login
Search
| From | Subject | Date |
|---|---|---|
| Justin Mattock | nopage mm.h
Hello; I have a question about the nopage function that was removed in
the kernel.
What would be the best way for me to build a module that has a nopage
funtion in it?
Right now I went and undid what you had done to the kernel,to atleast
have the module not give me errors,
but, am wondering if I should try to take the module and remove the
nopage function in it , and add the functions that is relevent.
to keep the module up-to date.
regards;
--
Justin P. Mattock
--
| May 3, 7:29 pm 2008 |
| Adrian Bunk | Re: nopage mm.h
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
--
| May 3, 7:38 pm 2008 |
| Justin Mattock | Re: nopage mm.h
The module is fglrx.
regards;
--
Justin P. Mattock
--
| May 3, 7:48 pm 2008 |
| Adrian Bunk | Re: nopage mm.h
Please ask AMD for support with your problem.
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
--
| May 3, 7:56 pm 2008 |
| Malte | Hard lockup with kernel 2.6.25 and 2.6.25.1
--MP_/gVrG5uwNqv5Relro=m_idHP
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Hello,
my Athlon XP locks up hard with the current stable kernel. Output of
lspci, the .config and dump from serial console is attached.
This happens without warning or special actions.=20
This incident happened only a few hours after booting the new kernel,
SysRq-keys didn't work. 2.6.24.5 has been running fine.
The problem starts at line 390 of the dum...
| May 3, 6:52 pm 2008 |
| Greg KH | Re: Hard lockup with kernel 2.6.25 and 2.6.25.1
Hm, this doesn't look good. Is it repeatable? Your log shows all the
PCI devices pretty much going out to lunch, which is very odd.
If it's repeatable, it would be interesting if you are able to do a
Does these lines also show up in 2.6.24.5?
thanks,
greg k-h
--
| May 3, 7:18 pm 2008 |
| Christian Kujau | Re: Hard lockup with kernel 2.6.25 and 2.6.25.1
What's with the I/O errors - did they show up in 2.6.24.5? If not, what
has changed in your .config with .25.1?
C.
--
BOFH excuse #202:
kernel panic: write-only-memory (/dev/wom0) capacity exceeded.
--
| May 3, 7:05 pm 2008 |
| Marcin Slusarz | [PATCH] xfrm: convert empty xfrm_audit_* macros to functions
it removes these warnings when CONFIG_AUDITSYSCALL is unset:
net/xfrm/xfrm_user.c: In function 'xfrm_add_sa':
net/xfrm/xfrm_user.c:412: warning: unused variable 'sid'
net/xfrm/xfrm_user.c:411: warning: unused variable 'sessionid'
net/xfrm/xfrm_user.c:410: warning: unused variable 'loginuid'
net/xfrm/xfrm_user.c: In function 'xfrm_del_sa':
net/xfrm/xfrm_user.c:485: warning: unused variable 'sid'
net/xfrm/xfrm_user.c:484: warning: unused variable 'sessionid'
net/xfrm/xfrm_user.c:483: warning: unused...
| May 3, 6:42 pm 2008 |
| devzero | Re: modprobe aha1542 oops with git17
thanks for the pointers !
i`m new to git and i thought i had fetched the latest git (just fetched via http some hours ago) - wondering why my copy doesn`t have that fix.....
_________________________________________________________________________
In 5 Schritten zur eigenen Homepage. Jetzt Domain sichern und gestalten!
Nur 3,99 EUR/Monat! http://www.maildomain.web.de/?mc=021114
--
| May 3, 6:38 pm 2008 |
| anthonyedem35 | Hello my good friend.
Hello my good friend.
I am very happy to inform you about my success in getting
that fund,Now I want you to contact my Account officer on
the information below:
CONTACT PERSON: Mr James Adam
His email address: jamesadam1@hotmail.fr
Ask him to send you the total sum of $950,000.00 Bank
Draft, which I kept for your compensation I have left
instrution to him on your behalf to recive the draft,
Also reconfirm him with your name,full addres and
cellphone number.
Regards,
Dr Anth...
| May 3, 6:35 pm 2008 |
| Natalie Protasevich | Kernel Bugzilla: Bug stats
Here some current numbers of bugs by subsystem and overall, in
bugzilla categories.
The number of open/solved bugs varies by subsystem, however, large
number of bugs is not always indication of bad maintenance. Some
factors are: "popularity" of the component and number of users, its
size etc., and also complexity involved, time and effort needed to
identify/solve a problem.
Dynamics of the bug resolution reflects quality and standing of the
module the best.
Are there any subsystems that don't...
| May 3, 6:32 pm 2008 |
| Ingo Molnar | [patch] irda: fix !PNP support for drivers/net/irda/smsc-irc...
x86.git testing found this build bug on v2.6.26-rc1:
ERROR: "pnp_get_resource" [drivers/net/irda/smsc-ircc2.ko] undefined!
make[1]: *** [__modpost] Error 1
make: *** [modules] Error 2
the driver did not anticipate the case of !CONFIG_PNP which is rare but
still possible. Instead of restricting the driver to PNP-only in the
Kconfig space, add the (trivial) dummy struct pnp_driver - this is that
other drivers use in the !PNP case too.
The driver itself can in theory be initialized on !P...
| May 3, 6:26 pm 2008 |
| Balbir Singh | [-mm][PATCH 0/4] Add rlimit controller to cgroups (v3)
This is the third version of the address space control patches. These
patches are against 2.6.25-mm1 and have been tested using KVM in SMP mode,
both with and without the config enabled.
The first patch adds the user interface. The second patch fixes the
cgroup mm_owner_changed callback to pass the task struct, so that
accounting can be adjusted on owner changes. The thrid patch adds accounting
and control. The fourth patch updates documentation.
An earlier post of the patchset can be found at
...
| May 3, 5:37 pm 2008 |
| Balbir Singh | [-mm][PATCH 4/4] Add rlimit controller documentation
This is the documentation patch. It describes the rlimit controller and how
to build and use it.
Signed-off-by: Balbir Singh <balbir@linux.vnet.ibm.com>
---
Documentation/controllers/rlimit.txt | 29 +++++++++++++++++++++++++++++
1 file changed, 29 insertions(+)
diff -puN /dev/null Documentation/controllers/rlimit.txt
--- /dev/null 2008-05-03 22:12:13.033285313 +0530
+++ linux-2.6.25-balbir/Documentation/controllers/rlimit.txt 2008-05-04 03:06:06.000000000 +0530
@@ -0,0 +1,29 @@
...
| May 3, 5:38 pm 2008 |
| Balbir Singh | [-mm][PATCH 3/4] Add rlimit controller accounting and control
This patch adds support for accounting and control of virtual address space
limits. The accounting is done via the rlimit_cgroup_(un)charge_as functions.
The core of the accounting takes place during fork time in copy_process(),
may_expand_vm(), remove_vma_list() and exit_mmap(). There are some special
cases that are handled here as well (arch/ia64/kernel/perform.c,
arch/x86/kernel/ptrace.c, insert_special_mapping())
Signed-off-by: Balbir Singh <balbir@linux.vnet.ibm.com>
---
arch/ia6...
| May 3, 5:38 pm 2008 |
| Balbir Singh | [-mm][PATCH 2/4] Enhance cgroup mm_owner_changed callback to...
This patch adds an additional field to the mm_owner callbacks. This field
is required to get to the mm that changed.
Signed-off-by: Balbir Singh <balbir@linux.vnet.ibm.com>
---
include/linux/cgroup.h | 3 ++-
kernel/cgroup.c | 2 +-
2 files changed, 3 insertions(+), 2 deletions(-)
diff -puN kernel/cgroup.c~cgroup-add-task-to-mm--owner-callbacks kernel/cgroup.c
--- linux-2.6.25/kernel/cgroup.c~cgroup-add-task-to-mm--owner-callbacks 2008-05-04 02:53:05.000000000 +0530
++...
| May 3, 5:38 pm 2008 |
| Balbir Singh | [-mm][PATCH 1/4] Setup the rlimit controller
This patch sets up the rlimit cgroup controller. It adds the basic create,
destroy and populate functionality. The user interface provided is very
similar to the memory resource controller. The rlimit controller can be
enhanced easily in the future to control mlocked pages.
Signed-off-by: Balbir Singh <balbir@linux.vnet.ibm.com>
---
include/linux/cgroup_subsys.h | 4 +
include/linux/rlimitcgroup.h | 19 +++++
init/Kconfig | 10 ++
mm/Makefile ...
| May 3, 5:37 pm 2008 |
| Ingo Molnar | [patch] video: build fix for drivers/media/video/au0828
x86.git testing found the following build failure in v2.6.26-rc1:
MODPOST 424 modules
ERROR: "tveeprom_hauppauge_analog" [drivers/media/video/au0828/au0828.ko] undefined!
ERROR: "tveeprom_read" [drivers/media/video/au0828/au0828.ko] undefined!
with this config:
http://redhat.com/~mingo/misc/config-Sat_May__3_22_28_58_CEST_2008.bad
this patch does what other video drivers do to utilize the
VIDEO_TVEEPROM functionality (and this resolves the build problem) - but
i have not checked ...
| May 3, 5:23 pm 2008 |
| Vefa Bicakci | [Bisected Regression] snd-intel-hda no longer produces sound...
Hello,
I am using Debian Sid on a Toshiba Satellite A100 series laptop.
This laptop has an Intel HDA sound card. It uses the Realtec codecs.
After modprobe'ing snd-intel-hda, the following appears in dmesg:
"hda_codec: Unknown model for ALC861, trying auto-probe from BIOS..."
The sound output used to work fine with 2.6.24.x and 2.6.25. In an
unrelated move, I decided to give 2.6.25-gitX series a try; however,
I was very sad to notice that sound output became disabled. Programs
such as ogg123 or ...
| May 3, 4:57 pm 2008 |
| Adrian Bunk | [2.6 patch] fix asm-alpha/types.h breakage
This patch fixes the following compile error on alpha caused by
commit 3726c23df8e4d95b6f2b335dfa90e3f4850a8a00
(alpha: types: use <asm-generic/int-*.h> for the alpha architecture):
<-- snip -->
...
CC arch/alpha/kernel/asm-offsets.s
In file included from include2/asm/topology.h:6,
from /home/bunk/linux/kernel-2.6/git/linux-2.6/include/linux/topology.h:34,
from /home/bunk/linux/kernel-2.6/git/linux-2.6/include/linux/mmzone.h:683,
...
| May 3, 4:51 pm 2008 |
| devzero | Re: modprobe aha1542 oops with git17
this also happens on real hardware with latest git - no aha1542 attached
didn`t happen before - looks like a regression to me
List: linux-kernel
Subject: modprobe aha1542 oops with git17
From: "Roland" <devzero () web ! de>
Date: 2008-05-01 23:49:54
2.6.25-git17-default on vmware - i`m getting oops when modprobe aha1542 (ok,
makes no sense in vmware, but i think this is a regression because it never
happend before. i did the same test with older kernels)
B...
| May 3, 4:29 pm 2008 |
| James Bottomley | Re: modprobe aha1542 oops with git17
Actually, if you check with our tool sets:
http://www.kerneloops.org/search.php?search=scsi_destroy_command_freelist&btnG=Fun...
They'll point you to the bugzilla entry:
http://bugzilla.kernel.org/show_bug.cgi?id=10598
And the fix (which is now upstream):
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=61d7416a28...
James
--
| May 3, 6:23 pm 2008 |
| Vegard Nossum | [PATCH] headerdep: a tool for detecting inclusion cycles in ...
Hi,
Here comes v2 of this tool. I've addressed most of your nitpicks and specific
comments. Running 'make headerdep', will go through all headers (instead of
source files as v1 did). Running with the --graph option will produce a
dependency graph in the 'dot' language (GraphViz).
Note that I am not really making an attempt to include this in the kernel
itself; I am submitting it in the hope that somebody will find it useful.
For Sam: Writing this in sparse is a waste of time. Sparse is good for...
| May 3, 4:27 pm 2008 |
| Bartlomiej Zolnierki... | [PATCH] 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>
---
drivers/ide/ide-io.c | 32 +++++---------------------------
1 file changed, 5 insertions(+), 27 deletions(-)
Index: b/drivers/ide/ide-io.c
===================================================================
--- a/drivers/ide/...
| May 3, 4:39 pm 2008 |
| devzero | [PATCH] mdacon messing up default vc's, set default to
mdacon incorrectly detects MDA hardware on systems without such graphics card.
one may load this module by chance, for example when doing some systematical
module-testing, and if there is no Monochrome Display Adapter attached , module init
renders vc1-16 completely unusable.
me and others have run into this more than once. see [Bug 224522 - modprobe mdacon
freezes machine -> https://bugzilla.novell.com/show_bug.cgi?id=224522 ] for example
apparently, proper MDA detection seems to be brok...
| May 3, 4:09 pm 2008 |
| Thomas Gleixner | [GIT pull} hrtimer updates
Linus,
please pull hrtimer updates from
ssh://master.kernel.org/pub/scm/linux/kernel/git/tglx/linux-2.6-hrt.git master
- permission fixups for /sys/devices/.../clocksource
- fold duplicated inline function into hrtimer.h
Thanks,
tglx
---
Heiko Carstens (2):
clocksource: Fix permissions for available_clocksource
clocksource: allow read access to available/current_clocksource
Oliver Hartkopp (1):
hrtimer: remove duplicate helper function
include/linux/hrtimer...
| May 3, 3:44 pm 2008 |
| Linus Torvalds | Linux 2.6.26-rc1
So this merge window was somewhat rocky in the sense that there was a lot
of arguments about it, but at the same time I at least personally think
that from a technical angle, we had somewhat less scary stuff going on
than has been almost the rule lately.
Lots of changes, but nothing that really feels all that fragile to me.
Famous last words. I expect that the x86 PAT support (which has been long
in the making) has the potential to have some issues, but the obvious
problems were hashed out...
| May 3, 3:38 pm 2008 |
| Alistair John Strachan | Re: Linux 2.6.26-rc1
drivers/built-in.o: In function `xc2028_get_reg':
tuner-xc2028.c:(.text+0x6b613): undefined reference to `i2c_transfer'
drivers/built-in.o: In function `load_firmware':
tuner-xc2028.c:(.text+0x6bf90): undefined reference to `i2c_transfer'
drivers/built-in.o: In function `generic_set_freq':
tuner-xc2028.c:(.text+0x6cdff): undefined reference to `i2c_transfer'
tuner-xc2028.c:(.text+0x6ce65): undefined reference to `i2c_transfer'
tuner-xc2028.c:(.text+0x6ceea): undefined reference to `i2c_transfer'
d...
| May 3, 5:47 pm 2008 |
| Alistair John Strachan | Re: Linux 2.6.26-rc1
Amazingly, even if I say Y to this option and disable all tuners, it still
tries (and fails) to compile this file. What is pulling in this tuner-core
dependency?
drivers/built-in.o: In function `v4l2_i2c_drv_detach_legacy':
tuner-core.c:(.text+0x72ba9): undefined reference to `i2c_detach_client'
drivers/built-in.o: In function `v4l2_i2c_drv_probe_legacy':
tuner-core.c:(.text+0x72bff): undefined reference to `i2c_probe'
drivers/built-in.o: In function `v4l2_i2c_drv_attach_legacy':
tuner-core.c:...
| May 3, 5:53 pm 2008 |
| Robert Hancock | Re: Remove Asus EEE UDMA/33 limitation.
Given that the bootup ACPI _GTM results will be used by Windows to
determine the DMA mode setting with the standard Microsoft atapi.sys
driver which PATA controllers would typically use, and Windows is what
the majority of this hardware was designed to run, I think it should be
quite trustworthy..
--
| May 3, 3:34 pm 2008 |
| Ingo Molnar | [patch] irda: fix !PNP support in drivers/net/irda/nsc-ircc.c
x86.git testing found the following build failure in latest -git:
drivers/built-in.o: In function `nsc_ircc_pnp_probe':
nsc-ircc.c:(.text+0xdf1b6): undefined reference to `pnp_get_resource'
nsc-ircc.c:(.text+0xdf1d4): undefined reference to `pnp_get_resource'
nsc-ircc.c:(.text+0xdf1ee): undefined reference to `pnp_get_resource'
nsc-ircc.c:(.text+0xdf237): undefined reference to `pnp_get_resource'
nsc-ircc.c:(.text+0xdf24c): undefined reference to `pnp_get_resource'
drivers/built-in.o:nsc-i...
| May 3, 3:32 pm 2008 |
| Ulrich Drepper | [PATCH] tiny mq_open optimization
A very small cleanup for mq_open. We do not have to call set_close_on_exit
if we create the file descriptor right away with the flag set. We have a
function for this now. The resulting code is smaller and a tiny bit faster.
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 94fd3b0..b3b69fd 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -673,7 +673,7 @@ asmlinkage long sys_mq_open(const char __user *u_name, int oflag, mode_t mode,
if ...
| May 3, 3:28 pm 2008 |
| Adrian Bunk | [2.6 patch] fix asm-mips/types.h syntax error
This patch fixes the following compile error caused by
commit 23cf11ddb5099f8c7f7cb3eb154bff0faf31cae9
(mips: types: use <asm-generic/int-*.h> for the mips architecture):
<-- snip -->
...
CC kernel/bounds.s
In file included from /home/bunk/linux/kernel-2.6/git/linux-2.6/include/linux/types.h:12,
from /home/bunk/linux/kernel-2.6/git/linux-2.6/include/linux/page-flags.h:8,
from /home/bunk/linux/kernel-2.6/git/linux-2.6/kernel/bounds.c:9:
incl...
| May 3, 3:26 pm 2008 |
| H. Peter Anvin | Re: [2.6 patch] fix asm-mips/types.h syntax error
Acked-by: H. Peter Anvin <hpa@zytor.com>
Went through the other patches, too, to verify none of the other
architectures were similarly affected.
-hpa
--
| May 3, 3:36 pm 2008 |
| Harvey Harrison | [RFC PATCH] lib: add ascii hex helper functions
Everyone rolls their own version around the tree, centralize
in lib/hexdump.c. Move the only existing users of hex_asc over
to the hi/lo helpers.
Add helpers for reading binary values from a stream of hex chars.
Most existing users want to know how many chars were read as well.
This implementation chooses to return the number of chars read
and takes a pointer to the value being read.
As an alternative, a char **buf could be passed and the value could
be returned and the pointer updated by refer...
| May 3, 3:27 pm 2008 |
| Gerhard Mack | [2.6.25.1] qla2xxx can't see more than 4 drives
I have an array witjh 6 drives in it and if use the card's utilities on
boot it sees all of them but for some reason Linux only sees 4 drives.
Am I running into a limit? I can see both drives if I remove other drives
from the array.
[0:0:0:0] disk MAXTOR ATLAS10K4_36SCA DFM0 /dev/sda
[0:0:1:0] disk MAXTOR ATLAS10K4_36SCA DFM0 /dev/sdb
[0:0:2:0] disk MAXTOR ATLAS10K4_36SCA DFM0 /dev/sdc
[0:0:6:0] process PE/PV 1x3 SCSI BP 0.28 -
fiber channel drives:
...
| May 3, 3:17 pm 2008 |
| Ulrich Drepper | [PATCH v3] unified (weak) sys_pipe implementation
[Empty message]
| May 3, 3:10 pm 2008 |
| Adrian Bunk | nommu compile error
Commit 9f3acc3140444a900ab280de942291959f0f615d (split linux/file.h)
causes the following compile error:
<-- snip -->
...
