mirror of
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
synced 2025-09-04 20:19:47 +08:00

Make the architecture-optimized CRC code do its CPU feature checks in subsys_initcalls instead of arch_initcalls. This makes it consistent with arch/*/lib/crypto/ and ensures that it runs after initcalls that possibly could be a prerequisite for kernel-mode FPU, such as x86's xfd_update_static_branch() and loongarch's init_euen_mask(). Note: as far as I can tell, x86's xfd_update_static_branch() isn't *actually* needed for kernel-mode FPU. loongarch's init_euen_mask() is needed to enable save/restore of the vector registers, but loongarch doesn't yet have any CRC or crypto code that uses vector registers anyway. Regardless, let's be consistent with arch/*/lib/crypto/ and robust against any potential future dependency on an arch_initcall. Link: https://lore.kernel.org/r/20250510035959.87995-1-ebiggers@kernel.org Signed-off-by: Eric Biggers <ebiggers@google.com>
124 lines
3.0 KiB
C
124 lines
3.0 KiB
C
// SPDX-License-Identifier: GPL-2.0-only
|
|
/*
|
|
* Accelerated CRC32(C) using ARM CRC, NEON and Crypto Extensions instructions
|
|
*
|
|
* Copyright (C) 2016 Linaro Ltd <ard.biesheuvel@linaro.org>
|
|
*/
|
|
|
|
#include <linux/cpufeature.h>
|
|
#include <linux/crc32.h>
|
|
#include <linux/init.h>
|
|
#include <linux/kernel.h>
|
|
#include <linux/module.h>
|
|
#include <linux/string.h>
|
|
|
|
#include <crypto/internal/simd.h>
|
|
|
|
#include <asm/hwcap.h>
|
|
#include <asm/neon.h>
|
|
#include <asm/simd.h>
|
|
|
|
static __ro_after_init DEFINE_STATIC_KEY_FALSE(have_crc32);
|
|
static __ro_after_init DEFINE_STATIC_KEY_FALSE(have_pmull);
|
|
|
|
#define PMULL_MIN_LEN 64 /* min size of buffer for pmull functions */
|
|
|
|
asmlinkage u32 crc32_pmull_le(const u8 buf[], u32 len, u32 init_crc);
|
|
asmlinkage u32 crc32_armv8_le(u32 init_crc, const u8 buf[], u32 len);
|
|
|
|
asmlinkage u32 crc32c_pmull_le(const u8 buf[], u32 len, u32 init_crc);
|
|
asmlinkage u32 crc32c_armv8_le(u32 init_crc, const u8 buf[], u32 len);
|
|
|
|
static u32 crc32_le_scalar(u32 crc, const u8 *p, size_t len)
|
|
{
|
|
if (static_branch_likely(&have_crc32))
|
|
return crc32_armv8_le(crc, p, len);
|
|
return crc32_le_base(crc, p, len);
|
|
}
|
|
|
|
u32 crc32_le_arch(u32 crc, const u8 *p, size_t len)
|
|
{
|
|
if (len >= PMULL_MIN_LEN + 15 &&
|
|
static_branch_likely(&have_pmull) && crypto_simd_usable()) {
|
|
size_t n = -(uintptr_t)p & 15;
|
|
|
|
/* align p to 16-byte boundary */
|
|
if (n) {
|
|
crc = crc32_le_scalar(crc, p, n);
|
|
p += n;
|
|
len -= n;
|
|
}
|
|
n = round_down(len, 16);
|
|
kernel_neon_begin();
|
|
crc = crc32_pmull_le(p, n, crc);
|
|
kernel_neon_end();
|
|
p += n;
|
|
len -= n;
|
|
}
|
|
return crc32_le_scalar(crc, p, len);
|
|
}
|
|
EXPORT_SYMBOL(crc32_le_arch);
|
|
|
|
static u32 crc32c_scalar(u32 crc, const u8 *p, size_t len)
|
|
{
|
|
if (static_branch_likely(&have_crc32))
|
|
return crc32c_armv8_le(crc, p, len);
|
|
return crc32c_base(crc, p, len);
|
|
}
|
|
|
|
u32 crc32c_arch(u32 crc, const u8 *p, size_t len)
|
|
{
|
|
if (len >= PMULL_MIN_LEN + 15 &&
|
|
static_branch_likely(&have_pmull) && crypto_simd_usable()) {
|
|
size_t n = -(uintptr_t)p & 15;
|
|
|
|
/* align p to 16-byte boundary */
|
|
if (n) {
|
|
crc = crc32c_scalar(crc, p, n);
|
|
p += n;
|
|
len -= n;
|
|
}
|
|
n = round_down(len, 16);
|
|
kernel_neon_begin();
|
|
crc = crc32c_pmull_le(p, n, crc);
|
|
kernel_neon_end();
|
|
p += n;
|
|
len -= n;
|
|
}
|
|
return crc32c_scalar(crc, p, len);
|
|
}
|
|
EXPORT_SYMBOL(crc32c_arch);
|
|
|
|
u32 crc32_be_arch(u32 crc, const u8 *p, size_t len)
|
|
{
|
|
return crc32_be_base(crc, p, len);
|
|
}
|
|
EXPORT_SYMBOL(crc32_be_arch);
|
|
|
|
static int __init crc32_arm_init(void)
|
|
{
|
|
if (elf_hwcap2 & HWCAP2_CRC32)
|
|
static_branch_enable(&have_crc32);
|
|
if (elf_hwcap2 & HWCAP2_PMULL)
|
|
static_branch_enable(&have_pmull);
|
|
return 0;
|
|
}
|
|
subsys_initcall(crc32_arm_init);
|
|
|
|
static void __exit crc32_arm_exit(void)
|
|
{
|
|
}
|
|
module_exit(crc32_arm_exit);
|
|
|
|
u32 crc32_optimizations(void)
|
|
{
|
|
if (elf_hwcap2 & (HWCAP2_CRC32 | HWCAP2_PMULL))
|
|
return CRC32_LE_OPTIMIZATION | CRC32C_OPTIMIZATION;
|
|
return 0;
|
|
}
|
|
EXPORT_SYMBOL(crc32_optimizations);
|
|
|
|
MODULE_AUTHOR("Ard Biesheuvel <ard.biesheuvel@linaro.org>");
|
|
MODULE_DESCRIPTION("Accelerated CRC32(C) using ARM CRC, NEON and Crypto Extensions");
|
|
MODULE_LICENSE("GPL v2");
|