FAQ Search Today's Posts Mark Forums Read
» Video Reviews

» Linux Archive

Linux-archive is a website aiming to archive linux email lists and to make them easily accessible for linux users/developers.


» Sponsor

» Partners

» Sponsor

Go Back   Linux Archive > Ubuntu > Ubuntu Kernel Team

 
 
LinkBack Thread Tools
 
Old 12-18-2009, 08:56 AM
Stefan Bader
 
Default 2.6.31.9 stable kernel update

Hi all,

I thought as I am not working, err well theoretically, I thought
I post my notes on the upcoming 2.6.31.9 update (review period
end today).

There are another 92 patches, 4 of which I just added as pre-stable
and 2 more we asked to be in stable. There is even one that has a
reference to Launchpad that is not coming from us.
(I added BugLink annotations to all patches I immediately could tell)

Steve, there are also a bunch (3) drm/i915 changes which we (well at
least I) did not explicitly request. Could you check your notes/lists
and in case you got bugs for those problems reply with them?

Happy $WHATEVER_YOU_DO,
-Stefan

commit 386a08c279bc8cef5cb11e32784ff027646a184e
Author: zhao.ming9@zte.com.cn <zhao.ming9@zte.com.cn>

USB: option: add pid for ZTE

Type: Enablement
- Just adds a bunch of new IDs to one driver.

commit d968e2b293428f7d50b7c309e8fe3658353e2828
Author: Alan Stern <stern@rowland.harvard.edu>

USB: usb-storage: fix bug in fill_inquiry

Type: Security/Data consistency
- Initializes string structure passed back to userspace which would
otherwise contain random kernel data.
- Also optimizes repeated calls to strlen.

commit cec505ca34edc7472d930901e519b905fa4da5be
Author: Jay Fenlason <fenlason@redhat.com>

firewire: ohci: handle receive packets with a data length of zero

Type: Crash
- Avoids a crash in a certain transfer mode by checking for zero sized
packets.
- Also optimizes a loop

commit 45582402feed8522a6e823a3a0adcb7bdfc9f7f9
Author: Jaroslav Kysela <perex@perex.cz>

ALSA: hda - Terradici HDA controllers does not support 64-bit mode

Type: Enablement
- Disables 64bit address mode for a specific audio controller.
- The description does not describe too much about the failure mode, but
from the Redhat bug report it looks like it just would not work.

commit 3a80740b4471da73c787fd569e3c86cddfa6372f
Author: Takashi Iwai <tiwai@suse.de>

ALSA: hrtimer - Fix lock-up

Type: Hang
- Avoids a recursive lock-up by moving some code and adding a bit to
toggle for testing a running timer.
! Slightly uneasy on atomic operations without other protection but
could be ok.

commit 9528f19cb4da0a5d25074291bb8c33c090c8b973
Author: Bob Copeland <me@bobcopeland.com>

ath5k: allow setting txpower to 0

Type: Cleanup
- Removes a previous quirk which sets the default transmit power
when the argument is 0.

commit 58e696fd90ae0398759530d900f2fbfea5b82eb1
Author: Luis R. Rodriguez <lrodriguez@atheros.com>

ath5k: enable EEPROM checksum check

Type: Debug
- Enables a commented out section which causes the EEPROM checksum
to be verified and if it not matches, is returning an error.
! There is a chance that some devices will fail with this now. On
the other hand, those might be the ones which gave problems.

commit f5715134382265376115555441159532b3a63cfd
Author: Alexey Dobriyan <adobriyan@gmail.com>

bsdacct: fix uid/gid misreporting

Type: Accounting
- Corrects the uid/gid returned on process accounting (sa -m)
! This fixes up a problem introduced by 2.6.31.5

commit 72bf6cbd140be281fcdc2015f353164ca56d183a
Author: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>

debugfs: fix create mutex racy fops and private data

Type: Race
- Moved setting data and file operations inside protected area by passing
them as arguments to inner functions.

commit a3e276a5ad3c2f614af7c977db41da0d9cac51eb
Author: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com>

devpts_get_tty() should validate inode

Type: Crash
- Validates that tty has not already been invalidated by looking up the
dentry.
- Fixes a regression introduced by 2.6.28

commit 8989f8ee39acc5812e30d480819e29232c781996
Author: Andi Kleen <andi@firstfloor.org>

futex: Take mmap_sem for get_user_pages in fault_in_user_writeable

Type: Cash/Data inconsistency (???)
- Ensures the mmap_sem is held while calling get_user_pages
! Patch looks ok, but what effects does it fix?

commit 50f8a845bc0955b5a9ef26632082675026e97ea5
Author: Amerigo Wang <amwang@redhat.com>

hfs: fix a potential buffer overflow

Type: Security
- A length argument supplied from the filesystem was not checked against
the actually allocated element buffer, so a filesystem could be crafted
to cause a buffer overflow and execute alien code.
- Adds checks against the length to several places. Looks ok, but not
checked for completeness.

commit ac55d501b6e10f44e646a5015386f405e590caad
Author: Feng Tang <feng.tang@intel.com>

hrtimer: Fix /proc/timer_list regression

Type: Regression/Correctness
- Removes on check and corrects another check which were introduced
by 2.6.31 that was causing /proc/timer_stats not to display the
start function.

commit 2c73ab050e7cc8ffc2a794acbee4fa617c49d59d
Author: Theodore Ts'o <tytso@mit.edu>

jbd2: don't wipe the journal on a failed journal checksum

Type: Data recovery
- Adds a check to return an error when an invalid journal transaction
is detected. This allow the administrator/userspace to decide how
to handle the situation.

commit 274a3287a74c5e35f788a3ac50fadfe955bbd67d
Author: Carsten Otte <carsteno@de.ibm.com>

KVM: s390: Make psw available on all exits, not just a subset

Type: Enhancment/Architecture specific
- Adds additional data fields to the kvm run data to store the the psw
data on exist. Change is limited to the S390 architecture.

commit 02ee954e459a91eb763ee4558a1819c5def3a58b
Author: NeilBrown <neilb@suse.de>

md/bitmap: protect against bitmap removal while being updated.

Type: Crash
- Adds mutex to protect against removing a structure while working
on it.
- ABI bumper(?)

commit 75d84dc70d9661292ce64b89059e4e40ee8c74f9
Author: Sergei Shtylyov <sshtylyov@ru.mvista.com>

pata_hpt{37x|3x2n}: fix timing register masks (take 2)

Type: Enablement(?)
- Mostly (apart from version numbers) modifies masks used for bitfield
operations. Have to trust the author there.

commit d03d545f9d18e023c5e3f614ea7e1c638a5afb45
Author: Igor Grinberg <grinberg@compulab.co.il>

pxa/em-x270: fix usb hub power up/reset sequence

Type: Enablement
- Changes around the sequence used in hub init for a specific driver
on ARM.

commit 4611c5ae5ff5e5d8ef12e023f307e8d81e6958e4
Author: Michael Buesch <mb@bu3sch.de>

ssb: Fix range check in sprom write

Type: Enablement/Checking
- Fixes an incorrect length assumption and add more checking before
writing to the sprom (better prevents bricking devices).

commit 8ff8264c72f906c833928134393f989bc1fd0919
Author: Roel Kluin <roel.kluin@gmail.com>

SUNRPC: IS_ERR/PTR_ERR confusion

Type: Checking
- Changes code to use the correct PTR macro to convert to an error
number. The symptom is not mentioned but hopefulle the comment
about this normally not being expected is true.

commit 8a03a3776f92be6026a71c654574eb6066c7faa3
Author: Russ Dill <Russ.Dill@gmail.com>

USB: Close usb_find_interface race v3

Type: Race
- Fix a lookup race by walking a different structure list when looking
for a device.

commit 7bcd8b7d0a98b2b6c4bba4a9cf2ae0b49c3a0d49
Author: Sergei Shtylyov <sshtylyov@ru.mvista.com>

USB: musb_gadget_ep0: fix unhandled endpoint 0 IRQs, again

Type: Cleanup
- Remove one line to change the internal state machine in one case that
is causing unhandled interrupt messages.
- Fixup for fix that went into 2.6.31

commit fd9351441f1af6965ef060e0f6adbab852aa7127
Author: Zhang Le <r0bertz@gentoo.org>

USB: option.c: add support for D-Link DWM-162-U5

Type: Enablement
- Adds a ID.

commit 506ece0302357374de4301db7b706e3d2f7bb2e4
Author: Alan Stern <stern@rowland.harvard.edu>

USB: usb-storage: add BAD_SENSE flag

Type: Enablement
- Add special code to handle devices that fail when asked for more
than 18b of sense data (thought claiming to support this).
- One device is quirked by this patch otherwise this only gets
enabled by new device entries or kernel arguments.

commit 441e390c5092472164d368b342133ec3e7a60e79
Author: Andre Herms <andre.herms@tec-venture.de>

USB: usbtmc: repeat usb_bulk_msg until whole message is transfered

Type: Hang
- Adds retry code to the usbtmc driver, which can otherwise cause
devices to hang until reset or replugged.

commit 37200d1ba5eb8065fe96cfdc8fb25b1446b7f3ac
Author: Roel Kluin <roel.kluin@gmail.com>

V4L/DVB: Fix test in copy_reg_bits()

Type: ???
- Fixes one half of a test in a loop. So potentially not everything
was correctly copied...

commit f3a7b5b59480177b49b38913209c35029103aeea
Author: Dave Jones <davej@redhat.com>

x86: Add new Intel CPU cache size descriptors

Type: Enablement
- Adds new Intel cache types.

commit 710313be627f17d57e2a560657a66a34cd6ac8f9
Author: Joerg Roedel <joerg.roedel@amd.com>

x86/amd-iommu: attach devices to pre-allocated domains early

Type: Crash
- Attaches DMA domains sooner, to avoid IO page faults.

commit 0f1b67bb56591c9c57ca1b60a0c502d13d7deead
Author: Joerg Roedel <joerg.roedel@amd.com>

x86/amd-iommu: un__init iommu_setup_msi

Type: Crash(?)
- Removes the __init annotation from a function which needs to be called
again after resume.

commit b628e4119c625d45635de890d0cba2f7164a8c48
Author: Mikael Pettersson <mikpe@it.uu.se>

x86, apic: Enable lapic nmi watchdog on AMD Family 11h

Type: Enablement
- Include CPU family 0x11 for NMI watchdog operation.

commit d18e42833ab653006b84cfa98de9ec4ff924cbde
Author: Leann Ogasawara <leann.ogasawara@canonical.com>

x86: ASUS P4S800 reboot=bios quirk

Type: Enablement
- Reboot quirk from Leann.

commit 072e9c386e4cac60bf62cbfe92148d42e0e7452e
Author: Darrick J. Wong <djwong@us.ibm.com>

x86, Calgary IOMMU quirk: Find nearest matching Calgary while walking up the PCI tree

Type: Enhancement
- Changes the way PCI devices look for a certain IOMMU controller
(Calgary). This results in correct tree layout.

commit 60a13b1a01bd9c0af0c9aa25d1bb59cab42e110c
Author: Tejun Heo <tj@kernel.org>

x86: Fix iommu=nodac parameter handling

Type: Cleanup
- When iommu=nodac was used, this resulted in using it. Simple cleanup
fixes this.

