ABataev added a comment. In D71241#1787998 <https://reviews.llvm.org/D71241#1787998>, @jdoerfert wrote:
> In D71241#1787888 <https://reviews.llvm.org/D71241#1787888>, @ABataev wrote: > > > Hal, are we going to support something like this? > > > I'm not Hal but I will answer anyway. > > > void cpu() { asm("nop"); } > > > > #pragma omp declare variant(cpu) match(device = {kind(cpu)}) > > void wrong_asm() { > > asm ("xxx"); > > } > > > > > > Currently, there is an error when we try to emit the assembler output. > > While it is unclear to me what you think should happen, an error pointing at > `"xxx"` is to be expected without further information on how this is compiled. Shall we emit function `wrong_asm` or not? If we're not going to use it in our program, only `cpu` must be used, `wrong_asm` is not required and must not be emitted. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D71241/new/ https://reviews.llvm.org/D71241 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
