Bug 105068 - vulkaninfo gives an VK_ERROR_INITIALIZATION_FAILED
Summary: vulkaninfo gives an VK_ERROR_INITIALIZATION_FAILED
Status: RESOLVED FIXED
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/Vulkan/radeon (show other bugs)
Version: 17.2
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: mesa-dev
QA Contact: mesa-dev
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2018-02-13 03:38 UTC by Pietro Pesci Feltri
Modified: 2018-02-17 16:51 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Attachments
attachment-17476-0.html (3.27 KB, text/html)
2018-02-13 13:57 UTC, Pietro Pesci Feltri
Details
attachment-326-0.html (4.21 KB, text/html)
2018-02-13 14:50 UTC, Pietro Pesci Feltri
Details
attachment-4910-0.html (288 bytes, text/html)
2018-02-17 04:56 UTC, Pietro Pesci Feltri
Details
attachment-1586-0.html (1.33 KB, text/html)
2018-02-17 16:45 UTC, Pietro Pesci Feltri
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Pietro Pesci Feltri 2018-02-13 03:38:52 UTC
On Fedora 27, with mesa-vulkan-drivers-17.2.4-3.fc27.x86_64 installed, the output in vulkaninfo is:

===========
VULKAN INFO
===========

Vulkan API Version: 1.0.65

WARNING: [loader] Code 0 : Skipping manifest file /usr/share/vulkan/explicit_layer.d/VkLayer_object_tracker.json - The file has already been read once
WARNING: [loader] Code 0 : Skipping manifest file /usr/share/vulkan/explicit_layer.d/VkLayer_core_validation.json - The file has already been read once
WARNING: [loader] Code 0 : Skipping manifest file /usr/share/vulkan/explicit_layer.d/VkLayer_threading.json - The file has already been read once
WARNING: [loader] Code 0 : Skipping manifest file /usr/share/vulkan/explicit_layer.d/VkLayer_standard_validation.json - The file has already been read once
WARNING: [loader] Code 0 : Skipping manifest file /usr/share/vulkan/explicit_layer.d/VkLayer_parameter_validation.json - The file has already been read once
WARNING: [loader] Code 0 : Skipping manifest file /usr/share/vulkan/explicit_layer.d/VkLayer_unique_objects.json - The file has already been read once
WARNING: [loader] Code 0 : Skipping manifest file /usr/share/vulkan/icd.d/radeon_icd.x86_64.json - The file has already been read once
WARNING: [loader] Code 0 : Skipping manifest file /usr/share/vulkan/icd.d/intel_icd.x86_64.json - The file has already been read once

Instance Extensions:
====================
Instance Extensions     count = 9
        VK_KHR_surface                      : extension revision 25
        VK_KHR_xcb_surface                  : extension revision  6
        VK_KHR_xlib_surface                 : extension revision  6
        VK_KHR_wayland_surface              : extension revision  6
        VK_KHR_get_physical_device_properties2: extension revision  1
        VK_KHR_external_memory_capabilities : extension revision  1
        VK_KHR_external_semaphore_capabilities: extension revision  1
        VK_KHR_get_surface_capabilities2    : extension revision  1
        VK_EXT_debug_report                 : extension revision  9
/builddir/build/BUILD/Vulkan-LoaderAndValidationLayers-sdk-1.0.65.2/demos/vulkaninfo.c:1670: failed with VK_ERROR_INITIALIZATION_FAILED

using;
export VK_LOADER_DEBUG=all

gives a very long output where the interesting part at the end is:

ERROR: setupLoaderTermPhysDevs:  Failed to detect any valid GPUs in the current config
ERROR: setupLoaderTrampPhysDevs:  Failed during dispatch call of 'vkEnumeratePhysicalDevices' to lower layers or loader to get count.
/builddir/build/BUILD/Vulkan-LoaderAndValidationLayers-sdk-1.0.65.2/demos/vulkaninfo.c:1670: failed with VK_ERROR_INITIALIZATION_FAILED

My video card is an R7 260X and is supposed to work.

lspci output;
VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Bonaire XTX [Radeon R7 260X/360]
Comment 1 Bas Nieuwenhuizen 2018-02-13 09:01:05 UTC
What does 

glxinfo | grep "OpenGL renderer string"

say? If DRM version is < 3.0.0, you should make sure your kernel is at least 4.13 and then add 

radeon.si_support=0 radeon.cik_support=0 amdgpu.si_support=1 amdgpu.cik_support=1

to the kernel commandline.
Comment 2 Pietro Pesci Feltri 2018-02-13 13:57:44 UTC
Created attachment 137317 [details]
attachment-17476-0.html

OpenGL renderer string: AMD BONAIRE (DRM 2.50.0 / 4.14.16-300.fc27.x86_64,
LLVM 5.0.0)

I changed the file /etc/default/gub to:

cat /etc/default/grub
GRUB_TIMEOUT=5
GRUB_DISTRIBUTOR="$(sed 's, release .*$,,g' /etc/system-release)"
GRUB_DEFAULT=saved
GRUB_DISABLE_SUBMENU=true
GRUB_TERMINAL_OUTPUT="console"
GRUB_CMDLINE_LINUX="rhgb quiet radeon.si_support=0 radeon.cik_support=0
amdgpu.si_support=1 amdgpu.cik_support=1"
GRUB_DISABLE_RECOVERY="true"

and ran:
sudo grub2-mkconfig
reboot

Same result.

