summaryrefslogtreecommitdiffstats
path: root/drivers/oprofile/cpu_buffer.c
diff options
context:
space:
mode:
authorRobert Richter <robert.richter@amd.com>2008-12-09 01:21:32 +0100
committerRobert Richter <robert.richter@amd.com>2008-12-10 14:20:18 +0100
commit6dad828b76c7224a22ddc9ce7aa495d994f03b31 (patch)
tree364de7a3efd56d60c0742145e3a8d3d4f73bcde4 /drivers/oprofile/cpu_buffer.c
parente09373f22e76cc048ca5fe10a9ff9012f5d64309 (diff)
downloadop-kernel-dev-6dad828b76c7224a22ddc9ce7aa495d994f03b31.zip
op-kernel-dev-6dad828b76c7224a22ddc9ce7aa495d994f03b31.tar.gz
oprofile: port to the new ring_buffer
This patch replaces the current oprofile cpu buffer implementation with the ring buffer provided by the tracing framework. The motivation here is to leave the pain of implementing ring buffers to others. Oh, no, there are more advantages. Main reason is the support of different sample sizes that could be stored in the buffer. Use cases for this are IBS and Cell spu profiling. Using the new ring buffer ensures valid and complete samples and allows copying the cpu buffer stateless without knowing its content. Second it will use generic kernel API and also reduce code size. And hopefully, there are less bugs. Since the new tracing ring buffer implementation uses spin locks to protect the buffer during read/write access, it is difficult to use the buffer in an NMI handler. In this case, writing to the buffer by the NMI handler (x86) could occur also during critical sections when reading the buffer. To avoid this, there are 2 buffers for independent read and write access. Read access is in process context only, write access only in the NMI handler. If the read buffer runs empty, both buffers are swapped atomically. There is potentially a small window during swapping where the buffers are disabled and samples could be lost. Using 2 buffers is a little bit overhead, but the solution is clear and does not require changes in the ring buffer implementation. It can be changed to a single buffer solution when the ring buffer access is implemented as non-locking atomic code. The new buffer requires more size to store the same amount of samples because each sample includes an u32 header. Also, there is more code to execute for buffer access. Nonetheless, the buffer implementation is proven in the ftrace environment and worth to use also in oprofile. Patches that changes the internal IBS buffer usage will follow. Cc: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Robert Richter <robert.richter@amd.com>
Diffstat (limited to 'drivers/oprofile/cpu_buffer.c')
-rw-r--r--drivers/oprofile/cpu_buffer.c63
1 files changed, 48 insertions, 15 deletions
diff --git a/drivers/oprofile/cpu_buffer.c b/drivers/oprofile/cpu_buffer.c
index 5cf7efe..eb280ec 100644
--- a/drivers/oprofile/cpu_buffer.c
+++ b/drivers/oprofile/cpu_buffer.c
@@ -28,6 +28,25 @@
#include "buffer_sync.h"
#include "oprof.h"
+#define OP_BUFFER_FLAGS 0
+
+/*
+ * Read and write access is using spin locking. Thus, writing to the
+ * buffer by NMI handler (x86) could occur also during critical
+ * sections when reading the buffer. To avoid this, there are 2
+ * buffers for independent read and write access. Read access is in
+ * process context only, write access only in the NMI handler. If the
+ * read buffer runs empty, both buffers are swapped atomically. There
+ * is potentially a small window during swapping where the buffers are
+ * disabled and samples could be lost.
+ *
+ * Using 2 buffers is a little bit overhead, but the solution is clear
+ * and does not require changes in the ring buffer implementation. It
+ * can be changed to a single buffer solution when the ring buffer
+ * access is implemented as non-locking atomic code.
+ */
+struct ring_buffer *op_ring_buffer_read;
+struct ring_buffer *op_ring_buffer_write;
DEFINE_PER_CPU(struct oprofile_cpu_buffer, cpu_buffer);
static void wq_sync_buffer(struct work_struct *work);
@@ -37,12 +56,12 @@ static int work_enabled;
void free_cpu_buffers(void)
{
- int i;
-
- for_each_possible_cpu(i) {
- vfree(per_cpu(cpu_buffer, i).buffer);
- per_cpu(cpu_buffer, i).buffer = NULL;
- }
+ if (op_ring_buffer_read)
+ ring_buffer_free(op_ring_buffer_read);
+ op_ring_buffer_read = NULL;
+ if (op_ring_buffer_write)
+ ring_buffer_free(op_ring_buffer_write);
+ op_ring_buffer_write = NULL;
}
unsigned long oprofile_get_cpu_buffer_size(void)
@@ -64,14 +83,16 @@ int alloc_cpu_buffers(void)
unsigned long buffer_size = fs_cpu_buffer_size;
+ op_ring_buffer_read = ring_buffer_alloc(buffer_size, OP_BUFFER_FLAGS);
+ if (!op_ring_buffer_read)
+ goto fail;
+ op_ring_buffer_write = ring_buffer_alloc(buffer_size, OP_BUFFER_FLAGS);
+ if (!op_ring_buffer_write)
+ goto fail;
+
for_each_possible_cpu(i) {
struct oprofile_cpu_buffer *b = &per_cpu(cpu_buffer, i);
- b->buffer = vmalloc_node(sizeof(struct op_sample) * buffer_size,
- cpu_to_node(i));
- if (!b->buffer)
- goto fail;
-
b->last_task = NULL;
b->last_is_kernel = -1;
b->tracing = 0;
@@ -140,10 +161,22 @@ static inline void
add_sample(struct oprofile_cpu_buffer *cpu_buf,
unsigned long pc, unsigned long event)
{
- struct op_sample *entry = cpu_buffer_write_entry(cpu_buf);
- entry->eip = pc;
- entry->event = event;
- cpu_buffer_write_commit(cpu_buf);
+ struct op_entry entry;
+
+ if (cpu_buffer_write_entry(&entry))
+ goto Error;
+
+ entry.sample->eip = pc;
+ entry.sample->event = event;
+
+ if (cpu_buffer_write_commit(&entry))
+ goto Error;
+
+ return;
+
+Error:
+ cpu_buf->sample_lost_overflow++;
+ return;
}
static inline void
OpenPOWER on IntegriCloud