Re: [Spice-devel] spice libraries for windows client

Hi, On Sun, Apr 15, 2012 at 12:24:38PM +0300, Arnon Gilboa wrote: will be nice to keep latest x86 x64 wspice zips on the new dl page. otherwise, no way to run the legacy spicec on windows. These would be http://spice-space.org/download/stable/wspice-x86_20110308.zip and

Re: [Spice-devel] spice libraries for windows client

Christophe Fergeau wrote: Hi, On Sun, Apr 15, 2012 at 12:24:38PM +0300, Arnon Gilboa wrote: will be nice to keep latest x86 x64 wspice zips on the new dl page. otherwise, no way to run the legacy spicec on windows. These would be

Re: [Spice-devel] qxl: add optinal 64bit vram bar patch significantly slows down Spice

On 04/06/12 15:54, Hans de Goede wrote: Hi Gerd, While cherry picking some spice patches into my own qemu-kvm-1.0 branch, so that we can add them to Fedora-17 I noticed a significant slowdown after I was done cherry picking. Investigation has shown your qxl: add optinal 64bit vram bar patch

Re: [Spice-devel] qxl: add optinal 64bit vram bar patch significantly slows down Spice

On 04/06/12 15:54, Hans de Goede wrote: Hi Gerd, While cherry picking some spice patches into my own qemu-kvm-1.0 branch, so that we can add them to Fedora-17 I noticed a significant slowdown after I was done cherry picking. Investigation has shown your qxl: add optinal 64bit vram bar patch

Re: [Spice-devel] qxl: add optinal 64bit vram bar patch significantly slows down Spice

Hi, On 04/16/2012 11:39 AM, Gerd Hoffmann wrote: On 04/06/12 15:54, Hans de Goede wrote: Hi Gerd, While cherry picking some spice patches into my own qemu-kvm-1.0 branch, so that we can add them to Fedora-17 I noticed a significant slowdown after I was done cherry picking. Investigation has

Re: [Spice-devel] qxl: add optinal 64bit vram bar patch significantly slows down Spice

On 04/16/2012 12:39 PM, Gerd Hoffmann wrote: On 04/06/12 15:54, Hans de Goede wrote: Hi Gerd, While cherry picking some spice patches into my own qemu-kvm-1.0 branch, so that we can add them to Fedora-17 I noticed a significant slowdown after I was done cherry picking. Investigation

Re: [Spice-devel] qxl: add optinal 64bit vram bar patch significantly slows down Spice

Hi, On 04/16/2012 11:59 AM, Avi Kivity wrote: On 04/16/2012 12:39 PM, Gerd Hoffmann wrote: On 04/06/12 15:54, Hans de Goede wrote: Hi Gerd, While cherry picking some spice patches into my own qemu-kvm-1.0 branch, so that we can add them to Fedora-17 I noticed a significant slowdown after I

Re: [Spice-devel] qxl: add optinal 64bit vram bar patch significantly slows down Spice

On 04/16/2012 03:02 PM, Hans de Goede wrote: Could it be that the patch somehow causes seabios and/or Xorg to see the region as different wrt being cachable / uncachable or some such? In general we ignore cacheability in kvm. Most likely a but in the memory core caused the region to be

Re: [Spice-devel] qxl: add optinal 64bit vram bar patch significantly slows down Spice

On 04/16/2012 12:39 PM, Gerd Hoffmann wrote: On 04/06/12 15:54, Hans de Goede wrote: Hi Gerd, While cherry picking some spice patches into my own qemu-kvm-1.0 branch, so that we can add them to Fedora-17 I noticed a significant slowdown after I was done cherry picking. Investigation

Re: [Spice-devel] qxl: add optinal 64bit vram bar patch significantly slows down Spice

Hi, Looks like both the BAR and it's backing region were initialized to 4k. The number 4096 does show up in the patch, probably it sneaked into the BAR somehow. Yea, one page is the minimum size, that shouldn't have been the default though. Alon has a fix for it (attached), still in the

Re: [Spice-devel] qxl: add optinal 64bit vram bar patch significantly slows down Spice

On 04/16/2012 03:43 PM, Gerd Hoffmann wrote: Hi, Looks like both the BAR and it's backing region were initialized to 4k. The number 4096 does show up in the patch, probably it sneaked into the BAR somehow. Yea, one page is the minimum size, that shouldn't have been the default though.

Re: [Spice-devel] qxl: add optinal 64bit vram bar patch significantly slows down Spice

Hi, On 04/16/2012 02:43 PM, Gerd Hoffmann wrote: Hi, Looks like both the BAR and it's backing region were initialized to 4k. The number 4096 does show up in the patch, probably it sneaked into the BAR somehow. Yea, one page is the minimum size, that shouldn't have been the default

[Spice-devel] [PATCH spice-common] Add a return value to the default case in create_bitmap to satisfy VC++

--- common/gdi_canvas.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/common/gdi_canvas.c b/common/gdi_canvas.c index 4c5daea..0a04bf8 100644 --- a/common/gdi_canvas.c +++ b/common/gdi_canvas.c @@ -564,7 +564,7 @@ static uint8_t *create_bitmap(HBITMAP *bitmap,

[Spice-devel] [PATCH spice] update windows client generate.bat and generate1.bat

the python script is now in spice-common and client_marshallers.h needs to be included instead of marshallers.h --- client/windows/generate.bat |4 ++-- client/windows/generate1.bat |4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/client/windows/generate.bat

Re: [Spice-devel] [PATCH spice 00/12] Add more compiler warnings

Hi Daniel, Did you forgot about your patches? Should I take care of doing the small change in 6/12 and push for you? cheers -- Marc-André Lureau ___ Spice-devel mailing list Spice-devel@lists.freedesktop.org

[Spice-devel] [PATCH spice] Copy getopt.h from old common/win/my_getopt-1.5/my_getopt

This patch completes the copy from 4d8f39020ac83602c1647d4af04e8b19bf74ed6e which missed this file. --- client/windows/getopt.h | 56 +++ 1 files changed, 56 insertions(+), 0 deletions(-) create mode 100644 client/windows/getopt.h diff --git

[Spice-devel] Introduction and Hello

Hello, My name is Aric Stewart. Primarily a WINE hacker, due to interest from a client I am starting work on spice. I will mostly be focusing on the win32 client to begin with but we will see where work progresses. I am happy to be helping out here and hope to be able to be productive.