1
linux/arch/arm64/crypto/poly1305-glue.c

233 lines
6.0 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0
/*
* OpenSSL/Cryptogams accelerated Poly1305 transform for arm64
*
* Copyright (C) 2019 Linaro Ltd. <ard.biesheuvel@linaro.org>
*/
#include <asm/hwcap.h>
#include <asm/neon.h>
#include <asm/simd.h>
#include <linux/unaligned.h>
#include <crypto/algapi.h>
#include <crypto/internal/hash.h>
#include <crypto/internal/poly1305.h>
#include <crypto/internal/simd.h>
#include <linux/cpufeature.h>
#include <linux/crypto.h>
#include <linux/jump_label.h>
#include <linux/module.h>
asmlinkage void poly1305_init_arm64(void *state, const u8 *key);
asmlinkage void poly1305_blocks(void *state, const u8 *src, u32 len, u32 hibit);
asmlinkage void poly1305_blocks_neon(void *state, const u8 *src, u32 len, u32 hibit);
asmlinkage void poly1305_emit(void *state, u8 *digest, const u32 *nonce);
static __ro_after_init DEFINE_STATIC_KEY_FALSE(have_neon);
void poly1305_init_arch(struct poly1305_desc_ctx *dctx, const u8 key[POLY1305_KEY_SIZE])
{
poly1305_init_arm64(&dctx->h, key);
dctx->s[0] = get_unaligned_le32(key + 16);
dctx->s[1] = get_unaligned_le32(key + 20);
dctx->s[2] = get_unaligned_le32(key + 24);
dctx->s[3] = get_unaligned_le32(key + 28);
dctx->buflen = 0;
}
EXPORT_SYMBOL(poly1305_init_arch);
static int neon_poly1305_init(struct shash_desc *desc)
{
struct poly1305_desc_ctx *dctx = shash_desc_ctx(desc);
dctx->buflen = 0;
dctx->rset = 0;
dctx->sset = false;
return 0;
}
static void neon_poly1305_blocks(struct poly1305_desc_ctx *dctx, const u8 *src,
u32 len, u32 hibit, bool do_neon)
{
if (unlikely(!dctx->sset)) {
if (!dctx->rset) {
crypto: arm64/poly1305 - fix a read out-of-bound A kasan error was reported during fuzzing: BUG: KASAN: slab-out-of-bounds in neon_poly1305_blocks.constprop.0+0x1b4/0x250 [poly1305_neon] Read of size 4 at addr ffff0010e293f010 by task syz-executor.5/1646715 CPU: 4 PID: 1646715 Comm: syz-executor.5 Kdump: loaded Not tainted 5.10.0.aarch64 #1 Hardware name: Huawei TaiShan 2280 /BC11SPCD, BIOS 1.59 01/31/2019 Call trace: dump_backtrace+0x0/0x394 show_stack+0x34/0x4c arch/arm64/kernel/stacktrace.c:196 __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x158/0x1e4 lib/dump_stack.c:118 print_address_description.constprop.0+0x68/0x204 mm/kasan/report.c:387 __kasan_report+0xe0/0x140 mm/kasan/report.c:547 kasan_report+0x44/0xe0 mm/kasan/report.c:564 check_memory_region_inline mm/kasan/generic.c:187 [inline] __asan_load4+0x94/0xd0 mm/kasan/generic.c:252 neon_poly1305_blocks.constprop.0+0x1b4/0x250 [poly1305_neon] neon_poly1305_do_update+0x6c/0x15c [poly1305_neon] neon_poly1305_update+0x9c/0x1c4 [poly1305_neon] crypto_shash_update crypto/shash.c:131 [inline] shash_finup_unaligned+0x84/0x15c crypto/shash.c:179 crypto_shash_finup+0x8c/0x140 crypto/shash.c:193 shash_digest_unaligned+0xb8/0xe4 crypto/shash.c:201 crypto_shash_digest+0xa4/0xfc crypto/shash.c:217 crypto_shash_tfm_digest+0xb4/0x150 crypto/shash.c:229 essiv_skcipher_setkey+0x164/0x200 [essiv] crypto_skcipher_setkey+0xb0/0x160 crypto/skcipher.c:612 skcipher_setkey+0x3c/0x50 crypto/algif_skcipher.c:305 alg_setkey+0x114/0x2a0 crypto/af_alg.c:220 alg_setsockopt+0x19c/0x210 crypto/af_alg.c:253 __sys_setsockopt+0x190/0x2e0 net/socket.c:2123 __do_sys_setsockopt net/socket.c:2134 [inline] __se_sys_setsockopt net/socket.c:2131 [inline] __arm64_sys_setsockopt+0x78/0x94 net/socket.c:2131 __invoke_syscall