io_uring: fix soft lockup in io_submit_sqes()

Offering: HULK
hulk inclusion
category: bugfix
bugzilla: 186692,https://gitee.com/openeuler/kernel/issues/I5930C

-------------------

when set up sq ring size with IORING_MAX_ENTRIES, io_submit_sqes may
looping ~32768 times which may trigger soft lockups. add cond_resched
condition to avoid this bad situation.

set sq ring size 32768 to perform stress test
as follows:

watchdog: BUG: soft lockup - CPU#3 stuck for 26s! [poc:691]
Modules linked in:
CPU: 3 PID: 691 Comm: poc Not tainted 5.18.0+ #9
Hardware name: linux,dummy-virt (DT)
pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
pc : arch_local_irq_enable+0xc/0x28
lr : io_issue_sqe+0x870/0x28e8
sp : ffff80000e0f7800
x29: ffff80000e0f7800 x28: ffff0000cf850dd0 x27: ffff0000cf2c2000
x26: ffff8000096f4b20 x25: ffff0000cd769c00 x24: ffff0000c12b6650
x23: ffff800009dad958 x22: 00000006fc23ac00 x21: ffff0000cd769c08
x20: 1ffff00001c1ef1a x19: ffff0000cd767e00 x18: 0000000000000000
x17: ffff800008032b74 x16: ffff800008636448 x15: 0000fffff7166568
x14: ffff80000861edf0 x13: ffff600019e58449 x12: 1fffe00019e58448
x11: 1fffe00019e58448 x10: ffff600019e58448 x9 : dfff800000000000
x8 : ffff0000cf2c2244 x7 : 0000000000000001 x6 : ffff600019e58449
x5 : ffff600019e58449 x4 : ffff600019e58449 x3 : ffff8000086306c0
x2 : 0000000000000001 x1 : ffff0000cf2c2244 x0 : 00000000000000e0
Call trace:
 arch_local_irq_enable+0xc/0x28
 io_submit_sqes+0x530/0x29d8
 __arm64_sys_io_uring_enter+0x380/0xd18
 invoke_syscall+0x64/0x180
 el0_svc_common.constprop.2+0x178/0x208
 do_el0_svc+0x84/0xa0
 el0_svc+0x48/0x1a0
 el0t_64_sync_handler+0x90/0xb8
 el0t_64_sync+0x180/0x184
Kernel panic - not syncing: softlockup: hung tasks
CPU: 3 PID: 691 Comm: poc Tainted: G             L    5.18.0+ #9
Hardware name: linux,dummy-virt (DT)
Call trace:
 dump_backtrace+0x218/0x228
 show_stack+0x20/0x68
 dump_stack_lvl+0x68/0x84
 dump_stack+0x1c/0x38
 panic+0x1ec/0x3ec
 watchdog_timer_fn+0x28c/0x300
 __hrtimer_run_queues+0x1d8/0x498
 hrtimer_interrupt+0x238/0x558
 arch_timer_handler_virt+0x48/0x60
 handle_percpu_devid_irq+0xdc/0x270
 generic_handle_domain_irq+0x50/0x70
 gic_handle_irq+0x8c/0x4bc
 call_on_irq_stack+0x2c/0x38
 do_interrupt_handler+0xc4/0xc8
 el1_interrupt+0x48/0xb0
 el1h_64_irq_handler+0x18/0x28
 el1h_64_irq+0x74/0x78
 arch_local_irq_enable+0xc/0x28
 io_submit_sqes+0x530/0x29d8
 __arm64_sys_io_uring_enter+0x380/0xd18
 invoke_syscall+0x64/0x180
 el0_svc_common.constprop.2+0x178/0x208
 do_el0_svc+0x84/0xa0
 el0_svc+0x48/0x1a0
 el0t_64_sync_handler+0x90/0xb8
 el0t_64_sync+0x180/0x184
SMP: stopping secondary CPUs
Kernel Offset: disabled
CPU features: 0x110,00008f09,00001006
Memory Limit: none
---[ end Kernel panic - not syncing: softlockup: hung tasks ]---

Link: https://lore.kernel.org/all/d4bc3afb-02d5-1793-cffa-e15b2bdb0028@huawei.com/Signed-off-by: NGuo Xuenan <guoxuenan@huawei.com>

Conflict:
	io_uring/io_uring.c
Signed-off-by: NLi Lingfeng <lilingfeng3@huawei.com>
Reviewed-by: NZhang Yi <yi.zhang@huawei.com>
Reviewed-by: NWang Weiyang <wangweiyang2@huawei.com>
Signed-off-by: NJialin Zhang <zhangjialin11@huawei.com>

io_uring/io_uring.c

@@ -7190,6 +7190,10 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr)
 		submitted++;
 		if (io_submit_sqe(ctx, req, sqe))
 			break;
+
+		/* to avoid doing too much in one submit round */
+		if (submitted > IORING_MAX_ENTRIES / 2)
+			cond_resched();
 	}
 
 	if (unlikely(submitted != nr)) {