From: Manish Varma <varmam@google.com>
To: Alexander Viro <viro@zeniv.linux.org.uk>,
Thomas Gleixner <tglx@linutronix.de>
Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
Manish Varma <varmam@google.com>,
Kelly Rossmoyer <krossmo@google.com>
Subject: [PATCH] fs: Improve eventpoll logging to stop indicting timerfd
Date: Mon, 1 Mar 2021 19:49:28 -0800 [thread overview]
Message-ID: <20210302034928.3761098-1-varmam@google.com> (raw)
timerfd doesn't create any wakelocks, but eventpoll can. When it does,
it names them after the underlying file descriptor, and since all
timerfd file descriptors are named "[timerfd]" (which saves memory on
systems like desktops with potentially many timerfd instances), all
wakesources created as a result of using the eventpoll-on-timerfd idiom
are called... "[timerfd]".
However, it becomes impossible to tell which "[timerfd]" wakesource is
affliated with which process and hence troubleshooting is difficult.
This change addresses this problem by changing the way eventpoll and
timerfd file descriptors (and thus the eventpoll-on-timerfd wakesources)
are named:
1) timerfd descriptors are now named "[timerfdN:P]", where N is a
monotonically increasing integer for each timerfd instance created and P
is the command string of the creating process.
2) the top-level per-process eventpoll wakesource is now named "epoll:P"
(instead of just "eventpoll"), where P, again, is the command string of
the creating process
3) individual per-underlying-filedescriptor eventpoll wakesources are
now named "epollitem:P.F", where P is the command string of the creating
process and F is the name of the underlying file descriptor.
All together, that will give us names like the following:
1) timerfd file descriptor: [timerfd14:system_server]
2) eventpoll top-level per-process wakesource: epoll:system_server
3) eventpoll-on-timerfd per-descriptor wakesource:
epollitem:system_server.[timerfd14:system_server]
Co-developed-by: Kelly Rossmoyer <krossmo@google.com>
Signed-off-by: Kelly Rossmoyer <krossmo@google.com>
Signed-off-by: Manish Varma <varmam@google.com>
---
fs/eventpoll.c | 10 ++++++++--
fs/timerfd.c | 11 ++++++++++-
2 files changed, 18 insertions(+), 3 deletions(-)
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 185252d8f4c7..af28be4285f8 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -1451,15 +1451,21 @@ static int ep_create_wakeup_source(struct epitem *epi)
{
struct name_snapshot n;
struct wakeup_source *ws;
+ char task_comm_buf[sizeof(current->comm)];
+ char buf[64];
+
+ get_task_comm(task_comm_buf, current);
if (!epi->ep->ws) {
- epi->ep->ws = wakeup_source_register(NULL, "eventpoll");
+ snprintf(buf, sizeof(buf), "epoll:%s", task_comm_buf);
+ epi->ep->ws = wakeup_source_register(NULL, buf);
if (!epi->ep->ws)
return -ENOMEM;
}
take_dentry_name_snapshot(&n, epi->ffd.file->f_path.dentry);
- ws = wakeup_source_register(NULL, n.name.name);
+ snprintf(buf, sizeof(buf), "epollitem:%s.%s", task_comm_buf, n.name.name);
+ ws = wakeup_source_register(NULL, buf);
release_dentry_name_snapshot(&n);
if (!ws)
diff --git a/fs/timerfd.c b/fs/timerfd.c
index c5509d2448e3..4249e8c9a38c 100644
--- a/fs/timerfd.c
+++ b/fs/timerfd.c
@@ -46,6 +46,8 @@ struct timerfd_ctx {
bool might_cancel;
};
+static atomic_t instance_count = ATOMIC_INIT(0);
+
static LIST_HEAD(cancel_list);
static DEFINE_SPINLOCK(cancel_lock);
@@ -391,6 +393,9 @@ SYSCALL_DEFINE2(timerfd_create, int, clockid, int, flags)
{
int ufd;
struct timerfd_ctx *ctx;
+ char task_comm_buf[sizeof(current->comm)];
+ char file_name_buf[32];
+ int instance;
/* Check the TFD_* constants for consistency. */
BUILD_BUG_ON(TFD_CLOEXEC != O_CLOEXEC);
@@ -427,7 +432,11 @@ SYSCALL_DEFINE2(timerfd_create, int, clockid, int, flags)
ctx->moffs = ktime_mono_to_real(0);
- ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx,
+ instance = atomic_inc_return(&instance_count);
+ get_task_comm(task_comm_buf, current);
+ snprintf(file_name_buf, sizeof(file_name_buf), "[timerfd%d:%s]",
+ instance, task_comm_buf);
+ ufd = anon_inode_getfd(file_name_buf, &timerfd_fops, ctx,
O_RDWR | (flags & TFD_SHARED_FCNTL_FLAGS));
if (ufd < 0)
kfree(ctx);
--
2.30.0.617.g56c4b15f3c-goog
next reply other threads:[~2021-03-02 8:12 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-02 3:49 Manish Varma [this message]
2021-03-18 13:04 ` [PATCH] fs: Improve eventpoll logging to stop indicting timerfd Thomas Gleixner
2021-03-22 17:15 ` Manish Varma
2021-03-22 21:40 ` Thomas Gleixner
2021-03-25 5:18 ` Manish Varma
2021-03-25 7:06 ` Thomas Gleixner
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210302034928.3761098-1-varmam@google.com \
--to=varmam@google.com \
--cc=krossmo@google.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=tglx@linutronix.de \
--cc=viro@zeniv.linux.org.uk \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).