From: Boqun Feng <boqun.feng@gmail.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: panxinhui <xinhui@linux.vnet.ibm.com>,
Pan Xinhui <xinhui.pan@linux.vnet.ibm.com>,
linux-kernel@vger.kernel.org, mingo@redhat.com,
dave@stgolabs.net, will.deacon@arm.com, Waiman.Long@hpe.com,
benh@kernel.crashing.org
Subject: Re: [PATCH] locking/osq: Drop the overload of osq lock
Date: Mon, 27 Jun 2016 18:31:25 +0800 [thread overview]
Message-ID: <20160627103125.GF6512@insomnia> (raw)
In-Reply-To: <20160627080959.GU30154@twins.programming.kicks-ass.net>
[-- Attachment #1: Type: text/plain, Size: 2549 bytes --]
On Mon, Jun 27, 2016 at 10:09:59AM +0200, Peter Zijlstra wrote:
[snip]
>
> No, this is entirely insane, also broken.
>
> No vectors, no actual function calls, nothing like that. You want the
> below to completely compile away and generate the exact 100% same code
> it does today.
>
Point taken.
As Xinhui also posted something similar, which worked better on not
effecting the generated code if not enabled. I think I'd better to drop
this workload to him ;-)
> > +++ b/kernel/locking/osq_lock.c
> > @@ -1,6 +1,7 @@
> > #include <linux/percpu.h>
> > #include <linux/sched.h>
> > #include <linux/osq_lock.h>
> > +#include <linux/vcpu_preempt.h>
> >
> > /*
> > * An MCS like lock especially tailored for optimistic spinning for sleeping
> > @@ -87,6 +88,8 @@ bool osq_lock(struct optimistic_spin_queue *lock)
> > struct optimistic_spin_node *prev, *next;
> > int curr = encode_cpu(smp_processor_id());
> > int old;
> > + int loops;
> > + long vpc;
> >
> > node->locked = 0;
> > node->next = NULL;
> > @@ -106,6 +109,9 @@ bool osq_lock(struct optimistic_spin_queue *lock)
> > node->prev = prev;
> > WRITE_ONCE(prev->next, node);
> >
> > + old = old - 1;
>
> That's just nasty, and could result in an unconditional decrement being
> issues, even though its never used.
>
Right, better to calculate this decrement at the argument of
vcpu_is_preempted() callsite, and define the primitive of host code as
#define vcpu_is_preempted(cpu) false
, which could probably be optimized out.
Regards,
Boqun
> > + vpc = vcpu_preempt_count();
> > +
> > /*
> > * Normally @prev is untouchable after the above store; because at that
> > * moment unlock can proceed and wipe the node element from stack.
> > @@ -118,8 +124,14 @@ bool osq_lock(struct optimistic_spin_queue *lock)
> > while (!READ_ONCE(node->locked)) {
> > /*
> > * If we need to reschedule bail... so we can block.
> > + * An over-committed guest with more vCPUs than pCPUs
> > + * might fall in this loop and cause a huge overload.
> > + * This is because vCPU A(prev) hold the osq lock and yield out,
> > + * vCPU B(node) wait ->locked to be set, IOW, wait till
> > + * vCPU A run and unlock the osq lock.
> > + * NOTE that vCPU A and vCPU B might run on same physical cpu.
> > */
> > - if (need_resched())
> > + if (need_resched() || vcpu_is_preempted(old) || vcpu_has_preempted(vpc))
> > goto unqueue;
> >
> > cpu_relax_lowlatency();
> >
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 473 bytes --]
prev parent reply other threads:[~2016-06-27 10:27 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-25 17:42 [PATCH] locking/osq: Drop the overload of osq lock Pan Xinhui
2016-06-25 14:24 ` Peter Zijlstra
2016-06-25 15:21 ` Boqun Feng
2016-06-25 16:09 ` Peter Zijlstra
2016-06-25 16:13 ` Peter Zijlstra
2016-06-25 17:27 ` panxinhui
2016-06-25 19:12 ` Peter Zijlstra
2016-06-26 4:59 ` panxinhui
2016-06-27 7:55 ` Peter Zijlstra
2016-06-27 10:19 ` xinhui
2016-06-25 16:28 ` Boqun Feng
2016-06-25 18:28 ` Peter Zijlstra
2016-06-25 16:15 ` Peter Zijlstra
2016-06-25 16:45 ` Boqun Feng
2016-06-25 17:27 ` panxinhui
2016-06-25 19:20 ` Peter Zijlstra
2016-06-25 19:29 ` Peter Zijlstra
2016-06-26 2:26 ` Boqun Feng
2016-06-26 5:21 ` panxinhui
2016-06-26 6:10 ` Boqun Feng
2016-06-26 6:58 ` panxinhui
2016-06-26 14:11 ` Boqun Feng
2016-06-26 15:54 ` panxinhui
2016-06-26 6:59 ` Boqun Feng
2016-06-26 7:08 ` panxinhui
2016-06-26 14:29 ` Boqun Feng
2016-06-26 15:11 ` panxinhui
2016-06-27 6:45 ` Boqun Feng
2016-06-27 7:36 ` xinhui
2016-06-27 8:09 ` Peter Zijlstra
2016-06-27 10:31 ` Boqun Feng [this message]
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=20160627103125.GF6512@insomnia \
--to=boqun.feng@gmail.com \
--cc=Waiman.Long@hpe.com \
--cc=benh@kernel.crashing.org \
--cc=dave@stgolabs.net \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=will.deacon@arm.com \
--cc=xinhui.pan@linux.vnet.ibm.com \
--cc=xinhui@linux.vnet.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).