LKML Archive mirror
 help / color / mirror / Atom feed
From: Oded Gabbay <ogabbay@kernel.org>
To: linux-kernel@vger.kernel.org
Cc: Ohad Sharabi <osharabi@habana.ai>
Subject: [PATCH 3/7] habanalabs: report EQ fault during heartbeat
Date: Wed,  9 Jun 2021 18:03:39 +0300	[thread overview]
Message-ID: <20210609150343.4712-3-ogabbay@kernel.org> (raw)
In-Reply-To: <20210609150343.4712-1-ogabbay@kernel.org>

From: Ohad Sharabi <osharabi@habana.ai>

In case we have EQ fault we would like to know about it.
For this, a status bitmask was added in which EQ_FAULT bit is
set by FW in case of EQ fault.

Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
---
 drivers/misc/habanalabs/common/firmware_if.c  |  8 +++++++-
 .../misc/habanalabs/include/common/cpucp_if.h | 20 +++++++++++++++++++
 2 files changed, 27 insertions(+), 1 deletion(-)

diff --git a/drivers/misc/habanalabs/common/firmware_if.c b/drivers/misc/habanalabs/common/firmware_if.c
index 9412e6707906..d5a3c786d4c9 100644
--- a/drivers/misc/habanalabs/common/firmware_if.c
+++ b/drivers/misc/habanalabs/common/firmware_if.c
@@ -362,7 +362,7 @@ void hl_fw_cpu_accessible_dma_pool_free(struct hl_device *hdev, size_t size,
 
 int hl_fw_send_heartbeat(struct hl_device *hdev)
 {
-	struct cpucp_packet hb_pkt = {};
+	struct cpucp_packet hb_pkt = {0};
 	u64 result;
 	int rc;
 
@@ -374,7 +374,13 @@ int hl_fw_send_heartbeat(struct hl_device *hdev)
 						sizeof(hb_pkt), 0, &result);
 
 	if ((rc) || (result != CPUCP_PACKET_FENCE_VAL))
+		return -EIO;
+
+	if (le32_to_cpu(hb_pkt.status_mask) &
+					CPUCP_PKT_HB_STATUS_EQ_FAULT_MASK) {
+		dev_warn(hdev->dev, "FW reported EQ fault during heartbeat\n");
 		rc = -EIO;
+	}
 
 	return rc;
 }
diff --git a/drivers/misc/habanalabs/include/common/cpucp_if.h b/drivers/misc/habanalabs/include/common/cpucp_if.h
index d4dc189a6c92..80b1d5a9d9f1 100644
--- a/drivers/misc/habanalabs/include/common/cpucp_if.h
+++ b/drivers/misc/habanalabs/include/common/cpucp_if.h
@@ -404,6 +404,20 @@ enum cpucp_packet_id {
 #define CPUCP_PKT_RES_PLL_OUT3_SHIFT	48
 #define CPUCP_PKT_RES_PLL_OUT3_MASK	0xFFFF000000000000ull
 
+#define CPUCP_PKT_VAL_PFC_IN1_SHIFT	0
+#define CPUCP_PKT_VAL_PFC_IN1_MASK	0x0000000000000001ull
+#define CPUCP_PKT_VAL_PFC_IN2_SHIFT	1
+#define CPUCP_PKT_VAL_PFC_IN2_MASK	0x000000000000001Eull
+
+#define CPUCP_PKT_VAL_LPBK_IN1_SHIFT	0
+#define CPUCP_PKT_VAL_LPBK_IN1_MASK	0x0000000000000001ull
+#define CPUCP_PKT_VAL_LPBK_IN2_SHIFT	1
+#define CPUCP_PKT_VAL_LPBK_IN2_MASK	0x000000000000001Eull
+
+/* heartbeat status bits */
+#define CPUCP_PKT_HB_STATUS_EQ_FAULT_SHIFT		0
+#define CPUCP_PKT_HB_STATUS_EQ_FAULT_MASK		0x00000001
+
 struct cpucp_packet {
 	union {
 		__le64 value;	/* For SET packets */
@@ -445,6 +459,12 @@ struct cpucp_packet {
 
 		/* For get CpuCP info/EEPROM data/NIC info */
 		__le32 data_max_size;
+
+		/*
+		 * For any general status bitmask. Shall be used whenever the
+		 * result cannot be used to hold general purpose data.
+		 */
+		__le32 status_mask;
 	};
 
 	__le32 reserved;
-- 
2.25.1


  parent reply	other threads:[~2021-06-09 15:04 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-09 15:03 [PATCH 1/7] habanalabs/gaudi: use standard error codes Oded Gabbay
2021-06-09 15:03 ` [PATCH 2/7] habanalabs: small code refactoring Oded Gabbay
2021-06-09 15:03 ` Oded Gabbay [this message]
2021-06-09 15:03 ` [PATCH 4/7] habanalabs: enable stop on error for all QMANs and engines Oded Gabbay
2021-06-09 15:03 ` [PATCH 5/7] habanalabs: enable dram scramble before linux f/w Oded Gabbay
2021-06-09 15:03 ` [PATCH 6/7] habanalabs: add hard reset timeout for PLDM Oded Gabbay
2021-06-09 15:03 ` [PATCH 7/7] habanalabs: print firmware versions Oded Gabbay

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=20210609150343.4712-3-ogabbay@kernel.org \
    --to=ogabbay@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=osharabi@habana.ai \
    /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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).