Re: [PATCH] IB/ehca: Serialize HCA-related hCalls on POWER5

From: Roland Dreier <rdreier@cisco.com>
To: Arnd Bergmann <arnd@arndb.de>
Cc: linuxppc-dev@ozlabs.org, Joachim Fenkes <fenkes@de.ibm.com>,
	LKML <linux-kernel@vger.kernel.org>,
	"OF-General" <general@lists.openfabrics.org>,
	Roland Dreier <rolandd@cisco.com>,
	"OF-EWG" <ewg@lists.openfabrics.org>,
	Stefan Roscher <stefan.roscher@de.ibm.com>,
	Christoph Raisch <raisch@de.ibm.com>,
	Marcus Eder <meder@de.ibm.com>
Subject: Re: [PATCH] IB/ehca: Serialize HCA-related hCalls on POWER5
Date: Thu, 06 Dec 2007 10:27:09 -0800	[thread overview]
Message-ID: <ada7ijrd6gy.fsf@cisco.com> (raw)
In-Reply-To: <200712061648.24806.arnd@arndb.de> (Arnd Bergmann's message of "Thu, 6 Dec 2007 16:48:23 +0100")

 > > +               ehca_lock_hcalls = !(cur_cpu_spec->cpu_user_features
 > > +                                    & PPC_FEATURE_ARCH_2_05);

 > We already talked about this yesterday, but I still feel that checking the
 > instruction set of the CPU should not be used to determine whether a
 > specific device driver implementation is used int hypervisor.

I had the same reaction... is testing cpu_user_features really the
best way to detect this issue?

I'll hold off applying this for a few days so you guys can decide the
best thing to do.  We'll definitely get some fix into 2.6.24 but we
have time to make a good decision.

 > Regarding the performance problem, have you checked whether converting all
 > your spin_lock_irqsave to spin_lock/spin_lock_irq improves your performance
 > on the older machines? Maybe it's already fast enough that way.

It does seem that the only places that the hcall_lock is taken also
use msleep, so they must always be in process context.  So you can
safely just use spin_lock(), right?

 - R.