From: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
To: Takashi Iwai <tiwai-l3A5Bk7waGM@public.gmane.org>
Cc: "Konrad Rzeszutek Wilk"
<konrad.wilk-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
"Christian König" <christian.koenig-5C7GfCeVMHo@public.gmane.org>,
iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org,
"Christoph Hellwig" <hch-jcswGhMUV9g@public.gmane.org>
Subject: Re: [PATCH] swiotlb: Fix unexpected swiotlb_alloc_coherent() failures
Date: Thu, 12 Apr 2018 08:02:27 +0200 [thread overview]
Message-ID: <20180412060227.GA30248@lst.de> (raw)
In-Reply-To: <s5h4lkib3rd.wl-tiwai-l3A5Bk7waGM@public.gmane.org>
On Wed, Apr 11, 2018 at 09:28:54AM +0200, Takashi Iwai wrote:
> > But we should try a GFP_DMA32 allocation first, so this is a bit
> > surprising.
>
> Hm, do we really try that?
> Through a quick glance, dma_alloc_coherent_gfp_flags() gives GFP_DMA32
> only when coherent mask <= DMA_BIT_MASK(32); in the case of iwlwifi,
> it's 36bit, so GFP_DMA isn't set.
Oh, yes - it is using an odd dma mask, and amdgpu seems to use an
just as odd 40-bit dma mask.
> We had a fallback allocation with GFP_DMA32 in the past, but this
> seems gone long time ago along with cleanups (commit c647c3bb2d16).
>
> But I haven't followed about this topic for long time, so I might have
> missed obviously...
I think a fallback would be much better here rather than relying on the
limited swiotlb buffer bool. dma_direct_alloc (which in 4.17 is also
used for x86) already has a GFP_DMA fallback, so extending this for
GFP_DMA32 as well would seem reasonable.
Any volunteers?
WARNING: multiple messages have this Message-ID (diff)
From: Christoph Hellwig <hch@lst.de>
To: Takashi Iwai <tiwai@suse.de>
Cc: "Christoph Hellwig" <hch@lst.de>,
"Robin Murphy" <robin.murphy@arm.com>,
linux-kernel@vger.kernel.org, iommu@lists.linux-foundation.org,
"Christian König" <christian.koenig@amd.com>,
"Konrad Rzeszutek Wilk" <konrad.wilk@oracle.com>
Subject: Re: [PATCH] swiotlb: Fix unexpected swiotlb_alloc_coherent() failures
Date: Thu, 12 Apr 2018 08:02:27 +0200 [thread overview]
Message-ID: <20180412060227.GA30248@lst.de> (raw)
In-Reply-To: <s5h4lkib3rd.wl-tiwai@suse.de>
On Wed, Apr 11, 2018 at 09:28:54AM +0200, Takashi Iwai wrote:
> > But we should try a GFP_DMA32 allocation first, so this is a bit
> > surprising.
>
> Hm, do we really try that?
> Through a quick glance, dma_alloc_coherent_gfp_flags() gives GFP_DMA32
> only when coherent mask <= DMA_BIT_MASK(32); in the case of iwlwifi,
> it's 36bit, so GFP_DMA isn't set.
Oh, yes - it is using an odd dma mask, and amdgpu seems to use an
just as odd 40-bit dma mask.
> We had a fallback allocation with GFP_DMA32 in the past, but this
> seems gone long time ago along with cleanups (commit c647c3bb2d16).
>
> But I haven't followed about this topic for long time, so I might have
> missed obviously...
I think a fallback would be much better here rather than relying on the
limited swiotlb buffer bool. dma_direct_alloc (which in 4.17 is also
used for x86) already has a GFP_DMA fallback, so extending this for
GFP_DMA32 as well would seem reasonable.
Any volunteers?
next prev parent reply other threads:[~2018-04-12 6:02 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-04-10 17:05 [PATCH] swiotlb: Fix unexpected swiotlb_alloc_coherent() failures Takashi Iwai
2018-04-10 17:05 ` Takashi Iwai
2018-04-10 17:06 ` Christoph Hellwig
[not found] ` <20180410170615.GA27589-jcswGhMUV9g@public.gmane.org>
2018-04-10 17:07 ` Takashi Iwai
2018-04-10 17:07 ` Takashi Iwai
[not found] ` <s5hzi2bat25.wl-tiwai-l3A5Bk7waGM@public.gmane.org>
2018-04-10 17:50 ` Robin Murphy
2018-04-10 17:50 ` Robin Murphy
[not found] ` <40618b9c-786d-7f5f-3f5b-fc1dd8375846-5wv7dgnIgG8@public.gmane.org>
2018-04-10 18:10 ` Christoph Hellwig
2018-04-10 18:10 ` Christoph Hellwig
[not found] ` <20180410181020.GA28565-jcswGhMUV9g@public.gmane.org>
2018-04-11 7:28 ` Takashi Iwai
2018-04-11 7:28 ` Takashi Iwai
[not found] ` <s5h4lkib3rd.wl-tiwai-l3A5Bk7waGM@public.gmane.org>
2018-04-12 6:02 ` Christoph Hellwig [this message]
2018-04-12 6:02 ` Christoph Hellwig
[not found] ` <20180412060227.GA30248-jcswGhMUV9g@public.gmane.org>
2018-04-12 8:03 ` Takashi Iwai
2018-04-12 8:03 ` Takashi Iwai
[not found] ` <s5h3700am1f.wl-tiwai-l3A5Bk7waGM@public.gmane.org>
2018-04-12 8:19 ` Takashi Iwai
2018-04-12 8:19 ` Takashi Iwai
[not found] ` <s5hy3hs96rq.wl-tiwai-l3A5Bk7waGM@public.gmane.org>
2018-04-12 8:27 ` Takashi Iwai
2018-04-12 8:27 ` Takashi Iwai
[not found] ` <s5hwoxc96d5.wl-tiwai-l3A5Bk7waGM@public.gmane.org>
2018-04-12 10:32 ` Robin Murphy
2018-04-12 10:32 ` Robin Murphy
[not found] ` <81b9eae2-11d4-162d-9d3a-12e06eff9fea-5wv7dgnIgG8@public.gmane.org>
2018-04-15 8:43 ` Takashi Iwai
2018-04-15 8:43 ` Takashi Iwai
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=20180412060227.GA30248@lst.de \
--to=hch-jcswghmuv9g@public.gmane.org \
--cc=christian.koenig-5C7GfCeVMHo@public.gmane.org \
--cc=iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \
--cc=konrad.wilk-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=tiwai-l3A5Bk7waGM@public.gmane.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.