From: Oded Gabbay <oded.gabbay@gmail.com>
To: linux-kernel@vger.kernel.org, SW_Drivers@habana.ai
Subject: [PATCH] habanalabs: count dropped CS because max CS in-flight
Date: Fri, 4 Sep 2020 18:11:19 +0300 [thread overview]
Message-ID: <20200904151119.24224-1-oded.gabbay@gmail.com> (raw)
There is a case where the user reaches the maximum number of CS in-flight.
In that case, the driver rejects the new CS of the user with EAGAIN. Count
that event so the user can query the driver later to see if it happened.
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
---
drivers/misc/habanalabs/common/command_submission.c | 5 ++++-
include/uapi/misc/habanalabs.h | 2 ++
2 files changed, 6 insertions(+), 1 deletion(-)
diff --git a/drivers/misc/habanalabs/common/command_submission.c b/drivers/misc/habanalabs/common/command_submission.c
index a811a9fdf13b..470bffbe9bdc 100644
--- a/drivers/misc/habanalabs/common/command_submission.c
+++ b/drivers/misc/habanalabs/common/command_submission.c
@@ -252,6 +252,8 @@ static void cs_counters_aggregate(struct hl_device *hdev, struct hl_ctx *ctx)
ctx->cs_counters.parsing_drop_cnt;
hdev->aggregated_cs_counters.queue_full_drop_cnt +=
ctx->cs_counters.queue_full_drop_cnt;
+ hdev->aggregated_cs_counters.max_cs_in_flight_drop_cnt +=
+ ctx->cs_counters.max_cs_in_flight_drop_cnt;
}
static void cs_do_release(struct kref *ref)
@@ -431,8 +433,9 @@ static int allocate_cs(struct hl_device *hdev, struct hl_ctx *ctx,
(hdev->asic_prop.max_pending_cs - 1)];
if (other && !completion_done(&other->completion)) {
- dev_dbg(hdev->dev,
+ dev_dbg_ratelimited(hdev->dev,
"Rejecting CS because of too many in-flights CS\n");
+ ctx->cs_counters.max_cs_in_flight_drop_cnt++;
rc = -EAGAIN;
goto free_fence;
}
diff --git a/include/uapi/misc/habanalabs.h b/include/uapi/misc/habanalabs.h
index a2dcad29340f..69fb44d35292 100644
--- a/include/uapi/misc/habanalabs.h
+++ b/include/uapi/misc/habanalabs.h
@@ -401,12 +401,14 @@ struct hl_info_sync_manager {
* @parsing_drop_cnt: dropped due to error in packet parsing
* @queue_full_drop_cnt: dropped due to queue full
* @device_in_reset_drop_cnt: dropped due to device in reset
+ * @max_cs_in_flight_drop_cnt: dropped due to maximum CS in-flight
*/
struct hl_cs_counters {
__u64 out_of_mem_drop_cnt;
__u64 parsing_drop_cnt;
__u64 queue_full_drop_cnt;
__u64 device_in_reset_drop_cnt;
+ __u64 max_cs_in_flight_drop_cnt;
};
struct hl_info_cs_counters {
--
2.17.1
next reply other threads:[~2020-09-04 15:11 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-04 15:11 Oded Gabbay [this message]
2020-09-07 15:15 ` [PATCH] habanalabs: count dropped CS because max CS in-flight Tomer Tayar
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200904151119.24224-1-oded.gabbay@gmail.com \
--to=oded.gabbay@gmail.com \
--cc=SW_Drivers@habana.ai \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox