riscv: lib: add optimized memcmp() and extend KUnit tests
Subsystems (2)
v4 cover riscv: lib: add optimized memcmp() and extend KUnit tests
This v4 series introduces an optimized RISC-V memcmp() implementation and extends the KUnit string tests to cover both functional correctness and benchmarking of memcmp().
The memcmp() implementation incorporates review feedback from earlier versions. It features improved alignment handling, optimized loop structures, efficient use of the Zbb extension, and ensures overall correctness.
The KUnit updates provide comprehensive memcmp() test coverage and a benchmark mode. This v4 specifically fixes a build error regarding an undeclared function call (STRING_BENCH).
---
v4 changes:
- Fixed build error: call to undeclared function 'STRING_BENCH'
- Link to v3: https://lore.kernel.org/all/[email protected]/
v3 changes:
- Split memcmp benchmark into wrapper (string_bench_memcmp) and worker function (do_string_bench_memcmp).
- Removed all C99 mixed declarations; moved all variable declarations
to the top of each function.
- Converted len, iterations and loop counters in the benchmark to u64
to avoid implicit casts.
- Cleaned up spacing, indentation and minor style issues.
- Added #if defined(CONFIG_RISCV_ISA_ZBB)... in memcmp.S
- Link to v2: https://lore.kernel.org/all/[email protected]/
v2 changes:
- Added alignment checks for buffers to avoid expensive misaligned loads.
- Optimized the loop using end-pointers to reduce per-iteration overhead.
- Implemented word-aligned tail handling using ZBB shifts.
- Removed redundant pointer equality (a0 == a1) check.
- Retained BE support via #ifndef; ZBB rev8 is used for the LE fast-path.
- Fixed KUnit build failures for Clang and non-benchmark configs.
- Link to v1: https://lore.kernel.org/all/[email protected]/
Milan Tripkovic (2):
riscv: lib: add memcmp() implementation lib/string_kunit: extend benchmarks and unit test to memcmp()
arch/riscv/include/asm/string.h | 2 + arch/riscv/lib/Makefile | 1 + arch/riscv/lib/memcmp.S | 125 ++++++++++++++++++++++++++++++++ arch/riscv/purgatory/Makefile | 5 +- lib/tests/string_kunit.c | 120 ++++++++++++++++++++++++++++++ 5 files changed, 252 insertions(+), 1 deletion(-) create mode 100644 arch/riscv/lib/memcmp.S
Signature · 1 line
2.43.0
PATCH v4 1/2 riscv: lib: add memcmp() implementation
From: Milan Tripkovic <[email protected]>
Add an assembly implementation of memcmp() for RISC-V. The implementation uses the ZBB extension for word-at-a-time comparison and an assembly fallback for non-ZBB systems.
Benchmark results (QEMU TCG, rv64, Aligned):
Len | Default | NoZBB | ZBB | %NoZBB | %ZBB ------|---------|--------|--------|--------|------- 1 B | 20.3 | 25.0 | 20.9 | +23.2% | +3.0% 7 B | 88.9 | 107.5 | 155.7 | +20.9% | +75.1% 8 B | 89.6 | 110.9 | 176.2 | +23.8% | +96.7% 16 B | 134.4 | 172.4 | 334.8 | +28.3% | +149.1% 31 B | 163.5 | 220.5 | 606.2 | +34.9% | +270.8% 64 B | 203.8 | 235.9 | 968.6 | +15.8% | +375.3% 127 B | 224.6 | 268.7 | 1362.8 | +19.6% | +506.8% 512 B | 235.7 | 271.1 | 1913.7 | +15.0% | +711.9% 1024 B| 256.8 | 290.6 | 2123.6 | +13.2% | +726.9% 4096 B| 263.8 | 302.9 | 2290.4 | +14.8% | +768.2%
Benchmark results (QEMU TCG, rv64, Unaligned - Offset 3):
Len | Default | NoZBB | ZBB | %NoZBB | %ZBB ------|---------|--------|--------|--------|------- 1 B | 20.7 | 21.7 | 21.5 | +4.8% | +3.9% 7 B | 96.2 | 99.1 | 96.9 | +3.0% | +0.7% 8 B | 97.5 | 118.5 | 110.5 | +21.5% | +13.3% 16 B | 136.7 | 166.6 | 172.8 | +21.9% | +26.4% 31 B | 167.6 | 206.5 | 211.9 | +23.2% | +26.4% 64 B | 204.4 | 229.9 | 240.3 | +12.5% | +17.6% 127 B | 229.6 | 261.7 | 269.0 | +14.0% | +17.2% 512 B | 245.5 | 260.8 | 269.9 | +6.2% | +9.9% 1024 B| 246.9 | 261.2 | 283.5 | +5.8% | +14.8% 4096 B| 250.7 | 295.8 | 299.7 | +18.0% | +19.5%
--- arch/riscv/include/asm/string.h | 2 + arch/riscv/lib/Makefile | 1 + arch/riscv/lib/memcmp.S | 125 ++++++++++++++++++++++++++++++++ arch/riscv/purgatory/Makefile | 5 +- 4 files changed, 132 insertions(+), 1 deletion(-) create mode 100644 arch/riscv/lib/memcmp.S
arch/riscv/include/asm/string.h +2 -0
@@ -18,6 +18,8 @@ extern asmlinkage void *__memcpy(void *, const void *, size_t);#define __HAVE_ARCH_MEMMOVE extern asmlinkage void *memmove(void *, const void *, size_t); extern asmlinkage void *__memmove(void *, const void *, size_t); +#define __HAVE_ARCH_MEMCMP +extern asmlinkage int memcmp(const void *, const void *, size_t);#if !(defined(CONFIG_KASAN_GENERIC ) || defined(CONFIG_KASAN_SW_TAGS ))#define __HAVE_ARCH_STRCMP
arch/riscv/lib/Makefile +1 -0
@@ -3,6 +3,7 @@ lib-y += delay.o lib-y += memcpy.o lib-y += memset.o lib-y += memmove.o +lib-y += memcmp.o ifeq ($(CONFIG_KASAN_GENERIC)$(CONFIG_KASAN_SW_TAGS),) lib-y += strcmp.o lib-y += strlen.o
arch/riscv/lib/memcmp.S new file +125 -0
125 lines not on this page · raw message
arch/riscv/purgatory/Makefile +4 -1
@@ -1,6 +1,6 @@ # SPDX-License-Identifier: GPL-2.0 -purgatory-y := purgatory.o sha256.o entry.o string.o ctype.o memcpy.o memset.o +purgatory-y := purgatory.o sha256.o entry.o string.o ctype.o memcpy.o memset.o memcmp.o ifeq ($(CONFIG_KASAN_GENERIC)$(CONFIG_KASAN_SW_TAGS),) purgatory-y += strcmp.o strlen.o strncmp.o strnlen.o strchr.o strrchr.o endif
@@ -41,6 +41,9 @@ $(obj)/strchr.o: $(srctree)/arch/riscv/lib/strchr.S FORCE $(obj)/strrchr.o: $(srctree)/arch/riscv/lib/strrchr.S FORCE $(call if_changed_rule,as_o_S) +$(obj)/memcmp.o: $(srctree)/arch/riscv/lib/memcmp.S FORCE + $(call if_changed_rule,as_o_S) + CFLAGS_sha256.o := -D__DISABLE_EXPORTS -D__NO_FORTIFY CFLAGS_string.o := -D__DISABLE_EXPORTS CFLAGS_ctype.o := -D__DISABLE_EXPORTS
Signature · 1 line
2.43.0
Re: [PATCH v4 1/2] riscv: lib: add memcmp() implementation
Quoted 5 lines · On Mon, 18 May 2026 15:14:06 +0200 Milan Tripkovic <[email protected]> wrote:
From: Milan Tripkovic <[email protected]>
Add an assembly implementation of
memcmp()for RISC-V. The implementation uses the ZBB extension for word-at-a-time comparison and an assembly fallback for non-ZBB systems.
I think I mentioned before that the only ZBB bit I can see is the byte reverse at the end needed to get the correct sign. For non-ZBB it would be better to fall back to a byte compare at that point.
Oh - and there should be change info for this patch in this email.
-- David
Quoted 218 lines
Benchmark results (QEMU TCG, rv64, Aligned):
Len | Default | NoZBB | ZBB | %NoZBB | %ZBB ------|---------|--------|--------|--------|------- 1 B | 20.3 | 25.0 | 20.9 | +23.2% | +3.0% 7 B | 88.9 | 107.5 | 155.7 | +20.9% | +75.1% 8 B | 89.6 | 110.9 | 176.2 | +23.8% | +96.7% 16 B | 134.4 | 172.4 | 334.8 | +28.3% | +149.1% 31 B | 163.5 | 220.5 | 606.2 | +34.9% | +270.8% 64 B | 203.8 | 235.9 | 968.6 | +15.8% | +375.3% 127 B | 224.6 | 268.7 | 1362.8 | +19.6% | +506.8% 512 B | 235.7 | 271.1 | 1913.7 | +15.0% | +711.9% 1024 B| 256.8 | 290.6 | 2123.6 | +13.2% | +726.9% 4096 B| 263.8 | 302.9 | 2290.4 | +14.8% | +768.2%Benchmark results (QEMU TCG, rv64, Unaligned - Offset 3):
Len | Default | NoZBB | ZBB | %NoZBB | %ZBB ------|---------|--------|--------|--------|------- 1 B | 20.7 | 21.7 | 21.5 | +4.8% | +3.9% 7 B | 96.2 | 99.1 | 96.9 | +3.0% | +0.7% 8 B | 97.5 | 118.5 | 110.5 | +21.5% | +13.3% 16 B | 136.7 | 166.6 | 172.8 | +21.9% | +26.4% 31 B | 167.6 | 206.5 | 211.9 | +23.2% | +26.4% 64 B | 204.4 | 229.9 | 240.3 | +12.5% | +17.6% 127 B | 229.6 | 261.7 | 269.0 | +14.0% | +17.2% 512 B | 245.5 | 260.8 | 269.9 | +6.2% | +9.9% 1024 B| 246.9 | 261.2 | 283.5 | +5.8% | +14.8% 4096 B| 250.7 | 295.8 | 299.7 | +18.0% | +19.5%Signed-off-by Milan Tripkovic <[email protected]>--- arch/riscv/include/asm/string.h | 2 + arch/riscv/lib/Makefile | 1 + arch/riscv/lib/memcmp.S | 125 ++++++++++++++++++++++++++++++++ arch/riscv/purgatory/Makefile | 5 +- 4 files changed, 132 insertions(+), 1 deletion(-) create mode 100644 arch/riscv/lib/memcmp.Sarch/riscv/include/asm/string.h +2
-0@@ -18,6 +18,8 @@ extern asmlinkage void *__memcpy(void *, const void *, size_t);#define __HAVE_ARCH_MEMMOVE extern asmlinkage void *memmove(void *, const void *, size_t); extern asmlinkage void *__memmove(void *, const void *, size_t); +#define __HAVE_ARCH_MEMCMP +extern asmlinkage int memcmp(const void *, const void *, size_t);#if !(defined(CONFIG_KASAN_GENERIC ) || defined(CONFIG_KASAN_SW_TAGS ))#define __HAVE_ARCH_STRCMParch/riscv/lib/Makefile +1
-0@@ -3,6 +3,7 @@ lib-y += delay.o lib-y += memcpy.o lib-y += memset.o lib-y += memmove.o +lib-y += memcmp.o ifeq ($(CONFIG_KASAN_GENERIC)$(CONFIG_KASAN_SW_TAGS),) lib-y += strcmp.o lib-y += strlen.oarch/riscv/lib/memcmp.S new file +125
-0@@ -0,0 +1,125 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ + +#include <linux/linkage.h> +#include <asm/asm.h> +#include <asm/alternative-macros.h> +#include <asm/hwcap.h> + +/* int memcmp(const void *cs, const void *ct, size_t n) */ +SYM_FUNC_START(memcmp) + + __ALTERNATIVE_CFG("nop", "j memcmp_zbb", 0, RISCV_ISA_EXT_ZBB, + IS_ENABLED(CONFIG_RISCV_ISA_ZBB) && IS_ENABLED(CONFIG_TOOLCHAIN_HAS_ZBB)) +/* + * Parameters + * a0 - Pointer to first memory block (cs), also return value + * a1 - Pointer to second memory block (ct) + * a2 - Number of bytes to compare (n), transformed to end pointer (a0 + n) + * + * Returns + * a0 - 0 if equal, positive if cs > ct, negative if cs < ct + * + * Clobbers + * t0, t1 + */ + beqz a2, 2f + add a2, a0, a2 +1: + lbu t0, 0(a0) + lbu t1, 0(a1) + bne t0, t1, 3f + addi a0, a0, 1 + addi a1, a1, 1 + bne a0, a2, 1b +2: + li a0, 0 + ret +3: + sub a0, t0, t1 + ret + +#if defined(CONFIG_RISCV_ISA_ZBB) && defined(CONFIG_TOOLCHAIN_HAS_ZBB) +memcmp_zbb: + +.option push +.option arch,+zbb +/* + * Parameters + * a0 - Pointer to first memory block (cs), also return value + * a1 - Pointer to second memory block (ct) + * a2 - Number of bytes to compare (n), decremented during loop + * + * Returns + * a0 - 0 if equal, positive if cs > ct, negative if cs < ct + * + * Clobbers + * t0, t1, t2, t3, t4 + */ + add t3, a0, a2 + or t0, a0, a1 + andi t0, t0, (SZREG - 1) + bnez t0, 5f + + addi t4, t3, -SZREG + bltu t4, a0, 7f + +1: + REG_L t1, 0(a0) + REG_L t2, 0(a1) + bne t1, t2, 2f + addi a0, a0, SZREG + addi a1, a1, SZREG + bleu a0, t4, 1b + +7: + beq a0, t3, 4f + REG_L t1, 0(a0) + REG_L t2, 0(a1) + + sub t0, t3, a0 + li t4, SZREG + sub t0, t4, t0 + slli t0, t0, 3 + +#ifndef CONFIG_CPU_BIG_ENDIAN + rev8 t1, t1 + rev8 t2, t2 +#endif + srl t1, t1, t0 + srl t2, t2, t0 + + bne t1, t2, 8f + li a0, 0 + ret +5: + beq a0, t3, 4f +6: + lbu t1, 0(a0) + lbu t2, 0(a1) + bne t1, t2, 3f + addi a0, a0, 1 + addi a1, a1, 1 + bne a0, t3, 6b + +4: li a0, 0 + ret +2: +#ifndef CONFIG_CPU_BIG_ENDIAN + rev8 t1, t1 + rev8 t2, t2 +#endif +8: + sltu a0, t2, t1 + sltu t0, t1, t2 + sub a0, a0, t0 + ret + +3: + sub a0, t1, t2 + ret + +.option pop +#endif +SYM_FUNC_END(memcmp) +SYM_FUNC_ALIAS(__pi_memcmp, memcmp) +EXPORT_SYMBOL(memcmp)arch/riscv/purgatory/Makefile +4
-1@@ -1,6 +1,6 @@ # SPDX-License-Identifier: GPL-2.0 -purgatory-y := purgatory.o sha256.o entry.o string.o ctype.o memcpy.o memset.o +purgatory-y := purgatory.o sha256.o entry.o string.o ctype.o memcpy.o memset.o memcmp.o ifeq ($(CONFIG_KASAN_GENERIC)$(CONFIG_KASAN_SW_TAGS),) purgatory-y += strcmp.o strlen.o strncmp.o strnlen.o strchr.o strrchr.o endif@@ -41,6 +41,9 @@ $(obj)/strchr.o: $(srctree)/arch/riscv/lib/strchr.S FORCE $(obj)/strrchr.o: $(srctree)/arch/riscv/lib/strrchr.S FORCE $(call if_changed_rule,as_o_S) +$(obj)/memcmp.o: $(srctree)/arch/riscv/lib/memcmp.S FORCE + $(call if_changed_rule,as_o_S) + CFLAGS_sha256.o := -D__DISABLE_EXPORTS -D__NO_FORTIFY CFLAGS_string.o := -D__DISABLE_EXPORTS CFLAGS_ctype.o := -D__DISABLE_EXPORTS
PATCH v4 2/2 lib/string_kunit: extend benchmarks and unit test to memcmp()
In reply to Milan TripkovicFrom: Milan Tripkovic <[email protected]>
Extend the string benchmarking suite to include memcmp(). Extend the string unit test to include memcmp().
