From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
patches@lists.linux.dev, Chase Hiltz <chase@path.net>,
Jens Axboe <axboe@kernel.dk>
Subject: [PATCH 6.1 69/92] io_uring: always arm linked timeouts prior to issue
Date: Mon, 12 May 2025 19:45:44 +0200 [thread overview]
Message-ID: <20250512172025.936349590@linuxfoundation.org> (raw)
In-Reply-To: <20250512172023.126467649@linuxfoundation.org>
6.1-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jens Axboe <axboe@kernel.dk>
Commit b53e523261bf058ea4a518b482222e7a277b186b upstream.
There are a few spots where linked timeouts are armed, and not all of
them adhere to the pre-arm, attempt issue, post-arm pattern. This can
be problematic if the linked request returns that it will trigger a
callback later, and does so before the linked timeout is fully armed.
Consolidate all the linked timeout handling into __io_issue_sqe(),
rather than have it spread throughout the various issue entry points.
Cc: stable@vger.kernel.org
Link: https://github.com/axboe/liburing/issues/1390
Reported-by: Chase Hiltz <chase@path.net>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
io_uring/io_uring.c | 53 +++++++++++++++-------------------------------------
1 file changed, 16 insertions(+), 37 deletions(-)
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -372,24 +372,6 @@ static struct io_kiocb *__io_prep_linked
return req->link;
}
-static inline struct io_kiocb *io_prep_linked_timeout(struct io_kiocb *req)
-{
- if (likely(!(req->flags & REQ_F_ARM_LTIMEOUT)))
- return NULL;
- return __io_prep_linked_timeout(req);
-}
-
-static noinline void __io_arm_ltimeout(struct io_kiocb *req)
-{
- io_queue_linked_timeout(__io_prep_linked_timeout(req));
-}
-
-static inline void io_arm_ltimeout(struct io_kiocb *req)
-{
- if (unlikely(req->flags & REQ_F_ARM_LTIMEOUT))
- __io_arm_ltimeout(req);
-}
-
static void io_prep_async_work(struct io_kiocb *req)
{
const struct io_op_def *def = &io_op_defs[req->opcode];
@@ -437,7 +419,6 @@ static void io_prep_async_link(struct io
static void io_queue_iowq(struct io_kiocb *req)
{
- struct io_kiocb *link = io_prep_linked_timeout(req);
struct io_uring_task *tctx = req->task->io_uring;
BUG_ON(!tctx);
@@ -462,8 +443,6 @@ static void io_queue_iowq(struct io_kioc
trace_io_uring_queue_async_work(req, io_wq_is_hashed(&req->work));
io_wq_enqueue(tctx->io_wq, &req->work);
- if (link)
- io_queue_linked_timeout(link);
}
static __cold void io_queue_deferred(struct io_ring_ctx *ctx)
@@ -1741,17 +1720,24 @@ static bool io_assign_file(struct io_kio
return !!req->file;
}
+#define REQ_ISSUE_SLOW_FLAGS (REQ_F_CREDS | REQ_F_ARM_LTIMEOUT)
+
static int io_issue_sqe(struct io_kiocb *req, unsigned int issue_flags)
{
const struct io_op_def *def = &io_op_defs[req->opcode];
const struct cred *creds = NULL;
+ struct io_kiocb *link = NULL;
int ret;
if (unlikely(!io_assign_file(req, issue_flags)))
return -EBADF;
- if (unlikely((req->flags & REQ_F_CREDS) && req->creds != current_cred()))
- creds = override_creds(req->creds);
+ if (unlikely(req->flags & REQ_ISSUE_SLOW_FLAGS)) {
+ if ((req->flags & REQ_F_CREDS) && req->creds != current_cred())
+ creds = override_creds(req->creds);
+ if (req->flags & REQ_F_ARM_LTIMEOUT)
+ link = __io_prep_linked_timeout(req);
+ }
if (!def->audit_skip)
audit_uring_entry(req->opcode);
@@ -1761,8 +1747,12 @@ static int io_issue_sqe(struct io_kiocb
if (!def->audit_skip)
audit_uring_exit(!ret, ret);
- if (creds)
- revert_creds(creds);
+ if (unlikely(creds || link)) {
+ if (creds)
+ revert_creds(creds);
+ if (link)
+ io_queue_linked_timeout(link);
+ }
if (ret == IOU_OK) {
if (issue_flags & IO_URING_F_COMPLETE_DEFER)
@@ -1809,8 +1799,6 @@ void io_wq_submit_work(struct io_wq_work
else
req_ref_get(req);
- io_arm_ltimeout(req);
-
/* either cancelled or io-wq is dying, so don't touch tctx->iowq */
if (work->flags & IO_WQ_WORK_CANCEL) {
fail:
@@ -1908,15 +1896,11 @@ struct file *io_file_get_normal(struct i
static void io_queue_async(struct io_kiocb *req, int ret)
__must_hold(&req->ctx->uring_lock)
{
- struct io_kiocb *linked_timeout;
-
if (ret != -EAGAIN || (req->flags & REQ_F_NOWAIT)) {
io_req_complete_failed(req, ret);
return;
}
- linked_timeout = io_prep_linked_timeout(req);
-
switch (io_arm_poll_handler(req, 0)) {
case IO_APOLL_READY:
io_kbuf_recycle(req, 0);
@@ -1929,9 +1913,6 @@ static void io_queue_async(struct io_kio
case IO_APOLL_OK:
break;
}
-
- if (linked_timeout)
- io_queue_linked_timeout(linked_timeout);
}
static inline void io_queue_sqe(struct io_kiocb *req)
@@ -1945,9 +1926,7 @@ static inline void io_queue_sqe(struct i
* We async punt it if the file wasn't marked NOWAIT, or if the file
* doesn't support non-blocking read/write attempts
*/
- if (likely(!ret))
- io_arm_ltimeout(req);
- else
+ if (unlikely(ret))
io_queue_async(req, ret);
}
next prev parent reply other threads:[~2025-05-12 17:55 UTC|newest]
Thread overview: 102+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-05-12 17:44 [PATCH 6.1 00/92] 6.1.139-rc1 review Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 01/92] dm: add missing unlock on in dm_keyslot_evict() Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 02/92] arm64: dts: imx8mm-verdin: Link reg_usdhc2_vqmmc to usdhc2 Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 03/92] can: mcan: m_can_class_unregister(): fix order of unregistration calls Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 04/92] can: mcp251xfd: mcp251xfd_remove(): " Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 05/92] ksmbd: prevent out-of-bounds stream writes by validating *pos Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 06/92] openvswitch: Fix unsafe attribute parsing in output_userspace() Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 07/92] ksmbd: fix memory leak in parse_lease_state() Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 08/92] sch_htb: make htb_deactivate() idempotent Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 09/92] gre: Fix again IPv6 link-local address generation Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 10/92] can: mcp251xfd: fix TDC setting for low data bit rates Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 11/92] rcu/kvfree: Add kvfree_rcu_mightsleep() and kfree_rcu_mightsleep() Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 12/92] can: gw: fix RCU/BH usage in cgw_create_job() Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 13/92] ipv4: Drop tos parameter from flowi4_update_output() Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 14/92] ipvs: fix uninit-value for saddr in do_output_route4 Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 15/92] netfilter: ipset: fix region locking in hash types Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 16/92] bpf: Scrub packet on bpf_redirect_peer Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 17/92] net: dsa: b53: allow leaky reserved multicast Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 18/92] net: dsa: b53: fix clearing PVID of a port Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 19/92] net: dsa: b53: fix flushing old pvid VLAN on pvid change Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 20/92] net: dsa: b53: fix VLAN ID for untagged vlan on bridge leave Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 21/92] net: dsa: b53: always rejoin default untagged VLAN " Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 22/92] net: dsa: b53: fix learning on VLAN unaware bridges Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 23/92] Input: mtk-pmic-keys - fix possible null pointer dereference Greg Kroah-Hartman
2025-05-12 17:44 ` [PATCH 6.1 24/92] Input: synaptics - enable InterTouch on Dynabook Portege X30-D Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 25/92] Input: synaptics - enable InterTouch on Dynabook Portege X30L-G Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 26/92] Input: synaptics - enable InterTouch on Dell Precision M3800 Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 27/92] Input: synaptics - enable SMBus for HP Elitebook 850 G1 Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 28/92] Input: synaptics - enable InterTouch on TUXEDO InfinityBook Pro 14 v5 Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 29/92] staging: iio: adc: ad7816: Correct conditional logic for store mode Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 30/92] staging: axis-fifo: Remove hardware resets for user errors Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 31/92] staging: axis-fifo: Correct handling of tx_fifo_depth for size validation Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 32/92] x86/mm: Eliminate window where TLB flushes may be inadvertently skipped Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 33/92] drm/amd/display: Shift DMUB AUX reply command if necessary Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 34/92] iio: adc: ad7606: fix serial register access Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 35/92] iio: adis16201: Correct inclinometer channel resolution Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 36/92] iio: imu: st_lsm6dsx: fix possible lockup in st_lsm6dsx_read_fifo Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 37/92] iio: imu: st_lsm6dsx: fix possible lockup in st_lsm6dsx_read_tagged_fifo Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 38/92] drm/v3d: Add job to pending list if the reset was skipped Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 39/92] drm/amd/display: Fix the checking condition in dmub aux handling Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 40/92] drm/amd/display: Remove incorrect checking in dmub aux handler Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 41/92] drm/amd/display: Fix wrong handling for AUX_DEFER case Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 42/92] drm/amd/display: Copy AUX read reply data whenever length > 0 Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 43/92] drm/amdgpu/hdp5.2: use memcfg register to post the write for HDP flush Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 44/92] usb: uhci-platform: Make the clock really optional Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 45/92] xenbus: Use kref to track req lifetime Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 46/92] module: ensure that kobject_put() is safe for module type kobjects Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 47/92] ocfs2: switch osb->disable_recovery to enum Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 48/92] ocfs2: implement handshaking with ocfs2 recovery thread Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 49/92] ocfs2: stop quota recovery before disabling quotas Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 50/92] usb: cdnsp: Fix issue with resuming from L1 Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 51/92] usb: cdnsp: fix L1 resume issue for RTL_REVISION_NEW_LPM version Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 52/92] usb: gadget: tegra-xudc: ACK ST_RC after clearing CTRL_RUN Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 53/92] usb: host: tegra: Prevent host controller crash when OTG port is used Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 54/92] usb: typec: tcpm: delay SNK_TRY_WAIT_DEBOUNCE to SRC_TRYWAIT transition Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 55/92] usb: typec: ucsi: displayport: Fix NULL pointer access Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 56/92] USB: usbtmc: use interruptible sleep in usbtmc_read Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 57/92] usb: usbtmc: Fix erroneous get_stb ioctl error returns Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 58/92] usb: usbtmc: Fix erroneous wait_srq ioctl return Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 59/92] usb: usbtmc: Fix erroneous generic_read " Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 60/92] iio: accel: adxl367: fix setting odr for activity time update Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 61/92] iio: temp: maxim-thermocouple: Fix potential lack of DMA safe buffer Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 62/92] types: Complement the aligned types with signed 64-bit one Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 63/92] iio: accel: adxl355: Make timestamp 64-bit aligned using aligned_s64 Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 64/92] iio: adc: dln2: Use aligned_s64 for timestamp Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 65/92] MIPS: Fix MAX_REG_OFFSET Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 66/92] drm/panel: simple: Update timings for AUO G101EVN010 Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 67/92] nvme: unblock ctrl state transition for firmware update Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 68/92] do_umount(): add missing barrier before refcount checks in sync case Greg Kroah-Hartman
2025-05-12 17:45 ` Greg Kroah-Hartman [this message]
2025-05-12 17:45 ` [PATCH 6.1 70/92] io_uring: ensure deferred completions are posted for multishot Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 71/92] Revert "net: phy: microchip: force IRQ polling mode for lan88xx" Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 72/92] arm64: insn: Add support for encoding DSB Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 73/92] arm64: proton-pack: Expose whether the platform is mitigated by firmware Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 74/92] arm64: proton-pack: Expose whether the branchy loop k value Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 75/92] arm64: bpf: Add BHB mitigation to the epilogue for cBPF programs Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 76/92] arm64: bpf: Only mitigate cBPF programs loaded by unprivileged users Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 77/92] arm64: proton-pack: Add new CPUs k values for branch mitigation Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 78/92] x86/bpf: Call branch history clearing sequence on exit Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 79/92] x86/bpf: Add IBHF call at end of classic BPF Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 80/92] x86/bhi: Do not set BHI_DIS_S in 32-bit mode Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 81/92] x86/speculation: Simplify and make CALL_NOSPEC consistent Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 82/92] x86/speculation: Add a conditional CS prefix to CALL_NOSPEC Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 83/92] x86/speculation: Remove the extra #ifdef around CALL_NOSPEC Greg Kroah-Hartman
2025-05-12 17:45 ` [PATCH 6.1 84/92] Documentation: x86/bugs/its: Add ITS documentation Greg Kroah-Hartman
2025-05-12 17:46 ` [PATCH 6.1 85/92] x86/its: Enumerate Indirect Target Selection (ITS) bug Greg Kroah-Hartman
2025-05-12 17:46 ` [PATCH 6.1 86/92] x86/its: Add support for ITS-safe indirect thunk Greg Kroah-Hartman
2025-05-12 17:46 ` [PATCH 6.1 87/92] x86/its: Add support for ITS-safe return thunk Greg Kroah-Hartman
2025-05-12 17:46 ` [PATCH 6.1 88/92] x86/its: Enable Indirect Target Selection mitigation Greg Kroah-Hartman
2025-05-12 17:46 ` [PATCH 6.1 89/92] x86/its: Add "vmexit" option to skip mitigation on some CPUs Greg Kroah-Hartman
2025-05-12 17:46 ` [PATCH 6.1 90/92] x86/its: Align RETs in BHB clear sequence to avoid thunking Greg Kroah-Hartman
2025-05-12 17:46 ` [PATCH 6.1 91/92] x86/ibt: Keep IBT disabled during alternative patching Greg Kroah-Hartman
2025-05-12 17:46 ` [PATCH 6.1 92/92] x86/its: Use dynamic thunks for indirect branches Greg Kroah-Hartman
2025-05-12 20:56 ` [PATCH 6.1 00/92] 6.1.139-rc1 review Jon Hunter
2025-05-13 6:45 ` Pavel Machek
2025-05-13 9:43 ` Florian Fainelli
2025-05-13 9:48 ` Mark Brown
2025-05-13 10:04 ` Ron Economos
2025-05-13 11:39 ` Peter Schneider
2025-05-13 17:19 ` Naresh Kamboju
2025-05-13 17:32 ` Shuah Khan
2025-05-14 17:11 ` Hardik Garg
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=20250512172025.936349590@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=axboe@kernel.dk \
--cc=chase@path.net \
--cc=patches@lists.linux.dev \
--cc=stable@vger.kernel.org \
/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