Bug#767148: linux-image-3.16-3-amd64: OpenCL doesn't work on Intel GPU

2014-10-30 Thread Ben Hutchings
On Wed, 29 Oct 2014 20:15:04 + "Rebecca N. Palmer" 
 wrote:
> Control: tags -1 patch
> 
> I have now tested that upstream patch (note I earlier accidentally sent 
> you a link showing only one file's changes; we need the whole commit, 
> https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=c9224faa59c3071ecfa2d4b24592f4eb61e57069
>  
> , as attached), and it does fix this bug.

Thanks, I'll include this in the next update.

> While installing the fixed kernel (i.e. before actually using it), the 
> system froze at the update-grub stage (photo attached) and wouldn't 
> reboot even with Alt+SysRq+b (but didn't flash CapsLock); is that 
> because installing a kernel over the running one isn't supposed to work 
> (I thought the -3 was the ABI so didn't increment it) or is something 
> else wrong?

Even if this patch changed the kernel module ABI (which it doesn't) that
should only result in being unable to load new modules until you
rebooted into the new kernel.  Each symbol that's exported to modules
has a 'version' that is a hash of related type definitions, and the
kernel's module linker requires both name and version to match.

This certainly seems like another bug.  However I don't think we'll be
able to do anything about it without some kind of crash log.  If your
computer boots with EFI then the kernel might have been able to save a
crash log in flash.  The reportbug script for the kernel knows how to
read this back (it appears in the /sys/fs/pstore directory), and will
prompt you to include it.

Ben.

-- 
Ben Hutchings
The world is coming to an end.  Please log off.


signature.asc
Description: This is a digitally signed message part


Processed: Re: Bug#767148: linux-image-3.16-3-amd64: OpenCL doesn't work on Intel GPU

2014-10-29 Thread Debian Bug Tracking System
Processing control commands:

> tags -1 patch
Bug #767148 [src:linux] linux-image-3.16-3-amd64: OpenCL doesn't work on Intel 
GPU
Added tag(s) patch.

-- 
767148: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=767148
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems


--
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/handler.s.b767148.141461375924747.transcr...@bugs.debian.org



Bug#767148: linux-image-3.16-3-amd64: OpenCL doesn't work on Intel GPU

2014-10-28 Thread Rebecca N. Palmer

Package: src:linux
Version: 3.16.5-1
Severity: important
Control: affects -1 beignet
Control: tags -1 fixed-upstream
X-Debbugs-CC: s...@debian.org,pkg-opencl-de...@lists.alioth.debian.org

In current jessie, beignet (OpenCL for Intel GPUs, 0.8-1.1) is 
non-functional:

$ sudo apt-get install beignet python3-pyopencl
$ python3 beignet_test.py
python3: /tmp/buildd/beignet-0.8/src/intel/intel_gpgpu.c:427: 
intel_gpgpu_check_binded_buf_address: Assertion 
`gpgpu->binded_buf[i]->offset != 0' failed.

Aborted

It had previously worked in Ubuntu 14.04 (beignet 0.8-1, linux 3.13.0) 
on the same hardware (i5-3230M).  (beignet isn't in Wheezy.)


beignet's upstream documentation [0] says this is a known issue with 
3.15/3.16 kernels and suggested the workaround

$ sudo sh -c "echo -n 0 > /sys/module/i915/parameters/enable_cmd_parser"
which made it work, but the description [1] suggests this will break 
some OpenGL functionality and might be a security risk.


This is fixed upstream by [2]; I haven't yet tried this.

[0] http://www.freedesktop.org/wiki/Software/Beignet/
[1] 
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/drivers/gpu/drm/i915/i915_cmd_parser.c
[2] 
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/drivers/gpu/drm/i915/i915_cmd_parser.c?id=c9224faa59c3071ecfa2d4b24592f4eb61e57069



-- Package-specific info:
** Version:
Linux version 3.16-3-amd64 (debian-kernel@lists.debian.org) (gcc version 
4.8.3 (Debian 4.8.3-12) ) #1 SMP Debian 3.16.5-1 (2014-10-10)


** Command line:
BOOT_IMAGE=/vmlinuz-3.16-3-amd64 
root=UUID=1ebca590-0a18-44f5-8d75-f96f5ca7c36e ro quiet


** Not tainted

** Kernel log:
[   27.077954] ACPI Warning: SystemIO range 
0x0530-0x053f conflicts with OpRegion 
0x0500-0x057f (\GPXX) (20140424/utaddress-258)
[   27.077959] ACPI: If an ACPI driver is available for this device, you 
should use it instead of the native driver
[   27.077961] ACPI Warning: SystemIO range 
0x0500-0x052f conflicts with OpRegion 
0x0500-0x0563 (\GPIO) (20140424/utaddress-258)
[   27.077965] ACPI Warning: SystemIO range 
0x0500-0x052f conflicts with OpRegion 
0x0500-0x057f (\GPXX) (20140424/utaddress-258)
[   27.077970] ACPI: If an ACPI driver is available for this device, you 
should use it instead of the native driver

[   27.077972] lpc_ich: Resource conflict(s) found affecting gpio_ich
[   27.248942] shpchp: Standard Hot Plug PCI Controller Driver version: 0.4
[   27.284399] toshiba_bluetooth: Detected Toshiba ACPI Bluetooth device 
- installing RFKill handler

[   27.284453] toshiba_bluetooth: Re-enabling Toshiba Bluetooth
[   27.502477] [drm] Memory usable by graphics device = 2048M
[   27.502492] [drm] Replacing VGA console driver
[   27.502508] checking generic (b000 408000) vs hw (b000 1000)
[   27.502512] fb: switching to inteldrmfb from simple
[   27.502568] Console: switching to colour dummy device 80x25
[   27.524376] i915 :00:02.0: irq 43 for MSI/MSI-X
[   27.524393] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013).
[   27.524394] [drm] Driver supports precise vblank timestamp query.
[   27.524506] vgaarb: device changed decodes: 
PCI::00:02.0,olddecodes=io+mem,decodes=io+mem:owns=io+mem

[   27.565932] fbcon: inteldrmfb (fb0) is primary device
[   27.582152] wmi: Mapper loaded
[   27.694480] Bluetooth: Core ver 2.19
[   27.694494] NET: Registered protocol family 31
[   27.694495] Bluetooth: HCI device and connection manager initialized
[   27.694501] Bluetooth: HCI socket layer initialized
[   27.694502] Bluetooth: L2CAP socket layer initialized
[   27.694509] Bluetooth: SCO socket layer initialized
[   27.833463] usbcore: registered new interface driver btusb
[   27.907603] cfg80211: Calling CRDA to update world regulatory domain
[   27.951039] EFI Variables Facility v0.08 2004-May-17
[   27.959345] pstore: Registered efi as persistent store backend
[   28.248411] atkbd serio0: Unknown key released (translated set 2, 
code 0x7c on isa0060/serio0).
[   28.248413] atkbd serio0: Use 'setkeycodes 7c ' to make it 
known.

[   28.279982] input: PC Speaker as /devices/platform/pcspkr/input/input14
[   28.377421] Console: switching to colour frame buffer device 170x48
[   28.380569] i915 :00:02.0: fb0: inteldrmfb frame buffer device
[   28.380570] i915 :00:02.0: registered panic notifier
[   28.401883] ACPI: Video Device [GFX0] (multi-head: yes  rom: no 
post: no)
[   28.401985] input: Video Bus as 
/devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A08:00/LNXVIDEO:00/input/input15
[   28.402061] [drm] Initialized i915 1.6.0 20080730 for :00:02.0 on 
minor 0

[   28.402441] snd_hda_intel :00:1b.0: irq 44 for MSI/MSI-X
[   28.421461] ath: phy0: ASPM enabled: 0x43
[   28.421465] ath: EEPROM regdomain: 0x6a
[   28.421466] ath: EEPROM indicates we should expect a direct regpair map
[   28.421468