From: "Paul E. McKenney" <[email protected]>
To: Dan Carpenter <[email protected]>
Cc: [email protected], [email protected],
[email protected],
Ammar Faizi <[email protected]>,
GNU/Weeb Mailing List <[email protected]>
Subject: Re: [ammarfaizi2-block:paulmck/linux-rcu/dev.2023.01.19a 55/61] kernel/rcu/rcutorture.c:3548 rcu_torture_init_srcu_lockdep() warn: for statement not indented
Date: Tue, 24 Jan 2023 06:51:16 -0800 [thread overview]
Message-ID: <20230124145116.GH2948950@paulmck-ThinkPad-P17-Gen-1> (raw)
In-Reply-To: <[email protected]>
On Tue, Jan 24, 2023 at 12:32:14PM +0300, Dan Carpenter wrote:
> tree: https://github.com/ammarfaizi2/linux-block paulmck/linux-rcu/dev.2023.01.19a
> head: e51eadce38349e7edb506a61e325cc562d49409f
> commit: a5ae869a4ab737494a3c5c167d782b8eab3ba724 [55/61] squash! rcutorture: Add SRCU deadlock scenarios
> config: x86_64-randconfig-m001-20230123 (https://download.01.org/0day-ci/archive/20230124/[email protected]/config)
> compiler: gcc-11 (Debian 11.3.0-8) 11.3.0
>
> If you fix the issue, kindly add following tag where applicable
> | Reported-by: kernel test robot <[email protected]>
> | Reported-by: Dan Carpenter <[email protected]>
>
> smatch warnings:
> kernel/rcu/rcutorture.c:3548 rcu_torture_init_srcu_lockdep() warn: for statement not indented
>
> vim +3548 kernel/rcu/rcutorture.c
>
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3514 static void rcu_torture_init_srcu_lockdep(void)
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3515 {
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3516 int cyclelen;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3517 int cyclelenmax;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3518 int deadlock;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3519 bool err = false;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3520 int i;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3521 int j;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3522 int idx;
> a5ae869a4ab737 Paul E. McKenney 2023-01-23 3523 struct mutex *muts[] = { &mut0, &mut1, &mut2, &mut3, &mut4,
> a5ae869a4ab737 Paul E. McKenney 2023-01-23 3524 &mut5, &mut6, &mut7, &mut8, &mut9 };
> a5ae869a4ab737 Paul E. McKenney 2023-01-23 3525 struct srcu_struct *srcus[] = { &srcu0, &srcu1, &srcu2, &srcu3, &srcu4,
> a5ae869a4ab737 Paul E. McKenney 2023-01-23 3526 &srcu5, &srcu6, &srcu7, &srcu8, &srcu9 };
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3527 int testtype;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3528
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3529 if (!test_srcu_lockdep)
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3530 return;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3531
> a5ae869a4ab737 Paul E. McKenney 2023-01-23 3532 deadlock = test_srcu_lockdep / 1000;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3533 testtype = (test_srcu_lockdep / 10) % 100;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3534 cyclelen = test_srcu_lockdep % 10;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3535 cyclelenmax = min(ARRAY_SIZE(muts), ARRAY_SIZE(srcus));
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3536 WARN_ON_ONCE(ARRAY_SIZE(muts) != ARRAY_SIZE(srcus));
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3537 if (WARN_ONCE(deadlock != !!deadlock,
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3538 "%s: test_srcu_lockdep=%d and deadlock digit %d must be zero or one.\n",
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3539 __func__, test_srcu_lockdep, deadlock))
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3540 err = true;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3541 if (WARN_ONCE(cyclelen <= 0,
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3542 "%s: test_srcu_lockdep=%d and cycle-length digit %d must be greater than zero.\n",
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3543 __func__, test_srcu_lockdep, cyclelen))
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3544 err = true;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3545 if (err)
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3546 goto err_out;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3547
> a5ae869a4ab737 Paul E. McKenney 2023-01-23 @3548 for (i = 0; i < cyclelen; i++)
>
> This line looks like it should just be deleted. The real loops are
> later so i == cyclelen by the time it loops back here. Should probably
> create a static checker warning for nested loops where the inside loop
> resets the outside iterator to zero.
