On Tue, Jun 25, 2013 at 12:26 AM, Alan Stern wrote:
>
> Other than usbtest, the only driver using SG that I know of is
> usb-storage, and it does make that assumption. It works because the
Another example is usbfs driver, which sets the SG size as 16KB and also
makes the assumption.
> block lay
On Mon, 24 Jun 2013, Felipe Balbi wrote:
> > MUSB shouldn't be copying data to and from bounce buffers just to be
> > able to handle SG tranfers.
>
> well, if we can assume all SG elements to be wMaxPacketSize aligned, we
> could just break it up into smaller URBs and avoid the copy, but I
> wasn
On Tue, Jun 25, 2013 at 12:09:38AM +0800, Ming Lei wrote:
> On Mon, Jun 24, 2013 at 11:44 PM, Felipe Balbi wrote:
> >
> > BTW, EHCI can handle "short" SGs in the first page. See page 97 of EHCI
> > 1.0 spec.
>
> That isn't enough, EHCI requires that size of the 3 middle buffers are 4K.
right, I
On Mon, 24 Jun 2013, Felipe Balbi wrote:
> BTW, EHCI can handle "short" SGs in the first page. See page 97 of EHCI
> 1.0 spec.
I'm not sure what you're talking about. SG is used for bulk transfers,
whereas p. 97 is about interrupt transfers.
It's true that the hardware's abilities aren't fully
Hi,
On Mon, Jun 24, 2013 at 12:04:19PM -0400, Alan Stern wrote:
> On Mon, 24 Jun 2013, Felipe Balbi wrote:
> > > > you wouldn't notice the difference. The DMA engine is the one which
> > > > would read the sgtable to figure where the data is scattered, at the end
> > > > of the day, SW only knows
On Mon, Jun 24, 2013 at 11:44 PM, Felipe Balbi wrote:
>
> BTW, EHCI can handle "short" SGs in the first page. See page 97 of EHCI
> 1.0 spec.
That isn't enough, EHCI requires that size of the 3 middle buffers are 4K.
Thanks,
--
Ming Lei
--
To unsubscribe from this list: send the line "unsubscr
On Mon, 24 Jun 2013, Felipe Balbi wrote:
> > > you wouldn't notice the difference. The DMA engine is the one which
> > > would read the sgtable to figure where the data is scattered, at the end
> > > of the day, SW only knows of a single 1044bytes URB and controller is
> > > required to generate p
Hi,
On Mon, Jun 24, 2013 at 05:53:25PM +0200, Sebastian Andrzej Siewior wrote:
> On 06/24/2013 05:46 PM, Felipe Balbi wrote:
> >> The DMA engines (intern or extern) do not merge transfers. Some
> >> of them are clever enough to transfer a single 1024 as two 512
> >> requests. But two 256 bytes r
On 06/24/2013 05:46 PM, Felipe Balbi wrote:
>> The DMA engines (intern or extern) do not merge transfers. Some
>> of them are clever enough to transfer a single 1024 as two 512
>> requests. But two 256 bytes requests are not merged into one 512
>> request. I believe the device side of dwc3 is ab
On Mon, Jun 24, 2013 at 05:35:42PM +0200, Sebastian Andrzej Siewior wrote:
> On 06/24/2013 05:18 PM, Felipe Balbi wrote:
> >> It is a little. The first USB packet has 512 vs 502 bytes on the
> >> wire.
> >
> > you wouldn't notice the difference. The DMA engine is the one
> > which would read the s
On Mon, Jun 24, 2013 at 06:42:05PM +0300, Felipe Balbi wrote:
> On Mon, Jun 24, 2013 at 11:32:16AM -0400, Alan Stern wrote:
> > > > >> - the change violates USB spec(1.1/2.0/3.0)
> > > > >
> > > > > I can't see how this would violate USB spec. USB specifications
> > > > > have no knowledge of scat
Hi,
On Mon, Jun 24, 2013 at 11:32:16AM -0400, Alan Stern wrote:
> > > >> - the change violates USB spec(1.1/2.0/3.0)
> > > >
> > > > I can't see how this would violate USB spec. USB specifications
> > > > have no knowledge of scatter-gather.
> > > >
> > > > It really doesn't matter how the data
On 06/24/2013 05:18 PM, Felipe Balbi wrote:
>> It is a little. The first USB packet has 512 vs 502 bytes on the
>> wire.
>
> you wouldn't notice the difference. The DMA engine is the one
> which would read the sgtable to figure where the data is scattered,
> at the end of the day, SW only knows of
Hi,
On Mon, Jun 24, 2013 at 11:20:32PM +0800, Ming Lei wrote:
> On Mon, Jun 24, 2013 at 11:04 PM, Felipe Balbi wrote:
> > On Mon, Jun 24, 2013 at 10:56:30PM +0800, Ming Lei wrote:
> >> On Mon, Jun 24, 2013 at 10:33 PM, Denis V. Lunev wrote:
> >> > From: Konstantin Filatov
> >> >
> >> > This pat
On Mon, 24 Jun 2013, Felipe Balbi wrote:
> Hi,
>
> On Mon, Jun 24, 2013 at 05:12:50PM +0200, Sebastian Andrzej Siewior wrote:
> > On 06/24/2013 05:04 PM, Felipe Balbi wrote:
> > >> - the change violates USB spec(1.1/2.0/3.0)
> > >
> > > I can't see how this would violate USB spec. USB specificat
On Mon, 24 Jun 2013, Sebastian Andrzej Siewior wrote:
> On 06/24/2013 05:04 PM, Felipe Balbi wrote:
> >> - the change violates USB spec(1.1/2.0/3.0)
> >
> > I can't see how this would violate USB spec. USB specifications
> > have no knowledge of scatter-gather.
> >
> > It really doesn't matter h
On Mon, Jun 24, 2013 at 11:04 PM, Felipe Balbi wrote:
> On Mon, Jun 24, 2013 at 10:56:30PM +0800, Ming Lei wrote:
>> On Mon, Jun 24, 2013 at 10:33 PM, Denis V. Lunev wrote:
>> > From: Konstantin Filatov
>> >
>> > This patch shortens TD's packet not only for the last TD in sg list,
>> > but also
Hi,
On Mon, Jun 24, 2013 at 05:12:50PM +0200, Sebastian Andrzej Siewior wrote:
> On 06/24/2013 05:04 PM, Felipe Balbi wrote:
> >> - the change violates USB spec(1.1/2.0/3.0)
> >
> > I can't see how this would violate USB spec. USB specifications
> > have no knowledge of scatter-gather.
> >
> > I
On 06/24/2013 05:04 PM, Felipe Balbi wrote:
>> - the change violates USB spec(1.1/2.0/3.0)
>
> I can't see how this would violate USB spec. USB specifications
> have no knowledge of scatter-gather.
>
> It really doesn't matter how the data gets into the HW's FIFO, as
> long as it *does* get there
On Mon, Jun 24, 2013 at 10:56:30PM +0800, Ming Lei wrote:
> On Mon, Jun 24, 2013 at 10:33 PM, Denis V. Lunev wrote:
> > From: Konstantin Filatov
> >
> > This patch shortens TD's packet not only for the last TD in sg list,
> > but also for the last TD in sg element.
> >
> > Signed-off-by: Konstant
On Mon, Jun 24, 2013 at 10:33 PM, Denis V. Lunev wrote:
> From: Konstantin Filatov
>
> This patch shortens TD's packet not only for the last TD in sg list,
> but also for the last TD in sg element.
>
> Signed-off-by: Konstantin Filatov
> Signed-off-by: Denis V. Lunev
Considered that:
- the ch
21 matches
Mail list logo
