> Thanks for bringing this to my attention. A while back I had different > management of forked tasks and merged it with PF_NONSLEEP. Since then I've > changed the management of NONSLEEP tasks and didn't realise it had adversely > affected the accounting of forking tasks. This patch should rectify it. [snip] At least here this patch solves the "previously working" (i.e. starving) testcase, i.e. now all eatm-processes continue to work. Best, Michael -- Vote against SPAM - see http://www.politik-digital.de/spam/ Michael Gerdau email: [email protected] GPG-keys available on request or at public keyserver
Attachment:
pgp6BBohEuqfp.pgp
Description: PGP signature
- References:
- Re: [patch][rfc] quell interactive feeding frenzy
- From: Al Boldi <[email protected]>
- Re: [patch][rfc] quell interactive feeding frenzy
- From: Con Kolivas <[email protected]>
- Re: [patch][rfc] quell interactive feeding frenzy
- Prev by Date: [PATCH] modpost: relax driver data name
- Next by Date: LTP test failures on current kernels (latest -git and -mm)
- Previous by thread: Re: [patch][rfc] quell interactive feeding frenzy
- Next by thread: Re: [patch][rfc] quell interactive feeding frenzy
- Index(es):
 |