From: Matthew Wilcox <willy@infradead.org>
To: Jaewon Kim <jaewon31.kim@gmail.com>
Cc: gregkh@linuxfoundation.org, m.szyprowski@samsung.com,
linux-mm@kvack.org, linux-usb@vger.kernel.org,
linux-kernel@vger.kernel.org,
Jaewon Kim <jaewon31.kim@samsung.com>,
ytk.lee@samsung.com
Subject: Re: [RFC PATCH] usb: host: xhci: allow __GFP_FS in dma allocation
Date: Fri, 17 May 2019 09:34:20 -0700 [thread overview]
Message-ID: <20190517163420.GG31704@bombadil.infradead.org> (raw)
In-Reply-To: <CAJrd-UuMRdWHky4gkmiR0QYozfXW0O35Ohv6mJPFx2TLa8hRKg@mail.gmail.com>
On Sat, May 18, 2019 at 01:02:28AM +0900, Jaewon Kim wrote:
> Hello I don't have enough knowledge on USB core but I've wondered
> why GFP_NOIO has been used in xhci_alloc_dev for
> xhci_alloc_virt_device. I found commit ("a6d940dd759b xhci: Use
> GFP_NOIO during device reset"). But can we just change GFP_NOIO
> to __GFP_RECLAIM | __GFP_FS ?
No. __GFP_FS implies __GFP_IO; you can't set __GFP_FS and clear __GFP_IO.
It seems like the problem you have is using the CMA to do DMA allocation.
Why would you do such a thing?
> Please refer to below case.
>
> I got a report from Lee YongTaek <ytk.lee@samsung.com> that the
> xhci_alloc_virt_device was too slow over 2 seconds only for one page
> allocation.
>
> 1) It was because kernel version was v4.14 and DMA allocation was
> done from CMA(Contiguous Memory Allocator) where CMA region was
> almost filled with file page and CMA passes GFP down to page
> isolation. And the page isolation only allows file page isolation only to
> requests having __GFP_FS.
>
> 2) Historically CMA was changed at v4.19 to use GFP_KERNEL
> regardless of GFP passed to DMA allocation through the
> commit 6518202970c1 "(mm/cma: remove unsupported gfp_mask
> parameter from cma_alloc()".
>
> I think pre v4.19 the xhci_alloc_virt_device could be very slow
> depending on CMA situation but free to USB deadlock issue. But as of
> v4.19, I think, it will be fast but can face the deadlock issue.
> Consequently I think to meet the both cases, I think USB can pass
> __GFP_FS without __GFP_IO.
>
> If __GFP_FS is passed from USB core, of course, the CMA patch also
> need to be changed to pass GFP.
next prev parent reply other threads:[~2019-05-17 16:34 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-05-17 16:02 [RFC PATCH] usb: host: xhci: allow __GFP_FS in dma allocation Jaewon Kim
2019-05-17 16:34 ` Matthew Wilcox [this message]
2019-05-18 1:53 ` Jaewon Kim
2019-05-20 5:56 ` Christoph Hellwig
2019-05-20 9:09 ` Oliver Neukum
2019-05-20 10:12 ` Christoph Hellwig
2019-05-20 14:16 ` Alan Stern
2019-05-20 14:23 ` Christoph Hellwig
2019-05-21 8:54 ` Oliver Neukum
2019-05-21 13:27 ` Christoph Hellwig
2019-05-21 13:11 ` Oliver Neukum
2019-05-21 14:00 ` Alan Stern
2019-05-22 6:31 ` Oliver Neukum
2019-05-22 14:56 ` Alan Stern
2019-05-22 20:47 ` Oliver Neukum
2019-05-23 14:01 ` Alan Stern
2019-05-28 12:34 ` Oliver Neukum
2019-05-28 14:25 ` Alan Stern
2019-05-23 12:32 ` Oliver Neukum
2019-05-23 16:35 ` Christoph Hellwig
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=20190517163420.GG31704@bombadil.infradead.org \
--to=willy@infradead.org \
--cc=gregkh@linuxfoundation.org \
--cc=jaewon31.kim@gmail.com \
--cc=jaewon31.kim@samsung.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-usb@vger.kernel.org \
--cc=m.szyprowski@samsung.com \
--cc=ytk.lee@samsung.com \
/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).