chacha20: extend ppc64le support to ppc64

This requires fixing an incorrect save of the counter. It is a word value. It happens to work on LE because length is limited to u32. Refactor the constant table to load correctly independent of byte ordering. Add byte order swapping where output needs converted to LE ordering for storage. Change-Id: Ic7e09bd1c769bb77dd6e817f5a8639ba765f4c0f Reviewed-on: https://go-review.googlesource.com/c/crypto/+/614297 Reviewed-by: Cherry Mui <[email protected]> Reviewed-by: Michael Knyszek <[email protected]> Reviewed-by: Archana Ravindar <[email protected]> LUCI-TryBot-Result: Go LUCI <[email protected]>
golang · Oct 11, 2024 · b61b08d · b61b08d
1 parent 6c21748
commit b61b08d
Show file tree

Hide file tree

Showing 3 changed files with 88 additions and 30 deletions.
diff --git a/chacha20/chacha_noasm.go b/chacha20/chacha_noasm.go
@@ -2,7 +2,7 @@
 // Use of this source code is governed by a BSD-style
 // license that can be found in the LICENSE file.
 
-//go:build (!arm64 && !s390x && !ppc64le) || !gc || purego
+//go:build (!arm64 && !s390x && !ppc64 && !ppc64le) || !gc || purego
 
 package chacha20
 

diff --git a/chacha20/chacha_ppc64le.go → chacha20/chacha_ppc64x.go b/chacha20/chacha_ppc64le.go → chacha20/chacha_ppc64x.go
@@ -2,7 +2,7 @@
 // Use of this source code is governed by a BSD-style
 // license that can be found in the LICENSE file.
 
-//go:build gc && !purego
+//go:build gc && !purego && (ppc64 || ppc64le)
 
 package chacha20
 

diff --git a/chacha20/chacha_ppc64le.s → chacha20/chacha_ppc64x.s b/chacha20/chacha_ppc64le.s → chacha20/chacha_ppc64x.s
@@ -19,7 +19,7 @@
 // The differences in this and the original implementation are
 // due to the calling conventions and initialization of constants.
 
-//go:build gc && !purego
+//go:build gc && !purego && (ppc64 || ppc64le)
 
 #include "textflag.h"
 
@@ -36,32 +36,68 @@
 // for VPERMXOR
 #define MASK  R18
 
-DATA consts<>+0x00(SB)/8, $0x3320646e61707865
-DATA consts<>+0x08(SB)/8, $0x6b20657479622d32
-DATA consts<>+0x10(SB)/8, $0x0000000000000001
-DATA consts<>+0x18(SB)/8, $0x0000000000000000
-DATA consts<>+0x20(SB)/8, $0x0000000000000004
-DATA consts<>+0x28(SB)/8, $0x0000000000000000
-DATA consts<>+0x30(SB)/8, $0x0a0b08090e0f0c0d
-DATA consts<>+0x38(SB)/8, $0x0203000106070405
-DATA consts<>+0x40(SB)/8, $0x090a0b080d0e0f0c
-DATA consts<>+0x48(SB)/8, $0x0102030005060704
-DATA consts<>+0x50(SB)/8, $0x6170786561707865
-DATA consts<>+0x58(SB)/8, $0x6170786561707865
-DATA consts<>+0x60(SB)/8, $0x3320646e3320646e
-DATA consts<>+0x68(SB)/8, $0x3320646e3320646e
-DATA consts<>+0x70(SB)/8, $0x79622d3279622d32
-DATA consts<>+0x78(SB)/8, $0x79622d3279622d32
-DATA consts<>+0x80(SB)/8, $0x6b2065746b206574
-DATA consts<>+0x88(SB)/8, $0x6b2065746b206574
-DATA consts<>+0x90(SB)/8, $0x0000000100000000
-DATA consts<>+0x98(SB)/8, $0x0000000300000002
-DATA consts<>+0xa0(SB)/8, $0x5566774411223300
-DATA consts<>+0xa8(SB)/8, $0xddeeffcc99aabb88
-DATA consts<>+0xb0(SB)/8, $0x6677445522330011
-DATA consts<>+0xb8(SB)/8, $0xeeffccddaabb8899
+DATA consts<>+0x00(SB)/4, $0x61707865
+DATA consts<>+0x04(SB)/4, $0x3320646e
+DATA consts<>+0x08(SB)/4, $0x79622d32
+DATA consts<>+0x0c(SB)/4, $0x6b206574
+DATA consts<>+0x10(SB)/4, $0x00000001
+DATA consts<>+0x14(SB)/4, $0x00000000
+DATA consts<>+0x18(SB)/4, $0x00000000
+DATA consts<>+0x1c(SB)/4, $0x00000000
+DATA consts<>+0x20(SB)/4, $0x00000004
+DATA consts<>+0x24(SB)/4, $0x00000000
+DATA consts<>+0x28(SB)/4, $0x00000000
+DATA consts<>+0x2c(SB)/4, $0x00000000
+DATA consts<>+0x30(SB)/4, $0x0e0f0c0d
+DATA consts<>+0x34(SB)/4, $0x0a0b0809
+DATA consts<>+0x38(SB)/4, $0x06070405
+DATA consts<>+0x3c(SB)/4, $0x02030001
+DATA consts<>+0x40(SB)/4, $0x0d0e0f0c
+DATA consts<>+0x44(SB)/4, $0x090a0b08
+DATA consts<>+0x48(SB)/4, $0x05060704
+DATA consts<>+0x4c(SB)/4, $0x01020300
+DATA consts<>+0x50(SB)/4, $0x61707865
+DATA consts<>+0x54(SB)/4, $0x61707865
+DATA consts<>+0x58(SB)/4, $0x61707865
+DATA consts<>+0x5c(SB)/4, $0x61707865
+DATA consts<>+0x60(SB)/4, $0x3320646e
+DATA consts<>+0x64(SB)/4, $0x3320646e
+DATA consts<>+0x68(SB)/4, $0x3320646e
+DATA consts<>+0x6c(SB)/4, $0x3320646e
+DATA consts<>+0x70(SB)/4, $0x79622d32
+DATA consts<>+0x74(SB)/4, $0x79622d32
+DATA consts<>+0x78(SB)/4, $0x79622d32
+DATA consts<>+0x7c(SB)/4, $0x79622d32
+DATA consts<>+0x80(SB)/4, $0x6b206574
+DATA consts<>+0x84(SB)/4, $0x6b206574
+DATA consts<>+0x88(SB)/4, $0x6b206574
+DATA consts<>+0x8c(SB)/4, $0x6b206574
+DATA consts<>+0x90(SB)/4, $0x00000000
+DATA consts<>+0x94(SB)/4, $0x00000001
+DATA consts<>+0x98(SB)/4, $0x00000002
+DATA consts<>+0x9c(SB)/4, $0x00000003
+DATA consts<>+0xa0(SB)/4, $0x11223300
+DATA consts<>+0xa4(SB)/4, $0x55667744
+DATA consts<>+0xa8(SB)/4, $0x99aabb88
+DATA consts<>+0xac(SB)/4, $0xddeeffcc
+DATA consts<>+0xb0(SB)/4, $0x22330011
+DATA consts<>+0xb4(SB)/4, $0x66774455
+DATA consts<>+0xb8(SB)/4, $0xaabb8899
+DATA consts<>+0xbc(SB)/4, $0xeeffccdd
 GLOBL consts<>(SB), RODATA, $0xc0
 
