All of lore.kernel.org
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: Catalin Marinas <catalin.marinas@arm.com>
Cc: Russell King <rmk@arm.linux.org.uk>,
	Paul Mundt <lethal@linux-sh.org>,
	Andrea Arcangeli <aarcange@redhat.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Rik van Riel <riel@redhat.com>, Ingo Molnar <mingo@elte.hu>,
	"akpm@linux-foundation.org" <akpm@linux-foundation.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-arch@vger.kernel.org" <linux-arch@vger.kernel.org>,
	"linux-mm@kvack.org" <linux-mm@kvack.org>,
	Benjamin Herrenschmidt <benh@kernel.crashing.org>,
	David Miller <davem@davemloft.net>,
	Hugh Dickins <hugh.dickins@tiscali.co.uk>,
	Mel Gorman <mel@csn.ul.ie>, Nick Piggin <npiggin@kernel.dk>,
	Chris Metcalf <cmetcalf@tilera.com>,
	Martin Schwidefsky <schwidefsky@de.ibm.com>
Subject: Re: [RFC][PATCH 4/6] arm, mm: Convert arm to generic tlb
Date: Thu, 17 May 2012 18:59:19 +0200	[thread overview]
Message-ID: <1337273959.4281.62.camel@twins> (raw)
In-Reply-To: <1337273053.4281.50.camel@twins>

On Thu, 2012-05-17 at 18:44 +0200, Peter Zijlstra wrote:
> 
> So the RCU code can from ppc in commit
> 267239116987d64850ad2037d8e0f3071dc3b5ce, which has similar behaviour.
> Also I suspect the mm_users < 2 test will be incorrect for ARM since
> even the one user can be concurrent with your speculation engine.
> 
> 
Right, last mail, I promise, I've confused myself enough already! :-)

OK, so ppc/sparc are special (forgot all about s390) I think by the time
they are done with unmap_page_range() their hardware hash-tables are
empty and nobody but software page-table walkers will still access the
linux page tables.

So when we do free_pgtables() to clean up the actual page-tables.
Power/Sparc need to RCU free this to allow concurrent software
page-table walkers like gup_fast.

Thus I don't think they need to tlb flush again because their hardware
doesn't actually walk the link page-tables, it walks hash-tables, which
by this time are empty.

Now if x86/Xen were to use this, it would indeed also need to TLB flush
when freeing the page-tables, since its hardware walkers do indeed
traverse these pages and we need to sync against them.

So my first patch in the tlb-unify tree is actually buggy.

Humm,. what to do adding a tlb flush in there might slow down ppc/sparc
unnecessarily.. dave/ben? I guess we need more knobs :-(


Now its quite possible I've utterly confused myself and everybody
reading, apologies for that, I shall rest and purge all from memory and
start over before commenting more.. 

WARNING: multiple messages have this Message-ID (diff)
From: Peter Zijlstra <peterz@infradead.org>
To: Catalin Marinas <catalin.marinas@arm.com>
Cc: Russell King <rmk@arm.linux.org.uk>,
	Paul Mundt <lethal@linux-sh.org>,
	Andrea Arcangeli <aarcange@redhat.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Rik van Riel <riel@redhat.com>, Ingo Molnar <mingo@elte.hu>,
	"akpm@linux-foundation.org" <akpm@linux-foundation.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-arch@vger.kernel.org" <linux-arch@vger.kernel.org>,
	"linux-mm@kvack.org" <linux-mm@kvack.org>,
	Benjamin Herrenschmidt <benh@kernel.crashing.org>,
	David Miller <davem@davemloft.net>,
	Hugh Dickins <hugh.dickins@tiscali.co.uk>,
	Mel Gorman <mel@csn.ul.ie>, Nick Piggin <npiggin@kernel.dk>,
	Chris Metcalf <cmetcalf@tilera.com>,
	Martin Schwidefsky <schwidefsky@de.ibm.com>
Subject: Re: [RFC][PATCH 4/6] arm, mm: Convert arm to generic tlb
Date: Thu, 17 May 2012 18:59:19 +0200	[thread overview]
Message-ID: <1337273959.4281.62.camel@twins> (raw)
In-Reply-To: <1337273053.4281.50.camel@twins>

