[gentoo-commits] gentoo-x86 commit in sys-kernel/spl/files: spl-0.6.0_rc14-no-cond_resched.patch

[gentoo-commits] gentoo-x86 commit in sys-kernel/spl/files: spl-0.6.0_rc14-no-cond_resched.patch

[Richard Yao (ryao)]

ryao        13/03/23 23:48:08

  Added:                spl-0.6.0_rc14-no-cond_resched.patch
  Log:
  Free memory more quickly under memory pressure; do not dodoc INSTALL (deprecated upstream); fix /usr/src/spl symlink in 9999 ebuild (thanks likewhoa)
  
  (Portage version: 2.2.0_alpha169/cvs/Linux x86_64, signed Manifest commit with key 0xBEE84C64)

Revision  Changes    Path
1.1                  sys-kernel/spl/files/spl-0.6.0_rc14-no-cond_resched.patch

file : http://sources.gentoo.org/viewvc.cgi/gentoo-x86/sys-kernel/spl/files/spl-0.6.0_rc14-no-cond_resched.patch?rev=1.1&view=markup
plain: http://sources.gentoo.org/viewvc.cgi/gentoo-x86/sys-kernel/spl/files/spl-0.6.0_rc14-no-cond_resched.patch?rev=1.1&content-type=text/plain

Index: spl-0.6.0_rc14-no-cond_resched.patch
===================================================================
From 58a382c73ad3393d7591421950624e75d3c4aea1 Mon Sep 17 00:00:00 2001
From: Richard Yao <ryao@cs.stonybrook.edu>
Date: Thu, 21 Mar 2013 13:21:11 -0400
Subject: [PATCH] Do not call cond_resched() in spl_slab_reclaim()

Calling cond_resched() after each object is freed and then after each
slab is freed can cause slabs of objects to live for excessive periods
of time following reclaimation. This interferes with the kernel's own
memory management when called from kswapd and can cause direct reclaim
to occur in response to memory pressure that should have been resolved.

Signed-off-by: Richard Yao <ryao@cs.stonybrook.edu>
---
 module/spl/spl-kmem.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/module/spl/spl-kmem.c b/module/spl/spl-kmem.c
index f9c1114..a0ca2d2 100644
--- a/module/spl/spl-kmem.c
+++ b/module/spl/spl-kmem.c
@@ -1112,14 +1112,11 @@ char *kvasprintf(gfp_t gfp, const char *fmt, va_list ap)
 
 		if (skc->skc_flags & KMC_OFFSLAB)
 			kv_free(skc, sko->sko_addr, size);
-
-		cond_resched();
 	}
 
 	list_for_each_entry_safe(sks, m, &sks_list, sks_list) {
 		ASSERT(sks->sks_magic == SKS_MAGIC);
 		kv_free(skc, sks, skc->skc_slab_size);
-		cond_resched();
 	}
 
 	SEXIT;
-- 
1.8.1.5