diff mbox series

[net-next,v3] eth: fbnic: Add PCIe hardware statistics

Message ID 20241108204640.3165724-1-sanman.p211993@gmail.com (mailing list archive)
State Superseded
Delegated to: Netdev Maintainers
Headers show
Series [net-next,v3] eth: fbnic: Add PCIe hardware statistics | expand

Checks

Context Check Description
netdev/series_format success Single patches do not need cover letters
netdev/tree_selection success Clearly marked for net-next
netdev/ynl success Generated files up to date; no warnings/errors; no diff in generated;
netdev/fixes_present success Fixes tag not required for -next series
netdev/header_inline success No static functions without inline keyword in header files
netdev/build_32bit success Errors and warnings before: 3 this patch: 3
netdev/build_tools success No tools touched, skip
netdev/cc_maintainers success CCed 13 of 13 maintainers
netdev/build_clang success Errors and warnings before: 4 this patch: 4
netdev/verify_signedoff success Signed-off-by tag matches author and committer
netdev/deprecated_api success None detected
netdev/check_selftest success No net selftest shell script
netdev/verify_fixes success No Fixes tag
netdev/build_allmodconfig_warn success Errors and warnings before: 8 this patch: 8
netdev/checkpatch warning WARNING: added, moved or deleted file(s), does MAINTAINERS need updating? WARNING: line length of 81 exceeds 80 columns WARNING: line length of 83 exceeds 80 columns WARNING: line length of 85 exceeds 80 columns WARNING: line length of 93 exceeds 80 columns WARNING: line length of 95 exceeds 80 columns WARNING: line length of 98 exceeds 80 columns
netdev/build_clang_rust success No Rust files in patch. Skipping build
netdev/kdoc success Errors and warnings before: 0 this patch: 0
netdev/source_inline success Was 0 now: 0
netdev/contest success net-next-2024-11-09--03-00 (tests: 784)

Commit Message

Sanman Pradhan Nov. 8, 2024, 8:46 p.m. UTC
Add PCIe hardware statistics support to the fbnic driver. These stats
provide insight into PCIe transaction performance and error conditions.

Which includes, read/write and completion TLP counts and DWORD counts and
debug counters for tag, completion credit and NP credit exhaustion

The stats are exposed via debugfs and can be used to monitor PCIe
performance and debug PCIe issues.

Signed-off-by: Sanman Pradhan <sanman.p211993@gmail.com>
---
 .../device_drivers/ethernet/meta/fbnic.rst    |  26 ++++
 drivers/net/ethernet/meta/fbnic/fbnic_csr.h   |  37 ++++++
 .../net/ethernet/meta/fbnic/fbnic_debugfs.c   |  46 +++++++
 .../net/ethernet/meta/fbnic/fbnic_hw_stats.c  | 114 ++++++++++++++++++
 .../net/ethernet/meta/fbnic/fbnic_hw_stats.h  |  12 ++
 .../net/ethernet/meta/fbnic/fbnic_netdev.c    |   3 +
 6 files changed, 238 insertions(+)
 create mode 100644 drivers/net/ethernet/meta/fbnic/fbnic_debugfs.c

--
2.43.5

Comments

Bagas Sanjaya Nov. 9, 2024, 3:30 a.m. UTC | #1
On Fri, Nov 08, 2024 at 12:46:40PM -0800, Sanman Pradhan wrote:
> diff --git a/Documentation/networking/device_drivers/ethernet/meta/fbnic.rst b/Documentation/networking/device_drivers/ethernet/meta/fbnic.rst
> index 32ff114f5c26..13ebcdbb5f22 100644
> --- a/Documentation/networking/device_drivers/ethernet/meta/fbnic.rst
> +++ b/Documentation/networking/device_drivers/ethernet/meta/fbnic.rst
> @@ -27,3 +27,29 @@ driver takes over.
>  devlink dev info provides version information for all three components. In
>  addition to the version the hg commit hash of the build is included as a
>  separate entry.
> +
> +PCIe Statistics
> +---------------
> +
> +The fbnic driver exposes PCIe hardware performance statistics through debugfs
> +(``pcie_stats``). These statistics provide insights into PCIe transaction
> +behavior and potential performance bottlenecks.
> +
> +Statistics Categories
> +
> +1. PCIe Transaction Counters:
> +
> +   These counters track PCIe transaction activity:
> +        - pcie_ob_rd_tlp: Outbound read Transaction Layer Packets count
> +        - pcie_ob_rd_dword: DWORDs transferred in outbound read transactions
> +        - pcie_ob_wr_tlp: Outbound write Transaction Layer Packets count
> +        - pcie_ob_wr_dword: DWORDs transferred in outbound write transactions
> +        - pcie_ob_cpl_tlp: Outbound completion TLP count
> +        - pcie_ob_cpl_dword: DWORDs transferred in outbound completion TLPs
> +
> +2. PCIe Resource Monitoring:
> +
> +   These counters indicate PCIe resource exhaustion events:
> +        - pcie_ob_rd_no_tag: Read requests dropped due to tag unavailability
> +        - pcie_ob_rd_no_cpl_cred: Read requests dropped due to completion credit exhaustion
> +        - pcie_ob_rd_no_np_cred: Read requests dropped due to non-posted credit exhaustion

