Messages in this thread Patch in this message | | | Date | Mon, 28 Feb 2011 10:53:16 +0100 | From | Michal Hocko <> | Subject | Re: [RFC PATCH] page_cgroup: Reduce allocation overhead for page_cgroup array for CONFIG_SPARSEMEM v4 |
| |
On Mon 28-02-11 18:23:22, KAMEZAWA Hiroyuki wrote: [...] > > From 84a9555741b59cb2a0a67b023e4bd0f92c670ca1 Mon Sep 17 00:00:00 2001 > > From: Michal Hocko <mhocko@suse.cz> > > Date: Thu, 24 Feb 2011 11:25:44 +0100 > > Subject: [PATCH] page_cgroup: Reduce allocation overhead for page_cgroup array for CONFIG_SPARSEMEM > > > > Currently we are allocating a single page_cgroup array per memory > > section (stored in mem_section->base) when CONFIG_SPARSEMEM is selected. > > This is correct but memory inefficient solution because the allocated > > memory (unless we fall back to vmalloc) is not kmalloc friendly: > > - 32b - 16384 entries (20B per entry) fit into 327680B so the > > 524288B slab cache is used > > - 32b with PAE - 131072 entries with 2621440B fit into 4194304B > > - 64b - 32768 entries (40B per entry) fit into 2097152 cache > > > > This is ~37% wasted space per memory section and it sumps up for the > > whole memory. On a x86_64 machine it is something like 6MB per 1GB of > > RAM. > > > > We can reduce the internal fragmentation by using alloc_pages_exact > > which allocates PAGE_SIZE aligned blocks so we will get down to <4kB > > wasted memory per section which is much better. > > > > We still need a fallback to vmalloc because we have no guarantees that > > we will have a continuous memory of that size (order-10) later on during > > the hotplug events. > > > > Signed-off-by: Michal Hocko <mhocko@suse.cz> > > CC: Dave Hansen <dave@linux.vnet.ibm.com> > > CC: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> > > Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Thanks. I will repost it with Andrew in the CC.
> > But...nitpick, it may be from my fault.. [...] > > +static void free_page_cgroup(void *addr) > > +{ > > + if (is_vmalloc_addr(addr)) { > > + vfree(addr); > > + } else { > > + struct page *page = virt_to_page(addr); > > + if (!PageReserved(page)) { /* Is bootmem ? */ > > I think we never see PageReserved if we just use alloc_pages_exact()/vmalloc().
I have checked that and we really do not (unless I am missing some subtle side effects). Anyway, I think we still should at least BUG_ON on that.
> Maybe my old patch was not enough and this kind of junks are remaining in > the original code.
Should I incorporate it into the patch. I think that a separate one would be better for readability.
--- From e7a897a42b526620eb4afada2d036e1c9ff9e62a Mon Sep 17 00:00:00 2001 From: Michal Hocko <mhocko@suse.cz> Date: Mon, 28 Feb 2011 10:43:12 +0100 Subject: [PATCH] page_cgroup array is never stored on reserved pages
KAMEZAWA Hiroyuki noted that free_pages_cgroup doesn't have to check for PageReserved because we never store the array on reserved pages (neither alloc_pages_exact nor vmalloc use those pages).
So we can replace the check by a BUG_ON.
Signed-off-by: Michal Hocko <mhocko@suse.cz> CC: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> --- mm/page_cgroup.c | 8 ++++---- 1 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/mm/page_cgroup.c b/mm/page_cgroup.c index eae3cd2..dd5f789 100644 --- a/mm/page_cgroup.c +++ b/mm/page_cgroup.c @@ -125,10 +125,10 @@ static void free_page_cgroup(void *addr) vfree(addr); } else { struct page *page = virt_to_page(addr); - if (!PageReserved(page)) { /* Is bootmem ? */ - size_t table_size = sizeof(struct page_cgroup) * PAGES_PER_SECTION; - free_pages_exact(addr, table_size); - } + size_t table_size = sizeof(struct page_cgroup) * PAGES_PER_SECTION; + + BUG_ON(PageReserved(page)); + free_pages_exact(addr, table_size); } } -- 1.7.2.3 -- Michal Hocko SUSE Labs SUSE LINUX s.r.o. Lihovarska 1060/12 190 00 Praha 9 Czech Republic
| |