Hi Linus,
Can you pull the drm-forlinus branch from
git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6.git
This is a pretty major merge over of DRM CVS, and every driver in the DRM
is brought up to latest versions.... along with a lot of interface
renaming that makes the driver interface easier to understand (it used to
use the same names as the old macro interface)..
Regards,
Dave.
b/drivers/char/drm/Makefile | 4
b/drivers/char/drm/ati_pcigart.c | 23
b/drivers/char/drm/drm.h | 4
b/drivers/char/drm/drmP.h | 122 +++--
b/drivers/char/drm/drm_agpsupport.c | 133 +++--
b/drivers/char/drm/drm_bufs.c | 49 +-
b/drivers/char/drm/drm_context.c | 2
b/drivers/char/drm/drm_core.h | 4
b/drivers/char/drm/drm_drv.c | 152 +++---
b/drivers/char/drm/drm_fops.c | 317 +++++++------
b/drivers/char/drm/drm_ioctl.c | 27 -
b/drivers/char/drm/drm_lock.c | 1
b/drivers/char/drm/drm_memory.c | 8
b/drivers/char/drm/drm_memory_debug.h | 269 +++++------
b/drivers/char/drm/drm_os_linux.h | 1
b/drivers/char/drm/drm_pciids.h | 12
b/drivers/char/drm/drm_proc.c | 16
b/drivers/char/drm/drm_stub.c | 63 --
b/drivers/char/drm/drm_sysfs.c | 66 +-
b/drivers/char/drm/i810_dma.c | 49 +-
b/drivers/char/drm/i810_drv.c | 60 --
b/drivers/char/drm/i810_drv.h | 10
b/drivers/char/drm/i830_dma.c | 47 +
b/drivers/char/drm/i830_drv.c | 59 --
b/drivers/char/drm/i830_drv.h | 8
b/drivers/char/drm/i915_dma.c | 52 +-
b/drivers/char/drm/i915_drm.h | 6
b/drivers/char/drm/i915_drv.c | 66 --
b/drivers/char/drm/i915_drv.h | 44 +
b/drivers/char/drm/i915_irq.c | 48 +-
b/drivers/char/drm/i915_mem.c | 5
b/drivers/char/drm/mga_dma.c | 160 ++++--
b/drivers/char/drm/mga_drv.c | 58 --
b/drivers/char/drm/mga_drv.h | 14
b/drivers/char/drm/mga_state.c | 26 -
b/drivers/char/drm/r128_cce.c | 15
b/drivers/char/drm/r128_drm.h | 4
b/drivers/char/drm/r128_drv.c | 48 --
b/drivers/char/drm/r128_drv.h | 8
b/drivers/char/drm/r128_irq.c | 4
b/drivers/char/drm/r128_state.c | 42 -
b/drivers/char/drm/r300_cmdbuf.c | 38 -
b/drivers/char/drm/r300_reg.h | 1
b/drivers/char/drm/radeon_cp.c | 106 ++--
b/drivers/char/drm/radeon_drm.h | 6
b/drivers/char/drm/radeon_drv.c | 62 +-
b/drivers/char/drm/radeon_drv.h | 41 -
b/drivers/char/drm/radeon_state.c | 246 ++++------
b/drivers/char/drm/savage_bci.c | 81 +--
b/drivers/char/drm/savage_drv.c | 50 --
b/drivers/char/drm/savage_drv.h | 29 -
b/drivers/char/drm/savage_state.c | 324 ++++++-------
b/drivers/char/drm/sis_drm.h | 25 +
b/drivers/char/drm/sis_drv.c | 42 -
b/drivers/char/drm/sis_drv.h | 4
b/drivers/char/drm/sis_ds.h | 7
b/drivers/char/drm/sis_mm.c | 30 -
b/drivers/char/drm/tdfx_drv.c | 42 -
b/drivers/char/drm/tdfx_drv.h | 7
b/drivers/char/drm/via_dma.c | 38 +
b/drivers/char/drm/via_dmablit.c | 805 ++++++++++++++++++++++++++++++++++
b/drivers/char/drm/via_dmablit.h | 140 +++++
b/drivers/char/drm/via_drm.h | 58 +-
b/drivers/char/drm/via_drv.c | 63 --
b/drivers/char/drm/via_drv.h | 56 +-
b/drivers/char/drm/via_ds.c | 9
b/drivers/char/drm/via_irq.c | 53 +-
b/drivers/char/drm/via_map.c | 47 +
b/drivers/char/drm/via_mm.c | 20
b/drivers/char/drm/via_verifier.c | 6
b/drivers/char/drm/via_verifier.h | 4
b/drivers/char/drm/via_video.c | 7
drivers/char/drm/drm_init.c | 53 --
73 files changed, 2794 insertions(+), 1812 deletions(-)
commit 9c7d462eda13ca211b7b4a62f191f4cfda135e2d
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Thu Jan 12 20:44:30 2006 +1100
drm: fix issues with systems with no MTRR
On systems with no MTRR we should still define the interface.
