[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20210711125418.17533-4-ogabbay@kernel.org>
Date: Sun, 11 Jul 2021 15:54:15 +0300
From: Oded Gabbay <ogabbay@...nel.org>
To: linux-kernel@...r.kernel.org
Cc: Tomer Tayar <ttayar@...ana.ai>
Subject: [PATCH 4/7] habanalabs: mark linux image as not loaded after hw_fini
From: Tomer Tayar <ttayar@...ana.ai>
If hard reset fails after the call to hw_fini and before loading the
linux image to the device, a subsequent call to hw_fini should
communicate via COMMS (or MSG_TO_CPU regs for old FW versions).
However, the driver still tries in this case to communicate via the GIC,
and thus no hard reset is actually done.
To avoid that, the patch clears the linux_loaded flag after every call
to hw_fini.
Signed-off-by: Tomer Tayar <ttayar@...ana.ai>
Reviewed-by: Oded Gabbay <ogabbay@...nel.org>
Signed-off-by: Oded Gabbay <ogabbay@...nel.org>
---
drivers/misc/habanalabs/common/device.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/drivers/misc/habanalabs/common/device.c b/drivers/misc/habanalabs/common/device.c
index 28fe15a28361..4fcd24e5a609 100644
--- a/drivers/misc/habanalabs/common/device.c
+++ b/drivers/misc/habanalabs/common/device.c
@@ -1075,9 +1075,12 @@ int hl_device_reset(struct hl_device *hdev, u32 flags)
hdev->asic_funcs->hw_fini(hdev, hard_reset);
if (hard_reset) {
+ hdev->fw_loader.linux_loaded = false;
+
/* Release kernel context */
if (hdev->kernel_ctx && hl_ctx_put(hdev->kernel_ctx) == 1)
hdev->kernel_ctx = NULL;
+
hl_vm_fini(hdev);
hl_mmu_fini(hdev);
hl_eq_reset(hdev, &hdev->event_queue);
@@ -1611,6 +1614,8 @@ void hl_device_fini(struct hl_device *hdev)
/* Reset the H/W. It will be in idle state after this returns */
hdev->asic_funcs->hw_fini(hdev, true);
+ hdev->fw_loader.linux_loaded = false;
+
/* Release kernel context */
if ((hdev->kernel_ctx) && (hl_ctx_put(hdev->kernel_ctx) != 1))
dev_err(hdev->dev, "kernel ctx is still alive\n");
--
2.25.1
Powered by blists - more mailing lists