6,534
edits
Changes
created
== Changes ==
* Bug fixes
=== Configs ===
Same as {{kernel link|2.6.8|022stab050.1}} plus:<br/>
Removed:
* -<code>CONFIG_SCSI_DPT_I2O=y</code>
Added:
* +<code>CONFIG_I2O_SCSI=m</code>
* +<code>CONFIG_LBD=y</code>
<includeonly>[[{{PAGENAME}}/changes#Patches|{{Long changelog message}}]]</includeonly><noinclude>
=== Patches ===
==== diff-ve-fsareacheck-20051129 ====
<div class="change">Patch from Pavel:
Due to numerous complains temporarily turn off area access checking.
Problems which caused such a decision are:
* use of simfs for trusted templates and executing binaries from it without VPS starting
: Bug #55131.
* non-virtualized pipe-fs and sock-fs which brokes stat on some /proc/pid/fd entries. SuSe templates don't work due to this.
: Bug #55178.
</div>
==== diff-ve-sysctl-sem-20051129 ====
<div class="change">Patch from Dmitry:<br/>
Fixed inability to change ve0 sem parameters.
proc_dointvec_minmax() can't handle vectors.
</div>
==== diff-ve-ip-conntrack-debug-20051129 ====
<div class="change">Patch from Alexander:<br/>
just a misprints when NETFILTER_DEBUG is on.
</div>
==== diff-ve-fsareacheck-msg-20051130 ====
<div class="change">Patch from Pavel:<br/>
Missed newline in fs-area-check spoils ouptut.
</div>
==== diff-ve-sock-warn-20051130 ====
<div class="change">Patch from Pavel:<br/>
User-triggerable printk()'s in get/setsockopt with SO_BSDCOMPAT
</div>
==== diff-vzdq-debug2-20051202 ====
<div class="change">Patch from Dmitry:<br/>
Apply 3 seconds timeout for "detached inode not in creation"
messages too.
Bug #49285.
</div>
==== diff-fairsched-loadbal-20051201 ====
<div class="change">Patch from Kirill:<br/>
This patch removes warning on load balance.
Balancing is racy, and it is not fatal anyhow if
busiest rq is the current one. printk() under
rq->lock can be the much more fatal thing here...
Bug #55290.
</div>
==== diff-ve-netlink-20051210 ====
<div class="change">Patch from Dmitry:
this patch adds owner_env to primary key for selecting
from nl_table in order to avoid collisions due to virt pids.
Bug #55602.
</div>
==== diff-venet-inetrelease-20051205 ====
<div class="change">Patch from Alexey:<br/>
[PATCH] switch exec_env to ve while inet_release
fput() on a file owned by VPS can be called in context of VPS0. It is critical
for TCP to call socket operations in context of correct exec_env.
Bug #55393.
</div>
==== diff-venet-srccheck-fix-20051208 ====
<div class="change">Patch from Denis and Pavel:
When ip_rt_src_check is off we still can try
to route multicast/broadcast packets.
Bug #55476.
</div>
==== diff-ve-vpid-netlink-20051210 ====
<div class="change">Patch from Denis:
This patch fixes netlink_autobind to make `ip rule show`
works inside VE as it checks the destination PID
on the message against value obtained via getpid().
Bug #55602.
</div>
==== diff-mainstream-dcache-race4 ====
<div class="change">Patch from Dmitry:
fixed race between dput_recursive and generic_shutdown_super
on root dentry.
Bug #55453.
</div>
==== diff-ve-procstat-20051214 ====
<div class="change">Patch from Vasiliy:<br/>
Fix small typo in /proc/stat
</div>
==== diff-ms-mm-pageprivate-20051213 ====
<div class="change">Patch from Dmitry:<br/>
fixed race in pagevec_strip
Bug #56040.<br/>
Bug #55235.
</div>
==== diff-ms-jbd-checkpoint-20050602 ====
<div class="change">Patch from mainstream:<br/>
[PATCH] ext3: fix log_do_checkpoint() assertion failure
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. The patch was
confirmed to fix observed assertion failures for several users.
When we flushed some buffers we need to retry scanning the list.
Otherwise we can fail to detect our progress.
Signed-off-by: Jan Kara <jack@suse.cz><br/>
Signed-off-by: Andrew Morton <akpm@osdl.org><br/>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
GIT: 00ea81459c279f14a7b344320a71c94f60f88929<br/>
http://linux.bkbits.net:8080/linux-2.6/cset@1.3104.1.1
Bug #56042.
</div>
==== diff-ve-procvmstat-20051215 ====
<div class="change">Patch from Pavel:
Show /proc/vmstat entry inside vps, but keep it empty.
required for latest vmstat from RHEL4 :(
Bug #56186.
</div>
</noinclude>
* Bug fixes
=== Configs ===
Same as {{kernel link|2.6.8|022stab050.1}} plus:<br/>
Removed:
* -<code>CONFIG_SCSI_DPT_I2O=y</code>
Added:
* +<code>CONFIG_I2O_SCSI=m</code>
* +<code>CONFIG_LBD=y</code>
<includeonly>[[{{PAGENAME}}/changes#Patches|{{Long changelog message}}]]</includeonly><noinclude>
=== Patches ===
==== diff-ve-fsareacheck-20051129 ====
<div class="change">Patch from Pavel:
Due to numerous complains temporarily turn off area access checking.
Problems which caused such a decision are:
* use of simfs for trusted templates and executing binaries from it without VPS starting
: Bug #55131.
* non-virtualized pipe-fs and sock-fs which brokes stat on some /proc/pid/fd entries. SuSe templates don't work due to this.
: Bug #55178.
</div>
==== diff-ve-sysctl-sem-20051129 ====
<div class="change">Patch from Dmitry:<br/>
Fixed inability to change ve0 sem parameters.
proc_dointvec_minmax() can't handle vectors.
</div>
==== diff-ve-ip-conntrack-debug-20051129 ====
<div class="change">Patch from Alexander:<br/>
just a misprints when NETFILTER_DEBUG is on.
</div>
==== diff-ve-fsareacheck-msg-20051130 ====
<div class="change">Patch from Pavel:<br/>
Missed newline in fs-area-check spoils ouptut.
</div>
==== diff-ve-sock-warn-20051130 ====
<div class="change">Patch from Pavel:<br/>
User-triggerable printk()'s in get/setsockopt with SO_BSDCOMPAT
</div>
==== diff-vzdq-debug2-20051202 ====
<div class="change">Patch from Dmitry:<br/>
Apply 3 seconds timeout for "detached inode not in creation"
messages too.
Bug #49285.
</div>
==== diff-fairsched-loadbal-20051201 ====
<div class="change">Patch from Kirill:<br/>
This patch removes warning on load balance.
Balancing is racy, and it is not fatal anyhow if
busiest rq is the current one. printk() under
rq->lock can be the much more fatal thing here...
Bug #55290.
</div>
==== diff-ve-netlink-20051210 ====
<div class="change">Patch from Dmitry:
this patch adds owner_env to primary key for selecting
from nl_table in order to avoid collisions due to virt pids.
Bug #55602.
</div>
==== diff-venet-inetrelease-20051205 ====
<div class="change">Patch from Alexey:<br/>
[PATCH] switch exec_env to ve while inet_release
fput() on a file owned by VPS can be called in context of VPS0. It is critical
for TCP to call socket operations in context of correct exec_env.
Bug #55393.
</div>
==== diff-venet-srccheck-fix-20051208 ====
<div class="change">Patch from Denis and Pavel:
When ip_rt_src_check is off we still can try
to route multicast/broadcast packets.
Bug #55476.
</div>
==== diff-ve-vpid-netlink-20051210 ====
<div class="change">Patch from Denis:
This patch fixes netlink_autobind to make `ip rule show`
works inside VE as it checks the destination PID
on the message against value obtained via getpid().
Bug #55602.
</div>
==== diff-mainstream-dcache-race4 ====
<div class="change">Patch from Dmitry:
fixed race between dput_recursive and generic_shutdown_super
on root dentry.
Bug #55453.
</div>
==== diff-ve-procstat-20051214 ====
<div class="change">Patch from Vasiliy:<br/>
Fix small typo in /proc/stat
</div>
==== diff-ms-mm-pageprivate-20051213 ====
<div class="change">Patch from Dmitry:<br/>
fixed race in pagevec_strip
Bug #56040.<br/>
Bug #55235.
</div>
==== diff-ms-jbd-checkpoint-20050602 ====
<div class="change">Patch from mainstream:<br/>
[PATCH] ext3: fix log_do_checkpoint() assertion failure
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. The patch was
confirmed to fix observed assertion failures for several users.
When we flushed some buffers we need to retry scanning the list.
Otherwise we can fail to detect our progress.
Signed-off-by: Jan Kara <jack@suse.cz><br/>
Signed-off-by: Andrew Morton <akpm@osdl.org><br/>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
GIT: 00ea81459c279f14a7b344320a71c94f60f88929<br/>
http://linux.bkbits.net:8080/linux-2.6/cset@1.3104.1.1
Bug #56042.
</div>
==== diff-ve-procvmstat-20051215 ====
<div class="change">Patch from Pavel:
Show /proc/vmstat entry inside vps, but keep it empty.
required for latest vmstat from RHEL4 :(
Bug #56186.
</div>
</noinclude>