From mboxrd@z Thu Jan  1 00:00:00 1970
From: andy@greyhouse.net (Andy Gospodarek)
Date: Fri, 17 Mar 2017 13:43:35 -0400
Subject: [RFC arm64] samples/bpf: explicitly exclude sysreg sections with
 asm macros
In-Reply-To: <383e1626-335b-01fe-5921-5956d7201cf4@arm.com>
References: <20170310192656.GA13928@C02RW35GFVH8>
 <20170310204112.GA21382@leverpostej>
 <20170315183130.GB32819@C02RW35GFVH8.dhcp.broadcom.net>
 <20170315191741.GB29452@leverpostej>
 <20170315205404.GA43995@C02RW35GFVH8.dhcp.broadcom.net>
 <20170316104204.GA31101@leverpostej>
 <20170316210454.GA53792@C02RW35GFVH8.dhcp.broadcom.net>
 <e842a237-9206-99ed-624b-4f3cf15cc557@arm.com>
 <20170317135836.GA58086@C02RW35GFVH8.dhcp.broadcom.net>
 <383e1626-335b-01fe-5921-5956d7201cf4@arm.com>
Message-ID: <20170317174335.GA63516@C02RW35GFVH8.dhcp.broadcom.net>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

On Fri, Mar 17, 2017 at 04:57:20PM +0000, Robin Murphy wrote:
> On 17/03/17 13:58, Andy Gospodarek wrote:
> > On Fri, Mar 17, 2017 at 11:11:59AM +0000, Robin Murphy wrote:
> >> On 16/03/17 21:04, Andy Gospodarek wrote:
> >>>
> >>> While I generally agree isn't the entire reason this inline asm exists
> >>> -- to fix a binutils issue?  :-)
> >>
> >> Binutils is required to build the arm64 kernel; if certain versions of
> >> binutils packaged by arm64 distros can't cope with certain instructions
> >> we need to use, we have to do what we can to make it build. However, if
> >> LLVM cannot cope with inline assembly on arm64, then it's pretty clear
> >> that the kernel cannot be built with LLVM anyway, thus there is nothing
> >> to work around in that regard. Sorry, you've got a straw man there ;)
> > 
> > I know, right?  I'll see if I can come up with something that appears to
> > be less fragile and/or intrusive to the arm64 arch code as I'd really
> > like to stop carrying a patch in my devel tree for this.
> 
> I'm honestly not sure if I'm serious or not, but:
> 
> diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
> index 72c58675973e..f52ebad7b096 100644
> --- a/samples/bpf/Makefile
> +++ b/samples/bpf/Makefile
> @@ -154,6 +154,6 @@ $(src)/*.c: verify_target_bpf
>  # useless for BPF samples.
>  $(obj)/%.o: $(src)/%.c
>         $(CLANG) $(NOSTDINC_FLAGS) $(LINUXINCLUDE) $(EXTRA_CFLAGS) \
> -               -D__KERNEL__ -D__ASM_SYSREG_H -Wno-unused-value
> -Wno-pointer-sign \
> +               -D__KERNEL__ -D'asm(...)' -Wno-unused-value
> -Wno-pointer-sign \
>                 -Wno-compare-distinct-pointer-types \
>                 -O2 -emit-llvm -c $< -o -| $(LLC) -march=bpf
> -filetype=obj -o $@
> 

Hahahaha, I know what you mean.

Scarily, I had a similar thought.  :-)