Bug#639691: updating build-deps for nfs-utils to ease backporting to squeeze
On Tue, Aug 30, 2011 at 07:25:58AM +0200, Luk Claes wrote: On 08/30/2011 06:16 AM, Daniel Kahn Gillmor wrote: I concur with Sean Finney that nfs-utils should Build-Depend on libnfsidmap-dev = 0.24 to ease backporting. I'm hoping to prepare nfs-utils 1.2.4 as a backport for squeeze, and it'd be nice to modify the source package as minimally as possible. What features do you need/want from 1.2.4? also, for those using kerberos + nvfsv4, it fixes some rather annoying bugs (like not working at all) for environments where users have a large list of group memberships (see 5604b35a and to a lesser extent 9274e94d). i shoudl add that minus the nfsidmap not building (and thus breaking after the dh additions to install the files), the package would otherwise build fine with no other needed modifications with the existing build-depends. so i guess it's a choice between whether you want to substitute uniformity (i.e. all builds having nfsidmapd by requiring also a backport of libnfsidmap) vs flexibility (builds are slightly more non-deterministic, based on what version of libnfsidmap is installed). i don't have a strong opinion either way :) sean -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110831085341.ga31...@cobija.connexer.com
Bug#639691: nfs-utils: Needs versioned build-dep on libnfsidmap-dev = 0.24
Package: nfs-utils Version: 1.2.4-1 Severity: normal The upstream m4/autofoo only enables nfsidmapd related stuff if it finds certain api functions available, which after some investigation appears to have been introduced in 0.24. This will otherwise cause FTBFS's for anyone wanting to backport to an older debian/ubuntu release, as I've just found out, due to commit 3d3bc354 which unconditionally tries to install manpages and binaries for it. I suggest eitehr adding the strict build-deps, or alternatively making the dh logic a bit smarter for whether it should attempt to install them. -- System Information: Debian Release: 6.0.2 APT prefers stable APT policy: (500, 'stable') Architecture: amd64 (x86_64) Kernel: Linux 2.6.32-5-xen-amd64 (SMP w/1 CPU core) Locale: LANG=en_US.UTF8, LC_CTYPE=en_US.UTF8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110829121916.25276.46234.reportbug@localhost.localdomain
Re: anyone running a PS3
(d-k folks, please let me know if you're feeling spammed by the cross-posting) okay, reporting back in on this... using the 2.6.26.5 kernel from the debian kernel snapshot repo, the system successfully boots up! however, there seems to be another, perhaps more subtle problem lurking in this kernel too. the symptoms are a system which gradually slows down, to the point of being unresponsive and needing a hard-reboot, and individual processes hanging in an unkillable state. typically this takes about an hour or two. the first time it happened i was in the middle of installing about a gig of packages (apt-get install gnome...), another time it was while trying to play a video. occassionally kernel log messages pop up regarding what seems to be the wireless component of the gelic module (i'm not using it, but maybe n-m is tickling it), and sometimes specific proccesses (apt-get in the previous paragraph) will hang; several minutes after a kill -9 the kernel will issue some kind of message about the task not responding. sorry this isn't really specific, i didn't have any kind of console logging set up (i'll get a copy of it hte next time it happens). i'll also try installing some vanilla upstream kernels to see if they suffer from the same problems... sean On Sat, Oct 11, 2008 at 03:36:08PM -0400, Norberto Feliberty wrote: Date: Sat, 11 Oct 2008 11:10:44 +0200 From: [EMAIL PROTECTED] To: [EMAIL PROTECTED] CC: [EMAIL PROTECTED]; debian-kernel@lists.debian.org Subject: Re: anyone running a PS3 (cc'ing debian-kernel now too, for some feedback) d-k folks: the issue is that the ps3 fails to boot with the latest lenny 2.6.26 kernels, due to what appears to be a kernel bug which has been fixed upstream). my original post: http://lists.debian.org/debian-powerpc/2008/10/msg00029.html so where was i... yes, it seems that it's in fact a kernel bug. i updated to the latest kboot, which also had the same problem booting the kernel. then googling for ps3 2.6.26 leads to the following: http://www.yellowdog-board.com/viewtopic.php?f=19t=4063start=0st=0sk=tsd=a (scroll down a bit searching for 2.6.26) which points to: http://www.gossamer-threads.com/lists/linux/kernel/972382 which mentions an obscure intterupt handling bug causing the problem, and which points to: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff_plain;h=de85422b94ddb23c021126815ea49414047c13dc which was accepted into the -stable queue for 2.6, which was then released as part of 2.6.26.6: http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.26.y.git;a=commit;h=90e21dd5346538810cff7f5fa2d3b0ae4c88989d so thanks for the pointer. d-k folks: is there any feasibility for getting this fix into lenny, or am I just SoL? sean _ See how Windows connects the people, information, and fun that are part of your life. http://clk.atdmt.com/MRT/go/msnnkwxp1020093175mrt/direct/01/ -- signature.asc Description: Digital signature
Re: anyone running a PS3
(cc'ing debian-kernel now too, for some feedback) d-k folks: the issue is that the ps3 fails to boot with the latest lenny 2.6.26 kernels, due to what appears to be a kernel bug which has been fixed upstream). my original post: http://lists.debian.org/debian-powerpc/2008/10/msg00029.html so where was i... yes, it seems that it's in fact a kernel bug. i updated to the latest kboot, which also had the same problem booting the kernel. then googling for ps3 2.6.26 leads to the following: http://www.yellowdog-board.com/viewtopic.php?f=19t=4063start=0st=0sk=tsd=a (scroll down a bit searching for 2.6.26) which points to: http://www.gossamer-threads.com/lists/linux/kernel/972382 which mentions an obscure intterupt handling bug causing the problem, and which points to: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff_plain;h=de85422b94ddb23c021126815ea49414047c13dc which was accepted into the -stable queue for 2.6, which was then released as part of 2.6.26.6: http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.26.y.git;a=commit;h=90e21dd5346538810cff7f5fa2d3b0ae4c88989d so thanks for the pointer. d-k folks: is there any feasibility for getting this fix into lenny, or am I just SoL? sean signature.asc Description: Digital signature
Re: anyone running a PS3
On Sat, Oct 11, 2008 at 01:35:33PM +0200, maximilian attems wrote: yes 2.6.26.6 will be in the next upload use sid branch snapshots for now - http://wiki.debian.org/DebianKernel ooh, i didn't know about this service! i'll use that for the time being, thanks! sean -- signature.asc Description: Digital signature
Bug#501923: linux-image-2.6-powerpc64: PS3 Currently unbootable (fixed in 2.6.26.6)
Package: linux-image-2.6-powerpc64 Version: 2.6.26+16 Severity: important -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 hi there, this is a placeholder bug to document/track the regression with ps3's being unbootable with anything from 2.6.26 before 2.6.26.6. the issue was brought up on the -powerpc and -kernel lists: http://lists.debian.org/debian-powerpc/2008/10/msg00029.html and the fix is included in the upcoming upload of 2.6.26 from the d-k snapshot repo, information for which is available at: http://wiki.debian.org/DebianKernel sean -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (GNU/Linux) iD8DBQFI8NOJynjLPm522B0RAmAUAJwKu+TohkIsZDKLwLHVaywpDZrsVQCcCpBd NeFKqfP+YBicC9Ere10OjYc= =LusJ -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: HighPoint- GPL Licensed Controller wants To be Include In Debian Distribution
Hi May, (This should all be prefaced with the statement that i don't have a great deal of experience with kernel module packaging, someone from debian-kernel may have more insight than me) First I should say it is very thoughtful of you to contact the debian community regarding your drivers, if only more members in the industry would do so! Regarding packaging the kernel module: i suppose the best way to package the module for kernel versions before the mainline inclusion of your driver is to make a module-assistant-compatible package, which is about the easiest way one can install kernel modules in debian systems. take a look at the documentation for module-assistant, specifically: /usr/share/doc/module-assistant/HOWTO-DEVEL.gz (though this assumes a certain level of familiarity with debian packaging). and maybe use a couple other module source packages for reference as well. more information is also available at: http://wiki.debian.org/ModuleAssistant though this is more user-oriented. after you have created an initial version of your package, you can always feel free to request feedback or review (via debian-devel/debian-kernel, or via IRC), and eventually sponsorship of your package for inclusion into debian if you feel it's relevant. typically this last step is done by filing an RFS bug against the pseudo package wnpp via reportbug. if you need help with packaging basics, there's also the debian-mentors list, which is also a good place to find a package sponsor. anyway, hope that information is helpful, sean On Monday 28 January 2008 05:59:10 pm May Hwang wrote: Dear Debian Developer, This is May Hwang, Product Manager from HighPoint Technologies. HighPoint have launched a new series of H/W RAID controllers based on Intel 2nd generation PCI-express I/O processor, one main advantage is we are the only manufacturer integrate this Intel Fastest SATA I/O processor on HighPoint 4 and 8 ports controllers compares to others RAID controllers use 1st generation PCI-express of I/O Processor. These Hardware RAID controllers offer GPL licensed Linux Open Source driver and have been accepted into 2.6.25 main kernel tree but since this is not a stable kernel version yet so our Debian system integrators still have to go through driver compilation in every installation process. Therefore, our Debian customers request HighPoint must work with Debian developer to include our linux drivers into the latest Debian Distribution. HighPoint is always looking ways to provide friendly use experience for customers so we will assign a dedicated firmware interface to work with Debian developer on this project. We are looking forward to provide better support for Debian customers, let us know how to move forward? Best Regards, May Hwang HighPoint Technologies,Inc. Tel:408-240-6118 Fax-408-942-5800 http://www.highpoint-tech.com www.highpoint-tech.com http://www.hptmac.com www.hptmac.com Distribution Partners: ASI, BellMicro, DH, Malabs RocketRAID - Terabyte Storage Technologies signature.asc Description: This is a digitally signed message part.