On Tue, Feb 13, 2018 at 5:01 AM, <bugzilla-daemon@freedesktop.org> wrote:

> Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl> changed bug 105068
> <https://bugs.freedesktop.org/show_bug.cgi?id=105068>
> What Removed Added
> Status NEW NEEDINFO
>
> *Comment # 1 <https://bugs.freedesktop.org/show_bug.cgi?id=105068#c1> on
> bug 105068 <https://bugs.freedesktop.org/show_bug.cgi?id=105068> from Bas
> Nieuwenhuizen <bas@basnieuwenhuizen.nl> *
>
> What does
>
> glxinfo | grep "OpenGL renderer string"
>
> say? If DRM version is < 3.0.0, you should make sure your kernel is at least
> 4.13 and then add
>
> radeon.si_support=0 radeon.cik_support=0 amdgpu.si_support=1
> amdgpu.cik_support=1
>
> to the kernel commandline.
>
> ------------------------------
> You are receiving this mail because:
>
>    - You reported the bug.
>
>
Comment 3 Pietro Pesci Feltri 2018-02-13 14:50:19 UTC
Created attachment 137320 [details]
attachment-326-0.html

BTW:
uname -a
Linux xxxxxxxxxxxxxxxxxx.dsl.cantv.net 4.14.16-300.fc27.x86_64 #1 SMP Wed
Jan 31 19:24:27 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux



On Tue, Feb 13, 2018 at 9:57 AM, Pietro Pesci Feltri <ppesci@gmail.com>
wrote:

> OpenGL renderer string: AMD BONAIRE (DRM 2.50.0 / 4.14.16-300.fc27.x86_64,
> LLVM 5.0.0)
>
> I changed the file /etc/default/gub to:
>
> cat /etc/default/grub
> GRUB_TIMEOUT=5
> GRUB_DISTRIBUTOR="$(sed 's, release .*$,,g' /etc/system-release)"
> GRUB_DEFAULT=saved
> GRUB_DISABLE_SUBMENU=true
> GRUB_TERMINAL_OUTPUT="console"
> GRUB_CMDLINE_LINUX="rhgb quiet radeon.si_support=0 radeon.cik_support=0
> amdgpu.si_support=1 amdgpu.cik_support=1"
> GRUB_DISABLE_RECOVERY="true"
>
> and ran:
> sudo grub2-mkconfig
> reboot
>
> Same result.
>
> On Tue, Feb 13, 2018 at 5:01 AM, <bugzilla-daemon@freedesktop.org> wrote:
>
>> Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl> changed bug 105068
>> <https://bugs.freedesktop.org/show_bug.cgi?id=105068>
>> What Removed Added
>> Status NEW NEEDINFO
>>
>> *Comment # 1 <https://bugs.freedesktop.org/show_bug.cgi?id=105068#c1> on
>> bug 105068 <https://bugs.freedesktop.org/show_bug.cgi?id=105068> from Bas
>> Nieuwenhuizen <bas@basnieuwenhuizen.nl> *
>>
>> What does
>>
>> glxinfo | grep "OpenGL renderer string"
>>
>> say? If DRM version is < 3.0.0, you should make sure your kernel is at least
>> 4.13 and then add
>>
>> radeon.si_support=0 radeon.cik_support=0 amdgpu.si_support=1
>> amdgpu.cik_support=1
>>
>> to the kernel commandline.
>>
>> ------------------------------
>> You are receiving this mail because:
>>
>>    - You reported the bug.
>>
>>
>
Comment 4 Bas Nieuwenhuizen 2018-02-16 21:49:32 UTC
Looks like you need to call grub2-mkconfig with an argument. Looking at the wiki it should be 

grub2-mkconfig -o /boot/grub2/grub.cfg

(https://fedoraproject.org/wiki/GRUB_2)

Could you try that.
Comment 5 Pietro Pesci Feltri 2018-02-17 04:56:46 UTC
Created attachment 137412 [details]
attachment-4910-0.html

I add the parameters manually pressing "e" in the grub menu 2 days ago. One
hour later the initialization process was not ended.
I was forced to reboot the machine.

Anyway thanks.
Comment 6 Pietro Pesci Feltri 2018-02-17 16:45:32 UTC
Created attachment 137418 [details]
attachment-1586-0.html

Problem solved :).

As stated in
https://wiki.archlinux.org/index.php/AMDGPU#Selecting_the_right_driver

The parameters must be added to the kernel but the parameters depend on
what cart you have installed in the case the card is Southern Islands
**or** Sea Islands.

For Southern Islands parameters are: radeon.si_support=0 amdgpu.si_support=1
For Sea Islands parameters are: radeon.cik_support=0 amdgpu.cik_support=1

but not both, or you will end with a frozen display during the boot as in
my case.

I hope this help others.

On Sat, Feb 17, 2018 at 12:56 AM, Pietro Pesci Feltri <ppesci@gmail.com>
wrote:

> I add the parameters manually pressing "e" in the grub menu 2 days ago.
> One hour later the initialization process was not ended.
> I was forced to reboot the machine.
>
> Anyway thanks.
>
>
>
Comment 7 Pietro Pesci Feltri 2018-02-17 16:51:18 UTC
The problem appear in machines containing Southern Islands cards and Sea Islands cards.

As stated in
https://wiki.archlinux.org/index.php/AMDGPU#Selecting_the_right_driver

The parameters must be added to the kernel but the parameters depend on
what cart you have installed in the case the card is Southern Islands
**or** Sea Islands.

For Southern Islands parameters are: radeon.si_support=0 amdgpu.si_support=1
For Sea Islands parameters are: radeon.cik_support=0 amdgpu.cik_support=1

but not all parameters, or you will end with a frozen display during the boot as in
my case


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.