[SRM] kernel updates for oldstable
As previously noted, a 2.6 kernel security update is pending for
sarge that changes the ABI. This update has now been NEW-processed,
and a DSA should be released soon. I've spoken with Frans, and the
plan is to do a spin of d-i ASAP as oldstable installs are
currently rather broken. This respin would also include this ABI
Since we're spinning d-i anyway, I thought I'd see if we could get
some other non-security changes in there anyway. We've had some stuff
queued for way too long (since before sarge released even). I
re-examined each of the issues yesterday, and ended up with the
following changes, most of which are things Horms had queued for a
stable update (though I did drop 3 changes that I felt were of less
than important severity), and a few that we've identified since.
I have started builds across all of the architectures - I think the
final one will complete tomorrow morning. If there are no objections
to any of the proposed changes (or requests for an extended review
cycle), I'll plan to upload tomorrow evening after some testing.
kernel-source-2.6.8 (2.6.8-17) oldstable; urgency=high
[ Simon Horman ]
* drivers-net-via-rhine-wol-oops.dpatch (removed):
This patch breaks the via-rhine driver and 2.6.8 and is
completely bogus for this version of the kernel
Allow Leadtek WinFast VC100 XP cards to work.
Fix possible false assertion failure in log_do_checkpoint(). We might fail
to detect that we actually made a progress when cleaning up the checkpoint
lists if we don't retry after writing something to disk.
Stop try_to_unmap_cluster() passing out-of-bounds pte to pte_unmap()
Fix deadlock with ip_queue and tcp local input path.
Make sure gcc doesn't reorder memory accesses in strncmp and friends on
Make sure we call acpi_register_gsi() even for default PCI interrupt
assignment. That's the part that keeps track of the ELCR register, and we
want to make sure that the PCI interrupts are properly marked level/low.
[ dann frazier ]
* Merge in applicable fixes from 184.108.40.206
[ Simon Horman ]
Add SATAII TX2 and TX2/TX4 support to sata promise driver
Module per-cpu alignment cannot always be met
fix gl_skb/skb type error in genelink driver in usbnet
Backported From 220.127.116.11
Make sure BLK_DEV_IDEDMA_PCI is defined for pmac ide driver builds
ext3 file systems mounted over nfs may lookup .. in dx directories
causing an oops.
Use SA_SHIRQ in sparc specific code.
forcedeth: Initialize link settings in every nv_open()
Fix up more strange byte writes to the PCI_ROM_ADDRESS config word
yenta oops fix
Fix fs/exec.c:788 (de_thread()) BUG_ON
fix IPv6 per-socket multicast filtering in exact-match case
ipvs: ip_vs_ftp breaks connections using persistence
ieee1394/sbp2: fixes for hot-unplug and module unloading
[SPARC64]: Fix userland FPU state corruption.
[ dann frazier ]
* drivers-block-raw-ioctl2.dpatch, drivers-block-ioctl-enotty.dpatch:
Fix a bug in the block layer that causes a bootloader installation
error under certain conditions - breaks installation on cciss devices.
* Fix data corruption with dm-crypt over RAID5 (closes: #336153)
* Fix VLAN support for 3c59x/90x series hardware (closes: #349774)
* Fix erroneous calculation of 'len' parameter to NLMSG_PUT resulting in
bogus 'error during NLMSG_PUT' messages (closes: #372621)
* hp-diva-rmp3.dpatch, hp-diva-hurricane.dpatch:
Add PCI IDs for newer Diva console ports
-- dann frazier <email@example.com> Sat, 26 May 2007 01:56:24 -0600