Skip to content

Commit

Permalink
tracing: Fix outputting formats of x86-tsc and counter when use trace…
Browse files Browse the repository at this point in the history
…_clock

Outputting formats of x86-tsc and counter should be a raw format, but after
applying the patch(2b6080f), the format was
changed to nanosec. This is because the global variable trace_clock_id was used.
When we use multiple buffers, clock_id of each sub-buffer should be used. Then,
this patch uses tr->clock_id instead of the global variable trace_clock_id.

[ Basically, this fixes a regression where the multibuffer code changed the
  trace_clock file to update tr->clock_id but the traces still use the old
  global trace_clock_id variable, negating the file's effect. The global
  trace_clock_id variable is obsolete and removed. - SR ]

Link: http://lkml.kernel.org/r/20130423013239.22334.7394.stgit@yunodevel

Signed-off-by: Yoshihiro YUNOMAE <[email protected]>
Signed-off-by: Steven Rostedt <[email protected]>
  • Loading branch information
Yoshihiro YUNOMAE authored and rostedt committed Jun 11, 2013
1 parent f17a519 commit 58e8eed
Show file tree
Hide file tree
Showing 2 changed files with 3 additions and 7 deletions.
8 changes: 3 additions & 5 deletions kernel/trace/trace.c
Original file line number Diff line number Diff line change
Expand Up @@ -652,8 +652,6 @@ static struct {
ARCH_TRACE_CLOCKS
};

int trace_clock_id;

/*
* trace_parser_get_init - gets the buffer for trace parser
*/
Expand Down Expand Up @@ -2826,7 +2824,7 @@ __tracing_open(struct inode *inode, struct file *file, bool snapshot)
iter->iter_flags |= TRACE_FILE_ANNOTATE;

/* Output in nanoseconds only if we are using a clock in nanoseconds. */
if (trace_clocks[trace_clock_id].in_ns)
if (trace_clocks[tr->clock_id].in_ns)
iter->iter_flags |= TRACE_FILE_TIME_IN_NS;

/* stop the trace while dumping if we are not opening "snapshot" */
Expand Down Expand Up @@ -3825,7 +3823,7 @@ static int tracing_open_pipe(struct inode *inode, struct file *filp)
iter->iter_flags |= TRACE_FILE_LAT_FMT;

/* Output in nanoseconds only if we are using a clock in nanoseconds. */
if (trace_clocks[trace_clock_id].in_ns)
if (trace_clocks[tr->clock_id].in_ns)
iter->iter_flags |= TRACE_FILE_TIME_IN_NS;

iter->cpu_file = tc->cpu;
Expand Down Expand Up @@ -5095,7 +5093,7 @@ tracing_stats_read(struct file *filp, char __user *ubuf,
cnt = ring_buffer_bytes_cpu(trace_buf->buffer, cpu);
trace_seq_printf(s, "bytes: %ld\n", cnt);

if (trace_clocks[trace_clock_id].in_ns) {
if (trace_clocks[tr->clock_id].in_ns) {
/* local or global for trace_clock */
t = ns2usecs(ring_buffer_oldest_event_ts(trace_buf->buffer, cpu));
usec_rem = do_div(t, USEC_PER_SEC);
Expand Down
2 changes: 0 additions & 2 deletions kernel/trace/trace.h
Original file line number Diff line number Diff line change
Expand Up @@ -700,8 +700,6 @@ enum print_line_t print_trace_line(struct trace_iterator *iter);

extern unsigned long trace_flags;

extern int trace_clock_id;

/* Standard output formatting function used for function return traces */
#ifdef CONFIG_FUNCTION_GRAPH_TRACER

Expand Down

0 comments on commit 58e8eed

Please sign in to comment.