Original bug from apkm.
Signed-off-by: Dave Airlie <[email protected]>
commit aab8df141fdc4c4c9587521a24b6865390eaeb79
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Wed Jan 11 22:32:51 2006 +1100
drm: cleanup properly on drm module unload
Cleanup multiple cards properly
Signed-off-by: Dave Airlie <[email protected]>
commit 8d2ea6258123d7a92a1f6ec638a8cad4a0604c43
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Wed Jan 11 20:48:09 2006 +1100
drm: fixup drm bufs being just under the EOM
If the mapping was just under the end of memory it would fail.
Lets DRM start on my PCI card.
Signed-off-by: Dave Airlie <[email protected]>
commit f8e0f2905bf0a7cb5ef2baaf009f0c26f80c3056
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Tue Jan 10 19:56:17 2006 +1100
drm: fix radeon warnings on 64-bit
From: Andrew Morton <[email protected]>
drivers/char/drm/radeon_state.c: In function `radeon_cp_dispatch_texture':
drivers/char/drm/radeon_state.c:1653: warning: int format, different type arg
(arg 3)
drivers/char/drm/radeon_state.c:1661: warning: int format, different type arg
(arg 3)
drivers/char/drm/radeon_state.c:1689: warning: int format, different type arg
(arg 3)
sizeof() doesn't return an int.
Cc: Dave Airlie <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
commit 2185200cd2a910ca7f4e3fa0370c6ed8a2bdc49c
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Tue Jan 10 19:53:54 2006 +1100
drm: fix warning on alpha
From: Andrew Morton <[email protected]>
On alpha:
drivers/char/drm/via_dmablit.h:44: error: field `direction' has incomplete type
Cc: Dave Airlie <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
commit 4e4c62bd45c40e8e04d09a5f383bffb149abaa63
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Tue Jan 3 22:25:29 2006 +1100
drm: remove is_pci flag completely...
this snuck back in, in the last merge.
Signed-off-by: Dave Airlie <[email protected]>
commit 97f2aab6698f3ab2552c41c1024a65ffd0763a6d
Merge: d985c1088146607532093d9eaaaf99758f6a4d21 88026842b0a760145aa71d69e74fbc9ec118ca44
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Tue Jan 3 18:18:01 2006 +1100
drm: merge in Linus mainline
commit d985c1088146607532093d9eaaaf99758f6a4d21
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Mon Jan 2 21:32:48 2006 +1100
drm: major update from CVS for radeon and core
This patch pull in a lot of changes from CVS to the main core DRM,
and updates the radeon driver to 1.21.0 that supports r300 texrect
and radeon card type ioctl.
