diff for duplicates of <20110503061156.GC10278@cmpxchg.org> diff --git a/a/1.txt b/N1/1.txt index 2d5371f..df1c786 100644 --- a/a/1.txt +++ b/N1/1.txt @@ -63,3 +63,10 @@ are checked. Good point, it should probably bail earlier on a zone that does not really contribute to the soft limit excess. + +-- +To unsubscribe, send a message with 'unsubscribe linux-mm' in +the body to majordomo@kvack.org. For more info on Linux MM, +see: http://www.linux-mm.org/ . +Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ +Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a> diff --git a/a/content_digest b/N1/content_digest index ef3f2e2..168f312 100644 --- a/a/content_digest +++ b/N1/content_digest @@ -80,6 +80,13 @@ "> contributing much to the global reclaim.\n" "\n" "Good point, it should probably bail earlier on a zone that does not\n" - really contribute to the soft limit excess. + "really contribute to the soft limit excess.\n" + "\n" + "--\n" + "To unsubscribe, send a message with 'unsubscribe linux-mm' in\n" + "the body to majordomo@kvack.org. For more info on Linux MM,\n" + "see: http://www.linux-mm.org/ .\n" + "Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/\n" + "Don't email: <a href=mailto:\"dont@kvack.org\"> email@kvack.org </a>" -c95c187945f6875734f326da63b42e546c77cfb2e360e8072dd6c79127be0e59 +bb5bf620ba90da6b88fe802dbc9b3812f0afdbe398c0b4f1e05a4e1e46ec3abf
diff --git a/a/1.txt b/N2/1.txt index 2d5371f..d1d7427 100644 --- a/a/1.txt +++ b/N2/1.txt @@ -9,20 +9,20 @@ On Mon, May 02, 2011 at 04:14:09PM -0700, Ying Han wrote: > >> > >> Is mitigateable if we prevent the cgroups code being so aggressive in > >> its zone shrinking (by reducing it's default shrink from 0 [everything] -> >> to DEF_PRIORITY [some things]). This will have an obvious knock on +> >> to DEF_PRIORITY [some things]). This will have an obvious knock on > >> effect to cgroup accounting, but it's better than hanging systems. > > -> > Actually, it's not that obvious. At least not to me. I added Balbir, +> > Actually, it's not that obvious. At least not to me. I added Balbir, > > who added said comment and code in the first place, to CC: Here is the > > comment in full quote: > > -> > /* -> > * NOTE: Although we can get the priority field, using it -> > * here is not a good idea, since it limits the pages we can scan. -> > * if we don't reclaim here, the shrink_zone from balance_pgdat -> > * will pick up pages from other mem cgroup's as well. We hack -> > * the priority and make it zero. -> > */ +> > /* +> > * NOTE: Although we can get the priority field, using it +> > * here is not a good idea, since it limits the pages we can scan. +> > * if we don't reclaim here, the shrink_zone from balance_pgdat +> > * will pick up pages from other mem cgroup's as well. We hack +> > * the priority and make it zero. +> > */ > > > > The idea is that if one memcg is above its softlimit, we prefer > > reducing pages from this memcg over reclaiming random other pages, @@ -30,16 +30,16 @@ On Mon, May 02, 2011 at 04:14:09PM -0700, Ying Han wrote: > > > > But the code flow looks like this: > > -> > balance_pgdat -> > mem_cgroup_soft_limit_reclaim -> > mem_cgroup_shrink_node_zone -> > shrink_zone(0, zone, &sc) -> > shrink_zone(prio, zone, &sc) +> > balance_pgdat +> > mem_cgroup_soft_limit_reclaim +> > mem_cgroup_shrink_node_zone +> > shrink_zone(0, zone, &sc) +> > shrink_zone(prio, zone, &sc) > > > > so the success of the inner memcg shrink_zone does at least not > > explicitely result in the outer, global shrink_zone steering clear of -> > other memcgs' pages. It just tries to move the pressure of balancing -> > the zones to the memcg with the biggest soft limit excess. That can +> > other memcgs' pages. It just tries to move the pressure of balancing +> > the zones to the memcg with the biggest soft limit excess. That can > > only really work if the memcg is a large enough contributor to the > > zone's total number of lru pages, though, and looks very likely to hit > > the exceeding memcg too hard in other cases. @@ -63,3 +63,10 @@ are checked. Good point, it should probably bail earlier on a zone that does not really contribute to the soft limit excess. + +-- +To unsubscribe, send a message with 'unsubscribe linux-mm' in +the body to