Re: [PATCH RESEND] perf inject --jit: Remove //anon mmap events

2019-10-19 Thread Jiri Olsa
On Thu, Oct 17, 2019 at 11:23:20AM -0700, Steve MacLean wrote:
> From: Steve MacLean 
> 
> While a JIT is jitting code it will eventually need to commit more pages and
> change these pages to executable permissions.
> 
> Typically the JIT will want these colocated to minimize branch displacements.
> 
> The kernel will coalesce these anonymous mapping with identical permissions
> before sending an MMAP event for the new pages. This means the mmap event for
> the new pages will include the older pages.
> 
> These anonymous mmap events will obscure the jitdump injected pseudo events.
> This means that the jitdump generated symbols, machine code, debugging info,
> and unwind info will no longer be used.
> 
> Observations:
> 
> When a process emits a jit dump marker and a jitdump file, the perf-xxx.map
> file represents inferior information which has been superceded by the
> jitdump jit-xxx.dump file.
> 
> Further the '//anon*' mmap events are only required for the legacy
> perf-xxx.map mapping.
> 
> Summary:
> 
> Add rbtree to track which pids have sucessfully injected a jitdump file.
> 
> During "perf inject --jit", discard "//anon*" mmap events for any pid which
> has sucessfully processed a jitdump file.
> 
> Committer testing:
> 
> // jitdump case
> perf record 
> perf inject --jit --input perf.data --output perfjit.data
> 
> // verify mmap "//anon" events present initially
> perf script --input perf.data --show-mmap-events | grep '//anon'
> // verify mmap "//anon" events removed
> perf script --input perfjit.data --show-mmap-events | grep '//anon'
> 
> // no jitdump case
> perf record 
> perf inject --jit --input perf.data --output perfjit.data
> 
> // verify mmap "//anon" events present initially
> perf script --input perf.data --show-mmap-events | grep '//anon'
> // verify mmap "//anon" events not removed
> perf script --input perfjit.data --show-mmap-events | grep '//anon'
> 
> Repro:
> 
> This issue was discovered while testing the initial CoreCLR jitdump
> implementation. https://github.com/dotnet/coreclr/pull/26897.

I posted some questions for previous version in here,
but can't find answers:
  https://lore.kernel.org/lkml/20191003105716.GB23291@krava/

thanks,
jirka



[PATCH RESEND] perf inject --jit: Remove //anon mmap events

2019-10-17 Thread Steve MacLean
From: Steve MacLean 

While a JIT is jitting code it will eventually need to commit more pages and
change these pages to executable permissions.

Typically the JIT will want these colocated to minimize branch displacements.

The kernel will coalesce these anonymous mapping with identical permissions
before sending an MMAP event for the new pages. This means the mmap event for
the new pages will include the older pages.

These anonymous mmap events will obscure the jitdump injected pseudo events.
This means that the jitdump generated symbols, machine code, debugging info,
and unwind info will no longer be used.

Observations:

When a process emits a jit dump marker and a jitdump file, the perf-xxx.map
file represents inferior information which has been superceded by the
jitdump jit-xxx.dump file.

Further the '//anon*' mmap events are only required for the legacy
perf-xxx.map mapping.

Summary:

Add rbtree to track which pids have sucessfully injected a jitdump file.

During "perf inject --jit", discard "//anon*" mmap events for any pid which
has sucessfully processed a jitdump file.

Committer testing:

// jitdump case
perf record 
perf inject --jit --input perf.data --output perfjit.data

// verify mmap "//anon" events present initially
perf script --input perf.data --show-mmap-events | grep '//anon'
// verify mmap "//anon" events removed
perf script --input perfjit.data --show-mmap-events | grep '//anon'

// no jitdump case
perf record 
perf inject --jit --input perf.data --output perfjit.data

// verify mmap "//anon" events present initially
perf script --input perf.data --show-mmap-events | grep '//anon'
// verify mmap "//anon" events not removed
perf script --input perfjit.data --show-mmap-events | grep '//anon'

Repro:

This issue was discovered while testing the initial CoreCLR jitdump
implementation. https://github.com/dotnet/coreclr/pull/26897.

Cc: Peter Zijlstra 
Cc: Ingo Molnar 
Cc: Arnaldo Carvalho de Melo 
Cc: Mark Rutland 
Cc: Alexander Shishkin 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Stephane Eranian 
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Steve MacLean 
---
 tools/perf/builtin-inject.c |  4 +--
 tools/perf/util/jitdump.c   | 63 +
 2 files changed, 65 insertions(+), 2 deletions(-)

diff --git a/tools/perf/builtin-inject.c b/tools/perf/builtin-inject.c
index 372ecb3..0f38862 100644
--- a/tools/perf/builtin-inject.c
+++ b/tools/perf/builtin-inject.c
@@ -263,7 +263,7 @@ static int perf_event__jit_repipe_mmap(struct perf_tool 
*tool,
 * if jit marker, then inject jit mmaps and generate ELF images
 */
ret = jit_process(inject->session, >output, machine,
- event->mmap.filename, sample->pid, );
+ event->mmap.filename, event->mmap.pid, );
if (ret < 0)
return ret;
if (ret) {
@@ -301,7 +301,7 @@ static int perf_event__jit_repipe_mmap2(struct perf_tool 
*tool,
 * if jit marker, then inject jit mmaps and generate ELF images
 */
ret = jit_process(inject->session, >output, machine,
- event->mmap2.filename, sample->pid, );
+ event->mmap2.filename, event->mmap2.pid, );
if (ret < 0)
return ret;
if (ret) {
diff --git a/tools/perf/util/jitdump.c b/tools/perf/util/jitdump.c
index e3ccb0c..6d891d1 100644
--- a/tools/perf/util/jitdump.c
+++ b/tools/perf/util/jitdump.c
@@ -749,6 +749,59 @@ static int jit_repipe_debug_info(struct jit_buf_desc *jd, 
union jr_entry *jr)
return 0;
 }
 
+struct pid_rbtree
+{
+   struct rb_node node;
+   pid_t pid;
+};
+
+static void jit_add_pid(struct rb_root *root, pid_t pid)
+{
+   struct rb_node **new = &(root->rb_node), *parent = NULL;
+   struct pid_rbtree* data = NULL;
+
+   /* Figure out where to put new node */
+   while (*new) {
+   struct pid_rbtree *this = container_of(*new, struct pid_rbtree, 
node);
+   pid_t nodePid = this->pid;
+
+   parent = *new;
+   if (pid < nodePid)
+   new = &((*new)->rb_left);
+   else if (pid > nodePid)
+   new = &((*new)->rb_right);
+   else
+   return;
+   }
+
+   data = malloc(sizeof(struct pid_rbtree));
+   data->pid = pid;
+
+   /* Add new node and rebalance tree. */
+   rb_link_node(>node, parent, new);
+   rb_insert_color(>node, root);
+
+   return;
+}
+
+static bool jit_has_pid(struct rb_root *root, pid_t pid)
+{
+   struct rb_node *node = root->rb_node;
+
+   while (node) {
+   struct pid_rbtree *this = container_of(node, struct pid_rbtree, 
node);
+   pid_t nodePid = this->pid;
+
+   if (pid < nodePid)
+   node = node->rb_left;
+   else if (pid > nodePid)
+   node = node->rb_right;
+