Signed-off-by: Dave Airlie <[email protected]>
commit b0cae664ebc85f2431c5a7c9e192a2a2ef72ecc7
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Mon Jan 2 21:23:07 2006 +1100
drm: update drm pci ids list
Signed-off-by: Dave Airlie <[email protected]>
commit fe34765be1ee9465b10406e6e5dddbd43ddc4fbe
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Mon Jan 2 21:19:39 2006 +1100
drm: drm_ioctl.c sync with fixes from CVS
Apply the fixes from CVS that were outstanding for this file
Signed-off-by: Dave Airlie <[email protected]>
commit b3a80a223d5f1af1e1713383376e5472cec4e20c
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Mon Jan 2 21:15:01 2006 +1100
drm: update lock flags from userspace
Signed-off-by: Dave Airlie <[email protected]>
commit 0d6aa60b4ac9689b750e35cd66f5d7c053aff0f4
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Mon Jan 2 20:14:23 2006 +1100
drm: update to i915 1.3.0
Add support for vblank ioctls to i915 driver
From: Dave Airlie <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
commit f0c408b564ddefa0959ada4e2c2248f4a57f1842
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Mon Jan 2 19:52:09 2006 +1100
drm: update drm_memory_debug.h
Update from DRM CVS for drm memory debug
From: Jon Smirl <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
commit 1e7d51902a8bd08e37113aaf182d12233b157151
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Mon Jan 2 19:25:35 2006 +1100
drm: proper fix for drm_context
Bad patch in last version
Signed-off-by: Dave Airlie <[email protected]>
commit b5e9fc13dd0f25a2f422000c185f491bfd4f7335
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Mon Jan 2 19:23:44 2006 +1100
drm: fix issue with contexts running out of RAM
Signed-off-by: Dave Airlie <[email protected]>
commit f26c473cdf557ea6e8f267d34eee82d30473a363
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Mon Jan 2 17:18:39 2006 +1100
drm: update PCIGART support from CVS
In order to work on FreeBSD the gart needed to use a local mapping
This patch moves the mainline to the new code and aligns some comment
changes
From: Eric Anholt <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
commit 0a406877e638a6f43ed4591bb08d528415d7d53a
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Mon Jan 2 16:49:02 2006 +1100
drm: remove old reclaim_buffers from ix0 drivers
From: Dave Airlie <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
commit 269dc51296f4e985741d2fd567e7be4e7a0a9f29
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Mon Jan 2 16:23:01 2006 +1100
drm: bring savage inline with latest CVS
apply some whitespace cleanup and add wrappers for MTRR for OS calls
From: Eric Anholt <[email protected]> + Dave Airlie <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
commit 3528af1b189d0fbb4c7a3f121f46d9987b9af5b6
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Mon Jan 2 16:11:44 2006 +1100
drm: fix a LOR issue on FreeBSD for savage driver
Correct a LOR issue on FreeBSD by allocating temporary space and doing a single
DRM_COPY_FROM_USER rather than DRM_VERIFYAREA_READ followed by tons of
DRM_COPY_FROM_USER_UNCHECKED. I don't like the look of the temporary space
allocation, but I like the simplification in the rest of the file. Tested
with glxgears, tuxracer, and q3 on a savage4.
From: Eric Anholt <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
commit 952d751a140e961f7ac67f743cf94d1a37c736e8
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Mon Jan 2 14:44:12 2006 +1100
drm: bring sis + tdfx up to latest CVS
Cleanup SIS + TDFX drivers with latest changes from CVS.
From: Eric Anholt <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
commit 443448d05468277abe99c9b24b9df538dd840f35
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Mon Jan 2 14:26:20 2006 +1100
drm: via driver update to CVS version
This updates the DRM via driver to the latest CVS version, which contains
support for DMA blitting.
It also contains some whitespace and other minor fixes
From: Thomas Hellstrom <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
commit a7a2cc315c8a5e51b08538d102ec3229c966ac87
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Mon Jan 2 13:54:04 2006 +1100
drm: move ioctl flags to a bit field of flags
From: Dave Airlie
Signed-off-by: Dave Airlie <[email protected]>
commit 925142431bd653175b80ae153bd7a8bc13628bde
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Sat Nov 12 21:52:46 2005 +1100
drm: update VIA driver to 2.7.2
Add PCI DMA blitengine to VIA DRM
Add portability code for porting VIA to FreeBSD.
Sync via_drm.h with 3d driver
From: Thomas Hellstrom <[email protected]>, Eric Anholt <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
commit 792d2b9a12594522111fbe2a7f17460a4d7edff7
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Fri Nov 11 23:30:27 2005 +1100
drm: drop mtrr from i915
Alan Hourihane wants to set MTRR in the DDX only as otherwise
we get problems with the shared memory chipset.
From: Alan Hourihane <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
commit 7ccf800e9415daf9214eb667318e356f9a3d81fc
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Fri Nov 11 23:11:34 2005 +1100
drm: update mga driver for new bootstrap code
The MGA driver needs to use the full AGP interface.
From: Ian Romanick <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
commit c0be4d240483f3ebd138db467b5e8fbe15c520e2
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Fri Nov 11 23:10:18 2005 +1100
drm: remove exports that modules shouldn't use.
Modules should go via the new drm_agp_ functions.
Signed-off-by: Dave Airlie <[email protected]>
commit efa58395bee82dc5a87805e7eb7c710e88eb4bd7
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Fri Nov 11 22:33:39 2005 +1100
drm: add in-kernel entry points for rest of AGP ioctls
Allow DRM modules to call AGP internally in the kernel.
From: Ian Romanick <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
commit 732052ed3e7539d87136dd833be523747af3fb3e
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Fri Nov 11 22:07:35 2005 +1100
drm: simplify sysfs code for drm
This simplifies the sysfs code for the drm and add a dri_library_name
attribute which can be used by a userspace app to figure out which
library to load.
From: Jon Smirl <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
commit e96e33eeb8b876c7ec009c557ca5269328eceda0
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Fri Nov 11 20:27:35 2005 +1100
drm: fixup drm_proc.c struct table
Signed-off-by: Dave Airlie <[email protected]>
commit 61d04160ff89514919ef95b0d10bee706f569925
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Fri Nov 11 19:52:22 2005 +1100
drm: remove old backwards compatibilty stuff
Signed-off-by: Dave Airlie <[email protected]>
commit 8f5f39f77f5a6053ae287d4673028e7a69335f5e
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Fri Nov 11 19:40:52 2005 +1100
drm: remove drm_flush
drm_flush is no longer needed remove.
Signed-off-by: Dave Airlie <[email protected]>
commit 7052cff984ba575926bb7d2ae5454cce531a97e1
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Fri Nov 11 19:34:47 2005 +1100
drm: cleanup via_ds.c includes
Remove the linux includes from via_ds.c
Signed-off-by: Dave Airlie <[email protected]>
commit 9d6160137a8ef8bd25266ccc0f97d55863708fc6
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Fri Nov 11 19:34:10 2005 +1100
drm: remove remnamt of old DRM code from tdfx
Signed-off-by: Dave Airlie <[email protected]>
commit 3f9df54d6386bb632ffc00665489bb3bb3bf6ff2
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Thu Nov 10 22:28:56 2005 +1100
drm: remove drm_init.c it is no longer needed
Move drm_cpu_valid into drm_fops.c
Signed-off-by: Dave Airlie <[email protected]>
commit 22eae947bf76e236ba972f2f11cfd1b083b736ad
Author: Dave Airlie <airlied@starflyer.(none)>
Date: Thu Nov 10 22:16:34 2005 +1100
drm: rename driver hooks more understandably
Rename the driver hooks in the DRM to something a little more understandable:
preinit -> load
postinit -> (removed)
presetup -> firstopen
postsetup -> (removed)
open_helper -> open
prerelease -> preclose
free_filp_priv -> postclose
pretakedown -> lastclose
postcleanup -> unload
release -> reclaim_buffers_locked
version -> (removed)
postinit and version were replaced with generic code in the Linux DRM (drivers
now set their version numbers and description in the driver structure, like on
BSD). postsetup wasn't used at all. Fixes the savage hooks for
initializing and tearing down mappings at the right times. Testing involved at
least starting X, running glxgears, killing glxgears, exiting X, and repeating.
Tested on: FreeBSD (g200, g400, r200, r128)
Linux (r200, savage4)
From: Eric Anholt <[email protected]>
Signed-off-by: Dave Airlie <[email protected]>
--
David Airlie, Software Engineer
http://www.skynet.ie/~airlied / airlied at skynet.ie
Linux kernel - DRI, VAX / pam_smb / ILUG
On Thu, 12 Jan 2006, Dave Airlie wrote:
>
> Hi Linus,
> Can you pull the drm-forlinus branch from
> git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6.git
>
> This is a pretty major merge over of DRM CVS, and every driver in the DRM
> is brought up to latest versions....
I'm actually somewhat inclined to not pull any more. We've had lots of
(hopefully minor) issues for the last few days, and I know that people
had DRM issues with the -mm tree (which I assume tracked this tree) not
more than a week or so ago.
IOW, I want to make sure that my tree is somewhat stable again. I don't
want -rc1 to be horrible.
Linus
> >
> > Hi Linus,
> > Can you pull the drm-forlinus branch from
> > git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6.git
> >
> > This is a pretty major merge over of DRM CVS, and every driver in the DRM
> > is brought up to latest versions....
>
> I'm actually somewhat inclined to not pull any more. We've had lots of
> (hopefully minor) issues for the last few days, and I know that people
> had DRM issues with the -mm tree (which I assume tracked this tree) not
> more than a week or so ago.
>
Actually none of this tree was causing problems in -mm it was the
agpgart tree, this tree is fully tested on all my available hardware,
it is also the first tree that supports multiple cards properly
without oopsing on module remove....
Dave.
Linus Torvalds <[email protected]> wrote:
>
>
>
> On Thu, 12 Jan 2006, Dave Airlie wrote:
> >
> > Hi Linus,
> > Can you pull the drm-forlinus branch from
> > git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6.git
> >
> > This is a pretty major merge over of DRM CVS, and every driver in the DRM
> > is brought up to latest versions....
>
> I'm actually somewhat inclined to not pull any more. We've had lots of
> (hopefully minor) issues for the last few days, and I know that people
> had DRM issues with the -mm tree (which I assume tracked this tree) not
> more than a week or so ago.
iirc that was AGP. DRM had a few won't-compile problems, now fixed.
So I think we could squeeze DRM in, but yes, it's getting to that time.
> IOW, I want to make sure that my tree is somewhat stable again. I don't
> want -rc1 to be horrible.
Merge status:
Size in bytes tree
(including changlog)
666832 git-acpi.patch
2534 git-agpgart.patch
98909 git-audit.patch
32930 git-cfq.patch
46766 git-cifs.patch
289519 git-drm.patch
465768 git-infiniband.patch
1045 git-ntfs.patch
9244 git-ocfs2.patch
40442 git-pcmcia.patch
24191 git-sym2.patch
31765 git-watchdog.patch
acpi: A few recent reports of AML-level unaligned accesses in that tree.
audit: we're tracking one oops which seems to be coming out of the audit code
cfq: OK
CIFS: no problems of which I'm aware
DRM: no problems of which I'm aware
infiniband: Roland, you need to resend the pull request asap, please.
ntfs: no problems of which I'm aware
ocfs2: it's a small update
pcmcia: had a problem but I think that's now fixed. But this seems
to be fairly fresh code?
sym2: no problems of which I'm aware
watchdog: Wim has been very quiet in recent months. No problems
of which I'm aware.
On Fri, 13 Jan 2006, Dave Airlie wrote:
>
> Actually none of this tree was causing problems in -mm it was the
> agpgart tree, this tree is fully tested on all my available hardware,
> it is also the first tree that supports multiple cards properly
> without oopsing on module remove....
Ok, since Andrew backs you up on that one, I'll merge asap.
Linus
Hi,
On Thu, Jan 12, 2006 at 01:42:55PM -0800, Andrew Morton wrote:
> 40442 git-pcmcia.patch
> pcmcia: had a problem but I think that's now fixed. But this seems
> to be fairly fresh code?
For the two important fixes which are part of this tree, (add new IDs for
serial_cs.c, runtime powermanagement interface), please pull from
git://git.kernel.org/pub/scm/linux/kernel/git/brodo/pcmcia-fixes-2.6.git/
The other patches are indeed fresh code, and are meant for 2.6.17.
Thanks,
Dominik
On Thu, 2006-01-12 at 13:42 -0800, Andrew Morton wrote:
> audit: we're tracking one oops which seems to be coming out of the
> audit code
Are we? I recall one oops which was tracked down to an inode which had
i_sb == 0x00000008 so it didn't seem to be audit-related. Was there
something else I should be looking at?
--
dwmw2
> 465768 git-infiniband.patch
> infiniband: Roland, you need to resend the pull request asap, please.
I am planning on sending a request by Friday, but note that I think
you're looking my patches for -mm. That includes the driver for IBM
eHCA hardware, which is not going upstream now. Without that, the
queue is much smaller -- my branch for 2.6.16-rc1 looks like:
drivers/infiniband/core/cm.c | 29 +----
drivers/infiniband/core/sysfs.c | 22 +---
drivers/infiniband/core/uverbs_cmd.c | 2
drivers/infiniband/hw/mthca/mthca_eq.c | 28 +++--
drivers/infiniband/hw/mthca/mthca_provider.c | 122 ++++++++++++++-----------
drivers/infiniband/ulp/ipoib/ipoib_multicast.c | 28 +++--
drivers/infiniband/ulp/srp/ib_srp.c | 23 ----
include/rdma/ib_verbs.h | 2
8 files changed, 119 insertions(+), 137 deletions(-)
with a few more patches still to decide on.
- R.
David Woodhouse <[email protected]> wrote:
>
> On Thu, 2006-01-12 at 13:42 -0800, Andrew Morton wrote:
> > audit: we're tracking one oops which seems to be coming out of the
> > audit code
>
> Are we? I recall one oops which was tracked down to an inode which had
> i_sb == 0x00000008 so it didn't seem to be audit-related. Was there
> something else I should be looking at?
>
Well it's oopsing in the audit code, and might not oops without audit.
Perhaps the audit code is being called before i_sb is fully set up or
something. We won't know until we know.
Did we work out why i_sb is crazy?
On 1/12/06, Andrew Morton <[email protected]> wrote:
> Linus Torvalds <[email protected]> wrote:
> >
> >
> >
> > On Thu, 12 Jan 2006, Dave Airlie wrote:
> > >
> > > Hi Linus,
> > > Can you pull the drm-forlinus branch from
> > > git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6.git
> > >
> > > This is a pretty major merge over of DRM CVS, and every driver in the DRM
> > > is brought up to latest versions....
> >
> > I'm actually somewhat inclined to not pull any more. We've had lots of
> > (hopefully minor) issues for the last few days, and I know that people
> > had DRM issues with the -mm tree (which I assume tracked this tree) not
> > more than a week or so ago.
>
> iirc that was AGP. DRM had a few won't-compile problems, now fixed.
>
> So I think we could squeeze DRM in, but yes, it's getting to that time.
>
I would like to merge psmouse resync patch before -rc1 is out (I have
fixed issues with Frank Sorenson's touchpad and we really need it so
KVMs will be finally useful). The only thing is that I feel like shit
so I am not sure if I will be able to prepare the pull/patch today.
--
Dmitry
On Thu, Jan 12, 2006 at 10:17:52AM -0800, Linus Torvalds wrote:
> I'm actually somewhat inclined to not pull any more. We've had lots of
> (hopefully minor) issues for the last few days, and I know that people
> had DRM issues with the -mm tree (which I assume tracked this tree) not
> more than a week or so ago.
>
> IOW, I want to make sure that my tree is somewhat stable again. I don't
> want -rc1 to be horrible.
FYI - I have one large merge still to come, which is the ARM EABI code.
This allows us to build the kernel with the new ARM ELF ABI toolchain,
and maintain a certain level of compatibility with existing userspace.
(Some ioctls end up breaking, but so far we've only found some ALSA
mixer related ones.)
Of course, if you build with the existing non-EABI toolchain, there's
almost no change to the binary (well, it unfortunately grows by about
160 bytes because the stat64 structure is handled a very slightly
different but 100% compatible way.)
I'm expecting the patches to be re-spun hopefully before -rc1, but this
is dependent on others doing so - and of course it will only impact ARM.
--
Russell King
Linux kernel 2.6 ARM Linux - http://www.arm.linux.org.uk/
maintainer of: 2.6 Serial core
On Thu, 2006-01-12 at 14:07 -0800, Andrew Morton wrote:
> Well it's oopsing in the audit code, and might not oops without audit.
> Perhaps the audit code is being called before i_sb is fully set up or
> something. We won't know until we know.
>
> Did we work out why i_sb is crazy?
No, we didn't. He disabled auditing (which he never really wanted in the
first place, but Fedora Core 4 stupidly enables it by default), then I
believe his last mail reported a BUG() in skb_under_panic(). It all
looks fairly random to me.
--
dwmw2
Andrew,
> watchdog: Wim has been very quiet in recent months. No problems
> of which I'm aware.
No problems at this moment, except that the i8xx_tco driver doesn't
seem to work on ICH6. Apparently Intel changed something after the
ICH5 chipsets... Still need to look at that...
There is also a SuperMicro update that needs to go in for the
i8xx_tco driver.
For the rest I have been working on the Berkshire PC-Watchdog ISA card.
Greetings,
Wim.