On Thu, 2012-05-17 at 18:44 +0200, Peter Zijlstra wrote:
> 
> So the RCU code can from ppc in commit
> 267239116987d64850ad2037d8e0f3071dc3b5ce, which has similar behaviour.
> Also I suspect the mm_users < 2 test will be incorrect for ARM since
> even the one user can be concurrent with your speculation engine.
> 
> 
Right, last mail, I promise, I've confused myself enough already! :-)

OK, so ppc/sparc are special (forgot all about s390) I think by the time
they are done with unmap_page_range() their hardware hash-tables are
empty and nobody but software page-table walkers will still access the
linux page tables.

So when we do free_pgtables() to clean up the actual page-tables.
Power/Sparc need to RCU free this to allow concurrent software
page-table walkers like gup_fast.

Thus I don't think they need to tlb flush again because their hardware
doesn't actually walk the link page-tables, it walks hash-tables, which
by this time are empty.

Now if x86/Xen were to use this, it would indeed also need to TLB flush
when freeing the page-tables, since its hardware walkers do indeed
traverse these pages and we need to sync against them.

So my first patch in the tlb-unify tree is actually buggy.

Humm,. what to do adding a tlb flush in there might slow down ppc/sparc
unnecessarily.. dave/ben? I guess we need more knobs :-(


Now its quite possible I've utterly confused myself and everybody
reading, apologies for that, I shall rest and purge all from memory and
start over before commenting more.. 

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  reply	other threads:[~2012-05-17 16:59 UTC|newest]

