From: Guo Ren <ren_guo@c-sky.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: linux-arch@vger.kernel.org, linux-kernel@vger.kernel.org,
tglx@linutronix.de, daniel.lezcano@linaro.org,
jason@lakedaemon.net, arnd@arndb.de, devicetree@vger.kernel.org,
andrea.parri@amarulasolutions.com, c-sky_gcc_upstream@c-sky.com,
gnu-csky@mentor.com, thomas.petazzoni@bootlin.com,
wbx@uclibc-ng.org, green.hu@gmail.com
Subject: Re: [PATCH V3 11/27] csky: Atomic operations
Date: Mon, 17 Sep 2018 23:05:32 +0800 [thread overview]
Message-ID: <20180917150532.GC2612@guoren-Inspiron-7460> (raw)
In-Reply-To: <20180917081755.GO24124@hirez.programming.kicks-ass.net>
On Mon, Sep 17, 2018 at 10:17:55AM +0200, Peter Zijlstra wrote:
> On Sat, Sep 15, 2018 at 10:55:13PM +0800, Guo Ren wrote:
> > > > +#define ATOMIC_OP_RETURN(op, c_op) \
>
> > > > +#define ATOMIC_FETCH_OP(op, c_op) \
>
> > > For these you could generate _relaxed variants and not provide smp_mb()
> > > inside them.
> > Ok, but I'll modify it in next commit.
>
> That's fine. Just wanted to let you know about _relaxed() since it will
> benefit your platform.
Thank you.
> > > > +#define ATOMIC_OP(op, c_op) \
> > > > +static inline void atomic_##op(int i, atomic_t *v) \
> > > > +{ \
> > > > + unsigned long tmp, flags; \
> > > > + \
> > > > + raw_local_irq_save(flags); \
> > > > + \
> > > > + asm volatile ( \
> > > > + " ldw %0, (%2) \n" \
> > > > + " " #op " %0, %1 \n" \
> > > > + " stw %0, (%2) \n" \
> > > > + : "=&r" (tmp) \
> > > > + : "r" (i), "r"(&v->counter) \
> > > > + : "memory"); \
> > > > + \
> > > > + raw_local_irq_restore(flags); \
> > > > +}
> > >
> > > Is this really 'better' than the generic UP fallback implementation?
> > There is a lock irq instruction "idly4" with out irq_save. eg:
> > asm volatile ( \
> > " idly4 \n" \
> > " ldw %0, (%2) \n" \
> > " " #op " %0, %1 \n" \
> > " stw %0, (%2) \n" \
> > I'll change to that after full tested.
>
> That is pretty nifty, could you explain (or reference me to a arch doc
> that does) the exact semantics of that "idly4" instruction?
The idly4 allows the 4 instructions behind it to not respond to interrupts.
When ldw got exception, it will cause the carry to be 1. So I need
prepare the assemble like this:
1: cmpne r0, r0
idly4
ldw %0, (%2)
bt 1b
" #op " ...
stw ...
I need more stress test on it and then I'll change to it.
> > > > +static inline void arch_spin_lock(arch_spinlock_t *lock)
> > > > +{
> > > > + arch_spinlock_t lockval;
> > > > + u32 ticket_next = 1 << TICKET_NEXT;
> > > > + u32 *p = &lock->lock;
> > > > + u32 tmp;
> > > > +
> > > > + smp_mb();
> > >
> > > spin_lock() doesn't need smp_mb() before.
> > read_lock and write_lock also needn't smp_mb() before, isn't it?
>
> Correct. The various *_lock() functions only need imply an ACQUIRE
> barrier, such that the critical section happens after the lock is taken.
>
> > > > +
> > > > +static inline void arch_spin_unlock(arch_spinlock_t *lock)
> > > > +{
> > > > + smp_mb();
> > > > + lock->tickets.owner++;
> > > > + smp_mb();
> > >
> > > spin_unlock() doesn't need smp_mb() after.
> > read_unlock and write_unlock also needn't smp_mb() after, isn't it?
>
> Indeed so, the various *_unlock() functions only need imply a RELEASE
> barrier, such that the critical section happend before the lock is
> released.
>
> In both cases (lock and unlock) there is a great amount of subtle
> details, but most of that is irrelevant if all you have is smp_mb().
Got it, Thx for the explanation.
>
>
> > > > +/*
> > > > + * Test-and-set spin-locking.
> > > > + */
> > >
> > > Why retain that?
> > >
> > > same comments; it has far too many smp_mb()s in.
> > I'm not sure about queued_rwlocks and just for 2-cores-smp test-and-set is
> > faster and simpler, isn't it?
>
> Even on 2 cores I think you can create starvation cases with
> test-and-set spinlocks. And the maintenace overhead of carrying two lock
> implementations is non trivial.
>
> As to performance; I cannot say, but the ticket lock isn't very
> expensive, you could benchmark of course.
Ticket lock is good.
But How about queued_rwlocks v.s my_test_set_rwlock?
I'm not sure about the queued_rwlocks. I just implement the ticket-spinlock.
Best Regards
Guo Ren
prev parent reply other threads:[~2018-09-17 15:06 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-09-12 13:24 [PATCH V4 00/27] C-SKY(csky) Linux Kernel Port Guo Ren
2018-09-12 13:24 ` [PATCH V3 01/27] csky: Build infrastructure Guo Ren
2018-09-12 13:24 ` [PATCH V3 02/27] csky: defconfig Guo Ren
2018-09-12 13:24 ` [PATCH V3 03/27] csky: Kernel booting Guo Ren
2018-09-12 13:24 ` [PATCH V3 04/27] csky: Exception handling Guo Ren
2018-09-12 13:24 ` [PATCH V3 05/27] csky: System Call Guo Ren
2018-09-12 13:24 ` [PATCH V3 06/27] csky: Cache and TLB routines Guo Ren
2018-09-12 13:24 ` [PATCH V3 07/27] csky: MMU and page table management Guo Ren
2018-09-12 13:24 ` [PATCH V3 08/27] csky: Process management and Signal Guo Ren
2018-09-12 13:24 ` [PATCH V3 09/27] csky: VDSO and rt_sigreturn Guo Ren
2018-09-12 13:24 ` [PATCH V3 10/27] csky: IRQ handling Guo Ren
2018-09-12 13:24 ` [PATCH V3 12/27] csky: ELF and module probe Guo Ren
2018-09-12 13:24 ` [PATCH V3 13/27] csky: Library functions Guo Ren
2018-09-12 13:24 ` [PATCH V3 14/27] csky: User access Guo Ren
2018-09-12 13:24 ` [PATCH V3 15/27] csky: Debug and Ptrace GDB Guo Ren
2018-09-12 13:24 ` [PATCH V3 16/27] csky: SMP support Guo Ren
2018-09-12 13:24 ` [PATCH V3 18/27] dt-bindings: csky CPU Bindings Guo Ren
2018-09-12 13:24 ` [PATCH V3 19/27] dt-bindings: timer: gx6605s SOC timer Guo Ren
2018-09-12 13:24 ` [PATCH V3 20/27] dt-bindings: timer: C-SKY Multi-processor timer Guo Ren
2018-09-12 13:24 ` [PATCH V3 22/27] dt-bindings: interrupt-controller: C-SKY SMP intc Guo Ren
2018-09-12 13:24 ` [PATCH V3 23/27] clocksource: add gx6605s SOC system timer Guo Ren
[not found] ` <abb46d366b513b814f7af234d560306a818b7324.1536757532.git.ren_guo@c-sky.com>
2018-09-12 14:22 ` [PATCH V3 17/27] csky: Misc headers Arnd Bergmann
2018-09-12 14:30 ` [PATCH V4 00/27] C-SKY(csky) Linux Kernel Port Arnd Bergmann
2018-09-14 14:37 ` Guo Ren
2018-09-14 14:46 ` Arnd Bergmann
2018-09-14 16:02 ` Guo Ren
2018-09-14 16:09 ` Arnd Bergmann
2018-09-14 23:28 ` Guo Ren
2018-09-20 17:52 ` Palmer Dabbelt
2018-09-21 5:18 ` Arnd Bergmann
2018-09-21 23:48 ` Guo Ren
2018-09-24 7:21 ` Geert Uytterhoeven
2018-09-24 8:47 ` Arnd Bergmann
2018-09-16 1:07 ` Guo Ren
2018-09-16 4:53 ` Guo Ren
2018-09-17 11:54 ` Stephen Rothwell
2018-09-17 12:03 ` Stephen Rothwell
2018-09-17 14:50 ` Guo Ren
2018-09-17 14:37 ` Guo Ren
[not found] ` <93e8b592e429c156ad4d4ca5d85ef48fd0ab8b70.1536757532.git.ren_guo@c-sky.com>
2018-09-12 15:55 ` [PATCH V3 11/27] csky: Atomic operations Peter Zijlstra
2018-09-15 14:55 ` Guo Ren
2018-09-17 8:17 ` Peter Zijlstra
2018-09-17 15:05 ` Guo Ren [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=20180917150532.GC2612@guoren-Inspiron-7460 \
--to=ren_guo@c-sky.com \
--cc=andrea.parri@amarulasolutions.com \
--cc=arnd@arndb.de \
--cc=c-sky_gcc_upstream@c-sky.com \
--cc=daniel.lezcano@linaro.org \
--cc=devicetree@vger.kernel.org \
--cc=gnu-csky@mentor.com \
--cc=green.hu@gmail.com \
--cc=jason@lakedaemon.net \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--cc=thomas.petazzoni@bootlin.com \
--cc=wbx@uclibc-ng.org \
/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).