[PATCH v4 5/5] RISC-V: add zbb support to string functions

[Andrew Jones]

On Mon, Jan 09, 2023 at 07:17:55PM +0100, Heiko Stuebner wrote:
> From: Heiko Stuebner <heiko.stuebner at vrull.eu>
> 
> Add handling for ZBB extension and add support for using it as a
> variant for optimized string functions.
> 
> Support for the Zbb-str-variants is limited to the GNU-assembler
> for now, as LLVM has not yet acquired the functionality to
> selectively change the arch option in assembler code.
> This is still under review at
>     https://reviews.llvm.org/D123515
> 
> Co-developed-by: Christoph Muellner <christoph.muellner at vrull.eu>
> Signed-off-by: Christoph Muellner <christoph.muellner at vrull.eu>
> Signed-off-by: Heiko Stuebner <heiko.stuebner at vrull.eu>
> ---
>  arch/riscv/Kconfig                   |  24 ++++++
>  arch/riscv/include/asm/errata_list.h |   3 +-
>  arch/riscv/include/asm/hwcap.h       |   1 +
>  arch/riscv/include/asm/string.h      |   2 +
>  arch/riscv/kernel/cpu.c              |   1 +
>  arch/riscv/kernel/cpufeature.c       |  18 +++++
>  arch/riscv/lib/strcmp.S              |  94 ++++++++++++++++++++++
>  arch/riscv/lib/strlen.S              | 114 +++++++++++++++++++++++++++
>  arch/riscv/lib/strncmp.S             | 111 ++++++++++++++++++++++++++
>  9 files changed, 367 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
> index e2b656043abf..7c814fbf9527 100644
> --- a/arch/riscv/Kconfig
> +++ b/arch/riscv/Kconfig
> @@ -416,6 +416,30 @@ config RISCV_ISA_SVPBMT
>  
>  	   If you don't know what to do here, say Y.
>  
> +config TOOLCHAIN_HAS_ZBB
> +	bool
> +	default y
> +	depends on !64BIT || $(cc-option,-mabi=lp64 -march=rv64ima_zbb)
> +	depends on !32BIT || $(cc-option,-mabi=ilp32 -march=rv32ima_zbb)
> +	depends on LLD_VERSION >= 150000 || LD_VERSION >= 23900
> +	depends on AS_IS_GNU
> +
> +config RISCV_ISA_ZBB
> +	bool "Zbb extension support for bit manipulation instructions"
> +	depends on TOOLCHAIN_HAS_ZBB
> +	depends on !XIP_KERNEL && MMU
> +	select RISCV_ALTERNATIVE
> +	default y
> +	help
> +	   Adds support to dynamically detect the presence of the ZBB
> +	   extension (basic bit manipulation) and enable its usage.
> +
> +	   The Zbb extension provides instructions to accelerate a number
> +	   of bit-specific operations (count bit population, sign extending,
> +	   bitrotation, etc).
> +
> +	   If you don't know what to do here, say Y.
> +
>  config TOOLCHAIN_HAS_ZICBOM
>  	bool
>  	default y
> diff --git a/arch/riscv/include/asm/errata_list.h b/arch/riscv/include/asm/errata_list.h
> index 4180312d2a70..95e626b7281e 100644
> --- a/arch/riscv/include/asm/errata_list.h
> +++ b/arch/riscv/include/asm/errata_list.h
> @@ -24,7 +24,8 @@
>  
>  #define	CPUFEATURE_SVPBMT 0
>  #define	CPUFEATURE_ZICBOM 1
> -#define	CPUFEATURE_NUMBER 2
> +#define	CPUFEATURE_ZBB 2
> +#define	CPUFEATURE_NUMBER 3
>  
>  #ifdef __ASSEMBLY__
>  
> diff --git a/arch/riscv/include/asm/hwcap.h b/arch/riscv/include/asm/hwcap.h
> index 86328e3acb02..b727491fb100 100644
> --- a/arch/riscv/include/asm/hwcap.h
> +++ b/arch/riscv/include/asm/hwcap.h
> @@ -59,6 +59,7 @@ enum riscv_isa_ext_id {
>  	RISCV_ISA_EXT_ZIHINTPAUSE,
>  	RISCV_ISA_EXT_SSTC,
>  	RISCV_ISA_EXT_SVINVAL,
> +	RISCV_ISA_EXT_ZBB,
>  	RISCV_ISA_EXT_ID_MAX
>  };
>  static_assert(RISCV_ISA_EXT_ID_MAX <= RISCV_ISA_EXT_MAX);
> diff --git a/arch/riscv/include/asm/string.h b/arch/riscv/include/asm/string.h
> index a96b1fea24fe..17dfc4ab4c80 100644
> --- a/arch/riscv/include/asm/string.h
> +++ b/arch/riscv/include/asm/string.h
> @@ -6,6 +6,8 @@
>  #ifndef _ASM_RISCV_STRING_H
>  #define _ASM_RISCV_STRING_H
>  
> +#include <asm/alternative-macros.h>
> +#include <asm/errata_list.h>

Why are these includes getting added? Shouldn't they be getting directly
added to whatever is including asm/string.h instead?

>  #include <linux/types.h>
>  #include <linux/linkage.h>
>  
> diff --git a/arch/riscv/kernel/cpu.c b/arch/riscv/kernel/cpu.c
> index 1b9a5a66e55a..c4d1aa166f8b 100644
> --- a/arch/riscv/kernel/cpu.c
> +++ b/arch/riscv/kernel/cpu.c
> @@ -162,6 +162,7 @@ arch_initcall(riscv_cpuinfo_init);
>   *    extensions by an underscore.
>   */
>  static struct riscv_isa_ext_data isa_ext_arr[] = {
> +	__RISCV_ISA_EXT_DATA(zbb, RISCV_ISA_EXT_ZBB),
>  	__RISCV_ISA_EXT_DATA(sscofpmf, RISCV_ISA_EXT_SSCOFPMF),
>  	__RISCV_ISA_EXT_DATA(sstc, RISCV_ISA_EXT_SSTC),
>  	__RISCV_ISA_EXT_DATA(svinval, RISCV_ISA_EXT_SVINVAL),

Huh, this array still doesn't appear to be in order... Zbb should
be getting inserted between the Zi* extensions (which should be first)
and the S* extensions and each of those three categories should be
in alphabetical order.

> diff --git a/arch/riscv/kernel/cpufeature.c b/arch/riscv/kernel/cpufeature.c
> index 205bbd6b1fce..bf3a791d7110 100644
> --- a/arch/riscv/kernel/cpufeature.c
> +++ b/arch/riscv/kernel/cpufeature.c
> @@ -222,6 +222,7 @@ void __init riscv_fill_hwcap(void)
>  					set_bit(nr, this_isa);
>  				}
>  			} else {
> +				SET_ISA_EXT_MAP("zbb", RISCV_ISA_EXT_ZBB);
>  				SET_ISA_EXT_MAP("sscofpmf", RISCV_ISA_EXT_SSCOFPMF);
>  				SET_ISA_EXT_MAP("svpbmt", RISCV_ISA_EXT_SVPBMT);
>  				SET_ISA_EXT_MAP("zicbom", RISCV_ISA_EXT_ZICBOM);

I think we wanted this one in alphabetical order...

> @@ -301,6 +302,20 @@ static bool __init_or_module cpufeature_probe_zicbom(unsigned int stage)
>  	return true;
>  }
>  
> +static bool __init_or_module cpufeature_probe_zbb(unsigned int stage)
> +{
> +	if (!IS_ENABLED(CONFIG_RISCV_ISA_ZBB))
> +		return false;
> +
> +	if (stage == RISCV_ALTERNATIVES_EARLY_BOOT)
> +		return false;
> +
> +	if (!riscv_isa_extension_available(NULL, ZBB))
> +		return false;
> +
> +	return true;
> +}
> +
>  /*
>   * Probe presence of individual extensions.
>   *
> @@ -318,6 +333,9 @@ static u32 __init_or_module cpufeature_probe(unsigned int stage)
>  	if (cpufeature_probe_zicbom(stage))
>  		cpu_req_feature |= BIT(CPUFEATURE_ZICBOM);
>  
> +	if (cpufeature_probe_zbb(stage))
> +		cpu_req_feature |= BIT(CPUFEATURE_ZBB);
> +
>  	return cpu_req_feature;
>  }
>

I'm skipping the asm review for now since I'm still not clear on the plan
for non-optimized string functions vs. optimized vs. compiler builtins.

Thanks,
drew