While I was in vacation, I noticed that more "tsk == current" check were added to the signal logic because of the way signalfd fetches other task signals. I'm playing at the moment with this patch, that recall Ben's idea of attaching to the sighand only during read/poll, and calling dequeue_signal() only with "current". This simplifies the signalfd logic quite a bit. If this patch is applied, a task calling signalfd can read its own private signals, and its own group signals. Comments? -...

You miss the whole point of dual licencing: Sam has stated in the licence that the code can be distributed under the terms of the BSD licence, or alternatively it can be distributed under the terms of the GPLv2. Noone removed Sam's licence. Sam has offered a choice, and if you choose one of the two offered licences when distributing his code that complies with what he stated in his copyright notice. IANAL, but if reyk contributed to dual licenced code keeping the file dual licenced it...

Package: linux-image-2.6.22-2-amd64 Followup-For: Bug #438663 see: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=438663 /var/log/dmesg - = - = - = - = - = - = - = - = - = - = - = - = - = - = - = - = - = - Linux version 2.6.22-2-amd64 (Debian 2.6.22-4snapshot.9420) (waldi@debian.org) (gcc version 4.1.3 20070718 (prerelease) (Debian 4.1.2-14+1)) #1 SMP Fri Aug 31 08:43:40 UTC 2007 Command line: root=/dev/sdb2 ro video=nvidiafb:ywrap,mtrr vga=791 BIOS-provided physical RAM map: BIOS-e820: 000...

Hi, today I switched from 2.6.22.3 to 2.6.23-rc5 (skipped quite a few -rc versions due to lack of time), and the box keeps panicking under certain circumstances. I suspected disk related problems, because: when the box is up, I usually resume ~10 bittorrent files. When doing this, each file (~200MB...1GB) is checked and disk activity is pretty high (20MB/s or so), and after 1 minute of doing so the box panicks. Every time. However, I could not reproduce it while generating disk-io with say tar...

Added thanks. I guess it's .23 material. -Andi -

Daniel, Thanks for your help tracking down this bug. Maybe we can close the bugzilla report now. -- -Stephane -

Hi all! [ I'm not subscribed to this list so please CC me on your replies. ] This patch is against kernel-2.6.23-rc4, and aims to enhance the "make xconfig" search dialog. At the moment what it does is allow searching on the "help" field of the menus as well as the "name" of the symbol. Try searching for "you will get a watchdog" before and after the patch to see what I mean. I'm next planning to add a radio button for choosing between regex match and substring match, as well as implement...

with the macro max provided by <linux/kernel.h>, so changed its name to a more proper one: limit Signed-off-by: Denis Cheng <crquan@gmail.com> --- net/netlink/af_netlink.c | 12 ++++++------ 1 files changed, 6 insertions(+), 6 deletions(-) diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 8bb14e3..ac3b100 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1885,7 +1885,7 @@ static int __init netlink_proto_init(void) { struct sk_buff ...

Hi, Just wondering if there is a newer version of intel-rng out of tree or whether modern Intel chipsets have a usable RNG. I haven't been able to get intel-rng loading (no such device) on anything from the p965, p975 or p35 chipsets. Best, Shane -- http://www.cm.nu/~shane/ -

Which is a major lossage :( -hpa -

Oh dear - Theo, go talk to a lawyer, or do a course on licencing. The owner generally starts with the rights to control who performs acts covered by copyright law. They pass some of those rights on to others by contract, licence or statutory means. It is quite normal for the owner to pass on the right to relicence or modify the licencing of a work. In many cases the owner actually hands on all such rights to a third party (eg an evil music company). [Owner and author often differ as many legal ...

What myth? The myth that Theo understands dual licensing? Jeff -

It's no longer dual licenced in the FreeBSD tree because the FreeBSD people removed the GPL choice of the dual licenced code 3 months ago. So all of Theo's accusations of people breaking the law by making this Both the FreeBSD and Linux people draw the logical conclusion that this "Alternatively" means everyone can always choose to remove one of the two choices alternatively offered. cu Adrian -- "Is there not promise of rain?" Ling Tan asked suddenly out of the d...

Sorry, this has been a thinko on my side: If noone except you and onoe-san made any contributions to this code that were non-trivial enough for automatically giving its author a copyright on his contributions (whatever this means in various 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 - Dr...

OK, I begin to understand this, there seem to be three different types of files changed by Jiri's patch: 1. dual licenced files planned to make GPL-only 2. previously dual licenced files with a too recent version used planned to make GPL-only 3. never dual licenced files planned to make GPL-only For files under 1. and 2. Reyk did contribute to dual licenced code without touching the licence, but I missed that there's also code unter 3. So there is a problem, but not with the code under 1...

I urge developers to not bait into this and just leave this alone. Those involved know what they are doing and have a strong team of attorneys watching their backs. Any *necessary* discussions are be done privately. Luis -

Where exactly do you see Reyk's copyright in the above quote? http://www.openbsd.org/cgi-bin/man.cgi?query=ath&sektion=4#AUTHORS C. -

This is more of a user question; please direct me to a kernel user list if that's more appropriate. I am trying to find out if NCQ and port multiplier are suppored in the nVidia nForce 590 chipset? Specifically, for the Asus Crosshair mobo. <http://www.asus.com/products4.aspx?modelmenu=2&model=1283&l1=3&l2=117&l3=0&l4=0> Thanks, --Yan -- o__ ,>/'_ o__ (_)\(_) ,>/'_ o__ Yan Seiner (_)\(_) ,>/'_ o__ ...

The warnings, drivers/char/hpet.c:72: warning: 'clocksource_hpet' defined but not used drivers/char/hpet.c:81: warning: 'hpet_clocksource' defined but not used from, commit 3b2b64fd311c92f2137eb7cee7025794cd854057 Author: Linus Torvalds <torvalds@woody.linux-foundation.org> Date: Fri Aug 31 20:13:57 2007 -0700 This patch eliminates the warnings when the clocksoure isn't used. It also removes some other unused stuff that goes along with the clocksource .. I don't have access to ...

Although it seems to pick the right version the following namespace conflict should be resolved: <-- snip --> ... CC kernel/sched.o /home/bunk/linux/kernel-2.6/linux-2.6.23-rc4-mm1/kernel/sched.c:676:1: warning: "RSR" redefined In file included from include2/asm/bitops.h:18, from /home/bunk/linux/kernel-2.6/linux-2.6.23-rc4-mm1/include/linux/bitops.h:17, from /home/bunk/linux/kernel-2.6/linux-2.6.23-rc4-mm1/include/linux/thread_info.h:20, ...

Hi Linus, Please pull from the "for_linus" branch of the repository at git pull git://git.linux-nfs.org/pub/linux/nfs-2.6.git for_linus This will update the following files through the appended changesets. Cheers, Trond ---- fs/nfs/file.c | 2 +- fs/nfs/namespace.c | 2 +- fs/nfs/nfs4proc.c | 4 ++-- fs/nfs/super.c | 20 ++++++++++---------- fs/nfs/write.c | 44 ++++++++++++++++++++++++++++++++++++++++++++ include/linux/nfs_fs.h ...

I just added three branches to 'ALL' branch of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/misc-2.6.git dmi-const DMI constification pciseg x86 PCI domain support, newly updated for the latest Muli stuff upstream isdn-cleanups Obvious and simple cleanups for ISDN All are targetted for 2.6.24. Just sent out dmi-const and pciseg to LKML. Sent isdn-cleanups to LKML, ISDN list and Karsten a while ago. The "super mondo ISDN cleanups", converting ISDN HiSax drivers to PCI/P...

I don't think this should be a config option. The second function is redundant? -Andi -

Then the other function is redundant. -Andi -

commit 457b6eb3bf3341d2e143518a0bb99ffbb8d754c4 Author: Jeff Garzik <jeff@garzik.org> Date: Sat Sep 1 10:16:45 2007 -0400 drivers/firmware: const-ify DMI API and internals Three main sets of changes: 1) dmi_get_system_info() return value should have been marked const, since callers should not be changing that data. 2) const-ify DMI internals, since DMI firmware tables should, whenever possible, be marked const to ensure we never ever write ...