Re: [PATCH 01/13] [APX EGPR] middle-end: Add insn argument to base_reg_class

2023-10-07 Thread Hongyu Wang
> It would be nice to add to the documentation that INSN_BASE_REG_CLASS, > INSN_INDEX_REG_CLASS, and REGNO_OK_FOR_INSN_BASE_P if defined have > priority over older corresponding macros as it is already documented for > REGNO_MODE_CODE_OK_FOR_BASE_P relating to REGNO_OK_FOR_BASE_P. But this > small

Re: [PATCH 01/13] [APX EGPR] middle-end: Add insn argument to base_reg_class

2023-09-22 Thread Vladimir Makarov
On 9/22/23 06:56, Hongyu Wang wrote: From: Kong Lingling Current reload infrastructure does not support selective base_reg_class for backend insn. Add new macros with insn parameters to base_reg_class for lra/reload usage. gcc/ChangeLog: * addresses.h (base_reg_class): Add insn argu

Re: [PATCH 01/13] [APX EGPR] middle-end: Add insn argument to base_reg_class

2023-09-14 Thread Vladimir Makarov via Gcc-patches
On 9/10/23 00:49, Hongyu Wang wrote: Vladimir Makarov via Gcc-patches 于2023年9月9日周六 01:04写道: On 8/31/23 04:20, Hongyu Wang wrote: @@ -2542,6 +2542,8 @@ the code of the immediately enclosing expression (@code{MEM} for the top level of an address, @code{ADDRESS} for something that occurs i

Re: [PATCH 01/13] [APX EGPR] middle-end: Add insn argument to base_reg_class

2023-09-09 Thread Hongyu Wang via Gcc-patches
Vladimir Makarov via Gcc-patches 于2023年9月9日周六 01:04写道: > > > On 8/31/23 04:20, Hongyu Wang wrote: > > @@ -2542,6 +2542,8 @@ the code of the immediately enclosing expression > > (@code{MEM} for the top level > > of an address, @code{ADDRESS} for something that occurs in an > > @code{address_op

Re: [PATCH 01/13] [APX EGPR] middle-end: Add insn argument to base_reg_class

2023-09-08 Thread Vladimir Makarov via Gcc-patches
On 8/31/23 04:20, Hongyu Wang wrote: @@ -2542,6 +2542,8 @@ the code of the immediately enclosing expression (@code{MEM} for the top level of an address, @code{ADDRESS} for something that occurs in an @code{address_operand}). @var{index_code} is the code of the corresponding index expres

Re: [PATCH 01/13] [APX EGPR] middle-end: Add insn argument to base_reg_class

2023-09-07 Thread Vladimir Makarov via Gcc-patches
On 9/7/23 02:23, Uros Bizjak wrote: On Wed, Sep 6, 2023 at 9:43 PM Vladimir Makarov wrote: On 9/1/23 05:07, Hongyu Wang wrote: I think the approach proposed by Intel developers is better. In some way we already use such approach when we pass memory mode to get the base reg class. Althou

Re: [PATCH 01/13] [APX EGPR] middle-end: Add insn argument to base_reg_class

2023-09-06 Thread Uros Bizjak via Gcc-patches
On Wed, Sep 6, 2023 at 9:43 PM Vladimir Makarov wrote: > > > On 9/1/23 05:07, Hongyu Wang wrote: > > Uros Bizjak via Gcc-patches 于2023年8月31日周四 18:16写道: > >> On Thu, Aug 31, 2023 at 10:20 AM Hongyu Wang wrote: > >>> From: Kong Lingling > >>> > >>> Current reload infrastructure does not support s

Re: [PATCH 01/13] [APX EGPR] middle-end: Add insn argument to base_reg_class

2023-09-06 Thread Vladimir Makarov via Gcc-patches
On 9/1/23 05:07, Hongyu Wang wrote: Uros Bizjak via Gcc-patches 于2023年8月31日周四 18:16写道: On Thu, Aug 31, 2023 at 10:20 AM Hongyu Wang wrote: From: Kong Lingling Current reload infrastructure does not support selective base_reg_class for backend insn. Add insn argument to base_reg_class for

Re: [PATCH 01/13] [APX EGPR] middle-end: Add insn argument to base_reg_class

2023-09-01 Thread Hongyu Wang via Gcc-patches
Uros Bizjak via Gcc-patches 于2023年8月31日周四 18:16写道: > > On Thu, Aug 31, 2023 at 10:20 AM Hongyu Wang wrote: > > > > From: Kong Lingling > > > > Current reload infrastructure does not support selective base_reg_class > > for backend insn. Add insn argument to base_reg_class for > > lra/reload usag

Re: [PATCH 01/13] [APX EGPR] middle-end: Add insn argument to base_reg_class

2023-08-31 Thread Uros Bizjak via Gcc-patches
On Thu, Aug 31, 2023 at 10:20 AM Hongyu Wang wrote: > > From: Kong Lingling > > Current reload infrastructure does not support selective base_reg_class > for backend insn. Add insn argument to base_reg_class for > lra/reload usage. I don't think this is the correct approach. Ideally, a memory co