arch/arm64/kernel/syscall.c:36 [inline] invoke_syscall+0x64/0x100 arch/arm64/kernel/syscall.c:48 el0_svc_common.constprop.0+0x220/0x230 arch/arm64/kernel/syscall.c:155 do_el0_svc+0xb4/0xd4 arch/arm64/kernel/syscall.c:217 el0_svc+0x24/0x3c arch/arm64/kernel/entry-common.c:353 el0_sync_handler+0x160/0x164 arch/arm64/kernel/entry-common.c:369 el0_sync+0x160/0x180 arch/arm64/kernel/entry.S:683 This error can be reproduced by the following code compiled as ko on a system with kasan enabled: #include <linux/module.h> #include <linux/crypto.h> #include <crypto/hash.h> #include <crypto/poly1305.h> char test_data[] = "\x00\x01\x02\x03\x04\x05\x06\x07" "\x08\x09\x0a\x0b\x0c\x0d\x0e\x0f" "\x10\x11\x12\x13\x14\x15\x16\x17" "\x18\x19\x1a\x1b\x1c\x1d\x1e"; int init(void) { struct crypto_shash *tfm = NULL; char *data = NULL, *out = NULL; tfm = crypto_alloc_shash("poly1305", 0, 0); data = kmalloc(POLY1305_KEY_SIZE - 1, GFP_KERNEL); out = kmalloc(POLY1305_DIGEST_SIZE, GFP_KERNEL); memcpy(data, test_data, POLY1305_KEY_SIZE - 1); crypto_shash_tfm_digest(tfm, data, POLY1305_KEY_SIZE - 1, out); kfree(data); kfree(out); return 0; } void deinit(void) { } module_init(init) module_exit(deinit) MODULE_LICENSE("GPL"); The root cause of the bug sits in neon_poly1305_blocks. The logic neon_poly1305_blocks() performed is that if it was called with both s[] and r[] uninitialized, it will first try to initialize them with the data from the first "block" that it believed to be 32 bytes in length. First 16 bytes are used as the key and the next 16 bytes for s[]. This would lead to the aforementioned read out-of-bound. However, after calling poly1305_init_arch(), only 16 bytes were deducted from the input and s[] is initialized yet again with the following 16 bytes. The second initialization of s[] is certainly redundent which indicates that the first initialization should be for r[] only. This patch fixes the issue by calling poly1305_init_arm64() instead of poly1305_init_arch(). This is also the implementation for the same algorithm on arm platform. Fixes: f569ca164751 ("crypto: arm64/poly1305 - incorporate OpenSSL/CRYPTOGAMS NEON implementation") Cc: stable@vger.kernel.org Signed-off-by: GUO Zihua <guozihua@huawei.com> Reviewed-by: Eric Biggers <ebiggers@google.com> Acked-by: Will Deacon <will@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2022-07-21 23:31:57 -07:00
poly1305_init_arm64(&dctx->h, src);
src += POLY1305_BLOCK_SIZE;
len -= POLY1305_BLOCK_SIZE;
dctx->rset = 1;
}
if (len >= POLY1305_BLOCK_SIZE) {
dctx->s[0] = get_unaligned_le32(src + 0);
dctx->s[1] = get_unaligned_le32(src + 4);
dctx->s[2] = get_unaligned_le32(src + 8);
dctx->s[3] = get_unaligned_le32(src + 12);
src += POLY1305_BLOCK_SIZE;
len -= POLY1305_BLOCK_SIZE;
dctx->sset = true;
}
if (len < POLY1305_BLOCK_SIZE)
return;
}
len &= ~(POLY1305_BLOCK_SIZE - 1);
if (static_branch_likely(&have_neon) && likely(do_neon))
poly1305_blocks_neon(&dctx->h, src, len, hibit);
else
poly1305_blocks(&dctx->h, src, len, hibit);
}
static void neon_poly1305_do_update(struct poly1305_desc_ctx *dctx,
const u8 *src, u32 len, bool do_neon)
{
if (unlikely(dctx->buflen)) {
u32 bytes = min(len, POLY1305_BLOCK_SIZE - dctx->buflen);
memcpy(dctx->buf + dctx->buflen, src, bytes);
src += bytes;
len -= bytes;
dctx->buflen += bytes;
if (dctx->buflen == POLY1305_BLOCK_SIZE) {
neon_poly1305_blocks(dctx, dctx->buf,
POLY1305_BLOCK_SIZE, 1, false);
dctx->buflen = 0;
}
}
if (likely(len >= POLY1305_BLOCK_SIZE)) {
neon_poly1305_blocks(dctx, src, len, 1, do_neon);
src += round_down(len, POLY1305_BLOCK_SIZE);
len %= POLY1305_BLOCK_SIZE;
}
if (unlikely(len)) {
dctx->buflen = len;
memcpy(dctx->buf, src, len);
}
}
static int neon_poly1305_update(struct shash_desc *desc,
const u8 *src, unsigned int srclen)
{
bool do_neon = crypto_simd_usable() && srclen > 128;
struct poly1305_desc_ctx *dctx = shash_desc_ctx(desc);
if (static_branch_likely(&have_neon) && do_neon)
kernel_neon_begin();
neon_poly1305_do_update(dctx, src, srclen, do_neon);
if (static_branch_likely(&have_neon) && do_neon)
kernel_neon_end();
return 0;
}
void poly1305_update_arch(struct poly1305_desc_ctx *dctx, const u8 *src,
unsigned int nbytes)
{
if (unlikely(dctx->buflen)) {
u32 bytes = min(nbytes, POLY1305_BLOCK_SIZE - dctx->buflen);
memcpy(dctx->buf + dctx->buflen, src, bytes);
src += bytes;
nbytes -= bytes;
dctx->buflen += bytes;
if (dctx->buflen == POLY1305_BLOCK_SIZE) {
poly1305_blocks(&dctx->h, dctx->buf, POLY1305_BLOCK_SIZE, 1);
dctx->buflen = 0;
}
}
if (likely(nbytes >= POLY1305_BLOCK_SIZE)) {
unsigned int len = round_down(nbytes, POLY1305_BLOCK_SIZE);
if (static_branch_likely(&have_neon) && crypto_simd_usable()) {
crypto: arch/lib - limit simd usage to 4k chunks The initial Zinc patchset, after some mailing list discussion, contained code to ensure that kernel_fpu_enable would not be kept on for more than a 4k chunk, since it disables preemption. The choice of 4k isn't totally scientific, but it's not a bad guess either, and it's what's used in both the x86 poly1305, blake2s, and nhpoly1305 code already (in the form of PAGE_SIZE, which this commit corrects to be explicitly 4k for the former two). Ard did some back of the envelope calculations and found that at 5 cycles/byte (overestimate) on a 1ghz processor (pretty slow), 4k means we have a maximum preemption disabling of 20us, which Sebastian confirmed was probably a good limit. Unfortunately the chunking appears to have been left out of the final patchset that added the glue code. So, this commit adds it back in. Fixes: 84e03fa39fbe ("crypto: x86/chacha - expose SIMD ChaCha routine as library function") Fixes: b3aad5bad26a ("crypto: arm64/chacha - expose arm64 ChaCha routine as library function") Fixes: a44a3430d71b ("crypto: arm/chacha - expose ARM ChaCha routine as library function") Fixes: d7d7b8535662 ("crypto: x86/poly1305 - wire up faster implementations for kernel") Fixes: f569ca164751 ("crypto: arm64/poly1305 - incorporate OpenSSL/CRYPTOGAMS NEON implementation") Fixes: a6b803b3ddc7 ("crypto: arm/poly1305 - incorporate OpenSSL/CRYPTOGAMS NEON implementation") Fixes: ed0356eda153 ("crypto: blake2s - x86_64 SIMD implementation") Cc: Eric Biggers <ebiggers@google.com> Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Cc: stable@vger.kernel.org Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com> Reviewed-by: Ard Biesheuvel <ardb@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2020-04-22 16:18:53 -07:00
do {
unsigned int todo = min_t(unsigned int, len, SZ_4K);
kernel_neon_begin();
poly1305_blocks_neon(&dctx->h, src, todo, 1);
kernel_neon_end();
len -= todo;
src += todo;
} while (len);
} else {
poly1305_blocks(&dctx->h, src, len, 1);
crypto: arch/lib - limit simd usage to 4k chunks The initial Zinc patchset, after some mailing list discussion, contained code to ensure that kernel_fpu_enable would not be kept on for more than a 4k chunk, since it disables preemption. The choice of 4k isn't totally scientific, but it's not a bad guess either, and it's what's used in both the x86 poly1305, blake2s, and nhpoly1305 code already (in the form of PAGE_SIZE, which this commit corrects to be explicitly 4k for the former two). Ard did some back of the envelope calculations and found that at 5 cycles/byte (overestimate) on a 1ghz processor (pretty slow), 4k means we have a maximum preemption disabling of 20us, which Sebastian confirmed was probably a good limit. Unfortunately the chunking appears to have been left out of the final patchset that added the glue code. So, this commit adds it back in. Fixes: 84e03fa39fbe ("crypto: x86/chacha - expose SIMD ChaCha routine as library function") Fixes: b3aad5bad26a ("crypto: arm64/chacha - expose arm64 ChaCha routine as library function") Fixes: a44a3430d71b ("crypto: arm/chacha - expose ARM ChaCha routine as library function") Fixes: d7d7b8535662 ("crypto: x86/poly1305 - wire up faster implementations for kernel") Fixes: f569ca164751 ("crypto: arm64/poly1305 - incorporate OpenSSL/CRYPTOGAMS NEON implementation") Fixes: a6b803b3ddc7 ("crypto: arm/poly1305 - incorporate OpenSSL/CRYPTOGAMS NEON implementation") Fixes: ed0356eda153 ("crypto: blake2s - x86_64 SIMD implementation") Cc: Eric Biggers <ebiggers@google.com> Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Cc: stable@vger.kernel.org Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com> Reviewed-by: Ard Biesheuvel <ardb@kernel.org> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
2020-04-22 16:18:53 -07:00
src += len;
}
nbytes %= POLY1305_BLOCK_SIZE;
}
if (unlikely(nbytes)) {
dctx->buflen = nbytes;
memcpy(dctx->buf, src, nbytes);
}
}
EXPORT_SYMBOL(poly1305_update_arch);
void poly1305_final_arch(struct poly1305_desc_ctx *dctx, u8 *dst)
{
if (unlikely(dctx->buflen)) {
dctx->buf[dctx->buflen++] = 1;
memset(dctx->buf + dctx->buflen, 0,
POLY1305_BLOCK_SIZE - dctx->buflen);
poly1305_blocks(&dctx->h, dctx->buf, POLY1305_BLOCK_SIZE, 0);
}
poly1305_emit(&dctx->h, dst, dctx->s);
memzero_explicit(dctx, sizeof(*dctx));
}
EXPORT_SYMBOL(poly1305_final_arch);
static int neon_poly1305_final(struct shash_desc *desc, u8 *dst)
{
struct poly1305_desc_ctx *dctx = shash_desc_ctx(desc);
if (unlikely(!dctx->sset))
return -ENOKEY;
poly1305_final_arch(dctx, dst);
return 0;
}
static struct shash_alg neon_poly1305_alg = {
.init = neon_poly1305_init,
.update = neon_poly1305_update,
.final = neon_poly1305_final,
.digestsize = POLY1305_DIGEST_SIZE,
.descsize = sizeof(struct poly1305_desc_ctx),
.base.cra_name = "poly1305",
.base.cra_driver_name = "poly1305-neon",
.base.cra_priority = 200,
.base.cra_blocksize = POLY1305_BLOCK_SIZE,
.base.cra_module = THIS_MODULE,
};
static int __init neon_poly1305_mod_init(void)
{
if (!cpu_have_named_feature(ASIMD))
return 0;
static_branch_enable(&have_neon);
return IS_REACHABLE(CONFIG_CRYPTO_HASH) ?
crypto_register_shash(&neon_poly1305_alg) : 0;
}
static void __exit neon_poly1305_mod_exit(void)
{
if (IS_REACHABLE(CONFIG_CRYPTO_HASH) && cpu_have_named_feature(ASIMD))
crypto_unregister_shash(&neon_poly1305_alg);
}
module_init(neon_poly1305_mod_init);
module_exit(neon_poly1305_mod_exit);
MODULE_DESCRIPTION("Poly1305 transform using NEON instructions");
MODULE_LICENSE("GPL v2");
MODULE_ALIAS_CRYPTO("poly1305");
MODULE_ALIAS_CRYPTO("poly1305-neon");