The docs LGTM, thanks!

Reviewed-by: Bagas Sanjaya <bagasdotme@gmail.com>
Andrew Lunn Nov. 9, 2024, 5:55 p.m. UTC | #2
> +static void fbnic_hw_stat_rst64(struct fbnic_dev *fbd, u32 reg, s32 offset,
> +				struct fbnic_stat_counter *stat)
> +{
> +	/* Record initial counter values and compute deltas from there to ensure
> +	 * stats start at 0 after reboot/reset. This avoids exposing absolute
> +	 * hardware counter values to userspace.

Now you are in debugfs, this convention from ethtool -S no longer
applies. You could simply this code a lot by exposing the absolute
values.

	Andrew
diff mbox series

Patch

diff --git a/Documentation/networking/device_drivers/ethernet/meta/fbnic.rst b/Documentation/networking/device_drivers/ethernet/meta/fbnic.rst
index 32ff114f5c26..13ebcdbb5f22 100644
--- a/Documentation/networking/device_drivers/ethernet/meta/fbnic.rst
+++ b/Documentation/networking/device_drivers/ethernet/meta/fbnic.rst
@@ -27,3 +27,29 @@  driver takes over.
 devlink dev info provides version information for all three components. In
 addition to the version the hg commit hash of the build is included as a
 separate entry.
+
+PCIe Statistics
+---------------
+
+The fbnic driver exposes PCIe hardware performance statistics through debugfs
+(``pcie_stats``). These statistics provide insights into PCIe transaction
+behavior and potential performance bottlenecks.
+
+Statistics Categories
+
+1. PCIe Transaction Counters:
+
+   These counters track PCIe transaction activity:
+        - pcie_ob_rd_tlp: Outbound read Transaction Layer Packets count
+        - pcie_ob_rd_dword: DWORDs transferred in outbound read transactions
+        - pcie_ob_wr_tlp: Outbound write Transaction Layer Packets count
+        - pcie_ob_wr_dword: DWORDs transferred in outbound write transactions
+        - pcie_ob_cpl_tlp: Outbound completion TLP count
+        - pcie_ob_cpl_dword: DWORDs transferred in outbound completion TLPs
+
+2. PCIe Resource Monitoring:
+
+   These counters indicate PCIe resource exhaustion events:
+        - pcie_ob_rd_no_tag: Read requests dropped due to tag unavailability
+        - pcie_ob_rd_no_cpl_cred: Read requests dropped due to completion credit exhaustion
+        - pcie_ob_rd_no_np_cred: Read requests dropped due to non-posted credit exhaustion
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
index e78745332d82..463fd5d54d35 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_csr.h
@@ -917,6 +917,43 @@  enum {
 #define FBNIC_MAX_QUEUES		128
 #define FBNIC_CSR_END_QUEUE	(0x40000 + 0x400 * FBNIC_MAX_QUEUES - 1)

+/* PUL User Registers*/
+#define FBNIC_PUL_USER_OB_RD_TLP_CNT_31_0 \
+					0x3106e		/* 0xc41b8 */
+#define FBNIC_PUL_USER_OB_RD_DWORD_CNT_31_0 \
+					0x31070		/* 0xc41c0 */
+#define FBNIC_PUL_USER_OB_RD_DWORD_CNT_63_32 \
+					0x31071		/* 0xc41c4 */
+#define FBNIC_PUL_USER_OB_WR_TLP_CNT_31_0 \
+					0x31072		/* 0xc41c8 */
+#define FBNIC_PUL_USER_OB_WR_TLP_CNT_63_32 \
+					0x31073		/* 0xc41cc */
+#define FBNIC_PUL_USER_OB_WR_DWORD_CNT_31_0 \
+					0x31074		/* 0xc41d0 */
+#define FBNIC_PUL_USER_OB_WR_DWORD_CNT_63_32 \
+					0x31075		/* 0xc41d4 */
+#define FBNIC_PUL_USER_OB_CPL_TLP_CNT_31_0 \
+					0x31076		/* 0xc41d8 */
+#define FBNIC_PUL_USER_OB_CPL_TLP_CNT_63_32 \
+					0x31077		/* 0xc41dc */
+#define FBNIC_PUL_USER_OB_CPL_DWORD_CNT_31_0 \
+					0x31078		/* 0xc41e0 */
+#define FBNIC_PUL_USER_OB_CPL_DWORD_CNT_63_32 \
+					0x31079		/* 0xc41e4 */
+#define FBNIC_PUL_USER_OB_RD_DBG_CNT_CPL_CRED_31_0 \
+					0x3107a		/* 0xc41e8 */
+#define FBNIC_PUL_USER_OB_RD_DBG_CNT_CPL_CRED_63_32 \
+					0x3107b		/* 0xc41ec */
+#define FBNIC_PUL_USER_OB_RD_DBG_CNT_TAG_31_0 \
+					0x3107c		/* 0xc41f0 */
+#define FBNIC_PUL_USER_OB_RD_DBG_CNT_TAG_63_32 \
+					0x3107d		/* 0xc41f4 */
+#define FBNIC_PUL_USER_OB_RD_DBG_CNT_NP_CRED_31_0 \
+					0x3107e		/* 0xc41f8 */
+#define FBNIC_PUL_USER_OB_RD_DBG_CNT_NP_CRED_63_32 \
+					0x3107f		/* 0xc41fc */
+#define FBNIC_CSR_END_PUL_USER	0x31080	/* CSR section delimiter */
+
 /* BAR 4 CSRs */

