mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Christophe Leroy <christophe.leroy@csgroup.eu>
To: Zhen Lei <thunder.leizhen@huawei.com>,
	Petr Mladek <pmladek@suse.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Luis Chamberlain <mcgrof@kernel.org>,
	"linux-modules@vger.kernel.org" <linux-modules@vger.kernel.org>
Cc: Anders Roxell <anders.roxell@linaro.org>
Subject: Re: [PATCH] kallsyms: Fix sleeping function called from invalid context when CONFIG_KALLSYMS_SELFTEST=y
Date: Tue, 20 Dec 2022 08:15:40 +0000	[thread overview]
Message-ID: <df75bb4e-6cf8-7f41-b053-9619c13d1c72@csgroup.eu> (raw)
In-Reply-To: <20221220063923.1937-1-thunder.leizhen@huawei.com>



Le 20/12/2022 à 07:39, Zhen Lei a écrit :
> [T58] BUG: sleeping function called from invalid context at kernel/kallsyms.c:305
> [T58] in_atomic(): 0, irqs_disabled(): 128, non_block: 0, pid: 58, name: kallsyms_test
> [T58] preempt_count: 0, expected: 0
> [T58] RCU nest depth: 0, expected: 0
> [T58] no locks held by kallsyms_test/58.
> [T58] irq event stamp: 18899904
> [T58] hardirqs last enabled at (18899903): finish_task_switch.isra.0 (core.c:?)
> [T58] hardirqs last disabled at (18899904): test_perf_kallsyms_on_each_symbol (kallsyms_selftest.c:?)
> [T58] softirqs last enabled at (18899886): __do_softirq (??:?)
> [T58] softirqs last disabled at (18899879): ____do_softirq (irq.c:?)
> [T58] CPU: 0 PID: 58 Comm: kallsyms_test Tainted: G T  6.1.0-next-20221215 #2
> [T58] Hardware name: linux,dummy-virt (DT)
> [T58] Call trace:
> [T58] dump_backtrace (??:?)
> [T58] show_stack (??:?)
> [T58] dump_stack_lvl (??:?)
> [T58] dump_stack (??:?)
> [T58] __might_resched (??:?)
> [T58] kallsyms_on_each_symbol (??:?)
> [T58] test_perf_kallsyms_on_each_symbol (kallsyms_selftest.c:?)
> [T58] test_entry (kallsyms_selftest.c:?)
> [T58] kthread (kthread.c:?)
> [T58] ret_from_fork (??:?)
> [T58] kallsyms_selftest: kallsyms_on_each_symbol() traverse all: 5744310840 ns
> [T58] kallsyms_selftest: kallsyms_on_each_match_symbol() traverse all: 1164580 ns
> [T58] kallsyms_selftest: finish
> 
> Functions kallsyms_on_each_symbol() and kallsyms_on_each_match_symbol()
> call the user-registered hook function for each symbol that meets the
> requirements. Because it is uncertain how long that hook function will
> execute, they call cond_resched() to avoid consuming CPU resources for a
> long time. However, irqs need to be disabled during the performance test
> to ensure the accuracy of test data. Because the performance test hook is
> very clear, very simple function, let's do not call cond_resched() when
> CONFIG_KALLSYMS_SELFTEST=y.

I don't think it is appropriate to change the behaviour of a core 
function based on whether a compile time option related to tests is 
selected or not, because you will change the behaviour for all users, 
not only for the tests.

If the problem is that IRQs are disabled, maybe the solution is

	if (!irqs_disabled())
		cond_resched();

Or try to disable the call to cond_resched() in a way or another during 
the run of selftests.

> 
> Fixes: 30f3bb09778d ("kallsyms: Add self-test facility")
> Reported-by: Anders Roxell <anders.roxell@linaro.org>
> Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com>
> ---
>   kernel/kallsyms.c | 6 ++++--
>   1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c
> index 83f499182c9aa31..a49e344a686517b 100644
> --- a/kernel/kallsyms.c
> +++ b/kernel/kallsyms.c
> @@ -302,7 +302,8 @@ int kallsyms_on_each_symbol(int (*fn)(void *, const char *, struct module *,
>   		ret = fn(data, namebuf, NULL, kallsyms_sym_address(i));
>   		if (ret != 0)
>   			return ret;
> -		cond_resched();
> +		if (!IS_ENABLED(CONFIG_KALLSYMS_SELFTEST))
> +			cond_resched();
>   	}
>   	return 0;
>   }
> @@ -319,7 +320,8 @@ int kallsyms_on_each_match_symbol(int (*fn)(void *, unsigned long),
>   
>   	for (i = start; !ret && i <= end; i++) {
>   		ret = fn(data, kallsyms_sym_address(get_symbol_seq(i)));
> -		cond_resched();
> +		if (!IS_ENABLED(CONFIG_KALLSYMS_SELFTEST))
> +			cond_resched();
>   	}
>   
>   	return ret;

  reply	other threads:[~2022-12-20  8:15 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-12-20  6:39 Zhen Lei
2022-12-20  8:15 ` Christophe Leroy [this message]
2022-12-20 11:01   ` Petr Mladek
2022-12-20 14:29     ` Leizhen (ThunderTown)
2022-12-20 15:43       ` Petr Mladek
2022-12-21 12:37         ` Leizhen (ThunderTown)
2022-12-27  7:16           ` Leizhen (ThunderTown)

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=df75bb4e-6cf8-7f41-b053-9619c13d1c72@csgroup.eu \
    --to=christophe.leroy@csgroup.eu \
    --cc=anders.roxell@linaro.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-modules@vger.kernel.org \
    --cc=mcgrof@kernel.org \
    --cc=pmladek@suse.com \
    --cc=thunder.leizhen@huawei.com \
    /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

all inboxes | Powered by JetHome®