From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: John Sperbeck <jsperbeck@google.com>,
Dennis Zhou <dennis@kernel.org>, Sasha Levin <sashal@kernel.org>,
linux-mm@kvack.org, netdev@vger.kernel.org, bpf@vger.kernel.org
Subject: [PATCH AUTOSEL 4.19 043/141] percpu: remove spurious lock dependency between percpu and sched
Date: Sat, 1 Jun 2019 09:20:19 -0400 [thread overview]
Message-ID: <20190601132158.25821-43-sashal@kernel.org> (raw)
In-Reply-To: <20190601132158.25821-1-sashal@kernel.org>
From: John Sperbeck <jsperbeck@google.com>
[ Upstream commit 198790d9a3aeaef5792d33a560020861126edc22 ]
In free_percpu() we sometimes call pcpu_schedule_balance_work() to
queue a work item (which does a wakeup) while holding pcpu_lock.
This creates an unnecessary lock dependency between pcpu_lock and
the scheduler's pi_lock. There are other places where we call
pcpu_schedule_balance_work() without hold pcpu_lock, and this case
doesn't need to be different.
Moving the call outside the lock prevents the following lockdep splat
when running tools/testing/selftests/bpf/{test_maps,test_progs} in
sequence with lockdep enabled:
======================================================
WARNING: possible circular locking dependency detected
5.1.0-dbg-DEV #1 Not tainted
------------------------------------------------------
kworker/23:255/18872 is trying to acquire lock:
000000000bc79290 (&(&pool->lock)->rlock){-.-.}, at: __queue_work+0xb2/0x520
but task is already holding lock:
00000000e3e7a6aa (pcpu_lock){..-.}, at: free_percpu+0x36/0x260
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #4 (pcpu_lock){..-.}:
lock_acquire+0x9e/0x180
_raw_spin_lock_irqsave+0x3a/0x50
pcpu_alloc+0xfa/0x780
__alloc_percpu_gfp+0x12/0x20
alloc_htab_elem+0x184/0x2b0
__htab_percpu_map_update_elem+0x252/0x290
bpf_percpu_hash_update+0x7c/0x130
__do_sys_bpf+0x1912/0x1be0
__x64_sys_bpf+0x1a/0x20
do_syscall_64+0x59/0x400
entry_SYSCALL_64_after_hwframe+0x49/0xbe
-> #3 (&htab->buckets[i].lock){....}:
lock_acquire+0x9e/0x180
_raw_spin_lock_irqsave+0x3a/0x50
htab_map_update_elem+0x1af/0x3a0
-> #2 (&rq->lock){-.-.}:
lock_acquire+0x9e/0x180
_raw_spin_lock+0x2f/0x40
task_fork_fair+0x37/0x160
sched_fork+0x211/0x310
copy_process.part.43+0x7b1/0x2160
_do_fork+0xda/0x6b0
kernel_thread+0x29/0x30
rest_init+0x22/0x260
arch_call_rest_init+0xe/0x10
start_kernel+0x4fd/0x520
x86_64_start_reservations+0x24/0x26
x86_64_start_kernel+0x6f/0x72
secondary_startup_64+0xa4/0xb0
-> #1 (&p->pi_lock){-.-.}:
lock_acquire+0x9e/0x180
_raw_spin_lock_irqsave+0x3a/0x50
try_to_wake_up+0x41/0x600
wake_up_process+0x15/0x20
create_worker+0x16b/0x1e0
workqueue_init+0x279/0x2ee
kernel_init_freeable+0xf7/0x288
kernel_init+0xf/0x180
ret_from_fork+0x24/0x30
-> #0 (&(&pool->lock)->rlock){-.-.}:
__lock_acquire+0x101f/0x12a0
lock_acquire+0x9e/0x180
_raw_spin_lock+0x2f/0x40
__queue_work+0xb2/0x520
queue_work_on+0x38/0x80
free_percpu+0x221/0x260
pcpu_freelist_destroy+0x11/0x20
stack_map_free+0x2a/0x40
bpf_map_free_deferred+0x3c/0x50
process_one_work+0x1f7/0x580
worker_thread+0x54/0x410
kthread+0x10f/0x150
ret_from_fork+0x24/0x30
other info that might help us debug this:
Chain exists of:
&(&pool->lock)->rlock --> &htab->buckets[i].lock --> pcpu_lock
Possible unsafe locking scenario:
CPU0 CPU1
---- ----
lock(pcpu_lock);
lock(&htab->buckets[i].lock);
lock(pcpu_lock);
lock(&(&pool->lock)->rlock);
*** DEADLOCK ***
3 locks held by kworker/23:255/18872:
#0: 00000000b36a6e16 ((wq_completion)events){+.+.},
at: process_one_work+0x17a/0x580
#1: 00000000dfd966f0 ((work_completion)(&map->work)){+.+.},
at: process_one_work+0x17a/0x580
#2: 00000000e3e7a6aa (pcpu_lock){..-.},
at: free_percpu+0x36/0x260
stack backtrace:
CPU: 23 PID: 18872 Comm: kworker/23:255 Not tainted 5.1.0-dbg-DEV #1
Hardware name: ...
Workqueue: events bpf_map_free_deferred
Call Trace:
dump_stack+0x67/0x95
print_circular_bug.isra.38+0x1c6/0x220
check_prev_add.constprop.50+0x9f6/0xd20
__lock_acquire+0x101f/0x12a0
lock_acquire+0x9e/0x180
_raw_spin_lock+0x2f/0x40
__queue_work+0xb2/0x520
queue_work_on+0x38/0x80
free_percpu+0x221/0x260
pcpu_freelist_destroy+0x11/0x20
stack_map_free+0x2a/0x40
bpf_map_free_deferred+0x3c/0x50
process_one_work+0x1f7/0x580
worker_thread+0x54/0x410
kthread+0x10f/0x150
ret_from_fork+0x24/0x30
Signed-off-by: John Sperbeck <jsperbeck@google.com>
Signed-off-by: Dennis Zhou <dennis@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
mm/percpu.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/mm/percpu.c b/mm/percpu.c
index 41e58f3d8fbf2..c66149ce1fe6a 100644
--- a/mm/percpu.c
+++ b/mm/percpu.c
@@ -1721,6 +1721,7 @@ void free_percpu(void __percpu *ptr)
struct pcpu_chunk *chunk;
unsigned long flags;
int off;
+ bool need_balance = false;
if (!ptr)
return;
@@ -1742,7 +1743,7 @@ void free_percpu(void __percpu *ptr)
list_for_each_entry(pos, &pcpu_slot[pcpu_nr_slots - 1], list)
if (pos != chunk) {
- pcpu_schedule_balance_work();
+ need_balance = true;
break;
}
}
@@ -1750,6 +1751,9 @@ void free_percpu(void __percpu *ptr)
trace_percpu_free_percpu(chunk->base_addr, off, ptr);
spin_unlock_irqrestore(&pcpu_lock, flags);
+
+ if (need_balance)
+ pcpu_schedule_balance_work();
}
EXPORT_SYMBOL_GPL(free_percpu);
--
2.20.1
next prev parent reply other threads:[~2019-06-01 13:35 UTC|newest]
Thread overview: 56+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-06-01 13:19 [PATCH AUTOSEL 4.19 001/141] rapidio: fix a NULL pointer dereference when create_workqueue() fails Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 002/141] fs/fat/file.c: issue flush after the writeback of FAT Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 003/141] sysctl: return -EINVAL if val violates minmax Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 004/141] ipc: prevent lockup on alloc_msg and free_msg Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 005/141] drm/pl111: Initialize clock spinlock early Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 006/141] ARM: prevent tracing IPI_CPU_BACKTRACE Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 007/141] mm/hmm: select mmu notifier when selecting HMM Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 008/141] hugetlbfs: on restore reserve error path retain subpool reservation Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 009/141] mem-hotplug: fix node spanned pages when we have a node with only ZONE_MOVABLE Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 010/141] mm/cma.c: fix crash on CMA allocation if bitmap allocation fails Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 011/141] initramfs: free initrd memory if opening /initrd.image fails Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 012/141] mm/cma.c: fix the bitmap status to show failed allocation reason Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 013/141] mm: page_mkclean vs MADV_DONTNEED race Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 014/141] mm/cma_debug.c: fix the break condition in cma_maxchunk_get() Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 015/141] mm/slab.c: fix an infinite loop in leaks_show() Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 016/141] kernel/sys.c: prctl: fix false positive in validate_prctl_map() Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 017/141] thermal: rcar_gen3_thermal: disable interrupt in .remove Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 018/141] drivers: thermal: tsens: Don't print error message on -EPROBE_DEFER Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 019/141] mfd: tps65912-spi: Add missing of table registration Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 020/141] mfd: intel-lpss: Set the device in reset state when init Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 021/141] drm/nouveau/disp/dp: respect sink limits when selecting failsafe link configuration Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 022/141] mfd: twl6040: Fix device init errors for ACCCTL register Sasha Levin
2019-06-01 13:19 ` [PATCH AUTOSEL 4.19 023/141] perf/x86/intel: Allow PEBS multi-entry in watermark mode Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 024/141] drm/nouveau/kms/gf119-gp10x: push HeadSetControlOutputResource() mthd when encoders change Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 025/141] drm/bridge: adv7511: Fix low refresh rate selection Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 026/141] objtool: Don't use ignore flag for fake jumps Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 027/141] drm/nouveau/kms/gv100-: fix spurious window immediate interlocks Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 028/141] bpf: fix undefined behavior in narrow load handling Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 029/141] EDAC/mpc85xx: Prevent building as a module Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 030/141] NFS4: Fix v4.0 client state corruption when mount Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 031/141] pwm: meson: Use the spin-lock only to protect register modifications Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 032/141] mailbox: stm32-ipcc: check invalid irq Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 033/141] ntp: Allow TAI-UTC offset to be set to zero Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 034/141] f2fs: fix to avoid panic in do_recover_data() Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 035/141] f2fs: fix to avoid panic in f2fs_inplace_write_data() Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 036/141] f2fs: fix to avoid panic in f2fs_remove_inode_page() Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 037/141] f2fs: fix to do sanity check on free nid Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 038/141] f2fs: fix to clear dirty inode in error path of f2fs_iget() Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 039/141] f2fs: fix to avoid panic in dec_valid_block_count() Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 040/141] f2fs: fix to use inline space only if inline_xattr is enable Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 041/141] f2fs: fix to do sanity check on valid block count of segment Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 042/141] f2fs: fix to do checksum even if inode page is uptodate Sasha Levin
2019-06-01 13:20 ` Sasha Levin [this message]
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 044/141] tracing: Fix partial reading of trace event's id file Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 045/141] configfs: fix possible use-after-free in configfs_register_group Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 046/141] uml: fix a boot splat wrt use of cpu_all_mask Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 047/141] PCI: dwc: Free MSI in dw_pcie_host_init() error path Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 048/141] PCI: dwc: Free MSI IRQ page in dw_pcie_free_msi() Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 049/141] ovl: do not generate duplicate fsnotify events for "fake" path Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 050/141] mmc: mmci: Prevent polling for busy detection in IRQ context Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 051/141] netfilter: nf_flow_table: fix missing error check for rhashtable_insert_fast Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 052/141] netfilter: nf_conntrack_h323: restore boundary check correctness Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 053/141] mips: Make sure dt memory regions are valid Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 054/141] netfilter: nf_tables: fix base chain stat rcu_dereference usage Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 055/141] watchdog: Use depends instead of select for pretimeout governors Sasha Levin
2019-06-01 13:20 ` [PATCH AUTOSEL 4.19 056/141] watchdog: imx2_wdt: Fix set_timeout for big timeout values Sasha Levin
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=20190601132158.25821-43-sashal@kernel.org \
--to=sashal@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=dennis@kernel.org \
--cc=jsperbeck@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=netdev@vger.kernel.org \
--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;
as well as URLs for NNTP newsgroup(s).