Re: [PATCH,rs6000] GCC -7, no swap optimization for vpermxor instruction

2018-03-21 Thread Peter Bergner
On 3/21/18 5:16 PM, Carl Love wrote: > The following patch is a back port from mainlin of a fix for [snip] > > 2018-03-21 Carl Love > > * config/rs6000/r6000.c (rtx_is_swappable_p): Add case UNSPEC_VPERMXOR. > > gcc/testsuite/ChangeLog: > > 2018-03-21 Carl Love > * gcc.target/

C++ PATCH for c++/81311, wrong C++17 overload resolution

2018-03-21 Thread Jason Merrill
Here, the code in build_special_member_function to avoid involving the copy constructor in initialization from a prvalue was causing wrongly different overload resolution. To fix it, instead of trying to produce a prvalue there, we now go through normal overload resolution and then specifically av

Re: [PR c++/84789] do not resolve typename into template-independent

2018-03-21 Thread Jason Merrill
OK, thanks. On Wed, Mar 21, 2018 at 11:48 AM, Jason Merrill wrote: > On Tue, Mar 20, 2018 at 11:27 PM, Alexandre Oliva wrote: >> On Mar 20, 2018, Jason Merrill wrote: >> >>> On Tue, Mar 20, 2018 at 6:07 PM, Alexandre Oliva wrote: On Mar 20, 2018, Jason Merrill wrote: > that doesn't m

Re: [og7] vector_length extension part 2: Generalize state propagation and synchronization

2018-03-21 Thread Cesar Philippidis
On 03/21/2018 10:10 AM, Tom de Vries wrote: > On 03/02/2018 05:55 PM, Cesar Philippidis wrote: >> In addition, nvptx_cta_sync and the corresponding nvptx_barsync insn, >> have been extended to take a barrier ID and a thread count. The idea >> here is to assign one barrier for each logical vector. W

<    1   2