i915 lockup / extreme delay
Subsystems (2)
I have been experiencing a very sluggish X server right after starting my laptop. The effect goes away after a few minutes, so it was rather hard to track down.
Recently after updating WINE to a newer release, I've been able to trigger the effect by running a certain game under wine on a remote computer and using VirtualGL to transport the GL display to my laptop. As I first thought the issue was somewhere with wine/virtualgl, I posted the question to the virtualgl mailing list. The thread is here:
http://thread.gmane.org/gmane.comp.video.opengl.virtualgl.user/170
But after some further investigation, it seems the issue is with the i915 driver.
The 'effect' is that only the mouse pointer works in the X server. The cpu usage on the laptop during the sluggishness is minimal. When I suspend the game with winedbg, the X server slowly becomes responsive again.
The output from latencytop seems to point to i915 being the culprit:
Cause Maximum Percentage
[i915_gem_sw_finish_ioctl] 3880.2 msec 51.0 % Throttling GPU while waiting for commands 1337.6 msec 12.1 % [i915_gem_set_domain_ioctl] 739.6 msec 10.3 % mmaping memory 275.0 msec 0.9 % Scheduler: waiting for cpu 106.1 msec 3.8 % Executing raw SCSI command 105.4 msec 0.8 % fsync() on a file (type 'F' for details) 84.5 msec 0.5 % [i915_gem_do_execbuffer] 19.7 msec 0.4 % [i915_gem_pwrite_ioctl] 19.7 msec 0.4 %
--
Cause Maximum Percentage
[i915_gem_sw_finish_ioctl] 4181.0 msec 57.5 % [i915_gem_set_domain_ioctl] 3395.2 msec 35.0 % Throttling GPU while waiting for commands 742.0 msec 2.7 % [i915_gem_create_ioctl] 19.7 msec 0.1 % [i915_gem_busy_ioctl] 19.7 msec 0.4 % [i915_gem_pwrite_ioctl] 19.6 msec 0.3 % [i915_gem_madvise_ioctl] 19.6 msec 1.4 % [i915_gem_do_execbuffer] 19.4 msec 0.1 % [i915_do_wait_request] 5.0 msec 0.2 %
--
Cause Maximum Percentage
[i915_gem_set_domain_ioctl] 4609.5 msec 32.8 % [i915_gem_sw_finish_ioctl] 3316.7 msec 52.0 % [i915_gem_create_ioctl] 744.1 msec 3.0 % synchronous write 345.9 msec 1.3 % Executing a program 109.7 msec 0.4 % fsync() on a file (type 'F' for details) 87.6 msec 1.0 % Page fault 86.1 msec 0.8 % Writing a page to disk 20.1 msec 0.4 % [i915_gem_madvise_ioctl] 19.7 msec 3.4 %
These were taken on a Fedora 12 kernel v2.6.32.9-70
I also tried to reproduce it with the latest kernel v2.6.34-rc2, but it's much worse with that one, since it causes the laptop to completely lockup when I start the game!
I've tried to enable some extra kernel debug options and also setup netconsole to see if there is anything reported before the lockup, but unfortunately there's no output.
