Re: [Qemu-devel] new patches?

[Johannes Martin <jmartin@notamusica.com>]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 912 bytes --]

Hi,

I have attached a patch I wrote that fixes sound problems in OS/2. As far
as I know, it hasn't made it into CVS yet.

	Johannes


On Mon, 19 Jul 2004, Hetz Ben Hamo wrote:

> Hi,
>
> I'm doing some cleaning on my web site tonight and I was wondering if there's
> any patch that I didn't put on my web site yet. If you're writing a patch
> which was not commited yet, please send it to me (hetz@dad-answers.com, not
> tto this email please)
>
> Does anyone have any comment on any patch on my web site which should be
> removed (old, outdated, committed already, etc..)
>
> Thanks,
> Hetz
>
>
> _______________________________________________
> Qemu-devel mailing list
> Qemu-devel@nongnu.org
> http://lists.nongnu.org/mailman/listinfo/qemu-devel
>

---
Mit zwei Euro im Monat helfen:
	http://www.2-euro-helfen.de/
Help to fight hunger and injustice in our world:
	http://www.devp.org/
	http://www.oxfam.org/

[-- Attachment #2: Type: TEXT/PLAIN, Size: 641 bytes --]

*** oss.c       Sun Jul 11 18:02:19 2004
--- oss.c.~1.5.~        Mon Jul  5 23:25:11 2004
***************
*** 410,420 ****
                  return;
              }
              else {
!                 int _errno = errno;
!                 fprintf (stderr, "oss: write audio\n");
!                 fprintf (stderr, "system error: %s\n", strerror (_errno));
!                 // ERRFail ("write audio");
!                 written = play;
              }
          }

--- 410,416 ----
                  return;
              }
              else {
!                 ERRFail ("write audio");
              }
          }

[-- Attachment #3: Type: TEXT/PLAIN, Size: 506 bytes --]

qemu -enable-sound sometimes aborts with a message saying
        oss: write audio
        system error: Unknown error 572
	(the actual error number changes)
For some reason that I haven't understood yet, write() to the dsp device
failes sometimes and qemu aborts on the first failure. It seems to be safe
to ignore these failures and keep going instead.

The problem first occured to me with an OS/2 guest. I have had a report
that it also happened in Windows.

The attached patch fixes this.