ExternalVendorCode/openwrt
1
0
Fork 0
mirror of https://github.com/openwrt/openwrt.git synced 2024-12-26 17:01:14 +00:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity
7ae0f74bcd
openwrt/target/linux/generic/backport-5.4/080-wireguard-0014-crypto-x86-poly1305-unify-Poly1305-state-struct-with.patch
Jason A. Donenfeld c0cb86e1d5 kernel: 5.4: import wireguard backport 
Rather than using the clunky, old, slower wireguard-linux-compat out of
tree module, this commit does a patch-by-patch backport of upstream's
wireguard to 5.4. This specific backport is in widespread use, being
part of SUSE's enterprise kernel, Oracle's enterprise kernel, Google's
Android kernel, Gentoo's distro kernel, and probably more I've forgotten
about. It's definately the "more proper" way of adding wireguard to a
kernel than the ugly compat.h hell of the wireguard-linux-compat repo.
And most importantly for OpenWRT, it allows using the same module
configuration code for 5.10 as for 5.4, with no need for bifurcation.

These patches are from the backport tree which is maintained in the
open here: https://git.zx2c4.com/wireguard-linux/log/?h=backport-5.4.y
I'll be sending PRs to update this as needed.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
(cherry picked from commit 3888fa7880)
(cherry picked from commit d540725871)
(cherry picked from commit 196f3d586f)
(cherry picked from commit 3500fd7938)
(cherry picked from commit 23b801d3ba)
(cherry picked from commit 0c0cb97da7)
(cherry picked from commit 2a27f6f90a)
Signed-off-by: Ilya Lipnitskiy <ilya.lipnitskiy@gmail.com>
2021-04-10 14:21:32 +02:00

252 lines
7.8 KiB
Diff
Raw Blame History

From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Ard Biesheuvel <ardb@kernel.org>
Date: Fri, 8 Nov 2019 13:22:20 +0100
Subject: [PATCH] crypto: x86/poly1305 - unify Poly1305 state struct with
generic code
commit ad8f5b88383ea685f2b8df2a12ee3e08089a1287 upstream.
In preparation of exposing a Poly1305 library interface directly from
the accelerated x86 driver, align the state descriptor of the x86 code
with the one used by the generic driver. This is needed to make the
library interface unified between all implementations.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
---
arch/x86/crypto/poly1305_glue.c | 88 ++++++++++--------------------
crypto/poly1305_generic.c | 6 +-
include/crypto/internal/poly1305.h | 4 +-
include/crypto/poly1305.h | 18 +++---
4 files changed, 43 insertions(+), 73 deletions(-)
--- a/arch/x86/crypto/poly1305_glue.c
+++ b/arch/x86/crypto/poly1305_glue.c
@@ -14,40 +14,14 @@
#include <linux/module.h>
#include <asm/simd.h>
-struct poly1305_simd_desc_ctx {
- struct poly1305_desc_ctx base;
- /* derived key u set? */
- bool uset;
-#ifdef CONFIG_AS_AVX2
- /* derived keys r^3, r^4 set? */
- bool wset;
-#endif
- /* derived Poly1305 key r^2 */
- u32 u[5];
- /* ... silently appended r^3 and r^4 when using AVX2 */
-};
-
asmlinkage void poly1305_block_sse2(u32 *h, const u8 *src,
const u32 *r, unsigned int blocks);
asmlinkage void poly1305_2block_sse2(u32 *h, const u8 *src, const u32 *r,
unsigned int blocks, const u32 *u);
-#ifdef CONFIG_AS_AVX2
asmlinkage void poly1305_4block_avx2(u32 *h, const u8 *src, const u32 *r,
unsigned int blocks, const u32 *u);
-static bool poly1305_use_avx2;
-#endif
-static int poly1305_simd_init(struct shash_desc *desc)
-{
- struct poly1305_simd_desc_ctx *sctx = shash_desc_ctx(desc);
-
- sctx->uset = false;
-#ifdef CONFIG_AS_AVX2
- sctx->wset = false;
-#endif
-
- return crypto_poly1305_init(desc);
-}
+static bool poly1305_use_avx2 __ro_after_init;
static void poly1305_simd_mult(u32 *a, const u32 *b)
{
@@ -63,53 +37,49 @@ static void poly1305_simd_mult(u32 *a, c
static unsigned int poly1305_simd_blocks(struct poly1305_desc_ctx *dctx,
const u8 *src, unsigned int srclen)
{
- struct poly1305_simd_desc_ctx *sctx;
unsigned int blocks, datalen;
- BUILD_BUG_ON(offsetof(struct poly1305_simd_desc_ctx, base));
- sctx = container_of(dctx, struct poly1305_simd_desc_ctx, base);
-
if (unlikely(!dctx->sset)) {
datalen = crypto_poly1305_setdesckey(dctx, src, srclen);
src += srclen - datalen;
srclen = datalen;
}
-#ifdef CONFIG_AS_AVX2
- if (poly1305_use_avx2 && srclen >= POLY1305_BLOCK_SIZE * 4) {
- if (unlikely(!sctx->wset)) {
- if (!sctx->uset) {
- memcpy(sctx->u, dctx->r.r, sizeof(sctx->u));
- poly1305_simd_mult(sctx->u, dctx->r.r);
- sctx->uset = true;
+ if (IS_ENABLED(CONFIG_AS_AVX2) &&
+ poly1305_use_avx2 &&
+ srclen >= POLY1305_BLOCK_SIZE * 4) {
+ if (unlikely(dctx->rset < 4)) {
+ if (dctx->rset < 2) {
+ dctx->r[1] = dctx->r[0];
+ poly1305_simd_mult(dctx->r[1].r, dctx->r[0].r);
}
- memcpy(sctx->u + 5, sctx->u, sizeof(sctx->u));
- poly1305_simd_mult(sctx->u + 5, dctx->r.r);
- memcpy(sctx->u + 10, sctx->u + 5, sizeof(sctx->u));
- poly1305_simd_mult(sctx->u + 10, dctx->r.r);
- sctx->wset = true;
+ dctx->r[2] = dctx->r[1];
+ poly1305_simd_mult(dctx->r[2].r, dctx->r[0].r);
+ dctx->r[3] = dctx->r[2];
+ poly1305_simd_mult(dctx->r[3].r, dctx->r[0].r);
+ dctx->rset = 4;
}
blocks = srclen / (POLY1305_BLOCK_SIZE * 4);
- poly1305_4block_avx2(dctx->h.h, src, dctx->r.r, blocks,
- sctx->u);
+ poly1305_4block_avx2(dctx->h.h, src, dctx->r[0].r, blocks,
+ dctx->r[1].r);
src += POLY1305_BLOCK_SIZE * 4 * blocks;
srclen -= POLY1305_BLOCK_SIZE * 4 * blocks;
}
-#endif
+
if (likely(srclen >= POLY1305_BLOCK_SIZE * 2)) {
- if (unlikely(!sctx->uset)) {
- memcpy(sctx->u, dctx->r.r, sizeof(sctx->u));
- poly1305_simd_mult(sctx->u, dctx->r.r);
- sctx->uset = true;
+ if (unlikely(dctx->rset < 2)) {
+ dctx->r[1] = dctx->r[0];
+ poly1305_simd_mult(dctx->r[1].r, dctx->r[0].r);
+ dctx->rset = 2;
}
blocks = srclen / (POLY1305_BLOCK_SIZE * 2);
- poly1305_2block_sse2(dctx->h.h, src, dctx->r.r, blocks,
- sctx->u);
+ poly1305_2block_sse2(dctx->h.h, src, dctx->r[0].r,
+ blocks, dctx->r[1].r);
src += POLY1305_BLOCK_SIZE * 2 * blocks;
srclen -= POLY1305_BLOCK_SIZE * 2 * blocks;
}
if (srclen >= POLY1305_BLOCK_SIZE) {
- poly1305_block_sse2(dctx->h.h, src, dctx->r.r, 1);
+ poly1305_block_sse2(dctx->h.h, src, dctx->r[0].r, 1);
srclen -= POLY1305_BLOCK_SIZE;
