping code
  [GFS2] kernel panic mounting volume
  [GFS2] Revise readpage locking
  [GFS2] Fix ordering of args for list_add
  [GFS2] trivial sparse lock annotations
  [GFS2] No lock_nolock
  [GFS2] Fix ordering bug in lock_dlm
  [GFS2] Clean up the glock core

16 years agoMerge branch 'core/topology' of git://git.kernel.org/pub/scm/linux/kernel/git/tip...
Linus Torvalds [Tue, 15 Jul 2008 17:32:39 +0000 (10:32 -0700)]
Merge branch 'core/topology' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'core/topology' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  cputopology: always define CPU topology information, clean up
  cpu topology: always define CPU topology information

16 years agoMerge branch 'core/stacktrace' of git://git.kernel.org/pub/scm/linux/kernel/git/tip...
Linus Torvalds [Tue, 15 Jul 2008 17:31:35 +0000 (10:31 -0700)]
Merge branch 'core/stacktrace' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'core/stacktrace' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  generic-ipi: powerpc/generic-ipi tree build failure
  stacktrace: fix build failure on sparc64
  stacktrace: export save_stack_trace[_tsk]
  stacktrace: fix modular build, export print_stack_trace and save_stack_trace
  backtrace: replace timer with tasklet + completions
  stacktrace: add saved stack traces to backtrace self-test
  stacktrace: print_stack_trace() cleanup
  debugging: make stacktrace independent from DEBUG_KERNEL
  stacktrace: don't crash on invalid stack trace structs

16 years agoMerge branch 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Tue, 15 Jul 2008 15:41:43 +0000 (08:41 -0700)]
Merge branch 'x86-fixes-for-linus' of git://git./linux/kernel/git/tip/linux-2.6-tip

* 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  x86: Fix compile error with CONFIG_AS_CFI=n
  Documentation: document debugpat commandline option
  x86: sanitize Kconfig
  x86, suspend, acpi: correct and add comments about Big Real Mode
  x86, suspend, acpi: enter Big Real Mode

Fixed trivial conflict in include/asm-x86/dwarf2.h due to just using
different names for "cfi_ignore" (vs "__cfi_ignore") macro.

16 years agoMerge branch 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4
Linus Torvalds [Tue, 15 Jul 2008 15:36:38 +0000 (08:36 -0700)]
Merge branch 'for_linus' of git://git./linux/kernel/git/tytso/ext4

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4: (61 commits)
  ext4: Documention update for new ordered mode and delayed allocation
  ext4: do not set extents feature from the kernel
  ext4: Don't allow nonextenst mount option for large filesystem
  ext4: Enable delalloc by default.
  ext4: delayed allocation i_blocks fix for stat
  ext4: fix delalloc i_disksize early update issue
  ext4: Handle page without buffers in ext4_*_writepage()
  ext4: Add ordered mode support for delalloc
  ext4: Invert lock ordering of page_lock and transaction start in delalloc
  mm: Add range_cont mode for writeback
  ext4: delayed allocation ENOSPC handling
  percpu_counter: new function percpu_counter_sum_and_set
  ext4: Add delayed allocation support in data=writeback mode
  vfs: add hooks for ext4's delayed allocation support
  jbd2: Remove data=ordered mode support using jbd buffer heads
  ext4: Use new framework for data=ordered mode in JBD2
  jbd2: Implement data=ordered mode handling via inodes
  vfs: export filemap_fdatawrite_range()
  ext4: Fix lock inversion in ext4_ext_truncate()
  ext4: Invert the locking order of page_lock and transaction start
  ...

16 years agoMerge branch 'drm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied...
Linus Torvalds [Tue, 15 Jul 2008 15:35:28 +0000 (08:35 -0700)]
Merge branch 'drm-fixes' of git://git./linux/kernel/git/airlied/drm-2.6

* 'drm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6:
  drm/radeon: fixup issue with radeon and PAT support.

16 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris...
Linus Torvalds [Tue, 15 Jul 2008 15:16:48 +0000 (08:16 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/jmorris/security-testing-2.6

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/security-testing-2.6:
  Revert "SELinux: allow fstype unknown to policy to use xattrs if present"

16 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland...
Linus Torvalds [Tue, 15 Jul 2008 15:01:15 +0000 (08:01 -0700)]
Merge branch 'for-linus' of git://git./linux/kernel/git/roland/infiniband