--
Some hardware details:
00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07) (prog-if 00 [VGA controller])
Subsystem: Dell Device 024f
Flags: bus master, fast devsel, latency 0, IRQ 31
Memory at f6c00000 (64-bit, non-prefetchable) [size=4M]
Memory at e0000000 (64-bit, prefetchable) [size=256M]
I/O ports at ef98 [size=8]
Expansion ROM at <unassigned> [disabled]
Capabilities: <access denied>
Kernel driver in use: i915
Kernel modules: i915
00:02.1 Display controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07)
Subsystem: Dell Device 024f
Flags: bus master, fast devsel, latency 0
Memory at f6b00000 (64-bit, non-prefetchable) [size=1M]
Capabilities: <access denied>
X.Org X Server 1.7.5.902 (1.7.6 RC 2)
Release Date: 2010-03-12
X Protocol Version 11, Revision 0
(--) PCI:*(0:0:2:0) 8086:2a42:1028:024f Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller rev 7, Mem @ 0xf6c00000/4194304, 0xe0000000/268435456, I/O @ 0x0000ef98/8, BIOS @ 0x????????/131072 (--) PCI: (0:0:2:1) 8086:2a43:1028:024f Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller rev 7, Mem @ 0xf6b00000/1048576, BIOS @ 0x????????/65536
(II) LoadModule: "intel"
(II) Loading /usr/lib64/xorg/modules/drivers/intel_drv.so (II) Module intel: vendor="X.Org Foundation"
compiled for 1.7.0, module version = 2.9.1 Module class: X.Org Video Driver
ABI class: X.Org Video Driver, version 6.0
(II) intel: Driver for Intel Integrated Graphics Chipsets: i810,
i810-dc100, i810e, i815, i830M, 845G, 852GM/855GM, 865G, 915G,
E7221 (i915), 915GM, 945G, 945GM, 945GME, Pineview GM, Pineview G,
965G, G35, 965Q, 946GZ, 965GM, 965GME/GLE, G33, Q35, Q33, GM45,
4 Series, G45/G43, Q45/Q43, G41, B43, Clarkdale, Arrandale(II) Primary Device is: PCI 00@00:02:0
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is 9, (OK)
drmOpenByBusid: Searching for BusID pci:0000:00:02.0
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is 9, (OK)
drmOpenByBusid: drmOpenMinor returns 9
drmOpenByBusid: drmGetBusid reports pci:0000:00:02.0
(**) intel(0): Depth 24, (--) framebuffer bpp 32
(==) intel(0): RGB weight 888
(==) intel(0): Default visual is TrueColor
(II) intel(0): Integrated Graphics Chipset: Intel(R) GM45
(--) intel(0): Chipset: "GM45"
(II) intel(0): Output LVDS1 has no monitor section
(II) intel(0): found backlight control interface /sys/class/backlight/acpi_video0 (II) intel(0): Output VGA1 has no monitor section
(II) intel(0): Output HDMI1 has no monitor section
(II) intel(0): Output DP1 has no monitor section
(II) intel(0): Output HDMI2 has no monitor section
(II) intel(0): Output DP2 has no monitor section
(II) intel(0): Output DP3 has no monitor section
(II) intel(0): Output TV1 has no monitor section
(II) intel(0): EDID for output LVDS1
(II) intel(0): Manufacturer: SEC Model: 5443 Serial#: 0
(II) intel(0): Year: 2008 Week: 0
(II) intel(0): EDID Version: 1.3
(II) intel(0): Digital Display Input
(II) intel(0): Max Image Size [cm]: horiz.: 33 vert.: 21
(II) intel(0): Gamma: 2.20
(II) intel(0): No DPMS capabilities specified
(II) intel(0): Supported color encodings: RGB 4:4:4 YCrCb 4:4:4
(II) intel(0): First detailed timing is preferred mode
(II) intel(0): redX: 0.580 redY: 0.340 greenX: 0.310 greenY: 0.550
(II) intel(0): blueX: 0.155 blueY: 0.155 whiteX: 0.313 whiteY: 0.329
(II) intel(0): Manufacturer's mask: 0
(II) intel(0): Supported detailed timing:
(II) intel(0): clock: 164.2 MHz Image Size: 331 x 207 mm
(II) intel(0): h_active: 1920 h_sync: 2020 h_sync_end 2052 h_blank_end 2216 h_border: 0 (II) intel(0): v_active: 1200 v_sync: 1202 v_sync_end 1208 v_blanking: 1235 v_border: 0 (II) intel(0): Supported detailed timing: (II) intel(0): clock: 109.5 MHz Image Size: 331 x 207 mm (II) intel(0): h_active: 1920 h_sync: 2020 h_sync_end 2052 h_blank_end 2216 h_border: 0 (II) intel(0): v_active: 1200 v_sync: 1202 v_sync_end 1208 v_blanking: 1235 v_border: 0
(II) intel(0): RX392<80>154CT
(II) intel(0):
[Adding CCs]
Quoted 166 lines · Karl Vogel
I have been experiencing a very sluggish X server right after starting my laptop. The effect goes away after a few minutes, so it was rather hard to track down.
