RE: [PATCH] Kbuild: Elevate sha1sum to sha256sum for atomic headers check

From: David Laight
Date: Fri Mar 17 2023 - 05:00:30 EST


From: Bhaskar Chowdhury
> Sent: 15 March 2023 13:05
>
> Thought it would be a good idea to use a elevated mechanism i.e sha256sum

How can this change possibly work.
It is just a list of definitions read by another makefile.
You've changed the name of a definition without changing where it is used.
Also if the code is looking for a change, you'd need to change
what it is compared against.
In any case no one is worried about malicious attempts to change
things without being noticed, even sha1 is OTT.

David

>
> Signed-off-by: Bhaskar Chowdhury <unixbhaskar@xxxxxxxxx>
> ---
> Kbuild | 10 +++++-----
> 1 file changed, 5 insertions(+), 5 deletions(-)
>
> diff --git a/Kbuild b/Kbuild
> index 464b34a08f51..b74040346d76 100644
> --- a/Kbuild
> +++ b/Kbuild
> @@ -45,14 +45,14 @@ missing-syscalls: scripts/checksyscalls.sh $(offsets-file)
>
> # Check the manual modification of atomic headers
>
> -quiet_cmd_check_sha1 = CHKSHA1 $<
> - cmd_check_sha1 = \
> - if ! command -v sha1sum >/dev/null; then \
> - echo "warning: cannot check the header due to sha1sum missing"; \
> +quiet_cmd_check_sha256 = CHKSHA256 $<
> + cmd_check_sha256 = \
> + if ! command -v sha256sum >/dev/null; then \
> + echo "warning: cannot check the header due to sha256sum missing"; \
> exit 0; \
> fi; \
> if [ "$$(sed -n '$$s:// ::p' $<)" != \
> - "$$(sed '$$d' $< | sha1sum | sed 's/ .*//')" ]; then \
> + "$$(sed '$$d' $< | sha256sum | sed 's/ .*//')" ]; then \
> echo "error: $< has been modified." >&2; \
> exit 1; \
> fi; \
> --
> 2.39.2

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)