linux-kernel - Re: [PATCH 1/2] dma: Add interface to calculate data transferred

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] [day] [month] [year] [list]

Date:	Wed, 16 Oct 2013 11:17:33 -0700
From:	Dan Williams <dan.j.williams@...el.com>
To:	Vinod Koul <vinod.koul@...el.com>
Cc:	Greg KH <gregkh@...uxfoundation.org>,
	"Shevchenko, Andriy" <andriy.shevchenko@...el.com>,
	Youquan Song <youquan.song@...ux.jf.intel.com>,
	"Song, Youquan" <youquan.song@...el.com>,
	"Westerberg, Mika" <mika.westerberg@...el.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Dave Jiang <dave.jiang@...el.com>
Subject: Re: [PATCH 1/2] dma: Add interface to calculate data transferred

On Wed, Oct 16, 2013 at 8:07 AM, Vinod Koul <vinod.koul@...el.com> wrote:
> On Wed, Oct 16, 2013 at 07:12:47AM -0700, Greg KH wrote:
>> On Wed, Oct 16, 2013 at 09:13:20AM +0000, Shevchenko, Andriy wrote:
>> > On Wed, 2013-10-16 at 13:27 +0530, Vinod Koul wrote:
>> > > On Wed, Oct 16, 2013 at 02:06:30PM +0530, Shevchenko, Andriy wrote:
>> > > > On Wed, 2013-10-16 at 11:08 +0530, Vinod Koul wrote:
>> >
>> > []
>> >
>> > > > > Well am going to take it a bit further and ask you why do we need the
>> > > > > /sys/class/dma? I have never used it for slave work.
>> > > >
>> > > > How user (who, f.e., would like to run dmatest) will know names of the
>> > > > channels provided?
>> > > Ok dmatest requires this, I overlooked that part
>> > >
>> > > > How could we see what channels of certain dma controller are requested /
>> > > > busy from userspace?
>> > > But do end user care or need to know about this?
>> >
>> > The idea is to have some test cases. So, end user probably doesn't
>> > really need this.
>>
>> So could this just be debugfs files instead?
> Yes I was thinking on same lines and would agree with you on that. This _really_
> doesn't need sysfs unless Dan some async usage dependent on it. While at it we
> can review the fields and remove the ones which are not required...
>
> Dan, let me know if you are okay with it, I will try to get that done over the
> weekend...
>

Channel naming is the primary use case and several drivers rely on the
channel name in their print messages, some for naming resources, and
of course dmatest for identifying channels.  ioatdma uses it as a
parent device for allocating its per-channel 'quickdata' attributes to
advertise the capabilities and the ring_size which is useful for
determining channel loading and verifying the BIOS configuration.

An argument could be made to move the ioatdma attributes to debugfs,
but I don't see a clean/worthwhile way of ripping and replacing the
channel naming scheme for drivers.

--
Dan
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/