Hello community,

here is the log from the commit of package suse-prime for openSUSE:Factory 
checked in at 2019-01-15 09:16:47
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Comparing /work/SRC/openSUSE:Factory/suse-prime (Old)
 and      /work/SRC/openSUSE:Factory/.suse-prime.new.28833 (New)
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Package is "suse-prime"

Tue Jan 15 09:16:47 2019 rev:4 rq:664721 version:0.3

Changes:
--------
--- /work/SRC/openSUSE:Factory/suse-prime/suse-prime.changes    2019-01-11 
14:06:42.947725990 +0100
+++ /work/SRC/openSUSE:Factory/.suse-prime.new.28833/suse-prime.changes 
2019-01-15 09:16:48.758211087 +0100
@@ -1,0 +2,10 @@
+Fri Jan 11 16:45:45 UTC 2019 - [email protected]
+
+- U_prime-select.sh-bail-out-if-nvidia-xconfig-fails-use.patch 
+  * 'nvidia-xconfig --query-gpu-info' may easily fail due to missing
+    NVIDIA kernel modules or alike. Therefore bail out in this case
+    with exit 1.
+  * There could be more than one NVIDIA card/GPU in the system. Use
+    the first detected one for X configuration in that case.
+
+-------------------------------------------------------------------

New:
----
  U_prime-select.sh-bail-out-if-nvidia-xconfig-fails-use.patch

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Other differences:
------------------
++++++ suse-prime.spec ++++++
--- /var/tmp/diff_new_pack.PJkbIt/_old  2019-01-15 09:16:49.190210686 +0100
+++ /var/tmp/diff_new_pack.PJkbIt/_new  2019-01-15 09:16:49.190210686 +0100
@@ -24,6 +24,7 @@
 Group:          System/X11/Utilities
 Url:            https://github.com/michalsrb/SUSEPrime
 Source0:        
https://github.com/michalsrb/SUSEPrime/archive/%{version}.tar.gz#/SUSEPrime-%{version}.tar.gz
+Patch0:         U_prime-select.sh-bail-out-if-nvidia-xconfig-fails-use.patch
 Recommends:     nvidia_driver
 Supplements:    modalias(nvidia_driver:pci:v00008086d*sv*sd*bc03sc*i*)
 Conflicts:      suse-prime-alt
@@ -36,6 +37,7 @@
 
 %prep
 %setup -n SUSEPrime-%{version}
+%patch0 -p1
 
 %build
 :

++++++ U_prime-select.sh-bail-out-if-nvidia-xconfig-fails-use.patch ++++++
>From 8be917026263539230b470e2c05cc46955e618c4 Mon Sep 17 00:00:00 2001
From: Stefan Dirsch <[email protected]>
Date: Fri, 11 Jan 2019 17:35:35 +0100
Subject: [PATCH] prime-select.sh: bail out if nvidia-xconfig fails; use PCI
 BusID of first NVIDIA GPU/card

'nvidia-xconfig --query-gpu-info' may easily fail due to missing NVIDIA
kernel modules or alike. Therefore bail out in this case with exit 1.

There could be more than one NVIDIA card/GPU in the system. Use the
first detected one for X configuration in that case.
---
 prime-select.sh | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/prime-select.sh b/prime-select.sh
index de51c2b..6613328 100755
--- a/prime-select.sh
+++ b/prime-select.sh
@@ -27,7 +27,13 @@ case $type in
       clean_files 
 
       gpu_info=`nvidia-xconfig --query-gpu-info`
-      nvidia_busid=`echo "$gpu_info" |grep -i "PCI BusID"|sed 's/PCI BusID 
://'|sed 's/ //g'`
+      # This may easily fail, if no NVIDIA kernel module is available or alike
+      if [ $? -ne 0 ]; then
+         echo "PCI BusID of NVIDIA card could not be detected!"
+         exit 1
+      fi
+      # There could be more than on NVIDIA card/GPU; use the first one in that 
case
+      nvidia_busid=`echo "$gpu_info" |grep -i "PCI BusID"|head -n 1|sed 's/PCI 
BusID ://'|sed 's/ //g'`
       libglx_nvidia=`update-alternatives --list libglx.so|grep 
nvidia-libglx.so`
 
       update-alternatives --set libglx.so $libglx_nvidia
-- 
2.16.4


Reply via email to