From: Michael Ellerman <mpe@ellerman.id.au>
To: Nicholas Piggin <npiggin@gmail.com>,
linux-arch@vger.kernel.org, Will Deacon <will.deacon@arm.com>
Cc: Andrea Parri <andrea.parri@amarulasolutions.com>,
Arnd Bergmann <arnd@arndb.de>,
Benjamin Herrenschmidt <benh@kernel.crashing.org>,
Rich Felker <dalias@libc.org>,
David Howells <dhowells@redhat.com>,
Daniel Lustig <dlustig@nvidia.com>,
linux-kernel@vger.kernel.org,
"Maciej W. Rozycki" <macro@linux-mips.org>,
Ingo Molnar <mingo@kernel.org>,
Palmer Dabbelt <palmer@sifive.com>,
Paul Burton <paul.burton@mips.com>,
"Paul E. McKenney" <paulmck@linux.ibm.com>,
Peter Zijlstra <peterz@infradead.org>,
Alan Stern <stern@rowland.harvard.edu>,
Tony Luck <tony.luck@intel.com>,
Linus Torvalds <torvalds@linux-foundation.org>,
Yoshinori Sato <ysato@users.sourceforge.jp>
Subject: Re: [PATCH 01/20] asm-generic/mmiowb: Add generic implementation of mmiowb() tracking
Date: Sun, 03 Mar 2019 20:26:35 +1100 [thread overview]
Message-ID: <87tvgkia0k.fsf@concordia.ellerman.id.au> (raw)
In-Reply-To: <1551575210.6lwpiqtg5k.astroid@bobo.none>
Nicholas Piggin <npiggin@gmail.com> writes:
> Will Deacon's on March 2, 2019 12:03 am:
>> In preparation for removing all explicit mmiowb() calls from driver
>> code, implement a tracking system in asm-generic based loosely on the
>> PowerPC implementation. This allows architectures with a non-empty
>> mmiowb() definition to have the barrier automatically inserted in
>> spin_unlock() following a critical section containing an I/O write.
>
> Is there a reason to call this "mmiowb"? We already have wmb that
> orders cacheable stores vs mmio stores don't we?
>
> Yes ia64 "sn2" is broken in that case, but that can be fixed (if
> anyone really cares about the platform any more). Maybe that's
> orthogonal to what you're doing here, I just don't like seeing
> "mmiowb" spread.
>
> This series works for spin locks, but you would want a driver to
> be able to use wmb() to order locks vs mmio when using a bit lock
> or a mutex or whatever else. Calling your wmb-if-io-is-pending
> version io_mb_before_unlock() would kind of match with existing
> patterns.
>
>> +static inline void mmiowb_set_pending(void)
>> +{
>> + struct mmiowb_state *ms = __mmiowb_state();
>> + ms->mmiowb_pending = ms->nesting_count;
>> +}
>> +
>> +static inline void mmiowb_spin_lock(void)
>> +{
>> + struct mmiowb_state *ms = __mmiowb_state();
>> + ms->nesting_count++;
>> +}
>> +
>> +static inline void mmiowb_spin_unlock(void)
>> +{
>> + struct mmiowb_state *ms = __mmiowb_state();
>> +
>> + if (unlikely(ms->mmiowb_pending)) {
>> + ms->mmiowb_pending = 0;
>> + mmiowb();
>> + }
>> +
>> + ms->nesting_count--;
>> +}
>
> Humour me for a minute and tell me what this algorithm is doing, or
> what was broken about the powerpc one, which is basically:
>
> static inline void mmiowb_set_pending(void)
> {
> struct mmiowb_state *ms = __mmiowb_state();
> ms->mmiowb_pending = 1;
> }
>
> static inline void mmiowb_spin_lock(void)
> {
> }
The current powerpc code clears io_sync in spin_lock().
ie, it would be equivalent to:
static inline void mmiowb_spin_lock(void)
{
ms->mmiowb_pending = 0;
}
Which means that:
spin_lock(a);
writel(x, y);
spin_lock(b);
...
spin_unlock(b);
spin_unlock(a);
Does no barrier.
cheers
next prev parent reply other threads:[~2019-03-03 9:26 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-03-01 14:03 [PATCH 00/20] Remove Mysterious Macro Intended to Obscure Weird Behaviours (mmiowb()) Will Deacon
2019-03-01 14:03 ` [PATCH 01/20] asm-generic/mmiowb: Add generic implementation of mmiowb() tracking Will Deacon
2019-03-03 1:43 ` Nicholas Piggin
2019-03-03 2:18 ` Linus Torvalds
2019-03-03 3:34 ` Nicholas Piggin
[not found] ` <CAHk-=whVN58nWh29jvXx+X-Yx9dCC6BeAZOtKak+d01y_UVg=A@mail.gmail.com>
2019-03-03 10:05 ` Nicholas Piggin
2019-03-03 18:48 ` Linus Torvalds
2019-03-05 0:21 ` Nicholas Piggin
2019-03-05 0:33 ` Linus Torvalds
2019-03-03 9:26 ` Michael Ellerman [this message]
2019-03-03 10:07 ` Nicholas Piggin
2019-03-04 1:01 ` Michael Ellerman
2019-03-05 0:21 ` Nicholas Piggin
2019-03-04 10:24 ` Michael Ellerman
2019-03-05 0:19 ` Linus Torvalds
2019-03-07 0:47 ` Michael Ellerman
2019-03-07 1:13 ` Linus Torvalds
2019-03-07 9:13 ` Peter Zijlstra
2019-03-01 14:03 ` [PATCH 02/20] arch: Use asm-generic header for asm/mmiowb.h Will Deacon
2019-03-01 14:03 ` [PATCH 03/20] mmiowb: Hook up mmiowb helpers to spinlocks and generic I/O accessors Will Deacon
2019-03-03 1:47 ` Nicholas Piggin
2019-03-01 14:03 ` [PATCH 04/20] ARM/io: Remove useless definition of mmiowb() Will Deacon
2019-03-01 14:03 ` [PATCH 05/20] arm64/io: " Will Deacon
2019-03-01 14:03 ` [PATCH 06/20] x86/io: " Will Deacon
2019-03-01 14:03 ` [PATCH 07/20] nds32/io: " Will Deacon
2019-03-01 14:03 ` [PATCH 08/20] m68k/io: " Will Deacon
2019-03-01 14:03 ` [PATCH 09/20] sh/mmiowb: Add unconditional mmiowb() to arch_spin_unlock() Will Deacon
2019-03-01 14:03 ` [PATCH 10/20] mips/mmiowb: " Will Deacon
2019-03-01 22:16 ` Paul Burton
2019-03-01 14:03 ` [PATCH 11/20] ia64/mmiowb: " Will Deacon
2019-03-01 14:03 ` [PATCH 12/20] powerpc/mmiowb: Hook up mmwiob() implementation to asm-generic code Will Deacon
2019-03-02 12:46 ` Michael Ellerman
2019-03-01 14:03 ` [PATCH 13/20] riscv/mmiowb: " Will Deacon
2019-03-01 21:13 ` Palmer Dabbelt
2019-03-01 14:03 ` [PATCH 14/20] Documentation: Kill all references to mmiowb() Will Deacon
2019-03-01 14:03 ` [PATCH 15/20] drivers: Remove useless trailing comments from mmiowb() invocations Will Deacon
2019-03-01 14:03 ` [PATCH 16/20] drivers: Remove explicit invocations of mmiowb() Will Deacon
2019-03-01 14:03 ` [PATCH 17/20] scsi/qla1280: Remove stale comment about mmiowb() Will Deacon
2019-03-01 14:03 ` [PATCH 18/20] i40iw: Redefine i40iw_mmiowb() to do nothing Will Deacon
2019-03-01 14:03 ` [PATCH 19/20] net/ethernet/silan/sc92031: Remove stale comment about mmiowb() Will Deacon
2019-03-01 14:03 ` [PATCH 20/20] arch: Remove dummy mmiowb() definitions from arch code Will Deacon
2019-03-01 16:41 ` [PATCH 00/20] Remove Mysterious Macro Intended to Obscure Weird Behaviours (mmiowb()) Linus Torvalds
2019-03-02 12:56 ` Michael Ellerman
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=87tvgkia0k.fsf@concordia.ellerman.id.au \
--to=mpe@ellerman.id.au \
--cc=andrea.parri@amarulasolutions.com \
--cc=arnd@arndb.de \
--cc=benh@kernel.crashing.org \
--cc=dalias@libc.org \
--cc=dhowells@redhat.com \
--cc=dlustig@nvidia.com \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=macro@linux-mips.org \
--cc=mingo@kernel.org \
--cc=npiggin@gmail.com \
--cc=palmer@sifive.com \
--cc=paul.burton@mips.com \
--cc=paulmck@linux.ibm.com \
--cc=peterz@infradead.org \
--cc=stern@rowland.harvard.edu \
--cc=tony.luck@intel.com \
--cc=torvalds@linux-foundation.org \
--cc=will.deacon@arm.com \
--cc=ysato@users.sourceforge.jp \
/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).