Messages in this thread | | | Date | Tue, 22 Jul 2008 14:47:56 +0200 | From | "Michael Kerrisk" <> | Subject | Re: [PATCH -mm] coredump: format_corename: don't append .%pid if multi-threaded |
| |
On Tue, Jul 22, 2008 at 2:43 PM, Oleg Nesterov <oleg@tv-sign.ru> wrote: > On 07/22, Michael Kerrisk wrote: >> >> On Tue, Jul 22, 2008 at 2:18 PM, Oleg Nesterov <oleg@tv-sign.ru> wrote: >> > If the coredumping is multi-threaded, format_corename() appends .%pid >> > to the corename. This was needed before the proper multi-thread core >> > dump support, now all the threads in the mm go into a single unified >> > core file. >> > >> > Remove this special case, it is not even documented and we have "%p" >> > and core_uses_pid. >> >> Hi Oleg, >> >> I have not thought about this at any length, but one question that >> jumps to mind: could this feature still be useful for LinuxThreads, >> where each thread does indeed have a separate PID? > > As far as I know, LinuxThreads use CLONE_VM, right?
Yes.
> The coredump will create the single core file for all processes because > they have the same ->mm, the "threads" won't dump all over each other.
Yes, looks like you are right. I had this vague idea that there were circumstances where a dump of a LinuxThreads m-t process could produce multipl core files, distinguished by the .PID, but I think I must have misremembered.
> And, just in case, this patch doesn't make any difference if core_uses_pid > is set or pid_in_pattern is true. > > That said, this is the user-visible change...
True. Not sure how important that is in this case though. What is the reason for making this change (other than tidiness)?
-- Michael Kerrisk Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/ man-pages online: http://www.kernel.org/doc/man-pages/online_pages.html Found a bug? http://www.kernel.org/doc/man-pages/reporting_bugs.html
| |