[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20180222133643.GJ30681@dhcp22.suse.cz>
Date: Thu, 22 Feb 2018 14:36:43 +0100
From: Michal Hocko <mhocko@...nel.org>
To: Matthew Wilcox <willy@...radead.org>
Cc: Dave Hansen <dave.hansen@...el.com>,
Konstantin Khlebnikov <khlebnikov@...dex-team.ru>,
linux-kernel@...r.kernel.org,
Christoph Hellwig <hch@...radead.org>, linux-mm@...ck.org,
Andy Lutomirski <luto@...nel.org>,
Andrew Morton <akpm@...ux-foundation.org>,
"Kirill A. Shutemov" <kirill@...temov.name>
Subject: Re: Use higher-order pages in vmalloc
On Thu 22-02-18 04:22:54, Matthew Wilcox wrote:
> On Thu, Feb 22, 2018 at 07:59:43AM +0100, Michal Hocko wrote:
> > On Wed 21-02-18 09:01:29, Matthew Wilcox wrote:
> > > Right. It helps with fragmentation if we can keep higher-order
> > > allocations together.
> >
> > Hmm, wouldn't it help if we made vmalloc pages migrateable instead? That
> > would help the compaction and get us to a lower fragmentation longterm
> > without playing tricks in the allocation path.
>
> I was wondering about that possibility. If we want to migrate a page
> then we have to shoot down the PTE across all CPUs, copy the data to the
> new page, and insert the new PTE. Copying 4kB doesn't take long; if you
> have 12GB/s (current example on Wikipedia: dual-channel memory and one
> DDR2-800 module per channel gives a theoretical bandwidth of 12.8GB/s)
> then we should be able to copy a page in 666ns). So there's no problem
> holding a spinlock for it.
>
> But we can't handle a fault in vmalloc space today. It's handled in
> arch-specific code, see vmalloc_fault() in arch/x86/mm/fault.c
> If we're going to do this, it'll have to be something arches opt into
> because I'm not taking on the job of fixing every architecture!
yes.
> > Maybe we should consider kvmalloc for the kernel stack?
>
> We'd lose the guard page, so it'd have to be something we let the
> sysadmin decide to do.
ohh, right, I forgot about the guard page.
--
Michal Hocko
SUSE Labs
Powered by blists - more mailing lists