Recently after updating WINE to a newer release, I've been able to trigger the effect by running a certain game under wine on a remote computer and using VirtualGL to transport the GL display to my laptop. As I first thought the issue was somewhere with wine/virtualgl, I posted the question to the virtualgl mailing list. The thread is here:
http://thread.gmane.org/gmane.comp.video.opengl.virtualgl.user/170
But after some further investigation, it seems the issue is with the i915 driver.
The 'effect' is that only the mouse pointer works in the X server. The cpu usage on the laptop during the sluggishness is minimal. When I suspend the game with winedbg, the X server slowly becomes responsive again.
The output from latencytop seems to point to i915 being the culprit:
Cause Maximum Percentage
[i915_gem_sw_finish_ioctl] 3880.2 msec 51.0 % Throttling GPU while waiting for commands 1337.6 msec 12.1 % [i915_gem_set_domain_ioctl] 739.6 msec 10.3 % mmaping memory 275.0 msec 0.9 % Scheduler: waiting for cpu 106.1 msec 3.8 % Executing raw SCSI command 105.4 msec 0.8 % fsync() on a file (type 'F' for details) 84.5 msec 0.5 % [i915_gem_do_execbuffer] 19.7 msec 0.4 % [i915_gem_pwrite_ioctl] 19.7 msec 0.4 %--
Cause Maximum Percentage[i915_gem_sw_finish_ioctl] 4181.0 msec 57.5 % [i915_gem_set_domain_ioctl] 3395.2 msec 35.0 % Throttling GPU while waiting for commands 742.0 msec 2.7 % [i915_gem_create_ioctl] 19.7 msec 0.1 % [i915_gem_busy_ioctl] 19.7 msec 0.4 % [i915_gem_pwrite_ioctl] 19.6 msec 0.3 % [i915_gem_madvise_ioctl] 19.6 msec 1.4 % [i915_gem_do_execbuffer] 19.4 msec 0.1 % [i915_do_wait_request] 5.0 msec 0.2 %--
Cause Maximum Percentage[i915_gem_set_domain_ioctl] 4609.5 msec 32.8 % [i915_gem_sw_finish_ioctl] 3316.7 msec 52.0 % [i915_gem_create_ioctl] 744.1 msec 3.0 % synchronous write 345.9 msec 1.3 % Executing a program 109.7 msec 0.4 % fsync() on a file (type 'F' for details) 87.6 msec 1.0 % Page fault 86.1 msec 0.8 % Writing a page to disk 20.1 msec 0.4 % [i915_gem_madvise_ioctl] 19.7 msec 3.4 %These were taken on a Fedora 12 kernel v2.6.32.9-70
I also tried to reproduce it with the latest kernel v2.6.34-rc2, but it's much worse with that one, since it causes the laptop to completely lockup when I start the game!I've tried to enable some extra kernel debug options and also setup netconsole to see if there is anything reported before the lockup, but unfortunately there's no output.
--
Some hardware details:00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07) (prog-if 00 [VGA controller])
Subsystem: Dell Device 024f
Flags: bus master, fast devsel, latency 0, IRQ 31 Memory atf6c00000(64-bit, non-prefetchable) [size=4M] Memory ate0000000(64-bit, prefetchable) [size=256M] I/O ports at ef98 [size=8]
Expansion ROM at <unassigned> [disabled]
Capabilities: <access denied>
Kernel driver in use: i915
Kernel modules: i91500:02.1 Display controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 07)
Subsystem: Dell Device 024f
Flags: bus master, fast devsel, latency 0
Memory atf6b00000(64-bit, non-prefetchable) [size=1M]
Capabilities: <access denied>X.Org X Server 1.7.5.902 (1.7.6 RC 2)
Release Date: 2010-03-12
X Protocol Version 11, Revision 0(--) PCI:*(0:0:2:0) 8086:2a42:1028:024f Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller rev 7, Mem @ 0xf6c00000/4194304, 0xe0000000/268435456, I/O @ 0x0000ef98/8, BIOS @ 0x????????/131072
(--) PCI: (0:0:2:1) 8086:2a43:1028:024f Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller rev 7, Mem @ 0xf6b00000/1048576, BIOS @ 0x????????/65536(II) LoadModule: "intel"
(II) Loading/usr/lib64/xorg/modules/drivers/intel_drv.so(II) Module intel: vendor="X.Org Foundation"
compiled for 1.7.0, module version = 2.9.1 Module class: X.Org Video Driver
ABI class: X.Org Video Driver, version 6.0
(II) intel: Driver for Intel Integrated Graphics Chipsets: i810,i810-dc100, i810e, i815, i830M, 845G, 852GM/855GM, 865G, 915G, E7221 (i915), 915GM, 945G, 945GM, 945GME, Pineview GM, Pineview G, 965G, G35, 965Q, 946GZ, 965GM, 965GME/GLE, G33, Q35, Q33, GM45, 4 Series, G45/G43, Q45/Q43, G41, B43, Clarkdale, Arrandale(II) Primary Device is: PCI 00@00:02:0
drmOpenDevice: node name is/dev/dri/card0
drmOpenDevice: open result is 9, (OK)
drmOpenByBusid: Searching for BusID pci:0000:00:02.0
drmOpenDevice: node name is/dev/dri/card0
drmOpenDevice: open result is 9, (OK)
drmOpenByBusid: drmOpenMinor returns 9
drmOpenByBusid: drmGetBusid reports pci:0000:00:02.0
(**) intel(0): Depth 24, (--) framebuffer bpp 32
(==) intel(0): RGB weight 888
(==) intel(0): Default visual is TrueColor
(II) intel(0): Integrated Graphics Chipset: Intel(R) GM45 (--) intel(0): Chipset: "GM45"
(II) intel(0): Output LVDS1 has no monitor section
(II) intel(0): found backlight control interface/sys/class/backlight/acpi_video0
(II) intel(0): Output VGA1 has no monitor section
(II) intel(0): Output HDMI1 has no monitor section
(II) intel(0): Output DP1 has no monitor section
(II) intel(0): Output HDMI2 has no monitor section
(II) intel(0): Output DP2 has no monitor section
(II) intel(0): Output DP3 has no monitor section
(II) intel(0): Output TV1 has no monitor section
(II) intel(0): EDID for output LVDS1
(II) intel(0): Manufacturer: SEC Model: 5443 Serial#: 0 (II) intel(0): Year: 2008 Week: 0
(II) intel(0): EDID Version: 1.3
(II) intel(0): Digital Display Input
(II) intel(0): Max Image Size [cm]: horiz.: 33 vert.: 21 (II) intel(0): Gamma: 2.20
(II) intel(0): No DPMS capabilities specified
(II) intel(0): Supported color encodings: RGB 4:4:4 YCrCb 4:4:4 (II) intel(0): First detailed timing is preferred mode
(II) intel(0): redX: 0.580 redY: 0.340 greenX: 0.310 greenY: 0.550
(II) intel(0): blueX: 0.155 blueY: 0.155 whiteX: 0.313 whiteY: 0.329
(II) intel(0): Manufacturer's mask: 0
(II) intel(0): Supported detailed timing:
(II) intel(0): clock: 164.2 MHz Image Size: 331 x 207 mm
(II) intel(0):h_active: 1920h_sync: 2020h_sync_end2052h_blank_end2216h_border: 0
(II) intel(0):v_active: 1200v_sync: 1202v_sync_end1208v_blanking:
1235v_border: 0
(II) intel(0): Supported detailed timing:
(II) intel(0): clock: 109.5 MHz Image Size: 331 x 207 mm
(II) intel(0):h_active: 1920h_sync: 2020h_sync_end2052h_blank_end2216h_border: 0
(II) intel(0):v_active: 1200v_sync: 1202v_sync_end1208v_blanking:
1235v_border: 0
(II) intel(0): RX392<80>154CT
(II) intel(0):
Quoted 12 lines · Karl Vogel
I have been experiencing a very sluggish X server right after starting my laptop. The effect goes away after a few minutes, so it was rather hard to track down.
Recently after updating WINE to a newer release, I've been able to trigger the effect by running a certain game under wine on a remote computer and using VirtualGL to transport the GL display to my laptop. As I first thought the issue was somewhere with wine/virtualgl, I posted the question to the virtualgl mailing list. The thread is here:
http://thread.gmane.org/gmane.comp.video.opengl.virtualgl.user/170
Quoted 8 lines
But after some further investigation, it seems the issue is with the i915 driver.
The 'effect' is that only the mouse pointer works in the X server. The cpu usage on the laptop during the sluggishness is minimal. When I suspend the game with winedbg, the X server slowly becomes responsive again.
The output from latencytop seems to point to i915 being the culprit:
If there's some code doing glFlush()es, it's probably that code at fault. You don't need to do that unless you're doing frontbuffer rendering, and if you're doing frontbuffer rendering you should really be doing backbuffer rendering. I don't see a kernel issue here.
Quoted 11 lines · Eric Anholt
Quoted 5 lines · Karl Vogel
The 'effect' is that only the mouse pointer works in the X server. The cpu usage on the laptop during the sluggishness is minimal. When I suspend the game with winedbg, the X server slowly becomes responsive again.
The output from latencytop seems to point to i915 being the culprit:
If there's some code doing glFlush()es, it's probably that code at fault. You don't need to do that unless you're doing frontbuffer rendering, and if you're doing frontbuffer rendering you should really be doing backbuffer rendering. I don't see a kernel issue here.
That doesnt explain why the box completely locks up on 2.6.34-rc2 though, where only a cold reboot works.
It also seems a bit odd that 1 opengl application is able to entirely lockup user interaction of the X server, then again that might be more of an X issue than a kernel one.. or is it expected behaviour? (the only thing you can do is wiggle the pointer, nothing else.. no window selection, no window dragging, not even start new apps that open windows)
Quoted 8 lines · Karl Vogel
Quoted 7 lines · Eric Anholt
Quoted 5 lines · Karl Vogel
The 'effect' is that only the mouse pointer works in the X server. The cpu usage on the laptop during the sluggishness is minimal. When I suspend the game with winedbg, the X server slowly becomes responsive again.
The output from latencytop seems to point to i915 being the culprit:
Quoted 7 lines
Quoted 4 lines
If there's some code doing glFlush()es, it's probably that code at fault. You don't need to do that unless you're doing frontbuffer rendering, and if you're doing frontbuffer rendering you should really be doing backbuffer rendering. I don't see a kernel issue here.
That doesnt explain why the box completely locks up on 2.6.34-rc2 though, where only a cold reboot works.
Missed that part of the message. If there's a regression, bisect please.
On Mon, Mar 22, 2010 at 4:34 PM, Eric Anholt <[email protected]> wrote:> On Mon, 22 Mar 2010 09:11:06 +0100, Karl Vogel <[email protected]> wrote:>> On Mon, Mar 22, 2010 at 5:20 AM, Eric Anholt <[email protected]> wrote:>> > On Sat, 20 Mar 2010 14:41:41 +0100, Karl Vogel <[email protected]> wrote:>> >> The 'effect' is that only the mouse pointer works in the X server. The>> >> cpu usage on the laptop during the sluggishness is minimal. When I>> >> suspend the game with winedbg, the X server slowly becomes responsive again.>> >>>> >> The output from latencytop seems to point to i915 being the culprit:>> >>> > If there's some code doing glFlush()es, it's probably that code at>> > fault. You don't need to do that unless you're doing frontbuffer>> > rendering, and if you're doing frontbuffer rendering you should really>> > be doing backbuffer rendering. I don't see a kernel issue here.>>>> That doesnt explain why the box completely locks up on 2.6.34-rc2>> though, where only a cold reboot works.>> Missed that part of the message. If there's a regression, bisect> please.
Apparently the crash was caused by a hardware bug in the intel chipsetwhich is 8086:2a40 rev 07. While doing the bisect I got an error:
DRHD: handling fault status reg 2DMAR:[DMA Write] Request device [00:02.0] fault addr dd69a000DMAR:[fault reason 05] PTE Write access is not set
After some googling around, I found this bugzilla entry which explains it:
https://bugzilla.redhat.com/show_bug.cgi?id=538163#c58
The issue appears that the graphics chip is corrupting memory:
"Unfortunately, this particular chipset sometimes reads from the GTT, does thetranslation, then writes the translated address back to the _original_ GTTinstead of to the shadow GTT. That's why you're seeing real physical addresseswhere you should have 'virtual DMA addresses', and you get the faults. "
Adding "intel_iommu=igfx_off" to the kernel command line resolved the issue.The fedora kernel automatically disables this when it detects this particularchipset revision.
As for the freeze/slowdown right after booting, sysprof shows that more than 77%of the time is spent inside: drm_mode_getconnector
[/usr/bin/Xorg] 0.00% 80.29% ioctl 0.00% 78.47% - - kernel - - 0.01% 78.47% system_call_fastpath 0.00% 77.15% sys_ioctl 0.00% 77.15% do_vfs_ioctl 0.01% 77.15% vfs_ioctl 0.00% 77.14% drm_ioctl 0.01% 77.14% drm_mode_getconnector 0.00% 77.02% drm_helper_probe_single_connector_modes 0.00% 77.02% intel_lvds_get_modes 0.00% 62.46% intel_ddc_get_modes 0.00% 62.46% drm_get_edid 0.00% 62.45% drm_ddc_read_edid 0.00% 62.45% drm_do_probe_ddc_edid 0.00% 62.45% i2c_transfer 0.00% 62.45% bit_xfer 0.01% 62.44% sclhi 0.00% 25.76% set_clock 0.08% 12.86% __udelay 0.00% 12.78% get_clock 0.11% 0.11% __const_udelay 0.01% 0.01% set_clock 0.12% 12.54% __const_udelay 0.01% 12.42% __delay 0.00% 0.00% __udelay 0.00% 12.08% acknak 0.00% 7.68% sdahi 0.01% 2.30% try_address 0.00% 1.20% i2c_outb 0.00% 0.59% get_data 0.11% 0.11% i2c_stop 0.00% 0.09% i2c_repstart 0.00% 0.07% i2c_start 0.00% 0.01% __const_udelay 0.01% 0.01% __udelay 0.01% 0.01% drm_add_edid_modes 0.00% 0.00% drm_mode_connector_update_edid_property 0.00% 0.00% intel_hdmi_detect 0.00% 8.01% intel_dp_detect 0.00% 4.91% intel_tv_detect 0.00% 1.23% intel_lvds_detect 0.00% 0.38% intel_crt_detect 0.00% 0.03% drm_get_connector_name 0.00% 0.01% i915_gem_execbuffer 0.00% 0.06% i915_gem_mmap_ioctl 0.00% 0.01% i915_gem_set_domain_ioctl 0.00% 0.01% drm_gem_close_ioctl 0.00% 0.01% drm_mode_object_find 0.00% 0.00% i915_gem_busy_ioctl 0.00% 0.00% lock_kernel 0.00% 0.00% i915_gem_create_ioctl 0.00% 0.00% copy_to_user 0.00% 0.00% copy_user_generic_string 0.00% 0.00%ÿôèº{.nÇ+‰·Ÿ®‰†+%ŠËÿ±éݶ¥Šwÿº{.nÇ+‰·¥Š{±þG«�éÿŠ{ayºÊ‡Ú™ë,j¢f£¢·hš�ï�êÿ‘êçz_è®(階ŠÝ¢j"�ú¶m§ÿÿ¾«þG«�éÿ¢¸?™¨èÚ&£ø§~�á¶iO•æ¬z·švØ^¶m§ÿÿÃÿ¶ìÿ¢¸?–I¥
Quoted 36 lines · Karl Vogel
Quoted 8 lines · Eric Anholt
Quoted 7 lines · Karl Vogel
Quoted 6 lines · Eric Anholt
Quoted 5 lines · Karl Vogel
The 'effect' is that only the mouse pointer works in the X server. The cpu usage on the laptop during the sluggishness is minimal. When I suspend the game with winedbg, the X server slowly becomes responsive again.