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: (52 commits)
  IB/mlx4: Use kzalloc() for new QPs so flags are initialized to 0
  mlx4_core: Use MOD_STAT_CFG command to get minimal page size
  RDMA/cma: Simplify locking needed for serialization of callbacks
  RDMA/addr: Keep pointer to netdevice in struct rdma_dev_addr
  RDMA/cxgb3: Fixes for zero STag
  RDMA/core: Add local DMA L_Key support
  IB/mthca: Fix check of max_send_sge for special QPs
  IB/mthca: Use round_jiffies() for catastrophic error polling timer
  IB/mthca: Remove "stop" flag for catastrophic error polling timer
  IPoIB: Double default RX/TX ring sizes
  IPoIB/cm: Reduce connected mode TX object size
  IB/ipath: Use IEEE OUI for vendor_id reported by ibv_query_device()
  IPoIB: Use dev_set_mtu() to change mtu
  IPoIB: Use rtnl lock/unlock when changing device flags
  IPoIB: Get rid of ipoib_mcast_detach() wrapper
  IPoIB: Only set Q_Key once: after joining broadcast group
  IPoIB: Remove priv->mcast_mutex
  IPoIB: Remove unused IPOIB_MCAST_STARTED code
  RDMA/cxgb3: Set rkey field for new memory windows in iwch_alloc_mw()
  RDMA/nes: Get rid of ring_doorbell parameter of nes_post_cqp_request()
  ...

16 years agox86: Fix compile error with CONFIG_AS_CFI=n
Kevin Winchester [Tue, 15 Jul 2008 00:36:13 +0000 (21:36 -0300)]
x86: Fix compile error with CONFIG_AS_CFI=n

   AS      arch/x86/lib/csum-copy_64.o
arch/x86/lib/csum-copy_64.S: Assembler messages:
arch/x86/lib/csum-copy_64.S:48: Error: Macro `ignore' was already defined
make[1]: *** [arch/x86/lib/csum-copy_64.o] Error 1
make: *** [arch/x86/lib] Error 2

It appears that csum-copy_64.S and dwarf2.h both define an ignore macro.
I would expect one of them can be renamed quite easily, unless they
are references elsewhere.

Caused-by-commit: 392a0fc96bd059b38564f5f8fb58327460cb5a9d
    x86: merge dwarf2 headers

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
16 years agoDocumentation: document debugpat commandline option
Thomas Gleixner [Tue, 15 Jul 2008 13:04:56 +0000 (15:04 +0200)]
Documentation: document debugpat commandline option

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
16 years agox86: sanitize Kconfig
Thomas Gleixner [Tue, 15 Jul 2008 12:48:48 +0000 (14:48 +0200)]
x86: sanitize Kconfig

Set default n for MEMTEST and MTRR_SANITIZER and fix the help texts.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
16 years agoMerge branch 'linus' into x86/urgent
Ingo Molnar [Tue, 15 Jul 2008 11:45:59 +0000 (13:45 +0200)]
Merge branch 'linus' into x86/urgent

16 years agoRevert "SELinux: allow fstype unknown to policy to use xattrs if present"
James Morris [Tue, 15 Jul 2008 08:32:49 +0000 (18:32 +1000)]
Revert "SELinux: allow fstype unknown to policy to use xattrs if present"

This reverts commit 811f3799279e567aa354c649ce22688d949ac7a9.

From Eric Paris:

"Please drop this patch for now.  It deadlocks on ntfs-3g.  I need to
rework it to handle fuse filesystems better.  (casey was right)"

16 years agoIB/mlx4: Use kzalloc() for new QPs so flags are initialized to 0
Eli Cohen [Tue, 15 Jul 2008 06:48:53 +0000 (23:48 -0700)]
IB/mlx4: Use kzalloc() for new QPs so flags are initialized to 0

Current code uses kmalloc() and then just does a bitwise OR operation on
qp->flags in create_qp_common(), which means that qp->flags may
potentially have some unintended bits set.  This patch uses kzalloc()
and avoids further explicit clearing of structure members, which also
shrinks the code:

add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-65 (-65)
function                                     old     new   delta
create_qp_common                            2024    1959     -65

Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
16 years agomlx4_core: Use MOD_STAT_CFG command to get minimal page size
Vladimir Sokolovsky [Tue, 15 Jul 2008 06:48:53 +0000 (23:48 -0700)]
mlx4_core: Use MOD_STAT_CFG command to get minimal page size

There was a bug in some versions of the mlx4 driver in
mlx4_alloc_fmr(), which hardcoded the minimum acceptable page_shift to
be 12.  However, new ConnectX firmware can support a minimum
page_shift of 9 (log_pg_sz of 9 returned by QUERY_DEV_LIM) -- so with
old drivers, ib_fmr_alloc() would fail for ULPs using the device
minimum when creating FMRs.

To preserve firmware compatibility with released mlx4 drivers, the
firmware will continue to return 12 as before for log_page_sz in
QUERY_DEV_CAP for these drivers.  However, to enable new drivers to
take advantage of the available smaller page size, the mlx4 driver now
first sets the log_pg_sz to the device minimum by setting a
log_page_sz value to 0 via the MOD_STAT_CFG command and then reading
the real minimum via QUERY_DEV_CAP.

Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il>
Signed-off-by: Vladimir Sokolovsky <vlad@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>