--- lib/tests/string_kunit.c | 120 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 120 insertions(+)
lib/tests/string_kunit.c +120 -0
@@ -881,6 +881,124 @@ static void string_bench_strrchr(struct kunit *test)STRING_BENCH_BUF (test, buf, len, strrchr, buf,'\0' ); } +static void string_test_memcmp (struct kunit *test) +{ +const unsigned int max_offset =16 ; +const unsigned int max_len =32 ; +const unsigned int buf_size = max_offset + max_len +32 ; +u8 *buf1, *buf2; +unsigned int i, j, len, k; +int res; + + buf1 =kunit_kzalloc (test, buf_size,GFP_KERNEL ); + buf2 =kunit_kzalloc (test, buf_size,GFP_KERNEL ); +KUNIT_ASSERT_NOT_ERR_OR_NULL (test, buf1); +KUNIT_ASSERT_NOT_ERR_OR_NULL (test, buf2); + +for (i =0 ; i < max_offset; i++) { +for (j =0 ; j < max_offset; j++) { +for (len =0 ; len <= max_len; len++) { +memset (buf1,'A' , buf_size); +memset (buf2,'A' , buf_size); +KUNIT_EXPECT_EQ_MSG (test,memcmp (buf1 + i, buf2 + j, len),0 , +"Should be equal: i:%u j:%u len:%u" , i, j, len); +for (k =0 ; k < len; k++) { +memset (buf1,'A' , buf_size); +memset (buf2,'A' , buf_size); + buf2[j + k] ='B' ; + res =memcmp (buf1 + i, buf2 + j, len); +KUNIT_EXPECT_NE_MSG (test, res,0 , +"Should detect difference at k:%u (i:%u j:%u len:%u)" , + k, i, j, len); +if (buf1[i + k] < buf2[j + k]) +KUNIT_EXPECT_LT (test, res,0 ); +else +KUNIT_EXPECT_GT (test, res,0 ); + } + } + } + } +} + +#ifndef STRING_BENCH +#define STRING_BENCH (...) 0 +#endif + +static void do_string_bench_memcmp (struct kunit *test) +{ +char *buf1 =NULL ; +char *buf2 =NULL ; +const u64 lengths[] = {1 ,7 ,8 ,16 ,32 ,64 ,128 ,512 ,1024 ,4096 }; +const int offsets[] = {0 ,1 ,3 ,7 }; +const u64 max_len =4096 +64 ; +unsigned int w, o, i; +unsigned int off; +u64 len; +char *p1; +char *p2; +u64 iterations; +u64 elapsed; +u64 ns_per_call; +u64 mbps; +u64 j; + + buf1 =vmalloc (max_len); + buf2 =vmalloc (max_len); + +if (!buf1 || !buf2) { +vfree (buf1); +vfree (buf2); +kunit_err (test,"vmalloc failed\n" ); +return ; + } + +memset (buf1,'A' , max_len); +memset (buf2,'A' , max_len); + +for (w =0 ; w <100000U ; w++) + (void )memcmp (buf1, buf2,4096 ); + +for (o =0 ; o <ARRAY_SIZE (offsets); o++) { + off = offsets[o]; + +for (i =0 ; i <ARRAY_SIZE (lengths); i++) { + len = lengths[i]; + p1 = buf1; + p2 = buf2 + off; + iterations = (len <512 ) ?100000ULL :10000ULL ; + +for (j =0 ; j < iterations; j++) { + (void )memcmp (p1, p2, len); +barrier (); + } + + elapsed =STRING_BENCH (iterations, memcmp, p1, p2, len); + ns_per_call =div_u64 (elapsed, iterations); + mbps = len ?div_u64 (iterations * len * (NSEC_PER_SEC /MEGA ), elapsed) :0 ; + +if (off ==0 ) { +kunit_info (test,"bench_memcmp_aligned: len=%-4llu: %llu MB/s (%llu ns/call)\n" , + len, mbps, ns_per_call); + }else { +kunit_info (test,"bench_memcmp_unaligned(off=%u): len=%-4llu: %llu MB/s (%llu ns/call)\n" , + off, len, mbps, ns_per_call); + } + } + } + +vfree (buf1); +vfree (buf2); +} + +static void string_bench_memcmp (struct kunit *test) +{ +if (!IS_ENABLED (CONFIG_STRING_KUNIT_BENCH )) { +kunit_skip (test,"CONFIG_STRING_KUNIT_BENCH not enabled" ); +return ; + } +do_string_bench_memcmp (test); +} +static struct kunit_case string_test_cases[] = {KUNIT_CASE (string_test_memset16),KUNIT_CASE (string_test_memset32),
@@ -910,6 +1028,8 @@ static struct kunit_case string_test_cases[] = {KUNIT_CASE (string_bench_strnlen),KUNIT_CASE (string_bench_strchr),KUNIT_CASE (string_bench_strrchr), +KUNIT_CASE (string_test_memcmp), + KUNIT_CASE_SLOW(string_bench_memcmp), {} };
Signature · 1 line
2.43.0