gecko-dev/mfbt/SHA1.cpp
Chris Peterson 975ea3bc1e Bug 1806779 - Fix C++20 -Wdeprecated-volatile warnings in mfbt/SHA1.cpp. r=glandium
C++20 deprecated decrement/increment of object of volatile-qualified types, e.g. v++.

https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2019/p1152r3.html

mfbt/SHA1.cpp:400:9 [-Wdeprecated-volatile] compound assignment to object of volatile-qualified type 'volatile unsigned int' is deprecated
mfbt/SHA1.cpp:401:9 [-Wdeprecated-volatile] compound assignment to object of volatile-qualified type 'volatile unsigned int' is deprecated
mfbt/SHA1.cpp:402:9 [-Wdeprecated-volatile] compound assignment to object of volatile-qualified type 'volatile unsigned int' is deprecated
mfbt/SHA1.cpp:403:9 [-Wdeprecated-volatile] compound assignment to object of volatile-qualified type 'volatile unsigned int' is deprecated
mfbt/SHA1.cpp:404:9 [-Wdeprecated-volatile] compound assignment to object of volatile-qualified type 'volatile unsigned int' is deprecated

shaCompress() has a comment emphasizing the importance of the X array being volatile. I verified that changing `XH(0) += A` to `XH(0) = XH(0) + A` generates the same -S assembly code (for clang -O2 on Apple Silicon).

Whether this comment about the volatile code generated by gcc 3.4.3 -O3 in 2012 is still relevant for clang 15 -O2 in 2023 is a different question.

Differential Revision: https://phabricator.services.mozilla.com/D165268
2022-12-22 02:58:24 +00:00

406 lines
13 KiB
C++

/* -*- Mode: C++; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/* vim: set ts=8 sts=2 et sw=2 tw=80: */
/* This Source Code Form is subject to the terms of the Mozilla Public
* License, v. 2.0. If a copy of the MPL was not distributed with this
* file, You can obtain one at http://mozilla.org/MPL/2.0/. */
#include "mozilla/Assertions.h"
#include "mozilla/EndianUtils.h"
#include "mozilla/SHA1.h"
#include <string.h>
using mozilla::NativeEndian;
using mozilla::SHA1Sum;
static inline uint32_t SHA_ROTL(uint32_t aT, uint32_t aN) {
MOZ_ASSERT(aN < 32);
return (aT << aN) | (aT >> (32 - aN));
}
static void shaCompress(volatile unsigned* aX, const uint32_t* aBuf);
#define SHA_F1(X, Y, Z) ((((Y) ^ (Z)) & (X)) ^ (Z))
#define SHA_F2(X, Y, Z) ((X) ^ (Y) ^ (Z))
#define SHA_F3(X, Y, Z) (((X) & (Y)) | ((Z) & ((X) | (Y))))
#define SHA_F4(X, Y, Z) ((X) ^ (Y) ^ (Z))
#define SHA_MIX(n, a, b, c) XW(n) = SHA_ROTL(XW(a) ^ XW(b) ^ XW(c) ^ XW(n), 1)
SHA1Sum::SHA1Sum() : mSize(0), mDone(false) {
// Initialize H with constants from FIPS180-1.
mH[0] = 0x67452301L;
mH[1] = 0xefcdab89L;
mH[2] = 0x98badcfeL;
mH[3] = 0x10325476L;
mH[4] = 0xc3d2e1f0L;
}
/*
* Explanation of H array and index values:
*
* The context's H array is actually the concatenation of two arrays
* defined by SHA1, the H array of state variables (5 elements),
* and the W array of intermediate values, of which there are 16 elements.
* The W array starts at H[5], that is W[0] is H[5].
* Although these values are defined as 32-bit values, we use 64-bit
* variables to hold them because the AMD64 stores 64 bit values in
* memory MUCH faster than it stores any smaller values.
