On Tue, 2023-02-21 at 11:08 -0800, Evan Green wrote:
> This allows userspace to select various routines to use based on the
> performance of misaligned access on the target hardware.
[]
> diff --git a/arch/riscv/kernel/cpufeature.c b/arch/riscv/kernel/cpufeature.c
[]
> @@ -89,11 +94,11 @@ static bool riscv_isa_extension_check(int id)
> void __init riscv_fill_hwcap(void)
> {
> struct device_node *node;
> - const char *isa;
> + const char *isa, *misaligned;
> char print_str[NUM_ALPHA_EXTS + 1];
> int i, j, rc;
> unsigned long isa2hwcap[26] = {0};
> - unsigned long hartid;
> + unsigned long hartid, cpu;
>
> isa2hwcap['i' - 'a'] = COMPAT_HWCAP_ISA_I;
> isa2hwcap['m' - 'a'] = COMPAT_HWCAP_ISA_M;
> @@ -246,6 +251,28 @@ void __init riscv_fill_hwcap(void)
> bitmap_copy(riscv_isa, this_isa, RISCV_ISA_EXT_MAX);
> else
> bitmap_and(riscv_isa, riscv_isa, this_isa, RISCV_ISA_EXT_MAX);
> +
> + /*
> + * Check for the performance of misaligned accesses.
> + */
> + cpu = hartid_to_cpuid_map(hartid);
> + if (cpu < 0)
> + continue;
unsigned long can't be less than 0
Likely cpu should be long not unsigned long
It seems cpu is rather randomly either int or long.
Perhaps standardizing on int would be better.
