When executing in a PCI passthrough based virtuzliation environment, the
hypervisor will usually attempt to send a PCIe bus reset signal to the
ASIC when the VM reboots. In this scenario, the card is not correctly
initialized, but we still consider it to be posted. Therefore, in a
passthrough based environemnt we should always post the card to guarantee
it is in a good state for driver initialization.
Ported from amdgpu commit:
amdgpu: fix asic initialization for virtualized environments
Cc: Andres Rodriguez <andres.rodriguez@xxxxxxx>
Cc: Alex Williamson <alex.williamson@xxxxxxxxxx>
Signed-off-by: Alex Deucher <alexander.deucher@xxxxxxx>
Cc: stable@xxxxxxxxxxxxxxx
---
drivers/gpu/drm/radeon/radeon_device.c | 21 +++++++++++++++++++++
1 file changed, 21 insertions(+)
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index e61c763..21c44b2 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -630,6 +630,23 @@ void radeon_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc)
/*
* GPU helpers function.
*/
+
+/**
+ * radeon_device_is_virtual - check if we are running is a virtual environment
+ *
+ * Check if the asic has been passed through to a VM (all asics).
+ * Used at driver startup.
+ * Returns true if virtual or false if not.
+ */
+static bool radeon_device_is_virtual(void)
+{
+#ifdef CONFIG_X86
+ return boot_cpu_has(X86_FEATURE_HYPERVISOR);
+#else
+ return false;
+#endif
+}
+
/**
* radeon_card_posted - check if the hw has already been initialized
*
@@ -643,6 +660,10 @@ bool radeon_card_posted(struct radeon_device *rdev)
{
uint32_t reg;
+ /* for pass through, always force asic_init */
+ if (radeon_device_is_virtual())
+ return false;
+
/* required for EFI mode on macbook2,1 which uses an r5xx asic */
if (efi_enabled(EFI_BOOT) &&
(rdev->pdev->subsystem_vendor == PCI_VENDOR_ID_APPLE) &&
--
2.5.5
_______________________________________________
dri-devel mailing list
dri-devel@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/dri-devel
