perf: Add attr->mmap2 attribute to an event

Adds a new PERF_RECORD_MMAP2 record type which is essence
an expanded version of PERF_RECORD_MMAP.

Used to request mmap records with more information about
the mapping, including device major, minor and the inode
number and generation for mappings associated with files
or shared memory segments. Works for code and data
(with attr->mmap_data set).

Existing PERF_RECORD_MMAP record is unmodified by this patch.

Signed-off-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Link: http://lkml.kernel.org/r/1377079825-19057-2-git-send-email-eranian@google.com
[ Added Al to the Cc:. Are the ino, maj/min exports of vma->vm_file OK? ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>

2 files changed, 65 insertions, 5 deletions

diff --git a/include/uapi/linux/perf_event.h b/include/uapi/linux/perf_event.h
index 42cb7b62ca59..a77f43af72b8 100644
--- a/include/uapi/linux/perf_event.h
+++ b/include/uapi/linux/perf_event.h
@@ -276,8 +276,9 @@ struct perf_event_attr {
 
                                 exclude_callchain_kernel : 1, /* exclude kernel callchains */
                                 exclude_callchain_user   : 1, /* exclude user callchains */
+                                mmap2          :  1, /* include mmap with inode data     */
 
-                                __reserved_1   : 41;
+                                __reserved_1   : 40;
 
         union {
                 __u32           wakeup_events;    /* wakeup every n events */
@@ -651,6 +652,27 @@ enum perf_event_type {
          */
         PERF_RECORD_SAMPLE                      = 9,
 
+        /*
+         * The MMAP2 records are an augmented version of MMAP, they add
+         * maj, min, ino numbers to be used to uniquely identify each mapping
+         *
+         * struct {
+         *      struct perf_event_header        header;
+         *
+         *      u32                             pid, tid;
+         *      u64                             addr;
+         *      u64                             len;
+         *      u64                             pgoff;
+         *      u32                             maj;
+         *      u32                             min;
+         *      u64                             ino;
+         *      u64                             ino_generation;
+         *      char                            filename[];
+         *      struct sample_id                sample_id;
+         * };
+         */
+        PERF_RECORD_MMAP2                       = 10,
+
         PERF_RECORD_MAX,                        /* non-ABI */
 };
 
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 15d0f2418e54..c7ee497c39a7 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -4776,7 +4776,7 @@ next:
 /*
  * task tracking -- fork/exit
  *
- * enabled by: attr.comm | attr.mmap | attr.mmap_data | attr.task
+ * enabled by: attr.comm | attr.mmap | attr.mmap2 | attr.mmap_data | attr.task
  */
 
 struct perf_task_event {
@@ -4796,8 +4796,9 @@ struct perf_task_event {
 
 static int perf_event_task_match(struct perf_event *event)
 {
-        return event->attr.comm || event->attr.mmap ||
-               event->attr.mmap_data || event->attr.task;
+        return event->attr.comm  || event->attr.mmap ||
+               event->attr.mmap2 || event->attr.mmap_data ||
+               event->attr.task;
 }
 
 static void perf_event_task_output(struct perf_event *event,
@@ -4992,6 +4993,9 @@ struct perf_mmap_event {
 
         const char              *file_name;
         int                     file_size;
+        int                     maj, min;
+        u64                     ino;
+        u64                     ino_generation;
 
         struct {
                 struct perf_event_header        header;
@@ -5012,7 +5016,7 @@ static int perf_event_mmap_match(struct perf_event *event,
         int executable = vma->vm_flags & VM_EXEC;
 
         return (!executable && event->attr.mmap_data) ||
-               (executable && event->attr.mmap);
+               (executable && (event->attr.mmap || event->attr.mmap2));
 }
 
 static void perf_event_mmap_output(struct perf_event *event,
@@ -5027,6 +5031,13 @@ static void perf_event_mmap_output(struct perf_event *event,
         if (!perf_event_mmap_match(event, data))
                 return;
 
+        if (event->attr.mmap2) {
+                mmap_event->event_id.header.type = PERF_RECORD_MMAP2;
+                mmap_event->event_id.header.size += sizeof(mmap_event->maj);
+                mmap_event->event_id.header.size += sizeof(mmap_event->min);
+                mmap_event->event_id.header.size += sizeof(mmap_event->ino);
+        }
+
         perf_event_header__init_id(&mmap_event->event_id.header, &sample, event);
         ret = perf_output_begin(&handle, event,
                                 mmap_event->event_id.header.size);
@@ -5037,6 +5048,14 @@ static void perf_event_mmap_output(struct perf_event *event,
         mmap_event->event_id.tid = perf_event_tid(event, current);
 
         perf_output_put(&handle, mmap_event->event_id);
+
+        if (event->attr.mmap2) {
+                perf_output_put(&handle, mmap_event->maj);
+                perf_output_put(&handle, mmap_event->min);
+                perf_output_put(&handle, mmap_event->ino);
+                perf_output_put(&handle, mmap_event->ino_generation);
+        }
+
         __output_copy(&handle, mmap_event->file_name,
                                    mmap_event->file_size);
 
@@ -5051,6 +5070,8 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event)
 {
         struct vm_area_struct *vma = mmap_event->vma;
         struct file *file = vma->vm_file;
+        int maj = 0, min = 0;
+        u64 ino = 0, gen = 0;
         unsigned int size;
         char tmp[16];
         char *buf = NULL;
@@ -5059,6 +5080,8 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event)
         memset(tmp, 0, sizeof(tmp));
 
         if (file) {
+                struct inode *inode;
+                dev_t dev;
                 /*
                  * d_path works from the end of the rb backwards, so we
                  * need to add enough zero bytes after the string to handle
@@ -5074,6 +5097,13 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event)
                         name = strncpy(tmp, "//toolong", sizeof(tmp));
                         goto got_name;
                 }
+                inode = file_inode(vma->vm_file);
+                dev = inode->i_sb->s_dev;
+                ino = inode->i_ino;
+                gen = inode->i_generation;
+                maj = MAJOR(dev);
+                min = MINOR(dev);
+
         } else {
                 if (arch_vma_name(mmap_event->vma)) {
                         name = strncpy(tmp, arch_vma_name(mmap_event->vma),
@@ -5104,6 +5134,10 @@ got_name:
 
         mmap_event->file_name = name;
         mmap_event->file_size = size;
+        mmap_event->maj = maj;
+        mmap_event->min = min;
+        mmap_event->ino = ino;
+        mmap_event->ino_generation = gen;
 
         if (!(vma->vm_flags & VM_EXEC))
                 mmap_event->event_id.header.misc |= PERF_RECORD_MISC_MMAP_DATA;
@@ -5140,6 +5174,10 @@ void perf_event_mmap(struct vm_area_struct *vma)
                         .len    = vma->vm_end - vma->vm_start,
                         .pgoff  = (u64)vma->vm_pgoff << PAGE_SHIFT,
                 },
+                /* .maj (attr_mmap2 only) */
+                /* .min (attr_mmap2 only) */
+                /* .ino (attr_mmap2 only) */
+                /* .ino_generation (attr_mmap2 only) */
         };
 
         perf_event_mmap_event(&mmap_event);