 /* The IPC mailbox consists of 32 mailboxes, with each mailbox consisting
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_debugfs.c b/drivers/net/ethernet/meta/fbnic/fbnic_debugfs.c
new file mode 100644
index 000000000000..38736c4b42d0
--- /dev/null
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_debugfs.c
@@ -0,0 +1,46 @@ 
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (c) Meta Platforms, Inc. and affiliates. */
+
+#include <linux/debugfs.h>
+#include <linux/pci.h>
+#include <linux/rtnetlink.h>
+#include <linux/seq_file.h>
+#include "fbnic.h"
+
+static struct dentry *fbnic_dbg_root;
+
+static int fbnic_dbg_pcie_stats_show(struct seq_file *s, void *v)
+{
+	struct fbnic_dev *fbd = s->private;
+
+	rtnl_lock();
+
+	fbnic_get_hw_stats(fbd);
+
+	seq_printf(s, "ob_rd_tlp: %llu\n", fbd->hw_stats.pcie.ob_rd_tlp.value);
+	seq_printf(s, "ob_rd_dword: %llu\n", fbd->hw_stats.pcie.ob_rd_dword.value);
+	seq_printf(s, "ob_wr_tlp: %llu\n", fbd->hw_stats.pcie.ob_wr_tlp.value);
+	seq_printf(s, "ob_wr_dword: %llu\n", fbd->hw_stats.pcie.ob_wr_dword.value);
+	seq_printf(s, "ob_cpl_tlp: %llu\n", fbd->hw_stats.pcie.ob_cpl_tlp.value);
+	seq_printf(s, "ob_cpl_dword: %llu\n", fbd->hw_stats.pcie.ob_cpl_dword.value);
+	seq_printf(s, "ob_rd_no_tag: %llu\n", fbd->hw_stats.pcie.ob_rd_no_tag.value);
+	seq_printf(s, "ob_rd_no_cpl_cred: %llu\n", fbd->hw_stats.pcie.ob_rd_no_cpl_cred.value);
+	seq_printf(s, "ob_rd_no_np_cred: %llu\n", fbd->hw_stats.pcie.ob_rd_no_np_cred.value);
+
+	rtnl_unlock();
+	return 0;
+}
+
+DEFINE_SHOW_ATTRIBUTE(fbnic_dbg_pcie_stats);
+
+void fbnic_dbg_init(void)
+{
+	fbnic_dbg_root = debugfs_create_dir(fbnic_driver_name, NULL);
+	debugfs_create_file("pcie_stats", 0400, fbnic_dbg_root, NULL, &fbnic_dbg_pcie_stats_fops);
+}
+
+void fbnic_dbg_exit(void)
+{
+	debugfs_remove_recursive(fbnic_dbg_root);
+	fbnic_dbg_root = NULL;
+}
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_hw_stats.c b/drivers/net/ethernet/meta/fbnic/fbnic_hw_stats.c
index a0acc7606aa1..eb19b49fe306 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_hw_stats.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_hw_stats.c
@@ -25,3 +25,117 @@  u64 fbnic_stat_rd64(struct fbnic_dev *fbd, u32 reg, u32 offset)
 	 */
 	return ((u64)upper << 32);
 }
+
+static void fbnic_hw_stat_rst64(struct fbnic_dev *fbd, u32 reg, s32 offset,
+				struct fbnic_stat_counter *stat)
+{
+	/* Record initial counter values and compute deltas from there to ensure
+	 * stats start at 0 after reboot/reset. This avoids exposing absolute
+	 * hardware counter values to userspace.
+	 */
+	stat->u.old_reg_value_64 = fbnic_stat_rd64(fbd, reg, offset);
+}
+
+static void fbnic_hw_stat_rd64(struct fbnic_dev *fbd, u32 reg, s32 offset,
+			       struct fbnic_stat_counter *stat)
+{
+	u64 new_reg_value;
+
+	new_reg_value = fbnic_stat_rd64(fbd, reg, offset);
+	stat->value += new_reg_value - stat->u.old_reg_value_64;
+	stat->u.old_reg_value_64 = new_reg_value;
+}
+
+static void fbnic_reset_pcie_stats_asic(struct fbnic_dev *fbd,
+					struct fbnic_pcie_stats *pcie)
+{
+	fbnic_hw_stat_rst64(fbd,
+			    FBNIC_PUL_USER_OB_RD_TLP_CNT_31_0,
+			    1,
+			    &pcie->ob_rd_tlp);
+	fbnic_hw_stat_rst64(fbd,
+			    FBNIC_PUL_USER_OB_RD_DWORD_CNT_31_0,
+			    1,
+			    &pcie->ob_rd_dword);
+	fbnic_hw_stat_rst64(fbd,
+			    FBNIC_PUL_USER_OB_CPL_TLP_CNT_31_0,
+			    1,
+			    &pcie->ob_cpl_tlp);
+	fbnic_hw_stat_rst64(fbd,
+			    FBNIC_PUL_USER_OB_CPL_DWORD_CNT_31_0,
+			    1,
+			    &pcie->ob_cpl_dword);
+	fbnic_hw_stat_rst64(fbd,
+			    FBNIC_PUL_USER_OB_WR_TLP_CNT_31_0,
+			    1,
+			    &pcie->ob_wr_tlp);
+	fbnic_hw_stat_rst64(fbd,
+			    FBNIC_PUL_USER_OB_WR_DWORD_CNT_31_0,
+			    1,
+			    &pcie->ob_wr_dword);
+
+	fbnic_hw_stat_rst64(fbd,
+			    FBNIC_PUL_USER_OB_RD_DBG_CNT_TAG_31_0,
+			    1,
+			    &pcie->ob_rd_no_tag);
+	fbnic_hw_stat_rst64(fbd,
+			    FBNIC_PUL_USER_OB_RD_DBG_CNT_CPL_CRED_31_0,
+			    1,
+			    &pcie->ob_rd_no_cpl_cred);
+	fbnic_hw_stat_rst64(fbd,
+			    FBNIC_PUL_USER_OB_RD_DBG_CNT_NP_CRED_31_0,
+			    1,
+			    &pcie->ob_rd_no_np_cred);
+}
+
+static void fbnic_get_pcie_stats_asic64(struct fbnic_dev *fbd,
+					struct fbnic_pcie_stats *pcie)
+{
+	fbnic_hw_stat_rd64(fbd,
+			   FBNIC_PUL_USER_OB_RD_TLP_CNT_31_0,
+			   1,
+			   &pcie->ob_rd_tlp);
+	fbnic_hw_stat_rd64(fbd,
+			   FBNIC_PUL_USER_OB_RD_DWORD_CNT_31_0,
+			   1,
+			   &pcie->ob_rd_dword);
+	fbnic_hw_stat_rd64(fbd,
+			   FBNIC_PUL_USER_OB_WR_TLP_CNT_31_0,
+			   1,
+			   &pcie->ob_wr_tlp);
+	fbnic_hw_stat_rd64(fbd,
+			   FBNIC_PUL_USER_OB_WR_DWORD_CNT_31_0,
+			   1,
+			   &pcie->ob_wr_dword);
+	fbnic_hw_stat_rd64(fbd,
+			   FBNIC_PUL_USER_OB_CPL_TLP_CNT_31_0,
+			   1,
+			   &pcie->ob_cpl_tlp);
+	fbnic_hw_stat_rd64(fbd,
+			   FBNIC_PUL_USER_OB_CPL_DWORD_CNT_31_0,
+			   1,
+			   &pcie->ob_cpl_dword);
+
+	fbnic_hw_stat_rd64(fbd,
+			   FBNIC_PUL_USER_OB_RD_DBG_CNT_TAG_31_0,
+			   1,
+			   &pcie->ob_rd_no_tag);
+	fbnic_hw_stat_rd64(fbd,
+			   FBNIC_PUL_USER_OB_RD_DBG_CNT_CPL_CRED_31_0,
+			   1,
+			   &pcie->ob_rd_no_cpl_cred);
+	fbnic_hw_stat_rd64(fbd,
+			   FBNIC_PUL_USER_OB_RD_DBG_CNT_NP_CRED_31_0,
+			   1,
+			   &pcie->ob_rd_no_np_cred);
+}
+
+void fbnic_reset_hw_stats(struct fbnic_dev *fbd)
+{
+	fbnic_reset_pcie_stats_asic(fbd, &fbd->hw_stats.pcie);
+}
+
+void fbnic_get_hw_stats(struct fbnic_dev *fbd)
+{
+	fbnic_get_pcie_stats_asic64(fbd, &fbd->hw_stats.pcie);
+}
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_hw_stats.h b/drivers/net/ethernet/meta/fbnic/fbnic_hw_stats.h
index 30348904b510..036cc065a857 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_hw_stats.h
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_hw_stats.h
@@ -31,10 +31,22 @@  struct fbnic_mac_stats {
 	struct fbnic_eth_mac_stats eth_mac;
 };

+struct fbnic_pcie_stats {
+	struct fbnic_stat_counter ob_rd_tlp, ob_rd_dword;
+	struct fbnic_stat_counter ob_wr_tlp, ob_wr_dword;
+	struct fbnic_stat_counter ob_cpl_tlp, ob_cpl_dword;
+
+	struct fbnic_stat_counter ob_rd_no_tag;
+	struct fbnic_stat_counter ob_rd_no_cpl_cred;
+	struct fbnic_stat_counter ob_rd_no_np_cred;
+};
+
 struct fbnic_hw_stats {
 	struct fbnic_mac_stats mac;
+	struct fbnic_pcie_stats pcie;
 };

 u64 fbnic_stat_rd64(struct fbnic_dev *fbd, u32 reg, u32 offset);

+void fbnic_reset_hw_stats(struct fbnic_dev *fbd);
 void fbnic_get_hw_stats(struct fbnic_dev *fbd);
diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
index fc7d80db5fa6..04077649161b 100644
--- a/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
+++ b/drivers/net/ethernet/meta/fbnic/fbnic_netdev.c
@@ -628,6 +628,9 @@  struct net_device *fbnic_netdev_alloc(struct fbnic_dev *fbd)

 	fbnic_reset_queues(fbn, default_queues, default_queues);

+	/* Capture snapshot of hardware stats so netdev can calculate delta */
+	fbnic_reset_hw_stats(fbd);
+
 	fbnic_reset_indir_tbl(fbn);
 	fbnic_rss_key_fill(fbn->rss_key);
 	fbnic_rss_init_en_mask(fbn);