*
* Rather than passing the context structure to shaCompress, we pass
* this combined array of H and W values. We do not pass the address
* of the first element of this array, but rather pass the address of an
* element in the middle of the array, element X. Presently X[0] is H[11].
* So we pass the address of H[11] as the address of array X to shaCompress.
* Then shaCompress accesses the members of the array using positive AND
* negative indexes.
*
* Pictorially: (each element is 8 bytes)
* H | H0 H1 H2 H3 H4 W0 W1 W2 W3 W4 W5 W6 W7 W8 W9 Wa Wb Wc Wd We Wf |
* X |-11-10 -9 -8 -7 -6 -5 -4 -3 -2 -1 X0 X1 X2 X3 X4 X5 X6 X7 X8 X9 |
*
* The byte offset from X[0] to any member of H and W is always
* representable in a signed 8-bit value, which will be encoded
* as a single byte offset in the X86-64 instruction set.
* If we didn't pass the address of H[11], and instead passed the
* address of H[0], the offsets to elements H[16] and above would be
* greater than 127, not representable in a signed 8-bit value, and the
* x86-64 instruction set would encode every such offset as a 32-bit
* signed number in each instruction that accessed element H[16] or
* higher. This results in much bigger and slower code.
*/
#define H2X 11 /* X[0] is H[11], and H[0] is X[-11] */
#define W2X 6 /* X[0] is W[6], and W[0] is X[-6] */
/*
* SHA: Add data to context.
*/
void SHA1Sum::update(const void* aData, uint32_t aLen) {
MOZ_ASSERT(!mDone, "SHA1Sum can only be used to compute a single hash.");
const uint8_t* data = static_cast<const uint8_t*>(aData);
if (aLen == 0) {
return;
}
/* Accumulate the byte count. */
unsigned int lenB = static_cast<unsigned int>(mSize) & 63U;
mSize += aLen;
/* Read the data into W and process blocks as they get full. */
unsigned int togo;
if (lenB > 0) {
togo = 64U - lenB;
if (aLen < togo) {
togo = aLen;
}
memcpy(mU.mB + lenB, data, togo);
aLen -= togo;
data += togo;
lenB = (lenB + togo) & 63U;
if (!lenB) {
shaCompress(&mH[H2X], mU.mW);
}
}
while (aLen >= 64U) {
aLen -= 64U;
shaCompress(&mH[H2X], reinterpret_cast<const uint32_t*>(data));
data += 64U;
}
if (aLen > 0) {
memcpy(mU.mB, data, aLen);
}
}
/*
* SHA: Generate hash value
*/
void SHA1Sum::finish(SHA1Sum::Hash& aHashOut) {
MOZ_ASSERT(!mDone, "SHA1Sum can only be used to compute a single hash.");
uint64_t size = mSize;
uint32_t lenB = uint32_t(size) & 63;
static const uint8_t bulk_pad[64] = {
0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0};
/* Pad with a binary 1 (e.g. 0x80), then zeroes, then length in bits. */
update(bulk_pad, (((55 + 64) - lenB) & 63) + 1);
MOZ_ASSERT((uint32_t(mSize) & 63) == 56);
/* Convert size from bytes to bits. */
size <<= 3;
mU.mW[14] = NativeEndian::swapToBigEndian(uint32_t(size >> 32));
mU.mW[15] = NativeEndian::swapToBigEndian(uint32_t(size));
shaCompress(&mH[H2X], mU.mW);
/* Output hash. */
mU.mW[0] = NativeEndian::swapToBigEndian(mH[0]);
mU.mW[1] = NativeEndian::swapToBigEndian(mH[1]);
mU.mW[2] = NativeEndian::swapToBigEndian(mH[2]);
mU.mW[3] = NativeEndian::swapToBigEndian(mH[3]);
mU.mW[4] = NativeEndian::swapToBigEndian(mH[4]);
memcpy(aHashOut, mU.mW, 20);
mDone = true;
}
/*
* SHA: Compression function, unrolled.
*
* Some operations in shaCompress are done as 5 groups of 16 operations.
* Others are done as 4 groups of 20 operations.
* The code below shows that structure.
*
* The functions that compute the new values of the 5 state variables
* A-E are done in 4 groups of 20 operations (or you may also think
* of them as being done in 16 groups of 5 operations). They are
* done by the SHA_RNDx macros below, in the right column.
*
* The functions that set the 16 values of the W array are done in
* 5 groups of 16 operations. The first group is done by the
* LOAD macros below, the latter 4 groups are done by SHA_MIX below,
* in the left column.
*
* gcc's optimizer observes that each member of the W array is assigned
* a value 5 times in this code. It reduces the number of store
* operations done to the W array in the context (that is, in the X array)
* by creating a W array on the stack, and storing the W values there for
* the first 4 groups of operations on W, and storing the values in the
* context's W array only in the fifth group. This is undesirable.
* It is MUCH bigger code than simply using the context's W array, because
* all the offsets to the W array in the stack are 32-bit signed offsets,
* and it is no faster than storing the values in the context's W array.
*
* The original code for sha_fast.c prevented this creation of a separate
* W array in the stack by creating a W array of 80 members, each of
* whose elements is assigned only once. It also separated the computations
* of the W array values and the computations of the values for the 5
* state variables into two separate passes, W's, then A-E's so that the
* second pass could be done all in registers (except for accessing the W
* array) on machines with fewer registers. The method is suboptimal
* for machines with enough registers to do it all in one pass, and it
* necessitates using many instructions with 32-bit offsets.
*
* This code eliminates the separate W array on the stack by a completely
* different means: by declaring the X array volatile. This prevents
* the optimizer from trying to reduce the use of the X array by the
* creation of a MORE expensive W array on the stack. The result is
* that all instructions use signed 8-bit offsets and not 32-bit offsets.
*
* The combination of this code and the -O3 optimizer flag on GCC 3.4.3
* results in code that is 3 times faster than the previous NSS sha_fast
* code on AMD64.
*/
static void shaCompress(volatile unsigned* aX, const uint32_t* aBuf) {
unsigned A, B, C, D, E;
#define XH(n) aX[n - H2X]
#define XW(n) aX[n - W2X]
#define K0 0x5a827999L
#define K1 0x6ed9eba1L
#define K2 0x8f1bbcdcL
#define K3 0xca62c1d6L
#define SHA_RND1(a, b, c, d, e, n) \
a = SHA_ROTL(b, 5) + SHA_F1(c, d, e) + a + XW(n) + K0; \
c = SHA_ROTL(c, 30)
#define SHA_RND2(a, b, c, d, e, n) \
a = SHA_ROTL(b, 5) + SHA_F2(c, d, e) + a + XW(n) + K1; \
c = SHA_ROTL(c, 30)
#define SHA_RND3(a, b, c, d, e, n) \
a = SHA_ROTL(b, 5) + SHA_F3(c, d, e) + a + XW(n) + K2; \
c = SHA_ROTL(c, 30)
#define SHA_RND4(a, b, c, d, e, n) \
a = SHA_ROTL(b, 5) + SHA_F4(c, d, e) + a + XW(n) + K3; \
c = SHA_ROTL(c, 30)
#define LOAD(n) XW(n) = NativeEndian::swapToBigEndian(aBuf[n])
A = XH(0);
B = XH(1);
C = XH(2);
D = XH(3);
E = XH(4);
LOAD(0);
SHA_RND1(E, A, B, C, D, 0);
LOAD(1);
SHA_RND1(D, E, A, B, C, 1);
LOAD(2);
SHA_RND1(C, D, E, A, B, 2);
LOAD(3);
SHA_RND1(B, C, D, E, A, 3);
LOAD(4);
SHA_RND1(A, B, C, D, E, 4);
LOAD(5);
SHA_RND1(E, A, B, C, D, 5);
LOAD(6);
SHA_RND1(D, E, A, B, C, 6);
LOAD(7);
SHA_RND1(C, D, E, A, B, 7);
LOAD(8);
SHA_RND1(B, C, D, E, A, 8);
LOAD(9);
SHA_RND1(A, B, C, D, E, 9);
LOAD(10);
SHA_RND1(E, A, B, C, D, 10);
LOAD(11);
SHA_RND1(D, E, A, B, C, 11);
LOAD(12);
SHA_RND1(C, D, E, A, B, 12);
LOAD(13);
SHA_RND1(B, C, D, E, A, 13);
LOAD(14);
SHA_RND1(A, B, C, D, E, 14);
LOAD(15);
SHA_RND1(E, A, B, C, D, 15);
SHA_MIX(0, 13, 8, 2);
SHA_RND1(D, E, A, B, C, 0);
SHA_MIX(1, 14, 9, 3);
SHA_RND1(C, D, E, A, B, 1);
SHA_MIX(2, 15, 10, 4);
SHA_RND1(B, C, D, E, A, 2);
SHA_MIX(3, 0, 11, 5);
SHA_RND1(A, B, C, D, E, 3);
SHA_MIX(4, 1, 12, 6);
SHA_RND2(E, A, B, C, D, 4);
SHA_MIX(5, 2, 13, 7);
SHA_RND2(D, E, A, B, C, 5);
SHA_MIX(6, 3, 14, 8);
SHA_RND2(C, D, E, A, B, 6);
SHA_MIX(7, 4, 15, 9);
SHA_RND2(B, C, D, E, A, 7);
SHA_MIX(8, 5, 0, 10);
SHA_RND2(A, B, C, D, E, 8);
SHA_MIX(9, 6, 1, 11);
SHA_RND2(E, A, B, C, D, 9);
SHA_MIX(10, 7, 2, 12);
SHA_RND2(D, E, A, B, C, 10);
SHA_MIX(11, 8, 3, 13);
SHA_RND2(C, D, E, A, B, 11);
SHA_MIX(12, 9, 4, 14);
SHA_RND2(B, C, D, E, A, 12);
SHA_MIX(13, 10, 5, 15);
SHA_RND2(A, B, C, D, E, 13);
SHA_MIX(14, 11, 6, 0);
SHA_RND2(E, A, B, C, D, 14);
SHA_MIX(15, 12, 7, 1);
SHA_RND2(D, E, A, B, C, 15);
SHA_MIX(0, 13, 8, 2);
SHA_RND2(C, D, E, A, B, 0);
SHA_MIX(1, 14, 9, 3);
SHA_RND2(B, C, D, E, A, 1);
SHA_MIX(2, 15, 10, 4);
SHA_RND2(A, B, C, D, E, 2);