+#ifdef GOARCH_ppc64
+#define BE_XXBRW_INIT() \
+		LVSL (R0)(R0), V24 \
+		VSPLTISB $3, V25   \
+		VXOR V24, V25, V24 \
+
+#define BE_XXBRW(vr) VPERM vr, vr, V24, vr
+#else
+#define BE_XXBRW_INIT()
+#define BE_XXBRW(vr)
+#endif
+
 //func chaCha20_ctr32_vsx(out, inp *byte, len int, key *[8]uint32, counter *uint32)
 TEXT ·chaCha20_ctr32_vsx(SB),NOSPLIT,$64-40
 	MOVD out+0(FP), OUT
@@ -94,6 +130,8 @@ TEXT ·chaCha20_ctr32_vsx(SB),NOSPLIT,$64-40
 	// Clear V27
 	VXOR V27, V27, V27
 
+	BE_XXBRW_INIT()
+
 	// V28
 	LXVW4X (CONSTBASE)(R11), VS60
 
@@ -299,6 +337,11 @@ loop_vsx:
 	VADDUWM V8, V18, V8
 	VADDUWM V12, V19, V12
 
+	BE_XXBRW(V0)
+	BE_XXBRW(V4)
+	BE_XXBRW(V8)
+	BE_XXBRW(V12)
+
 	CMPU LEN, $64
 	BLT tail_vsx
 
@@ -327,15 +370,20 @@ loop_vsx:
 	VADDUWM V9, V18, V8
 	VADDUWM V13, V19, V12
 
+	BE_XXBRW(V0)
+	BE_XXBRW(V4)
+	BE_XXBRW(V8)
+	BE_XXBRW(V12)
+
 	CMPU  LEN, $64
 	BLT   tail_vsx
 
 	LXVW4X (INP)(R0), VS59
 	LXVW4X (INP)(R8), VS60
 	LXVW4X (INP)(R9), VS61
 	LXVW4X (INP)(R10), VS62
-	VXOR   V27, V0, V27
 
+	VXOR V27, V0, V27
 	VXOR V28, V4, V28
 	VXOR V29, V8, V29
 	VXOR V30, V12, V30
@@ -354,6 +402,11 @@ loop_vsx:
 	VADDUWM V10, V18, V8
 	VADDUWM V14, V19, V12
 
+	BE_XXBRW(V0)
+	BE_XXBRW(V4)
+	BE_XXBRW(V8)
+	BE_XXBRW(V12)
+
 	CMPU LEN, $64
 	BLT  tail_vsx
 
@@ -381,6 +434,11 @@ loop_vsx:
 	VADDUWM V11, V18, V8
 	VADDUWM V15, V19, V12
 
+	BE_XXBRW(V0)
+	BE_XXBRW(V4)
+	BE_XXBRW(V8)
+	BE_XXBRW(V12)
+
 	CMPU  LEN, $64
 	BLT   tail_vsx
 
@@ -408,9 +466,9 @@ loop_vsx:
 
 done_vsx:
 	// Increment counter by number of 64 byte blocks
-	MOVD (CNT), R14
+	MOVWZ (CNT), R14
 	ADD  BLOCKS, R14
-	MOVD R14, (CNT)
+	MOVWZ R14, (CNT)
 	RET
 
 tail_vsx: