Skip to contentSkip to conversation

iomap: trace iomap_appply results

PATCH In review Darrick J. Wong 2 participants 2 messages Nov 22, 2019

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 iomap: trace iomap_appply results

This is basically a debugging patch that I've been using to monitor what exactly xfs is sending back to iomap, and thought I ought to pitch it to the mailing list universe to see what reaction I get.

--D

---
From: Darrick J. Wong <[email protected]>

Add some tracepoints so that we can more easily debug what the filesystem is returning from ->iomap_begin.

---
 fs/iomap/apply.c |    7 ++++
 fs/iomap/trace.h |  103 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 110 insertions(+)
fs/iomap/apply.c +7 -0
index 484dd8eda861..76925b40b5fd 100644
@@ -7,6 +7,7 @@
 #include <linux/compiler.h>
 #include <linux/fs.h>
 #include <linux/iomap.h>
+#include "trace.h"
 
 /*
  * Execute a iomap write on a segment of the mapping that spans a
@@ -28,6 +29,8 @@ iomap_apply(struct inode *inode, loff_t pos, loff_t length, unsigned flags,
 	loff_t written = 0, ret;
 	u64 end;
 
+	trace_iomap_apply(inode, pos, length, flags, ops, actor, _RET_IP_);
+
 	/*
 	 * Need to map a range from start position for length bytes. This can
 	 * span multiple pages - it is only guaranteed to return a range of a
@@ -48,6 +51,10 @@ iomap_apply(struct inode *inode, loff_t pos, loff_t length, unsigned flags,
 	if (WARN_ON(iomap.length == 0))
 		return -EIO;
 
+	trace_iomap_apply_dstmap(inode, &iomap);
+	if (srcmap.type != IOMAP_HOLE)
+		trace_iomap_apply_srcmap(inode, &srcmap);
+
 	/*
 	 * Cut down the length to the one actually provided by the filesystem,
 	 * as it might not be able to give us the whole size that we requested.
fs/iomap/trace.h +103 -0

109 lines not on this page · raw message