From: Koby Elbaz <kel...@habana.ai> hl_device_status() returns the status of an acquired device. If a device is going down (following an rmmod cmd), it should be marked as an unusable/malfunctioning device, and hence should not be acquired. However, since this was not the case so far (i.e., a device going down would inaccurately return 'in reset' status allowing the user to acquire the device) it introduced a bug where as part of a reset flow, the driver could not kill processes that have not run yet, and since those processes aren't blocked from reacquiring a device, we get eventually a new flow of a driver attempting to kill all processes in a list that can't be ever really empty.
Signed-off-by: Koby Elbaz <kel...@habana.ai> Reviewed-by: Oded Gabbay <ogab...@kernel.org> Signed-off-by: Oded Gabbay <ogab...@kernel.org> --- drivers/accel/habanalabs/common/device.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/accel/habanalabs/common/device.c b/drivers/accel/habanalabs/common/device.c index 993305871292..5973e4d64e19 100644 --- a/drivers/accel/habanalabs/common/device.c +++ b/drivers/accel/habanalabs/common/device.c @@ -315,7 +315,9 @@ enum hl_device_status hl_device_status(struct hl_device *hdev) { enum hl_device_status status; - if (hdev->reset_info.in_reset) { + if (hdev->device_fini_pending) { + status = HL_DEVICE_STATUS_MALFUNCTION; + } else if (hdev->reset_info.in_reset) { if (hdev->reset_info.in_compute_reset) status = HL_DEVICE_STATUS_IN_RESET_AFTER_DEVICE_RELEASE; else @@ -343,9 +345,9 @@ bool hl_device_operational(struct hl_device *hdev, *status = current_status; switch (current_status) { + case HL_DEVICE_STATUS_MALFUNCTION: case HL_DEVICE_STATUS_IN_RESET: case HL_DEVICE_STATUS_IN_RESET_AFTER_DEVICE_RELEASE: - case HL_DEVICE_STATUS_MALFUNCTION: case HL_DEVICE_STATUS_NEEDS_RESET: return false; case HL_DEVICE_STATUS_OPERATIONAL: -- 2.40.1