Hi Laurent,

--- On Thu, 7/22/10, P F <[email protected]> wrote:
> 
> Also, without this workaround in place, I was able to
> extract two copies
> of a JPEG via two methods:
> 1. Through my application
> 2. Through an in-line hardware USB analyzer
> 
> 
> I've attached the two JPEGs mentioned above.
> 
While debugging further, I realized that the 32-byte chunks might not be
random data at all.  I extracted the previous JPEG in the test stream, and
lo and behold, the two corrupt 32-byte chunks were in fact duplicated from
this previous frame.

I thought that this duplication implied a dirty cache in the frame buffer,
so I tried a few changes before finally landing on this:

Adding:
vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
to uvc_v4l2_mmap.  This resolved the corruption.  gspca_zc3xx has the same
issue, and this patch worked there as well.

I don't know if this is The Right Way to fix it, but hopefully it gives you
a good clue.

Best Regards


      

<<attachment: previous.jpg>>

_______________________________________________
Linux-uvc-devel mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/linux-uvc-devel

Reply via email to