Bug#443373: linux-image-2.6.18-4-ixp4xx: kernel null paging request, crash from bitmapped md devices
hi martin, On Wednesday 17 October 2007 12:32:02 am Martin Michlmayr wrote: Sean, did you have a chance to tese these kernel images I prepared? If not, I suggest we close this bug saying that it has been fixed in newer kernels. i haven't yet, but i would really like to try, still. i worked a bit last week to see if i could reproduce this in qemu (so i wouldn't need to do the whole reboot/flash/reinstall-kernel thing so many times), but without luck. it could be that the qemu instance didn't work because it was a different kernel subarchitecture (it doesn't provide an ixp4xx one), or because the devices i setup (loopback based files that were a couple hundred megs) were not big enough to trigger it. i'll try and look at it more this weekend. sean signature.asc Description: This is a digitally signed message part.
Bug#443373: linux-image-2.6.18-4-ixp4xx: kernel null paging request, crash from bitmapped md devices
hi martin,
On Wednesday 26 September 2007 11:15:12 am Martin Michlmayr wrote:
* sean finney [EMAIL PROTECTED] [2007-09-21 09:10]:
i don't really have the resources to try compilingpatched arm kernels,
but if you do and want me to test something just let me know.
I put two test kernels at http://merkel.debian.org/~tbm/nslu2/
If the a variant works, can you please also try the b variant (if
the a doesn't work, there's no need to try b).
like i mentioned on irc, a doesn't work, unfortunately. with a it doesn't
oops/crash, and instead will hang when mounting an md device with a bitmap.
the system is still useable while it hangs, though it will require a hard
reboot due to the hung mount. i'm pretty sur ethe patch is still relevant,
but i think there's more that needs to be done. so i've taken the difference
between 2.6.18 and 2.6.21 from upstream git, and cherry picked the commits
that looked like they might be relevant. below i've attached the commit logs
incl description and commit id's, as well as a patch of all of these
cherry-picked into the 2.6.18 version.
shall we give it another try with this?
sean
diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index ecc5676..1814ac0 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -479,9 +479,12 @@ static int bitmap_read_sb(struct bitmap *bitmap)
int err = -EINVAL;
/* page 0 is the superblock, read it... */
- if (bitmap-file)
- bitmap-sb_page = read_page(bitmap-file, 0, bitmap, PAGE_SIZE);
- else {
+ if (bitmap-file) {
+ loff_t isize = i_size_read(bitmap-file-f_mapping-host);
+ int bytes = isize PAGE_SIZE ? PAGE_SIZE : isize;
+
+ bitmap-sb_page = read_page(bitmap-file, 0, bitmap, bytes);
+ } else {
bitmap-sb_page = read_sb_page(bitmap-mddev, bitmap-offset, 0);
}
if (IS_ERR(bitmap-sb_page)) {
@@ -536,7 +539,7 @@ static int bitmap_read_sb(struct bitmap *bitmap)
printk(KERN_INFO %s: bitmap file is out of date (%llu %llu)
-- forcing full recovery\n, bmname(bitmap), events,
(unsigned long long) bitmap-mddev-events);
- sb-state |= BITMAP_STALE;
+ sb-state |= cpu_to_le32(BITMAP_STALE);
}
success:
/* assign fields using values from superblock */
@@ -544,11 +547,11 @@ success:
bitmap-daemon_sleep = daemon_sleep;
bitmap-daemon_lastrun = jiffies;
bitmap-max_write_behind = write_behind;
- bitmap-flags |= sb-state;
+ bitmap-flags |= le32_to_cpu(sb-state);
if (le32_to_cpu(sb-version) == BITMAP_MAJOR_HOSTENDIAN)
bitmap-flags |= BITMAP_HOSTENDIAN;
bitmap-events_cleared = le64_to_cpu(sb-events_cleared);
- if (sb-state BITMAP_STALE)
+ if (sb-state cpu_to_le32(BITMAP_STALE))
bitmap-events_cleared = bitmap-mddev-events;
err = 0;
out:
@@ -578,9 +581,9 @@ static void bitmap_mask_state(struct bitmap *bitmap, enum bitmap_state bits,
spin_unlock_irqrestore(bitmap-lock, flags);
sb = (bitmap_super_t *)kmap_atomic(bitmap-sb_page, KM_USER0);
switch (op) {
- case MASK_SET: sb-state |= bits;
+ case MASK_SET: sb-state |= cpu_to_le32(bits);
break;
- case MASK_UNSET: sb-state = ~bits;
+ case MASK_UNSET: sb-state = cpu_to_le32(~bits);
break;
default: BUG();
}
@@ -858,9 +861,7 @@ static int bitmap_init_from_disk(struct bitmap *bitmap, sector_t start)
/* We need 4 bits per page, rounded up to a multiple of sizeof(unsigned long) */
bitmap-filemap_attr = kzalloc(
- (((num_pages*4/8)+sizeof(unsigned long)-1)
- /sizeof(unsigned long))
- *sizeof(unsigned long),
+ roundup( DIV_ROUND_UP(num_pages*4, 8), sizeof(unsigned long)),
GFP_KERNEL);
if (!bitmap-filemap_attr)
goto out;
@@ -875,7 +876,8 @@ static int bitmap_init_from_disk(struct bitmap *bitmap, sector_t start)
int count;
/* unmap the old page, we're done with it */
if (index == num_pages-1)
-count = bytes - index * PAGE_SIZE;
+count = bytes + sizeof(bitmap_super_t)
+ - index * PAGE_SIZE;
else
count = PAGE_SIZE;
if (index == 0) {
@@ -1154,6 +1156,22 @@ int bitmap_startwrite(struct bitmap *bitmap, sector_t offset, unsigned long sect
return 0;
}
+ if (unlikely((*bmc COUNTER_MAX) == COUNTER_MAX)) {
+ DEFINE_WAIT(__wait);
+ /* note that it is safe to do the prepare_to_wait
+ * after the test as long as we do it before dropping
+ * the spinlock.
+ */
+ prepare_to_wait(bitmap-overflow_wait, __wait,
+ TASK_UNINTERRUPTIBLE);
+ spin_unlock_irq(bitmap-lock);
+ bitmap-mddev-queue
+-unplug_fn(bitmap-mddev-queue);
+ schedule();
+ finish_wait(bitmap-overflow_wait, __wait);
+ continue;
+ }
+
switch(*bmc) {
case 0:
bitmap_file_set_bit(bitmap, offset);
@@ -1163,7 +1181,7 @@ int bitmap_startwrite(struct bitmap *bitmap, sector_t offset, unsigned long sect
case 1:
*bmc = 2;
}
- BUG_ON((*bmc COUNTER_MAX) == COUNTER_MAX);
+
(*bmc)++;
spin_unlock_irq(bitmap-lock);
@@ -1201,6 +1219,9 @@ void bitmap_endwrite(struct bitmap *bitmap, sector_t offset, unsigned long secto
if (!success ! (*bmc NEEDED_MASK
Bug#443373: linux-image-2.6.18-4-ixp4xx: kernel null paging request, crash from bitmapped md devices
hiya, On Friday 21 September 2007 08:49:52 am you wrote: * sean finney [EMAIL PROTECTED] [2007-09-20 23:45]: Package: linux-image-2.6.18-4-ixp4xx same story as this guy: http://www.mail-archive.com/[EMAIL PROTECTED]/msg06810.html Thanks for the report. I'll try to reproduce this with the latest kernel this weekend. fyi, i just installed a 2.6.21 backport like the reporter, and it seems fixed. if you want to test a backported patch like the one i provided (and have the resources to compile an arm kernel), i'll happily test it :) sean signature.asc Description: This is a digitally signed message part.
