You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
So far, I get an error from testing in-kernel amdgpu:
graphics/1_auto_switch_card_Vega_10_XT__Radeon_PRO_SSG_ not supported job cannot be started: resource expression "graphics_card.driver in ['nvidia', 'amdgpu-pro', 'pcieport']" evaluates to false
graphics/1_gl_support_Vega_10_XT__Radeon_PRO_SSG_ not supported job cannot be started: resource expression "package.name == 'nux-tools'" evaluates to false
On Somerville projects, the factory image always shipped in performance mode for estar.
This test job should be fixed and apply to the machines that installed with the factory image, in order to verify if there is a regression with dGPU.
This issue was migrated from https://bugs.launchpad.net/plainbox-provider-checkbox/+bug/1881861
Summary
Description
so far we use DRI_PRIME=1 to make all amdgpu driver (pro stack, non-pro stack, in-kernel amdgpu) rendering.
So, there's some part of automatical dgpu test cases [1][2] is waiting for revising otherwise in-kernel amdgpu will not be verified.
There're some related code there, but not yet understand why it works: https://git.launchpad.net/plainbox-provider-checkbox/tree/bin/graphics_env#n25
So far, I get an error from testing in-kernel amdgpu:
graphics/1_auto_switch_card_Vega_10_XT__Radeon_PRO_SSG_ not supported job cannot be started: resource expression "graphics_card.driver in ['nvidia', 'amdgpu-pro', 'pcieport']" evaluates to false
graphics/1_gl_support_Vega_10_XT__Radeon_PRO_SSG_ not supported job cannot be started: resource expression "package.name == 'nux-tools'" evaluates to false
[1] https://git.launchpad.net/plainbox-provider-checkbox/tree/units/graphics/jobs.pxu#n525
[2] https://git.launchpad.net/plainbox-provider-resource/tree/bin/graphics_card_resource#n132
Attachments
No attachments
Tags:
[]
The text was updated successfully, but these errors were encountered: