Skip to contentSkip to conversation

race in expand_stack in 2.6.7 ?

DISCUSSION Alexander Y. Fomichev 1 participant 1 message Jul 26, 2004

lore
Share & tools
lore
#0 lore raw

G'Day.

Some months ago i've reported about bug which Andrea defined as "race in expand_stack with threads" (http://lkml.org/lkml/2004/4/14/52),
but it was related to -aa branch (More precisely 2.6.5-aa5). After applied patch from Andrea (many tnx), i did not notice any troubles with 2.6.5-aa5 and later with 2.6.7 until today when i've caught very similar bug(?) on the same system. [Dual P4 Xeon 2.4GHz, 2G ECC RAM - (production web-server) Gentoo, kernel 2.6.7]

System behaviour and trace are nearly the same, system alive but i couldn't do ps, pkill etc. Here is a trace:

http://sysadminday.org.ru/expand_stack_race/trace-20040726

config: http://sysadminday.org.ru/expand_stack_race/config-2.6.7

cpuinfo:        http://sysadminday.org.ru/expand_stack_race/cpuinfo
lspci:          http://sysadminday.org.ru/expand_stack_race/lspci 

lspci -vvn http://sysadminday.org.ru/expand_stack_race/lspci-vvn

Signature · 3 lines
Best regards.
        Alexander Y. Fomichev <[email protected]>
        Public PGP key: http://sysadminday.org.ru/gluk.asc