Re: [PATCH 1/9] gensupport: Fix define_subst operand renumbering.

[Andreas Krebbel <krebbel@linux.vnet.ibm.com>]

On 03/01/2016 02:38 PM, James Greenhalgh wrote:
> On Tue, Mar 01, 2016 at 01:35:18PM +0100, Andreas Krebbel wrote:
>> On 03/01/2016 01:15 PM, James Greenhalgh wrote:
>>> On Tue, Mar 01, 2016 at 10:29:28AM +0100, Andreas Krebbel wrote:
>>>> On 02/29/2016 02:36 PM, Bernd Schmidt wrote:
>>>>> On 02/29/2016 09:46 AM, Andreas Krebbel wrote:
>>>>>> Ok for mainline?
>>>>>>
>>>>>> 	* gensupport.c (process_substs_on_one_elem): Split loop to
>>>>>> 	complete mark_operands_used_in_match_dup on all expressions in the
>>>>>> 	vector first.
>>>>>> 	(adjust_operands_numbers): Inline into process_substs_on_one_elem
>>>>>> 	and remove function.
>>>>>
>>>>> Didn't I approve this a while ago? Not sure it's appropriate for stage4 
>>>>> though; is this series fixing an important regression?
>>>>
>>>> Yes you did. I didn't commit it until the rest of the patch series was ready to commit.  The patch
>>>> series fixes a fundamental problem in the backend. The first iteration was posted before stage 4 but
>>>> it took me a few iterations to get it right.
>>>>
>>>> I've committed the patch now after retesting.
>>>
>>> This looks like it has caused failures in the following tests on an
>>> x86_64-none-linux-gnu build.
>>
>> I've regression tested the patch on x86_64 as well.  Are there specific
>> options required to enable these tests?
> 
> The bisect robot just builds a stage one compiler, configured with:
> 
>   ./configure --disable-bootstrap, --enable-languages=c,c++,fortran
>               --disable-multilib --disable-libsanitizer
> 
> My system GCC is a 5.2 from the release sources with:
> 
>   ../gcc-5.2.0/configure --with-bugurl='Good luck'
>                          --enable-languages=c,c++,go,fortran,objc,obj-c++
>                          --prefix=/work/install-gcc-5.2.0 --enable-shared
>                          --enable-linker-build-id --without-included-gettext
>                          --enable-threads=posix --enable-nls
> 			 --enable-clocale=gnu --enable-libstdcxx-debug
>                          --enable-libstdcxx-time=yes
>                          --enable-gnu-unique-object --disable-libmudflap
>                          --enable-plugin --with-system-zlib
>                          --disable-browser-plugin --enable-java-awt=gtk
>                          --enable-gtk-cairo --with-arch-directory=amd64
>                          --enable-objc-gc --enable-multiarch
>                          --disable-werror --with-arch-32=i686
>                          --with-abi=m64 --with-multilib-list=m32,m64,mx32
>                          --with-tune=native --enable-checking=release
>                          --build=x86_64-linux-gnu --host=x86_64-linux-gnu
>                          --target=x86_64-linux-gnu
> 
> I tried a full bootstrap at that revision and still see these failures.
> Who knows what state has been corrupted, or that you silently get away with,
> if this is an undefined behaviour somewhere :-). I haven't tried with a
> valgrind checking build to see what it can spot.

Ok. Thanks for the infos.  I'll try to have a look. I've reverted the patch now.

Bye,

-Andreas-