From: Rene Herman <rene.herman@keyaccess.nl>
To: Ingo Molnar <mingo@elte.hu>
Cc: Venki Pallipadi <venkatesh.pallipadi@intel.com>,
Dave Airlie <airlied@gmail.com>,
"Li, Shaohua" <shaohua.li@intel.com>,
Yinghai Lu <yhlu.kernel@gmail.com>,
Andreas Herrmann <andreas.herrmann3@amd.com>,
Arjan van de Ven <arjan@infradead.org>,
Linux Kernel <linux-kernel@vger.kernel.org>,
"Siddha, Suresh B" <suresh.b.siddha@intel.com>,
Thomas Gleixner <tglx@linutronix.de>,
"H. Peter Anvin" <hpa@zytor.com>,
Dave Jones <davej@codemonkey.org.uk>
Subject: [PATCH] x86: have set_memory_array_{uc,wb} coalesce memtypes.
Date: Fri, 22 Aug 2008 01:02:20 +0200 [thread overview]
Message-ID: <48ADF3FC.7070002@keyaccess.nl> (raw)
In-Reply-To: <48ADA2C2.8090905@keyaccess.nl>
[-- Attachment #1: Type: text/plain, Size: 1068 bytes --]
On 21-08-08 19:15, Rene Herman wrote:
> On 21-08-08 14:06, Ingo Molnar wrote:
>> Would be nice to test tip/master - it has both that patch included and
>> the latest pageattr-array API (with enablement in AGP drivers)
>> patchset, done by Shaohua Li, based on Dave's original patch.
>
> That patch by itself doesn't help any -- the new set_memory_array_uc()
> still calls reserve_memtype() for each single page in the array. To
> help, it needs to coalesce adjacent entries, which is itself easy to do
> were it not for the need to keep the list of reserved (base,end) pairs
> around for free_memtype() time (and halfway fail time).
Actually, might as well simply reconstruct the memtype list at free time
I guess. How is this for a coalescing version of the array functions?
NOTE: I am posting this because I'm going to bed but haven't stared
comfortably long at this and might be buggy. Compiles, boots and
provides me with:
root@7ixe4:~# wc -l /debug/x86/pat_memtype_list
53 /debug/x86/pat_memtype_list
otherwise (down from 16384+).
<snore>
Rene.
[-- Attachment #2: 0001-x86-have-set_memory_array_-uc-wb-coalesce-memtypes.patch --]
[-- Type: text/plain, Size: 2236 bytes --]
>From b5dc6e481b38cf4e7792bcb9a8f5dd9aab0e5590 Mon Sep 17 00:00:00 2001
From: Rene Herman <rene.herman@gmail.com>
Date: Fri, 22 Aug 2008 00:56:00 +0200
Subject: [PATCH] x86: have set_memory_array_{uc,wb} coalesce memtypes.
---
arch/x86/mm/pageattr.c | 38 ++++++++++++++++++++++++++++++++------
1 files changed, 32 insertions(+), 6 deletions(-)
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index d49e4db..a2a497a 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -942,21 +942,38 @@ EXPORT_SYMBOL(set_memory_uc);
int set_memory_array_uc(unsigned long *addr, int addrinarray)
{
+ unsigned long start;
+ unsigned long end;
int i;
/*
* for now UC MINUS. see comments in ioremap_nocache()
*/
for (i = 0; i < addrinarray; i++) {
- if (reserve_memtype(__pa(addr[i]), __pa(addr[i]) + PAGE_SIZE,
- _PAGE_CACHE_UC_MINUS, NULL))
+ start = __pa(addr[i]);
+ for (end = start + PAGE_SIZE; i < addrinarray - 1; end += PAGE_SIZE) {
+ if (end != __pa(addr[i + 1]))
+ break;
+ i++;
+ }
+ if (reserve_memtype(start, end, _PAGE_CACHE_UC_MINUS, NULL))
goto out;
}
return change_page_attr_set(addr, addrinarray,
__pgprot(_PAGE_CACHE_UC_MINUS), 1);
out:
- while (--i >= 0)
- free_memtype(__pa(addr[i]), __pa(addr[i]) + PAGE_SIZE);
+ for (i = 0; i < addrinarray; i++) {
+ unsigned long tmp = __pa(addr[i]);
+
+ if (tmp == start)
+ break;
+ for (end = start + PAGE_SIZE; i < addrinarray - 1; end += PAGE_SIZE) {
+ if (end != __pa(addr[i + 1]))
+ break;
+ i++;
+ }
+ free_memtype(tmp, end);
+ }
return -EINVAL;
}
EXPORT_SYMBOL(set_memory_array_uc);
@@ -997,9 +1014,18 @@ EXPORT_SYMBOL(set_memory_wb);
int set_memory_array_wb(unsigned long *addr, int addrinarray)
{
int i;
- for (i = 0; i < addrinarray; i++)
- free_memtype(__pa(addr[i]), __pa(addr[i]) + PAGE_SIZE);
+ for (i = 0; i < addrinarray; i++) {
+ unsigned long start = __pa(addr[i]);
+ unsigned long end;
+
+ for (end = start + PAGE_SIZE; i < addrinarray - 1; end += PAGE_SIZE) {
+ if (end != __pa(addr[i + 1]))
+ break;
+ i++;
+ }
+ free_memtype(start, end);
+ }
return change_page_attr_clear(addr, addrinarray,
__pgprot(_PAGE_CACHE_MASK), 1);
}
--
1.5.5
next prev parent reply other threads:[~2008-08-21 23:01 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-08-04 16:30 AGP and PAT (induced?) problem (on AMD family 6) Rene Herman
2008-08-06 13:51 ` Andreas Herrmann
2008-08-06 20:57 ` Rene Herman
2008-08-11 9:46 ` Rene Herman
2008-08-15 14:22 ` Ingo Molnar
2008-08-15 15:24 ` Rene Herman
2008-08-19 10:11 ` Rene Herman
2008-08-19 10:26 ` Ingo Molnar
2008-08-19 14:19 ` Rene Herman
2008-08-19 19:07 ` Venki Pallipadi
2008-08-19 19:22 ` Rene Herman
2008-08-19 23:28 ` Venki Pallipadi
2008-08-20 10:09 ` Ingo Molnar
2008-08-20 10:04 ` Ingo Molnar
2008-08-20 10:50 ` Rene Herman
2008-08-20 14:27 ` Rene Herman
2008-08-20 19:41 ` Venki Pallipadi
2008-08-20 21:40 ` Rene Herman
2008-08-20 21:46 ` Dave Airlie
2008-08-20 22:16 ` Venki Pallipadi
2008-08-21 3:42 ` Andi Kleen
2008-08-21 21:13 ` Suresh Siddha
2008-08-22 2:12 ` Andi Kleen
2008-08-21 12:06 ` Ingo Molnar
2008-08-21 17:15 ` Rene Herman
2008-08-21 22:10 ` [PATCH] x86: {reverve,free}_memtype() take a physical address Rene Herman
2008-08-21 22:16 ` Pallipadi, Venkatesh
2008-08-21 22:26 ` Rene Herman
2008-08-21 22:57 ` Pallipadi, Venkatesh
2008-08-21 23:06 ` Rene Herman
2008-08-21 23:02 ` Rene Herman [this message]
2008-08-22 4:15 ` [PATCH] x86: have set_memory_array_{uc,wb} coalesce memtypes Ingo Molnar
2008-08-22 19:08 ` Venki Pallipadi
2008-08-22 20:15 ` Rene Herman
2008-08-23 15:33 ` Ingo Molnar
2008-08-22 20:02 ` Rene Herman
2008-09-10 19:52 ` AGP PAT issue Rene Herman
2008-09-11 8:17 ` Ingo Molnar
2008-09-11 8:30 ` Rene Herman
2008-09-13 0:26 ` Pallipadi, Venkatesh
2008-09-13 0:44 ` Rene Herman
2008-10-09 15:53 ` Thomas Hellstrom
2008-10-13 17:10 ` Pallipadi, Venkatesh
2008-10-13 19:26 ` Thomas Hellström
2008-08-20 21:02 ` AGP and PAT (induced?) problem (on AMD family 6) Dave Airlie
2008-08-20 21:16 ` Rene Herman
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=48ADF3FC.7070002@keyaccess.nl \
--to=rene.herman@keyaccess.nl \
--cc=airlied@gmail.com \
--cc=andreas.herrmann3@amd.com \
--cc=arjan@infradead.org \
--cc=davej@codemonkey.org.uk \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=shaohua.li@intel.com \
--cc=suresh.b.siddha@intel.com \
--cc=tglx@linutronix.de \
--cc=venkatesh.pallipadi@intel.com \
--cc=yhlu.kernel@gmail.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