--- Begin Message ---
- To: submit@bugs.debian.org
- Subject: linux-image-2.6.26-2-686-bigmem: Does not recognize all CPUs on "summit" SMP machines
- From: "Andreas U. Trottmann" <andreas.trottmann@werft22.com>
- Date: Mon, 18 May 2009 17:11:41 +0200
- Message-id: <20090518151131.GA11118@werft22.net>
Package: linux-image-2.6.26-2-686-bigmem
Version: 2.6.26-15
Hello,
The Linux kernel sources upon which
linux-image-2.6.26-2-686-bigmem is based are suffering from a
bug, which only affects the (relatively rare) "summit" SMP
architecture. It is found in IBM xSeries servers such as the
x440 or x445.
On these servers, linux-image-2.6.26-2-686-bigmem only correctly
identifies and uses one CPU, regardless of the number of CPUs
actually installed.
The bug is documented and resolved on linux-kernel:
http://lkml.indiana.edu/hypermail/linux/kernel/0807.3/0493.html
The patch responsible for the resolution is the following:
--- a/include/asm-x86/mach-summit/mach_apic.h 2008-07-24 18:06:39.000000000 -0700
+++ b/include/asm-x86/mach-summit/mach_apic.h 2008-07-24 18:09:29.000000000 -0700
@@ -122,7 +122,7 @@ static inline physid_mask_t ioapic_phys_
static inline physid_mask_t apicid_to_cpu_present(int apicid)
{
- return physid_mask_of_physid(0);
+ return physid_mask_of_physid(apicid);
}
I report the bug against a lenny kernel, as I don't have any such system running
testing or unstable. The current linux kernel source in unstable, linux-2.6_2.6.29-5,
appears to have the file structure reorganized. The relevant file appears to be
arch/x86/include/asm/summit/apic.h, and the bug appears still to be present:
apicid_to_cpu_present still returns physid_mask_of_physid(0) instead of (apicid).
However, currently I cannot test whether the fix is still relevant.
If it is of help, I could try to boot an unpatched linux-2.6_2.6.29-5 kernel on a
summit machine and compare the number of identified CPUs with a patched linux-2.6_2.6.29-5
kernel.
In any case, it could probably be a good idea to include the fix in the next lenny
kernel update, as it does render those multi-CPU machines relatively useless, and
the fix does not affect anyone else.
With kind regards,
--
Andreas Trottmann
--- End Message ---
--- Begin Message ---
Source: user-mode-linux
Source-Version: 2.6.26-1um-2+16
We believe that the bug you reported is fixed in the latest version of
user-mode-linux, which is due to be installed in the Debian FTP archive:
user-mode-linux_2.6.26-1um-2+16.diff.gz
to pool/main/u/user-mode-linux/user-mode-linux_2.6.26-1um-2+16.diff.gz
user-mode-linux_2.6.26-1um-2+16.dsc
to pool/main/u/user-mode-linux/user-mode-linux_2.6.26-1um-2+16.dsc
user-mode-linux_2.6.26-1um-2+16_i386.deb
to pool/main/u/user-mode-linux/user-mode-linux_2.6.26-1um-2+16_i386.deb
A summary of the changes between this version and the previous one is
attached.
Thank you for reporting the bug, which will now be closed. If you
have further comments please address them to 529312@bugs.debian.org,
and the maintainer will reopen the bug report if appropriate.
Debian distribution maintenance software
pp.
dann frazier <dannf@debian.org> (supplier of updated user-mode-linux package)
(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing ftpmaster@debian.org)
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Format: 1.8
Date: Tue, 09 Jun 2009 16:47:27 -0600
Source: user-mode-linux
Binary: user-mode-linux
Architecture: source i386
Version: 2.6.26-1um-2+16
Distribution: stable
Urgency: high
Maintainer: User Mode Linux Maintainers <pkg-uml-pkgs@lists.alioth.debian.org>
Changed-By: dann frazier <dannf@debian.org>
Description:
user-mode-linux - User-mode Linux (kernel)
Closes: 466491 498271 511334 514268 514627 520551 521420 523359 523360 523364 524199 524300 525060 526836 527101 529312 529864 530884
Changes:
user-mode-linux (2.6.26-1um-2+16) stable; urgency=high
.
* Rebuild against linux-source-2.6.26 (2.6.26-16):
[ maximilian attems ]
* [openvz] 5dcfcf5 NETLINK: disable netns broadcast filtering.
