The patch titled
core_pattern: fix up a few miscellaneous bugs
has been added to the -mm tree. Its filename is
core_pattern-fix-up-a-few-miscellaneous-bugs.patch
*** Remember to use Documentation/SubmitChecklist when testing your code ***
See http://www.zip.com.au/~akpm/linux/patches/stuff/added-to-mm.txt to find
out what to do about this
------------------------------------------------------
Subject: core_pattern: fix up a few miscellaneous bugs
From: Neil Horman <nhorman@xxxxxxxxxxxxx>
- Fixes call_usermodehelper_pipe to not wait for program exit on the user
mode helper process. Since this context fill the pipe. The user mode
helper, if it reads stdin, will block waiting for the pipe to have data on
it, and we will wait forever for the usermodehelper to exit. Changed this
to wait only for a successful exec.
- Fixes do_coredump to detect a crash in the user mode helper process and
abort the attempt to recursively dump core to another copy of the helper
process, potentially ad-infinitum.
Signed-off-by: Neil Horman <nhorman@xxxxxxxxxxxxx>
Cc: <martin.pitt@xxxxxxxxxx>
Cc: <wwoods@xxxxxxxxxx>
Cc: Jeremy Fitzhardinge <jeremy@xxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
fs/exec.c | 14 +++++++++++++-
kernel/kmod.c | 2 +-
2 files changed, 14 insertions(+), 2 deletions(-)
diff -puN fs/exec.c~core_pattern-fix-up-a-few-miscellaneous-bugs fs/exec.c
--- a/fs/exec.c~core_pattern-fix-up-a-few-miscellaneous-bugs
+++ a/fs/exec.c
@@ -1768,12 +1768,24 @@ int do_coredump(long signr, int exit_cod
goto fail_unlock;
if (ispipe) {
- core_limit = RLIM_INFINITY;
helper_argv = argv_split(GFP_KERNEL, corename+1, &helper_argc);
/* Terminate the string before the first option */
delimit = strchr(corename, ' ');
if (delimit)
*delimit = '\0';
+ delimit = strrchr(helper_argv[0], '/');
+ if (delimit)
+ delimit++;
+ else
+ delimit = helper_argv[0];
+ if (!strcmp(delimit, current->comm))
+ {
+ printk(KERN_NOTICE "Recursive core dump detected, Aborting\n");
+ goto fail_unlock;
+ }
+
+ core_limit = RLIM_INFINITY;
+
/* SIGPIPE can happen, but it's just never processed */
if(call_usermodehelper_pipe(corename+1, helper_argv, NULL, &file)) {
printk(KERN_INFO "Core dump to %s pipe failed\n",
diff -puN kernel/kmod.c~core_pattern-fix-up-a-few-miscellaneous-bugs kernel/kmod.c
--- a/kernel/kmod.c~core_pattern-fix-up-a-few-miscellaneous-bugs
+++ a/kernel/kmod.c
@@ -505,7 +505,7 @@ int call_usermodehelper_pipe(char *path,
if (ret < 0)
goto out;
- return call_usermodehelper_exec(sub_info, 1);
+ return call_usermodehelper_exec(sub_info, UMH_WAIT_EXEC);
out:
call_usermodehelper_freeinfo(sub_info);
_
Patches currently in -mm which might be from nhorman@xxxxxxxxxxxxx are
core_pattern-ignore-rlimit_core-if-core_pattern-is-a-pipe.patch
core_pattern-ignore-rlimit_core-if-core_pattern-is-a-pipe-fix.patch
core_pattern-allow-passing-of-arguments-to-user-mode-helper-when-core_pattern-is-a-pipe.patch
core_pattern-fix-up-a-few-miscellaneous-bugs.patch
core_pattern-fix-up-a-few-miscellaneous-bugs-fix.patch
-
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
