Skip to content

L
linux
Switch branch/tag
  1. 22 May, 2009 7 commits
  3. 17 May, 2009 8 commits
  5. 16 May, 2009 4 commits
  7. 30 Apr, 2009 3 commits
  9. 23 Apr, 2009 4 commits
  11. 22 Apr, 2009 5 commits
    • Bartlomiej Zolnierkiewicz's avatar
      Merge branch 'for-linus' into for-next · 24fc484a
      Bartlomiej Zolnierkiewicz authored
      24fc484a
    • Bartlomiej Zolnierkiewicz's avatar
      mediabay: fix build for CONFIG_BLOCK=n · 83cff839
      Bartlomiej Zolnierkiewicz authored 
      On Tuesday 14 April 2009 20:31:21 Subrata Modak wrote:
> Observed the following build error:
> ---
> CC      drivers/macintosh/mediabay.o
> In file included from drivers/macintosh/mediabay.c:21:
> include/linux/ide.h:605: error: field ‘request_sense_rq’ has incomplete
> type
> make[2]: *** [drivers/macintosh/mediabay.o] Error 1
> make[1]: *** [drivers/macintosh] Error 2
> make: *** [drivers] Error 2
> ---

mediabay shouldn't include <linux/ide.h> unconditionally so
remove the superfluous include from mediabay.c (<asm/mediabay.h>
will pull <linux/ide.h> in for CONFIG_BLK_DEV_IDE_PMAC=y).
Reported-by: default avatarSubrata Modak <subrata@linux.vnet.ibm.com>
Cc: Paul Mackerras <paulus@samba.org>
Signed-off-by: default avatarBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
      83cff839
    • Bruno Prémont's avatar
      ide: Stop disks on reboot for laptop which cuts power · b0aedb04
      Bruno Prémont authored 
      My laptop (Acer Travelmate 660) always cuts the power when rebooting
which causes the disk to emergency-park it's head.

Add a dmi check to stop disk as for shutdown on this laptop.
Signed-off-by: default avatarBruno Prémont <bonbons@linux-vserver.org>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: default avatarBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
      b0aedb04
    • Helge Deller's avatar
      ide-cd: fix kernel crash on hppa regression · a1f9a89c
      Helge Deller authored 
      With 2.6.30-rc2 I face a kernel crash on the 32bit hppa architecture 
due to ide-cd when udev creates the device nodes at startup:

Kernel Fault: Code=26 regs=8ed34c40 (Addr=00000024)
IASQ: 00000000 00000000 IAOQ: 1034b5ac 1034b5b0
 IIR: 4ab30048    ISR: 00000000  IOR: 00000024
 CPU:        0   CR30: 8ed34000 CR31: ffff55ff
 ORIG_R28: 00000000
 IAOQ[0]: ide_complete_rq+0x2c/0x70
 IAOQ[1]: ide_complete_rq+0x30/0x70
 RP(r2): cdrom_newpc_intr+0x178/0x46c
Backtrace:
 [<1035c608>] cdrom_newpc_intr+0x178/0x46c
 [<1034c494>] ide_intr+0x1b0/0x214
 [<1016d284>] handle_IRQ_event+0x70/0x150
 [<1016d4b0>] __do_IRQ+0x14c/0x1cc
 [<102f7864>] superio_interrupt+0x88/0xbc
 [<1016d284>] handle_IRQ_event+0x70/0x150
 [<1016d4b0>] __do_IRQ+0x14c/0x1cc
 [<10112efc>] do_cpu_irq_mask+0x9c/0xd0
 [<10116068>] intr_return+0x0/0x4

This crash seems to happen due to an uninitialized variable "rc".
The compiler even warns about that:
  CC      drivers/ide/ide-cd.o                                                         
/mnt/sda4/home/cvs/parisc/git-kernel/linus-linux-2.6/drivers/ide/ide-cd.c: In function `cdrom_newpc_intr':
/mnt/sda4/home/cvs/parisc/git-kernel/linus-linux-2.6/drivers/ide/ide-cd.c:612: warning: `rc' might be used uninitialized in this function

After applying the trivial patch below, which just initializes 
the variable to zero, the kernel doesn't crash any longer:

Starting the hotplug events dispatcher: udevd.
Synthesizing the initial hotplug events...
hda: command error: status=0x51 { DriveReady SeekComplete Error }
hda: command error: error=0x54 <3>{ AbortedCommand LastFailedSense=0x05 }
ide: failed opcode was: unknown
done.
Signed-off-by: default avatarHelge Deller <deller@gmx.de>
Acked-by: default avatarBorislav Petkov <petkovbb@gmail.com>
Cc: Linus <torvalds@linux-foundation.org>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: default avatarBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
      a1f9a89c
    • David Brownell's avatar
      palm_bk3710: UDMA performance fix · db2f38c2
      David Brownell authored 
      Fix UDMA throughput bug:  tCYC averages t2CYCTYP/2, but the code
previously assumed it was the same as t2CYCTYP.  (That is, it was
using just one clock edge, not both.)  Move the table's type
declaration so it's adjacent to the table, making it more clear
what those numbers mean.

On one system this change increased throughput by almost 4x: UDMA/66
sometimes topped 23 MB/sec (on a drive known to do much better).  On
another system it was around a 10% win (UDMA/66 up to 7+ MB/sec).

The difference might be caused by the ratio between memory and IDE
clocks.  In the system with large speedup, this was exactly 2 (as a
workaround for a rev 1.1 silicon bug).  The other system used a more
standard ratio of 1.63 (and rev 2.1 silicon) ... clock domain synch
might have some issues, they're not unheard-of.
Signed-off-by: default avatarDavid Brownell <dbrownell@users.sourceforge.net>
Acked-by: default avatarSergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: default avatarBartlomiej Zolnierkiewicz <bzolnier@gmail.com>
      db2f38c2
  13. 21 Apr, 2009 1 commit
  15. 18 Apr, 2009 8 commits
    • Tejun Heo's avatar
      ide: drop rq->data handling from ide_map_sg() · 2c316bb5
      Tejun Heo authored 
      Impact: remove code path which is no longer necessary

All IDE data transfers now use rq->bio.  Simplify ide_map_sg()
accordingly.
Signed-off-by: default avatarTejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
      2c316bb5
    • Tejun Heo's avatar
      ide-atapi: kill unused fields and callbacks · 6d700387
      Tejun Heo authored 
      Impact: remove fields and code paths which are no longer necessary

Now that ide-tape uses standard mechanisms to transfer data, special
case handling for bh handling can be dropped from ide-atapi.  Drop the
followings.

* pc->cur_pos, b_count, bh and b_data
* drive->pc_update_buffers() and pc_io_buffers().
Signed-off-by: default avatarTejun Heo <tj@kernel.org>
      6d700387
    • Tejun Heo's avatar
      ide-tape: simplify read/write functions · 07bd9686
      Tejun Heo authored 
      Impact: cleanup

idetape_chrdev_read/write() functions are unnecessarily complex when
everything can be handled in a single loop.  Collapse
idetape_add_chrdev_read/write_request() into the rw functions and
simplify the implementation.
Signed-off-by: default avatarTejun Heo <tj@kernel.org>
      07bd9686
    • Tejun Heo's avatar
      ide-tape: use byte size instead of sectors on rw issue functions · 6bb11dd1
      Tejun Heo authored 
      Impact: cleanup

Byte size is what most issue functions deal with, make
idetape_queue_rw_tail() and its wrappers take byte size instead of
sector counts.  idetape_chrdev_read() and write() functions are
converted to use tape->buffer_size instead of ctl from tape->cap.

