From: Jens Axboe <[email protected]>
To: [email protected]
Cc: Jens Axboe <[email protected]>
Subject: [PATCH 12/18] io_uring: pass required context in as flags
Date: Fri, 16 Oct 2020 10:02:18 -0600 [thread overview]
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>
We have a number of bits that decide what context to inherit. Set up
io-wq flags for these instead. This is in preparation for always having
the various members set, but not always needing them for all requests.
No intended functional changes in this patch.
Signed-off-by: Jens Axboe <[email protected]>
---
fs/io-wq.c | 10 +++--
fs/io-wq.h | 6 +++
fs/io_uring.c | 100 ++++++++++++++++++++------------------------------
3 files changed, 52 insertions(+), 64 deletions(-)
diff --git a/fs/io-wq.c b/fs/io-wq.c
index 149fd2f0927e..e636898f8a1f 100644
--- a/fs/io-wq.c
+++ b/fs/io-wq.c
@@ -448,6 +448,8 @@ static inline void io_wq_switch_blkcg(struct io_worker *worker,
struct io_wq_work *work)
{
#ifdef CONFIG_BLK_CGROUP
+ if (!(work->flags & IO_WQ_WORK_BLKCG))
+ return;
if (work->blkcg_css != worker->blkcg_css) {
kthread_associate_blkcg(work->blkcg_css);
worker->blkcg_css = work->blkcg_css;
@@ -470,17 +472,17 @@ static void io_wq_switch_creds(struct io_worker *worker,
static void io_impersonate_work(struct io_worker *worker,
struct io_wq_work *work)
{
- if (work->files && current->files != work->files) {
+ if ((work->flags & IO_WQ_WORK_FILES) && current->files != work->files) {
task_lock(current);
current->files = work->files;
current->nsproxy = work->nsproxy;
task_unlock(current);
}
- if (work->fs && current->fs != work->fs)
+ if ((work->flags & IO_WQ_WORK_FS) && current->fs != work->fs)
current->fs = work->fs;
- if (work->mm != worker->mm)
+ if ((work->flags & IO_WQ_WORK_MM) && work->mm != worker->mm)
io_wq_switch_mm(worker, work);
- if (worker->cur_creds != work->creds)
+ if ((work->flags & IO_WQ_WORK_CREDS) && worker->cur_creds != work->creds)
io_wq_switch_creds(worker, work);
current->signal->rlim[RLIMIT_FSIZE].rlim_cur = work->fsize;
io_wq_switch_blkcg(worker, work);
diff --git a/fs/io-wq.h b/fs/io-wq.h
index 84bcf6a85523..31a29023605a 100644
--- a/fs/io-wq.h
+++ b/fs/io-wq.h
@@ -10,6 +10,12 @@ enum {
IO_WQ_WORK_NO_CANCEL = 8,
IO_WQ_WORK_CONCURRENT = 16,
+ IO_WQ_WORK_FILES = 32,
+ IO_WQ_WORK_FS = 64,
+ IO_WQ_WORK_MM = 128,
+ IO_WQ_WORK_CREDS = 256,
+ IO_WQ_WORK_BLKCG = 512,
+
IO_WQ_HASH_SHIFT = 24, /* upper 8 bits are used for hash key */
};
diff --git a/fs/io_uring.c b/fs/io_uring.c
index 2c83c2688ec5..6f6f6bcef82d 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -729,8 +729,6 @@ struct io_submit_state {
};
struct io_op_def {
- /* needs current->mm setup, does mm access */
- unsigned needs_mm : 1;
/* needs req->file assigned */
unsigned needs_file : 1;
/* don't fail if file grab fails */
@@ -741,10 +739,6 @@ struct io_op_def {
unsigned unbound_nonreg_file : 1;
/* opcode is not supported by this kernel */
unsigned not_supported : 1;
- /* needs file table */
- unsigned file_table : 1;
- /* needs ->fs */
- unsigned needs_fs : 1;
/* set if opcode supports polled "wait" */
unsigned pollin : 1;
unsigned pollout : 1;
@@ -754,45 +748,42 @@ struct io_op_def {
unsigned needs_fsize : 1;
/* must always have async data allocated */
unsigned needs_async_data : 1;
- /* needs blkcg context, issues async io potentially */
- unsigned needs_blkcg : 1;
/* size of async data needed, if any */
unsigned short async_size;
+ unsigned work_flags;
};
static const struct io_op_def io_op_defs[] = {
[IORING_OP_NOP] = {},
[IORING_OP_READV] = {
- .needs_mm = 1,
.needs_file = 1,
.unbound_nonreg_file = 1,
.pollin = 1,
.buffer_select = 1,
.needs_async_data = 1,
- .needs_blkcg = 1,
.async_size = sizeof(struct io_async_rw),
+ .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG,
},
[IORING_OP_WRITEV] = {
- .needs_mm = 1,
.needs_file = 1,
.hash_reg_file = 1,
.unbound_nonreg_file = 1,
.pollout = 1,
.needs_fsize = 1,
.needs_async_data = 1,
- .needs_blkcg = 1,
.async_size = sizeof(struct io_async_rw),
+ .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG,
},
[IORING_OP_FSYNC] = {
.needs_file = 1,
- .needs_blkcg = 1,
+ .work_flags = IO_WQ_WORK_BLKCG,
},
[IORING_OP_READ_FIXED] = {
.needs_file = 1,
.unbound_nonreg_file = 1,
.pollin = 1,
- .needs_blkcg = 1,
.async_size = sizeof(struct io_async_rw),
+ .work_flags = IO_WQ_WORK_BLKCG,
},
[IORING_OP_WRITE_FIXED] = {
.needs_file = 1,
@@ -800,8 +791,8 @@ static const struct io_op_def io_op_defs[] = {
.unbound_nonreg_file = 1,
.pollout = 1,
.needs_fsize = 1,
- .needs_blkcg = 1,
.async_size = sizeof(struct io_async_rw),
+ .work_flags = IO_WQ_WORK_BLKCG,
},
[IORING_OP_POLL_ADD] = {
.needs_file = 1,
@@ -810,137 +801,123 @@ static const struct io_op_def io_op_defs[] = {
[IORING_OP_POLL_REMOVE] = {},
[IORING_OP_SYNC_FILE_RANGE] = {
.needs_file = 1,
- .needs_blkcg = 1,
+ .work_flags = IO_WQ_WORK_BLKCG,
},
[IORING_OP_SENDMSG] = {
- .needs_mm = 1,
.needs_file = 1,
.unbound_nonreg_file = 1,
- .needs_fs = 1,
.pollout = 1,
.needs_async_data = 1,
- .needs_blkcg = 1,
.async_size = sizeof(struct io_async_msghdr),
+ .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG |
+ IO_WQ_WORK_FS,
},
[IORING_OP_RECVMSG] = {
- .needs_mm = 1,
.needs_file = 1,
.unbound_nonreg_file = 1,
- .needs_fs = 1,
.pollin = 1,
.buffer_select = 1,
.needs_async_data = 1,
- .needs_blkcg = 1,
.async_size = sizeof(struct io_async_msghdr),
+ .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG |
+ IO_WQ_WORK_FS,
},
[IORING_OP_TIMEOUT] = {
- .needs_mm = 1,
.needs_async_data = 1,
.async_size = sizeof(struct io_timeout_data),
+ .work_flags = IO_WQ_WORK_MM,
},
[IORING_OP_TIMEOUT_REMOVE] = {},
[IORING_OP_ACCEPT] = {
- .needs_mm = 1,
.needs_file = 1,
.unbound_nonreg_file = 1,
- .file_table = 1,
.pollin = 1,
+ .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_FILES,
},
[IORING_OP_ASYNC_CANCEL] = {},
[IORING_OP_LINK_TIMEOUT] = {
- .needs_mm = 1,
.needs_async_data = 1,
.async_size = sizeof(struct io_timeout_data),
+ .work_flags = IO_WQ_WORK_MM,
},
[IORING_OP_CONNECT] = {
- .needs_mm = 1,
.needs_file = 1,
.unbound_nonreg_file = 1,
.pollout = 1,
.needs_async_data = 1,
.async_size = sizeof(struct io_async_connect),
+ .work_flags = IO_WQ_WORK_MM,
},
[IORING_OP_FALLOCATE] = {
.needs_file = 1,
.needs_fsize = 1,
- .needs_blkcg = 1,
+ .work_flags = IO_WQ_WORK_BLKCG,
},
[IORING_OP_OPENAT] = {
- .file_table = 1,
- .needs_fs = 1,
- .needs_blkcg = 1,
+ .work_flags = IO_WQ_WORK_FILES | IO_WQ_WORK_BLKCG |
+ IO_WQ_WORK_FS,
},
[IORING_OP_CLOSE] = {
.needs_file = 1,
.needs_file_no_error = 1,
- .file_table = 1,
- .needs_blkcg = 1,
+ .work_flags = IO_WQ_WORK_FILES | IO_WQ_WORK_BLKCG,
},
[IORING_OP_FILES_UPDATE] = {
- .needs_mm = 1,
- .file_table = 1,
+ .work_flags = IO_WQ_WORK_FILES | IO_WQ_WORK_MM,
},
[IORING_OP_STATX] = {
- .needs_mm = 1,
- .needs_fs = 1,
- .file_table = 1,
- .needs_blkcg = 1,
+ .work_flags = IO_WQ_WORK_FILES | IO_WQ_WORK_MM |
+ IO_WQ_WORK_FS | IO_WQ_WORK_BLKCG,
},
[IORING_OP_READ] = {
- .needs_mm = 1,
.needs_file = 1,
.unbound_nonreg_file = 1,
.pollin = 1,
.buffer_select = 1,
- .needs_blkcg = 1,
.async_size = sizeof(struct io_async_rw),
+ .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG,
},
[IORING_OP_WRITE] = {
- .needs_mm = 1,
.needs_file = 1,
.unbound_nonreg_file = 1,
.pollout = 1,
.needs_fsize = 1,
- .needs_blkcg = 1,
.async_size = sizeof(struct io_async_rw),
+ .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG,
},
[IORING_OP_FADVISE] = {
.needs_file = 1,
- .needs_blkcg = 1,
+ .work_flags = IO_WQ_WORK_BLKCG,
},
[IORING_OP_MADVISE] = {
- .needs_mm = 1,
- .needs_blkcg = 1,
+ .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG,
},
[IORING_OP_SEND] = {
- .needs_mm = 1,
.needs_file = 1,
.unbound_nonreg_file = 1,
.pollout = 1,
- .needs_blkcg = 1,
+ .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG,
},
[IORING_OP_RECV] = {
- .needs_mm = 1,
.needs_file = 1,
.unbound_nonreg_file = 1,
.pollin = 1,
.buffer_select = 1,
- .needs_blkcg = 1,
+ .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG,
},
[IORING_OP_OPENAT2] = {
- .file_table = 1,
- .needs_fs = 1,
- .needs_blkcg = 1,
+ .work_flags = IO_WQ_WORK_FILES | IO_WQ_WORK_FS |
+ IO_WQ_WORK_BLKCG,
},
[IORING_OP_EPOLL_CTL] = {
.unbound_nonreg_file = 1,
- .file_table = 1,
+ .work_flags = IO_WQ_WORK_FILES,
},
[IORING_OP_SPLICE] = {
.needs_file = 1,
.hash_reg_file = 1,
.unbound_nonreg_file = 1,
- .needs_blkcg = 1,
+ .work_flags = IO_WQ_WORK_BLKCG,
},
[IORING_OP_PROVIDE_BUFFERS] = {},
[IORING_OP_REMOVE_BUFFERS] = {},
@@ -1031,7 +1008,7 @@ static int __io_sq_thread_acquire_mm(struct io_ring_ctx *ctx)
static int io_sq_thread_acquire_mm(struct io_ring_ctx *ctx,
struct io_kiocb *req)
{
- if (!io_op_defs[req->opcode].needs_mm)
+ if (!(io_op_defs[req->opcode].work_flags & IO_WQ_WORK_MM))
return 0;
return __io_sq_thread_acquire_mm(ctx);
}
@@ -1224,7 +1201,8 @@ static void io_prep_async_work(struct io_kiocb *req)
if (def->unbound_nonreg_file)
req->work.flags |= IO_WQ_WORK_UNBOUND;
}
- if (!req->work.files && io_op_defs[req->opcode].file_table &&
+ if (!req->work.files &&
+ (io_op_defs[req->opcode].work_flags & IO_WQ_WORK_FILES) &&
!(req->flags & REQ_F_NO_FILE_TABLE)) {
req->work.files = get_files_struct(current);
get_nsproxy(current->nsproxy);
@@ -1235,12 +1213,12 @@ static void io_prep_async_work(struct io_kiocb *req)
list_add(&req->inflight_entry, &ctx->inflight_list);
spin_unlock_irq(&ctx->inflight_lock);
}
- if (!req->work.mm && def->needs_mm) {
+ if (!req->work.mm && (def->work_flags & IO_WQ_WORK_MM)) {
mmgrab(current->mm);
req->work.mm = current->mm;
}
#ifdef CONFIG_BLK_CGROUP
- if (!req->work.blkcg_css && def->needs_blkcg) {
+ if (!req->work.blkcg_css && (def->work_flags & IO_WQ_WORK_BLKCG)) {
rcu_read_lock();
req->work.blkcg_css = blkcg_css();
/*
@@ -1254,7 +1232,7 @@ static void io_prep_async_work(struct io_kiocb *req)
#endif
if (!req->work.creds)
req->work.creds = get_current_cred();
- if (!req->work.fs && def->needs_fs) {
+ if (!req->work.fs && (def->work_flags & IO_WQ_WORK_FS)) {
spin_lock(¤t->fs->lock);
if (!current->fs->in_exec) {
req->work.fs = current->fs;
@@ -1268,6 +1246,8 @@ static void io_prep_async_work(struct io_kiocb *req)
req->work.fsize = rlimit(RLIMIT_FSIZE);
else
req->work.fsize = RLIM_INFINITY;
+
+ req->work.flags |= def->work_flags;
}
static void io_prep_async_link(struct io_kiocb *req)
--
2.28.0
next prev parent reply other threads:[~2020-10-16 16:02 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-10-16 16:02 [PATCHES 0/10] Fixes queued up for 5.10 Jens Axboe
2020-10-16 16:02 ` [PATCH 01/18] io_uring: Fix sizeof() mismatch Jens Axboe
2020-10-16 16:02 ` [PATCH 02/18] readahead: use limited read-ahead to satisfy read Jens Axboe
2020-10-16 16:02 ` [PATCH 03/18] io_uring: don't clear IOCB_NOWAIT for async buffered retry Jens Axboe
2020-10-16 16:02 ` [PATCH 04/18] io_uring: don't set COMP_LOCKED if won't put Jens Axboe
2020-10-16 16:02 ` [PATCH 05/18] io_uring: don't unnecessarily clear F_LINK_TIMEOUT Jens Axboe
2020-10-16 16:02 ` [PATCH 06/18] io_uring: don't put a poll req under spinlock Jens Axboe
2020-10-16 16:02 ` [PATCH 07/18] io_uring: dig out COMP_LOCK from deep call chain Jens Axboe
2020-10-16 16:02 ` [PATCH 08/18] io_uring: fix REQ_F_COMP_LOCKED by killing it Jens Axboe
2020-10-16 16:02 ` [PATCH 09/18] Revert "io_uring: mark io_uring_fops/io_op_defs as __read_mostly" Jens Axboe
2020-10-16 16:02 ` [PATCH 10/18] io_uring: fix error path cleanup in io_sqe_files_register() Jens Axboe
2020-10-16 16:02 ` [PATCH 11/18] io-wq: assign NUMA node locality if appropriate Jens Axboe
2020-10-16 16:02 ` Jens Axboe [this message]
2020-10-16 16:02 ` [PATCH 13/18] io_uring: rely solely on work flags to determine personality Jens Axboe
2020-10-16 16:02 ` [PATCH 14/18] io_uring: move io identity items into separate struct Jens Axboe
2020-10-16 16:02 ` [PATCH 15/18] io_uring: COW io_identity on mismatch Jens Axboe
2020-10-16 16:02 ` [PATCH 16/18] io_uring: store io_identity in io_uring_task Jens Axboe
2020-10-16 16:02 ` [PATCH 17/18] io_uring: assign new io_identity for task if members have changed Jens Axboe
2020-10-16 16:02 ` [PATCH 18/18] io_uring: use percpu counters to track inflight requests Jens Axboe
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 \
[email protected] \
[email protected] \
[email protected] \
/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