[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <87124c1d-5b95-4499-aa4b-22f2ecd376bc@suse.cz>
Date: Wed, 30 Oct 2024 12:03:10 +0100
From: Vlastimil Babka <vbabka@...e.cz>
To: John Hubbard <jhubbard@...dia.com>, Christoph Hellwig <hch@...radead.org>
Cc: Andrew Morton <akpm@...ux-foundation.org>,
LKML <linux-kernel@...r.kernel.org>, linux-mm@...ck.org,
linux-stable@...r.kernel.org, Vivek Kasireddy <vivek.kasireddy@...el.com>,
David Hildenbrand <david@...hat.com>, Dave Airlie <airlied@...hat.com>,
Gerd Hoffmann <kraxel@...hat.com>, Matthew Wilcox <willy@...radead.org>,
Jason Gunthorpe <jgg@...dia.com>, Peter Xu <peterx@...hat.com>,
Arnd Bergmann <arnd@...db.de>, Daniel Vetter <daniel.vetter@...ll.ch>,
Dongwon Kim <dongwon.kim@...el.com>, Hugh Dickins <hughd@...gle.com>,
Junxiao Chang <junxiao.chang@...el.com>,
Mike Kravetz <mike.kravetz@...cle.com>, Oscar Salvador <osalvador@...e.de>
Subject: Re: [PATCH] mm/gup: restore the ability to pin more than 2GB at a
time
On 10/30/24 05:39, John Hubbard wrote:
> On 10/29/24 9:33 PM, Christoph Hellwig wrote:
>> On Tue, Oct 29, 2024 at 09:30:41PM -0700, John Hubbard wrote:
>>> I do, yes. And what happens is that when you use GPUs, drivers like
>>> to pin system memory, and then point the GPU page tables to that
>>> memory. For older GPUs that don't support replayable page faults,
>>> that's required.
>>>
>>> So this behavior has been around forever.
>>>
>>> The customer was qualifying their software and noticed that before
>>> Linux 6.10, they could allocate >2GB, and with 6.11, they could
>>> not.
>>>
>>> Whether it is "wise" for user space to allocate that much at once
>>> is a reasonable question, but at least one place is (or was!) doing
>>> it.
>>
>> Still missing a callchain, which make me suspect that it is your weird
>> out of tree driver, in which case this simply does not matter.
>>
>
> I expect I could piece together something with Nouveau, given enough
> time and help from Ben Skeggs and Danillo and all...
>
> Yes, this originated with the out of tree driver. But it never occurred
> to me that upstream be uninterested in an obvious fix to an obvious
> regression.
It might be a regression even if you don't try to pin over 2GB. high-order
(>costly order) allocations can fail and/or cause disruptive
reclaim/compaction cycles even below MAX_PAGE_ORDER and it's better to use
kvmalloc if physical contiguity is not needed, it will attempt the physical
kmalloc() allocation with __GFP_NORETRY (little disruption) and fallback to
vmalloc() quickly.
Of course if there's a way to avoid the allocation completely, even beter.
>
> thanks,
Powered by blists - more mailing lists