qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Qemu-devel] Patches from pkgsrc for 0.9.1


From: Rumko
Subject: [Qemu-devel] Patches from pkgsrc for 0.9.1
Date: Tue, 15 Jan 2008 19:15:11 +0100
User-agent: KNode/0.10.5

Some patches are present in pkgsrc that should be sent upstream, so I have
attached the patches and some information about the patches below.

patch-aa: some machines have <machine/bswap.h> instead of <byteswap.h>

patch-ab: qemu (op.c) shouldn't be compiled with the stack-protector turned on,
so this patch adds the needed -fno-stack-protector if it is used by default

patch-ac: on NetBSD some operations are not supported

patch-ad: on DragonFly <sys/disk.h> must not be included by userland programs

patch-ae: I was unable to compile slavio_serial.c using gcc 3.4.6 because of
some inlining errors, but just reordering the functions makes it compile
cleanly

patch-ag: DragonFly support is added, a check for <machine/bswap.h> and the
stack protector are added as well

patch-am: I'm not quite sure what this is for ... but this patch has existed
for a bit more than 3 years ... maybe ioctl needs 3 args on some systems?

patch-ao: as the comment added by the patch says, this is a work around some
ugly #defines in NetBSD 2.x

patch-au: defines some possibly missing functionality on NetBSD

patch-av: quoted ... "Provide trunc() implementation for NetBSD 3 and
Dragonfly"

patch-aw: quoted ... "Workaround for x86_64 target, patch from Hubert Feyrer."

patch-ax: quoted from cvs revision history ... "Apply patches from Tobyas
Nigren (PR pkg/32464) to make this build on NetBSD/amd64. Thank you Tobias!"

patch-ay: quoted from cvs revision history ... "Update to 0.9.0. Thanks to
Michael Van Elst for sending me the missing patch to let this build on systems
without POSIX AIO."

patch-ba: quoted from cvs revision history ... "Add patches/patch-ba:
  Emulate the speaker port's refresh clock bit.  This is supposed
  to toggle between 0 and 1<<4 every 15 microseconds.  XXX: We use
  gettimeofday() in the real machine instead of a monotonic clock
  in the virtual machine, and we are a bit sloppy about the 15
  microseconds.  This should be good enough for crude loops that
  measure approximate delays by counting how often this line toggles."

Attachment: patch-aa
Description: Text Data

Attachment: patch-ab
Description: Text Data

Attachment: patch-ac
Description: Text Data

Attachment: patch-ad
Description: Text Data

Attachment: patch-ae
Description: Text Data

Attachment: patch-ag
Description: Text Data

Attachment: patch-am
Description: Text Data

Attachment: patch-ao
Description: Text Data

Attachment: patch-au
Description: Text Data

Attachment: patch-av
Description: Text Data

Attachment: patch-aw
Description: Text Data

Attachment: patch-ax
Description: Text Data

Attachment: patch-ay
Description: Text Data

Attachment: patch-ba
Description: Text Data


reply via email to

[Prev in Thread] Current Thread [Next in Thread]