lkml.org 
[lkml]   [2005]   [Nov]   [15]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
Subject[PATCH] mm redo ALLOC_* flag names again
More consistent mm/page_alloc.c ALLOC_* flag names.

Signed-off-by: Paul Jackson <pj@sgi.com>

---

mm/page_alloc.c | 28 ++++++++++++++--------------
1 files changed, 14 insertions(+), 14 deletions(-)

--- 2.6.14-mm2.orig/mm/page_alloc.c 2005-11-13 10:32:33.090792563 -0800
+++ 2.6.14-mm2/mm/page_alloc.c 2005-11-15 11:00:30.967888018 -0800
@@ -755,10 +755,10 @@ buffered_rmqueue(struct zone *zone, int
return page;
}

-#define ALLOC_DONT_DIP 0x01 /* don't dip into memory reserves */
-#define ALLOC_DIP_SOME 0x02 /* dip into reserves some */
-#define ALLOC_DIP_ALOT 0x04 /* dip into reserves further */
-#define ALLOC_MUSTHAVE 0x08 /* ignore all constraints */
+#define ALLOC_DIP_NONE 0x01 /* don't dip into memory reserves */
+#define ALLOC_DIP_LESS 0x02 /* dip into reserves some */
+#define ALLOC_DIP_MORE 0x04 /* dip into reserves further */
+#define ALLOC_DIP_FULL 0x08 /* ignore all constraints */

/*
* Return 1 if free pages are above 'mark'. This takes into account the order
@@ -771,9 +771,9 @@ int zone_watermark_ok(struct zone *z, in
long min = mark, free_pages = z->free_pages - (1 << order) + 1;
int o;

- if (alloc_flags & ALLOC_DIP_SOME)
+ if (alloc_flags & ALLOC_DIP_LESS)
min -= min / 2;
- if (alloc_flags & ALLOC_DIP_ALOT)
+ if (alloc_flags & ALLOC_DIP_MORE)
min -= min / 4;

if (free_pages <= min + z->lowmem_reserve[classzone_idx])
@@ -814,8 +814,8 @@ get_page_from_freelist(gfp_t gfp_mask, u
* See also cpuset_zone_allowed() comment in kernel/cpuset.c.
*/
do {
- if (!(alloc_flags & ALLOC_MUSTHAVE)) {
- if (alloc_flags == ALLOC_DONT_DIP)
+ if (!(alloc_flags & ALLOC_DIP_FULL)) {
+ if (alloc_flags == ALLOC_DIP_NONE)
gfp_mask |= __GFP_HARDWALL;
if (!cpuset_zone_allowed(*z, gfp_mask))
continue;
@@ -911,7 +911,7 @@ __alloc_pages(gfp_t gfp_mask, unsigned i
}
restart:
page = get_page_from_freelist(gfp_mask, order,
- zonelist, ALLOC_DONT_DIP);
+ zonelist, ALLOC_DIP_NONE);
if (page)
goto got_pg;

@@ -927,13 +927,13 @@ restart:
* The caller may dip into page reserves a bit more if the caller
* cannot run direct reclaim, or if the caller has realtime scheduling
* policy or is asking for __GFP_HIGH memory. GFP_ATOMIC requests will
- * set both ALLOC_DIP_ALOT (!wait) and ALLOC_DIP_SOME (__GFP_HIGH).
+ * set both ALLOC_DIP_MORE (!wait) and ALLOC_DIP_LESS (__GFP_HIGH).
*/
alloc_flags = 0;
if ((unlikely(rt_task(p)) && !in_interrupt()) || !wait)
- alloc_flags |= ALLOC_DIP_ALOT;
+ alloc_flags |= ALLOC_DIP_MORE;
if (gfp_mask & __GFP_HIGH)
- alloc_flags |= ALLOC_DIP_SOME;
+ alloc_flags |= ALLOC_DIP_LESS;

page = get_page_from_freelist(gfp_mask, order, zonelist, alloc_flags);
if (page)
@@ -947,7 +947,7 @@ restart:
nofail_alloc:
/* go through the zonelist yet again, ignoring mins */
page = get_page_from_freelist(gfp_mask, order,
- zonelist, ALLOC_MUSTHAVE);
+ zonelist, ALLOC_DIP_FULL);
if (page)
goto got_pg;
if (gfp_mask & __GFP_NOFAIL) {
@@ -991,7 +991,7 @@ rebalance:
* under heavy pressure.
*/
page = get_page_from_freelist(gfp_mask, order,
- zonelist, ALLOC_DONT_DIP);
+ zonelist, ALLOC_DIP_NONE);
if (page)
goto got_pg;

--
I won't rest till it's the best ...
Programmer, Linux Scalability
Paul Jackson <pj@sgi.com> 1.650.933.1373
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-11-15 20:25    [W:0.029 / U:0.052 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site