[PATCH] efi: libstub: add strnlen() implementation

[PATCH] efi: libstub: add strnlen() implementation

[Ard Biesheuvel]

Commit 91feabc2e224 ("scripts/dtc: Update to upstream commit
b06e55c88b9b") introduces references to the strnlen() string function,
which is not currently defined in the EFI stub. So add it to our string.c

Signed-off-by: Ard Biesheuvel <ard.biesheuvel-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
---
 drivers/firmware/efi/libstub/string.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/drivers/firmware/efi/libstub/string.c b/drivers/firmware/efi/libstub/string.c
index 09d5a0894343..d4055ab80116 100644
--- a/drivers/firmware/efi/libstub/string.c
+++ b/drivers/firmware/efi/libstub/string.c
@@ -55,3 +55,19 @@ int strncmp(const char *cs, const char *ct, size_t count)
 	return 0;
 }
 #endif
+
+#ifndef __HAVE_ARCH_STRNLEN
+/**
+ * strnlen - Find the length of a length-limited string
+ * @s: The string to be sized
+ * @count: The maximum number of bytes to search
+ */
+size_t strnlen(const char *s, size_t count)
+{
+	const char *sc;
+
+	for (sc = s; count-- && *sc != '\0'; ++sc)
+		/* nothing */;
+	return sc - s;
+}
+#endif
-- 
2.5.0

Re: [PATCH] efi: libstub: add strnlen() implementation

[Matt Fleming]

On Tue, 16 Feb, at 04:29:19PM, Ard Biesheuvel wrote:
> Commit 91feabc2e224 ("scripts/dtc: Update to upstream commit
> b06e55c88b9b") introduces references to the strnlen() string function,
> which is not currently defined in the EFI stub. So add it to our string.c
> 
> Signed-off-by: Ard Biesheuvel <ard.biesheuvel-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
> ---
>  drivers/firmware/efi/libstub/string.c | 16 ++++++++++++++++
>  1 file changed, 16 insertions(+)
 
Does this patch fix a build failure for ARM? Which tree is the above
commit in?

Re: [PATCH] efi: libstub: add strnlen() implementation

[Ard Biesheuvel]

On 18 February 2016 at 12:04, Matt Fleming <matt-mF/unelCI9GS6iBeEJttW/XRex20P6io@public.gmane.org> wrote:
> On Tue, 16 Feb, at 04:29:19PM, Ard Biesheuvel wrote:
>> Commit 91feabc2e224 ("scripts/dtc: Update to upstream commit
>> b06e55c88b9b") introduces references to the strnlen() string function,
>> which is not currently defined in the EFI stub. So add it to our string.c
>>
>> Signed-off-by: Ard Biesheuvel <ard.biesheuvel-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
>> ---
>>  drivers/firmware/efi/libstub/string.c | 16 ++++++++++++++++
>>  1 file changed, 16 insertions(+)
>
> Does this patch fix a build failure for ARM? Which tree is the above
> commit in?

It is in -next, and it will break the build once it hits upstream.

Re: [PATCH] efi: libstub: add strnlen() implementation

[Matt Fleming]

On Thu, 18 Feb, at 12:05:51PM, Ard Biesheuvel wrote:
> On 18 February 2016 at 12:04, Matt Fleming <matt-mF/unelCI9GS6iBeEJttW/XRex20P6io@public.gmane.org> wrote:
> > On Tue, 16 Feb, at 04:29:19PM, Ard Biesheuvel wrote:
> >> Commit 91feabc2e224 ("scripts/dtc: Update to upstream commit
> >> b06e55c88b9b") introduces references to the strnlen() string function,
> >> which is not currently defined in the EFI stub. So add it to our string.c
> >>
> >> Signed-off-by: Ard Biesheuvel <ard.biesheuvel-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
> >> ---
> >>  drivers/firmware/efi/libstub/string.c | 16 ++++++++++++++++
> >>  1 file changed, 16 insertions(+)
> >
> > Does this patch fix a build failure for ARM? Which tree is the above
> > commit in?
> 
> It is in -next, and it will break the build once it hits upstream.

OK, could you resend and include the build failure in the commit log
and make a reference to fixing a build issue? 

Re: [PATCH] efi: libstub: add strnlen() implementation

[Ard Biesheuvel]

On 18 February 2016 at 14:33, Matt Fleming <matt-mF/unelCI9GS6iBeEJttW/XRex20P6io@public.gmane.org> wrote:
> On Thu, 18 Feb, at 12:05:51PM, Ard Biesheuvel wrote:
>> On 18 February 2016 at 12:04, Matt Fleming <matt-mF/unelCI9GS6iBeEJttW/XRex20P6io@public.gmane.org> wrote:
>> > On Tue, 16 Feb, at 04:29:19PM, Ard Biesheuvel wrote:
>> >> Commit 91feabc2e224 ("scripts/dtc: Update to upstream commit
>> >> b06e55c88b9b") introduces references to the strnlen() string function,
>> >> which is not currently defined in the EFI stub. So add it to our string.c
>> >>
>> >> Signed-off-by: Ard Biesheuvel <ard.biesheuvel-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
>> >> ---
>> >>  drivers/firmware/efi/libstub/string.c | 16 ++++++++++++++++
>> >>  1 file changed, 16 insertions(+)
>> >
>> > Does this patch fix a build failure for ARM? Which tree is the above
>> > commit in?
>>
>> It is in -next, and it will break the build once it hits upstream.
>
> OK, could you resend and include the build failure in the commit log
> and make a reference to fixing a build issue?

Actually, I spoke too soon. While the issue does exist on arm64, ARM
already has a special version of strnlen() in its decompressor.

Sorry for the noise.