}
return srclen;
@@ -159,10 +129,10 @@ static int poly1305_simd_update(struct s
static struct shash_alg alg = {
.digestsize = POLY1305_DIGEST_SIZE,
- .init = poly1305_simd_init,
+ .init = crypto_poly1305_init,
.update = poly1305_simd_update,
.final = crypto_poly1305_final,
- .descsize = sizeof(struct poly1305_simd_desc_ctx),
+ .descsize = sizeof(struct poly1305_desc_ctx),
.base = {
.cra_name = "poly1305",
.cra_driver_name = "poly1305-simd",
@@ -177,14 +147,14 @@ static int __init poly1305_simd_mod_init
if (!boot_cpu_has(X86_FEATURE_XMM2))
return -ENODEV;
-#ifdef CONFIG_AS_AVX2
- poly1305_use_avx2 = boot_cpu_has(X86_FEATURE_AVX) &&
+ poly1305_use_avx2 = IS_ENABLED(CONFIG_AS_AVX2) &&
+ boot_cpu_has(X86_FEATURE_AVX) &&
boot_cpu_has(X86_FEATURE_AVX2) &&
cpu_has_xfeatures(XFEATURE_MASK_SSE | XFEATURE_MASK_YMM, NULL);
- alg.descsize = sizeof(struct poly1305_simd_desc_ctx);
+ alg.descsize = sizeof(struct poly1305_desc_ctx) + 5 * sizeof(u32);
if (poly1305_use_avx2)
alg.descsize += 10 * sizeof(u32);
-#endif
+
return crypto_register_shash(&alg);
}
--- a/crypto/poly1305_generic.c
+++ b/crypto/poly1305_generic.c
@@ -25,7 +25,7 @@ int crypto_poly1305_init(struct shash_de
poly1305_core_init(&dctx->h);
dctx->buflen = 0;
- dctx->rset = false;
+ dctx->rset = 0;
dctx->sset = false;
return 0;
@@ -43,7 +43,7 @@ static void poly1305_blocks(struct poly1
srclen = datalen;
}
- poly1305_core_blocks(&dctx->h, &dctx->r, src,
+ poly1305_core_blocks(&dctx->h, dctx->r, src,
srclen / POLY1305_BLOCK_SIZE, 1);
}
@@ -95,7 +95,7 @@ int crypto_poly1305_final(struct shash_d
dctx->buf[dctx->buflen++] = 1;
memset(dctx->buf + dctx->buflen, 0,
POLY1305_BLOCK_SIZE - dctx->buflen);
- poly1305_core_blocks(&dctx->h, &dctx->r, dctx->buf, 1, 0);
+ poly1305_core_blocks(&dctx->h, dctx->r, dctx->buf, 1, 0);
}
poly1305_core_emit(&dctx->h, digest);
--- a/include/crypto/internal/poly1305.h
+++ b/include/crypto/internal/poly1305.h
@@ -46,10 +46,10 @@ unsigned int crypto_poly1305_setdesckey(
{
if (!dctx->sset) {
if (!dctx->rset && srclen >= POLY1305_BLOCK_SIZE) {
- poly1305_core_setkey(&dctx->r, src);
+ poly1305_core_setkey(dctx->r, src);
src += POLY1305_BLOCK_SIZE;
srclen -= POLY1305_BLOCK_SIZE;
- dctx->rset = true;
+ dctx->rset = 1;
}
if (srclen >= POLY1305_BLOCK_SIZE) {
dctx->s[0] = get_unaligned_le32(src + 0);
--- a/include/crypto/poly1305.h
+++ b/include/crypto/poly1305.h
@@ -22,20 +22,20 @@ struct poly1305_state {
};
struct poly1305_desc_ctx {
- /* key */
- struct poly1305_key r;
- /* finalize key */
- u32 s[4];
- /* accumulator */
- struct poly1305_state h;
/* partial buffer */
u8 buf[POLY1305_BLOCK_SIZE];
/* bytes used in partial buffer */
unsigned int buflen;
- /* r key has been set */
- bool rset;
- /* s key has been set */
+ /* how many keys have been set in r[] */
+ unsigned short rset;
+ /* whether s[] has been set */
bool sset;
+ /* finalize key */
+ u32 s[4];
+ /* accumulator */
+ struct poly1305_state h;
+ /* key */
+ struct poly1305_key r[1];
};
#endif
Reference in New Issue View Git Blame Copy Permalink