* [PATCH v2 09/17] zinc: Poly1305 x86_64 implementation
From: Jason A. Donenfeld @ 2018-08-24 21:38 UTC (permalink / raw)
To: linux-kernel, netdev, davem
Cc: Jason A. Donenfeld, Andy Lutomirski, Greg KH, Samuel Neves,
Jean-Philippe Aumasson, Andy Polyakov, Thomas Gleixner,
Ingo Molnar, x86, linux-crypto
In-Reply-To: <20180824213849.23647-1-Jason@zx2c4.com>
This provides AVX, AVX-2, and AVX-512F implementations for Poly1305.
The AVX-512F implementation is disabled on Skylake, due to throttling.
These come from Andy Polyakov's implementation, with some heavy
modifications from Samuel Neves and me.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Greg KH <gregkh@linuxfoundation.org>
Cc: Samuel Neves <sneves@dei.uc.pt>
Cc: Jean-Philippe Aumasson <jeanphilippe.aumasson@gmail.com>
Cc: Andy Polyakov <appro@openssl.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: x86@kernel.org
Cc: linux-crypto@vger.kernel.org
---
lib/zinc/Makefile | 4 +
lib/zinc/poly1305/poly1305-x86_64-glue.h | 91 +
lib/zinc/poly1305/poly1305-x86_64.S | 2790 ++++++++++++++++++++++
3 files changed, 2885 insertions(+)
create mode 100644 lib/zinc/poly1305/poly1305-x86_64-glue.h
create mode 100644 lib/zinc/poly1305/poly1305-x86_64.S
diff --git a/lib/zinc/Makefile b/lib/zinc/Makefile
index db70ffedf7dd..57fe42b7f97c 100644
--- a/lib/zinc/Makefile
+++ b/lib/zinc/Makefile
@@ -24,6 +24,10 @@ endif
ifeq ($(CONFIG_ZINC_POLY1305),y)
zinc-y += poly1305/poly1305.o
+ifeq ($(CONFIG_X86_64),y)
+zinc-y += poly1305/poly1305-x86_64.o
+CFLAGS_poly1305.o += -include $(srctree)/$(src)/poly1305/poly1305-x86_64-glue.h
+endif
ifeq ($(CONFIG_ARM),y)
zinc-y += poly1305/poly1305-arm.o
CFLAGS_poly1305.o += -include $(srctree)/$(src)/poly1305/poly1305-arm-glue.h
diff --git a/lib/zinc/poly1305/poly1305-x86_64-glue.h b/lib/zinc/poly1305/poly1305-x86_64-glue.h
new file mode 100644
index 000000000000..dde3b0ec39d5
--- /dev/null
+++ b/lib/zinc/poly1305/poly1305-x86_64-glue.h
@@ -0,0 +1,91 @@
+/* SPDX-License-Identifier: GPL-2.0
+ *
+ * Copyright (C) 2015-2018 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include <zinc/poly1305.h>
+#include <asm/fpu/api.h>
+#include <asm/cpufeature.h>
+#include <asm/processor.h>
+#include <asm/intel-family.h>
+
+asmlinkage void poly1305_init_x86_64(void *ctx, const u8 key[POLY1305_KEY_SIZE]);
+asmlinkage void poly1305_blocks_x86_64(void *ctx, const u8 *inp, const size_t len, const u32 padbit);
+asmlinkage void poly1305_emit_x86_64(void *ctx, u8 mac[POLY1305_MAC_SIZE], const u32 nonce[4]);
+#ifdef CONFIG_AS_AVX
+asmlinkage void poly1305_emit_avx(void *ctx, u8 mac[POLY1305_MAC_SIZE], const u32 nonce[4]);
+asmlinkage void poly1305_blocks_avx(void *ctx, const u8 *inp, const size_t len, const u32 padbit);
+#endif
+#ifdef CONFIG_AS_AVX2
+asmlinkage void poly1305_blocks_avx2(void *ctx, const u8 *inp, const size_t len, const u32 padbit);
+#endif
+#ifdef CONFIG_AS_AVX512
+asmlinkage void poly1305_blocks_avx512(void *ctx, const u8 *inp, const size_t len, const u32 padbit);
+#endif
+
+static bool poly1305_use_avx __ro_after_init;
+static bool poly1305_use_avx2 __ro_after_init;
+static bool poly1305_use_avx512 __ro_after_init;
+
+void __init poly1305_fpu_init(void)
+{
+#ifndef CONFIG_UML
+ poly1305_use_avx = boot_cpu_has(X86_FEATURE_AVX) &&
+ cpu_has_xfeatures(XFEATURE_MASK_SSE | XFEATURE_MASK_YMM, NULL);
+ poly1305_use_avx2 = boot_cpu_has(X86_FEATURE_AVX) && boot_cpu_has(X86_FEATURE_AVX2) &&
+ cpu_has_xfeatures(XFEATURE_MASK_SSE | XFEATURE_MASK_YMM, NULL);
+ poly1305_use_avx512 = boot_cpu_has(X86_FEATURE_AVX) && boot_cpu_has(X86_FEATURE_AVX2) && boot_cpu_has(X86_FEATURE_AVX512F) &&
+ cpu_has_xfeatures(XFEATURE_MASK_SSE | XFEATURE_MASK_YMM | XFEATURE_MASK_AVX512, NULL) &&
+ boot_cpu_data.x86_model != INTEL_FAM6_SKYLAKE_X;
+#endif
+}
+
+static inline bool poly1305_init_arch(void *ctx, const u8 key[POLY1305_KEY_SIZE], simd_context_t simd_context)
+{
+ poly1305_init_x86_64(ctx, key);
+ return true;
+}
+
+static inline bool poly1305_blocks_arch(void *ctx, const u8 *inp, const size_t len, const u32 padbit, simd_context_t simd_context)
+{
+#ifdef CONFIG_AS_AVX512
+ if (poly1305_use_avx512 && simd_context == HAVE_FULL_SIMD)
+ poly1305_blocks_avx512(ctx, inp, len, padbit);
+ else
+#endif
+#ifdef CONFIG_AS_AVX2
+ if (poly1305_use_avx2 && simd_context == HAVE_FULL_SIMD)
+ poly1305_blocks_avx2(ctx, inp, len, padbit);
+ else
+#endif
+#ifdef CONFIG_AS_AVX
+ if (poly1305_use_avx && simd_context == HAVE_FULL_SIMD)
+ poly1305_blocks_avx(ctx, inp, len, padbit);
+ else
+#endif
+ poly1305_blocks_x86_64(ctx, inp, len, padbit);
+ return true;
+}
+
+static inline bool poly1305_emit_arch(void *ctx, u8 mac[POLY1305_MAC_SIZE], const u32 nonce[4], simd_context_t simd_context)
+{
+#ifdef CONFIG_AS_AVX512
+ if (poly1305_use_avx512 && simd_context == HAVE_FULL_SIMD)
+ poly1305_emit_avx(ctx, mac, nonce);
+ else
+#endif
+#ifdef CONFIG_AS_AVX2
+ if (poly1305_use_avx2 && simd_context == HAVE_FULL_SIMD)
+ poly1305_emit_avx(ctx, mac, nonce);
+ else
+#endif
+#ifdef CONFIG_AS_AVX
+ if (poly1305_use_avx && simd_context == HAVE_FULL_SIMD)
+ poly1305_emit_avx(ctx, mac, nonce);
+ else
+#endif
+ poly1305_emit_x86_64(ctx, mac, nonce);
+ return true;
+}
+
+#define HAVE_POLY1305_ARCH_IMPLEMENTATION
diff --git a/lib/zinc/poly1305/poly1305-x86_64.S b/lib/zinc/poly1305/poly1305-x86_64.S
new file mode 100644
index 000000000000..7c1b7363f487
--- /dev/null
+++ b/lib/zinc/poly1305/poly1305-x86_64.S
@@ -0,0 +1,2790 @@
+/* SPDX-License-Identifier: OpenSSL OR (BSD-3-Clause OR GPL-2.0)
+ *
+ * Copyright (C) 2017 Samuel Neves <sneves@dei.uc.pt>. All Rights Reserved.
+ * Copyright (C) 2015-2018 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ * Copyright 2016 The OpenSSL Project Authors. All Rights Reserved.
+ */
+
+#include <linux/linkage.h>
+
+.section .rodata.cst192.Lconst, "aM", @progbits, 192
+.align 64
+.Lconst:
+.long 0x0ffffff,0,0x0ffffff,0,0x0ffffff,0,0x0ffffff,0
+.long 16777216,0,16777216,0,16777216,0,16777216,0
+.long 0x3ffffff,0,0x3ffffff,0,0x3ffffff,0,0x3ffffff,0
+.long 2,2,2,3,2,0,2,1
+.long 0,0,0,1, 0,2,0,3, 0,4,0,5, 0,6,0,7
+
+.text
+
+.align 32
+ENTRY(poly1305_init_x86_64)
+ xorq %rax,%rax
+ movq %rax,0(%rdi)
+ movq %rax,8(%rdi)
+ movq %rax,16(%rdi)
+
+ cmpq $0,%rsi
+ je .Lno_key
+
+ movq $0x0ffffffc0fffffff,%rax
+ movq $0x0ffffffc0ffffffc,%rcx
+ andq 0(%rsi),%rax
+ andq 8(%rsi),%rcx
+ movq %rax,24(%rdi)
+ movq %rcx,32(%rdi)
+ movl $1,%eax
+.Lno_key:
+ ret
+ENDPROC(poly1305_init_x86_64)
+
+.align 32
+ENTRY(poly1305_blocks_x86_64)
+.Lblocks:
+ shrq $4,%rdx
+ jz .Lno_data
+
+ pushq %rbx
+ pushq %r12
+ pushq %r13
+ pushq %r14
+ pushq %r15
+ pushq %rdi
+
+.Lblocks_body:
+
+ movq %rdx,%r15
+
+ movq 24(%rdi),%r11
+ movq 32(%rdi),%r13
+
+ movq 0(%rdi),%r14
+ movq 8(%rdi),%rbx
+ movq 16(%rdi),%r10
+
+ movq %r13,%r12
+ shrq $2,%r13
+ movq %r12,%rax
+ addq %r12,%r13
+ jmp .Loop
+
+.align 32
+.Loop:
+
+ addq 0(%rsi),%r14
+ adcq 8(%rsi),%rbx
+ leaq 16(%rsi),%rsi
+ adcq %rcx,%r10
+ mulq %r14
+ movq %rax,%r9
+ movq %r11,%rax
+ movq %rdx,%rdi
+
+ mulq %r14
+ movq %rax,%r14
+ movq %r11,%rax
+ movq %rdx,%r8
+
+ mulq %rbx
+ addq %rax,%r9
+ movq %r13,%rax
+ adcq %rdx,%rdi
+
+ mulq %rbx
+ movq %r10,%rbx
+ addq %rax,%r14
+ adcq %rdx,%r8
+
+ imulq %r13,%rbx
+ addq %rbx,%r9
+ movq %r8,%rbx
+ adcq $0,%rdi
+
+ imulq %r11,%r10
+ addq %r9,%rbx
+ movq $-4,%rax
+ adcq %r10,%rdi
+
+ andq %rdi,%rax
+ movq %rdi,%r10
+ shrq $2,%rdi
+ andq $3,%r10
+ addq %rdi,%rax
+ addq %rax,%r14
+ adcq $0,%rbx
+ adcq $0,%r10
+
+ movq %r12,%rax
+ decq %r15
+ jnz .Loop
+
+ movq 0(%rsp),%rdi
+
+ movq %r14,0(%rdi)
+ movq %rbx,8(%rdi)
+ movq %r10,16(%rdi)
+
+ movq 8(%rsp),%r15
+ movq 16(%rsp),%r14
+ movq 24(%rsp),%r13
+ movq 32(%rsp),%r12
+ movq 40(%rsp),%rbx
+ leaq 48(%rsp),%rsp
+.Lno_data:
+.Lblocks_epilogue:
+ ret
+ENDPROC(poly1305_blocks_x86_64)
+
+.align 32
+ENTRY(poly1305_emit_x86_64)
+.Lemit:
+ movq 0(%rdi),%r8
+ movq 8(%rdi),%r9
+ movq 16(%rdi),%r10
+
+ movq %r8,%rax
+ addq $5,%r8
+ movq %r9,%rcx
+ adcq $0,%r9
+ adcq $0,%r10
+ shrq $2,%r10
+ cmovnzq %r8,%rax
+ cmovnzq %r9,%rcx
+
+ addq 0(%rdx),%rax
+ adcq 8(%rdx),%rcx
+ movq %rax,0(%rsi)
+ movq %rcx,8(%rsi)
+
+ ret
+ENDPROC(poly1305_emit_x86_64)
+
+.macro __poly1305_block
+ mulq %r14
+ movq %rax,%r9
+ movq %r11,%rax
+ movq %rdx,%rdi
+
+ mulq %r14
+ movq %rax,%r14
+ movq %r11,%rax
+ movq %rdx,%r8
+
+ mulq %rbx
+ addq %rax,%r9
+ movq %r13,%rax
+ adcq %rdx,%rdi
+
+ mulq %rbx
+ movq %r10,%rbx
+ addq %rax,%r14
+ adcq %rdx,%r8
+
+ imulq %r13,%rbx
+ addq %rbx,%r9
+ movq %r8,%rbx
+ adcq $0,%rdi
+
+ imulq %r11,%r10
+ addq %r9,%rbx
+ movq $-4,%rax
+ adcq %r10,%rdi
+
+ andq %rdi,%rax
+ movq %rdi,%r10
+ shrq $2,%rdi
+ andq $3,%r10
+ addq %rdi,%rax
+ addq %rax,%r14
+ adcq $0,%rbx
+ adcq $0,%r10
+.endm
+
+.macro __poly1305_init_avx
+ movq %r11,%r14
+ movq %r12,%rbx
+ xorq %r10,%r10
+
+ leaq 48+64(%rdi),%rdi
+
+ movq %r12,%rax
+ movq %rdi,0(%rsp)
+ __poly1305_block
+ movq 0(%rsp),%rdi
+
+ movl $0x3ffffff,%eax
+ movl $0x3ffffff,%edx
+ movq %r14,%r8
+ andl %r14d,%eax
+ movq %r11,%r9
+ andl %r11d,%edx
+ movl %eax,-64(%rdi)
+ shrq $26,%r8
+ movl %edx,-60(%rdi)
+ shrq $26,%r9
+
+ movl $0x3ffffff,%eax
+ movl $0x3ffffff,%edx
+ andl %r8d,%eax
+ andl %r9d,%edx
+ movl %eax,-48(%rdi)
+ leal (%rax,%rax,4),%eax
+ movl %edx,-44(%rdi)
+ leal (%rdx,%rdx,4),%edx
+ movl %eax,-32(%rdi)
+ shrq $26,%r8
+ movl %edx,-28(%rdi)
+ shrq $26,%r9
+
+ movq %rbx,%rax
+ movq %r12,%rdx
+ shlq $12,%rax
+ shlq $12,%rdx
+ orq %r8,%rax
+ orq %r9,%rdx
+ andl $0x3ffffff,%eax
+ andl $0x3ffffff,%edx
+ movl %eax,-16(%rdi)
+ leal (%rax,%rax,4),%eax
+ movl %edx,-12(%rdi)
+ leal (%rdx,%rdx,4),%edx
+ movl %eax,0(%rdi)
+ movq %rbx,%r8
+ movl %edx,4(%rdi)
+ movq %r12,%r9
+
+ movl $0x3ffffff,%eax
+ movl $0x3ffffff,%edx
+ shrq $14,%r8
+ shrq $14,%r9
+ andl %r8d,%eax
+ andl %r9d,%edx
+ movl %eax,16(%rdi)
+ leal (%rax,%rax,4),%eax
+ movl %edx,20(%rdi)
+ leal (%rdx,%rdx,4),%edx
+ movl %eax,32(%rdi)
+ shrq $26,%r8
+ movl %edx,36(%rdi)
+ shrq $26,%r9
+
+ movq %r10,%rax
+ shlq $24,%rax
+ orq %rax,%r8
+ movl %r8d,48(%rdi)
+ leaq (%r8,%r8,4),%r8
+ movl %r9d,52(%rdi)
+ leaq (%r9,%r9,4),%r9
+ movl %r8d,64(%rdi)
+ movl %r9d,68(%rdi)
+
+ movq %r12,%rax
+ movq %rdi,0(%rsp)
+ __poly1305_block
+ movq 0(%rsp),%rdi
+
+ movl $0x3ffffff,%eax
+ movq %r14,%r8
+ andl %r14d,%eax
+ shrq $26,%r8
+ movl %eax,-52(%rdi)
+
+ movl $0x3ffffff,%edx
+ andl %r8d,%edx
+ movl %edx,-36(%rdi)
+ leal (%rdx,%rdx,4),%edx
+ shrq $26,%r8
+ movl %edx,-20(%rdi)
+
+ movq %rbx,%rax
+ shlq $12,%rax
+ orq %r8,%rax
+ andl $0x3ffffff,%eax
+ movl %eax,-4(%rdi)
+ leal (%rax,%rax,4),%eax
+ movq %rbx,%r8
+ movl %eax,12(%rdi)
+
+ movl $0x3ffffff,%edx
+ shrq $14,%r8
+ andl %r8d,%edx
+ movl %edx,28(%rdi)
+ leal (%rdx,%rdx,4),%edx
+ shrq $26,%r8
+ movl %edx,44(%rdi)
+
+ movq %r10,%rax
+ shlq $24,%rax
+ orq %rax,%r8
+ movl %r8d,60(%rdi)
+ leaq (%r8,%r8,4),%r8
+ movl %r8d,76(%rdi)
+
+ movq %r12,%rax
+ movq %rdi,0(%rsp)
+ __poly1305_block
+ movq 0(%rsp),%rdi
+
+ movl $0x3ffffff,%eax
+ movq %r14,%r8
+ andl %r14d,%eax
+ shrq $26,%r8
+ movl %eax,-56(%rdi)
+
+ movl $0x3ffffff,%edx
+ andl %r8d,%edx
+ movl %edx,-40(%rdi)
+ leal (%rdx,%rdx,4),%edx
+ shrq $26,%r8
+ movl %edx,-24(%rdi)
+
+ movq %rbx,%rax
+ shlq $12,%rax
+ orq %r8,%rax
+ andl $0x3ffffff,%eax
+ movl %eax,-8(%rdi)
+ leal (%rax,%rax,4),%eax
+ movq %rbx,%r8
+ movl %eax,8(%rdi)
+
+ movl $0x3ffffff,%edx
+ shrq $14,%r8
+ andl %r8d,%edx
+ movl %edx,24(%rdi)
+ leal (%rdx,%rdx,4),%edx
+ shrq $26,%r8
+ movl %edx,40(%rdi)
+
+ movq %r10,%rax
+ shlq $24,%rax
+ orq %rax,%r8
+ movl %r8d,56(%rdi)
+ leaq (%r8,%r8,4),%r8
+ movl %r8d,72(%rdi)
+
+ leaq -48-64(%rdi),%rdi
+.endm
+
+#ifdef CONFIG_AS_AVX
+.align 32
+ENTRY(poly1305_blocks_avx)
+
+ movl 20(%rdi),%r8d
+ cmpq $128,%rdx
+ jae .Lblocks_avx
+ testl %r8d,%r8d
+ jz .Lblocks
+
+.Lblocks_avx:
+ andq $-16,%rdx
+ jz .Lno_data_avx
+
+ vzeroupper
+
+ testl %r8d,%r8d
+ jz .Lbase2_64_avx
+
+ testq $31,%rdx
+ jz .Leven_avx
+
+ pushq %rbx
+ pushq %r12
+ pushq %r13
+ pushq %r14
+ pushq %r15
+ pushq %rdi
+
+.Lblocks_avx_body:
+
+ movq %rdx,%r15
+
+ movq 0(%rdi),%r8
+ movq 8(%rdi),%r9
+ movl 16(%rdi),%r10d
+
+ movq 24(%rdi),%r11
+ movq 32(%rdi),%r13
+
+
+ movl %r8d,%r14d
+ andq $-2147483648,%r8
+ movq %r9,%r12
+ movl %r9d,%ebx
+ andq $-2147483648,%r9
+
+ shrq $6,%r8
+ shlq $52,%r12
+ addq %r8,%r14
+ shrq $12,%rbx
+ shrq $18,%r9
+ addq %r12,%r14
+ adcq %r9,%rbx
+
+ movq %r10,%r8
+ shlq $40,%r8
+ shrq $24,%r10
+ addq %r8,%rbx
+ adcq $0,%r10
+
+ movq $-4,%r9
+ movq %r10,%r8
+ andq %r10,%r9
+ shrq $2,%r8
+ andq $3,%r10
+ addq %r9,%r8
+ addq %r8,%r14
+ adcq $0,%rbx
+ adcq $0,%r10
+
+ movq %r13,%r12
+ movq %r13,%rax
+ shrq $2,%r13
+ addq %r12,%r13
+
+ addq 0(%rsi),%r14
+ adcq 8(%rsi),%rbx
+ leaq 16(%rsi),%rsi
+ adcq %rcx,%r10
+
+ movq %rdi,0(%rsp)
+ __poly1305_block
+ movq 0(%rsp),%rdi
+
+ testq %rcx,%rcx
+ jz .Lstore_base2_64_avx
+
+
+ movq %r14,%rax
+ movq %r14,%rdx
+ shrq $52,%r14
+ movq %rbx,%r11
+ movq %rbx,%r12
+ shrq $26,%rdx
+ andq $0x3ffffff,%rax
+ shlq $12,%r11
+ andq $0x3ffffff,%rdx
+ shrq $14,%rbx
+ orq %r11,%r14
+ shlq $24,%r10
+ andq $0x3ffffff,%r14
+ shrq $40,%r12
+ andq $0x3ffffff,%rbx
+ orq %r12,%r10
+
+ subq $16,%r15
+ jz .Lstore_base2_26_avx
+
+ vmovd %eax,%xmm0
+ vmovd %edx,%xmm1
+ vmovd %r14d,%xmm2
+ vmovd %ebx,%xmm3
+ vmovd %r10d,%xmm4
+ jmp .Lproceed_avx
+
+.align 32
+.Lstore_base2_64_avx:
+ movq %r14,0(%rdi)
+ movq %rbx,8(%rdi)
+ movq %r10,16(%rdi)
+ jmp .Ldone_avx
+
+.align 16
+.Lstore_base2_26_avx:
+ movl %eax,0(%rdi)
+ movl %edx,4(%rdi)
+ movl %r14d,8(%rdi)
+ movl %ebx,12(%rdi)
+ movl %r10d,16(%rdi)
+.align 16
+.Ldone_avx:
+ movq 8(%rsp),%r15
+ movq 16(%rsp),%r14
+ movq 24(%rsp),%r13
