[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <e015a144-fd3f-4a71-97bf-64c784406b21@ghiti.fr>
Date: Mon, 10 Mar 2025 10:40:16 +0100
From: Alexandre Ghiti <alex@...ti.fr>
To: Charlie Jenkins <charlie@...osinc.com>,
Paul Walmsley <paul.walmsley@...ive.com>, Palmer Dabbelt
<palmer@...belt.com>, Ard Biesheuvel <ardb@...nel.org>,
Ben Dooks <ben.dooks@...ethink.co.uk>,
Pasha Bouzarjomehri <pasha@...osinc.com>,
Emil Renner Berthing <emil.renner.berthing@...onical.com>
Cc: linux-riscv@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v8] riscv: Add runtime constant support
Hi Charlie,
On 05/03/2025 23:53, Charlie Jenkins wrote:
> Implement the runtime constant infrastructure for riscv. Use this
> infrastructure to generate constants to be used by the d_hash()
> function.
>
> This is the riscv variant of commit 94a2bc0f611c ("arm64: add 'runtime
> constant' support") and commit e3c92e81711d ("runtime constants: add
> x86 architecture support").
>
> Signed-off-by: Charlie Jenkins <charlie@...osinc.com>
> ---
> Ard brought this to my attention in this patch [1].
>
> I benchmarked this patch on the Nezha D1 (which does not contain Zba or
> Zbkb so it uses the default algorithm) by navigating through a large
> directory structure. I created a 1000-deep directory structure and then
> cd and ls through it. With this patch there was a 0.57% performance
> improvement.
>
> [1] https://lore.kernel.org/lkml/CAMj1kXE4DJnwFejNWQu784GvyJO=aGNrzuLjSxiowX_e7nW8QA@mail.gmail.com/
> ---
> Changes in v8:
> - EDITME: describe what is new in this series revision.
> - EDITME: use bulletpoints and terse descriptions.
> - Link to v7: https://lore.kernel.org/r/20250218-runtime_const_riscv-v7-1-e431763157ff@rivosinc.com
>
> Changes in v7:
> - Added benchmarking info
> - Added CONFIG_RISCV_ISA_ZBA and CONFIG_RISCV_ISA_ZBKB to check that the
> compiler supports the extensions.
> - Link to v6: https://lore.kernel.org/r/20250212-runtime_const_riscv-v6-1-3ef0146b310b@rivosinc.com
>
> Changes in v6:
> - .option arch only became officially supported by clang in version 17.
> Add a config to check that and guard the alternatives uses .option
> arch.
> - Link to v5: https://lore.kernel.org/r/20250203-runtime_const_riscv-v5-1-bc61736a3229@rivosinc.com
>
> Changes in v5:
> - Split instructions into 16-bit parcels to avoid alignment (Emil)
> - Link to v4: https://lore.kernel.org/r/20250130-runtime_const_riscv-v4-1-2d36c41b7b9c@rivosinc.com
>
> Changes in v4:
> - Add newlines after riscv32 assembler directives
> - Align instructions along 32-bit boundary (Emil)
> - Link to v3: https://lore.kernel.org/r/20250128-runtime_const_riscv-v3-1-11922989e2d3@rivosinc.com
>
> Changes in v3:
> - Leverage "pack" instruction for runtime_const_ptr() to reduce hot path
> by 3 instructions if Zbkb is supported. Suggested by Pasha Bouzarjomehri (pasha@...osinc.com)
> - Link to v2: https://lore.kernel.org/r/20250127-runtime_const_riscv-v2-1-95ae7cf97a39@rivosinc.com
>
> Changes in v2:
> - Treat instructions as __le32 and do proper conversions (Ben)
> - Link to v1: https://lore.kernel.org/r/20250127-runtime_const_riscv-v1-1-795b023ea20b@rivosinc.com
> ---
> arch/riscv/Kconfig | 22 +++
> arch/riscv/include/asm/runtime-const.h | 254 +++++++++++++++++++++++++++++++++
> arch/riscv/kernel/vmlinux.lds.S | 3 +
> 3 files changed, 279 insertions(+)
>
> diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
> index 7612c52e9b1e35607f1dd4603a596416d3357a71..c123f7c0579c1aca839e3c04bdb662d6856ae765 100644
> --- a/arch/riscv/Kconfig
> +++ b/arch/riscv/Kconfig
> @@ -783,6 +783,28 @@ config RISCV_ISA_ZBC
>
> If you don't know what to do here, say Y.
