login
Search
Willy Tarreau, the new maintainer of the Linux 2.4 kernel [story], released the 2.4.33.1 kernel. This is a security fix release for the 2.4.33 kernel. This is the first time a 2.4-series kernel is being released with 4 version fields. The fourth version field, introduced in the 2.6 kernel series, has now also been implemented for the 2.4 kernel.
About this release, Willy says
"As there were a few security fixes pending and 2.4.34-pre1 has not received enough validation, I've released 2.4.33.1 with the most important fixes. All those fixes are already in 2.4.34-pre1.
"Particularly important ones are :
- CVE-2006-1528 : local DoS via direct I/O from the sg driver to mmapped I/O space fix from Dann Frazier
- CVE-2006-4093 : possible local DoS on some PPC970.
fix from Olof Johansson"
List: linux-kernel
Subject: Linux 2.4.33.1
From: Willy Tarreau [email blocked]
Date: 2006-08-19 14:13:55
Hi !
As there were a few security fixes pending and 2.4.34-pre1 has not
received enough validation, I've released 2.4.33.1 with the most
important fixes. All those fixes are already in 2.4.34-pre1.
Particularly important ones are :
- CVE-2006-1528 : local DoS via direct I/O from the sg driver to mmapped I/O space
fix from Dann Frazier
- CVE-2006-4093 : possible local DoS on some PPC970.
fix from Olof Johansson
Hotfix patches for older versions should follow within a short time.
Regards,
Willy
Summary of changes from v2.4.33 to v2.4.33.1
============================================
dann frazier:
drivers/scsi/sg.c : fix CVE-2006-1528
Jeff Layton:
2.4 NFS client - update d_cache when server reports ENOENT on an NFS remove
Willy Tarreau:
[BLKMTD] : missing offset sometimes causes panics
[PKTGEN] : fix an oops when used with bonding driver (Tien ChenLi)
export memchr() which is used by smbfs and lp driver.
powerpc: Clear HID0 attention enable on PPC970 at boot time
Change VERSION to 2.4.33.1
2.4.x.y
As a proud user of the 2.4 kernels (IMHO, 2.6 is really a development kernel...), I was wondering if this new 2.4.x.y scheme is going to become the norm for 2.4, or is this just as aberration (i.e., 1 time security fix) ?
Trust me, these will not be t
Trust me, these will not be the last security fixes for 2.4.
New maintainer, new naming scheme ;)
I think when the new versionning scheme was discussed (1 year ago ?), the plan was to use it for both 2.6 and 2.4.
Maybe Marcelo did not want to change his habits, but I'm pretty sure that's what was decided by then... though I'm too lazy to search the archives. :)