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

Bug#639617: marked as done (libgl1-mesa-glx: libGL should give some indication when DRI2 fails)

Your message dated Wed, 11 Apr 2012 15:26:04 +0200 (CEST)
with message-id <8625015.15595941334150764157.JavaMail.defaultUser@defaultHost>
and subject line 
has caused the Debian Bug report #639617,
regarding libgl1-mesa-glx: libGL should give some indication when DRI2 fails
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)


-- 
639617: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=639617
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message --- 
Package: libgl1-mesa-glx
Version: 7.11-4
Severity: minor
Tags: patch

Recently I was trying to figure out why 3d was getting rendered by
software on my radeon X1300. The reason was that the packaged
r300_dri.so is a gallium driver and appears to be DRI2-only, but I had
DRI2 disabled in X (deliberately, for other reasons). This was very
non-obvious, since libGL gives no indication that anything is going on
with DRI2:

$ LIBGL_DEBUG=verbose glxinfo -b
libGL: OpenDriver: trying /usr/lib/i386-linux-gnu/dri/tls/r300_dri.so
libGL: OpenDriver: trying /usr/lib/i386-linux-gnu/dri/r300_dri.so
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is 4, (OK)
drmOpenByBusid: Searching for BusID pci:0000:01:00.0
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is 4, (OK)
drmOpenByBusid: drmOpenMinor returns 4
drmOpenByBusid: Interface 1.4 failed, trying 1.1
drmOpenByBusid: drmGetBusid reports pci:0000:01:00.0
libGL error: Calling driver entry point failedlibGL error: reverting to software direct rendering
libGL: OpenDriver: trying /usr/lib/i386-linux-gnu/dri/tls/swrast_dri.so
libGL: OpenDriver: trying /usr/lib/i386-linux-gnu/dri/swrast_dri.so 33

Arguably the "real" problem here is that DRI2 is disabled. So, after it
tried DRI2, libGL tried to use r300_dri.so with DRI, which doesn't work
(which is that 'Calling driver entry point failed' error). I spent a bit
of time trying to debug this, and it would have been easier if the debug
messages said a little more about what was happening.

Attached is a patch (which modifies libGL and the *_dri.so drivers) so
instead I get these messages:

$ LIBGL_DEBUG=verbose /usr/bin/glxinfo -b
libGL: screen 0 does not appear to be DRI2 capable
libGL: XF86DRIGetClientDriverName: 5.3.0 r300 (screen 0)
libGL: OpenDriver: trying /usr/lib/i386-linux-gnu/dri/tls/r300_dri.so
libGL: OpenDriver: trying /usr/lib/i386-linux-gnu/dri/r300_dri.so
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is 4, (OK)
drmOpenByBusid: Searching for BusID pci:0000:01:00.0
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is 4, (OK)
drmOpenByBusid: drmOpenMinor returns 4
drmOpenByBusid: Interface 1.4 failed, trying 1.1
drmOpenByBusid: drmGetBusid reports pci:0000:01:00.0
libGL: driver does not support DRI1
libGL error: Calling driver entry point failedlibGL error: reverting to software direct rendering
libGL: OpenDriver: trying /usr/lib/i386-linux-gnu/dri/tls/swrast_dri.so
libGL: OpenDriver: trying /usr/lib/i386-linux-gnu/dri/swrast_dri.so 33

which to me seems a bit more clear.

Technically this affects both libgl1-mesa-glx and libgl1-mesa-dri, but
this is such a little thing, I didn't want to clutter both packages with
a bug report.

-- 
Andrew Deason
adeason@dson.org

>From dd7e72d7883c63884f37020c38f89fa4e8ecc45c Mon Sep 17 00:00:00 2001
From: Andrew Deason <adeason@dson.org>
Date: Sun, 28 Aug 2011 11:46:02 -0500
Subject: [PATCH 1/2] GLX: Say something if we cannot connect via DRI2

---
 src/glx/dri2_glx.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/src/glx/dri2_glx.c b/src/glx/dri2_glx.c
index 80e4da3..1adea98 100644
--- a/src/glx/dri2_glx.c
+++ b/src/glx/dri2_glx.c
@@ -832,6 +832,7 @@ dri2CreateScreen(int screen, struct glx_display * priv)
 		    &driverName, &deviceName)) {
       glx_screen_cleanup(&psc->base);
       XFree(psc);
+      InfoMessageF("screen %d does not appear to be DRI2 capable\n", screen);
       return NULL;
    }
 
-- 
1.7.5.4


>From 12aeb060d1d924ba9b9da8cda1210b96f52ff234 Mon Sep 17 00:00:00 2001
From: Andrew Deason <adeason@dson.org>
Date: Sun, 28 Aug 2011 12:14:46 -0500
Subject: [PATCH 2/2] DRI: Log something if we don't support legacy DRI

If we are called via the legacy DRI interface, and we don't support
legacy DRI (InitScreen is NULL), print a debug message, so it is easy
to see why the driver fails to initialize.
---
 src/mesa/drivers/dri/common/dri_util.c |    4 +++-
 1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/src/mesa/drivers/dri/common/dri_util.c b/src/mesa/drivers/dri/common/dri_util.c
index 82638fa..725549b 100644
--- a/src/mesa/drivers/dri/common/dri_util.c
+++ b/src/mesa/drivers/dri/common/dri_util.c
@@ -806,8 +806,10 @@ driCreateNewScreen(int scrn,
     static const __DRIextension *emptyExtensionList[] = { NULL };
     __DRIscreen *psp;
 
-    if (driDriverAPI.InitScreen == NULL)
+    if (driDriverAPI.InitScreen == NULL) {
+	__driUtilMessage("driver does not support DRI1");
 	return NULL;
+    }
 
     psp = calloc(1, sizeof *psp);
     if (!psp)
-- 
1.7.5.4

--- End Message ---
--- Begin Message --- 
Version: 8.0-1

--- End Message ---
Reply to: