Subject: Re: [ofa-general] mpi failures on large ia64/ofed/IB clusters

From: Roland Dreier <rdreier@...> · Date: Friday, October 5, 2007 - 6:51 pm

> I don't really see anything racy in the FW command stuff, but it's
 > possible that there's something like an mmiowb() missing somewhere (I
 > have a hard time spotting that type of race for some reason).

Another possibility (independent of the hack I suggested before) would
be to add an mmiowb() before the mutex_unlock() in mthca_cmd_post().

I actually have a good feeling about this theory....

 - R.
_______________________________________________
general mailing list
general@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general

To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general


linux-kernel:
Andrew Morton	2.6.23-rc3-mm1
Tarkan Erimer	Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3
Yinghai Lu	Re: [PATCH RFC] x86: check for and defend against BIOS memory corruption
Frederik Deweerdt	[-mm patch] remove tcp header from tcp_v4_check (take #2)
openbsd-misc:

linux-netdev:
David Miller	[GIT]: Networking
Jarek Poplawski	Re: [PATCH] pkt_sched: Destroy gen estimators under rtnl_lock().
Gerrit Renker	[PATCH 27/37] dccp: Integration of dynamic feature activation - part 2 (server side)
Herbert Xu	Re: [PATCH 2/3][NET_BATCH] net core use batching
git: