Re: [PATCH v3 1/4] arm_mpam: Fix false positive assert failure during mpam_disable()

[Ben Horgan <ben.horgan@arm.com>]

Hi James,

On 5/8/26 17:23, James Morse wrote:
> mpam_assert_partid_sizes_fixed() is used to document that the caller
> doesn't expect the discovered PARTID size to change while it is walking
> a list sized by PARTID. Typically the MSC state is not written to until
> all the MSC have been discovered and this value is set.
> 
> However, if discovering the MSC fails and schedules mpam_disable(),
> then the MSC state is written to reset it. In this case the
> discovered PARTID size may be become smaller - but only PARTID 0
> will be used once resctrl_exit() has been called.
> 
> Skip the WARN_ON_ONCE() if mpam_disable_reason has been set.
> 
> Fixes: 3bd04fe7d807bb ("arm_mpam: Extend reset logic to allow devices to be reset at any time")
> Signed-off-by: James Morse <james.morse@arm.com>
> ---
>  drivers/resctrl/mpam_devices.c | 10 ++++++++--
>  1 file changed, 8 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/resctrl/mpam_devices.c b/drivers/resctrl/mpam_devices.c
> index 41b14344b16f..bef5e9e9e844 100644
> --- a/drivers/resctrl/mpam_devices.c
> +++ b/drivers/resctrl/mpam_devices.c
> @@ -164,11 +164,17 @@ static void mpam_free_garbage(void)
>  /*
>   * Once mpam is enabled, new requestors cannot further reduce the available
>   * partid. Assert that the size is fixed, and new requestors will be turned
> - * away.
> + * away. This is needed when walking over structures sized by PARTID.
> + *
> + * During mpam_disable() these structures are not fixed, but the MSC state
> + * is still reset using whatever sizes have been discovered so far. As only
> + * PARTID 0 will be used after mpam_disable(), any race would be benign.
> + * Skip the check if a mpam_disable_reason has been set.
>   */
>  static void mpam_assert_partid_sizes_fixed(void)
>  {
> -	WARN_ON_ONCE(!partid_max_published);
> +	if (!mpam_disable_reason)
> +		WARN_ON_ONCE(!partid_max_published);

A little bit nasty to use mpam_disable_reason this way but a simple better way eludes me.

Reviewed-by: Ben Horgan <ben.horgan@arm.com>

Thanks,

Ben


>  }
>  
>  static u32 __mpam_read_reg(struct mpam_msc *msc, u16 reg)