Soumya Koduri [skod...@redhat.com] wrote:
> Hi Kaleb/Malahal,
>
> Request you to merge below FSAL_GLUSTER patches into V2.2-stable branch -
>
> 366f71c - FSAL_GLUSTER: Fixed an issue with dereferencing a NULL ponter
I just looked at the patch from your other mail. I have few questions on
this patch.
1. I am not sure why you declared retval as int *. Does the up_thread
exists with an "int *"? Most threads just exit with an integer, so
declaring just "int retval" or a later casting would have been
suffice.
2. Technically, passed in OUT argument is updated by pthead_join() only
on success, so it should be valid only if pthead_join() returned
success (usually this should NOT fail though).
All in all, the code should be something like this:
int retval;
err = pthread_join(up_thread, (void**)&retval);
if (err)
LogCrit(XX, "pthread_join faild: %d", err);
else
LogDebug(XX, "upthread exited with: %d", retval);
Checking for '*retval' is only useful if your up_thread exited with
something like pthread_exit(&some_var), but I don't think it does that.
Regards, Malahal.
PS: I am just looking at the manpage, no direct experince! Did you ever
see the system printing a sane "Up_thread join returned value %d"
message?
------------------------------------------------------------------------------
_______________________________________________
Nfs-ganesha-devel mailing list
Nfs-ganesha-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel
