* [PATCH] riscv: Add support for XTheadBb in string-fz[a,i].h
@ 2023-08-23 5:46 Christoph Muellner
2023-08-24 17:21 ` Adhemerval Zanella Netto
0 siblings, 1 reply; 6+ messages in thread
From: Christoph Muellner @ 2023-08-23 5:46 UTC (permalink / raw)
To: libc-alpha, Palmer Dabbelt, Darius Rad, Andrew Waterman, Philipp Tomsich
Cc: Christoph Müllner
From: Christoph Müllner <christoph.muellner@vrull.eu>
XTheadBb has similar instructions like Zbb, which allow optimized
string processing:
* th.ff0: find-first zero is a CLZ instruction.
* th.tstnbz: Similar like orc.b, but with a bit-inverted result.
The instructions are documented here:
https://github.com/T-head-Semi/thead-extension-spec/tree/master/xtheadbb
These instructions can be found in the T-Head C906 and the C910.
Tested with the string tests.
Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
---
sysdeps/riscv/string-fza.h | 7 ++++++-
sysdeps/riscv/string-fzi.h | 2 +-
2 files changed, 7 insertions(+), 2 deletions(-)
diff --git a/sysdeps/riscv/string-fza.h b/sysdeps/riscv/string-fza.h
index 4429653a00..4958d5d151 100644
--- a/sysdeps/riscv/string-fza.h
+++ b/sysdeps/riscv/string-fza.h
@@ -19,7 +19,7 @@
#ifndef _RISCV_STRING_FZA_H
#define _RISCV_STRING_FZA_H 1
-#ifdef __riscv_zbb
+#if defined __riscv_zbb || defined __riscv_xtheadbb
/* With bitmap extension we can use orc.b to find all zero bytes. */
# include <string-misc.h>
# include <string-optype.h>
@@ -32,8 +32,13 @@ static __always_inline find_t
find_zero_all (op_t x)
{
find_t r;
+#ifdef __riscv_xtheadbb
+ asm ("th.tstnbz %0, %1" : "=r" (r) : "r" (x));
+ return r;
+#else
asm ("orc.b %0, %1" : "=r" (r) : "r" (x));
return ~r;
+#endif
}
/* This function returns 0xff for each byte that is equal between X1 and
diff --git a/sysdeps/riscv/string-fzi.h b/sysdeps/riscv/string-fzi.h
index 8f56c378ff..45d6367a10 100644
--- a/sysdeps/riscv/string-fzi.h
+++ b/sysdeps/riscv/string-fzi.h
@@ -19,7 +19,7 @@
#ifndef _STRING_RISCV_FZI_H
#define _STRING_RISCV_FZI_H 1
-#ifdef __riscv_zbb
+#if defined __riscv_zbb || defined __riscv_xtheadbb
# include <sysdeps/generic/string-fzi.h>
#else
/* Without bitmap clz/ctz extensions, it is faster to direct test the bits
--
2.41.0
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] riscv: Add support for XTheadBb in string-fz[a,i].h
2023-08-23 5:46 [PATCH] riscv: Add support for XTheadBb in string-fz[a,i].h Christoph Muellner
@ 2023-08-24 17:21 ` Adhemerval Zanella Netto
2023-08-24 17:36 ` Christoph Müllner
2023-09-06 10:34 ` Christoph Müllner
0 siblings, 2 replies; 6+ messages in thread
From: Adhemerval Zanella Netto @ 2023-08-24 17:21 UTC (permalink / raw)
To: Christoph Muellner, libc-alpha, Palmer Dabbelt, Darius Rad,
Andrew Waterman, Philipp Tomsich
On 23/08/23 02:46, Christoph Muellner wrote:
> From: Christoph Müllner <christoph.muellner@vrull.eu>
>
> XTheadBb has similar instructions like Zbb, which allow optimized
> string processing:
> * th.ff0: find-first zero is a CLZ instruction.
> * th.tstnbz: Similar like orc.b, but with a bit-inverted result.
>
> The instructions are documented here:
> https://github.com/T-head-Semi/thead-extension-spec/tree/master/xtheadbb
>
> These instructions can be found in the T-Head C906 and the C910.
>
> Tested with the string tests.
>
> Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
LGTM, thanks. Is th.tstnbz available as builtin by chance?