commit 41eb0435d174bd6107b80552eedb95024f2ea11b
Author: Dave Jones <davej@redhat.com>

x86: Fix typo in Intel CPU cache size descriptor

Type: Cleanup
- One of the entries in the cache info was wrong. Value changed.

commit a46bfd4f323250c025b148413c24b120bae6375c
Author: Joe Perches <joe@perches.com>

x86: GART: pci-gart_64.c: Use correct length in strncmp

Type: Cleanup
- String check forgot to compare the last character of a keyword.

commit 22d90dc228dada69876eaf58014f2276b3354e86
Author: Alex Deucher <alexdeucher@gmail.com>

drm/radeon/kms: Add quirk for HIS X1300 board

Type: Enablement
- Adds special treatment for a specific device.

commit 2b5b520ceb2450fc85d9466bc2b6c4bbf79dfeaf
Author: Alex Deucher <alexdeucher@gmail.com>

drm/radeon/kms: fix legacy crtc2 dpms

Type: Enhancment
- Fixes DPMS code in the Radeon legacy driver.

commit c0520b5d73781ff9f6168b83d51140a7e3136ce3
Author: Vasanthakumar Thiagarajan <vasanth@atheros.com>

mac80211: Fix bug in computing crc over dynamic IEs in beacon

Type: Enhancement
- Replace bit macro by shifting a 64bit 1 to work arount limitations
of the macro.

commit 9a71e7d4922b8ffda829d198f17dc6618de0c7b6
Author: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>

mm: hugetlb: fix hugepage memory leak in mincore()

Type: Memory leak
- Adds handling of hugetlb pages to mincode (whatever that is again)
to prevent leaking hugetlb pages.

commit dc853572f125b878bca17ce316ff050db0d9a6f8
Author: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>

mm: hugetlb: fix hugepage memory leak in walk_page_range()

Type: Memory leak
- Changes walk_page_range to skip over hugetlb pages to avoid leaking
those.

commit 5d73c813668a47c33336b644b1e5ac9b01324938
Author: Benjamin Herrenschmidt <benh@kernel.crashing.org>

powerpc: Fix usage of 64-bit instruction in 32-bit altivec code

Type: Crash(?)[powerpc]
- Replaces one assembly instruction in code used by 32 and 64 bit code
by one that is actually usable in both modes (was a 64bit instruction).
- Introduced in 2.6.31

commit 35b289cc49fcd20ef9a3b1d3447f0bf8e43385bf
Author: Ian Jackson <ian.jackson@eu.citrix.com>

Serial: Do not read IIR in serial8250_start_tx when UART_BUG_TXEN

Type: Data loss(?)
- Avoid accessing a register (interrupt information register?), as this
clears outstanding interrupts, before transmitting.

commit 06c3541e1835f0b48c186cef6b30289e9f129b31
Author: Mikulas Patocka <mpatocka@redhat.com>

ide: Serialize CMD643 and CMD646 to fix a hardware bug with SSD

Type: Data corruption
- Adds flags to two cmd64x IDE controllers to serialize access to them
to work around a bug that corrupts data for the first disk when a SSD
is connected to the second port.

commit 56747daf5d2e7906fa151698282cf66e8e05e049
Author: Hemant Pedanekar <hemantp@ti.com>

ide: fix ioctl to pass requested transfer mode to ide_find_dma_mode instead of UDMA6

Type: Enhancement
- Simple fix to replace a fix value by the actual argument of an ioctl call.
- Allows to actually set a DMA mode apart from the max or the one fix value.

commit 1625f593588445ddc2db1c3528b109fd251bdbc9
Author: David S. Miller <davem@davemloft.net>

Revert "ide: try to use PIO Mode 0 during probe if possible"

Type: Regression
- Reverts a patch from 2.6.31 which causes set_pio_mode to be called
before the media_type is correctly set and there are several drivers
that need this information in their implementation of set_pio_mode.
! Could have related bug reports.

