|Summary:||Xephyr gets BadDrawable on x86_64 because hostx.c is built with _XSERVER64|
|Product:||xorg||Reporter:||Aaron Plattner <aplattner>|
|Component:||Server/DDX/Xephyr||Assignee:||Dodji Seketeli <dodji>|
|Status:||RESOLVED FIXED||QA Contact:||Xorg Project Team <xorg-team>|
|Priority:||medium||CC:||brian, hub, martin, mpgritti, pascal|
|i915 platform:||i915 features:|
|Bug Depends on:|
Description Aaron Plattner 2008-05-05 16:42:54 UTC
Xephyr fails to run on a Linux x86_64 system because hostx.c is built with _XSERVER64 defined. It dies with this error: X Error of failed request: BadDrawable (invalid Pixmap or Window parameter) Major opcode of failed request: 55 (X_CreateGC) Resource id in failed request: 0x0 Serial number of failed request: 7 Current serial number in output stream: 12 This happens because RootWindow(HostX.dpy, HostX.screen) is returning None. This broke with this commit: commit 6a435b00003fb00930299a0e0810c93afc23a72e Author: Dodji Seketeli <email@example.com> Date: Wed Sep 19 15:35:51 2007 +0200 Xephyr: check presence of extensions in host X * hw/kdrive/ephyr/hostx.c,h: (hostx_has_xshape), (hostx_has_glx), (hostx_has_dri): added these new entry points * hw/kdrive/ephyr/ephyrdriext.c: (ephyrDRIExtensionInit): check presence of DRI and XShape extensions before trying to use them. * hw/kdrive/ephyr/ephyrglxext.c: (ephyrHijackGLXExtension): check presence of glx extension before we use it. which causes that file to include kdrive-config.h, which in turn includes dix-config.h, which defines _XSERVER64.
Comment 1 Martin Dengler 2008-06-09 22:08:32 UTC
I seem to have come across this in Fedora 9.
Comment 2 Dodji Seketeli 2008-06-20 07:35:29 UTC
Okay, if you don't mind, I will look at this.
Comment 3 Martin Dengler 2008-06-29 07:33:12 UTC
Thanks for looking at this!
Comment 4 Pascal Scheffers 2008-07-15 04:49:53 UTC
I'd love to see this fixed. It prevents me from running a working OLPC emulator on my Fedora 9 laptop
Comment 5 Julien Cristau 2008-07-22 12:54:44 UTC
Dodji, any progress on fixing this, or should we just revert the bad commit for 1.5?
Comment 6 Dodji Seketeli 2008-07-22 13:21:44 UTC
I have just got myself an x86_64 box to test this. Before that I didn't have such an environment. I will be able to test the problem and hopefully come to a solution. I am very sorry for the delay.
Comment 7 Dodji Seketeli 2008-07-22 16:13:32 UTC
So have committed http://gitweb.freedesktop.org/?p=xorg/xserver.git;a=commit;h=5de1867fbb0a336ff3fdc92cbf734849f6111b1b that seems to fix it for me. I have also committed a patch to fix various build issues on X86_64. I hope it works for you.
Comment 8 Hubert Figuiere 2008-07-22 16:54:42 UTC
I applied the patches and now it works here on x86_64 Thanks a lot.
Comment 9 Martin Dengler 2008-08-29 13:34:23 UTC
Just wanted to say thanks - I've been using the rpms at http://koji.fedoraproject.org/koji/buildinfo?buildID=57371 with this patch and it's great.