Contents
Changes
- Fixed hwclock hang on Dell hardware
- Security fixes
Configs
- CONFIG_LOG_BUF_SHIFT=17 (was 14)
- CONFIG_HPET_EMULATE_RTC=y (due to broken Dell hardware)
Patches
diff-ms-cpu-cachesize-1-20060109
Patch from mainstream, prepared by Kostja:
adds missing CPU cache size descriptions.
http://linux.bkbits.net:8080/linux-2.6/cset@1.1803.139.2
http://linux.bkbits.net:8080/linux-2.6/gnupatch@41290a27p2CnUBatiGkLq28NnGQ5lA
Bug 57479.
diff-ms-cpu-cachesize-2-20060109
Patch from mainstream, prepared by Kostja:
adds another Intel cache descriptor entry.
http://linux.bkbits.net:8080/linux-2.6/cset@1.1803.1.57
http://linux.bkbits.net:8080/linux-2.6/gnupatch@41291989HzqiFm_8Q_tN3u9TLJJICQ
Bug 57479.
diff-ms-hpetrtc-20050709
Patch from mainstream:
Some systems like:
- Dell Poweredge 1800/1850/SC420
- Dell Precision 370/670
- Dell Optiplex
- Dell Dimension 4700
- Dell GX280
are known to hang in rtc_read() due to broken RTC, when HPET and RTC are used together.
So, have to configure HPET_EMULATE_RTC to have RTC working.
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=149211
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=277298
http://bugzilla.kernel.org/show_bug.cgi?id=4904
http://bugzilla.kernel.org/attachment.cgi?id=5416&action=view
http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=c91096d85c95c6b7fe8d7065e2aa6825e0bdaca9
Bug 56720.
diff-psmouse-init-20060119
Patch from mainstream, backported by Vasily:
fixes oops in psmouse_interrupt.
Oops was faced by Rainer Duffner <rduffner@ip-tech.ch>
http://linux.bkbits.net:8080/linux-2.6/gnupatch@40fc93e600HS6BUpb8j23j70JM7_jQ
diff-ms-shrink-dcache-latency2
Patch from Pavel and Denis:
reduce scheduling latency in shrink_dcache_sb()
Bug 58107.
diff-ms-shrink-dcache-latency
[PATCH] sched: vfs: fix scheduling latencies in
prune_dcache() and select_parent() faced by Vasily
Bug 58107.
diff-venet-ipfrag-20060118
Patch from Denis:
This patch allows fragmented TCP packets passing through venet device.
diff-security-iptaddcnt-20051017
Patch from Kirill, issue found by Solar Designer:
This patch fixes buffer size check in do_add_counters().
Minor issue allowing to read some random kernel vmalloc memory.
diff-security-iptoverfl-20051014
Patch from Pavel:
This patch fixes security issues found by Solar Designer:
The bug I've triggered in practice is the first instance of ipt_table_info_alloc() and its corresponding copy_from_user(). To trigger the overflow on addition, tmp.size may be set to 0xffffffff such that SMP_ALIGN(tmp.size) would be 0.