2018-12-17 08:09:16

by peng yu

[permalink] [raw]
Subject: [PATCH v3 1/2] share nvme trace event helper functions with other modules

Currently the trace.c/trace.h could only be used by nvme-core. This
patch moves __assign_disk_name to genhd.h, exports nvme_trace_disk_name to
other modules. Thus other nvme modules could also trace disk name.

Signed-off-by: yupeng <[email protected]>
---
drivers/nvme/host/trace.c | 2 ++
drivers/nvme/host/trace.h | 10 ----------
include/linux/genhd.h | 8 ++++++++
3 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/nvme/host/trace.c b/drivers/nvme/host/trace.c
index 25b0e310f4a8..99a132a34403 100644
--- a/drivers/nvme/host/trace.c
+++ b/drivers/nvme/host/trace.c
@@ -139,3 +139,5 @@ const char *nvme_trace_disk_name(struct trace_seq *p, char *name)

return ret;
}
+
+EXPORT_SYMBOL_GPL(nvme_trace_disk_name);
diff --git a/drivers/nvme/host/trace.h b/drivers/nvme/host/trace.h
index 196d5bd56718..23f3fcf04fb6 100644
--- a/drivers/nvme/host/trace.h
+++ b/drivers/nvme/host/trace.h
@@ -82,16 +82,6 @@ const char *nvme_trace_disk_name(struct trace_seq *p, char *name);
#define __print_disk_name(name) \
nvme_trace_disk_name(p, name)

-#ifndef TRACE_HEADER_MULTI_READ
-static inline void __assign_disk_name(char *name, struct gendisk *disk)
-{
- if (disk)
- memcpy(name, disk->disk_name, DISK_NAME_LEN);
- else
- memset(name, 0, DISK_NAME_LEN);
-}
-#endif
-
TRACE_EVENT(nvme_setup_cmd,
TP_PROTO(struct request *req, struct nvme_command *cmd),
TP_ARGS(req, cmd),
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 70fc838e6773..468ab3143d77 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -749,6 +749,14 @@ static inline void blk_integrity_add(struct gendisk *disk) { }
static inline void blk_integrity_del(struct gendisk *disk) { }
#endif /* CONFIG_BLK_DEV_INTEGRITY */

+static inline void __assign_disk_name(char *name, struct gendisk *disk)
+{
+ if (disk)
+ memcpy(name, disk->disk_name, DISK_NAME_LEN);
+ else
+ memset(name, 0, DISK_NAME_LEN);
+}
+
#else /* CONFIG_BLOCK */

static inline void printk_all_partitions(void) { }
--
2.17.1



2018-12-17 08:20:31

by peng yu

[permalink] [raw]
Subject: [PATCH v3 2/2] trace nvme submit queue status

export nvme disk name, queue id, sq_head, sq_tail to trace event
usage example:
go to the event directory:
cd /sys/kernel/debug/tracing/events/nvme/nvme_sq
filter by disk name:
echo 'disk=="nvme1n1"' > filter
enable the event:
echo 1 > enable
check results from trace_pipe:
cat /sys/kernel/debug/tracing/trace_pipe
In practice, this patch help me debug hardware related
performant issue.

Signed-off-by: yupeng <[email protected]>
---
drivers/nvme/host/pci.c | 7 +++++
drivers/nvme/host/trace_pci.h | 57 +++++++++++++++++++++++++++++++++++
2 files changed, 64 insertions(+)
create mode 100644 drivers/nvme/host/trace_pci.h

diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index c33bb201b884..974cb05b3592 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -32,6 +32,9 @@
#include <linux/sed-opal.h>
#include <linux/pci-p2pdma.h>

+#define CREATE_TRACE_POINTS
+#include "trace_pci.h"
+
#include "nvme.h"

#define SQ_SIZE(depth) (depth * sizeof(struct nvme_command))
@@ -899,6 +902,10 @@ static inline void nvme_handle_cqe(struct nvme_queue *nvmeq, u16 idx)
}

req = blk_mq_tag_to_rq(*nvmeq->tags, cqe->command_id);
+ trace_nvme_sq(req->rq_disk,
+ nvmeq->qid,
+ le16_to_cpu(cqe->sq_head),
+ nvmeq->sq_tail);
nvme_end_request(req, cqe->status, cqe->result);
}

diff --git a/drivers/nvme/host/trace_pci.h b/drivers/nvme/host/trace_pci.h
new file mode 100644
index 000000000000..70fa853397f6
--- /dev/null
+++ b/drivers/nvme/host/trace_pci.h
@@ -0,0 +1,57 @@
+/*
+ * NVM Express device driver tracepoints
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM nvme
+
+#if !defined(_TRACE_NVME_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_NVME_H
+
+#include <linux/nvme.h>
+#include <linux/tracepoint.h>
+#include <linux/trace_seq.h>
+
+#include "nvme.h"
+
+const char *nvme_trace_disk_name(struct trace_seq *p, char *name);
+#define __print_disk_name(name) \
+ nvme_trace_disk_name(p, name)
+
+TRACE_EVENT(nvme_sq,
+ TP_PROTO(void *rq_disk, int qid, int sq_head, int sq_tail),
+ TP_ARGS(rq_disk, qid, sq_head, sq_tail),
+ TP_STRUCT__entry(
+ __array(char, disk, DISK_NAME_LEN)
+ __field(int, qid)
+ __field(int, sq_head)
+ __field(int, sq_tail)),
+ TP_fast_assign(
+ __entry->qid = qid;
+ __entry->sq_head = sq_head;
+ __entry->sq_tail = sq_tail;
+ __assign_disk_name(__entry->disk, rq_disk);
+ ),
+ TP_printk("nvme: %s qid=%d head=%d tail=%d",
+ __print_disk_name(__entry->disk),
+ __entry->qid, __entry->sq_head, __entry->sq_tail)
+);
+
+#endif /* _TRACE_NVME_H */
+
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE trace_pci
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
--
2.17.1


2018-12-17 08:34:57

by Johannes Thumshirn

[permalink] [raw]
Subject: Re: [PATCH v3 2/2] trace nvme submit queue status

Is the addition of a trace_pci.h really needed? Why can't you just put
it in trace.h?

Byte,
Johannes
--
Johannes Thumshirn SUSE Labs Filesystems
[email protected] +49 911 74053 689
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)
Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850

2018-12-17 11:14:33

by Christoph Hellwig

[permalink] [raw]
Subject: Re: [PATCH v3 2/2] trace nvme submit queue status

On Mon, Dec 17, 2018 at 09:33:24AM +0100, Johannes Thumshirn wrote:
> Is the addition of a trace_pci.h really needed? Why can't you just put
> it in trace.h?

Agreed. Especially as the concepts actually are generic, so we could
actually add the trace point to the rdma and fc transports as well
(not that they would be all that useful, so we probably shouldn't).