Em Wed, Jan 23, 2019 at 03:10:47PM -0800, Stephane Eranian escreveu:
> The xxx_mountpoint() interface provided by fs.c finds
> mount points for common pseudo filesystems. The first
> time xxx_mountpoint() is invoked, it scans the mount
> table (/proc/mounts) looking for a match. If found, it
> is cached. The price to scan /proc/mounts is paid once
> if the mount is found.
>
> When the mount point is not found, subsequent calls to
> xxx_mountpoint() scan /proc/mounts over and over again.
> There is no cacheing.
>
> This causes a scaling issue in perf record with hugeltbfs__mountpoint().
> The function is called for each process found in synthesize__mmap_events().
> If the machine has thousands of processes and if the /proc/mounts has many
> entries this could cause major overhead in perf record. We have observed
> multi-second slowdowns on some configurations.
>
> As an example on a laptop:
>
> Before:
> $ sudo umount /dev/hugepages
> $ strace -etrace=open -o /tmp/tt perf record -a ls
> $ fgrep mounts /tmp/tt
> 285
>
> After:
> $ sudo umount /dev/hugepages
> $ strace -etrace=open -o /tmp/tt perf record -a ls
> $ fgrep mounts /tmp/tt
> 1
>
> One could argue that the non-cacheing in case the moint point is not found
> is intentional. That way subsequent calls may discover a moint point if
> the syadmin mounts the filesystem. But the same argument could be made
> against cacheing the moint point. It could be unmounted causing errors.
> It all depends on the intent of the interface. This patch assumes it
> is expected to scan /proc/mounts once. The patch documents the cacheing
> behavior in the fs.h header file.
>
> An alternative would be to just fix perf record. But it would solve
> the problem with hugetlbs__mountpoint() but there could be similar
> issues (possibly down the line) with other xxx_mountpoint() calls
> in perf or other tools.
>
> Signed-off-by: Stephane Eranian <[email protected]>
> ---
> tools/lib/api/fs/fs.c | 19 +++++++++++++++++++
> tools/lib/api/fs/fs.h | 11 +++++++++++
> 2 files changed, 30 insertions(+)
>
> diff --git a/tools/lib/api/fs/fs.c b/tools/lib/api/fs/fs.c
> index 7aba8243a0e7..6934da54c96b 100644
> --- a/tools/lib/api/fs/fs.c
> +++ b/tools/lib/api/fs/fs.c
> @@ -90,6 +90,7 @@ struct fs {
> const char * const *mounts;
> char path[PATH_MAX];
> bool found;
> + bool checked;
> long magic;
> };
>
> @@ -111,31 +112,37 @@ static struct fs fs__entries[] = {
> .name = "sysfs",
> .mounts = sysfs__fs_known_mountpoints,
> .magic = SYSFS_MAGIC,
> + .checked= false,
No need for these initializations, 0 == false, and since we initialize
some of the other fields, the ones that haven't are set to zero.
> },
> [FS__PROCFS] = {
> .name = "proc",
> .mounts = procfs__known_mountpoints,
> .magic = PROC_SUPER_MAGIC,
> + .checked= false,
> },
> [FS__DEBUGFS] = {
> .name = "debugfs",
> .mounts = debugfs__known_mountpoints,
> .magic = DEBUGFS_MAGIC,
> + .checked= false,
> },
> [FS__TRACEFS] = {
> .name = "tracefs",
> .mounts = tracefs__known_mountpoints,
> .magic = TRACEFS_MAGIC,
> + .checked= false,
> },
> [FS__HUGETLBFS] = {
> .name = "hugetlbfs",
> .mounts = hugetlbfs__known_mountpoints,
> .magic = HUGETLBFS_MAGIC,
> + .checked= false,
> },
> [FS__BPF_FS] = {
> .name = "bpf",
> .mounts = bpf_fs__known_mountpoints,
> .magic = BPF_FS_MAGIC,
> + .checked= false,
> },
> };
>
> @@ -158,6 +165,7 @@ static bool fs__read_mounts(struct fs *fs)
> }
>
> fclose(fp);
> + fs->checked = true;
> return fs->found = found;
> }
>
> @@ -219,6 +227,7 @@ static bool fs__env_override(struct fs *fs)
> return false;
>
> fs->found = true;
> + fs->checked = true;
> strncpy(fs->path, override_path, sizeof(fs->path));
> return true;
> }
> @@ -244,6 +253,16 @@ static const char *fs__mountpoint(int idx)
> if (fs->found)
> return (const char *)fs->path;
>
> + /* the mount point was already checked for the mount point
Nit, we start with:
/*
* The mount point was already checked for the mount point
> + * but and did not exist, so return NULL to avoid scanning again.
> + * This makes the found and not found paths cost equivalent
> + * in case of multiple calls. This was not the case before
> + * and could cause significant scaling issues with callers.
> + * in case /proc/mounts need to be checked many times
.
> + */
> + if (fs->checked)
> + return NULL;
> +
> return fs__get_mountpoint(fs);
> }
>
> diff --git a/tools/lib/api/fs/fs.h b/tools/lib/api/fs/fs.h
> index 92d03b8396b1..00a5127b00e8 100644
> --- a/tools/lib/api/fs/fs.h
> +++ b/tools/lib/api/fs/fs.h
> @@ -18,6 +18,17 @@
> const char *name##__mount(void); \
> bool name##__configured(void); \
>
> +/*
> + * The xxxx__mountpoint() entry points find the first match mount point for
> each
Nicely written :-)
> + * filesystems listed below, where xxxx is the filesystem type.
> + *
> + * The interface is as follows:
> + * - If a mount point is found on first call, it is cached and used for all
> subsequent
> + * calls.
> + *
> + * - If a mount point is not found, NULL is returned on first call and all
> + * subsequent calls.
Ditto.
> + */
> FS(sysfs)
> FS(procfs)
> FS(debugfs)
> --
> 2.7.4