>
> +config TOOLCHAIN_HAS_ZBKB
> + bool
> + default y
> + depends on !64BIT || $(cc-option,-mabi=lp64 -march=rv64ima_zbkb)
> + depends on !32BIT || $(cc-option,-mabi=ilp32 -march=rv32ima_zbkb)
> + depends on LLD_VERSION >= 150000 || LD_VERSION >= 23900
> + depends on AS_HAS_OPTION_ARCH
> +
> +config RISCV_ISA_ZBKB
> + bool "Zbkb extension support for bit manipulation instructions"
> + depends on TOOLCHAIN_HAS_ZBKB
> + depends on RISCV_ALTERNATIVE
> + default y
> + help
> + Adds support to dynamically detect the presence of the ZBKB
> + extension (bit manipulation for cryptography) and enable its usage.
> +
> + The Zbkb extension provides instructions to accelerate a number
> + of common cryptography operations (pack, zip, etc).
> +
> + If you don't know what to do here, say Y.
> +
> config RISCV_ISA_ZICBOM
> bool "Zicbom extension support for non-coherent DMA operation"
> depends on MMU
> diff --git a/arch/riscv/include/asm/runtime-const.h b/arch/riscv/include/asm/runtime-const.h
> new file mode 100644
> index 0000000000000000000000000000000000000000..db2c253a4c37f58927179868d53db824f42e491e
> --- /dev/null
> +++ b/arch/riscv/include/asm/runtime-const.h
> @@ -0,0 +1,254 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +#ifndef _ASM_RISCV_RUNTIME_CONST_H
> +#define _ASM_RISCV_RUNTIME_CONST_H
> +
> +#include <asm/alternative.h>
> +#include <asm/cacheflush.h>
> +#include <asm/text-patching.h>
> +#include <linux/uaccess.h>
> +
> +#ifdef CONFIG_32BIT
> +#define runtime_const_ptr(sym) \
> +({ \
> + typeof(sym) __ret; \
> + asm_inline(".option push\n\t" \
> + ".option norvc\n\t" \
> + "1:\t" \
> + "lui %[__ret],0x89abd\n\t" \
> + "addi %[__ret],%[__ret],-0x211\n\t" \
> + ".option pop\n\t" \
> + ".pushsection runtime_ptr_" #sym ",\"a\"\n\t" \
> + ".long 1b - .\n\t" \
> + ".popsection" \
> + : [__ret] "=r" (__ret)); \
> + __ret; \
> +})
> +#else
> +/*
> + * Loading 64-bit constants into a register from immediates is a non-trivial
> + * task on riscv64. To get it somewhat performant, load 32 bits into two
> + * different registers and then combine the results.
> + *
> + * If the processor supports the Zbkb extension, we can combine the final
> + * "slli,slli,srli,add" into the single "pack" instruction. If the processor
> + * doesn't support Zbkb but does support the Zbb extension, we can
> + * combine the final "slli,srli,add" into one instruction "add.uw".
> + */
> +#define RISCV_RUNTIME_CONST_64_PREAMBLE \
> + ".option push\n\t" \
> + ".option norvc\n\t" \
> + "1:\t" \
> + "lui %[__ret],0x89abd\n\t" \
> + "lui %[__tmp],0x1234\n\t" \
> + "addiw %[__ret],%[__ret],-0x211\n\t" \
> + "addiw %[__tmp],%[__tmp],0x567\n\t" \
> +
> +#define RISCV_RUNTIME_CONST_64_BASE \
> + "slli %[__tmp],%[__tmp],32\n\t" \
> + "slli %[__ret],%[__ret],32\n\t" \
> + "srli %[__ret],%[__ret],32\n\t" \
> + "add %[__ret],%[__ret],%[__tmp]\n\t" \
> +
> +#define RISCV_RUNTIME_CONST_64_ZBA \
> + ".option push\n\t" \
> + ".option arch,+zba\n\t" \
> + "slli %[__tmp],%[__tmp],32\n\t" \
> + "add.uw %[__ret],%[__ret],%[__tmp]\n\t" \
> + "nop\n\t" \
> + "nop\n\t" \
> + ".option pop\n\t" \
> +
> +#define RISCV_RUNTIME_CONST_64_ZBKB \
> + ".option push\n\t" \
> + ".option arch,+zbkb\n\t" \