You are quite right, I intended to have deleted that statement.
And the fact that it works is quite coincidental. Thank you!
Thanx, Paul
> a5ae869a4ab737 Paul E. McKenney 2023-01-23 3549
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3550 if (testtype == 0) {
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3551 pr_info("%s: test_srcu_lockdep = %05d: SRCU %d-way %sdeadlock.\n",
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3552 __func__, test_srcu_lockdep, cyclelen, deadlock ? "" : "non-");
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3553 if (deadlock && cyclelen == 1)
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3554 pr_info("%s: Expect hang.\n", __func__);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3555 for (i = 0; i < cyclelen; i++) {
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3556 j = i + 1;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3557 if (i >= cyclelen - 1)
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3558 j = deadlock ? 0 : -1;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3559
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3560 if (j >= 0)
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3561 pr_info("%s: srcu_read_lock(%d), synchronize_srcu(%d), srcu_read_unlock(%d)\n",
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3562 __func__, i, j, i);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3563 else
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3564 pr_info("%s: srcu_read_lock(%d), srcu_read_unlock(%d)\n",
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3565 __func__, i, i);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3566 idx = srcu_read_lock(srcus[i]);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3567 if (j >= 0)
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3568 synchronize_srcu(srcus[j]);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3569 srcu_read_unlock(srcus[i], idx);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3570 }
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3571 return;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3572 }
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3573
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3574 if (testtype == 1) {
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3575 pr_info("%s: test_srcu_lockdep = %05d: SRCU/mutex %d-way %sdeadlock.\n",
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3576 __func__, test_srcu_lockdep, cyclelen, deadlock ? "" : "non-");
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3577 for (i = 0; i < cyclelen; i++) {
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3578 j = i + 1;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3579 if (i >= cyclelen - 1)
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3580 j = deadlock ? 0 : -1;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3581
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3582 pr_info("%s: srcu_read_lock(%d), mutex_lock(%d), mutex_unlock(%d), srcu_read_unlock(%d)\n",
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3583 __func__, i, i, i, i);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3584 idx = srcu_read_lock(srcus[i]);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3585 mutex_lock(muts[i]);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3586 mutex_unlock(muts[i]);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3587 srcu_read_unlock(srcus[i], idx);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3588
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3589 if (j >= 0) {
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3590 pr_info("%s: mutex_lock(%d), synchronize_srcu(%d), mutex_unlock(%d)\n",
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3591 __func__, i, j, i);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3592 mutex_lock(muts[i]);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3593 synchronize_srcu(srcus[j]);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3594 mutex_unlock(muts[i]);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3595 }
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3596 }
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3597 return;
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3598 }
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3599
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3600 err_out:
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3601 pr_info("%s: test_srcu_lockdep = %05d does nothing.\n", __func__, test_srcu_lockdep);
> a5ae869a4ab737 Paul E. McKenney 2023-01-23 3602 pr_info("%s: test_srcu_lockdep = DTTC.\n", __func__);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3603 pr_info("%s: D: Deadlock if nonzero.\n", __func__);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3604 pr_info("%s: TT: Test number, 0=SRCU, 1=SRCU/mutex.\n", __func__);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3605 pr_info("%s: C: Cycle length.\n", __func__);
> 81fba7e62c82be Paul E. McKenney 2023-01-13 3606 }
>
> --
> 0-DAY CI Kernel Test Service
> https://github.com/intel/lkp-tests
>
prev parent reply other threads:[~2023-01-24 14:51 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-01-24 9:32 [ammarfaizi2-block:paulmck/linux-rcu/dev.2023.01.19a 55/61] kernel/rcu/rcutorture.c:3548 rcu_torture_init_srcu_lockdep() warn: for statement not indented Dan Carpenter
2023-01-24 10:05 ` Dan Carpenter
2023-01-25 8:10 ` Dan Carpenter
2023-01-25 8:32 ` Dan Carpenter
2023-01-25 8:45 ` Ammar Faizi
2023-01-25 14:56 ` Paul E. McKenney
2023-01-24 14:51 ` Paul E. McKenney [this message]
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=20230124145116.GH2948950@paulmck-ThinkPad-P17-Gen-1 \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[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