majordomo@kvack.org. For more info on Linux MM, +see: http://www.linux-mm.org/ . +Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ +Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a> diff --git a/a/content_digest b/N2/content_digest index ef3f2e2..3996934 100644 --- a/a/content_digest +++ b/N2/content_digest @@ -27,20 +27,20 @@ "> >>\n" "> >> Is mitigateable if we prevent the cgroups code being so aggressive in\n" "> >> its zone shrinking (by reducing it's default shrink from 0 [everything]\n" - "> >> to DEF_PRIORITY [some things]). \302\240This will have an obvious knock on\n" + "> >> to DEF_PRIORITY [some things]). This will have an obvious knock on\n" "> >> effect to cgroup accounting, but it's better than hanging systems.\n" "> >\n" - "> > Actually, it's not that obvious. \302\240At least not to me. \302\240I added Balbir,\n" + "> > Actually, it's not that obvious. At least not to me. I added Balbir,\n" "> > who added said comment and code in the first place, to CC: Here is the\n" "> > comment in full quote:\n" "> >\n" - "> > \302\240 \302\240 \302\240 \302\240/*\n" - "> > \302\240 \302\240 \302\240 \302\240 * NOTE: Although we can get the priority field, using it\n" - "> > \302\240 \302\240 \302\240 \302\240 * here is not a good idea, since it limits the pages we can scan.\n" - "> > \302\240 \302\240 \302\240 \302\240 * if we don't reclaim here, the shrink_zone from balance_pgdat\n" - "> > \302\240 \302\240 \302\240 \302\240 * will pick up pages from other mem cgroup's as well. We hack\n" - "> > \302\240 \302\240 \302\240 \302\240 * the priority and make it zero.\n" - "> > \302\240 \302\240 \302\240 \302\240 */\n" + "> > /*\n" + "> > * NOTE: Although we can get the priority field, using it\n" + "> > * here is not a good idea, since it limits the pages we can scan.\n" + "> > * if we don't reclaim here, the shrink_zone from balance_pgdat\n" + "> > * will pick up pages from other mem cgroup's as well. We hack\n" + "> > * the priority and make it zero.\n" + "> > */\n" "> >\n" "> > The idea is that if one memcg is above its softlimit, we prefer\n" "> > reducing pages from this memcg over reclaiming random other pages,\n" @@ -48,16 +48,16 @@ "> >\n" "> > But the code flow looks like this:\n" "> >\n" - "> > \302\240 \302\240 \302\240 \302\240balance_pgdat\n" - "> > \302\240 \302\240 \302\240 \302\240 \302\240mem_cgroup_soft_limit_reclaim\n" - "> > \302\240 \302\240 \302\240 \302\240 \302\240 \302\240mem_cgroup_shrink_node_zone\n" - "> > \302\240 \302\240 \302\240 \302\240 \302\240 \302\240 \302\240shrink_zone(0, zone, &sc)\n" - "> > \302\240 \302\240 \302\240 \302\240 \302\240shrink_zone(prio, zone, &sc)\n" + "> > balance_pgdat\n" + "> > mem_cgroup_soft_limit_reclaim\n" + "> > mem_cgroup_shrink_node_zone\n" + "> > shrink_zone(0, zone, &sc)\n" + "> > shrink_zone(prio, zone, &sc)\n" "> >\n" "> > so the success of the inner memcg shrink_zone does at least not\n" "> > explicitely result in the outer, global shrink_zone steering clear of\n" - "> > other memcgs' pages. \302\240It just tries to move the pressure of balancing\n" - "> > the zones to the memcg with the biggest soft limit excess. \302\240That can\n" + "> > other memcgs' pages. It just tries to move the pressure of balancing\n" + "> > the zones to the memcg with the biggest soft limit excess. That can\n" "> > only really work if the memcg is a large enough contributor to the\n" "> > zone's total number of lru pages, though, and looks very likely to hit\n" "> > the exceeding memcg too hard in other cases.\n" @@ -80,6 +80,13 @@ "> contributing much to the global reclaim.\n" "\n" "Good point, it should probably bail earlier on a zone that does not\n" - really contribute to the soft limit excess. + "really contribute to the soft limit excess.\n" + "\n" + "--\n" + "To unsubscribe, send a message with 'unsubscribe linux-mm' in\n" + "the body to majordomo@kvack.org. For more info on Linux MM,\n" + "see: http://www.linux-mm.org/ .\n" + "Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/\n" + "Don't email: <a href=mailto:\"dont@kvack.org\"> email@kvack.org </a>" -c95c187945f6875734f326da63b42e546c77cfb2e360e8072dd6c79127be0e59 +9158c971a1591962acd759ab50f9817d3d68a5fc5668bd9ea6d4c228a858cf20
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.