The output from latencytop seems to point to i915 being the culprit:
Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
If there's some code doing glFlush()es, it's probably that code at fault. You don't need to do that unless you're doing frontbuffer rendering, and if you're doing frontbuffer rendering you should really be doing backbuffer rendering. I don't see a kernel issue here.
That doesnt explain why the box completely locks up on 2.6.34-rc2 though, where only a cold reboot works.
Missed that part of the message. If there's a regression, bisect please.
Apparently the crash was caused by a hardware bug in the intel chipset which is 8086:2a40 rev 07. While doing the bisect I got an error:
DRHD: handling fault status reg 2
DMAR:[DMA Write] Request device [00:02.0] fault addrdd69a000DMAR:[fault reason 05] PTE Write access is not setAfter some googling around, I found this bugzilla entry which explains it:
https://bugzilla.redhat.com/show_bug.cgi?id=538163#c58
The issue appears that the graphics chip is corrupting memory:
"Unfortunately, this particular chipset sometimes reads from the GTT, does the translation, then writes the translated address back to the _original_ GTT instead of to the shadow GTT. That's why you're seeing real physical addresses
where you should have 'virtual DMA addresses', and you get the faults. "
Quoted 6 lines
Adding "intel_iommu=igfx_off" to the kernel command line resolved the issue. The fedora kernel automatically disables this when it detects this particular chipset revision.
As for the freeze/slowdown right after booting, sysprof shows that more than 77% of the time is spent inside:
drm_mode_getconnector
http://lists.freedesktop.org/archives/intel-gfx/2010-February/005922.html
I'm waiting for the encoder/connector stuff to get merged before I either pester people about that bug again or try to fix it myself.
You can try the same hack I use (comment out the initialization of all digital outputs) if you don't use them -- that completely fixes it for me.
--Andy
Am Donnerstag 01 April 2010 schrieb Andy Lutomirski:
Quoted 44 lines
Quoted 36 lines · Karl Vogel
Quoted 8 lines · Eric Anholt
Quoted 7 lines · Karl Vogel
Quoted 6 lines · Eric Anholt
Quoted 5 lines · Karl Vogel
The 'effect' is that only the mouse pointer works in the X server. The cpu usage on the laptop during the sluggishness is minimal. When I suspend the game with winedbg, the X server slowly becomes responsive again.
The output from latencytop seems to point to i915 being the culprit:
Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
If there's some code doing glFlush()es, it's probably that code at fault. You don't need to do that unless you're doing frontbuffer rendering, and if you're doing frontbuffer rendering you should really be doing backbuffer rendering. I don't see a kernel issue here.
That doesnt explain why the box completely locks up on 2.6.34-rc2 though, where only a cold reboot works.
Missed that part of the message. If there's a regression, bisect please.
