When using some Thunderbolt hosts using BIOS-assisted PCI enumeration
with IO BAR assigned, we get an atombios timeout, such as:
[drm:atom_op_jump [amdgpu]] *ERROR* atombios stuck in loop for more than 20secs aborting
[drm:amdgpu_atom_execute_table_locked [amdgpu]] *ERROR* atombios stuck executing B456 (len 304, WS 4, PS 0) @ 0xB51B
[drm:amdgpu_atom_execute_table_locked [amdgpu]] *ERROR* atombios stuck executing B104 (len 183, WS 0, PS 8) @ 0xB17E
amdgpu 0000:08:00.0: amdgpu: gpu post error!
amdgpu 0000:08:00.0: amdgpu: Fatal error during GPU init
amdgpu: probe of 0000:08:00.0 failed with error -22
A workaround is to use MMIO to access ATOMBIOS when device is
Thunderbolt / USB4 attached.
---
drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c | 8 ++++++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c
index 86add0f4e..5d16ec10d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c
@@ -1999,11 +1999,15 @@ int amdgpu_atombios_init(struct amdgpu_device *adev)
atom_card_info->reg_read = cail_reg_read;
atom_card_info->reg_write = cail_reg_write;
/* needed for iio ops */
- if (adev->rio_mem) {
+ if (adev->rio_mem && !pci_is_thunderbolt_attached(adev->pdev)) {
atom_card_info->ioreg_read = cail_ioreg_read;
atom_card_info->ioreg_write = cail_ioreg_write;
} else {
- DRM_DEBUG("PCI I/O BAR is not found. Using MMIO to access ATOM BIOS\n");
+ if (pci_is_thunderbolt_attached(adev->pdev))
+ DRM_DEBUG("Device is attached via Thunderbolt / USB4. Using MMIO to access ATOM BIOS\n");
+ else
+ DRM_DEBUG("PCI I/O BAR is not found. Using MMIO to access ATOM BIOS\n");
+
atom_card_info->ioreg_read = cail_reg_read;
atom_card_info->ioreg_write = cail_reg_write;
}
--
2.30.2
_______________________________________________
amd-gfx mailing list
amd-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
