linux-kernel - Re: [PATCH 2/2] media: usb: pwc: Don't use coherent DMA buffers for ISO transfer

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <1556658.LS2rrRvGR3@avalon>
Date:   Mon, 30 Jul 2018 18:35:48 +0300
From:   Laurent Pinchart <laurent.pinchart@...asonboard.com>
To:     "Matwey V. Kornilov" <matwey@....msu.ru>
Cc:     Alan Stern <stern@...land.harvard.edu>,
        Tomasz Figa <tfiga@...omium.org>,
        Ezequiel Garcia <ezequiel@...labora.com>,
        Hans de Goede <hdegoede@...hat.com>,
        Hans Verkuil <hverkuil@...all.nl>,
        Mauro Carvalho Chehab <mchehab@...nel.org>,
        Steven Rostedt <rostedt@...dmis.org>, mingo@...hat.com,
        Mike Isely <isely@...ox.com>,
        Bhumika Goyal <bhumirks@...il.com>,
        Colin King <colin.king@...onical.com>,
        Linux Media Mailing List <linux-media@...r.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Kieran Bingham <kieran.bingham@...asonboard.com>,
        keiichiw@...omium.org
Subject: Re: [PATCH 2/2] media: usb: pwc: Don't use coherent DMA buffers for ISO transfer

Hi Matwey,

On Tuesday, 24 July 2018 21:56:09 EEST Matwey V. Kornilov wrote:
> 2018-07-23 21:57 GMT+03:00 Alan Stern:
> > On Mon, 23 Jul 2018, Matwey V. Kornilov wrote:
> >> I've tried to strategies:
> >> 
> >> 1) Use dma_unmap and dma_map inside the handler (I suppose this is
> >> similar to how USB core does when there is no URB_NO_TRANSFER_DMA_MAP)
> > 
> > Yes.
> > 
> >> 2) Use sync_cpu and sync_device inside the handler (and dma_map only
> >> once at memory allocation)
> >> 
> >> It is interesting that dma_unmap/dma_map pair leads to the lower
> >> overhead (+1us) than sync_cpu/sync_device (+2us) at x86_64 platform.
> >> At armv7l platform using dma_unmap/dma_map  leads to ~50 usec in the
> >> handler, and sync_cpu/sync_device - ~65 usec.
> >> 
> >> However, I am not sure is it mandatory to call
> >> dma_sync_single_for_device for FROM_DEVICE direction?
> > 
> > According to Documentation/DMA-API-HOWTO.txt, the CPU should not write
> > to a DMA_FROM_DEVICE-mapped area, so dma_sync_single_for_device() is
> > not needed.
> 
> Well, I measured the following at armv7l. The handler execution time
> (URB_NO_TRANSFER_DMA_MAP is used for all cases):
> 
> 1) coherent DMA: ~3000 usec (pwc is not functional)
> 2) explicit dma_unmap and dma_map in the handler: ~52 usec
> 3) explicit dma_sync_single_for_cpu (no dma_sync_single_for_device): ~56
> usec

I really don't understand why the sync option is slower. Could you please 
investigate ? Before doing anything we need to make sure we have a full 
understanding of the problem.

> So, I suppose that unfortunately Tomasz suggestion doesn't work. There
> is no performance improvement when dma_sync_single is used.
> 
> At x86_64 the following happens:
> 
> 1) coherent DMA: ~2 usec

What do you mean by coherent DMA for x86_64 ? Is that usb_alloc_coherent() ? 
Could you trace it to see how memory is allocated exactly, and how it's mapped 
to the CPU ? I suspect that it will end up in dma_direct_alloc() but I'd like 
a confirmation.

> 2) explicit dma_unmap and dma_map in the handler: ~3.5 usec
> 3) explicit dma_sync_single_for_cpu (no dma_sync_single_for_device): ~4 usec
> 
> So, whats to do next? Personally, I think that DMA streaming API
> introduces not so great overhead.

It might not be very large, but with USB3 cameras at high resolutions and 
framerates, it might still become noticeable. I wouldn't degrade performances 
on x86, especially if we can decide which option to use based on the platform 
(or perhaps even better based on Kconfig options such as DMA_NONCOHERENT).

> Does anybody happy with turning to streaming DMA or I'll introduce
> module-level switch as Ezequiel suggested?

A module-level switch isn't a good idea, it will just confuse users. We need 
to establish a strategy and come up with a good heuristic that can be applied 
at compile and/or runtime to automatically decide how to allocate buffers.

-- 
Regards,

Laurent Pinchart