Apparently the crash was caused by a hardware bug in the intel chipset which is 8086:2a40 rev 07. While doing the bisect I got an error:
DRHD: handling fault status reg 2
DMAR:[DMA Write] Request device [00:02.0] fault addrdd69a000DMAR:[fault reason 05] PTE Write access is not setAfter some googling around, I found this bugzilla entry which explains it:
https://bugzilla.redhat.com/show_bug.cgi?id=538163#c58
The issue appears that the graphics chip is corrupting memory:
"Unfortunately, this particular chipset sometimes reads from the GTT, does the translation, then writes the translated address back to the _original_ GTT instead of to the shadow GTT. That's why you're seeing real physical addresses
where you should have 'virtual DMA addresses', and you get the faults. "Quoted 6 lines
Adding "intel_iommu=igfx_off" to the kernel command line resolved the issue. The fedora kernel automatically disables this when it detects this particular chipset revision.
As for the freeze/slowdown right after booting, sysprof shows that more than 77% of the time is spent inside:
drm_mode_getconnector
Quoted 7 lines
http://lists.freedesktop.org/archives/intel-gfx/2010-February/005922.html
I'm waiting for the encoder/connector stuff to get merged before I either pester people about that bug again or try to fix it myself.
You can try the same hack I use (comment out the initialization of all digital outputs) if you don't use them -- that completely fixes it for me.
Andy, please try this patch.
Eric, please review.
Thanks,
Karsten
[RFC PATCH] drm/i915: Don't touch PORT_HOTPLUG_EN in intel_dp_detect()
Von: Karsten Wiese <[email protected]>
An: [email protected], Jesse Barnes <[email protected]>, Eric Anholt <[email protected]>
PORT_HOTPLUG_EN has allready been setup in i915_driver_irq_postinstall(), when intel_dp_detect() runs.
Delete the DP[BCD]_HOTPLUG_INT_EN defines, they are not referenced anymore.
I found this while searching for a fix for
https://bugzilla.redhat.com/show_bug.cgi?id=528312
---
See also
"drm/i915: only enable hotplug for detected outputs" b01f2c3a4a37d09a47ad73ccbb46d554d21cfeb0
drivers/gpu/drm/i915/i915_reg.h | 3 --- drivers/gpu/drm/i915/intel_dp.c | 10 ---------- 2 files changed, 0 insertions(+), 13 deletions(-)
drivers/gpu/drm/i915/i915_reg.h +0 -3
@@ -854,11 +854,8 @@/* Hotplug control (945+ only) */ #define PORT_HOTPLUG_EN 0x61110#define HDMIB_HOTPLUG_INT_EN (1 << 29) -#define DPB_HOTPLUG_INT_EN (1 << 29)#define HDMIC_HOTPLUG_INT_EN (1 << 28) -#define DPC_HOTPLUG_INT_EN (1 << 28)#define HDMID_HOTPLUG_INT_EN (1 << 27) -#define DPD_HOTPLUG_INT_EN (1 << 27)#define SDVOB_HOTPLUG_INT_EN (1 << 26)#define SDVOC_HOTPLUG_INT_EN (1 << 25)#define TV_HOTPLUG_INT_EN (1 << 18)
drivers/gpu/drm/i915/intel_dp.c +0 -10
@@ -1179,16 +1179,6 @@ intel_dp_detect(struct drm_connector *connector)if (IS_IRONLAKE (dev))return ironlake_dp_detect (connector); - temp =I915_READ (PORT_HOTPLUG_EN ); - -I915_WRITE (PORT_HOTPLUG_EN , - temp | -DPB_HOTPLUG_INT_EN | -DPC_HOTPLUG_INT_EN | -DPD_HOTPLUG_INT_EN ); - -POSTING_READ (PORT_HOTPLUG_EN ); -switch (dp_priv->output_reg) {case DP_B : bit =DPB_HOTPLUG_INT_STATUS ;
--