+ movq 32(%rsp),%r12
+ movq 40(%rsp),%rbx
+ leaq 48(%rsp),%rsp
+
+.Lno_data_avx:
+.Lblocks_avx_epilogue:
+ ret
+
+.align 32
+.Lbase2_64_avx:
+
+ pushq %rbx
+ pushq %r12
+ pushq %r13
+ pushq %r14
+ pushq %r15
+ pushq %rdi
+
+.Lbase2_64_avx_body:
+
+ movq %rdx,%r15
+
+ movq 24(%rdi),%r11
+ movq 32(%rdi),%r13
+
+ movq 0(%rdi),%r14
+ movq 8(%rdi),%rbx
+ movl 16(%rdi),%r10d
+
+ movq %r13,%r12
+ movq %r13,%rax
+ shrq $2,%r13
+ addq %r12,%r13
+
+ testq $31,%rdx
+ jz .Linit_avx
+
+ addq 0(%rsi),%r14
+ adcq 8(%rsi),%rbx
+ leaq 16(%rsi),%rsi
+ adcq %rcx,%r10
+ subq $16,%r15
+
+ movq %rdi,0(%rsp)
+ __poly1305_block
+ movq 0(%rsp),%rdi
+
+.Linit_avx:
+
+ movq %r14,%rax
+ movq %r14,%rdx
+ shrq $52,%r14
+ movq %rbx,%r8
+ movq %rbx,%r9
+ shrq $26,%rdx
+ andq $0x3ffffff,%rax
+ shlq $12,%r8
+ andq $0x3ffffff,%rdx
+ shrq $14,%rbx
+ orq %r8,%r14
+ shlq $24,%r10
+ andq $0x3ffffff,%r14
+ shrq $40,%r9
+ andq $0x3ffffff,%rbx
+ orq %r9,%r10
+
+ vmovd %eax,%xmm0
+ vmovd %edx,%xmm1
+ vmovd %r14d,%xmm2
+ vmovd %ebx,%xmm3
+ vmovd %r10d,%xmm4
+ movl $1,20(%rdi)
+
+ __poly1305_init_avx
+
+.Lproceed_avx:
+ movq %r15,%rdx
+
+ movq 8(%rsp),%r15
+ movq 16(%rsp),%r14
+ movq 24(%rsp),%r13
+ movq 32(%rsp),%r12
+ movq 40(%rsp),%rbx
+ leaq 48(%rsp),%rax
+ leaq 48(%rsp),%rsp
+
+.Lbase2_64_avx_epilogue:
+ jmp .Ldo_avx
+
+
+.align 32
+.Leven_avx:
+ vmovd 0(%rdi),%xmm0
+ vmovd 4(%rdi),%xmm1
+ vmovd 8(%rdi),%xmm2
+ vmovd 12(%rdi),%xmm3
+ vmovd 16(%rdi),%xmm4
+
+.Ldo_avx:
+ leaq 8(%rsp),%r10
+ andq $-32,%rsp
+ subq $8,%rsp
+ leaq -88(%rsp),%r11
+ subq $0x178,%rsp
+ subq $64,%rdx
+ leaq -32(%rsi),%rax
+ cmovcq %rax,%rsi
+
+ vmovdqu 48(%rdi),%xmm14
+ leaq 112(%rdi),%rdi
+ leaq .Lconst(%rip),%rcx
+
+ vmovdqu 32(%rsi),%xmm5
+ vmovdqu 48(%rsi),%xmm6
+ vmovdqa 64(%rcx),%xmm15
+
+ vpsrldq $6,%xmm5,%xmm7
+ vpsrldq $6,%xmm6,%xmm8
+ vpunpckhqdq %xmm6,%xmm5,%xmm9
+ vpunpcklqdq %xmm6,%xmm5,%xmm5
+ vpunpcklqdq %xmm8,%xmm7,%xmm8
+
+ vpsrlq $40,%xmm9,%xmm9
+ vpsrlq $26,%xmm5,%xmm6
+ vpand %xmm15,%xmm5,%xmm5
+ vpsrlq $4,%xmm8,%xmm7
+ vpand %xmm15,%xmm6,%xmm6
+ vpsrlq $30,%xmm8,%xmm8
+ vpand %xmm15,%xmm7,%xmm7
+ vpand %xmm15,%xmm8,%xmm8
+ vpor 32(%rcx),%xmm9,%xmm9
+
+ jbe .Lskip_loop_avx
+
+
+ vmovdqu -48(%rdi),%xmm11
+ vmovdqu -32(%rdi),%xmm12
+ vpshufd $0xEE,%xmm14,%xmm13
+ vpshufd $0x44,%xmm14,%xmm10
+ vmovdqa %xmm13,-144(%r11)
+ vmovdqa %xmm10,0(%rsp)
+ vpshufd $0xEE,%xmm11,%xmm14
+ vmovdqu -16(%rdi),%xmm10
+ vpshufd $0x44,%xmm11,%xmm11
+ vmovdqa %xmm14,-128(%r11)
+ vmovdqa %xmm11,16(%rsp)
+ vpshufd $0xEE,%xmm12,%xmm13
+ vmovdqu 0(%rdi),%xmm11
+ vpshufd $0x44,%xmm12,%xmm12
+ vmovdqa %xmm13,-112(%r11)
+ vmovdqa %xmm12,32(%rsp)
+ vpshufd $0xEE,%xmm10,%xmm14
+ vmovdqu 16(%rdi),%xmm12
+ vpshufd $0x44,%xmm10,%xmm10
+ vmovdqa %xmm14,-96(%r11)
+ vmovdqa %xmm10,48(%rsp)
+ vpshufd $0xEE,%xmm11,%xmm13
+ vmovdqu 32(%rdi),%xmm10
+ vpshufd $0x44,%xmm11,%xmm11
+ vmovdqa %xmm13,-80(%r11)
+ vmovdqa %xmm11,64(%rsp)
+ vpshufd $0xEE,%xmm12,%xmm14
+ vmovdqu 48(%rdi),%xmm11
+ vpshufd $0x44,%xmm12,%xmm12
+ vmovdqa %xmm14,-64(%r11)
+ vmovdqa %xmm12,80(%rsp)
+ vpshufd $0xEE,%xmm10,%xmm13
+ vmovdqu 64(%rdi),%xmm12
+ vpshufd $0x44,%xmm10,%xmm10
+ vmovdqa %xmm13,-48(%r11)
+ vmovdqa %xmm10,96(%rsp)
+ vpshufd $0xEE,%xmm11,%xmm14
+ vpshufd $0x44,%xmm11,%xmm11
+ vmovdqa %xmm14,-32(%r11)
+ vmovdqa %xmm11,112(%rsp)
+ vpshufd $0xEE,%xmm12,%xmm13
+ vmovdqa 0(%rsp),%xmm14
+ vpshufd $0x44,%xmm12,%xmm12
+ vmovdqa %xmm13,-16(%r11)
+ vmovdqa %xmm12,128(%rsp)
+
+ jmp .Loop_avx
+
+.align 32
+.Loop_avx:
+
+ vpmuludq %xmm5,%xmm14,%xmm10
+ vpmuludq %xmm6,%xmm14,%xmm11
+ vmovdqa %xmm2,32(%r11)
+ vpmuludq %xmm7,%xmm14,%xmm12
+ vmovdqa 16(%rsp),%xmm2
+ vpmuludq %xmm8,%xmm14,%xmm13
+ vpmuludq %xmm9,%xmm14,%xmm14
+
+ vmovdqa %xmm0,0(%r11)
+ vpmuludq 32(%rsp),%xmm9,%xmm0
+ vmovdqa %xmm1,16(%r11)
+ vpmuludq %xmm8,%xmm2,%xmm1
+ vpaddq %xmm0,%xmm10,%xmm10
+ vpaddq %xmm1,%xmm14,%xmm14
+ vmovdqa %xmm3,48(%r11)
+ vpmuludq %xmm7,%xmm2,%xmm0
+ vpmuludq %xmm6,%xmm2,%xmm1
+ vpaddq %xmm0,%xmm13,%xmm13
+ vmovdqa 48(%rsp),%xmm3
+ vpaddq %xmm1,%xmm12,%xmm12
+ vmovdqa %xmm4,64(%r11)
+ vpmuludq %xmm5,%xmm2,%xmm2
+ vpmuludq %xmm7,%xmm3,%xmm0
+ vpaddq %xmm2,%xmm11,%xmm11
+
+ vmovdqa 64(%rsp),%xmm4
+ vpaddq %xmm0,%xmm14,%xmm14
+ vpmuludq %xmm6,%xmm3,%xmm1
+ vpmuludq %xmm5,%xmm3,%xmm3
+ vpaddq %xmm1,%xmm13,%xmm13
+ vmovdqa 80(%rsp),%xmm2
+ vpaddq %xmm3,%xmm12,%xmm12
+ vpmuludq %xmm9,%xmm4,%xmm0
+ vpmuludq %xmm8,%xmm4,%xmm4
+ vpaddq %xmm0,%xmm11,%xmm11
+ vmovdqa 96(%rsp),%xmm3
+ vpaddq %xmm4,%xmm10,%xmm10
+
+ vmovdqa 128(%rsp),%xmm4
+ vpmuludq %xmm6,%xmm2,%xmm1
+ vpmuludq %xmm5,%xmm2,%xmm2
+ vpaddq %xmm1,%xmm14,%xmm14
+ vpaddq %xmm2,%xmm13,%xmm13
+ vpmuludq %xmm9,%xmm3,%xmm0
+ vpmuludq %xmm8,%xmm3,%xmm1
+ vpaddq %xmm0,%xmm12,%xmm12
+ vmovdqu 0(%rsi),%xmm0
+ vpaddq %xmm1,%xmm11,%xmm11
+ vpmuludq %xmm7,%xmm3,%xmm3
+ vpmuludq %xmm7,%xmm4,%xmm7
+ vpaddq %xmm3,%xmm10,%xmm10
+
+ vmovdqu 16(%rsi),%xmm1
+ vpaddq %xmm7,%xmm11,%xmm11
+ vpmuludq %xmm8,%xmm4,%xmm8
+ vpmuludq %xmm9,%xmm4,%xmm9
+ vpsrldq $6,%xmm0,%xmm2
+ vpaddq %xmm8,%xmm12,%xmm12
+ vpaddq %xmm9,%xmm13,%xmm13
+ vpsrldq $6,%xmm1,%xmm3
+ vpmuludq 112(%rsp),%xmm5,%xmm9
+ vpmuludq %xmm6,%xmm4,%xmm5
+ vpunpckhqdq %xmm1,%xmm0,%xmm4
+ vpaddq %xmm9,%xmm14,%xmm14
+ vmovdqa -144(%r11),%xmm9
+ vpaddq %xmm5,%xmm10,%xmm10
+
+ vpunpcklqdq %xmm1,%xmm0,%xmm0
+ vpunpcklqdq %xmm3,%xmm2,%xmm3
+
+
+ vpsrldq $5,%xmm4,%xmm4
+ vpsrlq $26,%xmm0,%xmm1
+ vpand %xmm15,%xmm0,%xmm0
+ vpsrlq $4,%xmm3,%xmm2
+ vpand %xmm15,%xmm1,%xmm1
+ vpand 0(%rcx),%xmm4,%xmm4
+ vpsrlq $30,%xmm3,%xmm3
+ vpand %xmm15,%xmm2,%xmm2
+ vpand %xmm15,%xmm3,%xmm3
+ vpor 32(%rcx),%xmm4,%xmm4
+
+ vpaddq 0(%r11),%xmm0,%xmm0
+ vpaddq 16(%r11),%xmm1,%xmm1
+ vpaddq 32(%r11),%xmm2,%xmm2
+ vpaddq 48(%r11),%xmm3,%xmm3
+ vpaddq 64(%r11),%xmm4,%xmm4
+
+ leaq 32(%rsi),%rax
+ leaq 64(%rsi),%rsi
+ subq $64,%rdx
+ cmovcq %rax,%rsi
+
+ vpmuludq %xmm0,%xmm9,%xmm5
+ vpmuludq %xmm1,%xmm9,%xmm6
+ vpaddq %xmm5,%xmm10,%xmm10
+ vpaddq %xmm6,%xmm11,%xmm11
+ vmovdqa -128(%r11),%xmm7
+ vpmuludq %xmm2,%xmm9,%xmm5
+ vpmuludq %xmm3,%xmm9,%xmm6
+ vpaddq %xmm5,%xmm12,%xmm12
+ vpaddq %xmm6,%xmm13,%xmm13
+ vpmuludq %xmm4,%xmm9,%xmm9
+ vpmuludq -112(%r11),%xmm4,%xmm5
+ vpaddq %xmm9,%xmm14,%xmm14
+
+ vpaddq %xmm5,%xmm10,%xmm10
+ vpmuludq %xmm2,%xmm7,%xmm6
+ vpmuludq %xmm3,%xmm7,%xmm5
+ vpaddq %xmm6,%xmm13,%xmm13
+ vmovdqa -96(%r11),%xmm8
+ vpaddq %xmm5,%xmm14,%xmm14
+ vpmuludq %xmm1,%xmm7,%xmm6
+ vpmuludq %xmm0,%xmm7,%xmm7
+ vpaddq %xmm6,%xmm12,%xmm12
+ vpaddq %xmm7,%xmm11,%xmm11
+
+ vmovdqa -80(%r11),%xmm9
+ vpmuludq %xmm2,%xmm8,%xmm5
+ vpmuludq %xmm1,%xmm8,%xmm6
+ vpaddq %xmm5,%xmm14,%xmm14
+ vpaddq %xmm6,%xmm13,%xmm13
+ vmovdqa -64(%r11),%xmm7
+ vpmuludq %xmm0,%xmm8,%xmm8
+ vpmuludq %xmm4,%xmm9,%xmm5
+ vpaddq %xmm8,%xmm12,%xmm12
+ vpaddq %xmm5,%xmm11,%xmm11
+ vmovdqa -48(%r11),%xmm8
+ vpmuludq %xmm3,%xmm9,%xmm9
+ vpmuludq %xmm1,%xmm7,%xmm6
+ vpaddq %xmm9,%xmm10,%xmm10
+
+ vmovdqa -16(%r11),%xmm9
+ vpaddq %xmm6,%xmm14,%xmm14
+ vpmuludq %xmm0,%xmm7,%xmm7
+ vpmuludq %xmm4,%xmm8,%xmm5
+ vpaddq %xmm7,%xmm13,%xmm13
+ vpaddq %xmm5,%xmm12,%xmm12
+ vmovdqu 32(%rsi),%xmm5
+ vpmuludq %xmm3,%xmm8,%xmm7
+ vpmuludq %xmm2,%xmm8,%xmm8
+ vpaddq %xmm7,%xmm11,%xmm11
+ vmovdqu 48(%rsi),%xmm6
+ vpaddq %xmm8,%xmm10,%xmm10
+
+ vpmuludq %xmm2,%xmm9,%xmm2
+ vpmuludq %xmm3,%xmm9,%xmm3
+ vpsrldq $6,%xmm5,%xmm7
+ vpaddq %xmm2,%xmm11,%xmm11
+ vpmuludq %xmm4,%xmm9,%xmm4
+ vpsrldq $6,%xmm6,%xmm8
+ vpaddq %xmm3,%xmm12,%xmm2
+ vpaddq %xmm4,%xmm13,%xmm3
+ vpmuludq -32(%r11),%xmm0,%xmm4
+ vpmuludq %xmm1,%xmm9,%xmm0
+ vpunpckhqdq %xmm6,%xmm5,%xmm9
+ vpaddq %xmm4,%xmm14,%xmm4
+ vpaddq %xmm0,%xmm10,%xmm0
+
+ vpunpcklqdq %xmm6,%xmm5,%xmm5
+ vpunpcklqdq %xmm8,%xmm7,%xmm8
+
+
+ vpsrldq $5,%xmm9,%xmm9
+ vpsrlq $26,%xmm5,%xmm6
+ vmovdqa 0(%rsp),%xmm14
+ vpand %xmm15,%xmm5,%xmm5
+ vpsrlq $4,%xmm8,%xmm7
+ vpand %xmm15,%xmm6,%xmm6
+ vpand 0(%rcx),%xmm9,%xmm9
+ vpsrlq $30,%xmm8,%xmm8
+ vpand %xmm15,%xmm7,%xmm7
+ vpand %xmm15,%xmm8,%xmm8
+ vpor 32(%rcx),%xmm9,%xmm9
+
+ vpsrlq $26,%xmm3,%xmm13
+ vpand %xmm15,%xmm3,%xmm3
+ vpaddq %xmm13,%xmm4,%xmm4
+
+ vpsrlq $26,%xmm0,%xmm10
+ vpand %xmm15,%xmm0,%xmm0
+ vpaddq %xmm10,%xmm11,%xmm1
+
+ vpsrlq $26,%xmm4,%xmm10
+ vpand %xmm15,%xmm4,%xmm4
+
+ vpsrlq $26,%xmm1,%xmm11
+ vpand %xmm15,%xmm1,%xmm1
+ vpaddq %xmm11,%xmm2,%xmm2
+
+ vpaddq %xmm10,%xmm0,%xmm0
+ vpsllq $2,%xmm10,%xmm10
+ vpaddq %xmm10,%xmm0,%xmm0
+
+ vpsrlq $26,%xmm2,%xmm12
+ vpand %xmm15,%xmm2,%xmm2
+ vpaddq %xmm12,%xmm3,%xmm3
+
+ vpsrlq $26,%xmm0,%xmm10
+ vpand %xmm15,%xmm0,%xmm0
+ vpaddq %xmm10,%xmm1,%xmm1
+
+ vpsrlq $26,%xmm3,%xmm13
+ vpand %xmm15,%xmm3,%xmm3
+ vpaddq %xmm13,%xmm4,%xmm4
+
+ ja .Loop_avx
+
+.Lskip_loop_avx:
+ vpshufd $0x10,%xmm14,%xmm14
+ addq $32,%rdx
+ jnz .Long_tail_avx
+
+ vpaddq %xmm2,%xmm7,%xmm7
+ vpaddq %xmm0,%xmm5,%xmm5
+ vpaddq %xmm1,%xmm6,%xmm6
+ vpaddq %xmm3,%xmm8,%xmm8
+ vpaddq %xmm4,%xmm9,%xmm9
+
+.Long_tail_avx:
+ vmovdqa %xmm2,32(%r11)
+ vmovdqa %xmm0,0(%r11)
+ vmovdqa %xmm1,16(%r11)
+ vmovdqa %xmm3,48(%r11)
+ vmovdqa %xmm4,64(%r11)
+
+ vpmuludq %xmm7,%xmm14,%xmm12
+ vpmuludq %xmm5,%xmm14,%xmm10
+ vpshufd $0x10,-48(%rdi),%xmm2
+ vpmuludq %xmm6,%xmm14,%xmm11
+ vpmuludq %xmm8,%xmm14,%xmm13
+ vpmuludq %xmm9,%xmm14,%xmm14
+
+ vpmuludq %xmm8,%xmm2,%xmm0
+ vpaddq %xmm0,%xmm14,%xmm14
+ vpshufd $0x10,-32(%rdi),%xmm3
+ vpmuludq %xmm7,%xmm2,%xmm1
+ vpaddq %xmm1,%xmm13,%xmm13
+ vpshufd $0x10,-16(%rdi),%xmm4
+ vpmuludq %xmm6,%xmm2,%xmm0
+ vpaddq %xmm0,%xmm12,%xmm12
+ vpmuludq %xmm5,%xmm2,%xmm2
+ vpaddq %xmm2,%xmm11,%xmm11
+ vpmuludq %xmm9,%xmm3,%xmm3
+ vpaddq %xmm3,%xmm10,%xmm10
+
+ vpshufd $0x10,0(%rdi),%xmm2
+ vpmuludq %xmm7,%xmm4,%xmm1
+ vpaddq %xmm1,%xmm14,%xmm14
+ vpmuludq %xmm6,%xmm4,%xmm0
+ vpaddq %xmm0,%xmm13,%xmm13
+ vpshufd $0x10,16(%rdi),%xmm3
+ vpmuludq %xmm5,%xmm4,%xmm4
+ vpaddq %xmm4,%xmm12,%xmm12
+ vpmuludq %xmm9,%xmm2,%xmm1
+ vpaddq %xmm1,%xmm11,%xmm11
+ vpshufd $0x10,32(%rdi),%xmm4
+ vpmuludq %xmm8,%xmm2,%xmm2
+ vpaddq %xmm2,%xmm10,%xmm10
+
+ vpmuludq %xmm6,%xmm3,%xmm0
+ vpaddq %xmm0,%xmm14,%xmm14
+ vpmuludq %xmm5,%xmm3,%xmm3
+ vpaddq %xmm3,%xmm13,%xmm13
+ vpshufd $0x10,48(%rdi),%xmm2
+ vpmuludq %xmm9,%xmm4,%xmm1
+ vpaddq %xmm1,%xmm12,%xmm12
+ vpshufd $0x10,64(%rdi),%xmm3
+ vpmuludq %xmm8,%xmm4,%xmm0
+ vpaddq %xmm0,%xmm11,%xmm11
+ vpmuludq %xmm7,%xmm4,%xmm4
+ vpaddq %xmm4,%xmm10,%xmm10
+
+ vpmuludq %xmm5,%xmm2,%xmm2
+ vpaddq %xmm2,%xmm14,%xmm14
+ vpmuludq %xmm9,%xmm3,%xmm1
+ vpaddq %xmm1,%xmm13,%xmm13
+ vpmuludq %xmm8,%xmm3,%xmm0
+ vpaddq %xmm0,%xmm12,%xmm12
+ vpmuludq %xmm7,%xmm3,%xmm1
+ vpaddq %xmm1,%xmm11,%xmm11
+ vpmuludq %xmm6,%xmm3,%xmm3
+ vpaddq %xmm3,%xmm10,%xmm10
+
+ jz .Lshort_tail_avx
+
+ vmovdqu 0(%rsi),%xmm0
+ vmovdqu 16(%rsi),%xmm1
+
+ vpsrldq $6,%xmm0,%xmm2
+ vpsrldq $6,%xmm1,%xmm3
+ vpunpckhqdq %xmm1,%xmm0,%xmm4
+ vpunpcklqdq %xmm1,%xmm0,%xmm0
+ vpunpcklqdq %xmm3,%xmm2,%xmm3
+
+ vpsrlq $40,%xmm4,%xmm4
+ vpsrlq $26,%xmm0,%xmm1
+ vpand %xmm15,%xmm0,%xmm0
+ vpsrlq $4,%xmm3,%xmm2
+ vpand %xmm15,%xmm1,%xmm1
+ vpsrlq $30,%xmm3,%xmm3
+ vpand %xmm15,%xmm2,%xmm2
+ vpand %xmm15,%xmm3,%xmm3
+ vpor 32(%rcx),%xmm4,%xmm4
+
+ vpshufd $0x32,-64(%rdi),%xmm9
+ vpaddq 0(%r11),%xmm0,%xmm0
+ vpaddq 16(%r11),%xmm1,%xmm1
+ vpaddq 32(%r11),%xmm2,%xmm2
+ vpaddq 48(%r11),%xmm3,%xmm3
+ vpaddq 64(%r11),%xmm4,%xmm4
+
+ vpmuludq %xmm0,%xmm9,%xmm5
+ vpaddq %xmm5,%xmm10,%xmm10
+ vpmuludq %xmm1,%xmm9,%xmm6
+ vpaddq %xmm6,%xmm11,%xmm11
+ vpmuludq %xmm2,%xmm9,%xmm5
+ vpaddq %xmm5,%xmm12,%xmm12
+ vpshufd $0x32,-48(%rdi),%xmm7
+ vpmuludq %xmm3,%xmm9,%xmm6
+ vpaddq %xmm6,%xmm13,%xmm13
+ vpmuludq %xmm4,%xmm9,%xmm9
+ vpaddq %xmm9,%xmm14,%xmm14
+
+ vpmuludq %xmm3,%xmm7,%xmm5
+ vpaddq %xmm5,%xmm14,%xmm14
+ vpshufd $0x32,-32(%rdi),%xmm8
+ vpmuludq %xmm2,%xmm7,%xmm6
+ vpaddq %xmm6,%xmm13,%xmm13
+ vpshufd $0x32,-16(%rdi),%xmm9
+ vpmuludq %xmm1,%xmm7,%xmm5
+ vpaddq %xmm5,%xmm12,%xmm12
+ vpmuludq %xmm0,%xmm7,%xmm7
+ vpaddq %xmm7,%xmm11,%xmm11
+ vpmuludq %xmm4,%xmm8,%xmm8
+ vpaddq %xmm8,%xmm10,%xmm10
+
+ vpshufd $0x32,0(%rdi),%xmm7
+ vpmuludq %xmm2,%xmm9,%xmm6
+ vpaddq %xmm6,%xmm14,%xmm14
+ vpmuludq %xmm1,%xmm9,%xmm5
+ vpaddq %xmm5,%xmm13,%xmm13
+ vpshufd $0x32,16(%rdi),%xmm8
+ vpmuludq %xmm0,%xmm9,%xmm9
+ vpaddq %xmm9,%xmm12,%xmm12
+ vpmuludq %xmm4,%xmm7,%xmm6
+ vpaddq %xmm6,%xmm11,%xmm11
+ vpshufd $0x32,32(%rdi),%xmm9
+ vpmuludq %xmm3,%xmm7,%xmm7
+ vpaddq %xmm7,%xmm10,%xmm10
+
+ vpmuludq %xmm1,%xmm8,%xmm5
+ vpaddq %xmm5,%xmm14,%xmm14
+ vpmuludq %xmm0,%xmm8,%xmm8
+ vpaddq %xmm8,%xmm13,%xmm13
+ vpshufd $0x32,48(%rdi),%xmm7
+ vpmuludq %xmm4,%xmm9,%xmm6
+ vpaddq %xmm6,%xmm12,%xmm12
+ vpshufd $0x32,64(%rdi),%xmm8
+ vpmuludq %xmm3,%xmm9,%xmm5
+ vpaddq %xmm5,%xmm11,%xmm11
+ vpmuludq %xmm2,%xmm9,%xmm9
+ vpaddq %xmm9,%xmm10,%xmm10
+
+ vpmuludq %xmm0,%xmm7,%xmm7
+ vpaddq %xmm7,%xmm14,%xmm14
+ vpmuludq %xmm4,%xmm8,%xmm6
+ vpaddq %xmm6,%xmm13,%xmm13
+ vpmuludq %xmm3,%xmm8,%xmm5
+ vpaddq %xmm5,%xmm12,%xmm12
+ vpmuludq %xmm2,%xmm8,%xmm6
+ vpaddq %xmm6,%xmm11,%xmm11
+ vpmuludq %xmm1,%xmm8,%xmm8
+ vpaddq %xmm8,%xmm10,%xmm10
+
+.Lshort_tail_avx:
+
+ vpsrldq $8,%xmm14,%xmm9
+ vpsrldq $8,%xmm13,%xmm8
+ vpsrldq $8,%xmm11,%xmm6
+ vpsrldq $8,%xmm10,%xmm5
+ vpsrldq $8,%xmm12,%xmm7
+ vpaddq %xmm8,%xmm13,%xmm13
+ vpaddq %xmm9,%xmm14,%xmm14
+ vpaddq %xmm5,%xmm10,%xmm10
+ vpaddq %xmm6,%xmm11,%xmm11
+ vpaddq %xmm7,%xmm12,%xmm12
+
+ vpsrlq $26,%xmm13,%xmm3
+ vpand %xmm15,%xmm13,%xmm13
+ vpaddq %xmm3,%xmm14,%xmm14
+
+ vpsrlq $26,%xmm10,%xmm0
+ vpand %xmm15,%xmm10,%xmm10
+ vpaddq %xmm0,%xmm11,%xmm11
+
+ vpsrlq $26,%xmm14,%xmm4
+ vpand %xmm15,%xmm14,%xmm14
+
+ vpsrlq $26,%xmm11,%xmm1
+ vpand %xmm15,%xmm11,%xmm11
+ vpaddq %xmm1,%xmm12,%xmm12
+
+ vpaddq %xmm4,%xmm10,%xmm10
+ vpsllq $2,%xmm4,%xmm4
+ vpaddq %xmm4,%xmm10,%xmm10
+
+ vpsrlq $26,%xmm12,%xmm2
+ vpand %xmm15,%xmm12,%xmm12
+ vpaddq %xmm2,%xmm13,%xmm13
+
+ vpsrlq $26,%xmm10,%xmm0
+ vpand %xmm15,%xmm10,%xmm10
+ vpaddq %xmm0,%xmm11,%xmm11
+
+ vpsrlq $26,%xmm13,%xmm3
+ vpand %xmm15,%xmm13,%xmm13
+ vpaddq %xmm3,%xmm14,%xmm14
+
+ vmovd %xmm10,-112(%rdi)
+ vmovd %xmm11,-108(%rdi)
+ vmovd %xmm12,-104(%rdi)
+ vmovd %xmm13,-100(%rdi)
+ vmovd %xmm14,-96(%rdi)
+ leaq -8(%r10),%rsp
+
+ vzeroupper
+ ret
+ENDPROC(poly1305_blocks_avx)
+
+.align 32
+ENTRY(poly1305_emit_avx)
+ cmpl $0,20(%rdi)
+ je .Lemit
+
+ movl 0(%rdi),%eax
+ movl 4(%rdi),%ecx
+ movl 8(%rdi),%r8d
+ movl 12(%rdi),%r11d
+ movl 16(%rdi),%r10d
+
+ shlq $26,%rcx
+ movq %r8,%r9
+ shlq $52,%r8
+ addq %rcx,%rax
+ shrq $12,%r9
+ addq %rax,%r8
+ adcq $0,%r9
+
+ shlq $14,%r11
+ movq %r10,%rax
+ shrq $24,%r10
+ addq %r11,%r9
+ shlq $40,%rax
+ addq %rax,%r9
+ adcq $0,%r10
+
+ movq %r10,%rax
+ movq %r10,%rcx
+ andq $3,%r10
+ shrq $2,%rax
+ andq $-4,%rcx
+ addq %rcx,%rax
+ addq %rax,%r8
+ adcq $0,%r9
+ adcq $0,%r10
+
+ movq %r8,%rax
+ addq $5,%r8
+ movq %r9,%rcx
+ adcq $0,%r9
+ adcq $0,%r10
+ shrq $2,%r10
+ cmovnzq %r8,%rax
+ cmovnzq %r9,%rcx
+
+ addq 0(%rdx),%rax
+ adcq 8(%rdx),%rcx
+ movq %rax,0(%rsi)
+ movq %rcx,8(%rsi)
+
+ ret
+ENDPROC(poly1305_emit_avx)
+#endif /* CONFIG_AS_AVX */
+
+#ifdef CONFIG_AS_AVX2
+.align 32
+ENTRY(poly1305_blocks_avx2)
+
+ movl 20(%rdi),%r8d
+ cmpq $128,%rdx
+ jae .Lblocks_avx2
+ testl %r8d,%r8d
+ jz .Lblocks
+
+.Lblocks_avx2:
+ andq $-16,%rdx
+ jz .Lno_data_avx2
+
+ vzeroupper
+
+ testl %r8d,%r8d
+ jz .Lbase2_64_avx2
+
+ testq $63,%rdx
+ jz .Leven_avx2
+
+ pushq %rbx
+ pushq %r12
+ pushq %r13
+ pushq %r14
+ pushq %r15
+ pushq %rdi
+
+.Lblocks_avx2_body:
+
+ movq %rdx,%r15
+
+ movq 0(%rdi),%r8
+ movq 8(%rdi),%r9
+ movl 16(%rdi),%r10d
+
+ movq 24(%rdi),%r11
+ movq 32(%rdi),%r13
+
+
+ movl %r8d,%r14d
+ andq $-2147483648,%r8
+ movq %r9,%r12
+ movl %r9d,%ebx
+ andq $-2147483648,%r9
+
+ shrq $6,%r8
+ shlq $52,%r12
+ addq %r8,%r14
+ shrq $12,%rbx
+ shrq $18,%r9
+ addq %r12,%r14
+ adcq %r9,%rbx
+
+ movq %r10,%r8
+ shlq $40,%r8
+ shrq $24,%r10
+ addq %r8,%rbx
+ adcq $0,%r10
+
+ movq $-4,%r9
+ movq %r10,%r8
+ andq %r10,%r9
+ shrq $2,%r8
+ andq $3,%r10
+ addq %r9,%r8
+ addq %r8,%r14
+ adcq $0,%rbx
+ adcq $0,%r10
+
+ movq %r13,%r12
+ movq %r13,%rax
+ shrq $2,%r13
+ addq %r12,%r13
+
+.Lbase2_26_pre_avx2:
+ addq 0(%rsi),%r14
+ adcq 8(%rsi),%rbx
+ leaq 16(%rsi),%rsi
+ adcq %rcx,%r10
+ subq $16,%r15
+
+ movq %rdi,0(%rsp)
+ __poly1305_block
+ movq 0(%rsp),%rdi
+ movq %r12,%rax
+
+ testq $63,%r15
+ jnz .Lbase2_26_pre_avx2
+
+ testq %rcx,%rcx
+ jz .Lstore_base2_64_avx2
+
+
+ movq %r14,%rax
+ movq %r14,%rdx
+ shrq $52,%r14
+ movq %rbx,%r11
+ movq %rbx,%r12
+ shrq $26,%rdx
+ andq $0x3ffffff,%rax
+ shlq $12,%r11
+ andq $0x3ffffff,%rdx
+ shrq $14,%rbx
+ orq %r11,%r14
+ shlq $24,%r10
+ andq $0x3ffffff,%r14
+ shrq $40,%r12
+ andq $0x3ffffff,%rbx
+ orq %r12,%r10
+
+ testq %r15,%r15
+ jz .Lstore_base2_26_avx2
+
+ vmovd %eax,%xmm0
+ vmovd %edx,%xmm1
+ vmovd %r14d,%xmm2
+ vmovd %ebx,%xmm3
+ vmovd %r10d,%xmm4
+ jmp .Lproceed_avx2
+
+.align 32
+.Lstore_base2_64_avx2:
+ movq %r14,0(%rdi)
+ movq %rbx,8(%rdi)
+ movq %r10,16(%rdi)
+ jmp .Ldone_avx2
+
+.align 16
+.Lstore_base2_26_avx2:
+ movl %eax,0(%rdi)
+ movl %edx,4(%rdi)
+ movl %r14d,8(%rdi)
+ movl %ebx,12(%rdi)
+ movl %r10d,16(%rdi)
+.align 16
+.Ldone_avx2:
+ movq 8(%rsp),%r15
+ movq 16(%rsp),%r14
+ movq 24(%rsp),%r13
+ movq 32(%rsp),%r12
+ movq 40(%rsp),%rbx
+ leaq 48(%rsp),%rsp
+
+.Lno_data_avx2:
+.Lblocks_avx2_epilogue:
+ ret
+
+
+.align 32
+.Lbase2_64_avx2:
+
+
+ pushq %rbx
+ pushq %r12
+ pushq %r13
+ pushq %r14
+ pushq %r15
+ pushq %rdi
+
+.Lbase2_64_avx2_body:
+
+ movq %rdx,%r15
+
+ movq 24(%rdi),%r11
+ movq 32(%rdi),%r13
+
+ movq 0(%rdi),%r14
+ movq 8(%rdi),%rbx
+ movl 16(%rdi),%r10d
+
+ movq %r13,%r12
+ movq %r13,%rax
+ shrq $2,%r13
+ addq %r12,%r13
+
+ testq $63,%rdx
+ jz .Linit_avx2
+
+.Lbase2_64_pre_avx2:
+ addq 0(%rsi),%r14
+ adcq 8(%rsi),%rbx
+ leaq 16(%rsi),%rsi
+ adcq %rcx,%r10
+ subq $16,%r15
+
+ movq %rdi,0(%rsp)
+ __poly1305_block
+ movq 0(%rsp),%rdi
+ movq %r12,%rax
+
+ testq $63,%r15
+ jnz .Lbase2_64_pre_avx2
+
+.Linit_avx2:
+
+ movq %r14,%rax
+ movq %r14,%rdx
+ shrq $52,%r14
+ movq %rbx,%r8
+ movq %rbx,%r9
+ shrq $26,%rdx
+ andq $0x3ffffff,%rax
+ shlq $12,%r8
+ andq $0x3ffffff,%rdx
+ shrq $14,%rbx
+ orq %r8,%r14
+ shlq $24,%r10
+ andq $0x3ffffff,%r14
+ shrq $40,%r9
+ andq $0x3ffffff,%rbx
+ orq %r9,%r10
+
+ vmovd %eax,%xmm0
+ vmovd %edx,%xmm1
+ vmovd %r14d,%xmm2
+ vmovd %ebx,%xmm3
+ vmovd %r10d,%xmm4
+ movl $1,20(%rdi)
+
+ __poly1305_init_avx
+
+.Lproceed_avx2:
+ movq %r15,%rdx
+
+ movq 8(%rsp),%r15
+ movq 16(%rsp),%r14
+ movq 24(%rsp),%r13
+ movq 32(%rsp),%r12
+ movq 40(%rsp),%rbx
+ leaq 48(%rsp),%rax
+ leaq 48(%rsp),%rsp
+
+.Lbase2_64_avx2_epilogue:
+ jmp .Ldo_avx2
+
+
+.align 32
+.Leven_avx2:
+
+ vmovd 0(%rdi),%xmm0
+ vmovd 4(%rdi),%xmm1
+ vmovd 8(%rdi),%xmm2
+ vmovd 12(%rdi),%xmm3
+ vmovd 16(%rdi),%xmm4
+
+.Ldo_avx2:
+ leaq 8(%rsp),%r10
+ subq $0x128,%rsp
+ leaq .Lconst(%rip),%rcx
+ leaq 48+64(%rdi),%rdi
+ vmovdqa 96(%rcx),%ymm7
+
+
+ vmovdqu -64(%rdi),%xmm9
+ andq $-512,%rsp
+ vmovdqu -48(%rdi),%xmm10
+ vmovdqu -32(%rdi),%xmm6
+ vmovdqu -16(%rdi),%xmm11
+ vmovdqu 0(%rdi),%xmm12
+ vmovdqu 16(%rdi),%xmm13
+ leaq 144(%rsp),%rax
+ vmovdqu 32(%rdi),%xmm14
+ vpermd %ymm9,%ymm7,%ymm9
+ vmovdqu 48(%rdi),%xmm15
+ vpermd %ymm10,%ymm7,%ymm10
+ vmovdqu 64(%rdi),%xmm5
+ vpermd %ymm6,%ymm7,%ymm6
+ vmovdqa %ymm9,0(%rsp)
+ vpermd %ymm11,%ymm7,%ymm11
+ vmovdqa %ymm10,32-144(%rax)
+ vpermd %ymm12,%ymm7,%ymm12
+ vmovdqa %ymm6,64-144(%rax)
+ vpermd %ymm13,%ymm7,%ymm13
+ vmovdqa %ymm11,96-144(%rax)
+ vpermd %ymm14,%ymm7,%ymm14
+ vmovdqa %ymm12,128-144(%rax)
+ vpermd %ymm15,%ymm7,%ymm15
+ vmovdqa %ymm13,160-144(%rax)
+ vpermd %ymm5,%ymm7,%ymm5
+ vmovdqa %ymm14,192-144(%rax)
+ vmovdqa %ymm15,224-144(%rax)
+ vmovdqa %ymm5,256-144(%rax)
+ vmovdqa 64(%rcx),%ymm5
+
+
+
+ vmovdqu 0(%rsi),%xmm7
+ vmovdqu 16(%rsi),%xmm8
+ vinserti128 $1,32(%rsi),%ymm7,%ymm7
+ vinserti128 $1,48(%rsi),%ymm8,%ymm8
+ leaq 64(%rsi),%rsi
+
+ vpsrldq $6,%ymm7,%ymm9
+ vpsrldq $6,%ymm8,%ymm10
+ vpunpckhqdq %ymm8,%ymm7,%ymm6
+ vpunpcklqdq %ymm10,%ymm9,%ymm9
+ vpunpcklqdq %ymm8,%ymm7,%ymm7
+
+ vpsrlq $30,%ymm9,%ymm10
+ vpsrlq $4,%ymm9,%ymm9
+ vpsrlq $26,%ymm7,%ymm8
+ vpsrlq $40,%ymm6,%ymm6
+ vpand %ymm5,%ymm9,%ymm9
+ vpand %ymm5,%ymm7,%ymm7
+ vpand %ymm5,%ymm8,%ymm8
+ vpand %ymm5,%ymm10,%ymm10
+ vpor 32(%rcx),%ymm6,%ymm6
+
+ vpaddq %ymm2,%ymm9,%ymm2
+ subq $64,%rdx
+ jz .Ltail_avx2
+ jmp .Loop_avx2
+
+.align 32
+.Loop_avx2:
+
+ vpaddq %ymm0,%ymm7,%ymm0
+ vmovdqa 0(%rsp),%ymm7
+ vpaddq %ymm1,%ymm8,%ymm1
+ vmovdqa 32(%rsp),%ymm8
+ vpaddq %ymm3,%ymm10,%ymm3
+ vmovdqa 96(%rsp),%ymm9
+ vpaddq %ymm4,%ymm6,%ymm4
+ vmovdqa 48(%rax),%ymm10
+ vmovdqa 112(%rax),%ymm5
+
+ vpmuludq %ymm2,%ymm7,%ymm13
+ vpmuludq %ymm2,%ymm8,%ymm14
+ vpmuludq %ymm2,%ymm9,%ymm15
+ vpmuludq %ymm2,%ymm10,%ymm11
+ vpmuludq %ymm2,%ymm5,%ymm12
+
+ vpmuludq %ymm0,%ymm8,%ymm6
+ vpmuludq %ymm1,%ymm8,%ymm2
+ vpaddq %ymm6,%ymm12,%ymm12
+ vpaddq %ymm2,%ymm13,%ymm13
+ vpmuludq %ymm3,%ymm8,%ymm6
+ vpmuludq 64(%rsp),%ymm4,%ymm2
+ vpaddq %ymm6,%ymm15,%ymm15
+ vpaddq %ymm2,%ymm11,%ymm11
+ vmovdqa -16(%rax),%ymm8
+
+ vpmuludq %ymm0,%ymm7,%ymm6
+ vpmuludq %ymm1,%ymm7,%ymm2
+ vpaddq %ymm6,%ymm11,%ymm11
+ vpaddq %ymm2,%ymm12,%ymm12
+ vpmuludq %ymm3,%ymm7,%ymm6
+ vpmuludq %ymm4,%ymm7,%ymm2
+ vmovdqu 0(%rsi),%xmm7
+ vpaddq %ymm6,%ymm14,%ymm14
+ vpaddq %ymm2,%ymm15,%ymm15
+ vinserti128 $1,32(%rsi),%ymm7,%ymm7
+
+ vpmuludq %ymm3,%ymm8,%ymm6
+ vpmuludq %ymm4,%ymm8,%ymm2
+ vmovdqu 16(%rsi),%xmm8
+ vpaddq %ymm6,%ymm11,%ymm11
+ vpaddq %ymm2,%ymm12,%ymm12
+ vmovdqa 16(%rax),%ymm2
+ vpmuludq %ymm1,%ymm9,%ymm6
+ vpmuludq %ymm0,%ymm9,%ymm9
+ vpaddq %ymm6,%ymm14,%ymm14
+ vpaddq %ymm9,%ymm13,%ymm13
+ vinserti128 $1,48(%rsi),%ymm8,%ymm8
+ leaq 64(%rsi),%rsi
+
+ vpmuludq %ymm1,%ymm2,%ymm6
+ vpmuludq %ymm0,%ymm2,%ymm2
+ vpsrldq $6,%ymm7,%ymm9
+ vpaddq %ymm6,%ymm15,%ymm15
+ vpaddq %ymm2,%ymm14,%ymm14
+ vpmuludq %ymm3,%ymm10,%ymm6
+ vpmuludq %ymm4,%ymm10,%ymm2
+ vpsrldq $6,%ymm8,%ymm10
+ vpaddq %ymm6,%ymm12,%ymm12
+ vpaddq %ymm2,%ymm13,%ymm13
+ vpunpckhqdq %ymm8,%ymm7,%ymm6
+
+ vpmuludq %ymm3,%ymm5,%ymm3
+ vpmuludq %ymm4,%ymm5,%ymm4
+ vpunpcklqdq %ymm8,%ymm7,%ymm7
+ vpaddq %ymm3,%ymm13,%ymm2
+ vpaddq %ymm4,%ymm14,%ymm3
+ vpunpcklqdq %ymm10,%ymm9,%ymm10
+ vpmuludq 80(%rax),%ymm0,%ymm4
+ vpmuludq %ymm1,%ymm5,%ymm0
+ vmovdqa 64(%rcx),%ymm5
+ vpaddq %ymm4,%ymm15,%ymm4
+ vpaddq %ymm0,%ymm11,%ymm0
+
+ vpsrlq $26,%ymm3,%ymm14
+ vpand %ymm5,%ymm3,%ymm3
+ vpaddq %ymm14,%ymm4,%ymm4
+
+ vpsrlq $26,%ymm0,%ymm11
+ vpand %ymm5,%ymm0,%ymm0
+ vpaddq %ymm11,%ymm12,%ymm1
+
+ vpsrlq $26,%ymm4,%ymm15
+ vpand %ymm5,%ymm4,%ymm4
+
+ vpsrlq $4,%ymm10,%ymm9
+
+ vpsrlq $26,%ymm1,%ymm12
+ vpand %ymm5,%ymm1,%ymm1
+ vpaddq %ymm12,%ymm2,%ymm2
+
+ vpaddq %ymm15,%ymm0,%ymm0
+ vpsllq $2,%ymm15,%ymm15
+ vpaddq %ymm15,%ymm0,%ymm0
+
+ vpand %ymm5,%ymm9,%ymm9
+ vpsrlq $26,%ymm7,%ymm8
+
+ vpsrlq $26,%ymm2,%ymm13
+ vpand %ymm5,%ymm2,%ymm2
+ vpaddq %ymm13,%ymm3,%ymm3
+
+ vpaddq %ymm9,%ymm2,%ymm2
+ vpsrlq $30,%ymm10,%ymm10
+
+ vpsrlq $26,%ymm0,%ymm11
+ vpand %ymm5,%ymm0,%ymm0
+ vpaddq %ymm11,%ymm1,%ymm1
+
+ vpsrlq $40,%ymm6,%ymm6
+
+ vpsrlq $26,%ymm3,%ymm14
+ vpand %ymm5,%ymm3,%ymm3
+ vpaddq %ymm14,%ymm4,%ymm4
+
+ vpand %ymm5,%ymm7,%ymm7
+ vpand %ymm5,%ymm8,%ymm8
+ vpand %ymm5,%ymm10,%ymm10
+ vpor 32(%rcx),%ymm6,%ymm6
+
+ subq $64,%rdx
+ jnz .Loop_avx2
+
+.byte 0x66,0x90
+.Ltail_avx2:
+
+ vpaddq %ymm0,%ymm7,%ymm0
+ vmovdqu 4(%rsp),%ymm7
+ vpaddq %ymm1,%ymm8,%ymm1
+ vmovdqu 36(%rsp),%ymm8
+ vpaddq %ymm3,%ymm10,%ymm3
+ vmovdqu 100(%rsp),%ymm9
+ vpaddq %ymm4,%ymm6,%ymm4
+ vmovdqu 52(%rax),%ymm10
+ vmovdqu 116(%rax),%ymm5
+
+ vpmuludq %ymm2,%ymm7,%ymm13
+ vpmuludq %ymm2,%ymm8,%ymm14
+ vpmuludq %ymm2,%ymm9,%ymm15
+ vpmuludq %ymm2,%ymm10,%ymm11
+ vpmuludq %ymm2,%ymm5,%ymm12
+
+ vpmuludq %ymm0,%ymm8,%ymm6
+ vpmuludq %ymm1,%ymm8,%ymm2
+ vpaddq %ymm6,%ymm12,%ymm12
+ vpaddq %ymm2,%ymm13,%ymm13
+ vpmuludq %ymm3,%ymm8,%ymm6
+ vpmuludq 68(%rsp),%ymm4,%ymm2
+ vpaddq %ymm6,%ymm15,%ymm15
+ vpaddq %ymm2,%ymm11,%ymm11
+
+ vpmuludq %ymm0,%ymm7,%ymm6
+ vpmuludq %ymm1,%ymm7,%ymm2
+ vpaddq %ymm6,%ymm11,%ymm11
+ vmovdqu -12(%rax),%ymm8
+ vpaddq %ymm2,%ymm12,%ymm12
+ vpmuludq %ymm3,%ymm7,%ymm6
+ vpmuludq %ymm4,%ymm7,%ymm2
+ vpaddq %ymm6,%ymm14,%ymm14
+ vpaddq %ymm2,%ymm15,%ymm15
+
+ vpmuludq %ymm3,%ymm8,%ymm6
+ vpmuludq %ymm4,%ymm8,%ymm2
+ vpaddq %ymm6,%ymm11,%ymm11
+ vpaddq %ymm2,%ymm12,%ymm12
+ vmovdqu 20(%rax),%ymm2
+ vpmuludq %ymm1,%ymm9,%ymm6
+ vpmuludq %ymm0,%ymm9,%ymm9
+ vpaddq %ymm6,%ymm14,%ymm14
+ vpaddq %ymm9,%ymm13,%ymm13
+
+ vpmuludq %ymm1,%ymm2,%ymm6
+ vpmuludq %ymm0,%ymm2,%ymm2
+ vpaddq %ymm6,%ymm15,%ymm15
+ vpaddq %ymm2,%ymm14,%ymm14
+ vpmuludq %ymm3,%ymm10,%ymm6
+ vpmuludq %ymm4,%ymm10,%ymm2
+ vpaddq %ymm6,%ymm12,%ymm12
+ vpaddq %ymm2,%ymm13,%ymm13
+
+ vpmuludq %ymm3,%ymm5,%ymm3
+ vpmuludq %ymm4,%ymm5,%ymm4
+ vpaddq %ymm3,%ymm13,%ymm2
+ vpaddq %ymm4,%ymm14,%ymm3
+ vpmuludq 84(%rax),%ymm0,%ymm4
+ vpmuludq %ymm1,%ymm5,%ymm0
+ vmovdqa 64(%rcx),%ymm5
+ vpaddq %ymm4,%ymm15,%ymm4
+ vpaddq %ymm0,%ymm11,%ymm0
+
+ vpsrldq $8,%ymm12,%ymm8
+ vpsrldq $8,%ymm2,%ymm9
+ vpsrldq $8,%ymm3,%ymm10
+ vpsrldq $8,%ymm4,%ymm6
+ vpsrldq $8,%ymm0,%ymm7
+ vpaddq %ymm8,%ymm12,%ymm12
+ vpaddq %ymm9,%ymm2,%ymm2
+ vpaddq %ymm10,%ymm3,%ymm3
+ vpaddq %ymm6,%ymm4,%ymm4
+ vpaddq %ymm7,%ymm0,%ymm0
+
+ vpermq $0x2,%ymm3,%ymm10
+ vpermq $0x2,%ymm4,%ymm6
+ vpermq $0x2,%ymm0,%ymm7
+ vpermq $0x2,%ymm12,%ymm8
+ vpermq $0x2,%ymm2,%ymm9
+ vpaddq %ymm10,%ymm3,%ymm3
+ vpaddq %ymm6,%ymm4,%ymm4
+ vpaddq %ymm7,%ymm0,%ymm0
+ vpaddq %ymm8,%ymm12,%ymm12
+ vpaddq %ymm9,%ymm2,%ymm2
+
+ vpsrlq $26,%ymm3,%ymm14
+ vpand %ymm5,%ymm3,%ymm3
+ vpaddq %ymm14,%ymm4,%ymm4
+
+ vpsrlq $26,%ymm0,%ymm11
+ vpand %ymm5,%ymm0,%ymm0
+ vpaddq %ymm11,%ymm12,%ymm1
+
+ vpsrlq $26,%ymm4,%ymm15
+ vpand %ymm5,%ymm4,%ymm4
+
+ vpsrlq $26,%ymm1,%ymm12
+ vpand %ymm5,%ymm1,%ymm1
+ vpaddq %ymm12,%ymm2,%ymm2
+
+ vpaddq %ymm15,%ymm0,%ymm0
+ vpsllq $2,%ymm15,%ymm15
+ vpaddq %ymm15,%ymm0,%ymm0
+
+ vpsrlq $26,%ymm2,%ymm13
+ vpand %ymm5,%ymm2,%ymm2
+ vpaddq %ymm13,%ymm3,%ymm3
+
+ vpsrlq $26,%ymm0,%ymm11
+ vpand %ymm5,%ymm0,%ymm0
+ vpaddq %ymm11,%ymm1,%ymm1
+
+ vpsrlq $26,%ymm3,%ymm14
+ vpand %ymm5,%ymm3,%ymm3
+ vpaddq %ymm14,%ymm4,%ymm4
+
+ vmovd %xmm0,-112(%rdi)
+ vmovd %xmm1,-108(%rdi)
+ vmovd %xmm2,-104(%rdi)
+ vmovd %xmm3,-100(%rdi)
+ vmovd %xmm4,-96(%rdi)
+ leaq -8(%r10),%rsp
+
+ vzeroupper
+ ret
+
+ENDPROC(poly1305_blocks_avx2)
+#endif /* CONFIG_AS_AVX2 */
+
+#ifdef CONFIG_AS_AVX512
+.align 32
+ENTRY(poly1305_blocks_avx512)
+
+ movl 20(%rdi),%r8d
+ cmpq $128,%rdx
+ jae .Lblocks_avx2_512
+ testl %r8d,%r8d
+ jz .Lblocks
+
+.Lblocks_avx2_512:
+ andq $-16,%rdx
+ jz .Lno_data_avx2_512
+
+ vzeroupper
+
+ testl %r8d,%r8d
+ jz .Lbase2_64_avx2_512
+
+ testq $63,%rdx
+ jz .Leven_avx2_512
+
+ pushq %rbx
+ pushq %r12
+ pushq %r13
+ pushq %r14
+ pushq %r15
+ pushq %rdi
+
+.Lblocks_avx2_body_512:
+
+ movq %rdx,%r15
+
+ movq 0(%rdi),%r8
+ movq 8(%rdi),%r9
+ movl 16(%rdi),%r10d
+
+ movq 24(%rdi),%r11
+ movq 32(%rdi),%r13
+
+
+ movl %r8d,%r14d
+ andq $-2147483648,%r8
+ movq %r9,%r12
+ movl %r9d,%ebx
+ andq $-2147483648,%r9
+
+ shrq $6,%r8
+ shlq $52,%r12
+ addq %r8,%r14
+ shrq $12,%rbx
+ shrq $18,%r9
+ addq %r12,%r14
+ adcq %r9,%rbx
+
+ movq %r10,%r8
+ shlq $40,%r8
+ shrq $24,%r10
+ addq %r8,%rbx
+ adcq $0,%r10
+
+ movq $-4,%r9
+ movq %r10,%r8
+ andq %r10,%r9
+ shrq $2,%r8
+ andq $3,%r10
+ addq %r9,%r8
+ addq %r8,%r14
+ adcq $0,%rbx
+ adcq $0,%r10
+
+ movq %r13,%r12
+ movq %r13,%rax
+ shrq $2,%r13
+ addq %r12,%r13
+
+.Lbase2_26_pre_avx2_512:
+ addq 0(%rsi),%r14
+ adcq 8(%rsi),%rbx
+ leaq 16(%rsi),%rsi
+ adcq %rcx,%r10
+ subq $16,%r15
+
+ movq %rdi,0(%rsp)
+ __poly1305_block
+ movq 0(%rsp),%rdi
+ movq %r12,%rax
+
+ testq $63,%r15
+ jnz .Lbase2_26_pre_avx2_512
+
+ testq %rcx,%rcx
+ jz .Lstore_base2_64_avx2_512
+
+
+ movq %r14,%rax
+ movq %r14,%rdx
+ shrq $52,%r14
+ movq %rbx,%r11
+ movq %rbx,%r12
+ shrq $26,%rdx
+ andq $0x3ffffff,%rax
+ shlq $12,%r11
+ andq $0x3ffffff,%rdx
+ shrq $14,%rbx
+ orq %r11,%r14
+ shlq $24,%r10
+ andq $0x3ffffff,%r14
+ shrq $40,%r12
+ andq $0x3ffffff,%rbx
+ orq %r12,%r10
+
+ testq %r15,%r15
+ jz .Lstore_base2_26_avx2_512
+
+ vmovd %eax,%xmm0
+ vmovd %edx,%xmm1
+ vmovd %r14d,%xmm2
+ vmovd %ebx,%xmm3
+ vmovd %r10d,%xmm4
+ jmp .Lproceed_avx2_512
+
+.align 32
+.Lstore_base2_64_avx2_512:
+ movq %r14,0(%rdi)
+ movq %rbx,8(%rdi)
+ movq %r10,16(%rdi)
+ jmp .Ldone_avx2_512
+
+.align 16
+.Lstore_base2_26_avx2_512:
+ movl %eax,0(%rdi)
+ movl %edx,4(%rdi)
+ movl %r14d,8(%rdi)
+ movl %ebx,12(%rdi)
+ movl %r10d,16(%rdi)
+.align 16
+.Ldone_avx2_512:
+ movq 8(%rsp),%r15
+ movq 16(%rsp),%r14
+ movq 24(%rsp),%r13
+ movq 32(%rsp),%r12
+ movq 40(%rsp),%rbx
+ leaq 48(%rsp),%rsp
+
+.Lno_data_avx2_512:
+.Lblocks_avx2_epilogue_512:
+ ret
+
+
+.align 32
+.Lbase2_64_avx2_512:
+
+ pushq %rbx
+ pushq %r12
+ pushq %r13
+ pushq %r14
+ pushq %r15
+ pushq %rdi
+
+.Lbase2_64_avx2_body_512:
+
+ movq %rdx,%r15
+
+ movq 24(%rdi),%r11
+ movq 32(%rdi),%r13
+
+ movq 0(%rdi),%r14
+ movq 8(%rdi),%rbx
+ movl 16(%rdi),%r10d
+
+ movq %r13,%r12
+ movq %r13,%rax
+ shrq $2,%r13
+ addq %r12,%r13
+
+ testq $63,%rdx
+ jz .Linit_avx2_512
+
+.Lbase2_64_pre_avx2_512:
+ addq 0(%rsi),%r14
+ adcq 8(%rsi),%rbx
+ leaq 16(%rsi),%rsi
+ adcq %rcx,%r10
+ subq $16,%r15
+
+ movq %rdi,0(%rsp)
+ __poly1305_block
+ movq 0(%rsp),%rdi
+ movq %r12,%rax
+
+ testq $63,%r15
+ jnz .Lbase2_64_pre_avx2_512
+
+.Linit_avx2_512:
+
+ movq %r14,%rax
+ movq %r14,%rdx
+ shrq $52,%r14
+ movq %rbx,%r8
+ movq %rbx,%r9
+ shrq $26,%rdx
+ andq $0x3ffffff,%rax
+ shlq $12,%r8
+ andq $0x3ffffff,%rdx
+ shrq $14,%rbx
+ orq %r8,%r14
+ shlq $24,%r10
+ andq $0x3ffffff,%r14
+ shrq $40,%r9
+ andq $0x3ffffff,%rbx
+ orq %r9,%r10
+
+ vmovd %eax,%xmm0
+ vmovd %edx,%xmm1
+ vmovd %r14d,%xmm2
+ vmovd %ebx,%xmm3
+ vmovd %r10d,%xmm4
+ movl $1,20(%rdi)
+
+ __poly1305_init_avx
+
+.Lproceed_avx2_512:
+ movq %r15,%rdx
+
+ movq 8(%rsp),%r15
+ movq 16(%rsp),%r14
+ movq 24(%rsp),%r13
+ movq 32(%rsp),%r12
+ movq 40(%rsp),%rbx
+ leaq 48(%rsp),%rax
+ leaq 48(%rsp),%rsp
+
+.Lbase2_64_avx2_epilogue_512:
+ jmp .Ldo_avx2_512
+
+
+.align 32
+.Leven_avx2_512:
+
+ vmovd 0(%rdi),%xmm0
+ vmovd 4(%rdi),%xmm1
+ vmovd 8(%rdi),%xmm2
+ vmovd 12(%rdi),%xmm3
+ vmovd 16(%rdi),%xmm4
+
+.Ldo_avx2_512:
+ cmpq $512,%rdx
+ jae .Lblocks_avx512
+.Lskip_avx512:
+ leaq 8(%rsp),%r10
+
+ subq $0x128,%rsp
+ leaq .Lconst(%rip),%rcx
+ leaq 48+64(%rdi),%rdi
+ vmovdqa 96(%rcx),%ymm7
+
+
+ vmovdqu -64(%rdi),%xmm9
+ andq $-512,%rsp
+ vmovdqu -48(%rdi),%xmm10
+ vmovdqu -32(%rdi),%xmm6
+ vmovdqu -16(%rdi),%xmm11
+ vmovdqu 0(%rdi),%xmm12
+ vmovdqu 16(%rdi),%xmm13
+ leaq 144(%rsp),%rax
+ vmovdqu 32(%rdi),%xmm14
+ vpermd %ymm9,%ymm7,%ymm9
+ vmovdqu 48(%rdi),%xmm15
+ vpermd %ymm10,%ymm7,%ymm10
+ vmovdqu 64(%rdi),%xmm5
+ vpermd %ymm6,%ymm7,%ymm6
+ vmovdqa %ymm9,0(%rsp)
+ vpermd %ymm11,%ymm7,%ymm11
+ vmovdqa %ymm10,32-144(%rax)
+ vpermd %ymm12,%ymm7,%ymm12
+ vmovdqa %ymm6,64-144(%rax)
+ vpermd %ymm13,%ymm7,%ymm13
+ vmovdqa %ymm11,96-144(%rax)
+ vpermd %ymm14,%ymm7,%ymm14
+ vmovdqa %ymm12,128-144(%rax)
+ vpermd %ymm15,%ymm7,%ymm15
+ vmovdqa %ymm13,160-144(%rax)
+ vpermd %ymm5,%ymm7,%ymm5
+ vmovdqa %ymm14,192-144(%rax)
+ vmovdqa %ymm15,224-144(%rax)
+ vmovdqa %ymm5,256-144(%rax)
+ vmovdqa 64(%rcx),%ymm5
+
+
+
+ vmovdqu 0(%rsi),%xmm7
+ vmovdqu 16(%rsi),%xmm8
+ vinserti128 $1,32(%rsi),%ymm7,%ymm7
+ vinserti128 $1,48(%rsi),%ymm8,%ymm8
+ leaq 64(%rsi),%rsi
+
+ vpsrldq $6,%ymm7,%ymm9
+ vpsrldq $6,%ymm8,%ymm10
+ vpunpckhqdq %ymm8,%ymm7,%ymm6
+ vpunpcklqdq %ymm10,%ymm9,%ymm9
+ vpunpcklqdq %ymm8,%ymm7,%ymm7
+
+ vpsrlq $30,%ymm9,%ymm10
+ vpsrlq $4,%ymm9,%ymm9
+ vpsrlq $26,%ymm7,%ymm8
+ vpsrlq $40,%ymm6,%ymm6
+ vpand %ymm5,%ymm9,%ymm9
+ vpand %ymm5,%ymm7,%ymm7
+ vpand %ymm5,%ymm8,%ymm8
+ vpand %ymm5,%ymm10,%ymm10
+ vpor 32(%rcx),%ymm6,%ymm6
+
+ vpaddq %ymm2,%ymm9,%ymm2
+ subq $64,%rdx
+ jz .Ltail_avx2_512
+ jmp .Loop_avx2_512
+
+.align 32
+.Loop_avx2_512:
+
+ vpaddq %ymm0,%ymm7,%ymm0
+ vmovdqa 0(%rsp),%ymm7
+ vpaddq %ymm1,%ymm8,%ymm1
+ vmovdqa 32(%rsp),%ymm8
+ vpaddq %ymm3,%ymm10,%ymm3
+ vmovdqa 96(%rsp),%ymm9
+ vpaddq %ymm4,%ymm6,%ymm4
+ vmovdqa 48(%rax),%ymm10
+ vmovdqa 112(%rax),%ymm5
+
+ vpmuludq %ymm2,%ymm7,%ymm13
+ vpmuludq %ymm2,%ymm8,%ymm14
+ vpmuludq %ymm2,%ymm9,%ymm15
+ vpmuludq %ymm2,%ymm10,%ymm11
+ vpmuludq %ymm2,%ymm5,%ymm12
+
+ vpmuludq %ymm0,%ymm8,%ymm6
+ vpmuludq %ymm1,%ymm8,%ymm2
+ vpaddq %ymm6,%ymm12,%ymm12
+ vpaddq %ymm2,%ymm13,%ymm13
+ vpmuludq %ymm3,%ymm8,%ymm6
+ vpmuludq 64(%rsp),%ymm4,%ymm2
+ vpaddq %ymm6,%ymm15,%ymm15
+ vpaddq %ymm2,%ymm11,%ymm11
+ vmovdqa -16(%rax),%ymm8
+
+ vpmuludq %ymm0,%ymm7,%ymm6
+ vpmuludq %ymm1,%ymm7,%ymm2
+ vpaddq %ymm6,%ymm11,%ymm11
+ vpaddq %ymm2,%ymm12,%ymm12
+ vpmuludq %ymm3,%ymm7,%ymm6
+ vpmuludq %ymm4,%ymm7,%ymm2
+ vmovdqu 0(%rsi),%xmm7
+ vpaddq %ymm6,%ymm14,%ymm14
+ vpaddq %ymm2,%ymm15,%ymm15
+ vinserti128 $1,32(%rsi),%ymm7,%ymm7
+
+ vpmuludq %ymm3,%ymm8,%ymm6
+ vpmuludq %ymm4,%ymm8,%ymm2
+ vmovdqu 16(%rsi),%xmm8
+ vpaddq %ymm6,%ymm11,%ymm11
+ vpaddq %ymm2,%ymm12,%ymm12
+ vmovdqa 16(%rax),%ymm2
+ vpmuludq %ymm1,%ymm9,%ymm6
+ vpmuludq %ymm0,%ymm9,%ymm9
+ vpaddq %ymm6,%ymm14,%ymm14
+ vpaddq %ymm9,%ymm13,%ymm13
+ vinserti128 $1,48(%rsi),%ymm8,%ymm8
+ leaq 64(%rsi),%rsi
+
+ vpmuludq %ymm1,%ymm2,%ymm6
+ vpmuludq %ymm0,%ymm2,%ymm2
+ vpsrldq $6,%ymm7,%ymm9
+ vpaddq %ymm6,%ymm15,%ymm15
+ vpaddq %ymm2,%ymm14,%ymm14
+ vpmuludq %ymm3,%ymm10,%ymm6
+ vpmuludq %ymm4,%ymm10,%ymm2
+ vpsrldq $6,%ymm8,%ymm10
+ vpaddq %ymm6,%ymm12,%ymm12
+ vpaddq %ymm2,%ymm13,%ymm13
+ vpunpckhqdq %ymm8,%ymm7,%ymm6
+
+ vpmuludq %ymm3,%ymm5,%ymm3
+ vpmuludq %ymm4,%ymm5,%ymm4
+ vpunpcklqdq %ymm8,%ymm7,%ymm7
+ vpaddq %ymm3,%ymm13,%ymm2
+ vpaddq %ymm4,%ymm14,%ymm3
+ vpunpcklqdq %ymm10,%ymm9,%ymm10
+ vpmuludq 80(%rax),%ymm0,%ymm4
+ vpmuludq %ymm1,%ymm5,%ymm0
+ vmovdqa 64(%rcx),%ymm5
+ vpaddq %ymm4,%ymm15,%ymm4
+ vpaddq %ymm0,%ymm11,%ymm0
+
+ vpsrlq $26,%ymm3,%ymm14
+ vpand %ymm5,%ymm3,%ymm3
+ vpaddq %ymm14,%ymm4,%ymm4
+
+ vpsrlq $26,%ymm0,%ymm11
+ vpand %ymm5,%ymm0,%ymm0
+ vpaddq %ymm11,%ymm12,%ymm1
+
+ vpsrlq $26,%ymm4,%ymm15
+ vpand %ymm5,%ymm4,%ymm4
+
+ vpsrlq $4,%ymm10,%ymm9
+
+ vpsrlq $26,%ymm1,%ymm12
+ vpand %ymm5,%ymm1,%ymm1
+ vpaddq %ymm12,%ymm2,%ymm2
+
+ vpaddq %ymm15,%ymm0,%ymm0
+ vpsllq $2,%ymm15,%ymm15
+ vpaddq %ymm15,%ymm0,%ymm0
+
+ vpand %ymm5,%ymm9,%ymm9
+ vpsrlq $26,%ymm7,%ymm8
+
