Hi Michael,
On 12/27/2018 04:35 PM, Gustavo Romero wrote:
> On 12/27/2018 12:41 PM, Michael Crusoe wrote:
>> https://buildd.debian.org/status/fetch.php?pkg=htslib&arch=ppc64el&ver=1.9-7&stamp=1545236716&raw=0
>>
>> Can I get some assistance here? Rebuilding using Qemu and the earlier source packages produces the same error, so maybe this is a regression in the compiler?
>
> I'm not sure if it's a compiler regression, but I can only reproduce that
> with gcc-8. Compiling with gcc-7 is fine.
I can see it fails with gcc-8 plus -O3, but not with -O2, because:
A (big) difference I see from the successful prior build are the following build-dep package versions:
successful combination:
binutils_2.31.1-10
g++-8_8.2.0-10
gcc-8_8.2.0-10
libc6-dev_2.28-1
libstdc++-8-dev_8.2.0-10
libstdc++6_8.2.0-10
linux-libc-dev_4.18.20-2
failed combination:
binutils_2.31.1-11
g++-8_8.2.0-13
gcc-8_8.2.0-13
libc6-dev_2.28-3
libstdc++-8-dev_8.2.0-13
libstdc++6_8.2.0-13
linux-libc-dev_4.19.12-1
So, the code for this lib seems a little bit suspicious with all those cast /
missing casts, union (to convert uint32_t array to float), and different ways to
convert uint8_t* to uint32_t, like in le_to_u32(). So some implicit truncation
might be working by luck on gcc7 and gcc8 plus -O2 but optimization (O3) "broke"
it.
I can take a closer look tomorrow if nobody else bites the bait.
That would be very welcome! Thanks for looking so far.
Using qemu I do get different failures with gcc-7 than gcc-8.
Cheers,
Gustavo
> Cheers,
> Gustavo
>
>> --
>> Michael R. Crusoe
>> Co-founder & Lead, Common Workflow Language project <http://www.commonwl.org/>
>> Direktorius, VšĮ "Darbo eigos", Vilnius, Lithuania
>> https://orcid.org/0000-0002-2961-9670 <https://impactstory.org/u/0000-0002-2961-9670>
>> mrc@commonwl.org <mailto:mrc@commonwl.org>
>> +1 480 627 9108 / +370 653 11125