Re: [PATCH] regmap: debugfs: Fix compilation warning

Re: [PATCH] regmap: debugfs: Fix compilation warning

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 379 bytes --]

On Mon, Jan 21, 2013 at 03:36:55PM +0100, Vincent Stehlé wrote:
> This fixes the following compilation warning:

> -	unsigned int i, ret;
> +	unsigned int i, ret = 0;

This sort of fix is not a good idea, you're just shutting the warning up
without any sort of analysis explaining why it's generated in error.  If
it's generating a spurious error that's a compiler bug.

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [PATCH] regmap: debugfs: Fix compilation warning

[Vincent Stehlé]

On 01/22/2013 07:41 AM, Mark Brown wrote:
(..)
> This sort of fix is not a good idea, you're just shutting the
> warning up without any sort of analysis explaining why it's
> generated in error.  If it's generating a spurious error that's a
> compiler bug.

In the regmap_debugfs_get_dump_start() function control flow, I think
the compiler cannot know for sure that the list_for_each_entry()
iterates at least once.:

	regmap_debugfs_get_dump_start()
	{
		unsigned int ret;
		<snip some code, which does not touch ret>

		list_for_each_entry() {
			<snip some code>
			ret = <value>;
		}

		return ret;
	}

Do you think there is a way to "mark" the list_for_each_entry() as
iterating at least once? an __attribute__ maybe?

Best regards,

V.

Re: [PATCH] regmap: debugfs: Fix compilation warning

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 227 bytes --]

On Tue, Jan 22, 2013 at 11:07:04AM +0100, Vincent Stehlé wrote:

> Do you think there is a way to "mark" the list_for_each_entry() as
> iterating at least once? an __attribute__ maybe?

No - but are you sure that's true?

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [PATCH] regmap: debugfs: Fix compilation warning

[Vincent Stehlé]

On 01/23/2013 04:58 PM, Mark Brown wrote:
> On Tue, Jan 22, 2013 at 11:07:04AM +0100, Vincent Stehlé wrote:
> 
>> Do you think there is a way to "mark" the list_for_each_entry()
>> as iterating at least once? an __attribute__ maybe?
> 
> No - but are you sure that's true?

If you mean "am I sure the loop iterates at least once", then yes, as
we have an explicit check just before the concerned list_for_each_entry():

	/*
	 * This should never happen; we return above if we fail to
	 * allocate and we should never be in this code if there are
	 * no registers at all.
	 */
	if (list_empty(&map->debugfs_off_cache)) {
		WARN_ON(list_empty(&map->debugfs_off_cache));
		return base;
	}

Best regards,

V.

Re: [PATCH] regmap: debugfs: Fix compilation warning

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 471 bytes --]

On Thu, Jan 24, 2013 at 11:12:24AM +0100, Vincent Stehlé wrote:

> If you mean "am I sure the loop iterates at least once", then yes, as
> we have an explicit check just before the concerned list_for_each_entry():

So, I hadn't realised this was triggered by a new compiler version -
Arnd just tried the same bodge.  I've committed a workaround for the
compiler which avoids the problems that setting 0 will have and allows
the flow analyis to run where it can.

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]