commit e4c68cde631ac28b011548c255be00de35556829
Author: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>

slc90e66: fix UDMA handling

Type: ???
- Fixes checking for current DMA speed. Effect?

commit 12af265bae2d3830444a75ac347632b6f068a04d
Author: Alexander Beregalov <a.beregalov@gmail.com>

Au1x00: fix crash when trying register_netdev()

Type: Crash
- Looks sensible to setup the device before calling register.

commit 3f25257a57dc590ad122a394b437414b7061fc36
Author: Stanislav Brabec <sbrabec@suse.cz>

b44 WOL setup: one-bit-off stack corruption kernel panic fix

Type: Stack corruption/Crash
- Fixes a one of bit write by using the bit value before incrementing it.
- Also seems to make WOL actually working.

commit e3655ba4624a8ffbbd683786989efe4a3df06a68
Author: David S. Miller <davem@davemloft.net>

b44: Fix wedge when using netconsole.

Type: Hang
- Removes a warning message which is said to be invalid and actually
harmful.

commit 9c7bf848388b43deaef046465d3f7b9dae46c9ba
Author: Herbert Xu <herbert@gondor.apana.org.au>

gro: Fix illegal merging of trailer trash

Type: Oops
- Adds an additional check for restrictions to avoid later BUG().

commit db24979c033c06859cd701aadab9e728c03aa88c
Author: Patrick McHardy <kaber@trash.net>

ip_fragment: also adjust skb->truesize for packets not owned by a socket

Type: ???
- Recycled skb might fail after being used for defragmentation. Adjust code
to update one variable in all cases instead of only for certain cases.

commit 46632b44ba7cbc2908deaf9d2e2eb6b866cdbd5c
Author: David S. Miller <davem@davemloft.net>

Revert "isdn: isdn_ppp: Use SKB list facilities instead of home-grown implementation."

Type: Regression
- Reverts patch that caused ISDN channel bundling to fail.
- Regression introduced with 2.6.28.

commit 5ec4898c1c8bf95ba9a8cdba89508e843f6e8351
Author: Eric Dumazet <eric.dumazet@gmail.com>

net: Fix the rollback test in dev_change_name()

Type: Enhancement
- Fixes rollback to be done in more cases (>= 0 instead of == 0).

commit 73aa4d5ca0b1b132571ac806e4aa49cec5b2f739
Author: Russell King - ARM Linux <linux@arm.linux.org.uk>

NET: smc91x: Fix irq flags

Type: Enablement[ARM]
- Adds an aditional check for handling special init values.

commit f6d522414505ebb5d37d91c45abeacd10b4510ca
Author: Steve Glendinning <steve.glendinning@smsc.com>

smsc9420: prevent BUG() if ethtool is called with interface down

Type: Oops
- Adds checking for interface down (NULL pointer).

commit ff608daa3c9d4c0d1f198fd1bb66e96a1b0b6b05
Author: Eric Dumazet <eric.dumazet@gmail.com>

vlan: Fix register_vlan_dev() error path

Type: Hang/Crash
- Adds hlist_del_rcu to remove a group before freeing it.

commit abdaad9e727fee85d59f6b351f1df6dac74f0686
Author: David S. Miller <davem@davemloft.net>

sparc64: Fix definition of VMEMMAP_SIZE.

Type: Memory corruption/Hang[sparc64]
- Changes allocation to used the size of the used struct and not pointers
for it.

commit ada3c0c8efce0e805658f90fe88daccc83df123b
Author: David S. Miller <davem@davemloft.net>

sunsab: Do not set sunsab_reg.cons right before registering minors.

Type: Cleanup[sun]
- Not setting this value prevents registering all minors als consoles.

commit dea14b09b7989e992c21a7e5e752ce25dfe91e72
Author: David S. Miller <davem@davemloft.net>

sunsu: Fix detection of SU ports which are RSC console or control.

Type: Cleanup[sun?]
- It seems to at an additional su compatible entry and a look to
match serial devices to su compatible ones.

commit 88f0f7dddcec2d586d7489d1f46ca9d610aa066d
Author: David S. Miller <davem@davemloft.net>

serial: suncore: Add 'ignore_line' argument to sunserial_console_match().

Type: Enablement
- This adds a new argument to one function to optionally get a different
logic. Though for that patch all callers use the old logic.
- ABI bumper.

commit 7d4e2f3fddf198a619e7f6c9127b03af77d72f1b
Author: David S. Miller <davem@davemloft.net>

serial: suncore: Fix RSC/LOM handling in sunserial_console_termios().

Type: Enablement
- Add additional device to handle into the sun serial driver

commit 5ab279af2d15781dcbde3148090a90ddccc5c829
Author: David S. Miller <davem@davemloft.net>

sunsu: Pass true 'ignore_line' to console match when RSC or LOM console.

Type: Enablement
- Now makes use of the ignore option added previously and uses that for
certain console devices.

commit 44829f7769b6feb0f40c5ced1c2a1f2b21d5a278
Author: David S. Miller <davem@davemloft.net>

sunsu: Use sunserial_console_termios() in sunsu_console_setup().

Type: Enablement
- Finally replaces hard coded IO parameter by values queried from the core.

commit 142b506aa61d5e5c73ee2192884f534625fd8332
Author: David S. Miller <davem@davemloft.net>

sparc64: Don't specify IRQF_SHARED for LDC interrupts.

Type: Cleanup[sparc64]
- Remove one option of two contradicting ones.

commit 955c6f2467de3351063ea072d277d2f62c68c3cd
Author: David S. Miller <davem@davemloft.net>

sparc64: Fix overly strict range type matching for PCI devices.

Type: Cleanup[sparc]
- Not understanding the restriction itself but it is made less restrictive
as the comment says.

commit b0c01c2186f7647098e665e8ca54aad3caab9e09
Author: David S. Miller <davem@davemloft.net>

sparc64: Fix stack debugging IRQ stack regression.

Type: Cleanup[sparc64]
- Changes the two issues mentioned in the patch comment.

commit 2362de7fe974966d023f08ef8cb61b64c16e595e
Author: David S. Miller <davem@davemloft.net>

sparc: Set UTS_MACHINE correctly.

Type: Cleanup[sparc]
- Corrects the value UTS_MACHINE is set to.

commit ae6f902b19a56b18c82095908807a5d0b552b81f
Author: Jan Beulich <jbeulich@novell.com>

x86/mce: Set up timer unconditionally

Type: Hang
- Move the call to setup_timer ahead, so it gets called in all cases.

commit 9f9d8ea7ceea9b224d6b69b76067f432733b6cb6
Author: Larry Finger <Larry.Finger@lwfinger.net>

b43legacy: avoid PPC fault during resume

Type: Crash
- This basically does the same change as has been done for the b43 driver
in 01a6a52a3cadb6a6db03199f3ec7ccb023aac45f (2.6.31.6). It avoids to
access the hw registers when the status in not started.
- BugLink: http://bugs.launchpad.net/bugs/476154

commit 12c73fff78d79060eec08820e64a07ec2c597ee4
Author: Larry Finger <Larry.Finger@lwfinger.net>

p54usb: Remove DMA buffer from stack

Type: StackWarning
- Moves a 4 byte buffer that was allocated as a stack variable to be
dynamically allocated.

commit cae0a1b5433da84fce3cd6782a02f6cf2eded401
Author: Yong Wang <yong.y.wang@linux.intel.com>

x86: Under BIOS control, restore AP's APIC_LVTTHMR to the BSP value

Type: Hang
- Patch adds a sort of quirking function to save the intitial thermal
interrupt handling and uses it later to setupr. Looks as it does what
it says in the description.

commit 8cb5bb38ae929eed055a3dcc223b38fec82e3488
Author: Zhao Yakui <yakui.zhao@intel.com>

ACPI: Use the ARB_DISABLE for the CPU which model id is less than 0x0f.

