block: introduce block_io_start/block_io_done tracepoints
authorHengqi Chen <hengqi.chen@gmail.com>
Sat, 20 May 2023 08:40:57 +0000 (08:40 +0000)
committerJens Axboe <axboe@kernel.dk>
Wed, 24 May 2023 14:38:59 +0000 (08:38 -0600)
Currently, several BCC ([0]) tools (biosnoop/biostacks/biotop) use
kprobes to blk_account_io_start/blk_account_io_done to implement
their functionalities. This is fragile because the target kernel
functions may be renamed ([1]) or inlined ([2]). So introduce two
new tracepoints for such use cases.

  [0]: https://github.com/iovisor/bcc
  [1]: https://github.com/iovisor/bcc/issues/3954
  [2]: https://github.com/iovisor/bcc/issues/4261

Tested-by: Francis Laniel <flaniel@linux.microsoft.com>
Signed-off-by: Hengqi Chen <hengqi.chen@gmail.com>
Tested-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/r/20230520084057.1467003-1-hengqi.chen@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
block/blk-mq.c
include/trace/events/block.h

index 551e776..1749f58 100644 (file)
@@ -962,6 +962,8 @@ EXPORT_SYMBOL_GPL(blk_update_request);
 
 static inline void blk_account_io_done(struct request *req, u64 now)
 {
+       trace_block_io_done(req);
+
        /*
         * Account IO completion.  flush_rq isn't accounted as a
         * normal IO on queueing nor completion.  Accounting the
@@ -981,6 +983,8 @@ static inline void blk_account_io_done(struct request *req, u64 now)
 
 static inline void blk_account_io_start(struct request *req)
 {
+       trace_block_io_start(req);
+
        if (blk_do_io_stat(req)) {
                /*
                 * All non-passthrough requests are created from a bio with one
index 7f4dfbd..40e60c3 100644 (file)
@@ -246,6 +246,32 @@ DEFINE_EVENT(block_rq, block_rq_merge,
 );
 
 /**
+ * block_io_start - insert a request for execution
+ * @rq: block IO operation request
+ *
+ * Called when block operation request @rq is queued for execution
+ */
+DEFINE_EVENT(block_rq, block_io_start,
+
+       TP_PROTO(struct request *rq),
+
+       TP_ARGS(rq)
+);
+
+/**
+ * block_io_done - block IO operation request completed
+ * @rq: block IO operation request
+ *
+ * Called when block operation request @rq is completed
+ */
+DEFINE_EVENT(block_rq, block_io_done,
+
+       TP_PROTO(struct request *rq),
+
+       TP_ARGS(rq)
+);
+
+/**
  * block_bio_complete - completed all work on the block operation
  * @q: queue holding the block operation
  * @bio: block operation completed