Bug#1038603: libisl23 0.25-1: amd64/i386 baseline violation: uses BMI instructions)

2023-07-22 Thread Roman Lebedev
Package: libisl23 Followup-For: Bug #1038603 X-Debbugs-Cc: Debian GCC Maintainers , debian-rele...@lists.debian.org -BEGIN PGP SIGNED MESSAGE- Hash: SHA512 While the issue has been resolved in isl 0.26-3, which was uploaded to unstable on 2023-06-19, nothing has been done for the stable

Bug#1029970: [wish] Please silence intentional unsigned overflow in an libstdc++ header

2023-01-29 Thread Roman Lebedev
Package: libstdc++-12-dev Version: 12.2.0-14 Severity: wishlist Tags: upstream patch -BEGIN PGP SIGNED MESSAGE- Hash: SHA512 Dear maintainer. As everyone knows, unsigned integer overflow is well-defined in C and C++. However, there are situations where you *know* that a particular code

Bug#1000927: fixed in unstable

2022-09-12 Thread Roman Lebedev
On Thu, 8 Sep 2022 15:20:23 +0200 Matthias Klose wrote: > fixed in unstable, now built using LLVM 13. When this bug was closed, it was titled "creduce: Please upgrade to llvm-toolchain-14", so i don't believe this is fixed, which is suboptimal because creduce is the only blocker for me to be

Bug#871250: g++-5: chokes on valid lambda code

2017-08-07 Thread Roman Lebedev
Package: g++-5 Version: 5.4.1-11 Severity: normal Tags: upstream -BEGIN PGP SIGNED MESSAGE- Hash: SHA512 The test-case is creduced. Only the GCC5 fails, gcc-4.9, 6 and 7 do not. clang-3.9, 4, 5, trunk is also fine with that code. $ g++-7 -std=c++14 -c main.cpp $ g++-6 -std=c++14 -c

Bug#859116: GCC miscompiles following code, unlike clang

2017-03-30 Thread Roman Lebedev
On Thu, Mar 30, 2017 at 4:51 PM, Gabriel Paubert <paub...@iram.es> wrote: > On Thu, Mar 30, 2017 at 04:23:41PM +0300, Roman Lebedev wrote: >> Package: g++-6 >> Version: 6.3.0-11 >> Severity: normal >> Tags: upstream >> >> -BEGIN PGP SIGNED MES

Bug#859116: GCC miscompiles following code, unlike clang

2017-03-30 Thread Roman Lebedev
Package: g++-6 Version: 6.3.0-11 Severity: normal Tags: upstream -BEGIN PGP SIGNED MESSAGE- Hash: SHA512 Hi. Yes, this issue should be sent upstream. But i'm unable to do that because their bugzilla does not want to reset my password. The attached testcase was minimized. With gcc (6),