LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Andi Kleen <ak@suse.de>
To: tglx@linutronix.de, mingo@elte.hu, linux-kernel@vger.kernel.org
Subject: [PATCH] [2/8] CPA: Flush the caches when setting pages not present
Date: Mon, 11 Feb 2008 10:34:30 +0100 (CET)	[thread overview]
Message-ID: <20080211093430.D198A1B41CE@basil.firstfloor.org> (raw)
In-Reply-To: <200802111034.764275766@suse.de>


e.g. the AMD64 pci-gart code sets pages not present to prevent potential 
cache coherency problems.  When doing this it is probably safer to flush the 
caches too. So consider clearing of the present bit as a cache flush indicator.

Note that debug pagealloc marks pages regularly not present either, but it won't
call this because it calls directly into a lower level function.

I have not actually seen failures from this, but it seems safer 
to do it this way.

Signed-off-by: Andi Kleen <ak@suse.de>

---
 arch/x86/mm/pageattr.c |   13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

Index: linux/arch/x86/mm/pageattr.c
===================================================================
--- linux.orig/arch/x86/mm/pageattr.c
+++ linux/arch/x86/mm/pageattr.c
@@ -670,9 +670,16 @@ static int __change_page_attr_set_clr(st
 	return 0;
 }
 
-static inline int cache_attr(pgprot_t attr)
+static inline int cache_attr(pgprot_t set, pgprot_t clr)
 {
-	return pgprot_val(attr) &
+	/*
+	 * Clearing pages is usually done for cache cohereny reasons
+	 * (except for pagealloc debug, but that doesn't call this anyways)
+	 * It's safer to flush the caches in this case too.
+	 */
+	if (pgprot_val(clr) & _PAGE_PRESENT)
+		return 1;
+	return pgprot_val(set) &
 		(_PAGE_PAT | _PAGE_PAT_LARGE | _PAGE_PWT | _PAGE_PCD);
 }
 
@@ -709,7 +716,7 @@ static int change_page_attr_set_clr(unsi
 	 * No need to flush, when we did not set any of the caching
 	 * attributes:
 	 */
-	cache = cache_attr(mask_set);
+	cache = cache_attr(mask_set, mask_clr);
 
 	/*
 	 * On success we use clflush, when the CPU supports it to

  parent reply	other threads:[~2008-02-11  9:34 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-02-11  9:34 [PATCH] [0/8] Various kernel mapping bug fixes Andi Kleen
2008-02-11  9:34 ` [PATCH] [1/8] CPA: Fix gbpages support in try_preserve_lage_page Andi Kleen
2008-02-11  9:45   ` Thomas Gleixner
2008-02-11 10:12     ` Ingo Molnar
2008-02-11 11:01       ` Andi Kleen
2008-02-11  9:34 ` Andi Kleen [this message]
2008-02-11 11:00   ` [PATCH] [2/8] CPA: Flush the caches when setting pages not present Ingo Molnar
2008-02-11 12:26     ` Andi Kleen
2008-02-11  9:34 ` [PATCH] [3/8] CPA: Test the correct mapping alias on x86-64 Andi Kleen
2008-02-11 11:49   ` Ingo Molnar
2008-02-11  9:34 ` [PATCH] [4/8] CPA: Fix set_memory_x for ioremap Andi Kleen
2008-02-11 12:27   ` Ingo Molnar
2008-02-11 12:45     ` Andi Kleen
2008-02-11  9:34 ` [PATCH] [5/8] Fix logic error in 64bit memory hotadd Andi Kleen
2008-02-11 12:46   ` Ingo Molnar
2008-02-11 13:05     ` Andi Kleen
2008-02-11 13:33       ` Ingo Molnar
2008-02-11 13:44         ` Andi Kleen
2008-02-12 10:35       ` Yasunori Goto
2008-02-12 11:20         ` Andi Kleen
2008-02-11  9:34 ` [PATCH] [6/8] Account overlapped mappings in end_pfn_map Andi Kleen
2008-02-11 13:08   ` Ingo Molnar
2008-02-11 13:27     ` Andi Kleen
2008-02-11 13:55       ` Ingo Molnar
2008-02-11 14:16       ` Peter Zijlstra
2008-02-11 14:24         ` Andi Kleen
2008-02-11 14:41           ` Sam Ravnborg
2008-02-11 15:12   ` Arjan van de Ven
2008-02-11  9:34 ` [PATCH] [7/8] Implement true end_pfn_mapped for 32bit Andi Kleen
2008-02-12 19:39   ` Thomas Gleixner
2008-02-12 19:49     ` Andi Kleen
2008-02-12 20:25       ` Thomas Gleixner
2008-02-11  9:34 ` [PATCH] [8/8] RFC: Fix some EFI problems Andi Kleen
2008-02-12 20:04   ` Thomas Gleixner
2008-02-12 20:23     ` Andi Kleen
2008-02-12 20:48       ` Thomas Gleixner
2008-02-13 11:05         ` Andi Kleen

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=20080211093430.D198A1B41CE@basil.firstfloor.org \
    --to=ak@suse.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=tglx@linutronix.de \
    --subject='Re: [PATCH] [2/8] CPA: Flush the caches when setting pages not present' \
    /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

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).