This cleans up code a little bit and will ease the next r/w
reimplementation.
Signed-off-by: default avatarTejun Heo <tj@kernel.org>
      6bb11dd1
    • Tejun Heo's avatar
      ide-tape: unify r/w init paths · 88f1b941
      Tejun Heo authored 
      Impact: cleanup

Read and write init paths are almost identical.  Unify them into
idetape_init_rw().
Signed-off-by: default avatarTejun Heo <tj@kernel.org>
      88f1b941
    • Tejun Heo's avatar
      ide-tape: kill idetape_bh · 963da55c
      Tejun Heo authored 
      Impact: kill now unnecessary idetape_bh

With everything using standard mechanisms, there is no need for
idetape_bh anymore.  Kill it and use tape->buf, cur and valid to
describe data buffer instead.

Changes worth mentioning are...

* idetape_queue_rq_tail() now always queue tape->buf and and adjusts
  buffer state properly before completion.

* idetape_pad_zeros() clears the buffer only once.
Signed-off-by: default avatarTejun Heo <tj@kernel.org>
      963da55c
    • Tejun Heo's avatar
      ide-tape: use standard data transfer mechanism · 21d9c5d2
      Tejun Heo authored 
      Impact: use standard way to transfer data

ide-tape uses rq in an interesting way.  For r/w requests, rq->special
is used to carry a private buffer management structure idetape_bh and
rq->nr_sectors and current_nr_sectors are initialized to the number of
idetape blocks which isn't necessary 512 bytes.  Also,
rq->current_nr_sectors is used to report back the residual count in
units of idetape blocks.

This peculiarity taxes both block layer and ide.  ide-atapi has
different paths and hooks to accomodate it and what a rq means becomes
quite confusing and making changes at the block layer becomes quite
difficult and error-prone.

This patch makes ide-tape use bio instead.  With the previous patch,
ide-tape currently is using single contiguos buffer so replacing it
isn't difficult.  Data buffer is mapped into bio using
blk_rq_map_kern() in idetape_queue_rw_tail().  idetape_io_buffers()
and idetape_update_buffers() are dropped and pc->bh is set to null to
tell ide-atapi to use standard data transfer mechanism and idetape_bh
byte counts are updated by the issuer on completion using the residual
count.

This change also nicely removes the FIXME in ide_pc_intr() where
ide-tape rqs need to be completed using ide_rq_bytes() instead of
blk_rq_bytes() (although this didn't really matter as the request
didn't have bio).
Signed-off-by: default avatarTejun Heo <tj@kernel.org>
Cc: Jens Axboe <jens.axboe@oracle.com>
      21d9c5d2
    • Tejun Heo's avatar
      ide-tape: use single continuous buffer · 35ab8d32
      Tejun Heo authored 
      Impact: simpler buffer allocation and handling, kills OOM, fix DMA transfers

ide-tape has its own multiple buffer mechanism using struct
idetape_bh.  It allocates buffer with decreasing order-of-two
allocations so that it results in minimum number of segments.
However, the implementation is quite complex and works in a way that
no other block or ide driver works necessitating a lot of special case
handling.

The benefit this complex allocation scheme brings is questionable as
PIO or DMA the number of segments (16 maximum) doesn't make any
noticeable difference and it also doesn't negate the need for multiple
order allocation which can fail under memory pressure or high
fragmentation although it does lower the highest order necessary by
one when the buffer size isn't power of two.

As the first step to remove the custom buffer management, this patch
makes ide-tape allocate single continous buffer.  The maximum order is
four.  I doubt the change would cause any trouble but if it ever
matters, it should be converted to regular sg mechanism like everyone
else and even in that case dropping custom buffer handling and moving
to standard mechanism first make sense as an intermediate step.

This patch makes the first bh to contain the whole buffer and drops
multi bh handling code.  Following patches will make further changes.

This patch has the side effect of killing OOM triggered by allocation
path and fixing DMA transfers.  Previously, bug in alloc path
triggered OOM on command issue and commands were passed to DMA engine
without DMA-mapping all the segments.
Signed-off-by: default avatarTejun Heo <tj@kernel.org>
      35ab8d32
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7