From: Oded Gabbay <oded.gabbay@xxxxxxxxx> [ Upstream commit 7a65ee046b2238e053f6ebb610e1a082cfc49490 ] When doing training, the DL framework (e.g. tensorflow) performs hundreds of thousands of memory allocations and mappings. In case the driver needs to perform hard-reset during training, the driver kills the application and unmaps all those memory allocations. Unfortunately, because of that large amount of mappings, the driver isn't able to do that in the current timeout (5 seconds). Therefore, increase the timeout significantly to 30 seconds to avoid situation where the driver resets the device with active mappings, which sometime can cause a kernel bug. BTW, it doesn't mean we will spend all the 30 seconds because the reset thread checks every one second if the unmap operation is done. Reviewed-by: Omer Shpigelman <oshpigelman@xxxxxxxxx> Signed-off-by: Oded Gabbay <oded.gabbay@xxxxxxxxx> Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx> --- drivers/misc/habanalabs/habanalabs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/misc/habanalabs/habanalabs.h b/drivers/misc/habanalabs/habanalabs.h index 31ebcf9458fe1..a6dd8e6ca594c 100644 --- a/drivers/misc/habanalabs/habanalabs.h +++ b/drivers/misc/habanalabs/habanalabs.h @@ -23,7 +23,7 @@ #define HL_MMAP_CB_MASK (0x8000000000000000ull >> PAGE_SHIFT) -#define HL_PENDING_RESET_PER_SEC 5 +#define HL_PENDING_RESET_PER_SEC 30 #define HL_DEVICE_TIMEOUT_USEC 1000000 /* 1 s */ -- 2.25.1