From: Gleb Natapov <gleb@redhat.com> To: Peter Zijlstra <peterz@infradead.org> Cc: kvm@vger.kernel.org, linux-mm@kvack.org, linux-kernel@vger.kernel.org, avi@redhat.com, mingo@elte.hu, tglx@linutronix.de, hpa@zytor.com, riel@redhat.com Subject: Re: [PATCH v2 10/12] Maintain preemptability count even for !CONFIG_PREEMPT kernels Date: Mon, 23 Nov 2009 17:58:51 +0200 [thread overview] Message-ID: <20091123155851.GU2999@redhat.com> (raw) In-Reply-To: <1258990455.4531.594.camel@laptop> On Mon, Nov 23, 2009 at 04:34:15PM +0100, Peter Zijlstra wrote: > On Mon, 2009-11-23 at 16:06 +0200, Gleb Natapov wrote: > > Do not preempt kernel. Just maintain counter to know if task can be rescheduled. > > Asynchronous page fault may be delivered while spinlock is held or current > > process can't be preempted for other reasons. KVM uses preempt_count() to check if preemptions is allowed and schedule other process if possible. This works > > with preemptable kernels since they maintain accurate information about > > preemptability in preempt_count. This patch make non-preemptable kernel > > maintain accurate information in preempt_count too. > > I'm thinking you're going to have to convince some people this won't > slow them down for no good. > I saw old discussions about this in mailing list archives. Usually someone wanted to use in_atomic() in driver code and this, of course, caused the resistant. In this case, I think, the use is legitimate. > Personally I always have PREEMPT=y, but other people seem to feel > strongly about not doing so. > It is possible to add one more config option to enable reliable preempt_count() without enabling preemption or make async pf be dependable on PREEMPT=y. Don't like both of this options especially first one. There are more then enough options already. -- Gleb.
WARNING: multiple messages have this Message-ID (diff)
From: Gleb Natapov <gleb@redhat.com> To: Peter Zijlstra <peterz@infradead.org> Cc: kvm@vger.kernel.org, linux-mm@kvack.org, linux-kernel@vger.kernel.org, avi@redhat.com, mingo@elte.hu, tglx@linutronix.de, hpa@zytor.com, riel@redhat.com Subject: Re: [PATCH v2 10/12] Maintain preemptability count even for !CONFIG_PREEMPT kernels Date: Mon, 23 Nov 2009 17:58:51 +0200 [thread overview] Message-ID: <20091123155851.GU2999@redhat.com> (raw) In-Reply-To: <1258990455.4531.594.camel@laptop> On Mon, Nov 23, 2009 at 04:34:15PM +0100, Peter Zijlstra wrote: > On Mon, 2009-11-23 at 16:06 +0200, Gleb Natapov wrote: > > Do not preempt kernel. Just maintain counter to know if task can be rescheduled. > > Asynchronous page fault may be delivered while spinlock is held or current > > process can't be preempted for other reasons. KVM uses preempt_count() to check if preemptions is allowed and schedule other process if possible. This works > > with preemptable kernels since they maintain accurate information about > > preemptability in preempt_count. This patch make non-preemptable kernel > > maintain accurate information in preempt_count too. > > I'm thinking you're going to have to convince some people this won't > slow them down for no good. > I saw old discussions about this in mailing list archives. Usually someone wanted to use in_atomic() in driver code and this, of course, caused the resistant. In this case, I think, the use is legitimate. > Personally I always have PREEMPT=y, but other people seem to feel > strongly about not doing so. > It is possible to add one more config option to enable reliable preempt_count() without enabling preemption or make async pf be dependable on PREEMPT=y. Don't like both of this options especially first one. There are more then enough options already. -- Gleb. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2009-11-23 15:59 UTC|newest] Thread overview: 64+ messages / expand[flat|nested] mbox.gz Atom feed top 2009-11-23 14:05 [PATCH v2 00/12] KVM: Add asynchronous page fault for PV guest Gleb Natapov 2009-11-23 14:05 ` Gleb Natapov 2009-11-23 14:05 ` [PATCH v2 01/12] Move kvm_smp_prepare_boot_cpu() from kvmclock.c to kvm.c Gleb Natapov 2009-11-23 14:05 ` Gleb Natapov 2009-11-23 14:05 ` [PATCH v2 02/12] Add PV MSR to enable asynchronous page faults delivery Gleb Natapov 2009-11-23 14:05 ` Gleb Natapov 2009-11-25 12:32 ` Avi Kivity 2009-11-25 12:32 ` Avi Kivity 2009-11-23 14:05 ` [PATCH v2 03/12] Add async PF initialization to PV guest Gleb Natapov 2009-11-23 14:05 ` Gleb Natapov 2009-11-23 14:05 ` [PATCH v2 04/12] Add "handle page fault" PV helper Gleb Natapov 2009-11-23 14:05 ` Gleb Natapov 2009-11-23 15:32 ` Peter Zijlstra 2009-11-23 15:32 ` Peter Zijlstra 2009-11-23 15:32 ` Peter Zijlstra 2009-11-23 14:06 ` [PATCH v2 05/12] Handle asynchronous page fault in a PV guest Gleb Natapov 2009-11-23 14:06 ` Gleb Natapov 2009-11-25 12:45 ` Avi Kivity 2009-11-25 12:45 ` Avi Kivity 2009-11-23 14:06 ` [PATCH v2 06/12] Export __get_user_pages_fast Gleb Natapov 2009-11-23 14:06 ` Gleb Natapov 2009-11-23 14:06 ` [PATCH v2 07/12] Add get_user_pages() variant that fails if major fault is required Gleb Natapov 2009-11-23 14:06 ` Gleb Natapov 2009-11-23 14:06 ` [PATCH v2 08/12] Inject asynchronous page fault into a guest if page is swapped out Gleb Natapov 2009-11-23 14:06 ` Gleb Natapov 2009-11-25 13:03 ` Avi Kivity 2009-11-25 13:03 ` Avi Kivity 2009-11-23 14:06 ` [PATCH v2 09/12] Retry fault before vmentry Gleb Natapov 2009-11-23 14:06 ` Gleb Natapov 2009-11-25 13:09 ` Avi Kivity 2009-11-25 13:09 ` Avi Kivity 2009-11-25 13:20 ` Gleb Natapov 2009-11-25 13:20 ` Gleb Natapov 2009-11-23 14:06 ` [PATCH v2 10/12] Maintain preemptability count even for !CONFIG_PREEMPT kernels Gleb Natapov 2009-11-23 14:06 ` Gleb Natapov 2009-11-23 15:34 ` Peter Zijlstra 2009-11-23 15:34 ` Peter Zijlstra 2009-11-23 15:34 ` Peter Zijlstra 2009-11-23 15:58 ` Gleb Natapov [this message] 2009-11-23 15:58 ` Gleb Natapov 2009-11-23 17:30 ` Christoph Lameter 2009-11-23 17:30 ` Christoph Lameter 2009-11-24 7:12 ` Gleb Natapov 2009-11-24 7:12 ` Gleb Natapov 2009-11-24 15:14 ` Christoph Lameter 2009-11-24 15:14 ` Christoph Lameter 2009-11-30 10:56 ` Gleb Natapov 2009-11-30 10:56 ` Gleb Natapov 2009-11-30 10:58 ` Gleb Natapov 2009-11-30 10:58 ` Gleb Natapov 2009-11-30 10:59 ` Peter Zijlstra 2009-11-30 10:59 ` Peter Zijlstra 2009-11-30 10:59 ` Peter Zijlstra 2009-11-30 11:01 ` Avi Kivity 2009-11-30 11:01 ` Avi Kivity 2009-11-30 11:05 ` Peter Zijlstra 2009-11-30 11:05 ` Peter Zijlstra 2009-11-30 11:05 ` Peter Zijlstra 2009-11-30 16:23 ` Christoph Lameter 2009-11-30 16:23 ` Christoph Lameter 2009-11-23 14:06 ` [PATCH v2 11/12] Handle async PF in non preemptable context Gleb Natapov 2009-11-23 14:06 ` Gleb Natapov 2009-11-23 14:06 ` [PATCH v2 12/12] Send async PF when guest is not in userspace too Gleb Natapov 2009-11-23 14:06 ` Gleb Natapov
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=20091123155851.GU2999@redhat.com \ --to=gleb@redhat.com \ --cc=avi@redhat.com \ --cc=hpa@zytor.com \ --cc=kvm@vger.kernel.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ --cc=mingo@elte.hu \ --cc=peterz@infradead.org \ --cc=riel@redhat.com \ --cc=tglx@linutronix.de \ /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: linkBe 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.