> + "pack %[__ret],%[__ret],%[__tmp]\n\t" \
> + "nop\n\t" \
> + "nop\n\t" \
> + "nop\n\t" \
> + ".option pop\n\t" \
> +
> +#define RISCV_RUNTIME_CONST_64_POSTAMBLE(sym) \
> + ".option pop\n\t" \
> + ".pushsection runtime_ptr_" #sym ",\"a\"\n\t" \
> + ".long 1b - .\n\t" \
> + ".popsection" \
> +
> +#if defined(CONFIG_RISCV_ISA_ZBA) && defined(CONFIG_RISCV_ISA_ZBKB)
> +#define runtime_const_ptr(sym) \
> +({ \
> + typeof(sym) __ret, __tmp; \
> + asm_inline(RISCV_RUNTIME_CONST_64_PREAMBLE \
> + ALTERNATIVE_2( \
> + RISCV_RUNTIME_CONST_64_BASE, \
> + RISCV_RUNTIME_CONST_64_ZBA, \
> + 0, RISCV_ISA_EXT_ZBA, 1, \
> + RISCV_RUNTIME_CONST_64_ZBKB, \
> + 0, RISCV_ISA_EXT_ZBKB, 1 \
> + ) \
> + RISCV_RUNTIME_CONST_64_POSTAMBLE(sym) \
> + : [__ret] "=r" (__ret), [__tmp] "=r" (__tmp)); \
> + __ret; \
> +})
> +#elif defined(CONFIG_RISCV_ISA_ZBA)
> +#define runtime_const_ptr(sym) \
> +({ \
> + typeof(sym) __ret, __tmp; \
> + asm_inline(RISCV_RUNTIME_CONST_64_PREAMBLE \
> + ALTERNATIVE( \
> + RISCV_RUNTIME_CONST_64_BASE, \
> + RISCV_RUNTIME_CONST_64_ZBA, \
> + 0, RISCV_ISA_EXT_ZBA, 1 \
> + ) \
> + RISCV_RUNTIME_CONST_64_POSTAMBLE(sym) \
> + : [__ret] "=r" (__ret), [__tmp] "=r" (__tmp)); \
> + __ret; \
> +})
> +#elif defined(CONFIG_RISCV_ISA_ZBKB)
> +#define runtime_const_ptr(sym) \
> +({ \
> + typeof(sym) __ret, __tmp; \
> + asm_inline(RISCV_RUNTIME_CONST_64_PREAMBLE \
> + ALTERNATIVE( \
> + RISCV_RUNTIME_CONST_64_BASE, \
> + RISCV_RUNTIME_CONST_64_ZBKB, \
> + 0, RISCV_ISA_EXT_ZBKB, 1 \
> + ) \
> + RISCV_RUNTIME_CONST_64_POSTAMBLE(sym) \
> + : [__ret] "=r" (__ret), [__tmp] "=r" (__tmp)); \
> + __ret; \
> +})
> +#else
> +#define runtime_const_ptr(sym) \
> +({ \
> + typeof(sym) __ret, __tmp; \
> + asm_inline(RISCV_RUNTIME_CONST_64_PREAMBLE \
> + RISCV_RUNTIME_CONST_64_BASE \
> + RISCV_RUNTIME_CONST_64_POSTAMBLE(sym) \
> + : [__ret] "=r" (__ret), [__tmp] "=r" (__tmp)); \
> + __ret; \
> +})
> +#endif
> +#endif
> +
> +#ifdef CONFIG_32BIT
> +#define SRLI "srli "
> +#else
> +#define SRLI "srliw "
> +#endif
Nit: you could move that to include/asm/asm.h where we define
architecture-width independent macro instructions (see REG_S for example).
> +
> +#define runtime_const_shift_right_32(val, sym) \
> +({ \
> + u32 __ret; \
> + asm_inline(".option push\n\t" \
> + ".option norvc\n\t" \
> + "1:\t" \
> + SRLI "%[__ret],%[__val],12\n\t" \
> + ".option pop\n\t" \
> + ".pushsection runtime_shift_" #sym ",\"a\"\n\t" \
> + ".long 1b - .\n\t" \
> + ".popsection" \
> + : [__ret] "=r" (__ret) \
> + : [__val] "r" (val)); \
> + __ret; \
> +})
> +
> +#define runtime_const_init(type, sym) do { \
> + extern s32 __start_runtime_##type##_##sym[]; \
> + extern s32 __stop_runtime_##type##_##sym[]; \
> + \
> + runtime_const_fixup(__runtime_fixup_##type, \
> + (unsigned long)(sym), \
> + __start_runtime_##type##_##sym, \
> + __stop_runtime_##type##_##sym); \
> +} while (0)
> +
> +static inline void __runtime_fixup_caches(void *where, unsigned int insns)
> +{
> + /* On riscv there are currently only cache-wide flushes so va is ignored. */
> + __always_unused uintptr_t va = (uintptr_t)where;
> +
> + flush_icache_range(va, va + 4 * insns);
> +}
> +
> +/*
> + * The 32-bit immediate is stored in a lui+addi pairing.
> + * lui holds the upper 20 bits of the immediate in the first 20 bits of the instruction.
> + * addi holds the lower 12 bits of the immediate in the first 12 bits of the instruction.
> + */
> +static inline void __runtime_fixup_32(__le16 *lui_parcel, __le16 *addi_parcel, unsigned int val)
> +{
> + unsigned int lower_immediate, upper_immediate;
> + __le32 lui_res, addi_res;
> + u32 lui_insn, addi_insn;
> +
> + lui_insn = (u32)le16_to_cpu(lui_parcel[0]) | (u32)le16_to_cpu(lui_parcel[1]) << 16;
> + addi_insn = (u32)le16_to_cpu(addi_parcel[0]) | (u32)le16_to_cpu(addi_parcel[1]) << 16;
> +
> + lower_immediate = sign_extend32(val, 11);
> + upper_immediate = (val - lower_immediate);
> +
> + if (upper_immediate & 0xfffff000) {
> + /* replace upper 20 bits of lui with upper immediate */
> + lui_insn &= 0x00000fff;
> + lui_insn |= upper_immediate & 0xfffff000;
> + } else {
> + /* replace lui with nop if immediate is small enough to fit in addi */
> + lui_insn = 0x00000013;
There exists a NOP4 macro defined in ftrace.h, can you try to use it?
> + }
> +
> + if (lower_immediate & 0x00000fff) {
> + /* replace upper 12 bits of addi with lower 12 bits of val */
> + addi_insn &= 0x000fffff;
> + addi_insn |= (lower_immediate & 0x00000fff) << 20;
> + } else {
> + /* replace addi with nop if lower_immediate is empty */
> + addi_insn = 0x00000013;
Ditto here
> + }
> +
> + addi_res = cpu_to_le32(addi_insn);
> + lui_res = cpu_to_le32(lui_insn);
> + patch_insn_write(addi_parcel, &addi_res, sizeof(addi_res));
> + patch_insn_write(lui_parcel, &lui_res, sizeof(lui_res));
> +}
> +
> +static inline void __runtime_fixup_ptr(void *where, unsigned long val)
> +{
> +#ifdef CONFIG_32BIT
> + __runtime_fixup_32(where, where + 4, val);
> + __runtime_fixup_caches(where, 2);
> +#else
> + __runtime_fixup_32(where, where + 8, val);
> + __runtime_fixup_32(where + 4, where + 12, val >> 32);
> + __runtime_fixup_caches(where, 4);
> +#endif
> +}
> +
> +/*
> + * Replace the least significant 5 bits of the srli/srliw immediate that is
> + * located at bits 20-24
> + */
> +static inline void __runtime_fixup_shift(void *where, unsigned long val)
> +{
> + __le16 *parcel = where;
> + __le32 res;
> + u32 insn;
> +
> + insn = (u32)le16_to_cpu(parcel[0]) | (u32)le16_to_cpu(parcel[1]) << 16;
> +
> + insn &= 0xfe0fffff;
> + insn |= (val & 0b11111) << 20;
> +
> + res = cpu_to_le32(insn);
> + patch_text_nosync(where, &res, sizeof(insn));
> +}
> +
> +static inline void runtime_const_fixup(void (*fn)(void *, unsigned long),
> + unsigned long val, s32 *start, s32 *end)
> +{
> + while (start < end) {
> + fn(*start + (void *)start, val);
> + start++;
> + }
> +}
> +
> +#endif /* _ASM_RISCV_RUNTIME_CONST_H */
> diff --git a/arch/riscv/kernel/vmlinux.lds.S b/arch/riscv/kernel/vmlinux.lds.S
> index 002ca58dd998cb78b662837b5ebac988fb6c77bb..61bd5ba6680a786bf1db7dc37bf1acda0639b5c7 100644
> --- a/arch/riscv/kernel/vmlinux.lds.S
> +++ b/arch/riscv/kernel/vmlinux.lds.S
> @@ -97,6 +97,9 @@ SECTIONS
> {
> EXIT_DATA
> }
> +
> + RUNTIME_CONST_VARIABLES
> +
> PERCPU_SECTION(L1_CACHE_BYTES)
>
> .rel.dyn : {
>
> ---
> base-commit: 2014c95afecee3e76ca4a56956a936e23283f05b
> change-id: 20250123-runtime_const_riscv-6cd854ee2817
Apart from the nits above that are up to you to fixup, you can add:
Reviewed-by: Alexandre Ghiti <alexghiti@...osinc.com>
Thanks,
Alex
Powered by blists - more mailing lists