From: Pavel Begunkov <[email protected]>
To: Jens Axboe <[email protected]>, [email protected]
Subject: [PATCH 1/1] io_uring: allow empty slots for reg buffers
Date: Wed, 28 Apr 2021 13:11:29 +0100 [thread overview]
Message-ID: <7e95e4d700082baaf010c648c72ac764c9cc8826.1619611868.git.asml.silence@gmail.com> (raw)
Allow empty reg buffer slots any request using which should fail. This
allows users to not register all buffers in advance, but do it lazily
and/or on demand via updates. That is achieved by setting iov_base and
iov_len to zero for registration and/or buffer updates. Empty buffer
can't have a non-zero tag.
Implementation details: to not add extra overhead to io_import_fixed(),
create a dummy buffer crafted to fail any request using it, and set it
to all empty buffer slots.
Signed-off-by: Pavel Begunkov <[email protected]>
---
fs/io_uring.c | 36 +++++++++++++++++++++++++++++-------
1 file changed, 29 insertions(+), 7 deletions(-)
diff --git a/fs/io_uring.c b/fs/io_uring.c
index a48b88b3e289..43b00077dbd3 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -456,6 +456,7 @@ struct io_ring_ctx {
spinlock_t rsrc_ref_lock;
struct io_rsrc_node *rsrc_node;
struct io_rsrc_node *rsrc_backup_node;
+ struct io_mapped_ubuf *dummy_ubuf;
struct io_restriction restrictions;
@@ -1158,6 +1159,12 @@ static struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p)
goto err;
__hash_init(ctx->cancel_hash, 1U << hash_bits);
+ ctx->dummy_ubuf = kzalloc(sizeof(*ctx->dummy_ubuf), GFP_KERNEL);
+ if (!ctx->dummy_ubuf)
+ goto err;
+ /* set invalid range, so io_import_fixed() fails meeting it */
+ ctx->dummy_ubuf->ubuf = -1UL;
+
if (percpu_ref_init(&ctx->refs, io_ring_ctx_ref_free,
PERCPU_REF_ALLOW_REINIT, GFP_KERNEL))
goto err;
@@ -1185,6 +1192,7 @@ static struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p)
INIT_LIST_HEAD(&ctx->submit_state.comp.locked_free_list);
return ctx;
err:
+ kfree(ctx->dummy_ubuf);
kfree(ctx->cancel_hash);
kfree(ctx);
return NULL;
@@ -8106,11 +8114,13 @@ static void io_buffer_unmap(struct io_ring_ctx *ctx, struct io_mapped_ubuf **slo
struct io_mapped_ubuf *imu = *slot;
unsigned int i;
- for (i = 0; i < imu->nr_bvecs; i++)
- unpin_user_page(imu->bvec[i].bv_page);
- if (imu->acct_pages)
- io_unaccount_mem(ctx, imu->acct_pages);
- kvfree(imu);
+ if (imu != ctx->dummy_ubuf) {
+ for (i = 0; i < imu->nr_bvecs; i++)
+ unpin_user_page(imu->bvec[i].bv_page);
+ if (imu->acct_pages)
+ io_unaccount_mem(ctx, imu->acct_pages);
+ kvfree(imu);
+ }
*slot = NULL;
}
@@ -8250,6 +8260,11 @@ static int io_sqe_buffer_register(struct io_ring_ctx *ctx, struct iovec *iov,
size_t size;
int ret, pret, nr_pages, i;
+ if (!iov->iov_base) {
+ *pimu = ctx->dummy_ubuf;
+ return 0;
+ }
+
ubuf = (unsigned long) iov->iov_base;
end = (ubuf + iov->iov_len + PAGE_SIZE - 1) >> PAGE_SHIFT;
start = ubuf >> PAGE_SHIFT;
@@ -8347,7 +8362,9 @@ static int io_buffer_validate(struct iovec *iov)
* constraints here, we'll -EINVAL later when IO is
* submitted if they are wrong.
*/
- if (!iov->iov_base || !iov->iov_len)
+ if (!iov->iov_base)
+ return iov->iov_len ? -EFAULT : 0;
+ if (!iov->iov_len)
return -EFAULT;
/* arbitrary limit, but we need something */
@@ -8397,6 +8414,8 @@ static int io_sqe_buffers_register(struct io_ring_ctx *ctx, void __user *arg,
ret = io_buffer_validate(&iov);
if (ret)
break;
+ if (!iov.iov_base && tag)
+ return -EINVAL;
ret = io_sqe_buffer_register(ctx, &iov, &ctx->user_bufs[i],
&last_hpage);
@@ -8446,12 +8465,14 @@ static int __io_sqe_buffers_update(struct io_ring_ctx *ctx,
err = io_buffer_validate(&iov);
if (err)
break;
+ if (!iov.iov_base && tag)
+ return -EINVAL;
err = io_sqe_buffer_register(ctx, &iov, &imu, &last_hpage);
if (err)
break;
i = array_index_nospec(offset, ctx->nr_user_bufs);
- if (ctx->user_bufs[i]) {
+ if (ctx->user_bufs[i] != ctx->dummy_ubuf) {
err = io_queue_rsrc_removal(ctx->buf_data, offset,
ctx->rsrc_node, ctx->user_bufs[i]);
if (unlikely(err)) {
@@ -8599,6 +8620,7 @@ static void io_ring_ctx_free(struct io_ring_ctx *ctx)
if (ctx->hash_map)
io_wq_put_hash(ctx->hash_map);
kfree(ctx->cancel_hash);
+ kfree(ctx->dummy_ubuf);
kfree(ctx);
}
--
2.31.1
next reply other threads:[~2021-04-28 12:11 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-04-28 12:11 Pavel Begunkov [this message]
2021-04-28 14:06 ` [PATCH 1/1] io_uring: allow empty slots for reg buffers 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 \
--in-reply-to=7e95e4d700082baaf010c648c72ac764c9cc8826.1619611868.git.asml.silence@gmail.com \
[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