From: "Benoît Thébaudeau" <benoit.thebaudeau@advansee.com>
To: Rusty Russell <rusty@rustcorp.com.au>
Cc: Pawel Moll <pawel.moll@arm.com>,
Artem Bityutskiy <dedekind1@gmail.com>,
Linus Torvalds <torvalds@linux-foundation.org>,
linux-kernel@vger.kernel.org, linux-mtd@lists.infradead.org,
benoit thebaudeau <benoit.thebaudeau@advansee.com>
Subject: Re: init: multi param parsing regression since 3.4
Date: Mon, 4 Jun 2012 14:38:51 +0200 (CEST) [thread overview]
Message-ID: <28190110.1966162.1338813531173.JavaMail.root@advansee.com> (raw)
In-Reply-To: <87ipf71ytx.fsf@rustcorp.com.au>
On Mon, Jun 4, 2012 at 6:32:42AM, Rusty Russell wrote:
> Erk, I tested it and you're right. Level 0 is 'early' initcalls, but
> traditional module parameters are called even earlier. Simplest fix
> is to move the default parameters to -1.
>
> This works for me:
>
> Subject: module_param: stop double-calling parameters.
> From: Rusty Russell <rusty@rustcorp.com.au>
>
> Commit 026cee0086fe1df4cf74691cf273062cc769617d "params:
> <level>_initcall-like kernel parameters" set old-style module
> parameters to level 0. And we call those level 0 calls where we used
> to, early in start_kernel().
>
> We also loop through the initcall levels and call the levelled
> module_params before the corresponding initcall. Unfortunately level
> 0 is early_init(), so we call the standard module_param calls twice.
>
> (Turns out most things don't care, but at least ubi.mtd does).
>
> Change the level to -1 for standard module_param calls.
>
> Reported-by: Benoît Thébaudeau <benoit.thebaudeau@advansee.com>
> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
> Cc: stable@kernel.org
>
> diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
> --- a/drivers/mtd/ubi/build.c
> +++ b/drivers/mtd/ubi/build.c
> @@ -1293,6 +1293,8 @@ static int __init ubi_mtd_param_parse(co
> char *pbuf = &buf[0];
> char *tokens[2] = {NULL, NULL};
>
> + printk("ubi_mtd_param_parse: val = %s\n", val);
> + WARN_ON(1);
> if (!val)
> return -EINVAL;
>
> diff --git a/include/linux/moduleparam.h
> b/include/linux/moduleparam.h
> --- a/include/linux/moduleparam.h
> +++ b/include/linux/moduleparam.h
> @@ -128,7 +128,7 @@ struct kparam_array
> * The ops can have NULL set or get functions.
> */
> #define module_param_cb(name, ops, arg, perm) \
> - __module_param_call(MODULE_PARAM_PREFIX, name, ops, arg, perm, 0)
> + __module_param_call(MODULE_PARAM_PREFIX, name, ops, arg, perm, -1)
>
> /**
> * <level>_param_cb - general callback for a module/cmdline
> parameter
> @@ -192,7 +192,7 @@ struct kparam_array
> { (void *)set, (void *)get }; \
> __module_param_call(MODULE_PARAM_PREFIX, \
> name, &__param_ops_##name, arg, \
> - (perm) + sizeof(__check_old_set_param(set))*0, 0)
> + (perm) + sizeof(__check_old_set_param(set))*0, -1)
>
> /* We don't get oldget: it's often a new-style param_get_uint, etc.
> */
> static inline int
> @@ -272,7 +272,7 @@ static inline void __kernel_param_unlock
> */
> #define core_param(name, var, type, perm) \
> param_check_##type(name, &(var)); \
> - __module_param_call("", name, ¶m_ops_##type, &var, perm, 0)
> + __module_param_call("", name, ¶m_ops_##type, &var, perm, -1)
> #endif /* !MODULE */
>
> /**
> @@ -290,7 +290,7 @@ static inline void __kernel_param_unlock
> = { len, string }; \
> __module_param_call(MODULE_PARAM_PREFIX, name, \
> ¶m_ops_string, \
> - .str = &__param_string_##name, perm, 0); \
> + .str = &__param_string_##name, perm, -1); \
> __MODULE_PARM_TYPE(name, "string")
>
> /**
> @@ -432,7 +432,7 @@ extern int param_set_bint(const char *va
> __module_param_call(MODULE_PARAM_PREFIX, name, \
> ¶m_array_ops, \
> .arr = &__param_arr_##name, \
> - perm, 0); \
> + perm, -1); \
> __MODULE_PARM_TYPE(name, "array of " #type)
>
> extern struct kernel_param_ops param_array_ops;
> diff --git a/init/main.c b/init/main.c
> --- a/init/main.c
> +++ b/init/main.c
> @@ -508,7 +508,7 @@ asmlinkage void __init start_kernel(void
> parse_early_param();
> parse_args("Booting kernel", static_command_line, __start___param,
> __stop___param - __start___param,
> - 0, 0, &unknown_bootoption);
> + -1, -1, &unknown_bootoption);
>
> jump_label_init();
>
>
It works for me too. But I think you can remove the 1st hunk, which seems to be
a leftover from your tests.
Regards,
Benoît
next prev parent reply other threads:[~2012-06-04 12:34 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <1002174455.1843950.1338570468781.JavaMail.root@advansee.com>
2012-06-01 17:26 ` init: multi param parsing regression since 3.4 Benoît Thébaudeau
2012-06-02 15:50 ` Benoît Thébaudeau
2012-06-04 4:32 ` Rusty Russell
2012-06-04 12:38 ` Benoît Thébaudeau [this message]
2012-06-04 20:39 ` Linus Torvalds
2012-06-05 0:17 ` Rusty Russell
2012-06-05 1:25 ` Linus Torvalds
2012-06-06 8:28 ` Artem Bityutskiy
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=28190110.1966162.1338813531173.JavaMail.root@advansee.com \
--to=benoit.thebaudeau@advansee.com \
--cc=dedekind1@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mtd@lists.infradead.org \
--cc=pawel.moll@arm.com \
--cc=rusty@rustcorp.com.au \
--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