lkml.org 
[lkml]   [2021]   [Nov]   [7]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Subject[PATCH 07/10] habanalabs/gaudi: Fix collective wait bug
Date
From: farah kassabri <fkassabri@habana.ai>

In Signaling-From-Graph case, the driver didn't set the hw_sob pointer
at the right place, which is needed for the cs completion
check prior to start sending all the master/slaves jobs to device.

Signed-off-by: farah kassabri <fkassabri@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
drivers/misc/habanalabs/gaudi/gaudi.c | 15 +++++++++++----
1 file changed, 11 insertions(+), 4 deletions(-)

diff --git a/drivers/misc/habanalabs/gaudi/gaudi.c b/drivers/misc/habanalabs/gaudi/gaudi.c
index b4814369062e..a9e279bfebae 100644
--- a/drivers/misc/habanalabs/gaudi/gaudi.c
+++ b/drivers/misc/habanalabs/gaudi/gaudi.c
@@ -1276,6 +1276,7 @@ static int gaudi_collective_wait_init_cs(struct hl_cs *cs)
container_of(cs->signal_fence, struct hl_cs_compl, base_fence);
struct hl_cs_compl *cs_cmpl =
container_of(cs->fence, struct hl_cs_compl, base_fence);
+ struct hl_cs_encaps_sig_handle *handle = cs->encaps_sig_hdl;
struct gaudi_collective_properties *cprop;
u32 stream, queue_id, sob_group_offset;
struct gaudi_device *gaudi;
@@ -1288,10 +1289,16 @@ static int gaudi_collective_wait_init_cs(struct hl_cs *cs)
gaudi = hdev->asic_specific;
cprop = &gaudi->collective_props;

- /* In encaps signals case the SOB info will be retrieved from
- * the handle in gaudi_collective_slave_init_job.
- */
- if (!cs->encaps_signals) {
+ if (cs->encaps_signals) {
+ cs_cmpl->hw_sob = handle->hw_sob;
+ /* at this checkpoint we only need the hw_sob pointer
+ * for the completion check before start going over the jobs
+ * of the master/slaves, the sob_value will be taken later on
+ * in gaudi_collective_slave_init_job depends on each
+ * job wait offset value.
+ */
+ cs_cmpl->sob_val = 0;
+ } else {
/* copy the SOB id and value of the signal CS */
cs_cmpl->hw_sob = signal_cs_cmpl->hw_sob;
cs_cmpl->sob_val = signal_cs_cmpl->sob_val;
--
2.25.1
\
 
 \ /
  Last update: 2021-11-07 21:53    [W:0.095 / U:0.040 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site