* [PATCH] init: Reject bare rodata parameter before architecture parsing
@ 2026-11-13 20:23 Thorsten Blum
2026-11-15 6:47 ` Mike Rapoport
0 siblings, 1 reply; 2+ messages in thread
From: Thorsten Blum @ 2026-11-13 20:23 UTC (permalink / raw)
To: Andrew Morton, Masami Hiramatsu (Google), Christian Brauner,
Breno Leitao, Sang-Heon Jeon, Wilson Felipe Pereira,
Mike Rapoport (Microsoft), Ryan Roberts
Cc: Thorsten Blum, linux-kernel
When a bare rodata parameter is specified on the kernel command line
without the '=' separator, early parameter parsing passes NULL to
set_debug_rodata().
Current hooks already handle NULL safely. Reject NULL before calling
arch_parse_debug_rodata() so architecture hooks always receive a
non-NULL string.
Return -EINVAL to report the malformed option instead of falling through
to the existing warning. Also remove the now-redundant NULL checks from
the "on" and "off" comparisons.
Signed-off-by: Thorsten Blum <blum@kernel.org>
---
init/main.c | 7 +++++--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/init/main.c b/init/main.c
index 7b0c8c34e6a9..768b65f2498f 100644
--- a/init/main.c
+++ b/init/main.c
@@ -1566,12 +1566,15 @@ static inline bool arch_parse_debug_rodata(char *str) { return false; }
static int __init set_debug_rodata(char *str)
{
+ if (!str)
+ return -EINVAL;
+
if (arch_parse_debug_rodata(str))
return 0;
- if (str && !strcmp(str, "on"))
+ if (!strcmp(str, "on"))
rodata_enabled = true;
- else if (str && !strcmp(str, "off"))
+ else if (!strcmp(str, "off"))
rodata_enabled = false;
else
pr_warn("Invalid option string for rodata: '%s'\n", str);
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] init: Reject bare rodata parameter before architecture parsing
2026-11-13 20:23 [PATCH] init: Reject bare rodata parameter before architecture parsing Thorsten Blum
@ 2026-11-15 6:47 ` Mike Rapoport
0 siblings, 0 replies; 2+ messages in thread
From: Mike Rapoport @ 2026-11-15 6:47 UTC (permalink / raw)
To: Thorsten Blum
Cc: Andrew Morton, Masami Hiramatsu (Google), Christian Brauner,
Breno Leitao, Sang-Heon Jeon, Wilson Felipe Pereira, Ryan Roberts,
linux-kernel
On Fri, Nov 13, 2026 at 09:23:19PM +0100, Thorsten Blum wrote:
> When a bare rodata parameter is specified on the kernel command line
> without the '=' separator, early parameter parsing passes NULL to
> set_debug_rodata().
>
> Current hooks already handle NULL safely. Reject NULL before calling
> arch_parse_debug_rodata() so architecture hooks always receive a
> non-NULL string.
>
> Return -EINVAL to report the malformed option instead of falling through
> to the existing warning. Also remove the now-redundant NULL checks from
> the "on" and "off" comparisons.
>
> Signed-off-by: Thorsten Blum <blum@kernel.org>
Reviewed-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
> ---
> init/main.c | 7 +++++--
> 1 file changed, 5 insertions(+), 2 deletions(-)
>
> diff --git a/init/main.c b/init/main.c
> index 7b0c8c34e6a9..768b65f2498f 100644
> --- a/init/main.c
> +++ b/init/main.c
> @@ -1566,12 +1566,15 @@ static inline bool arch_parse_debug_rodata(char *str) { return false; }
>
> static int __init set_debug_rodata(char *str)
> {
> + if (!str)
> + return -EINVAL;
> +
> if (arch_parse_debug_rodata(str))
> return 0;
>
> - if (str && !strcmp(str, "on"))
> + if (!strcmp(str, "on"))
> rodata_enabled = true;
> - else if (str && !strcmp(str, "off"))
> + else if (!strcmp(str, "off"))
> rodata_enabled = false;
> else
> pr_warn("Invalid option string for rodata: '%s'\n", str);
--
Sincerely yours,
Mike.
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-11-15 6:47 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-11-13 20:23 [PATCH] init: Reject bare rodata parameter before architecture parsing Thorsten Blum
2026-11-15 6:47 ` Mike Rapoport
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox