In training, there is a need for a large amount of patching to the recipe.
This results in many command buffers contains a lot of DMA packets. The
number of command buffers per CS is larger than the current maximum of 64,
which is an arbitrary number that is enough for inference, but it has no
real affect on the code and/or resources of the host machine.
Signed-off-by: Oded Gabbay <[email protected]>
---
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 4ff2da859653..0813041f669a 100644
--- a/drivers/misc/habanalabs/habanalabs.h
+++ b/drivers/misc/habanalabs/habanalabs.h
@@ -40,7 +40,7 @@
#define HL_MAX_QUEUES 128
-#define HL_MAX_JOBS_PER_CS 64
+#define HL_MAX_JOBS_PER_CS 512
/* MUST BE POWER OF 2 and larger than 1 */
#define HL_MAX_PENDING_CS 64
--
2.17.1
On Sun, Nov 10, 2019 at 11:56 PM Oded Gabbay wrote:
> In training, there is a need for a large amount of patching to the recipe.
> This results in many command buffers contains a lot of DMA packets. The
> number of command buffers per CS is larger than the current maximum of
> 64, which is an arbitrary number that is enough for inference, but it has no
> real affect on the code and/or resources of the host machine.
>
> Signed-off-by: Oded Gabbay <[email protected]>
Reviewed-by: Omer Shpigelman <[email protected]>