SHA_MIX(3, 0, 11, 5);
SHA_RND2(E, A, B, C, D, 3);
SHA_MIX(4, 1, 12, 6);
SHA_RND2(D, E, A, B, C, 4);
SHA_MIX(5, 2, 13, 7);
SHA_RND2(C, D, E, A, B, 5);
SHA_MIX(6, 3, 14, 8);
SHA_RND2(B, C, D, E, A, 6);
SHA_MIX(7, 4, 15, 9);
SHA_RND2(A, B, C, D, E, 7);
SHA_MIX(8, 5, 0, 10);
SHA_RND3(E, A, B, C, D, 8);
SHA_MIX(9, 6, 1, 11);
SHA_RND3(D, E, A, B, C, 9);
SHA_MIX(10, 7, 2, 12);
SHA_RND3(C, D, E, A, B, 10);
SHA_MIX(11, 8, 3, 13);
SHA_RND3(B, C, D, E, A, 11);
SHA_MIX(12, 9, 4, 14);
SHA_RND3(A, B, C, D, E, 12);
SHA_MIX(13, 10, 5, 15);
SHA_RND3(E, A, B, C, D, 13);
SHA_MIX(14, 11, 6, 0);
SHA_RND3(D, E, A, B, C, 14);
SHA_MIX(15, 12, 7, 1);
SHA_RND3(C, D, E, A, B, 15);
SHA_MIX(0, 13, 8, 2);
SHA_RND3(B, C, D, E, A, 0);
SHA_MIX(1, 14, 9, 3);
SHA_RND3(A, B, C, D, E, 1);
SHA_MIX(2, 15, 10, 4);
SHA_RND3(E, A, B, C, D, 2);
SHA_MIX(3, 0, 11, 5);
SHA_RND3(D, E, A, B, C, 3);
SHA_MIX(4, 1, 12, 6);
SHA_RND3(C, D, E, A, B, 4);
SHA_MIX(5, 2, 13, 7);
SHA_RND3(B, C, D, E, A, 5);
SHA_MIX(6, 3, 14, 8);
SHA_RND3(A, B, C, D, E, 6);
SHA_MIX(7, 4, 15, 9);
SHA_RND3(E, A, B, C, D, 7);
SHA_MIX(8, 5, 0, 10);
SHA_RND3(D, E, A, B, C, 8);
SHA_MIX(9, 6, 1, 11);
SHA_RND3(C, D, E, A, B, 9);
SHA_MIX(10, 7, 2, 12);
SHA_RND3(B, C, D, E, A, 10);
SHA_MIX(11, 8, 3, 13);
SHA_RND3(A, B, C, D, E, 11);
SHA_MIX(12, 9, 4, 14);
SHA_RND4(E, A, B, C, D, 12);
SHA_MIX(13, 10, 5, 15);
SHA_RND4(D, E, A, B, C, 13);
SHA_MIX(14, 11, 6, 0);
SHA_RND4(C, D, E, A, B, 14);
SHA_MIX(15, 12, 7, 1);
SHA_RND4(B, C, D, E, A, 15);
SHA_MIX(0, 13, 8, 2);
SHA_RND4(A, B, C, D, E, 0);
SHA_MIX(1, 14, 9, 3);
SHA_RND4(E, A, B, C, D, 1);
SHA_MIX(2, 15, 10, 4);
SHA_RND4(D, E, A, B, C, 2);
SHA_MIX(3, 0, 11, 5);
SHA_RND4(C, D, E, A, B, 3);
SHA_MIX(4, 1, 12, 6);
SHA_RND4(B, C, D, E, A, 4);
SHA_MIX(5, 2, 13, 7);
SHA_RND4(A, B, C, D, E, 5);
SHA_MIX(6, 3, 14, 8);
SHA_RND4(E, A, B, C, D, 6);
SHA_MIX(7, 4, 15, 9);
SHA_RND4(D, E, A, B, C, 7);
SHA_MIX(8, 5, 0, 10);
SHA_RND4(C, D, E, A, B, 8);
SHA_MIX(9, 6, 1, 11);
SHA_RND4(B, C, D, E, A, 9);
SHA_MIX(10, 7, 2, 12);
SHA_RND4(A, B, C, D, E, 10);
SHA_MIX(11, 8, 3, 13);
SHA_RND4(E, A, B, C, D, 11);
SHA_MIX(12, 9, 4, 14);
SHA_RND4(D, E, A, B, C, 12);
SHA_MIX(13, 10, 5, 15);
SHA_RND4(C, D, E, A, B, 13);
SHA_MIX(14, 11, 6, 0);
SHA_RND4(B, C, D, E, A, 14);
SHA_MIX(15, 12, 7, 1);
SHA_RND4(A, B, C, D, E, 15);
XH(0) = XH(0) + A;
XH(1) = XH(1) + B;
XH(2) = XH(2) + C;
XH(3) = XH(3) + D;
XH(4) = XH(4) + E;
}