Type: Hang/Regression
- Fixes a boot hang regression introduced by the code to not use ARB_DISABLE.
This is the second round of loosening those restrictions.
- pre-stable: a50ec6658391bbff827b93a0fb5593bed85a3406
BugLink: http://bugs.launchpad.net/bugs/481765

commit febf681a1085f7022826155ae7ed8ac41841e935
Author: Corentin Chary <corentincj@iksaif.net>

asus-laptop: change light sens default values.

Type: Cleanup
- Sets initial values as intended (and documented in the comments already).
- BugLink: https://bugs.launchpad.net/bugs/222171

commit 77ab99ad2929c9dcaf33385400e4e1d7072a394f
Author: Jean Delvare <khali@linux-fr.org>

backlight: lcd - Fix wrong sizeof

Type: Cleanup
- Replaces a sizeof of effectively the pointer by the sizeof the variable,
which makes sense.

commit 4c7afaa6b465c31db4ae38d57ad7ee2610a8f676
Author: Chris Wilson <chris@chris-wilson.co.uk>

drm/i915: Avoid NULL dereference with component_only tv_modes

Type: Oops
- Add checking for NULL pointer before access.
- pre-stable bdd3691a2bc4387c3f2981122fc6b645e79ddd06
BugLink: http://bugs.launchpad.net/bugs/494045

commit be344841de704ea6f2f53fd64f1e566841219f6f
Author: Zhenyu Wang <zhenyuw@linux.intel.com>

drm/i915: Fix CRT hotplug detect by checking really no channels attached

Type: Enablement
- Moves check from testing specific "good" values to exclude the one "bad"
value.
! Steve?

commit 35f22fbb50106ca7de6f9946506a41574b8f1331
Author: Zhenyu Wang <zhenyuw@linux.intel.com>

drm/i915: Fix LVDS stability issue on Ironlake

Type: Enablement
- Moves around and adds tests for various bits. Should be save to older
intel video devices as the only function changed is igdng_crtc_dpms,
where igdng indicates functions used only for/on the new devices.
! Steve?

commit 92447bdf43aa3bbd9d2cc6c77abad8b6dcc68162
Author: Jesse Barnes <jbarnes@virtuousgeek.org>

drm/i915: save/restore BLC histogram control reg across suspend/resume

Type: Regression
- Saves and restores one additional register on suspend resume which avoids
having no backlight on resume.
- pre-stable 5b78f8db9e1b72b8c63cc90ca74f7f37d8bf9999
BugLink: https://bugs.launchpad.net/bugs/292256

commit c88ba80453a6c534c58b8014624e2538c420e569
Author: Zhenyu Wang <zhenyuw@linux.intel.com>

drm/i915: PineView only has LVDS and CRT ports

Type: Enablement
- Adds test macros and replaces/adds them where the code tries to decide
based on the series of the device which features are present. Looks ok.
- Implications? Only providing some option that does not exist or more?
- Steve?

commit 1d8ef7a2d54b096bb7dc5bc94f573b1bb9243f63
Author: Jan Kara <jack@suse.cz>

ext3: Fix data / filesystem corruption when write fails to copy data

Type: Data corruption
- Replaces the call of ext3_truncate by a helper function which also
calls truncate_inode_pages, which matches the description.

commit b2ce04cdc440332cf307c2405a6086c900ba0fc3
Author: Simon Horman <horms@verge.net.au>

ipvs: zero usvc and udest

Type: Security
- Zeroes two data structures. Comment notes problems otherwise on
that netfilter.

commit f5469c117b1f05de16e260bc321ce8593441dd0b
Author: Zhu Yi <yi.zhu@intel.com>

ipw2100: fix rebooting hang with driver loaded

Type: Reboot/Power management
- Adds a shutdown function to that driver. Patch mentions reboot problems
on some machines but this might also improve power management.

commit 407d1fb47b4586c742bb0cbee21c7d199be1e76c
Author: David Woodhouse <David.Woodhouse@intel.com>

jffs2: Fix long-standing bug with symlink garbage collection.

