[PATCH v2] ahci: st: Switch from CONFIG_PM_SLEEP guards to pm_sleep_ptr()

[Raphaël Gallais-Pou]

Le 10/01/2025 à 13:02, Damien Le Moal a écrit :
> On 1/10/25 20:23, Niklas Cassel wrote:
>>>> -#ifdef CONFIG_PM_SLEEP
>>>>   static int st_ahci_suspend(struct device *dev)
>>>>   {
>>>>   	struct ata_host *host = dev_get_drvdata(dev);
>>>> @@ -221,9 +220,8 @@ static int st_ahci_resume(struct device *dev)
>>>>   
>>>>   	return ahci_platform_resume_host(dev);
>>>>   }
>>>> -#endif
>>>
>>> I do not think you can remove the ifdef here. Otherwise, there is going to be a
>>> compilation warning when CONFIG_PM_SLEEP is not enabled. No ?
>>
>> Look at the pm_sleep_ptr macro:
>> include/linux/pm.h:#define pm_sleep_ptr(_ptr) PTR_IF(IS_ENABLED(CONFIG_PM_SLEEP), (_ptr))
>>
>> I would expect the function should be optimized out by the compiler
>> using dead code elimination.
> 
> Indeed. Just tried and no warning. I was expecting a "defined but not used"
> warning, but none showed up. So all good.
> 
>> Raphael, perhaps you could show the before and after output
>> using ./scripts/bloat-o-meter ?
>> (When the config is not enabled: before and after your patch.)
Hi,

I have not used the bloat-o-meter until now, thanks ! :)
Here are my results:


   * with the configuration

$ ./scripts/bloat-o-meter ahci_st_no_patch_pm.o ahci_st_patch_pm.o
add/remove: 1/1 grow/shrink: 0/0 up/down: 4/-4 (0)
Function                                     old     new   delta
__initcall__kmod_ahci_st__384_241_st_ahci_driver_init6       -       4 
    +4
__initcall__kmod_ahci_st__384_243_st_ahci_driver_init6       4       - 
    -4
Total: Before=2200, After=2200, chg +0.00%


   * without the configuration

$ ./scripts/bloat-o-meter ahci_st_no_patch_no_pm.o ahci_st_patch_no_pm.o
add/remove: 1/2 grow/shrink: 0/0 up/down: 4/-96 (-92)
Function                                     old     new   delta
__initcall__kmod_ahci_st__383_241_st_ahci_driver_init6       -       4 
    +4
__initcall__kmod_ahci_st__383_243_st_ahci_driver_init6       4       - 
    -4
st_ahci_pm_ops                                92       -     -92
Total: Before=1904, After=1812, chg -4.83%


Looks like the patch shrinks a bit more the driver. I also tested, so we 
should be fine I think.
> 
> No need to do that I guess. But there are 17 other ata driver that set .pm
> operations. What about these ? Don't they need the same treatment as ahci_st ?
> 15 of these also use SIMPLE_DEV_PM_OPS() which can be replaced with
> DEFINE_SIMPLE_DEV_PM_OPS() also, no ?
> 
> Do you want us to do that cleanup ? (fine with me).

Regarding the other ata drivers, if you have the patience I can do this 
in a few weeks.  There is other things on the stove I would like to do.

Regards,
Raphaël
>