Re: [PATCH] gimple.cc: Adjust gimple_call_builtin_p and gimple_call_combined_fn [PR105150]

2022-04-06 Thread Richard Sandiford via Gcc-patches
Jakub Jelinek writes: > On Wed, Apr 06, 2022 at 11:52:23AM +0200, Richard Biener wrote: >> On Wed, 6 Apr 2022, Jakub Jelinek wrote: >> >> > On Wed, Apr 06, 2022 at 09:41:44AM +0100, Richard Sandiford wrote: >> > > But it seems like the magic incantation to detect “real” built-in >> > > function

Re: [PATCH] gimple.cc: Adjust gimple_call_builtin_p and gimple_call_combined_fn [PR105150]

2022-04-06 Thread Richard Biener via Gcc-patches
On Wed, 6 Apr 2022, Jakub Jelinek wrote: > On Wed, Apr 06, 2022 at 11:52:23AM +0200, Richard Biener wrote: > > On Wed, 6 Apr 2022, Jakub Jelinek wrote: > > > > > On Wed, Apr 06, 2022 at 09:41:44AM +0100, Richard Sandiford wrote: > > > > But it seems like the magic incantation to detect “real”

Re: [PATCH] gimple.cc: Adjust gimple_call_builtin_p and gimple_call_combined_fn [PR105150]

2022-04-06 Thread Jakub Jelinek via Gcc-patches
On Wed, Apr 06, 2022 at 11:52:23AM +0200, Richard Biener wrote: > On Wed, 6 Apr 2022, Jakub Jelinek wrote: > > > On Wed, Apr 06, 2022 at 09:41:44AM +0100, Richard Sandiford wrote: > > > But it seems like the magic incantation to detect “real” built-in > > > function calls is getting longer and

Re: [PATCH] gimple.cc: Adjust gimple_call_builtin_p and gimple_call_combined_fn [PR105150]

2022-04-06 Thread Richard Biener via Gcc-patches
On Wed, 6 Apr 2022, Jakub Jelinek wrote: > On Wed, Apr 06, 2022 at 09:41:44AM +0100, Richard Sandiford wrote: > > But it seems like the magic incantation to detect “real” built-in > > function calls is getting longer and longer. Can we not abstract this > > in a single place rather than have to

Re: [PATCH] gimple.cc: Adjust gimple_call_builtin_p and gimple_call_combined_fn [PR105150]

2022-04-06 Thread Jakub Jelinek via Gcc-patches
On Wed, Apr 06, 2022 at 09:41:44AM +0100, Richard Sandiford wrote: > But it seems like the magic incantation to detect “real” built-in > function calls is getting longer and longer. Can we not abstract this > in a single place rather than have to repeat the same long sequence in > multiple

Re: [PATCH] gimple.cc: Adjust gimple_call_builtin_p and gimple_call_combined_fn [PR105150]

2022-04-06 Thread Richard Sandiford via Gcc-patches
Richard Biener via Gcc-patches writes: > On Wed, 6 Apr 2022, Jakub Jelinek wrote: > >> On Wed, Apr 06, 2022 at 08:13:24AM +0200, Richard Biener wrote: >> > On Tue, 5 Apr 2022, Jakub Jelinek wrote: >> > >> > > On Tue, Apr 05, 2022 at 11:28:53AM +0200, Richard Biener wrote: >> > > > > In GIMPLE,

Re: [PATCH] gimple.cc: Adjust gimple_call_builtin_p and gimple_call_combined_fn [PR105150]

2022-04-06 Thread Richard Biener via Gcc-patches
On Wed, 6 Apr 2022, Jakub Jelinek wrote: > On Wed, Apr 06, 2022 at 08:13:24AM +0200, Richard Biener wrote: > > On Tue, 5 Apr 2022, Jakub Jelinek wrote: > > > > > On Tue, Apr 05, 2022 at 11:28:53AM +0200, Richard Biener wrote: > > > > > In GIMPLE, we call: > > > > > &&

Re: [PATCH] gimple.cc: Adjust gimple_call_builtin_p and gimple_call_combined_fn [PR105150]

2022-04-06 Thread Jakub Jelinek via Gcc-patches
On Wed, Apr 06, 2022 at 08:13:24AM +0200, Richard Biener wrote: > On Tue, 5 Apr 2022, Jakub Jelinek wrote: > > > On Tue, Apr 05, 2022 at 11:28:53AM +0200, Richard Biener wrote: > > > > In GIMPLE, we call: > > > > && gimple_builtin_call_types_compatible_p (stmt, fndecl) > > > > but that is

Re: [PATCH] gimple.cc: Adjust gimple_call_builtin_p and gimple_call_combined_fn [PR105150]

2022-04-06 Thread Richard Biener via Gcc-patches
On Tue, 5 Apr 2022, Jakub Jelinek wrote: > On Tue, Apr 05, 2022 at 11:28:53AM +0200, Richard Biener wrote: > > > In GIMPLE, we call: > > > && gimple_builtin_call_types_compatible_p (stmt, fndecl) > > > but that is insufficient, that verifies whether the arguments passed to > > > GIMPLE_CALL

[PATCH] gimple.cc: Adjust gimple_call_builtin_p and gimple_call_combined_fn [PR105150]

2022-04-05 Thread Jakub Jelinek via Gcc-patches
On Tue, Apr 05, 2022 at 11:28:53AM +0200, Richard Biener wrote: > > In GIMPLE, we call: > > && gimple_builtin_call_types_compatible_p (stmt, fndecl) > > but that is insufficient, that verifies whether the arguments passed to > > GIMPLE_CALL match the fndecl argument types. But that fndecl may