Type: Memory corruption/Data corruption?
- Changes the length indicator from a char to an int to avoid "interesting
and bad" things from happening.

commit 606faaa54b67a183eeec9db08674f3a20030a0ee
Author: Alan Cox <alan@linux.intel.com>

matroxfb: fix problems with display stability

Type: Regression
- Deals with a regression that apparently exists since 2.6.23 by commenting
out one line which seems to read out a DAC value.

commit 5177643e1344bb1f23d635bb004e687b61c103ee
Author: Eric W. Biederman <ebiederm@xmission.com>

net: Fix userspace RTM_NEWLINK notifications.

Type: Notification race
- Move user-space notification calls to places when the device is already
setup to avoid user-space trying to access before.

commit 8f6bc759edf4eab3c7da12c393e382392ec395ff
Author: Henrique de Moraes Holschuh <hmh@hmh.eng.br>

thinkpad-acpi: fix default brightness_mode for R50e/R51

Type: Enablement/Regression(?)
- Changes default for two TP models to control backlight by EC.

commit 43084a05f94c77aec087c43d14bf65dd7955b4cd
Author: Henrique de Moraes Holschuh <hmh@hmh.eng.br>

thinkpad-acpi: preserve rfkill state across suspend/resume

Type: Regression
- Removes own special handling and restores bluetooth and wireless
to the values provided by the BIOS.

commit 0c25fa2d014001239b0e4884c228ab18e6b91cd1
Author: Rafal Milecki <zajec5@gmail.com>

V4L/DVB (13116): gspca - ov519: Webcam 041e:4067 added.

Type: Enablement
- Adds one gspca webcam ID (submitted for stable by csurbhi)

commit 1496e89ae2a0962748e55165a590fa3209c6f158
Author: Darrick J. Wong <djwong@us.ibm.com>

powerpc/therm_adt746x: Record pwm invert bit at module load time]

Type: Regression
- Fixes up a patch introduced in 2.6.31 which unconditionally set the invert
bit, but sometimes it is supposed to be set. Now saves the initial state
and restores it.

commit 778c902640530371a169ad1c03566e7c51b09874
Author: Li Peng <peng.li@linux.intel.com>

Fix sync to vblank when VGA output is turned off

Type: Regression
- Somehwat larger but discussed and found it should be safe.
- pre-stable fad5557de503827f6b9934477e548cb1169c246a
BugLink: http://bugs.launchpad.net/bugs/494461

--
kernel-team mailing list
kernel-team@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/kernel-team
 
Old 12-18-2009, 08:58 AM
Stefan Bader
 
Default 2.6.31.9 stable kernel update

Stefan Bader wrote:
> Hi all,
>
> I thought as I am not working, err well theoretically, I thought
> I post my notes on the upcoming 2.6.31.9 update (review period
> end today).
>

> There are another 92 patches, 4 of which I just added as pre-stable

Actually it is only 90, there where 90 then two more but also two
dropped. :-P

-Stefan

> and 2 more we asked to be in stable. There is even one that has a
> reference to Launchpad that is not coming from us.
> (I added BugLink annotations to all patches I immediately could tell)
>
> Steve, there are also a bunch (3) drm/i915 changes which we (well at
> least I) did not explicitly request. Could you check your notes/lists
> and in case you got bugs for those problems reply with them?
>
> Happy $WHATEVER_YOU_DO,
> -Stefan
>
>


--
kernel-team mailing list
kernel-team@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/kernel-team
 

Thread Tools




All times are GMT. The time now is 03:58 AM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.
Copyright 2007 - 2008, www.linux-archive.org