Messages in this thread |  | | Date | Mon, 25 Jan 1999 13:15:27 -0500 (EST) | | From | David Mansfield <> | | Subject | Re: serious MM problem with pre8,9 - vmstat stuck in D state |
| |
> Hi, > > On Sat, 23 Jan 1999 18:33:05 -0500 (EST), David Mansfield > <david@cobite.com> said: > > > Additionally, ps and killall also hang in D state reading from proc. > > This is a side-effect of extra locking introduced into procfs to > safeguard certain operations against memory corruption. > > --Stephen >
I'm glad it's safe now but not being able to run 'ps' when heavy swapping is occuring, and not being able to run 'vmstat' to see how things are going is pretty terrible, IMHO... Those are the times where getting process information is most important.
The problem is centered around the call to read() from /proc/NNN/stat. I'll look at the changes and see if I can understand what you are referring to, but I'm really not in a position to provide any help here except as 'user' feedback :-( The VM's up to pre7 had seriously impressed me, and now even sending SIGTERM (by PID, not even using 'ps first) to a process seem orders of magnitude slower.
Thanks for the explanation, David
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
|  |