6,534
edits
Changes
m
<br/>[http://www.securityfocus.com/archive/1/427622/100/0/threaded http://www.securityfocus.com/archive/1/427622/100/0/threaded]
<br/>Signed-off-by: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru><br/>
</div>
<ul>
<li>* removed pb_add_list_ref - now pb_add_ref acts similar in both cases;</li><li>* changed logic on copying pages - if a page isn't accounted in pbc for parent it will not be in child. pb_dup_ref was introduced for it.</li></uldiv>
</div>
<br/>[http://forum.openvz.org/index.php?t=msg&th=465&start=0 http://forum.openvz.org/index.php?t=msg&th=465&start=0]
Protected "Download/kernel/2.6.16/026test009.1/changes": Robot: Protecting a list of files. [edit=autoconfirmed:move=autoconfirmed]
{{Kernel git log|2.6.16|026test009.1}}
<includeonly>[[{{PAGENAME}}/changes#Patches|{{Long changelog message}}]]</includeonly><noinclude>
=== Patches ===
<div class="change">
Patch from Alexey Kuznetsov <alexey@openvz.org>:<br/>
[TCP]: Do not use inet->id of global tcp_socket when sending RST.<br/> The problem is in ip_push_pending_frames(), which uses:<br/>
The problem is in ip_push_pending_frames(), which uses:<presource lang="c"> if (!df) { __ip_select_ident(iph, &rt->u.dst, 0); } else { iph->id = htons(inet->id++); }</presource>instead of ip_select_ident().<br/>
Right now I think the code is a nonsense. Most likely, I copied it from
old ip_build_xmit(), where it was really special, we had to decide
whether to generate unique ID when generating the first (well, the last)
fragment.<br/>
In ip_push_pending_frames() it does not make sense, it should use plain
ip_select_ident() instead.<br/>
Signed-off-by: David S. Miller <davem@davemloft.net>
</div>
==== diff-smp-nmi-show-regs ====
<div class="change">
Patch from Pavel Emelianov <xemul@openvz.org>:<br/>
Some fixes of page beancounters accounting.
* moved pb_alloc_all closer to pb_alloc and co;
==== diff-ubc-mlock-deadlock-20060412 ====
<div class="change">
Patch from Alexey Kuznetsov <alexey@openvz.org>:<br/>
Mlock was ported incorrectly<br/>Two bugs actually:<br/>1. In 2.6 mlock_fixup _MUST_ return correct prev VMA, otherwise mlockall deadlocks.<br/>2. Inaccurate port: in error path memory is not uncharged.
Two bugs actually:
# In 2.6 mlock_fixup _MUST_ return correct prev VMA, otherwise mlockall deadlocks.<br/>
# Inaccurate port: in error path memory is not uncharged.
</div>
==== diff-ubc-page-uncharge-race-20060414 ====
<div class="change">
Patch from Pavel Emelianov <xemul@openvz.org>:<br/>
Fix of race between ub_page_charge/_uncharge.<br/>
Pages were first put into global list of free pages and ther - uncharged.
During this gap page_alloc could charge not uncharged yet page thus
causing a BUG().<br/>
Moved ub_page_uncharge above freeing and out of local_irq_save to reduce
time spent with irqs off.
</div>
==== diff-ubc-pbfree-oops-20060417 ====
<div class="change">
