| Message ID | 20250402-fuse-io-uring-trace-points-v1-2-11b0211fa658@ddn.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | fuse: Improve ftraces, atomic req unique and code dup removal | expand |
On Wed, 2 Apr 2025 at 19:41, Bernd Schubert <bschubert@ddn.com> wrote: > > This is especially needed for better ftrace analysis, > for example to build histograms. So far the request unique > was missing, because it was added after the first trace message. > > IDs/req-unique now might not come up perfectly sequentially > anymore, but especially with cloned device or io-uring this > did not work perfectly anyway. Well, we can try in any case. It would be a pretty insane server that actually looks at the h->unique value, but not impossible. Thanks, Miklos
On 4/2/25 20:33, Miklos Szeredi wrote: > On Wed, 2 Apr 2025 at 19:41, Bernd Schubert <bschubert@ddn.com> wrote: >> >> This is especially needed for better ftrace analysis, >> for example to build histograms. So far the request unique >> was missing, because it was added after the first trace message. >> >> IDs/req-unique now might not come up perfectly sequentially >> anymore, but especially with cloned device or io-uring this >> did not work perfectly anyway. > > Well, we can try in any case. It would be a pretty insane server that > actually looks at the h->unique value, but not impossible. Should be easy to revert if someone complains. We could have another (and for sure per-cpu counter) and add that into ftrace output. Thanks, Bernd
diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index e9592ab092b948bacb5034018bd1f32c917d5c9f..1ccf5a9c61ae2b11bc1d0b799c08e6da908a9782 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -259,8 +259,6 @@ static void fuse_dev_queue_req(struct fuse_iqueue *fiq, struct fuse_req *req) { spin_lock(&fiq->lock); if (fiq->connected) { - if (req->in.h.opcode != FUSE_NOTIFY_REPLY) - req->in.h.unique = fuse_get_unique(fiq); list_add_tail(&req->list, &fiq->pending); fuse_dev_wake_and_unlock(fiq); } else { @@ -508,6 +506,9 @@ static void fuse_args_to_req(struct fuse_req *req, struct fuse_args *args) req->in.h.total_extlen = args->in_args[args->ext_idx].size / 8; if (args->end) __set_bit(FR_ASYNC, &req->flags); + + if (req->in.h.opcode != FUSE_NOTIFY_REPLY) + req->in.h.unique = fuse_get_unique(&req->fm->fc->iq); } ssize_t __fuse_simple_request(struct mnt_idmap *idmap, @@ -555,9 +556,6 @@ ssize_t __fuse_simple_request(struct mnt_idmap *idmap, static bool fuse_request_queue_background_uring(struct fuse_conn *fc, struct fuse_req *req) { - struct fuse_iqueue *fiq = &fc->iq; - - req->in.h.unique = fuse_get_unique(fiq); req->in.h.len = sizeof(struct fuse_in_header) + fuse_len_args(req->args->in_numargs, (struct fuse_arg *) req->args->in_args); diff --git a/fs/fuse/dev_uring.c b/fs/fuse/dev_uring.c index 82bf458fa9db5b2357ae2d1cf5621ed4db978892..5a05b76249d6fe6214e948955f23eed1e40bb751 100644 --- a/fs/fuse/dev_uring.c +++ b/fs/fuse/dev_uring.c @@ -1230,9 +1230,6 @@ void fuse_uring_queue_fuse_req(struct fuse_iqueue *fiq, struct fuse_req *req) if (!queue) goto err; - if (req->in.h.opcode != FUSE_NOTIFY_REPLY) - req->in.h.unique = fuse_get_unique(fiq); - spin_lock(&queue->lock); err = -ENOTCONN; if (unlikely(queue->stopped)) diff --git a/fs/fuse/virtio_fs.c b/fs/fuse/virtio_fs.c index 82afe78ec542358e2db6f4d955d521652ae363ec..ea13d57133c335554acae33b22e1604424886ac9 100644 --- a/fs/fuse/virtio_fs.c +++ b/fs/fuse/virtio_fs.c @@ -1482,9 +1482,6 @@ static void virtio_fs_send_req(struct fuse_iqueue *fiq, struct fuse_req *req) struct virtio_fs_vq *fsvq; int ret; - if (req->in.h.opcode != FUSE_NOTIFY_REPLY) - req->in.h.unique = fuse_get_unique(fiq); - clear_bit(FR_PENDING, &req->flags); fs = fiq->priv;
This is especially needed for better ftrace analysis, for example to build histograms. So far the request unique was missing, because it was added after the first trace message. IDs/req-unique now might not come up perfectly sequentially anymore, but especially with cloned device or io-uring this did not work perfectly anyway. Signed-off-by: Bernd Schubert <bschubert@ddn.com> --- fs/fuse/dev.c | 8 +++----- fs/fuse/dev_uring.c | 3 --- fs/fuse/virtio_fs.c | 3 --- 3 files changed, 3 insertions(+), 11 deletions(-)