Re: [PATCH 02/10] perf tool: fix: Don't discard prev in backward mode

From: Wangnan (F)
Date: Tue Oct 10 2017 - 14:30:21 EST




On 2017/10/11 1:20, kan.liang@xxxxxxxxx wrote:
From: Kan Liang <kan.liang@xxxxxxxxx>

Perf record can switch output. The new output should only store the data
after switching. However, in overwrite backward mode, the new output
still have the data from old output.

At the end of mmap_read, the position of processed ring buffer is saved
in md->prev. Next mmap_read should be end in md->prev.
However, the md->prev is discarded. So next mmap_read has to process
whole valid ring buffer, which definitely include the old processed
data.

Set the prev as the end of the range in backward mode.

Signed-off-by: Kan Liang <kan.liang@xxxxxxxxx>
---
tools/perf/util/evlist.c | 14 +++++++++++++-
1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index 33b8837..7d23cf5 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -742,13 +742,25 @@ static int
rb_find_range(void *data, int mask, u64 head, u64 old,
u64 *start, u64 *end, bool backward)
{
+ int ret;
+
if (!backward) {
*start = old;
*end = head;
return 0;
}
- return backward_rb_find_range(data, mask, head, start, end);
+ ret = backward_rb_find_range(data, mask, head, start, end);
+
+ /*
+ * The start and end from backward_rb_find_range is the range for all
+ * valid data in ring buffer.
+ * However, part of the data is processed previously.
+ * Reset the end to drop the processed data
+ */
+ *end = old;
+

I don't understand this patch. What rb_find_range() wants to do is to find start
and end pointer, so record__mmap_read() knows where to start reading and where to stop.
In backward mode, 'start' pointer is clearly 'head' (because 'head' is the header of the
last record kernel writes to the ring buffer), but the 'end' pointer is not very easy to
find ('end' pointer is the last (the earliest) available record in the ring buffer).
We have to parse the whole ring buffer to find the last available record and set its
tail into 'end', this is what backward_rb_find_range() tries hard to do. However,
your patch unconditionally overwrites *end (which is set by backward_rb_find_range()),
makes backward_rb_find_range() meaningless.

In my mind when you decide to use backward ring buffer you must make it overwrite
because you want the kernel silently overwrite old records without waking up perf, while
still able to parse the ring buffer even if overwriting happens. The use case
should be: perf runs in background, kernel silently ignores old records in the ring
buffer until something bad happens, then perf dumps 'events before the bad thing'.
Similar to fight recorder. In this use case, each dumping is independent. Even
if one record is appears in previous dumping, it is harmless (and useful) to make
it appears in following dumping again.

If you really want to avoid record duplication, you need to changes record__mmap_read()'s
logic. Now it complains "failed to keep up with mmap data" and avoid dumping data when
size of newly generated data is larger than the size of the ring buffer. It is reasonable
for forward ring buffer because in this case you lost the head of the first record, the
whole ring buffer is unparseable. However, it is wrong in backward case. What you
should do in this case is dumping the whole ring buffer.

+ return ret;
}
/*