On Monday, March 24, 2014 2:23 PM, Fred <heitk...@ameritech.net> wrote: On 03/15/14 04:34, stompdagg...@yahoo.com wrote:
here is a simple test case that reproduces the issue on my machine: >dagg@NCC-5001-D ~/workspace $ cat open_example.c >#include <unistd.h> >#include <fcntl.h> >#include <stdio.h> >#include <errno.h> >#include <string.h> > >int main() >{ > char *param1 = "ok"; > char *param2 = "."; > int filedesc = open("/dev/dri/card1", O_RDWR, 0); > if (filedesc) { > param1 = "ERROR: "; > param2 = strerror(errno); > } > printf("fd = %d, status = %s%s\n", filedesc, param1, param2); > close(filedesc); > > return 0; >}dagg@NCC-5001-D ~/workspace $ gcc open_example.c -o open_example && ./open_example >fd = -1, status = ERROR: Invalid argument > > > > >On Friday, March 14, 2014 4:22 PM, "stompdagg...@yahoo.com" ><stompdagg...@yahoo.com> wrote: > >On Tuesday, March 11, 2014 8:45 AM, "stompdagg...@yahoo.com" ><stompdagg...@yahoo.com> wrote: > >>On Tuesday, March 11, 2014 5:16 AM, Michel Dänzer <mic...@daenzer.net> wrote: > >>On Mon, 2014-03-10 at 08:23 -0700, stompdagg...@yahoo.com wrote: >> >>> >>> >>> >On Monday, March 10, 2014 5:58 AM, Michel Dänzer <mic...@daenzer.net> >>> wrote: >>> >>> >On Sam, 2014-03-08 at 11:41 -0800, stompdagg...@yahoo.com wrote: >>> > >>> >> >On Sat, Mar 8, 2014 at 2:56 AM, stompdagg...@yahoo.com >>> >> > >>> >> >> [ 51.829] (EE) RADEON(0): [drm] Failed to open DRM device for >>> >> >> pci:0000:06:00.0: No such file or directory >>> > >>> >'No such file or directory' sounds like something is deleting >>> >your /dev/dri/card* files, or not creating them in the first place. >>> >Maybe check your udev setup. >>> > >>> >If it only happens with a 3.13 kernel but not with a 3.12 one, >>> another >>> >option is to bisect the kernel. >.> >.> Greetings Michel, >.> >.> dri node do exists, see: >>> dagg@NCC-5001-D ~ $ ll /dev/dri >>> total 0 >>> crw-rw----+ 1 root video 226, 0 Mar 10 09:27 card0 >>> crw-rw----+ 1 root video 226, 1 Mar 10 09:27 card1 >>> crw-rw----+ 1 root video 226, 2 Mar 10 09:27 card2 >>> crw-rw---- 1 root video 226, 64 Mar 10 09:27 controlD64 >>> crw-rw---- 1 root video 226, 65 Mar 10 09:27 controlD65 >>> crw-rw---- 1 root video 226, 66 Mar 10 09:27 controlD66 >>> >>> I think that the way to find the cause on my part is to check the >>> origin of that msg in Xorg.log. >>> >>> is that a xserver msg or xf86-video-ati msg? > >> >>The message is from the radeon driver, but it's because drmOpen() in >>libdrm fails. It would indeed be interesting to track down where the 'No >>such file or directory' error originates from. > >will do, when I'll have more info, I'll update. > >Greetings Michel, > >I've tried to track the source if that error, I've enabled dbg prints in libdrm and found this: >(II) [KMS] Kernel modesetting enabled. >drmOpenDevice: node name is /dev/dri/card0 >drmOpenDevice: open result is 11, (OK) >drmOpenByBusid: Searching for BusID pci:0000:06:00.0 >drmOpenDevice: node name is /dev/dri/card0 >drmOpenDevice: open result is 11, (OK) >drmOpenByBusid: drmOpenMinor returns 11 >drmOpenByBusid: drmGetBusid reports pci:0000:00:02.0 >drmOpenDevice: node name is /dev/dri/card1 >drmOpenDevice: open result is -1, (Invalid argument) >drmOpenByBusid: drmOpenMinor returns -22 >drmOpenDevice: node name is /dev/dri/card2 >drmOpenDevice: open result is -1, (Invalid argument) > >looking in the code I see this: > fd = open(buf, O_RDWR, 0); > drmMsg("drmOpenDevice: open result is %d, (%s)\n", > fd, fd < 0 ? strerror(errno) : "OK"); > if (fd >= 0) > return fd; > >I think that the issue might be that the file cannot be opened for that mode as the path is printed above and it is ok. > >any ideas what might it be? > > I had a similar problem over the last weeks using kernel 3.13.x Not sure it happened before with older kernels. I am using a "from scratch" distro. It's using systemd and glibc-2.19, with relatively recent Xorg and Mesa development. I have a Radeon 4870. I don't think it's the video card. Tentatively what happens is I lose the /dev/dri/card0 and the display goes black. I get "no usable screens found" in the Xorg.0.log file. Anyway what seems to be causing the problem is If I leave a music CD in the CD drive, upon booting, timing issues or something cause the dri/card0 device to go away. There are a lot of messages in the kernel log, where it's trying to find out about the music CD. X seems to come up fine with no music CD in the drive. Fred _______________________________________________ >xorg@lists.x.org: X.Org support >Archives: http://lists.freedesktop.org/archives/xorg >Info: http://lists.x.org/mailman/listinfo/xorg >Your subscription address: %(user_address)s > > > > > > >_______________________________________________ xorg@lists.x.org: X.Org support Archives: http://lists.freedesktop.org/archives/xorg Info: http://lists.x.org/mailman/listinfo/xorg Your subscription address: %(user_address)s Greetings Fred, not sure how the cd is relevant for the issue but following patches fixes the issue on my setup: http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=9babd35ad72af631547c7ca294bc2e931cc40e58 http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=7848865914c6a63ead674f0f5604b77df7d3874f
_______________________________________________ xorg@lists.x.org: X.Org support Archives: http://lists.freedesktop.org/archives/xorg Info: http://lists.x.org/mailman/listinfo/xorg Your subscription address: %(user_address)s