2023-08-11 21:39:10

by Justin Stitt

[permalink] [raw]
Subject: [PATCH RFC 1/3] powerpc/ps3: refactor strncpy usage attempt 1

This approach simply replicates the implementation of `make_field` which
means we drop `strncpy` for `memcpy`.

Signed-off-by: Justin Stitt <[email protected]>
---
arch/powerpc/platforms/ps3/repository.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/powerpc/platforms/ps3/repository.c b/arch/powerpc/platforms/ps3/repository.c
index 205763061a2d..1abe33fbe529 100644
--- a/arch/powerpc/platforms/ps3/repository.c
+++ b/arch/powerpc/platforms/ps3/repository.c
@@ -73,9 +73,9 @@ static void _dump_node(unsigned int lpar_id, u64 n1, u64 n2, u64 n3, u64 n4,

static u64 make_first_field(const char *text, u64 index)
{
- u64 n;
+ u64 n = 0;

- strncpy((char *)&n, text, 8);
+ memcpy((char *)&n, text, strnlen(text, sizeof(n)));
return PS3_VENDOR_ID_NONE + (n >> 32) + index;
}


--
2.41.0.640.ga95def55d0-goog



2023-08-14 12:58:51

by Michael Ellerman

[permalink] [raw]
Subject: Re: [PATCH RFC 1/3] powerpc/ps3: refactor strncpy usage attempt 1

Justin Stitt <[email protected]> writes:
> This approach simply replicates the implementation of `make_field` which
> means we drop `strncpy` for `memcpy`.
>
> Signed-off-by: Justin Stitt <[email protected]>
> ---
> arch/powerpc/platforms/ps3/repository.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/arch/powerpc/platforms/ps3/repository.c b/arch/powerpc/platforms/ps3/repository.c
> index 205763061a2d..1abe33fbe529 100644
> --- a/arch/powerpc/platforms/ps3/repository.c
> +++ b/arch/powerpc/platforms/ps3/repository.c
> @@ -73,9 +73,9 @@ static void _dump_node(unsigned int lpar_id, u64 n1, u64 n2, u64 n3, u64 n4,
>
> static u64 make_first_field(const char *text, u64 index)
> {
> - u64 n;
> + u64 n = 0;
>
> - strncpy((char *)&n, text, 8);
> + memcpy((char *)&n, text, strnlen(text, sizeof(n)));
> return PS3_VENDOR_ID_NONE + (n >> 32) + index;
> }

I guess it's a slight improvement, because people generally know
memcpy's behaviour better than strncpy.

The change log should be a bit more verbose and mention that the result
is the same, including the NULL padding done my strncpy().

cheers