From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Tejun Heo <tj@kernel.org>,
Jann Horn <jannh@google.com>,
Kent Overstreet <kent.overstreet@gmail.com>,
Linus Torvalds <torvalds@linux-foundation.org>
Subject: [PATCH 3.18 64/68] fs/aio: Add explicit RCU grace period when freeing kioctx
Date: Mon, 19 Mar 2018 19:06:42 +0100 [thread overview]
Message-ID: <20180319171836.857305687@linuxfoundation.org> (raw)
In-Reply-To: <20180319171827.899658615@linuxfoundation.org>
3.18-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tejun Heo <tj@kernel.org>
commit a6d7cff472eea87d96899a20fa718d2bab7109f3 upstream.
While fixing refcounting, e34ecee2ae79 ("aio: Fix a trinity splat")
incorrectly removed explicit RCU grace period before freeing kioctx.
The intention seems to be depending on the internal RCU grace periods
of percpu_ref; however, percpu_ref uses a different flavor of RCU,
sched-RCU. This can lead to kioctx being freed while RCU read
protected dereferences are still in progress.
Fix it by updating free_ioctx() to go through call_rcu() explicitly.
v2: Comment added to explain double bouncing.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Jann Horn <jannh@google.com>
Fixes: e34ecee2ae79 ("aio: Fix a trinity splat")
Cc: Kent Overstreet <kent.overstreet@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: stable@vger.kernel.org # v3.13+
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
fs/aio.c | 23 +++++++++++++++++++----
1 file changed, 19 insertions(+), 4 deletions(-)
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -110,7 +110,8 @@ struct kioctx {
struct page **ring_pages;
long nr_pages;
- struct work_struct free_work;
+ struct rcu_head free_rcu;
+ struct work_struct free_work; /* see free_ioctx() */
/*
* signals when all in-flight requests are done
@@ -505,6 +506,12 @@ static int kiocb_cancel(struct kiocb *ki
return cancel(kiocb);
}
+/*
+ * free_ioctx() should be RCU delayed to synchronize against the RCU
+ * protected lookup_ioctx() and also needs process context to call
+ * aio_free_ring(), so the double bouncing through kioctx->free_rcu and
+ * ->free_work.
+ */
static void free_ioctx(struct work_struct *work)
{
struct kioctx *ctx = container_of(work, struct kioctx, free_work);
@@ -518,6 +525,14 @@ static void free_ioctx(struct work_struc
kmem_cache_free(kioctx_cachep, ctx);
}
+static void free_ioctx_rcufn(struct rcu_head *head)
+{
+ struct kioctx *ctx = container_of(head, struct kioctx, free_rcu);
+
+ INIT_WORK(&ctx->free_work, free_ioctx);
+ schedule_work(&ctx->free_work);
+}
+
static void free_ioctx_reqs(struct percpu_ref *ref)
{
struct kioctx *ctx = container_of(ref, struct kioctx, reqs);
@@ -526,8 +541,8 @@ static void free_ioctx_reqs(struct percp
if (ctx->requests_done)
complete(ctx->requests_done);
- INIT_WORK(&ctx->free_work, free_ioctx);
- schedule_work(&ctx->free_work);
+ /* Synchronize against RCU protected table->table[] dereferences */
+ call_rcu(&ctx->free_rcu, free_ioctx_rcufn);
}
/*
@@ -749,7 +764,7 @@ static int kill_ioctx(struct mm_struct *
table->table[ctx->id] = NULL;
spin_unlock(&mm->ioctx_lock);
- /* percpu_ref_kill() will do the necessary call_rcu() */
+ /* free_ioctx_reqs() will do the necessary RCU synchronization */
wake_up_all(&ctx->wait);
/*
next prev parent reply other threads:[~2018-03-19 18:06 UTC|newest]
Thread overview: 74+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-03-19 18:05 [PATCH 3.18 00/68] 3.18.101-stable review Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 01/68] Input: tsc2007 - check for presence and power down tsc2007 during probe Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 03/68] net: mvpp2: set dma mask and coherent dma mask on PPv2.2 Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 04/68] PCI/MSI: Stop disabling MSI/MSI-X in pci_device_shutdown() Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 05/68] selinux: check for address length in selinux_socket_bind() Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 06/68] perf tools: Make perf_event__synthesize_mmap_events() scale Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 07/68] drivers: net: xgene: Fix hardware checksum setting Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 09/68] ath10k: disallow DFS simulation if DFS channel is not enabled Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 10/68] HID: clamp input to logical range if no null state Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 11/68] ARM: dts: Adjust moxart IRQ controller and flags Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 12/68] batman-adv: handle race condition for claims between gateways Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 13/68] of: fix of_device_get_modalias returned length when truncating buffers Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 14/68] scsi: ipr: Fix missed EH wakeup Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 15/68] [media] media: i2c/soc_camera: fix ov6650 sensor getting wrong clock Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 16/68] timers, sched_clock: Update timeout for clock wrap Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 17/68] sched: act_csum: dont mangle TCP and UDP GSO packets Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 18/68] spi: omap2-mcspi: poll OMAP2_MCSPI_CHSTAT_RXS for PIO transfer Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 19/68] tcp: sysctl: Fix a race to avoid unexpected 0 window from space Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 20/68] mm: Fix false-positive VM_BUG_ON() in page_cache_{get,add}_speculative() Greg Kroah-Hartman
2018-03-19 18:05 ` [PATCH 3.18 21/68] blk-throttle: make sure expire time isnt too big Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 22/68] ARM: DRA7: hwmod_data: Prevent wait_target_disable error for usb_otg_ss Greg Kroah-Hartman
2018-03-21 10:37 ` Roger Quadros
2018-03-21 11:02 ` Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 23/68] braille-console: Fix value returned by _braille_console_setup Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 24/68] ARM: dts: r8a7790: Correct parent of SSI[0-9] clocks Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 25/68] ARM: dts: r8a7791: " Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 26/68] powerpc: Avoid taking a data miss on every userspace instruction miss Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 27/68] net/faraday: Add missing include of of.h Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 28/68] reiserfs: Make cancel_old_flush() reliable Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 29/68] fm10k: correctly check if interface is removed Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 30/68] apparmor: Make path_max parameter readonly Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 31/68] iommu/iova: Fix underflow bug in __alloc_and_insert_iova_range Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 32/68] video: ARM CLCD: fix dma allocation size Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 34/68] MIPS: BPF: Quit clobbering callee saved registers in JIT code Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 36/68] usb: gadget: dummy_hcd: Fix wrong power status bit clear/reset in dummy_hub_control() Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 37/68] perf inject: Copy events when reordering events in pipe mode Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 38/68] perf session: Dont rely on evlist " Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 39/68] scsi: sg: check for valid direction before starting the request Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 40/68] scsi: sg: close race condition in sg_remove_sfp_usercontext() Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 41/68] kprobes/x86: Fix kprobe-booster not to boost far call instructions Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 42/68] kprobes/x86: Set kprobes pages read-only Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 43/68] wil6210: fix memory access violation in wil_memcpy_from/toio_32 Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 44/68] HID: elo: clear BTN_LEFT mapping Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 45/68] sched: Stop resched_cpu() from sending IPIs to offline CPUs Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 46/68] net: xfrm: allow clearing socket xfrm policies Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 47/68] mtd: nand: fix interpretation of NAND_CMD_NONE in nand_command[_lp]() Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 48/68] ARM: dts: am335x-pepper: Fix the audio CODECs reset pin Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 49/68] ARM: dts: omap3-n900: " Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 50/68] mtd: nand: ifc: update bufnum mask for ver >= 2.0.0 Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 52/68] spi: sun6i: disable/unprepare clocks on remove Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 53/68] scsi: devinfo: apply to HP XP the same flags as Hitachi VSP Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 54/68] media: cpia2: Fix a couple off by one bugs Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 55/68] veth: set peer GSO values Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 56/68] mac80211: remove BUG() when interface type is invalid Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 57/68] ASoC: nuc900: Fix a loop timeout test Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 58/68] rcutorture/configinit: Fix build directory error message Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 59/68] ima: relax requiring a file signature for new files with zero length Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 60/68] ALSA: pcm: Fix UAF in snd_pcm_oss_get_formats() Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 61/68] ALSA: seq: Fix possible UAF in snd_seq_check_queue() Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 62/68] ALSA: seq: Clear client entry before deleting else at closing Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 63/68] lock_parent() needs to recheck if dentry got __dentry_killed under it Greg Kroah-Hartman
2018-03-19 18:06 ` Greg Kroah-Hartman [this message]
2018-03-19 18:06 ` [PATCH 3.18 65/68] fs/aio: Use RCU accessors for kioctx_table->table[] Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 66/68] scsi: sg: fix SG_DXFER_FROM_DEV transfers Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 67/68] scsi: sg: fix static checker warning in sg_is_valid_dxfer Greg Kroah-Hartman
2018-03-19 18:06 ` [PATCH 3.18 68/68] scsi: sg: only check for dxfer_len greater than 256M Greg Kroah-Hartman
2018-03-20 14:47 ` [PATCH 3.18 00/68] 3.18.101-stable review Guenter Roeck
2018-03-21 11:03 ` Greg Kroah-Hartman
2018-03-20 17:34 ` Shuah Khan
[not found] ` <CALpmF+Fu1EHkBrG__6u8mY2LNa4jpq5ka78Oaa97X14FPwLLQQ@mail.gmail.com>
2018-03-21 10:05 ` Greg Kroah-Hartman
2018-03-21 11:04 ` Greg Kroah-Hartman
2018-03-21 17:47 ` Guenter Roeck
2018-03-22 8:21 ` Greg Kroah-Hartman
2018-03-22 16:39 ` Guenter Roeck
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=20180319171836.857305687@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=jannh@google.com \
--cc=kent.overstreet@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=stable@vger.kernel.org \
--cc=tj@kernel.org \
--cc=torvalds@linux-foundation.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;
as well as URLs for NNTP newsgroup(s).