Re: [PATCH 05/11] arm64: Kconfig: Clean up tautological LLVM version checks

[Nathan Chancellor]

On Fri, Jan 26, 2024 at 12:59:55PM +0000, Mark Rutland wrote:
> On Thu, Jan 25, 2024 at 03:55:11PM -0700, Nathan Chancellor wrote:
> > Now that the minimum supported version of LLVM for building the kernel
> > has been bumped to 13.0.1, several conditions become tautologies, as
> > they will always be true because the build will fail during the
> > configuration stage for older LLVM versions. Drop them, as they are
> > unnecessary.
> > 
> > Signed-off-by: Nathan Chancellor <nathan@xxxxxxxxxx>
> > ---
> > Cc: catalin.marinas@xxxxxxx
> > Cc: will@xxxxxxxxxx
> > Cc: mark.rutland@xxxxxxx
> > Cc: linux-arm-kernel@xxxxxxxxxxxxxxxxxxx
> > ---
> >  arch/arm64/Kconfig | 5 +----
> >  1 file changed, 1 insertion(+), 4 deletions(-)
> > 
> > diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
> > index 5a8acca4dbf4..cb34e7d780c0 100644
> > --- a/arch/arm64/Kconfig
> > +++ b/arch/arm64/Kconfig
> > @@ -383,7 +383,7 @@ config BUILTIN_RETURN_ADDRESS_STRIPS_PAC
> >  	bool
> >  	# Clang's __builtin_return_adddress() strips the PAC since 12.0.0
> >  	# https://github.com/llvm/llvm-project/commit/2a96f47c5ffca84cd774ad402cacd137f4bf45e2
> > -	default y if CC_IS_CLANG && (CLANG_VERSION >= 120000)
> > +	default y if CC_IS_CLANG
> >  	# GCC's __builtin_return_address() strips the PAC since 11.1.0,
> >  	# and this was backported to 10.2.0, 9.4.0, 8.5.0, but not earlier
> >  	# https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94891
> > @@ -1387,7 +1387,6 @@ choice
> >  
> >  config CPU_BIG_ENDIAN
> >  	bool "Build big-endian kernel"
> > -	depends on !LD_IS_LLD || LLD_VERSION >= 130000
> >  	# https://github.com/llvm/llvm-project/commit/1379b150991f70a5782e9a143c2ba5308da1161c
> 
> We can delete the URL here, since that was just to describe why this depended
> upon LLVM 13+; it's weird for it to sit here on its own.

I think this is the URL for the fix for the problem brought up by
commit 146a15b87335 ("arm64: Restrict CPU_BIG_ENDIAN to GNU as or LLVM
IAS 15.x or newer"), so I think it should stay? It does not look like I
ever added a link or context for the LLD line, I definitely should have.

> The URL above for __builtin_return_address() can stay or go; it may as well
> stay since we have the comment aboout LLvm 12+ above it.

That's the conclusion I came to as well.

Thanks a lot for taking a look!

Cheers,
Nathan

> With that:
> 
> Acked-by: Mark Rutland <mark.rutland@xxxxxxx>
> 
> Mark.
> 
> >  	depends on AS_IS_GNU || AS_VERSION >= 150000
> >  	help
> > @@ -2018,8 +2017,6 @@ config ARM64_BTI_KERNEL
> >  	depends on !CC_IS_GCC || GCC_VERSION >= 100100
> >  	# https://gcc.gnu.org/bugzilla/show_bug.cgi?id=106671
> >  	depends on !CC_IS_GCC
> > -	# https://github.com/llvm/llvm-project/commit/a88c722e687e6780dcd6a58718350dc76fcc4cc9
> > -	depends on !CC_IS_CLANG || CLANG_VERSION >= 120000
> >  	depends on (!FUNCTION_GRAPH_TRACER || DYNAMIC_FTRACE_WITH_ARGS)
> >  	help
> >  	  Build the kernel with Branch Target Identification annotations
> > 
> > -- 
> > 2.43.0
> > 
>