From: Arnd Bergmann <arnd@arndb.de>
To: linuxppc-dev@ozlabs.org
Cc: Roland Dreier <rdreier@cisco.com>,
Joachim Fenkes <fenkes@de.ibm.com>,
LKML <linux-kernel@vger.kernel.org>,
OF-EWG <ewg@lists.openfabrics.org>,
Christoph Raisch <raisch@de.ibm.com>,
Marcus Eder <meder@de.ibm.com>,
OF-General <general@lists.openfabrics.org>,
Stefan Roscher <stefan.roscher@de.ibm.com>
Subject: Re: [PATCH] IB/ehca: Serialize HCA-related hCalls on POWER5
Date: Fri, 7 Dec 2007 10:58:37 +0100 [thread overview]
Message-ID: <200712071058.38416.arnd@arndb.de> (raw)
In-Reply-To: <ada7ijrd6gy.fsf@cisco.com>
On Thursday 06 December 2007, Roland Dreier wrote:
> =A0> Regarding the performance problem, have you checked whether converti=
ng all
> =A0> your spin_lock_irqsave to spin_lock/spin_lock_irq improves your perf=
ormance
> =A0> on the older machines? Maybe it's already fast enough that way.
>=20
> It does seem that the only places that the hcall_lock is taken also
> use msleep, so they must always be in process context. =A0So you can
> safely just use spin_lock(), right?
I think it needs some more inspection. The msleep in there is only called
for hcalls that return H_IS_LONG_BUSY(). In theory, you can call
ehca_plpar_hcall_norets() from inside an interrupt handler if the
hcall in question never returns long busy.
Arnd <><
WARNING: multiple messages have this Message-ID (diff)
From: Arnd Bergmann <arnd@arndb.de>
To: linuxppc-dev@ozlabs.org
Cc: Roland Dreier <rdreier@cisco.com>,
Joachim Fenkes <fenkes@de.ibm.com>,
LKML <linux-kernel@vger.kernel.org>,
OF-EWG <ewg@lists.openfabrics.org>,
Christoph Raisch <raisch@de.ibm.com>,
Marcus Eder <meder@de.ibm.com>,
OF-General <general@lists.openfabrics.org>,
Stefan Roscher <stefan.roscher@de.ibm.com>
Subject: Re: [PATCH] IB/ehca: Serialize HCA-related hCalls on POWER5
Date: Fri, 7 Dec 2007 10:58:37 +0100 [thread overview]
Message-ID: <200712071058.38416.arnd@arndb.de> (raw)
In-Reply-To: <ada7ijrd6gy.fsf@cisco.com>
On Thursday 06 December 2007, Roland Dreier wrote:
> > 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?
I think it needs some more inspection. The msleep in there is only called
for hcalls that return H_IS_LONG_BUSY(). In theory, you can call
ehca_plpar_hcall_norets() from inside an interrupt handler if the
hcall in question never returns long busy.
Arnd <><
next prev parent reply other threads:[~2007-12-07 9:58 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-12-06 15:07 [PATCH] IB/ehca: Serialize HCA-related hCalls on POWER5 Joachim Fenkes
2007-12-06 15:07 ` Joachim Fenkes
2007-12-06 15:48 ` Arnd Bergmann
2007-12-06 15:48 ` Arnd Bergmann
2007-12-06 18:27 ` Roland Dreier
2007-12-06 18:27 ` Roland Dreier
2007-12-07 9:58 ` Arnd Bergmann [this message]
2007-12-07 9:58 ` Arnd Bergmann
2007-12-09 23:22 ` Roland Dreier
2007-12-09 23:22 ` Roland Dreier
2007-12-10 17:41 ` Joachim Fenkes
2007-12-10 17:41 ` Joachim Fenkes
2007-12-10 21:47 ` Roland Dreier
2007-12-10 21:47 ` Roland Dreier
2007-12-11 8:38 ` Joachim Fenkes
2007-12-11 8:38 ` Joachim Fenkes
2007-12-12 12:14 ` [ewg] " Or Gerlitz
2007-12-12 12:14 ` Or Gerlitz
2007-12-12 16:02 ` Christoph Raisch
2007-12-12 16:02 ` Christoph Raisch
2007-12-12 19:09 ` Roland Dreier
2007-12-12 19:09 ` Roland Dreier
2007-12-13 8:30 ` Or Gerlitz
2007-12-13 8:30 ` Or Gerlitz
2007-12-13 19:22 ` [ofa-general] " Caitlin Bestler
2007-12-13 19:22 ` Caitlin Bestler
2007-12-13 20:59 ` Joachim Fenkes
2007-12-13 20:59 ` Joachim Fenkes
2007-12-13 21:08 ` Caitlin Bestler
2007-12-13 21:08 ` Caitlin Bestler
2007-12-13 21:35 ` Joachim Fenkes
2007-12-13 21:35 ` Joachim Fenkes
2007-12-13 21:48 ` [ofa-general] Re: [ewg] Re: [PATCH] IB/ehca: SerializeHCA-related " Sean Hefty
2007-12-13 21:48 ` Sean Hefty
2007-12-07 16:25 ` [PATCH] IB/ehca: Serialize HCA-related " Joachim Fenkes
2007-12-07 16:25 ` Joachim Fenkes
[not found] <OF85E31FAA.DADA6039-ONC12573AA.005439C8-C12573AA.005A132E@LocalDomain>
2007-12-10 17:59 ` Joachim Fenkes
2007-12-10 17:59 ` Joachim Fenkes
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=200712071058.38416.arnd@arndb.de \
--to=arnd@arndb.de \
--cc=ewg@lists.openfabrics.org \
--cc=fenkes@de.ibm.com \
--cc=general@lists.openfabrics.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linuxppc-dev@ozlabs.org \
--cc=meder@de.ibm.com \
--cc=raisch@de.ibm.com \
--cc=rdreier@cisco.com \
--cc=stefan.roscher@de.ibm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.