perand 5) are ignored.
> +
> @cindex @code{vec_set@var{m}} instruction pattern
> @item @samp{vec_set@var{m}}
> Set given field in the vector value. Operand 0 is the vector to modify,
>
> Thanks.
>
>
>
> juzhe.zh...@rivai.ai
>
> From: Richard Biener
> Date: 2023
> > mask_len_strided_load/mask_len_strided_store OPTABS/IFN
> > On Thu, 2 Nov 2023, ??? wrote:
> >
> > > Ok. So drop 'scale' and keep signed/unsigned argument, is that right?
> >
> > I don't think we need signed/unsigned. RTL expansion has the signedness
> > of the
. Operand 0 is the vector to modify,
Thanks.
juzhe.zh...@rivai.ai
From: Richard Biener
Date: 2023-11-03 15:40
To: juzhe.zh...@rivai.ai
CC: gcc-patches; jeffreyalaw; richard.sandiford; Robin Dapp
Subject: Re: Re: [PATCH V2] OPTABS/IFN: Add
mask_len_strided_load/mask_len_strided_store OPTABS/I
elements, I
am wondering whether it will cause issues.
juzhe.zh...@rivai.ai
From: Richard Biener
Date: 2023-11-03 15:40
To: juzhe.zh...@rivai.ai
CC: gcc-patches; jeffreyalaw; richard.sandiford; Robin Dapp
Subject: Re: Re: [PATCH V2] OPTABS/IFN: Add
mask_len_strided_load/mask_len_strided_stor
iate mode
> to offset a pointer.
>
> > And I wonder I should create the stride_type using size_type_node or
> > ptrdiff_type_node ?
> > Which is preferrable ?
>
> 'sizetype' - that's the type we require to be used for
> the POINTER_PLUS_EXPR offset operand.
>
Does it look reasonable ? Thanks.
juzhe.zh...@rivai.ai
From: Richard Biener
Date: 2023-11-02 22:34
To: 钟居哲
CC: gcc-patches; Jeff Law; richard.sandiford; rdapp.gcc
Subject: Re: Re: [PATCH V2] OPTABS/IFN: Add
mask_len_strided_load/mask_len_strided_store OPTABS/IFN
On Thu, 2 Nov 2023, ??? wrote:
&
gt; To: Juzhe-Zhong
> CC: gcc-patches; jeffreyalaw; richard.sandiford; rdapp.gcc
> Subject: Re: [PATCH V2] OPTABS/IFN: Add
> mask_len_strided_load/mask_len_strided_store OPTABS/IFN
> On Tue, 31 Oct 2023, Juzhe-Zhong wrote:
>
> > As previous Richard's suggested, we should su
ate: 2023-11-02 22:27
> To: ???
> CC: gcc-patches; Jeff Law; richard.sandiford; rdapp.gcc
> Subject: Re: Re: [PATCH V2] OPTABS/IFN: Add
> mask_len_strided_load/mask_len_strided_store OPTABS/IFN
> On Thu, 2 Nov 2023, ??? wrote:
>
> > Hi, Richi.
> >
> >
21:52
To: Juzhe-Zhong
CC: gcc-patches; jeffreyalaw; richard.sandiford; rdapp.gcc
Subject: Re: [PATCH V2] OPTABS/IFN: Add
mask_len_strided_load/mask_len_strided_store OPTABS/IFN
On Tue, 31 Oct 2023, Juzhe-Zhong wrote:
> As previous Richard's suggested, we should support strided load/store in
; richard.sandiford; rdapp.gcc
Subject: Re: Re: [PATCH V2] OPTABS/IFN: Add
mask_len_strided_load/mask_len_strided_store OPTABS/IFN
On Thu, 2 Nov 2023, ??? wrote:
> Hi, Richi.
>
> >> Do we really need to have two modes for the optab though or could we
> >> simply requi
On Tue, 31 Oct 2023, Juzhe-Zhong wrote:
> As previous Richard's suggested, we should support strided load/store in
> loop vectorizer instead hacking RISC-V backend.
>
> This patch adds MASK_LEN_STRIDED LOAD/STORE OPTABS/IFN.
>
> The GIMPLE IR is same as
Passed the x86 bootstrap and regression tests.
Pan
-Original Message-
From: Juzhe-Zhong
Sent: Tuesday, October 31, 2023 5:59 PM
To: gcc-patches@gcc.gnu.org
Cc: rguent...@suse.de; jeffreya...@gmail.com; richard.sandif...@arm.com;
rdapp@gmail.com; Juzhe-Zhong
Subject: [PATCH V2]
12 matches
Mail list logo