Ive worked around the problem by using get free pages to allocate the
URB buffer
Toby
Toby Collett wrote:
If I submit a single URB of PAGE_SIZE then everything appears to
succeed however the bulk endpoint is then stalled and I recieve an
EPIPE on the next URB.
If I clear the stall state on the
split
bulk requests? The snooping i did of the windows driver always uses a
single full size URB read.
Toby
Oliver Neukum wrote:
Am Dienstag, 6. Juli 2004 00:32 schrieb Toby Collett:
Hi,
sorry to be a pain about this but when I use rvmalloc I still have the
same issues as using vmalloc
Hi,
sorry to be a pain about this but when I use rvmalloc I still have the
same issues as using vmalloc, basically I submit my bulk read urb, it
comes back claiming success with the correct 0x4b000 bytes read, however
the data buffer has not been touched...
I looked at all of the drivers in the
the oc511 and usb-skeleton drivers as a base, however the
ov511 only allocates 32000 odd bytes using kmalloc)
Toby Collett
---
This SF.Net email sponsored by Black Hat Briefings & Training.
Attend Black Hat Briefings & Training, Las Vegas
Hi,
It turns out that I had failed to modify the usb skeleton driver
properly and it was changing my read size back to 512 after i set it to
0x4b000.
Thanks for your help anyway,
Toby
Toby Collett wrote:
Hi,
At the moment I just allocate it in kernel space during the usb probe...
dev
Hi,
At the moment I just allocate it in kernel space during the usb probe...
dev->bulk_in_buffer = kmalloc (buffer_size, GFP_KERNEL);
this may not be the best place for it, but the allocate doesnt fail...
Once Im able to get image data from the device ill be adding v4l support
and then ill rearran
Hi,
I am attempting to write a USB driver for a usb 2.0 camera device (from
vrmagic), I have captured the output of the setup stuff (control
commands) from the windows demo app and I am attempting to reproduce
this in linux
I have the basics all going and I get all the expected replies from
