strace trace the signal eventhough -e signal=!<sig-name> has been mentioned.

Srinivasa Ds srinivasa at in.ibm.com
Tue Feb 20 08:23:46 UTC 2007


I downloaded the latest strace-4.5.15.tar.bz2 from strace home page and
compiled it. But I saw one bug in that.

strace was tracing the signal eventhough I mentioned -e signal=\! sig.
This looks like a bug to me,according to man page of strace.Iam using
the latest 2.6.19 kernel.

So, I just wanted to ask,
 1)Is this a bug(as I taught)?
 2) If so, is there any patch to this??

Any comments to this would help me a lot.

===============================================
[root at llm27lp1 strace-4.5.15]# ./strace -o log -e signal=\!SIGALRM  who
root     hvc0         2007-02-19 19:37
srikar   pts/0        2007-02-20 11:52 (srdronam.in.ibm.com)
root     pts/1        2007-02-20 13:37 (k50wks273861wss.in.ibm.com)

[root at llm27lp1 strace-4.5.15]# grep SIGALRM log
rt_sigaction(SIGALRM, {0xff6a430, [], 0}, {SIG_DFL}, 8) = 0
rt_sigaction(SIGALRM, {SIG_DFL}, NULL, 8) = 0
rt_sigaction(SIGALRM, {0xff6a430, [], 0}, {SIG_DFL}, 8) = 0
rt_sigaction(SIGALRM, {SIG_DFL}, NULL, 8) = 0
rt_sigaction(SIGALRM, {0xff6a430, [], 0}, {SIG_DFL}, 8) = 0
rt_sigaction(SIGALRM, {SIG_DFL}, NULL, 8) = 0
rt_sigaction(SIGALRM, {0xff6a430, [], 0}, {SIG_DFL}, 8) = 0
rt_sigaction(SIGALRM, {SIG_DFL}, NULL, 8) = 0
rt_sigaction(SIGALRM, {0xff6a430, [], 0}, {SIG_DFL}, 8) = 0
rt_sigaction(SIGALRM, {SIG_DFL}, NULL, 8) = 0
rt_sigaction(SIGALRM, {0xff6a430, [], 0}, {SIG_DFL}, 8) = 0
rt_sigaction(SIGALRM, {SIG_DFL}, NULL, 8) = 0
rt_sigaction(SIGALRM, {0xff6a430, [], 0}, {SIG_DFL}, 8) = 0
rt_sigaction(SIGALRM, {SIG_DFL}, NULL, 8) = 0
rt_sigaction(SIGALRM, {0xff6a430, [], 0}, {SIG_DFL}, 8) = 0
rt_sigaction(SIGALRM, {SIG_DFL}, NULL, 8) = 0
===================================================================








More information about the Strace-devel mailing list