Bug#443373: linux-image-2.6.18-4-ixp4xx: kernel null paging request, crash from bitmapped md devices
On Friday 21 September 2007 08:49:52 am Martin Michlmayr wrote: * sean finney [EMAIL PROTECTED] [2007-09-20 23:45]: Package: linux-image-2.6.18-4-ixp4xx same story as this guy: http://www.mail-archive.com/[EMAIL PROTECTED]/msg06810.html Thanks for the report. I'll try to reproduce this with the latest kernel this weekend. in case it's helpful, there's supposed to be a 2.6.21 backport in backports.org, but there was a problem with the arm upload or something to that effect. i've spoken with the arm/bpo buildd admin (kmuto) and he's going to fix the problem soon. looking in upstream git, i can't be sure but i think this might be the fix: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=da6e1a32fb8d7539a27f699c8671f64d7fefd0cc or at least the BUG_ON is right about where i saw it (give or take a few lines). i don't really have the resources to try compilingpatched arm kernels, but if you do and want me to test something just let me know. sean signature.asc Description: This is a digitally signed message part.
Bug#430904: ipw3945-modules-2.6.21-1-amd64: no module shipped in package
Package: ipw3945-modules-2.6.21-1-amd64 Version: 2.6.21+1.2.1-1 Severity: grave Justification: renders package unusable -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 i opened my laptop this morning to discover i no longer had a network driver :( rangda[/home/sean] dpkg-deb -c /var/cache/apt/archives/ipw3945-modules-2.6.21-1-amd64_2.6.21+1.2.1-1_amd64.deb drwxr-xr-x root/root 0 2007-06-25 22:42 ./ drwxr-xr-x root/root 0 2007-06-25 22:42 ./lib/ drwxr-xr-x root/root 0 2007-06-25 22:42 ./lib/modules/ drwxr-xr-x root/root 0 2007-06-25 22:42 ./lib/modules/2.6.21-1-amd64/ drwxr-xr-x root/root 0 2007-06-25 22:42 ./lib/modules/2.6.21-1-amd64/contrib/ drwxr-xr-x root/root 0 2007-06-25 22:42 ./lib/modules/2.6.21-1-amd64/contrib/ipw3945/ drwxr-xr-x root/root 0 2007-06-25 22:42 ./usr/ drwxr-xr-x root/root 0 2007-06-25 22:42 ./usr/share/ drwxr-xr-x root/root 0 2007-06-25 22:42 ./usr/share/doc/ drwxr-xr-x root/root 0 2007-06-25 22:42 ./usr/share/doc/ipw3945-modules-2.6.21-1-amd64/ - -rw-r--r-- root/root 3641 2007-06-25 22:42 ./usr/share/doc/ipw3945-modules-2.6.21-1-amd64/copyright - -rw-r--r-- root/root 589 2007-06-17 21:05 ./usr/share/doc/ipw3945-modules-2.6.21-1-amd64/changelog.Debian.gz - -- System Information: Debian Release: lenny/sid APT prefers unstable APT policy: (500, 'unstable'), (500, 'testing'), (500, 'stable') Architecture: amd64 (x86_64) Kernel: Linux 2.6.21-1-amd64 (SMP w/2 CPU cores) Locale: LANG=en_US.utf8, LC_CTYPE=en_US.utf8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFGg1FEynjLPm522B0RAsAyAJ9TTmlz2+jKdm9Bc4PFgrLJ9NatkgCfdpL6 t7Rnb+Jd9TzKnjQr+dCEoB4= =FsWv -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#339327: linux-2.6: don't display lilo-related messages unless lilo is installed
Package: linux-2.6 Severity: minor -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 hi, i sometimes get messages like the following: The link /vmlinuz.old is a damaged link Removing symbolic link vmlinuz.old Unless you used the optional flag in lilo, you may need to re-run lilo The link /initrd.img.old is a damaged link Removing symbolic link initrd.img.old Unless you used the optional flag in lilo, you may need to re-run lilo a simple which lilo could probably determine whether mentioning lilo were appropriate. sean - -- System Information: Debian Release: testing/unstable APT prefers testing APT policy: (990, 'testing'), (500, 'unstable') Architecture: i386 (i686) Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.14-1-686 Locale: LANG=en_US, LC_CTYPE=en_US (charmap=ISO-8859-1) -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.2 (GNU/Linux) iD8DBQFDefRlynjLPm522B0RAt2vAJ9cdSjNHB+Tp/bKg5kisGF4neMeBACfYo3Y FnpeLzJ8gHKVQ5sk5tzy874= =2zcD -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
