Re: [RFC 0/8] Allow GFP_NOFS allocation to fail

[Tetsuo Handa]

Tetsuo Handa wrote:
> > Thoughts? Opinions?
> 
> To me, fixing callers (adding __GFP_NORETRY to callers) in a step-by-step
> fashion after adding proactive countermeasure sounds better than changing
> the default behavior (implicitly applying __GFP_NORETRY inside).
> 

Ping?

I showed you at http://marc.info/?l=linux-mm&m=144198479931388 that
changing the default behavior can not terminate the game of Whack-A-Mole.
As long as there are unkillable threads, we can't kill context-sensitive
moles.

I believe that what we need to do now is to add a proactive countermeasure
(e.g. kill more processes) than try to reduce the possibility of hitting
this issue (e.g. allow !__GFP_FS to fail).
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/