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
> ---
> sysdeps/riscv/string-fza.h | 7 ++++++-
> sysdeps/riscv/string-fzi.h | 2 +-
> 2 files changed, 7 insertions(+), 2 deletions(-)
>
> diff --git a/sysdeps/riscv/string-fza.h b/sysdeps/riscv/string-fza.h
> index 4429653a00..4958d5d151 100644
> --- a/sysdeps/riscv/string-fza.h
> +++ b/sysdeps/riscv/string-fza.h
> @@ -19,7 +19,7 @@
> #ifndef _RISCV_STRING_FZA_H
> #define _RISCV_STRING_FZA_H 1
>
> -#ifdef __riscv_zbb
> +#if defined __riscv_zbb || defined __riscv_xtheadbb
> /* With bitmap extension we can use orc.b to find all zero bytes. */
> # include <string-misc.h>
> # include <string-optype.h>
> @@ -32,8 +32,13 @@ static __always_inline find_t
> find_zero_all (op_t x)
> {
> find_t r;
> +#ifdef __riscv_xtheadbb
> + asm ("th.tstnbz %0, %1" : "=r" (r) : "r" (x));
> + return r;
> +#else
> asm ("orc.b %0, %1" : "=r" (r) : "r" (x));
> return ~r;
> +#endif
> }
>
> /* This function returns 0xff for each byte that is equal between X1 and
> diff --git a/sysdeps/riscv/string-fzi.h b/sysdeps/riscv/string-fzi.h
> index 8f56c378ff..45d6367a10 100644
> --- a/sysdeps/riscv/string-fzi.h
> +++ b/sysdeps/riscv/string-fzi.h
> @@ -19,7 +19,7 @@
> #ifndef _STRING_RISCV_FZI_H
> #define _STRING_RISCV_FZI_H 1
>
> -#ifdef __riscv_zbb
> +#if defined __riscv_zbb || defined __riscv_xtheadbb
> # include <sysdeps/generic/string-fzi.h>
> #else
> /* Without bitmap clz/ctz extensions, it is faster to direct test the bits
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] riscv: Add support for XTheadBb in string-fz[a,i].h
2023-08-24 17:21 ` Adhemerval Zanella Netto
@ 2023-08-24 17:36 ` Christoph Müllner
2023-09-06 10:34 ` Christoph Müllner
1 sibling, 0 replies; 6+ messages in thread
From: Christoph Müllner @ 2023-08-24 17:36 UTC (permalink / raw)
To: Adhemerval Zanella Netto
Cc: libc-alpha, Palmer Dabbelt, Darius Rad, Andrew Waterman, Philipp Tomsich
[-- Attachment #1: Type: text/plain, Size: 2552 bytes --]
On Thu, Aug 24, 2023 at 7:21 PM Adhemerval Zanella Netto <
adhemerval.zanella@linaro.org> wrote:
>
>
> On 23/08/23 02:46, Christoph Muellner wrote:
> > From: Christoph Müllner <christoph.muellner@vrull.eu>
> >
> > XTheadBb has similar instructions like Zbb, which allow optimized
> > string processing:
> > * th.ff0: find-first zero is a CLZ instruction.
> > * th.tstnbz: Similar like orc.b, but with a bit-inverted result.
> >
> > The instructions are documented here:
> >
> https://github.com/T-head-Semi/thead-extension-spec/tree/master/xtheadbb
> >
> > These instructions can be found in the T-Head C906 and the C910.
> >
> > Tested with the string tests.
> >
> > Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
>
>
> LGTM, thanks. Is th.tstnbz available as builtin by chance?
>
No, same situation like orc.b.
Thanks,
Christoph
>
> Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
>
>
> > ---
> > sysdeps/riscv/string-fza.h | 7 ++++++-
> > sysdeps/riscv/string-fzi.h | 2 +-
> > 2 files changed, 7 insertions(+), 2 deletions(-)
> >
> > diff --git a/sysdeps/riscv/string-fza.h b/sysdeps/riscv/string-fza.h
> > index 4429653a00..4958d5d151 100644
> > --- a/sysdeps/riscv/string-fza.h
> > +++ b/sysdeps/riscv/string-fza.h
> > @@ -19,7 +19,7 @@
> > #ifndef _RISCV_STRING_FZA_H
> > #define _RISCV_STRING_FZA_H 1
> >
> > -#ifdef __riscv_zbb
> > +#if defined __riscv_zbb || defined __riscv_xtheadbb
> > /* With bitmap extension we can use orc.b to find all zero bytes. */
> > # include <string-misc.h>
> > # include <string-optype.h>
> > @@ -32,8 +32,13 @@ static __always_inline find_t
> > find_zero_all (op_t x)
> > {
> > find_t r;
> > +#ifdef __riscv_xtheadbb
> > + asm ("th.tstnbz %0, %1" : "=r" (r) : "r" (x));
> > + return r;
> > +#else
> > asm ("orc.b %0, %1" : "=r" (r) : "r" (x));
> > return ~r;
> > +#endif
> > }
> >
> > /* This function returns 0xff for each byte that is equal between X1 and
> > diff --git a/sysdeps/riscv/string-fzi.h b/sysdeps/riscv/string-fzi.h
> > index 8f56c378ff..45d6367a10 100644
> > --- a/sysdeps/riscv/string-fzi.h
> > +++ b/sysdeps/riscv/string-fzi.h
> > @@ -19,7 +19,7 @@
> > #ifndef _STRING_RISCV_FZI_H
> > #define _STRING_RISCV_FZI_H 1
> >
> > -#ifdef __riscv_zbb
> > +#if defined __riscv_zbb || defined __riscv_xtheadbb
> > # include <sysdeps/generic/string-fzi.h>
> > #else
> > /* Without bitmap clz/ctz extensions, it is faster to direct test the
> bits
>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] riscv: Add support for XTheadBb in string-fz[a,i].h
2023-08-24 17:21 ` Adhemerval Zanella Netto
2023-08-24 17:36 ` Christoph Müllner
@ 2023-09-06 10:34 ` Christoph Müllner
2023-09-06 11:39 ` Adhemerval Zanella Netto
1 sibling, 1 reply; 6+ messages in thread
From: Christoph Müllner @ 2023-09-06 10:34 UTC (permalink / raw)
To: Adhemerval Zanella Netto
Cc: libc-alpha, Palmer Dabbelt, Darius Rad, Andrew Waterman, Philipp Tomsich
[-- Attachment #1: Type: text/plain, Size: 2568 bytes --]
On Thu, Aug 24, 2023 at 7:21 PM Adhemerval Zanella Netto <
adhemerval.zanella@linaro.org> wrote:
>
>
> On 23/08/23 02:46, Christoph Muellner wrote:
> > From: Christoph Müllner <christoph.muellner@vrull.eu>
> >
> > XTheadBb has similar instructions like Zbb, which allow optimized
> > string processing:
> > * th.ff0: find-first zero is a CLZ instruction.
> > * th.tstnbz: Similar like orc.b, but with a bit-inverted result.
> >
> > The instructions are documented here:
> >
> https://github.com/T-head-Semi/thead-extension-spec/tree/master/xtheadbb
> >
> > These instructions can be found in the T-Head C906 and the C910.
> >
> > Tested with the string tests.
> >
> > Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
>
>
> LGTM, thanks. Is th.tstnbz available as builtin by chance?
>
> Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
>
Is anything blocking this from getting merged?
Thanks,
Christoph
>
>
> > ---
> > sysdeps/riscv/string-fza.h | 7 ++++++-
> > sysdeps/riscv/string-fzi.h | 2 +-
> > 2 files changed, 7 insertions(+), 2 deletions(-)
> >
> > diff --git a/sysdeps/riscv/string-fza.h b/sysdeps/riscv/string-fza.h
> > index 4429653a00..4958d5d151 100644
> > --- a/sysdeps/riscv/string-fza.h
> > +++ b/sysdeps/riscv/string-fza.h
> > @@ -19,7 +19,7 @@
> > #ifndef _RISCV_STRING_FZA_H
> > #define _RISCV_STRING_FZA_H 1
> >
> > -#ifdef __riscv_zbb
> > +#if defined __riscv_zbb || defined __riscv_xtheadbb
> > /* With bitmap extension we can use orc.b to find all zero bytes. */
> > # include <string-misc.h>
> > # include <string-optype.h>
> > @@ -32,8 +32,13 @@ static __always_inline find_t
> > find_zero_all (op_t x)
> > {
> > find_t r;
> > +#ifdef __riscv_xtheadbb
> > + asm ("th.tstnbz %0, %1" : "=r" (r) : "r" (x));
> > + return r;
> > +#else
> > asm ("orc.b %0, %1" : "=r" (r) : "r" (x));
> > return ~r;
> > +#endif
> > }
> >
> > /* This function returns 0xff for each byte that is equal between X1 and
> > diff --git a/sysdeps/riscv/string-fzi.h b/sysdeps/riscv/string-fzi.h
> > index 8f56c378ff..45d6367a10 100644
> > --- a/sysdeps/riscv/string-fzi.h
> > +++ b/sysdeps/riscv/string-fzi.h
> > @@ -19,7 +19,7 @@
> > #ifndef _STRING_RISCV_FZI_H
> > #define _STRING_RISCV_FZI_H 1
> >
> > -#ifdef __riscv_zbb
> > +#if defined __riscv_zbb || defined __riscv_xtheadbb
> > # include <sysdeps/generic/string-fzi.h>
> > #else
> > /* Without bitmap clz/ctz extensions, it is faster to direct test the
> bits
>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] riscv: Add support for XTheadBb in string-fz[a,i].h
2023-09-06 10:34 ` Christoph Müllner
@ 2023-09-06 11:39 ` Adhemerval Zanella Netto
2023-09-06 13:52 ` Christoph Müllner
0 siblings, 1 reply; 6+ messages in thread
From: Adhemerval Zanella Netto @ 2023-09-06 11:39 UTC (permalink / raw)
To: Christoph Müllner
Cc: libc-alpha, Palmer Dabbelt, Darius Rad, Andrew Waterman, Philipp Tomsich
On 06/09/23 07:34, Christoph Müllner wrote:
>
>
> On Thu, Aug 24, 2023 at 7:21 PM Adhemerval Zanella Netto <adhemerval.zanella@linaro.org <mailto:adhemerval.zanella@linaro.org>> wrote:
>
>
>
> On 23/08/23 02:46, Christoph Muellner wrote:
> > From: Christoph Müllner <christoph.muellner@vrull.eu <mailto:christoph.muellner@vrull.eu>>
> >
> > XTheadBb has similar instructions like Zbb, which allow optimized
> > string processing:
> > * th.ff0: find-first zero is a CLZ instruction.
> > * th.tstnbz: Similar like orc.b, but with a bit-inverted result.
> >
> > The instructions are documented here:
> > https://github.com/T-head-Semi/thead-extension-spec/tree/master/xtheadbb <https://github.com/T-head-Semi/thead-extension-spec/tree/master/xtheadbb>
> >
> > These instructions can be found in the T-Head C906 and the C910.
> >
> > Tested with the string tests.
> >
> > Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu <mailto:christoph.muellner@vrull.eu>>
>
>
> LGTM, thanks. Is th.tstnbz available as builtin by chance?
>
> Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org <mailto:adhemerval.zanella@linaro.org>>
>
>
Someone to actually push it, since it is already reviewed. I will install it.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] riscv: Add support for XTheadBb in string-fz[a,i].h
2023-09-06 11:39 ` Adhemerval Zanella Netto
@ 2023-09-06 13:52 ` Christoph Müllner
0 siblings, 0 replies; 6+ messages in thread
From: Christoph Müllner @ 2023-09-06 13:52 UTC (permalink / raw)
To: Adhemerval Zanella Netto
Cc: libc-alpha, Palmer Dabbelt, Darius Rad, Andrew Waterman, Philipp Tomsich
[-- Attachment #1: Type: text/plain, Size: 1546 bytes --]
On Wed, Sep 6, 2023 at 1:39 PM Adhemerval Zanella Netto <
adhemerval.zanella@linaro.org> wrote:
>
>
> On 06/09/23 07:34, Christoph Müllner wrote:
> >
> >
> > On Thu, Aug 24, 2023 at 7:21 PM Adhemerval Zanella Netto <
> adhemerval.zanella@linaro.org <mailto:adhemerval.zanella@linaro.org>>
> wrote:
> >
> >
> >
> > On 23/08/23 02:46, Christoph Muellner wrote:
> > > From: Christoph Müllner <christoph.muellner@vrull.eu <mailto:
> christoph.muellner@vrull.eu>>
> > >
> > > XTheadBb has similar instructions like Zbb, which allow optimized
> > > string processing:
> > > * th.ff0: find-first zero is a CLZ instruction.
> > > * th.tstnbz: Similar like orc.b, but with a bit-inverted result.
> > >
> > > The instructions are documented here:
> > >
> https://github.com/T-head-Semi/thead-extension-spec/tree/master/xtheadbb <
> https://github.com/T-head-Semi/thead-extension-spec/tree/master/xtheadbb>
> > >
> > > These instructions can be found in the T-Head C906 and the C910.
> > >
> > > Tested with the string tests.
> > >
> > > Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu
> <mailto:christoph.muellner@vrull.eu>>
> >
> >
> > LGTM, thanks. Is th.tstnbz available as builtin by chance?
> >
> > Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org
> <mailto:adhemerval.zanella@linaro.org>>
> >
> >
>
> Someone to actually push it, since it is already reviewed. I will install
> it.
>
Thanks!
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2023-09-06 13:53 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-08-23 5:46 [PATCH] riscv: Add support for XTheadBb in string-fz[a,i].h Christoph Muellner
2023-08-24 17:21 ` Adhemerval Zanella Netto
2023-08-24 17:36 ` Christoph Müllner
2023-09-06 10:34 ` Christoph Müllner
2023-09-06 11:39 ` Adhemerval Zanella Netto
2023-09-06 13:52 ` Christoph Müllner
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).