(closes: #520551)
* Fix SQLite performance regression. (closes: #521420)
* [openvz] 0c295ff cfq link cfq_bc_data without bc io sched.
(closes: #523364)
* [openvz] 7e0f90d cfq: revalidate cached async queue.
(closes: #523359)
* [openvz] e4cea21 VE: fix idle time accounting.
* [openvz] 19b8e13 ptrace: ban ptracing of a container init from inside the
container. (closes: #523360)
* [openvz] 5b58141 ubc: uncharging too much for TCPSNDBUF.
* [openvz] 0ff728e ve: show task's vpid and veid even inside a container.
.
[ dann frazier ]
* [s390] Fix __div64_31 for CONFIG_MARCH_G5 (Closes: #511334)
* SUNRPC: Fix a performance regression in the RPC authentication code
(Closes: #524199)
* [x86] fix IBM Summit based systems' phys_cpu_present_map on 32-bit
kernels (closes: #529312)
* Fix soft lockups caused by one md resync blocking on another due
to sharing the same device (closes: #514627)
* [sparc64] Fix crash when reading /proc/iomem w/ heap memory checking
(CVE-2009-1914)
* splice: fix deadlock in ocfs2 (CVE-2009-1961)
* e1000: add missing length check to e1000 receive routine (CVE-2009-1385)
* r8169: fix crash when large packets are received (CVE-2009-1389)
.
[ Martin Michlmayr ]
* cdc-acm: Add quirk for MTK II GPS, such as Qstarz BT-Q1000X (closes:
#525060)
* USB: ftdi_sio: add vendor/product id for the Marvell SheevaPlug.
* [mipsel/r5k-cobalt] Enable SCSI_SYM53C8XX_2 (closes: #526836).
* [mips/r4k-ip22] Enable NET_ISA and various ISA network modules on
the request of Damian Dimmich since they might be useful on the
SGI Indigo2.
.
[ John Wright ]
* [x86] gettimeofday() vDSO: fix segfault when tv == NULL (Closes: #466491)
.
[ Ian Campbell ]
* [x86/xen] Apply missing syscall detection patch to -xen-amd64 image
(Closes: #527101)
* [xen] Add support for CDROM_GET_CAPABILITY to blkfront driver
(Closes: #529864)
.
[ Ben Hutchings ]
* sata_nv: avoid link reset on controllers where it's broken
(Closes: #498271)
* r8169: fix multicast filtering for RTL8101 and RTL8168 (Closes: #514268)
* asus_acpi: don't load asus-acpi if model is not supported
(Closes: #524300)
* iwl4965: avoid sleep in softirq context (Closes: #530884)
Checksums-Sha1:
5a838be0b6d9c80ef0e9cd3166b2dabb5e70122e 1242 user-mode-linux_2.6.26-1um-2+16.dsc
f8ad7fb460b70b1a7c35a6a145d8fe7c923bec59 14608 user-mode-linux_2.6.26-1um-2+16.diff.gz
8c92bd98f4f6dd7f9812316c334171bdf6ab5274 5354222 user-mode-linux_2.6.26-1um-2+16_i386.deb
Checksums-Sha256:
561c9d73d27efcd7086d864e7320386ceb09c14c48247cc8f464ef0b1a964c29 1242 user-mode-linux_2.6.26-1um-2+16.dsc
16deedc5ad410eb23463e8e6048321cd7b4f2700760553a0024ad52e214db981 14608 user-mode-linux_2.6.26-1um-2+16.diff.gz
e43ae359013ef61111792accc0fdb1f3373f6ce084428a432f7d1584312de23a 5354222 user-mode-linux_2.6.26-1um-2+16_i386.deb
Files:
1c330c7b90a949a8717f40b71522015a 1242 misc extra user-mode-linux_2.6.26-1um-2+16.dsc
c8f3820600488268d06b8ee3f68a2282 14608 misc extra user-mode-linux_2.6.26-1um-2+16.diff.gz
3e5b785dd366ad90035954cbab019d5b 5354222 misc extra user-mode-linux_2.6.26-1um-2+16_i386.deb
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
iD8DBQFKLvGshuANDBmkLRkRAlcaAJ0eirEC3KvBAzeIJyCgi2hpr5QwdQCeJ4JS
U/S7e49gS+B3d1gTy0zwndc=
=rxAF
-----END PGP SIGNATURE-----
--- End Message ---