On 12/10/20 3:58 AM, Alexandre Oliva wrote:
> On Dec 3, 2020, Alexandre Oliva wrote:
>
>> +local_define_builtin ("__builtin___clear_cache", ftype,
>> + BUILT_IN_CLEAR_CACHE,
>> + "__builtin___clear_cache",
>> + ECF_NOTHROW);
>
On Dec 3, 2020, Alexandre Oliva wrote:
> +local_define_builtin ("__builtin___clear_cache", ftype,
> + BUILT_IN_CLEAR_CACHE,
> + "__builtin___clear_cache",
> + ECF_NOTHROW);
Ugh, so that somehow worked for aarch64-linux-gnu-gf
On Dec 5, 2020, Jakub Jelinek wrote:
> On Sat, Dec 05, 2020 at 06:01:59PM -0300, Alexandre Oliva wrote:
>> On Dec 5, 2020, Andreas Schwab wrote:
>>
>> > ../../../../libffi/src/aarch64/ffi.c: In function 'ffi_prep_closure_loc':
>> > ../../../../libffi/src/aarch64/ffi.c:67:3: internal compiler
On Sat, Dec 05, 2020 at 06:01:59PM -0300, Alexandre Oliva wrote:
> On Dec 5, 2020, Andreas Schwab wrote:
>
> > ../../../../libffi/src/aarch64/ffi.c: In function 'ffi_prep_closure_loc':
> > ../../../../libffi/src/aarch64/ffi.c:67:3: internal compiler error: in
> > emit_library_call_value_1, at c
On Dec 5, 2020, Andreas Schwab wrote:
> ../../../../libffi/src/aarch64/ffi.c: In function 'ffi_prep_closure_loc':
> ../../../../libffi/src/aarch64/ffi.c:67:3: internal compiler error: in
> emit_library_call_value_1, at calls.c:5300
>67 | __builtin___clear_cache (start, end);
> | ^
../../../../libffi/src/aarch64/ffi.c: In function 'ffi_prep_closure_loc':
../../../../libffi/src/aarch64/ffi.c:67:3: internal compiler error: in
emit_library_call_value_1, at calls.c:5300
67 | __builtin___clear_cache (start, end);
| ^~~~
Andreas.
--
On 12/3/20 7:08 AM, Alexandre Oliva wrote:
> On Dec 3, 2020, Christophe Lyon wrote:
>
>> This patches causes a lot of regressions in fortran on arm and aarch64,
> On Dec 3, 2020, Andreas Schwab wrote:
>
>> On Dez 03 2020, Andreas Schwab wrote:
>>> ../../../../libffi/src/aarch64/ffi.c: In fun
On Dec 3, 2020, Christophe Lyon wrote:
> This patches causes a lot of regressions in fortran on arm and aarch64,
On Dec 3, 2020, Andreas Schwab wrote:
> On Dez 03 2020, Andreas Schwab wrote:
>> ../../../../libffi/src/aarch64/ffi.c: In function 'ffi_prep_closure_loc':
>> ../../../../libffi/s
On Dez 03 2020, Andreas Schwab wrote:
> ../../../../libffi/src/aarch64/ffi.c: In function 'ffi_prep_closure_loc':
> ../../../../libffi/src/aarch64/ffi.c:67:3: error: both arguments to
> '__builtin___clear_cache' must be pointers
>67 | __builtin___clear_cache (start, end);
> | ^~
../../../../libffi/src/aarch64/ffi.c: In function 'ffi_prep_closure_loc':
../../../../libffi/src/aarch64/ffi.c:67:3: error: both arguments to
'__builtin___clear_cache' must be pointers
67 | __builtin___clear_cache (start, end);
| ^~~~
Andreas.
--
And
On Dec 3, 2020, Christophe Lyon wrote:
> This patches causes a lot of regressions in fortran on arm and aarch64,
Thanks, on it.
--
Alexandre Oliva, happy hacker https://FSFLA.org/blogs/lxo/
Free Software Activist GNU Toolchain Engineer
Vim, Vi, Voltei pro Emacs -- GNUlius
Hi Alexandre,
On Wed, 2 Dec 2020 at 19:23, Jeff Law via Gcc-patches
wrote:
>
>
>
> On 11/10/20 7:35 PM, Alexandre Oliva wrote:
> > This patch introduces maybe_emit_call_builtin___clear_cache for the
> > builtin expander machinery and the trampoline initializers to use to
> > clear the instruction
On 11/10/20 7:35 PM, Alexandre Oliva wrote:
> This patch introduces maybe_emit_call_builtin___clear_cache for the
> builtin expander machinery and the trampoline initializers to use to
> clear the instruction cache, removing a source of inconsistencies and
> subtle errors in low-level machinery.
Hi Alex,
Thanks for your proposal on this!
> On 11 Nov 2020, at 03:35, Alexandre Oliva wrote:
>
> This patch introduces maybe_emit_call_builtin___clear_cache for the
> builtin expander machinery and the trampoline initializers to use to
> clear the instruction cache, removing a source of incons
This patch introduces maybe_emit_call_builtin___clear_cache for the
builtin expander machinery and the trampoline initializers to use to
clear the instruction cache, removing a source of inconsistencies and
subtle errors in low-level machinery.
I've adjusted all trampoline_init implementations t
15 matches
Mail list logo