Re: [PATCH v3] iomap: Remove pgoff from tracepoints

Re: [PATCH v3] iomap: Remove pgoff from tracepoints

[Matthew Wilcox]

From: Matthew Wilcox (Oracle) <willy@infradead.org>

The 'pgoff' displayed by the tracepoints wasn't a pgoff at all; it
was a byte offset from the start of the file.  We already emit that in
the form of the 'offset', so we can just remove pgoff.  That means we
can remove 'page' as an argument to the tracepoint, and rename this
type of tracepoint from being a page class to being a range class.

Fixes: 0b1b213fcf3a ("xfs: event tracing support")
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>

---
v2: Get rid of 'pgoff' instead of fixing it
v3: Fix releasepage and writepage to actually set offset/length.

 buffered-io.c |    7 ++++---
 trace.h       |   27 +++++++++++----------------
 2 files changed, 15 insertions(+), 19 deletions(-)

diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c
index 7057ef155a29..25ec29157ec8 100644
--- a/fs/iomap/buffered-io.c
+++ b/fs/iomap/buffered-io.c
@@ -487,7 +487,8 @@ EXPORT_SYMBOL_GPL(iomap_is_partially_uptodate);
 int
 iomap_releasepage(struct page *page, gfp_t gfp_mask)
 {
-	trace_iomap_releasepage(page->mapping->host, page, 0, 0);
+	trace_iomap_releasepage(page->mapping->host, page_offset(page),
+			PAGE_SIZE);
 
 	/*
 	 * mm accommodates an old ext3 case where clean pages might not have had
@@ -504,7 +505,7 @@ EXPORT_SYMBOL_GPL(iomap_releasepage);
 void
 iomap_invalidatepage(struct page *page, unsigned int offset, unsigned int len)
 {
-	trace_iomap_invalidatepage(page->mapping->host, page, offset, len);
+	trace_iomap_invalidatepage(page->mapping->host, offset, len);
 
 	/*
 	 * If we are invalidating the entire page, clear the dirty state from it
@@ -1503,7 +1504,7 @@ iomap_do_writepage(struct page *page, struct writeback_control *wbc, void *data)
 	u64 end_offset;
 	loff_t offset;
 
-	trace_iomap_writepage(inode, page, 0, 0);
+	trace_iomap_writepage(inode, page_offset(page), PAGE_SIZE);
 
 	/*
 	 * Refuse to write the page out if we are called from reclaim context.
diff --git a/fs/iomap/trace.h b/fs/iomap/trace.h
index d6ba705f938a..5693a39d52fb 100644
--- a/fs/iomap/trace.h
+++ b/fs/iomap/trace.h
@@ -41,14 +41,12 @@ DEFINE_EVENT(iomap_readpage_class, name,	\
 DEFINE_READPAGE_EVENT(iomap_readpage);
 DEFINE_READPAGE_EVENT(iomap_readahead);
 
-DECLARE_EVENT_CLASS(iomap_page_class,
-	TP_PROTO(struct inode *inode, struct page *page, unsigned long off,
-		 unsigned int len),
-	TP_ARGS(inode, page, off, len),
+DECLARE_EVENT_CLASS(iomap_range_class,
+	TP_PROTO(struct inode *inode, unsigned long off, unsigned int len),
+	TP_ARGS(inode, off, len),
 	TP_STRUCT__entry(
 		__field(dev_t, dev)
 		__field(u64, ino)
-		__field(pgoff_t, pgoff)
 		__field(loff_t, size)
 		__field(unsigned long, offset)
 		__field(unsigned int, length)
@@ -56,29 +54,26 @@ DECLARE_EVENT_CLASS(iomap_page_class,
 	TP_fast_assign(
 		__entry->dev = inode->i_sb->s_dev;
 		__entry->ino = inode->i_ino;
-		__entry->pgoff = page_offset(page);
 		__entry->size = i_size_read(inode);
 		__entry->offset = off;
 		__entry->length = len;
 	),
-	TP_printk("dev %d:%d ino 0x%llx pgoff 0x%lx size 0x%llx offset %lx "
+	TP_printk("dev %d:%d ino 0x%llx size 0x%llx offset %lx "
 		  "length %x",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		  __entry->ino,
-		  __entry->pgoff,
 		  __entry->size,
 		  __entry->offset,
 		  __entry->length)
 )
 
-#define DEFINE_PAGE_EVENT(name)		\
-DEFINE_EVENT(iomap_page_class, name,	\
-	TP_PROTO(struct inode *inode, struct page *page, unsigned long off, \
-		 unsigned int len),	\
-	TP_ARGS(inode, page, off, len))
-DEFINE_PAGE_EVENT(iomap_writepage);
-DEFINE_PAGE_EVENT(iomap_releasepage);
-DEFINE_PAGE_EVENT(iomap_invalidatepage);
+#define DEFINE_RANGE_EVENT(name)		\
+DEFINE_EVENT(iomap_range_class, name,	\
+	TP_PROTO(struct inode *inode, unsigned long off, unsigned int len),\
+	TP_ARGS(inode, off, len))
+DEFINE_RANGE_EVENT(iomap_writepage);
+DEFINE_RANGE_EVENT(iomap_releasepage);
+DEFINE_RANGE_EVENT(iomap_invalidatepage);
 
 #define IOMAP_TYPE_STRINGS \
 	{ IOMAP_HOLE,		"HOLE" }, \

Re: [PATCH v3] iomap: Remove pgoff from tracepoints

[Dave Chinner]

On Wed, Mar 04, 2020 at 09:54:29AM -0800, Matthew Wilcox wrote:
> From: Matthew Wilcox (Oracle) <willy@infradead.org>
> 
> The 'pgoff' displayed by the tracepoints wasn't a pgoff at all; it
> was a byte offset from the start of the file.  We already emit that in
> the form of the 'offset', so we can just remove pgoff.  That means we
> can remove 'page' as an argument to the tracepoint, and rename this
> type of tracepoint from being a page class to being a range class.
> 
> Fixes: 0b1b213fcf3a ("xfs: event tracing support")
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
> 
> ---
> v2: Get rid of 'pgoff' instead of fixing it
> v3: Fix releasepage and writepage to actually set offset/length.
> 
>  buffered-io.c |    7 ++++---
>  trace.h       |   27 +++++++++++----------------
>  2 files changed, 15 insertions(+), 19 deletions(-)

With the range updates for release/writeage it looks good to me.

Reviewed-by: Dave Chinner <dchinner@redhat.com>
-- 
Dave Chinner
david@fromorbit.com

Re: [PATCH v3] iomap: Remove pgoff from tracepoints

[Christoph Hellwig]

On Wed, Mar 04, 2020 at 09:54:29AM -0800, Matthew Wilcox wrote:
> From: Matthew Wilcox (Oracle) <willy@infradead.org>
> 
> The 'pgoff' displayed by the tracepoints wasn't a pgoff at all; it
> was a byte offset from the start of the file.  We already emit that in
> the form of the 'offset', so we can just remove pgoff.  That means we
> can remove 'page' as an argument to the tracepoint, and rename this
> type of tracepoint from being a page class to being a range class.
> 
> Fixes: 0b1b213fcf3a ("xfs: event tracing support")
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>

Looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>