Thread overview: 86+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-03-02 17:59 [RFC][PATCH 0/6] mm: Unify TLB gather implementations Peter Zijlstra
2011-03-02 17:59 ` Peter Zijlstra
2011-03-02 17:59 ` Peter Zijlstra
2011-03-02 17:59 ` [RFC][PATCH 1/6] mm: Optimize fullmm TLB flushing Peter Zijlstra
2011-03-02 17:59   ` Peter Zijlstra
2011-03-02 17:59   ` Peter Zijlstra
2011-03-02 17:59 ` [RFC][PATCH 2/6] mm: Change flush_tlb_range() to take an mm_struct Peter Zijlstra
2011-03-02 17:59   ` Peter Zijlstra
2011-03-02 17:59   ` Peter Zijlstra
2011-03-02 19:19   ` Linus Torvalds
2011-03-02 19:19     ` Linus Torvalds
2011-03-02 20:58     ` Rik van Riel
2011-03-02 20:58       ` Rik van Riel
2011-03-02 21:40     ` Peter Zijlstra
2011-03-02 21:40       ` Peter Zijlstra
2011-03-02 21:47       ` David Miller
2011-03-02 21:47         ` David Miller
2011-03-03 17:22         ` Chris Metcalf
2011-03-03 17:22           ` Chris Metcalf
2011-03-03 17:22           ` Chris Metcalf
2011-03-03 18:45           ` David Miller
2011-03-03 18:45             ` David Miller
2011-03-03 18:56             ` Chris Metcalf
2011-03-03 18:56               ` Chris Metcalf
2011-03-03 18:56               ` Chris Metcalf
2011-03-10 18:05           ` [PATCH] arch/tile: optimize icache flush Chris Metcalf
2011-03-10 18:05             ` Chris Metcalf
2011-03-10 18:05             ` Chris Metcalf
2011-03-10 23:19             ` Rik van Riel
2011-03-10 23:19               ` Rik van Riel
2011-03-02 17:59 ` [RFC][PATCH 3/6] mm: Provide generic range tracking and flushing Peter Zijlstra
2011-03-02 17:59   ` Peter Zijlstra
2011-03-02 17:59   ` Peter Zijlstra
2011-03-02 17:59 ` [RFC][PATCH 4/6] arm, mm: Convert arm to generic tlb Peter Zijlstra
2011-03-02 17:59   ` Peter Zijlstra
2011-03-02 17:59   ` Peter Zijlstra
2011-03-09 15:16   ` Catalin Marinas
2011-03-09 15:16     ` Catalin Marinas
2011-03-09 15:19     ` Peter Zijlstra
2011-03-09 15:19       ` Peter Zijlstra
2011-03-09 15:36       ` Catalin Marinas
2011-03-09 15:36         ` Catalin Marinas
2011-03-09 15:39         ` Peter Zijlstra
2011-03-09 15:39           ` Peter Zijlstra
2011-03-09 15:48           ` Peter Zijlstra
2011-03-09 15:48             ` Peter Zijlstra
2011-03-09 16:34             ` Catalin Marinas
2011-03-09 16:34               ` Catalin Marinas
2012-05-17  3:05   ` Paul Mundt
2012-05-17  3:05     ` Paul Mundt
2012-05-17  9:30     ` Catalin Marinas
2012-05-17  9:30       ` Catalin Marinas
2012-05-17  9:39       ` Catalin Marinas
2012-05-17  9:39         ` Catalin Marinas
2012-05-17  9:51       ` Russell King
2012-05-17  9:51         ` Russell King
2012-05-17 11:28         ` Peter Zijlstra
2012-05-17 11:28           ` Peter Zijlstra
2012-05-17 12:14           ` Catalin Marinas
2012-05-17 12:14             ` Catalin Marinas
2012-05-17 16:00           ` Catalin Marinas
2012-05-17 16:00             ` Catalin Marinas
2012-05-17 16:24             ` Peter Zijlstra
2012-05-17 16:24               ` Peter Zijlstra
2012-05-17 16:33               ` Peter Zijlstra
2012-05-17 16:33                 ` Peter Zijlstra
2012-05-17 16:44                 ` Peter Zijlstra
2012-05-17 16:44                   ` Peter Zijlstra
2012-05-17 16:59                   ` Peter Zijlstra [this message]
2012-05-17 16:59                     ` Peter Zijlstra
2012-05-17 17:01                   ` Catalin Marinas
2012-05-17 17:01                     ` Catalin Marinas
2012-05-17 17:11                     ` Peter Zijlstra
2012-05-17 17:11                       ` Peter Zijlstra
2012-05-21  7:47               ` Martin Schwidefsky
2012-05-21  7:47                 ` Martin Schwidefsky
2012-05-17 17:22             ` Russell King
2012-05-17 17:22               ` Russell King
2012-05-17 18:31               ` Catalin Marinas
2012-05-17 18:31                 ` Catalin Marinas
2011-03-02 17:59 ` [RFC][PATCH 5/6] ia64, mm: Convert ia64 " Peter Zijlstra
2011-03-02 17:59   ` Peter Zijlstra
2011-03-02 17:59   ` Peter Zijlstra
2011-03-02 17:59 ` [RFC][PATCH 6/6] sh, mm: Convert sh " Peter Zijlstra
2011-03-02 17:59   ` Peter Zijlstra
2011-03-02 17:59   ` Peter Zijlstra

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=1337273959.4281.62.camel@twins \
    --to=peterz@infradead.org \
    --cc=aarcange@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=benh@kernel.crashing.org \
    --cc=catalin.marinas@arm.com \
    --cc=cmetcalf@tilera.com \
    --cc=davem@davemloft.net \
    --cc=hugh.dickins@tiscali.co.uk \
    --cc=lethal@linux-sh.org \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mel@csn.ul.ie \
    --cc=mingo@elte.hu \
    --cc=npiggin@kernel.dk \
    --cc=riel@redhat.com \
    --cc=rmk@arm.linux.org.uk \
    --cc=schwidefsky@de.ibm.com \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.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 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.