Re: [PATCH] powerpc/boot: Delete unneeded .globl _zimage_start

[Fangrui Song <maskray@google.com>]

On 2020-03-25, Alan Modra wrote:
>On Wed, Mar 25, 2020 at 05:22:31AM +0000, Joel Stanley wrote:
>> On Wed, 25 Mar 2020 at 05:19, Fangrui Song <maskray@google.com> wrote:
>> >
>> > .globl sets the symbol binding to STB_GLOBAL while .weak sets the
>> > binding to STB_WEAK. They should not be used together. It is accidetal
>> > rather then intentional that GNU as let .weak override .globl while
>> > clang integrated assembler let the last win.
>
>No, it isn't accidental.  gas deliberately lets .weak override .globl.
>Since 1996-07-26, git commit 5ca547dc239

Fair. I am sadded by this commit.

% sed -n 'N;/.globl.*\n.*.weak/p;D' binutils-gdb/gas/testsuite/**/*(.)

I checked all occurrences and none is a real test excercising the behavior (.weak override .globl).
All seem accidental.


It is unclear that clang integrated assembler should copy this
behavior, though. For the record, I asked on binutils@sourceware.org
whether the assembler should error when .weak/.local can override a
previous binding directive. It was rejected
https://sourceware.org/pipermail/binutils/2020-March/110376.html

On the clang integrated assembler side, we may try building things with
an error or leave the overriding behavior as is.

>I'm fine with the patch so far as it is true that there is no need for
>both .globl and .weak (and it looks silly to have both), but the
>explanation isn't true.  The patch is needed because the clang
>assembler is incompatible with gas in this detail.

Since using one of .weak|.globl is nearly well-known, I'll send PATCH v2 with
the description updated.

On 2020-03-25, Segher Boessenkool wrote:
>Nothing is "overridden".
>
>The as manual says (.weak):
>
>  This directive sets the weak attribute on the comma separated list of
>  symbol 'names'.  If the symbols do not already exist, they will be
>  created.
>
>so this behaviour is obviously as intended (or was later documented in
>any case), so LLVM has a bug to fix (whether you like this (much saner)
>behaviour or not).

I will probably not call this a bug. I have recently discovered other
discrepancy for which I think copying gas behaviors can just clutter up the
code. We may need our own assembler documentation at some point.

>> > Fixes: cd197ffcf10b "[POWERPC] zImage: Cleanup and improve zImage entry point"
>> > Fixes: ee9d21b3b358 "powerpc/boot: Ensure _zimage_start is a weak symbol"
>> > Link: https://github.com/ClangBuiltLinux/linux/issues/937
>> > Signed-off-by: Fangrui Song <maskray@google.com>
>> > Cc: Joel Stanley <joel@jms.id.au>
>> > Cc: Michael Ellerman <mpe@ellerman.id.au>
>> > Cc: Nick Desaulniers <ndesaulniers@google.com>
>> > Cc: clang-built-linux@googlegroups.com
>> > ---
>> >  arch/powerpc/boot/crt0.S | 3 ---
>> >  1 file changed, 3 deletions(-)
>> >
>> > diff --git a/arch/powerpc/boot/crt0.S b/arch/powerpc/boot/crt0.S
>> > index 92608f34d312..1d83966f5ef6 100644
>> > --- a/arch/powerpc/boot/crt0.S
>> > +++ b/arch/powerpc/boot/crt0.S
>> > @@ -44,9 +44,6 @@ p_end:                .long   _end
>> >  p_pstack:      .long   _platform_stack_top
>> >  #endif
>> >
>> > -       .globl  _zimage_start
>> > -       /* Clang appears to require the .weak directive to be after the symbol
>> > -        * is defined. See https://bugs.llvm.org/show_bug.cgi?id=38921  */
>> >         .weak   _zimage_start
>> >  _zimage_start:
>>
>> Your explanation makes sense to me. I've added Alan to cc for his review.
>>
>> Reviewed-by: Joel Stanley <joel@jms.id.au>
>>
>> Thanks for the patch.
>>
>> Cheers,
>>
>> Joel
>>
>> >         .globl  _zimage_start_lib
>> > --
>> > 2.25.1.696.g5e7596f4ac-goog
>
>-- 
>Alan Modra
>Australia Development Lab, IBM