* [PATCH] riscv: hweight: relax assembly constraints
@ 2024-05-23 9:43 Qingfang Deng
2024-05-24 1:01 ` Wang, Xiao W
0 siblings, 1 reply; 4+ messages in thread
From: Qingfang Deng @ 2024-05-23 9:43 UTC (permalink / raw)
To: Paul Walmsley, Palmer Dabbelt, Albert Ou, linux-riscv,
linux-kernel
Cc: Xiao Wang, Qingfang Deng
From: Qingfang Deng <qingfang.deng@siflower.com.cn>
rd and rs don't have to be the same.
Signed-off-by: Qingfang Deng <qingfang.deng@siflower.com.cn>
---
arch/riscv/include/asm/arch_hweight.h | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/arch/riscv/include/asm/arch_hweight.h b/arch/riscv/include/asm/arch_hweight.h
index 85b2c443823e..613769b9cdc9 100644
--- a/arch/riscv/include/asm/arch_hweight.h
+++ b/arch/riscv/include/asm/arch_hweight.h
@@ -26,9 +26,9 @@ static __always_inline unsigned int __arch_hweight32(unsigned int w)
asm (".option push\n"
".option arch,+zbb\n"
- CPOPW "%0, %0\n"
+ CPOPW "%0, %1\n"
".option pop\n"
- : "+r" (w) : :);
+ : "=r" (w) : "r" (w) :);
return w;
@@ -57,9 +57,9 @@ static __always_inline unsigned long __arch_hweight64(__u64 w)
asm (".option push\n"
".option arch,+zbb\n"
- "cpop %0, %0\n"
+ "cpop %0, %1\n"
".option pop\n"
- : "+r" (w) : :);
+ : "=r" (w) : "r" (w) :);
return w;
--
2.34.1
^ permalink raw reply related [flat|nested] 4+ messages in thread
* RE: [PATCH] riscv: hweight: relax assembly constraints
2024-05-23 9:43 [PATCH] riscv: hweight: relax assembly constraints Qingfang Deng
@ 2024-05-24 1:01 ` Wang, Xiao W
2024-05-24 6:05 ` Qingfang Deng
0 siblings, 1 reply; 4+ messages in thread
From: Wang, Xiao W @ 2024-05-24 1:01 UTC (permalink / raw)
To: Qingfang Deng, Paul Walmsley, Palmer Dabbelt, Albert Ou,
linux-riscv@lists.infradead.org, linux-kernel@vger.kernel.org
Cc: Qingfang Deng
> -----Original Message-----
> From: Qingfang Deng <dqfext@gmail.com>
> Sent: Thursday, May 23, 2024 5:43 PM
> To: Paul Walmsley <paul.walmsley@sifive.com>; Palmer Dabbelt
> <palmer@dabbelt.com>; Albert Ou <aou@eecs.berkeley.edu>; linux-
> riscv@lists.infradead.org; linux-kernel@vger.kernel.org
> Cc: Wang, Xiao W <xiao.w.wang@intel.com>; Qingfang Deng
> <qingfang.deng@siflower.com.cn>
> Subject: [PATCH] riscv: hweight: relax assembly constraints
>
> From: Qingfang Deng <qingfang.deng@siflower.com.cn>
>
> rd and rs don't have to be the same.
>
> Signed-off-by: Qingfang Deng <qingfang.deng@siflower.com.cn>
> ---
> arch/riscv/include/asm/arch_hweight.h | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/arch/riscv/include/asm/arch_hweight.h
> b/arch/riscv/include/asm/arch_hweight.h
> index 85b2c443823e..613769b9cdc9 100644
> --- a/arch/riscv/include/asm/arch_hweight.h
> +++ b/arch/riscv/include/asm/arch_hweight.h
> @@ -26,9 +26,9 @@ static __always_inline unsigned int
> __arch_hweight32(unsigned int w)
>
> asm (".option push\n"
> ".option arch,+zbb\n"
> - CPOPW "%0, %0\n"
> + CPOPW "%0, %1\n"
> ".option pop\n"
> - : "+r" (w) : :);
> + : "=r" (w) : "r" (w) :);
The above code piece takes variable "w" as both input and output, so intuitively, the previous
patch made rs and rd the same.
Though rs and rd can be different, do you see performance difference with this change?
Or any analysis from assembly dump?
BRs,
Xiao
>
> return w;
>
> @@ -57,9 +57,9 @@ static __always_inline unsigned long
> __arch_hweight64(__u64 w)
>
> asm (".option push\n"
> ".option arch,+zbb\n"
> - "cpop %0, %0\n"
> + "cpop %0, %1\n"
> ".option pop\n"
> - : "+r" (w) : :);
> + : "=r" (w) : "r" (w) :);
>
> return w;
>
> --
> 2.34.1
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] riscv: hweight: relax assembly constraints
2024-05-24 1:01 ` Wang, Xiao W
@ 2024-05-24 6:05 ` Qingfang Deng
2024-05-24 8:07 ` Wang, Xiao W
0 siblings, 1 reply; 4+ messages in thread
From: Qingfang Deng @ 2024-05-24 6:05 UTC (permalink / raw)
To: Wang, Xiao W
Cc: Paul Walmsley, Palmer Dabbelt, Albert Ou,
linux-riscv@lists.infradead.org, linux-kernel@vger.kernel.org,
Qingfang Deng
Hi,
On Fri, May 24, 2024 at 9:02 AM Wang, Xiao W <xiao.w.wang@intel.com> wrote:
>
>
>
> > -----Original Message-----
> > From: Qingfang Deng <dqfext@gmail.com>
> > Sent: Thursday, May 23, 2024 5:43 PM
> > To: Paul Walmsley <paul.walmsley@sifive.com>; Palmer Dabbelt
> > <palmer@dabbelt.com>; Albert Ou <aou@eecs.berkeley.edu>; linux-
> > riscv@lists.infradead.org; linux-kernel@vger.kernel.org
> > Cc: Wang, Xiao W <xiao.w.wang@intel.com>; Qingfang Deng
> > <qingfang.deng@siflower.com.cn>
> > Subject: [PATCH] riscv: hweight: relax assembly constraints
> >
> > From: Qingfang Deng <qingfang.deng@siflower.com.cn>
> >
> > rd and rs don't have to be the same.
> >
> > Signed-off-by: Qingfang Deng <qingfang.deng@siflower.com.cn>
> > ---
> > arch/riscv/include/asm/arch_hweight.h | 8 ++++----
> > 1 file changed, 4 insertions(+), 4 deletions(-)
> >
> > diff --git a/arch/riscv/include/asm/arch_hweight.h
> > b/arch/riscv/include/asm/arch_hweight.h
> > index 85b2c443823e..613769b9cdc9 100644
> > --- a/arch/riscv/include/asm/arch_hweight.h
> > +++ b/arch/riscv/include/asm/arch_hweight.h
> > @@ -26,9 +26,9 @@ static __always_inline unsigned int
> > __arch_hweight32(unsigned int w)
> >
> > asm (".option push\n"
> > ".option arch,+zbb\n"
> > - CPOPW "%0, %0\n"
> > + CPOPW "%0, %1\n"
> > ".option pop\n"
> > - : "+r" (w) : :);
> > + : "=r" (w) : "r" (w) :);
>
> The above code piece takes variable "w" as both input and output, so intuitively, the previous
> patch made rs and rd the same.
> Though rs and rd can be different, do you see performance difference with this change?
> Or any analysis from assembly dump?
By making rs and rd different, we can save some `mv` instructions.
>
> BRs,
> Xiao
>
> >
> > return w;
> >
> > @@ -57,9 +57,9 @@ static __always_inline unsigned long
> > __arch_hweight64(__u64 w)
> >
> > asm (".option push\n"
> > ".option arch,+zbb\n"
> > - "cpop %0, %0\n"
> > + "cpop %0, %1\n"
> > ".option pop\n"
> > - : "+r" (w) : :);
> > + : "=r" (w) : "r" (w) :);
> >
> > return w;
> >
> > --
> > 2.34.1
>
^ permalink raw reply [flat|nested] 4+ messages in thread
* RE: [PATCH] riscv: hweight: relax assembly constraints
2024-05-24 6:05 ` Qingfang Deng
@ 2024-05-24 8:07 ` Wang, Xiao W
0 siblings, 0 replies; 4+ messages in thread
From: Wang, Xiao W @ 2024-05-24 8:07 UTC (permalink / raw)
To: Qingfang Deng
Cc: Paul Walmsley, Palmer Dabbelt, Albert Ou,
linux-riscv@lists.infradead.org, linux-kernel@vger.kernel.org,
Qingfang Deng
> -----Original Message-----
> From: Qingfang Deng <dqfext@gmail.com>
> Sent: Friday, May 24, 2024 2:05 PM
> To: Wang, Xiao W <xiao.w.wang@intel.com>
> Cc: Paul Walmsley <paul.walmsley@sifive.com>; Palmer Dabbelt
> <palmer@dabbelt.com>; Albert Ou <aou@eecs.berkeley.edu>; linux-
> riscv@lists.infradead.org; linux-kernel@vger.kernel.org; Qingfang Deng
> <qingfang.deng@siflower.com.cn>
> Subject: Re: [PATCH] riscv: hweight: relax assembly constraints
>
> Hi,
>
> On Fri, May 24, 2024 at 9:02 AM Wang, Xiao W <xiao.w.wang@intel.com>
> wrote:
> >
> >
> >
> > > -----Original Message-----
> > > From: Qingfang Deng <dqfext@gmail.com>
> > > Sent: Thursday, May 23, 2024 5:43 PM
> > > To: Paul Walmsley <paul.walmsley@sifive.com>; Palmer Dabbelt
> > > <palmer@dabbelt.com>; Albert Ou <aou@eecs.berkeley.edu>; linux-
> > > riscv@lists.infradead.org; linux-kernel@vger.kernel.org
> > > Cc: Wang, Xiao W <xiao.w.wang@intel.com>; Qingfang Deng
> > > <qingfang.deng@siflower.com.cn>
> > > Subject: [PATCH] riscv: hweight: relax assembly constraints
> > >
> > > From: Qingfang Deng <qingfang.deng@siflower.com.cn>
> > >
> > > rd and rs don't have to be the same.
> > >
> > > Signed-off-by: Qingfang Deng <qingfang.deng@siflower.com.cn>
> > > ---
> > > arch/riscv/include/asm/arch_hweight.h | 8 ++++----
> > > 1 file changed, 4 insertions(+), 4 deletions(-)
> > >
> > > diff --git a/arch/riscv/include/asm/arch_hweight.h
> > > b/arch/riscv/include/asm/arch_hweight.h
> > > index 85b2c443823e..613769b9cdc9 100644
> > > --- a/arch/riscv/include/asm/arch_hweight.h
> > > +++ b/arch/riscv/include/asm/arch_hweight.h
> > > @@ -26,9 +26,9 @@ static __always_inline unsigned int
> > > __arch_hweight32(unsigned int w)
> > >
> > > asm (".option push\n"
> > > ".option arch,+zbb\n"
> > > - CPOPW "%0, %0\n"
> > > + CPOPW "%0, %1\n"
> > > ".option pop\n"
> > > - : "+r" (w) : :);
> > > + : "=r" (w) : "r" (w) :);
> >
> > The above code piece takes variable "w" as both input and output, so
> intuitively, the previous
> > patch made rs and rd the same.
> > Though rs and rd can be different, do you see performance difference with
> this change?
> > Or any analysis from assembly dump?
>
> By making rs and rd different, we can save some `mv` instructions.
OK, I guess in some cases, the original data needs be saved for later usage.
Then, we can relax the assembly constraint here and gives flexibility to compiler
For optimization.
It's better to start the patch tile with " riscv: lib:", maybe you can make the title as:
riscv: lib: relax assembly constraints in hweight
BRs,
Xiao
>
> >
> > BRs,
> > Xiao
> >
> > >
> > > return w;
> > >
> > > @@ -57,9 +57,9 @@ static __always_inline unsigned long
> > > __arch_hweight64(__u64 w)
> > >
> > > asm (".option push\n"
> > > ".option arch,+zbb\n"
> > > - "cpop %0, %0\n"
> > > + "cpop %0, %1\n"
> > > ".option pop\n"
> > > - : "+r" (w) : :);
> > > + : "=r" (w) : "r" (w) :);
> > >
> > > return w;
> > >
> > > --
> > > 2.34.1
> >
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2024-05-24 8:07 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-05-23 9:43 [PATCH] riscv: hweight: relax assembly constraints Qingfang Deng
2024-05-24 1:01 ` Wang, Xiao W
2024-05-24 6:05 ` Qingfang Deng
2024-05-24 8:07 ` Wang, Xiao W
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox