Re: [PATCH] rs6000: Some builtins require IBM-128 long double format (PR103623)

2021-12-13 Thread Segher Boessenkool
Hi! On Mon, Dec 13, 2021 at 09:55:17AM -0600, Bill Schmidt wrote: > PR103623 shows that we ICE if __builtin_pack_longdouble or > __builtin_unpack_longdouble > is used when long double is not defined to be the IBM-128 (double-double) > format. > To solve this, I introduce a new built-in function

Re: [PATCH] rs6000: Some builtins require IBM-128 long double format (PR103623)

2021-12-13 Thread Bill Schmidt via Gcc-patches
Hi! On 12/13/21 2:15 PM, Martin Sebor wrote: > On 12/13/21 8:55 AM, Bill Schmidt via Gcc-patches wrote: >> Hi! >> >> PR103623 shows that we ICE if __builtin_pack_longdouble or >> __builtin_unpack_longdouble >> is used when long double is not defined to be the IBM-128 (double-double) >> format.

Re: [PATCH] rs6000: Some builtins require IBM-128 long double format (PR103623)

2021-12-13 Thread Martin Sebor via Gcc-patches
On 12/13/21 8:55 AM, Bill Schmidt via Gcc-patches wrote: Hi! PR103623 shows that we ICE if __builtin_pack_longdouble or __builtin_unpack_longdouble is used when long double is not defined to be the IBM-128 (double-double) format. To solve this, I introduce a new built-in function attribute

[PATCH] rs6000: Some builtins require IBM-128 long double format (PR103623)

2021-12-13 Thread Bill Schmidt via Gcc-patches
Hi! PR103623 shows that we ICE if __builtin_pack_longdouble or __builtin_unpack_longdouble is used when long double is not defined to be the IBM-128 (double-double) format. To solve this, I introduce a new built-in function attribute "ibmld" that enforces this requirement. Tested the fix on a