[PATCH 01/05] mm fix __alloc_pages cpuset ALLOC_* flags

From: Paul Jackson (pj_at_sgi.com)
Date: 11/14/05

  • Next message: Paul Jackson: "[PATCH 04/05] mm simplify __alloc_pages cpuset hardwall logic" 
    Date:	Sun, 13 Nov 2005 20:03:29 -0800 (PST)
To: akpm@osdl.org, linux-kernel@vger.kernel.org

    Two changes to the setting of the ALLOC_CPUSET flag in
    mm/page_alloc.c:__alloc_pages()

     1) A bug fix - the "ignoring mins" case should not be honoring
        ALLOC_CPUSET. This case of all cases, since it is handling a
        request that will free up more memory than is asked for (exiting
        tasks, e.g.) should be allowed to escape cpuset constraints
        when memory is tight.

     2) A logic change to make it simpler. Honor cpusets even on
        GFP_ATOMIC (!wait) requests. With this, cpuset confinement
        applies to all requests except ALLOC_NO_WATERMARKS, so that
        in a subsequent cleanup patch, I can remove the ALLOC_CPUSET
        flag entirely. Since I don't know any real reason this
        logic has to be either way, I am choosing the path of the
        simplest code.

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

    ---
 mm/page_alloc.c |    5 ++---
 1 files changed, 2 insertions(+), 3 deletions(-)
--- 2.6.14-mm2.orig/mm/page_alloc.c	2005-11-12 22:25:03.305301135 -0800
+++ 2.6.14-mm2/mm/page_alloc.c	2005-11-12 22:27:30.519813285 -0800
@@ -933,8 +933,7 @@ restart:
 		alloc_flags |= ALLOC_HARDER;
 	if (gfp_mask & __GFP_HIGH)
 		alloc_flags |= ALLOC_HIGH;
-	if (wait)
-		alloc_flags |= ALLOC_CPUSET;
+	alloc_flags |= ALLOC_CPUSET;
 
 	/*
 	 * Go through the zonelist again. Let __GFP_HIGH and allocations
@@ -956,7 +955,7 @@ restart:
 nofail_alloc:
 			/* go through the zonelist yet again, ignoring mins */
 			page = get_page_from_freelist(gfp_mask, order,
-				zonelist, ALLOC_NO_WATERMARKS|ALLOC_CPUSET);
+				zonelist, ALLOC_NO_WATERMARKS);
 			if (page)
 				goto got_pg;
 			if (gfp_mask & __GFP_NOFAIL) {
-- 
                          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/
  • Next message: Paul Jackson: "[PATCH 04/05] mm simplify __alloc_pages cpuset hardwall logic"

    Relevant Pages