Trying to use XV on the modesetting branch of the intel driver on an 855GME causes X to lockup. Running mplayer -vo xv yields the following in the Xorg.0.log: loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 loops (1) maxed out for buffer 1 (II) intel(0): [drm] removed 1 reserved context for kernel (II) intel(0): [drm] unmapping 8192 bytes of SAREA 0xdfbb8000 at 0xb7b42000 Error in I830WaitLpRing(), now is 280912022, start is 280910021 pgetbl_ctl: 0x1ffe0001 pgetbl_err: 0x0 ipeir: 0 iphdr: 1810000 LP ring tail: 870 head: 2ec len: 1f001 start 0 eir: 0 esr: 0 emr: ffff instdone: ffc0 instpm: 0 memmode: 108 instps: 0 hwstam: ffff ier: 0 imr: ffff iir: 0 space: 129652 wanted 131064 Fatal server error: lockup Backtrace: 0: X(xf86SigHandler+0x80) [0x80c0dd0] 1: [0xffffe420] 2: X(AbortDDX+0x87) [0x80a2777] 3: X(AbortServer+0x28) [0x81b8208] 4: X(FatalError+0x67) [0x81b8727] 5: /usr/local/src/xorg-modular/lib/xorg/modules//drivers/intel_drv.so(I830WaitLpRing+0x117) [0xb7b7e717] 6: /usr/local/src/xorg-modular/lib/xorg/modules//drivers/intel_drv.so(I830Sync+0xf9) [0xb7b7e949] 7: /usr/local/src/xorg-modular/lib/xorg/modules//drivers/intel_drv.so [0xb7b85d5c] 8: /usr/local/src/xorg-modular/lib/xorg/modules//drivers/intel_drv.so [0xb7b85e90] 9: /usr/local/src/xorg-modular/lib/xorg/modules//drivers/intel_drv.so [0xb7b86231] 10: /usr/local/src/xorg-modular/lib/xorg/modules//libshadow.so [0xb7b4546e] 11: X [0x815ab1e] 12: X [0x80c556f] 13: X [0x8134396] 14: X [0x8158365] 15: X(main+0x514) [0x806fe84] 16: /lib/tls/libc.so.6(__libc_start_main+0xc8) [0xb7d27ea8] 17: X(FontFileCompleteXLFD+0xad) [0x806f131] FatalError re-entered, aborting Caught signal 11. Server aborting
Created attachment 8951 [details] log showing XV lockup See also bug #10157 and http://lists.freedesktop.org/archives/xorg/2007-March/022229.html
Created attachment 8952 [details] xorg.conf used to generate the attached log
I should clarify that XV also failed to work in that no video actually made it as far as the display - instead only solid colour appeared. (I think it was pink overlay, but it may have been 'blank' YUV green - I'll need to check again).
Fixed a bug today that could easily cause serious problems like this. Please update your xserver to current master and try again.
I'm experiencing an XVideo crash with intel 2.0rc2 (1.9.92) on xserver 1.3rc2 (1.2.99.902), running on Intel 82852/855GM. It sounds similar to the one in this bug. Running a video using the XVideo extension (selected for example in vlc), the window displays only blue, and then after a few seconds the system crashes hard. When I change vlc to use "X11 video output" instead, then video plays okay. Eric, the only commit I could see of yours in master near 6 March was 024bbc7cbb924daaf3e305ddfc8e74509acd1e15 "Bug #9931: Fix linear allocations with a non-1-byte granularity." It doesn't seem to apply to the 1.3 branch.
Created attachment 9235 [details] XVideo crash log on intel 2.0rc2, xserver 1.3rc2
Could bug #10645 be related? I'll be marking it as depending on, as from where I'm standing (and from what I understand of the driver), what's actually happening and why it's crashing is that the driver is using only *one* pipe (pipe A) for XV, while the overlay is really in pipe B.
Hi, to those on this bug: could you try the patches attached to #10645 and see if that fixes it for you? (BTW: can I mark this as a duplicate of #10645?)
*** This bug has been marked as a duplicate of bug 10645 ***
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.