Re: [PATCH v3] perf/x86/amd/uncore: Use kcalloc*() instead of kzalloc*()

[Christophe JAILLET <christophe.jaillet@wanadoo.fr>]

Le 27/04/2024 à 18:45, Erick Archer a écrit :
> This is an effort to get rid of all multiplications from allocation
> functions in order to prevent integer overflows [1].
> 
> Here the multiplication is obviously safe. However, using kcalloc*()
> is more appropriate [2] and improves readability. This patch has no
> effect on runtime behavior.
> 
> Link: https://github.com/KSPP/linux/issues/162 [1]
> Link: https://www.kernel.org/doc/html/next/process/deprecated.html#open-coded-arithmetic-in-allocator-arguments [2]
> Reviewed-by: Gustavo A. R. Silva <gustavoars@kernel.org>
> Signed-off-by: Erick Archer <erick.archer@outlook.com>
> ---
> Changes in v3:
> - Update the commit message to better explain the changes.
> - Rebase against linux-next.
> 
> Changes in v2:
> - Add the "Reviewed-by:" tag.
> - Rebase against linux-next.
> 
> Previous versions:
> v1 -> https://lore.kernel.org/linux-hardening/20240116125813.3754-1-erick.archer@gmx.com
> v2 -> https://lore.kernel.org/linux-hardening/AS8PR02MB7237A07D73D6D15EBF72FD8D8B392@AS8PR02MB7237.eurprd02.prod.outlook.com
> 
> Hi,
> 
> This is a new try. In the v2 version Ingo explained that this change
> is nonsense since kzalloc() is a perfectly usable interface and there
> is no real overflow here.
> 
> Anyway, if we have the 2-factor form of the allocator, I think it is
> a good practice to use it.
> 
> In this version I have updated the commit message to explain that
> the code is obviusly safe in contrast with the last version where the
> impression was given that there was a real overlow bug.
> 
> I hope this patch can be applied this time.
> 
> Regards,
> Erick
> ---
>   arch/x86/events/amd/uncore.c | 6 +++---
>   1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/arch/x86/events/amd/uncore.c b/arch/x86/events/amd/uncore.c
> index 4ccb8fa483e6..61c0a2114183 100644
> --- a/arch/x86/events/amd/uncore.c
> +++ b/arch/x86/events/amd/uncore.c
> @@ -479,8 +479,8 @@ static int amd_uncore_ctx_init(struct amd_uncore *uncore, unsigned int cpu)
>   				goto fail;
>   
>   			curr->cpu = cpu;
> -			curr->events = kzalloc_node(sizeof(*curr->events) *
> -						    pmu->num_counters,
> +			curr->events = kcalloc_node(pmu->num_counters,
> +						    sizeof(*curr->events),
>   						    GFP_KERNEL, node);

Hi,

not related to your patch, but amd_uncore_ctx looks like a good 
candidate for a flexible array.

This would simplify allocation/freeing of the memory.

struct amd_uncore_ctx {
	int refcnt;
	int cpu;
	struct hlist_node node;
	struct perf_event *events[];
};

Just my 2c,

CJ

>   			if (!curr->events) {
>   				kfree(curr);
> @@ -928,7 +928,7 @@ int amd_uncore_umc_ctx_init(struct amd_uncore *uncore, unsigned int cpu)
>   		uncore->num_pmus += group_num_pmus[gid];
>   	}
>   
> -	uncore->pmus = kzalloc(sizeof(*uncore->pmus) * uncore->num_pmus,
> +	uncore->pmus = kcalloc(uncore->num_pmus, sizeof(*uncore->pmus),
>   			       GFP_KERNEL);
>   	if (!uncore->pmus) {
>   		uncore->num_pmus = 0;