From: "Steven Rostedt (VMware)" <rostedt@xxxxxxxxxxx>
When reading the ring buffers, a kbuffer is allocated and assigned to be
able to parse out the raw events from the kernel buffer. But it is never
freed and causes leaked memory.
Consolidate page_to_kbuf() into read_next_page(), and instead of
allocating the kbuffer descriptor each time, simply reload it with the new
page.
Signed-off-by: Steven Rostedt (VMware) <rostedt@xxxxxxxxxxx>
---
src/tracefs-events.c | 57 +++++++++++++++++++-------------------------
1 file changed, 24 insertions(+), 33 deletions(-)
diff --git a/src/tracefs-events.c b/src/tracefs-events.c
index 3a6196b..3e08571 100644
--- a/src/tracefs-events.c
+++ b/src/tracefs-events.c
@@ -20,37 +20,6 @@
#include "tracefs.h"
#include "tracefs-local.h"
-static struct kbuffer *
-page_to_kbuf(struct tep_handle *tep, void *page, int size)
-{
- enum kbuffer_long_size long_size;
- enum kbuffer_endian endian;
- struct kbuffer *kbuf;
-
- if (tep_is_file_bigendian(tep))
- endian = KBUFFER_ENDIAN_BIG;
- else
- endian = KBUFFER_ENDIAN_LITTLE;
-
- if (tep_get_header_page_size(tep) == 8)
- long_size = KBUFFER_LSIZE_8;
- else
- long_size = KBUFFER_LSIZE_4;
-
- kbuf = kbuffer_alloc(long_size, endian);
- if (!kbuf)
- return NULL;
-
- kbuffer_load_subbuffer(kbuf, page);
- if (kbuffer_subbuffer_size(kbuf) > size) {
- tracefs_warning("%s: page_size > size", __func__);
- kbuffer_free(kbuf);
- kbuf = NULL;
- }
-
- return kbuf;
-}
-
struct cpu_iterate {
struct tep_record record;
struct tep_event *event;
@@ -88,13 +57,34 @@ static int read_kbuf_record(struct cpu_iterate *cpu)
int read_next_page(struct tep_handle *tep, struct cpu_iterate *cpu)
{
+ enum kbuffer_long_size long_size;
+ enum kbuffer_endian endian;
+
cpu->rsize = read(cpu->fd, cpu->page, cpu->psize);
if (cpu->rsize <= 0)
return -1;
- cpu->kbuf = page_to_kbuf(tep, cpu->page, cpu->rsize);
- if (!cpu->kbuf)
+ if (!cpu->kbuf) {
+ if (tep_is_file_bigendian(tep))
+ endian = KBUFFER_ENDIAN_BIG;
+ else
+ endian = KBUFFER_ENDIAN_LITTLE;
+
+ if (tep_get_header_page_size(tep) == 8)
+ long_size = KBUFFER_LSIZE_8;
+ else
+ long_size = KBUFFER_LSIZE_4;
+
+ cpu->kbuf = kbuffer_alloc(long_size, endian);
+ if (!cpu->kbuf)
+ return -1;
+ }
+
+ kbuffer_load_subbuffer(cpu->kbuf, cpu->page);
+ if (kbuffer_subbuffer_size(cpu->kbuf) > cpu->rsize) {
+ tracefs_warning("%s: page_size > %d", __func__, cpu->rsize);
return -1;
+ }
return 0;
}
@@ -256,6 +246,7 @@ int tracefs_iterate_raw_events(struct tep_handle *tep,
out:
if (all_cpus) {
for (i = 0; i < count; i++) {
+ kbuffer_free(all_cpus[i].kbuf);
close(all_cpus[i].fd);
free(all_cpus[i].page);
}
--
2.29.2
|