From: Rik van Riel <riel@redhat.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Jerome Glisse <j.glisse@gmail.com>,
Peter Zijlstra <peterz@infradead.org>,
linux-mm <linux-mm@kvack.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
linux-fsdevel <linux-fsdevel@vger.kernel.org>,
Mel Gorman <mgorman@suse.de>, "H. Peter Anvin" <hpa@zytor.com>,
Andrew Morton <akpm@linux-foundation.org>,
Linda Wang <lwang@redhat.com>, Kevin E Martin <kem@redhat.com>,
Jerome Glisse <jglisse@redhat.com>,
Andrea Arcangeli <aarcange@redhat.com>,
Johannes Weiner <jweiner@redhat.com>,
Larry Woodman <lwoodman@redhat.com>,
Dave Airlie <airlied@redhat.com>, Jeff Law <law@redhat.com>,
Brendan Conoboy <blc@redhat.com>,
Joe Donohue <jdonohue@redhat.com>,
Duncan Poole <dpoole@nvidia.com>,
Sherry Cheung <SCheung@nvidia.com>,
Subhash Gutti <sgutti@nvidia.com>,
John Hubbard <jh
Subject: Re: [RFC] Heterogeneous memory management (mirror process address space on a device mmu).
Date: Tue, 06 May 2014 12:47:16 -0400 [thread overview]
Message-ID: <53691214.80906@redhat.com> (raw)
In-Reply-To: <CA+55aFwQWRKpcaR_-GvhMbUXE-n5yjEi_a_Um7=Bb_xbdQtFFg@mail.gmail.com>
On 05/06/2014 12:34 PM, Linus Torvalds wrote:
> On Tue, May 6, 2014 at 9:30 AM, Rik van Riel <riel@redhat.com> wrote:
>>
>> The GPU runs a lot faster when using video memory, instead
>> of system memory, on the other side of the PCIe bus.
>
> The nineties called, and they want their old broken model back.
>
> Get with the times. No high-performance future GPU will ever run
> behind the PCIe bus. We still have a few straggling historical
> artifacts, but everybody knows where the future is headed.
>
> They are already cache-coherent because flushing caches etc was too
> damn expensive. They're getting more so.
I suppose that VRAM could simply be turned into a very high
capacity CPU cache for the GPU, for the case where people
want/need an add-on card.
With a few hundred MB of "CPU cache" on the video card, we
could offload processing to the GPU very easily, without
having to worry about multiple address or page table formats
on the CPU side.
A new generation of GPU hardware seems to come out every
six months or so, so I guess we could live with TLB
invalidations to the first generations of hardware being
comically slow :)
--
All rights reversed
--
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/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2014-05-06 16:47 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-05-02 13:51 [RFC] Heterogeneous memory management (mirror process address space on a device mmu) j.glisse
2014-05-02 13:52 ` [PATCH 01/11] mm: differentiate unmap for vmscan from other unmap j.glisse
2014-05-02 13:52 ` [PATCH 02/11] mmu_notifier: add action information to address invalidation j.glisse
2014-05-02 13:52 ` [PATCH 03/11] mmu_notifier: pass through vma to invalidate_range and invalidate_page j.glisse
2014-05-02 13:52 ` [PATCH 04/11] interval_tree: helper to find previous item of a node in rb interval tree j.glisse
2014-05-02 13:52 ` [PATCH 05/11] mm/memcg: support accounting null page and transfering null charge to new page j.glisse
2014-05-02 13:52 ` [PATCH 06/11] hmm: heterogeneous memory management j.glisse
2014-05-02 13:52 ` [PATCH 07/11] hmm: support moving anonymous page to remote memory j.glisse
2014-05-02 13:52 ` [PATCH 08/11] hmm: support for migrate file backed pages " j.glisse
2014-05-02 13:52 ` [PATCH 09/11] fs/ext4: add support for hmm migration to remote memory of pagecache j.glisse
2014-05-02 13:52 ` [PATCH 10/11] hmm/dummy: dummy driver to showcase the hmm api j.glisse
2014-05-02 13:52 ` [PATCH 11/11] hmm/dummy_driver: add support for fake remote memory using pages j.glisse
2014-05-06 10:29 ` [RFC] Heterogeneous memory management (mirror process address space on a device mmu) Peter Zijlstra
2014-05-06 14:57 ` Linus Torvalds
2014-05-06 15:00 ` Jerome Glisse
2014-05-06 15:18 ` Linus Torvalds
2014-05-06 15:33 ` Jerome Glisse
2014-05-06 15:42 ` Rik van Riel
2014-05-06 15:47 ` Linus Torvalds
2014-05-06 16:18 ` Jerome Glisse
2014-05-06 16:32 ` Linus Torvalds
2014-05-06 16:49 ` Jerome Glisse
2014-05-06 17:28 ` Jerome Glisse
2014-05-06 17:43 ` Linus Torvalds
2014-05-06 18:13 ` Jerome Glisse
2014-05-06 18:22 ` Linus Torvalds
2014-05-06 18:38 ` Jerome Glisse
2014-05-07 7:18 ` Benjamin Herrenschmidt
2014-05-07 7:14 ` Benjamin Herrenschmidt
2014-05-07 12:39 ` Jerome Glisse
2014-05-09 1:26 ` Jerome Glisse
2014-05-10 4:28 ` Benjamin Herrenschmidt
2014-05-11 0:48 ` Jerome Glisse
2014-05-06 16:30 ` Rik van Riel
2014-05-06 16:34 ` Linus Torvalds
2014-05-06 16:47 ` Rik van Riel [this message]
2014-05-06 16:54 ` Jerome Glisse
2014-05-06 18:02 ` H. Peter Anvin
2014-05-06 18:26 ` Jerome Glisse
2014-05-06 22:44 ` David Airlie
2014-05-07 2:33 ` Davidlohr Bueso
2014-05-07 13:00 ` Peter Zijlstra
2014-05-07 17:34 ` Davidlohr Bueso
2014-05-07 16:21 ` Linus Torvalds
2014-05-08 16:47 ` sagi grimberg
2014-05-08 17:56 ` Jerome Glisse
2014-05-09 1:42 ` Davidlohr Bueso
2014-05-09 1:45 ` Jerome Glisse
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=53691214.80906@redhat.com \
--to=riel@redhat.com \
--cc=SCheung@nvidia.com \
--cc=aarcange@redhat.com \
--cc=airlied@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=blc@redhat.com \
--cc=dpoole@nvidia.com \
--cc=hpa@zytor.com \
--cc=j.glisse@gmail.com \
--cc=jdonohue@redhat.com \
--cc=jglisse@redhat.com \
--cc=jweiner@redhat.com \
--cc=kem@redhat.com \
--cc=law@redhat.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lwang@redhat.com \
--cc=lwoodman@redhat.com \
--cc=mgorman@suse.de \
--cc=peterz@infradead.org \
--cc=sgutti@nvidia.com \
--cc=torvalds@linux-foundation.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).