CC fs/proc/task_nommu.o
/home/bunk/linux/kernel-2.6/git/linux-2.6/fs/proc/task_nommu.c: In function 'task_mem':
/home/bunk/linux/kernel-2.6/git/linux-2.6/fs/proc/task_nommu.c:49: error: dereferencing pointer to incomplete type
make[3]: *** [fs/proc/task_nommu.o] Error 1
<-- snip -->
cu
Adrian
--
"Is there not promise of rain?" Ling ...
| May 3, 2:41 pm 2008 |
| Jay Cliburn | Need help debugging memory corruption
I'm trying to track down a memory corruption bug in the atl1 network
device driver that is exposed only when operating with 4GB or more
memory. (NB: The driver uses a 32-bit DMA mask.) The bug is hit under
certain conditions whenever the network interface is commanded down.
The information provided here is against recent -git, but this problem
also afflicts the atl1 driver in kernels 2.6.2[345].y. Dmesg and
config attached.
I can reproduce the bug at will by simply using scp to copy a few
hun...
| May 3, 2:09 pm 2008 |
| Ulrich Drepper | [PATCH v2] unify sys_pipe implementation
The second version of this patch with Linus' comments addressed. I've
changed the code for a few architectures to use a new name (<arch>_pipe
instead of sys_pipe) as per Linus' suggestion. I haven't touched MIPS
and Sparc, they use too much magic in this area. Should be easy enough
for the arch maintainer to clean up. sys_pipe is now unconditionally
created in fs/pipe.c and they will lead to link errors in case the
arch code isn't cleaned up.
Some arch maintainers might want to take a loo...
| May 3, 2:01 pm 2008 |
| Linus Torvalds | Re: [PATCH v2] unify sys_pipe implementation
No, definitely not.
That said, I think that in order to not break other architectures, and to
make it even easier to do this transformation, how about we just mark the
generic version with __weak?
That way, odd architectures can just continue to call their own versions
"sys_pipe()", and we don't break them unnecessarily.
Linus
--
| May 3, 2:40 pm 2008 |
| Adrian Bunk | s390 kvm_virtio.c build error
Commit c45a6816c19dee67b8f725e6646d428901a6dc24
(virtio: explicit advertisement of driver features)
and commit e976a2b997fc4ad70ccc53acfe62811c4aaec851
(s390: KVM guest: virtio device support, and kvm hypercalls)
don't like each other:
<-- snip -->
...
CC drivers/s390/kvm/kvm_virtio.o
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/s390/kvm/kvm_virtio.c:224: error: unknown field 'feature' specified in initializer
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/s390/kvm/kvm_vi...
| May 3, 1:47 pm 2008 |
| Oleg Nesterov | [PATCH 3/4] posix timers: timer_delete: remove the bogus "-&...
sys_timer_delete() and itimer_delete() check "timer->it_process != NULL", this
looks completely bogus. ->it_process == NULL means that this timer is already
under destruction or it is not fully initialized, this must not happen.
sys_timer_delete: the timer is locked, and lock_timer() can't succeed
if ->it_process == NULL.
itimer_delete: it is called by exit_itimers() when there are no other
threads which can play with signal_struct->posix_timers.
Signed-off-by: Oleg Nesterov <...
| May 3, 1:35 pm 2008 |
| Oleg Nesterov | [PATCH 0/4] posix timers: misc fixes
Untested, sorry, please review.
The patches don't depend on each other.
(I won't access to computers/internet from 4 evening till May 10).
Oleg.
--
| May 3, 1:35 pm 2008 |
| Oleg Nesterov | [PATCH 4/4] posix timers: release_posix_timer: kill the bogu...
release_posix_timer() can't be called with ->it_process != NULL. Once
sys_timer_create() sets ->it_process it must not call release_posix_timer(),
otherwise we can race with another thread doing sys_timer_delete(), this timer
is visible to idr_find() and unlocked.
The same is true for two other callers (actually, for any possible caller),
sys_timer_delete() and itimer_delete(). They must clear ->it_process before
unlock_timer() + release_posix_timer().
Signed-off-by: Oleg Nesterov <ol...
| May 3, 1:35 pm 2008 |
| Oleg Nesterov | [PATCH 2/4] posix timers: sigqueue_free: don't free sigqueue...
Currently sigqueue_free() removes sigqueue from list, but doesn't cancel the
pending signal. This is not consistent, the task should either receive the
"full" signal along with siginfo_t, or it shouldn't see the signal at all.
Change sigqueue_free() to clear SIGQUEUE_PREALLOC but leave sigqueue on list
if it is queued.
Note: I am not sure we shouldn't do the opposite, free sigqueue + cancel the
pending signal, but this needs some ugly changes. Perhaps we should reconsider
this change later. See a...
| May 3, 1:35 pm 2008 |
| Linus Torvalds | Re: [PATCH 2/4] posix timers: sigqueue_free: don't free sigq...
You know what, I think there might be an even simple solution.
How about just setting a bit saying it is canceled - and nothing more.
Then, the dequeue logic can be just taught to ignore those things.
Doesn't that sound like the simple way to cancel signals? Make
collect_signal() just do a "return 0" if the signal has been flushed..
Linus
--
| May 3, 1:51 pm 2008 |
| Oleg Nesterov | Re: [PATCH 2/4] posix timers: sigqueue_free: don't free sigq...
I see the light! Will return on May 10.
(Damn, you have proved I'm stupid twice on the same issue! OK, OK, thanks ;)
Oleg.
--
| May 3, 3:19 pm 2008 |
| Oleg Nesterov | Re: [PATCH 2/4] posix timers: sigqueue_free: don't free sigq...
In that case perhaps we can do something like (uncompiled, just for illustration)
patch below.
While I don't know what is the "right" behaviour, I hate this patch because
it "looks ugly" (even if correct).
Oleg.
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1709,7 +1709,7 @@ extern int send_sig(int, struct task_str
extern void zap_other_threads(struct task_struct *p);
extern int kill_proc(pid_t, int, int);
extern struct sigqueue *sigqueue_alloc(void);
-extern void sigqueu...
| May 3, 1:43 pm 2008 |
| Oleg Nesterov | [PATCH 1/4] posix timers: fix sigqueue_free() vs __exit_sign...
__exit_signal() does flush_sigqueue(tsk->pending) outside of ->siglock.
This can race with another thread doing sigqueue_free(), we can free
the same SIGQUEUE_PREALLOC sigqueue twice or corrupt the pending->list.
Note that even sys_exit_group() can trigger this race, not only
sys_timer_delete().
Move the callsite of flush_sigqueue(tsk->pending) under ->siglock.
This patch doesn't touch flush_sigqueue(->shared_pending) below, it is
called when there are no other threads which c...
| May 3, 1:35 pm 2008 |
| previous day | today | next day |
|---|---|---|
| May 2, 2008 | May 3, 2008 | May 4, 2008 |
| Jeremy Allison | Re: [RFC] Heads up on sys_fallocate() |
| Greg KH | [GIT PATCH] driver core patches against 2.6.24 |
| Joerg Roedel | [PATCH 03/34] AMD IOMMU: add defines and structures for ACPI scanning code |
| Eric W. Biederman | [PATCH] powerpc pseries eeh: Convert to kthread API |
| David Miller | [GIT]: Networking |
| Gerrit Renker | [PATCH 27/37] dccp: Integration of dynamic feature activation - part 2 (server side) |
| Natalie Protasevich | [BUG] New Kernel Bugs |
git: | |
