Re: [PATCH] RISC-V: Only provide the single-letter extensions in HWCAP

From: Conor Dooley <conor@kernel.org>
To: Palmer Dabbelt <palmer@rivosinc.com>
Cc: linux-riscv@lists.infradead.org
Subject: Re: [PATCH] RISC-V: Only provide the single-letter extensions in HWCAP
Date: Fri, 3 Feb 2023 07:11:06 +0000	[thread overview]
Message-ID: <Y9yzigjy+Nm7nsQz@dizzy> (raw)
In-Reply-To: <20230202233832.11036-1-palmer@rivosinc.com>

[-- Attachment #1.1: Type: text/plain, Size: 2894 bytes --]

On Thu, Feb 02, 2023 at 03:38:32PM -0800, Palmer Dabbelt wrote:
> The recent refactoring led to us leaking some HWCAP bits to userspace
> that didn't make much sense.  With any luck we'll have a better scheme
> soon, but for now just mask off those bits to avoid polluting userspace.
> 
> Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
> ---
>  arch/riscv/include/asm/elf.h   | 10 ++++++----
>  arch/riscv/include/asm/hwcap.h |  8 --------
>  2 files changed, 6 insertions(+), 12 deletions(-)
> 
> diff --git a/arch/riscv/include/asm/elf.h b/arch/riscv/include/asm/elf.h
> index e7acffdf21d2..30e7d2455960 100644
> --- a/arch/riscv/include/asm/elf.h
> +++ b/arch/riscv/include/asm/elf.h
> @@ -14,6 +14,7 @@
>  #include <asm/auxvec.h>
>  #include <asm/byteorder.h>
>  #include <asm/cacheinfo.h>
> +#include <asm/hwcap.h>
>  
>  /*
>   * These are used to set parameters in the core dumps.
> @@ -59,12 +60,13 @@ extern bool compat_elf_check_arch(Elf32_Ehdr *hdr);
>  #define STACK_RND_MASK		(0x3ffff >> (PAGE_SHIFT - 12))
>  #endif
>  #endif
> +
>  /*
> - * This yields a mask that user programs can use to figure out what
> - * instruction set this CPU supports.  This could be done in user space,
> - * but it's not easy, and we've already done it here.
> + * Provides information on the availiable set of ISA extensions to userspace,
> + * via a bitmap that coorespends to each single-letter ISA extension.  This is
> + * essentially defunct, but will remain for compatibility with userspace.
>   */
> -#define ELF_HWCAP	(elf_hwcap)
> +#define ELF_HWCAP	(elf_hwcap & ((1UL << RISCV_ISA_EXT_BASE) - 1))

Should that not be a GENMASK, or is there a reason not to use that here?
Away from a setup where I can poke at it though, so, since you say this
has only been exported in for-next, preventing us running out of hwcap
sounds like a great idea.

Acked-by: Conor Dooley <conor.dooley@microchip.com>

Thanks,
Conor.

>  extern unsigned long elf_hwcap;
>  
>  /*
> diff --git a/arch/riscv/include/asm/hwcap.h b/arch/riscv/include/asm/hwcap.h
> index ee9c80fe0062..831bebacb7fb 100644
> --- a/arch/riscv/include/asm/hwcap.h
> +++ b/arch/riscv/include/asm/hwcap.h
> @@ -52,18 +52,10 @@
>  
>  #include <linux/jump_label.h>
>  
> -/*
> - * This yields a mask that user programs can use to figure out what
> - * instruction set this cpu supports.
> - */
> -#define ELF_HWCAP		(elf_hwcap)
> -
>  enum {
>  	CAP_HWCAP = 1,
>  };
>  
> -extern unsigned long elf_hwcap;
> -
>  struct riscv_isa_ext_data {
>  	/* Name of the extension displayed to userspace via /proc/cpuinfo */
>  	char uprop[RISCV_ISA_EXT_NAME_LEN_MAX];
> -- 
> 2.39.0
> 
> 
> _______________________________________________
> linux-riscv mailing list
> linux-riscv@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-riscv
> 

[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]

[-- Attachment #2: Type: text/plain, Size: 161 bytes --]

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv