> > > So there is
> > > already an API functionality to do this; allocation and init are
> > > already separated.
> >
> > So let's drop the init in usb_alloc_urb()?
>
> FILL_* were originally intended to be just convenience macros, not a
> part of the API.
Hmm, I guess I tend not to see macros
> Excellent news!!! Good job.
Thanks!
>I'm happy the ISD guys were helpful.
Very much so. Also worth mentioning: Cypress, Intel, NetChip.
- Dave
___
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
http://lists.sourceforge.net/
I'm looking for someone who meets the following 3 criteria. If you
qualify, please let me know.
(1) Has a Sony MemoryStick reader that does not work properly when changing
media
(2) Has some coding experience
(3) Has a desire to help make this work
Matt
--
Matthew Dharm
Hello,
If I'm understanding everything correct you should do this to get a
feature report to a usb device:
/* set report_type and report_id in rinfo */
IOCGREPORTINFO(struct hiddev_report_info* rinfo)
IOCGREPORT(struct hiddev_report_info* rinfo)
/* set uref's report_id, report_type field_index
On Sun, Jun 17, 2001 at 08:22:51PM +0200, Wolfgang M?es wrote:
> On Sunday, 17. June 2001 16:18, David Brownell wrote:
>
> [spin_lock_init in usb_alloc_urb()]
>
> > I'm not sure I see what the deal is. The FILL_*_URB() macros
> > initialize URBs, including doing the spin lock init.
>
> I was n
Dave,
Excellent news!!! Good job. I'm happy the ISD guys were helpful.
Regards
John H.
>
> Message: 8
> Date: Sun, 17 Jun 2001 10:15:36 -0700
> From: David Brownell <[EMAIL PROTECTED]>
> To: [EMAIL PROTECTED]
> Subject: [linux-usb-devel] USB 2.0 and usb-storage -- works!
> Reply-To: [EMAIL PR
> [initialisation by hand]
>
> > Yet it is ugly. It'll break whenever the internal URB handling is
> > changed.
>
> Yes. Absolutely.
>
> > It isn't really that hard to add a function to the API to do that.
> > In fact the attached patch does just that.
>
> I don't like your patch. Because:
> - you
On Sunday, 17. June 2001 16:18, David Brownell wrote:
[spin_lock_init in usb_alloc_urb()]
> I'm not sure I see what the deal is. The FILL_*_URB() macros
> initialize URBs, including doing the spin lock init.
I was not aware of this...
So spin_lock_init() ist doubled: once in usb_alloc_urb(), o
Just thought I'd drop a quick status note. The current EHCI
driver is behaving with the In-System adapter that's found in
the current set of HighSpeed-capable USB storage units.
(And the NEC controller found in USB 2.0 PCI cards.)
This means is that you can get megabytes per second disk
transfer
> It isn't really that hard to add a function to the API to do that.
> In fact the attached patch does just that.
> Last time I proposed such an interface it was rejected.
I'm not sure I see what the deal is. The FILL_*_URB() macros
initialize URBs, including doing the spin lock init. So there
10 matches
Mail list logo