+ vpsrlq $26,%ymm2,%ymm13
+ vpand %ymm5,%ymm2,%ymm2
+ vpaddq %ymm13,%ymm3,%ymm3
+
+ vpaddq %ymm9,%ymm2,%ymm2
+ vpsrlq $30,%ymm10,%ymm10
+
+ vpsrlq $26,%ymm0,%ymm11
+ vpand %ymm5,%ymm0,%ymm0
+ vpaddq %ymm11,%ymm1,%ymm1
+
+ vpsrlq $40,%ymm6,%ymm6
+
+ vpsrlq $26,%ymm3,%ymm14
+ vpand %ymm5,%ymm3,%ymm3
+ vpaddq %ymm14,%ymm4,%ymm4
+
+ vpand %ymm5,%ymm7,%ymm7
+ vpand %ymm5,%ymm8,%ymm8
+ vpand %ymm5,%ymm10,%ymm10
+ vpor 32(%rcx),%ymm6,%ymm6
+
+ subq $64,%rdx
+ jnz .Loop_avx2_512
+
+.byte 0x66,0x90
+.Ltail_avx2_512:
+
+ vpaddq %ymm0,%ymm7,%ymm0
+ vmovdqu 4(%rsp),%ymm7
+ vpaddq %ymm1,%ymm8,%ymm1
+ vmovdqu 36(%rsp),%ymm8
+ vpaddq %ymm3,%ymm10,%ymm3
+ vmovdqu 100(%rsp),%ymm9
+ vpaddq %ymm4,%ymm6,%ymm4
+ vmovdqu 52(%rax),%ymm10
+ vmovdqu 116(%rax),%ymm5
+
+ vpmuludq %ymm2,%ymm7,%ymm13
+ vpmuludq %ymm2,%ymm8,%ymm14
+ vpmuludq %ymm2,%ymm9,%ymm15
+ vpmuludq %ymm2,%ymm10,%ymm11
+ vpmuludq %ymm2,%ymm5,%ymm12
+
+ vpmuludq %ymm0,%ymm8,%ymm6
+ vpmuludq %ymm1,%ymm8,%ymm2
+ vpaddq %ymm6,%ymm12,%ymm12
+ vpaddq %ymm2,%ymm13,%ymm13
+ vpmuludq %ymm3,%ymm8,%ymm6
+ vpmuludq 68(%rsp),%ymm4,%ymm2
+ vpaddq %ymm6,%ymm15,%ymm15
+ vpaddq %ymm2,%ymm11,%ymm11
+
+ vpmuludq %ymm0,%ymm7,%ymm6
+ vpmuludq %ymm1,%ymm7,%ymm2
+ vpaddq %ymm6,%ymm11,%ymm11
+ vmovdqu -12(%rax),%ymm8
+ vpaddq %ymm2,%ymm12,%ymm12
+ vpmuludq %ymm3,%ymm7,%ymm6
+ vpmuludq %ymm4,%ymm7,%ymm2
+ vpaddq %ymm6,%ymm14,%ymm14
+ vpaddq %ymm2,%ymm15,%ymm15
+
+ vpmuludq %ymm3,%ymm8,%ymm6
+ vpmuludq %ymm4,%ymm8,%ymm2
+ vpaddq %ymm6,%ymm11,%ymm11
+ vpaddq %ymm2,%ymm12,%ymm12
+ vmovdqu 20(%rax),%ymm2
+ vpmuludq %ymm1,%ymm9,%ymm6
+ vpmuludq %ymm0,%ymm9,%ymm9
+ vpaddq %ymm6,%ymm14,%ymm14
+ vpaddq %ymm9,%ymm13,%ymm13
+
+ vpmuludq %ymm1,%ymm2,%ymm6
+ vpmuludq %ymm0,%ymm2,%ymm2
+ vpaddq %ymm6,%ymm15,%ymm15
+ vpaddq %ymm2,%ymm14,%ymm14
+ vpmuludq %ymm3,%ymm10,%ymm6
+ vpmuludq %ymm4,%ymm10,%ymm2
+ vpaddq %ymm6,%ymm12,%ymm12
+ vpaddq %ymm2,%ymm13,%ymm13
+
+ vpmuludq %ymm3,%ymm5,%ymm3
+ vpmuludq %ymm4,%ymm5,%ymm4
+ vpaddq %ymm3,%ymm13,%ymm2
+ vpaddq %ymm4,%ymm14,%ymm3
+ vpmuludq 84(%rax),%ymm0,%ymm4
+ vpmuludq %ymm1,%ymm5,%ymm0
+ vmovdqa 64(%rcx),%ymm5
+ vpaddq %ymm4,%ymm15,%ymm4
+ vpaddq %ymm0,%ymm11,%ymm0
+
+ vpsrldq $8,%ymm12,%ymm8
+ vpsrldq $8,%ymm2,%ymm9
+ vpsrldq $8,%ymm3,%ymm10
+ vpsrldq $8,%ymm4,%ymm6
+ vpsrldq $8,%ymm0,%ymm7
+ vpaddq %ymm8,%ymm12,%ymm12
+ vpaddq %ymm9,%ymm2,%ymm2
+ vpaddq %ymm10,%ymm3,%ymm3
+ vpaddq %ymm6,%ymm4,%ymm4
+ vpaddq %ymm7,%ymm0,%ymm0
+
+ vpermq $0x2,%ymm3,%ymm10
+ vpermq $0x2,%ymm4,%ymm6
+ vpermq $0x2,%ymm0,%ymm7
+ vpermq $0x2,%ymm12,%ymm8
+ vpermq $0x2,%ymm2,%ymm9
+ vpaddq %ymm10,%ymm3,%ymm3
+ vpaddq %ymm6,%ymm4,%ymm4
+ vpaddq %ymm7,%ymm0,%ymm0
+ vpaddq %ymm8,%ymm12,%ymm12
+ vpaddq %ymm9,%ymm2,%ymm2
+
+ vpsrlq $26,%ymm3,%ymm14
+ vpand %ymm5,%ymm3,%ymm3
+ vpaddq %ymm14,%ymm4,%ymm4
+
+ vpsrlq $26,%ymm0,%ymm11
+ vpand %ymm5,%ymm0,%ymm0
+ vpaddq %ymm11,%ymm12,%ymm1
+
+ vpsrlq $26,%ymm4,%ymm15
+ vpand %ymm5,%ymm4,%ymm4
+
+ vpsrlq $26,%ymm1,%ymm12
+ vpand %ymm5,%ymm1,%ymm1
+ vpaddq %ymm12,%ymm2,%ymm2
+
+ vpaddq %ymm15,%ymm0,%ymm0
+ vpsllq $2,%ymm15,%ymm15
+ vpaddq %ymm15,%ymm0,%ymm0
+
+ vpsrlq $26,%ymm2,%ymm13
+ vpand %ymm5,%ymm2,%ymm2
+ vpaddq %ymm13,%ymm3,%ymm3
+
+ vpsrlq $26,%ymm0,%ymm11
+ vpand %ymm5,%ymm0,%ymm0
+ vpaddq %ymm11,%ymm1,%ymm1
+
+ vpsrlq $26,%ymm3,%ymm14
+ vpand %ymm5,%ymm3,%ymm3
+ vpaddq %ymm14,%ymm4,%ymm4
+
+ vmovd %xmm0,-112(%rdi)
+ vmovd %xmm1,-108(%rdi)
+ vmovd %xmm2,-104(%rdi)
+ vmovd %xmm3,-100(%rdi)
+ vmovd %xmm4,-96(%rdi)
+ leaq -8(%r10),%rsp
+
+ vzeroupper
+ ret
+
+.Lblocks_avx512:
+
+ movl $15,%eax
+ kmovw %eax,%k2
+ leaq 8(%rsp),%r10
+
+ subq $0x128,%rsp
+ leaq .Lconst(%rip),%rcx
+ leaq 48+64(%rdi),%rdi
+ vmovdqa 96(%rcx),%ymm9
+
+ vmovdqu32 -64(%rdi),%zmm16{%k2}{z}
+ andq $-512,%rsp
+ vmovdqu32 -48(%rdi),%zmm17{%k2}{z}
+ movq $0x20,%rax
+ vmovdqu32 -32(%rdi),%zmm21{%k2}{z}
+ vmovdqu32 -16(%rdi),%zmm18{%k2}{z}
+ vmovdqu32 0(%rdi),%zmm22{%k2}{z}
+ vmovdqu32 16(%rdi),%zmm19{%k2}{z}
+ vmovdqu32 32(%rdi),%zmm23{%k2}{z}
+ vmovdqu32 48(%rdi),%zmm20{%k2}{z}
+ vmovdqu32 64(%rdi),%zmm24{%k2}{z}
+ vpermd %zmm16,%zmm9,%zmm16
+ vpbroadcastq 64(%rcx),%zmm5
+ vpermd %zmm17,%zmm9,%zmm17
+ vpermd %zmm21,%zmm9,%zmm21
+ vpermd %zmm18,%zmm9,%zmm18
+ vmovdqa64 %zmm16,0(%rsp){%k2}
+ vpsrlq $32,%zmm16,%zmm7
+ vpermd %zmm22,%zmm9,%zmm22
+ vmovdqu64 %zmm17,0(%rsp,%rax,1){%k2}
+ vpsrlq $32,%zmm17,%zmm8
+ vpermd %zmm19,%zmm9,%zmm19
+ vmovdqa64 %zmm21,64(%rsp){%k2}
+ vpermd %zmm23,%zmm9,%zmm23
+ vpermd %zmm20,%zmm9,%zmm20
+ vmovdqu64 %zmm18,64(%rsp,%rax,1){%k2}
+ vpermd %zmm24,%zmm9,%zmm24
+ vmovdqa64 %zmm22,128(%rsp){%k2}
+ vmovdqu64 %zmm19,128(%rsp,%rax,1){%k2}
+ vmovdqa64 %zmm23,192(%rsp){%k2}
+ vmovdqu64 %zmm20,192(%rsp,%rax,1){%k2}
+ vmovdqa64 %zmm24,256(%rsp){%k2}
+
+ vpmuludq %zmm7,%zmm16,%zmm11
+ vpmuludq %zmm7,%zmm17,%zmm12
+ vpmuludq %zmm7,%zmm18,%zmm13
+ vpmuludq %zmm7,%zmm19,%zmm14
+ vpmuludq %zmm7,%zmm20,%zmm15
+ vpsrlq $32,%zmm18,%zmm9
+
+ vpmuludq %zmm8,%zmm24,%zmm25
+ vpmuludq %zmm8,%zmm16,%zmm26
+ vpmuludq %zmm8,%zmm17,%zmm27
+ vpmuludq %zmm8,%zmm18,%zmm28
+ vpmuludq %zmm8,%zmm19,%zmm29
+ vpsrlq $32,%zmm19,%zmm10
+ vpaddq %zmm25,%zmm11,%zmm11
+ vpaddq %zmm26,%zmm12,%zmm12
+ vpaddq %zmm27,%zmm13,%zmm13
+ vpaddq %zmm28,%zmm14,%zmm14
+ vpaddq %zmm29,%zmm15,%zmm15
+
+ vpmuludq %zmm9,%zmm23,%zmm25
+ vpmuludq %zmm9,%zmm24,%zmm26
+ vpmuludq %zmm9,%zmm17,%zmm28
+ vpmuludq %zmm9,%zmm18,%zmm29
+ vpmuludq %zmm9,%zmm16,%zmm27
+ vpsrlq $32,%zmm20,%zmm6
+ vpaddq %zmm25,%zmm11,%zmm11
+ vpaddq %zmm26,%zmm12,%zmm12
+ vpaddq %zmm28,%zmm14,%zmm14
+ vpaddq %zmm29,%zmm15,%zmm15
+ vpaddq %zmm27,%zmm13,%zmm13
+
+ vpmuludq %zmm10,%zmm22,%zmm25
+ vpmuludq %zmm10,%zmm16,%zmm28
+ vpmuludq %zmm10,%zmm17,%zmm29
+ vpmuludq %zmm10,%zmm23,%zmm26
+ vpmuludq %zmm10,%zmm24,%zmm27
+ vpaddq %zmm25,%zmm11,%zmm11
+ vpaddq %zmm28,%zmm14,%zmm14
+ vpaddq %zmm29,%zmm15,%zmm15
+ vpaddq %zmm26,%zmm12,%zmm12
+ vpaddq %zmm27,%zmm13,%zmm13
+
+ vpmuludq %zmm6,%zmm24,%zmm28
+ vpmuludq %zmm6,%zmm16,%zmm29
+ vpmuludq %zmm6,%zmm21,%zmm25
+ vpmuludq %zmm6,%zmm22,%zmm26
+ vpmuludq %zmm6,%zmm23,%zmm27
+ vpaddq %zmm28,%zmm14,%zmm14
+ vpaddq %zmm29,%zmm15,%zmm15
+ vpaddq %zmm25,%zmm11,%zmm11
+ vpaddq %zmm26,%zmm12,%zmm12
+ vpaddq %zmm27,%zmm13,%zmm13
+
+ vmovdqu64 0(%rsi),%zmm10
+ vmovdqu64 64(%rsi),%zmm6
+ leaq 128(%rsi),%rsi
+
+ vpsrlq $26,%zmm14,%zmm28
+ vpandq %zmm5,%zmm14,%zmm14
+ vpaddq %zmm28,%zmm15,%zmm15
+
+ vpsrlq $26,%zmm11,%zmm25
+ vpandq %zmm5,%zmm11,%zmm11
+ vpaddq %zmm25,%zmm12,%zmm12
+
+ vpsrlq $26,%zmm15,%zmm29
+ vpandq %zmm5,%zmm15,%zmm15
+
+ vpsrlq $26,%zmm12,%zmm26
+ vpandq %zmm5,%zmm12,%zmm12
+ vpaddq %zmm26,%zmm13,%zmm13
+
+ vpaddq %zmm29,%zmm11,%zmm11
+ vpsllq $2,%zmm29,%zmm29
+ vpaddq %zmm29,%zmm11,%zmm11
+
+ vpsrlq $26,%zmm13,%zmm27
+ vpandq %zmm5,%zmm13,%zmm13
+ vpaddq %zmm27,%zmm14,%zmm14
+
+ vpsrlq $26,%zmm11,%zmm25
+ vpandq %zmm5,%zmm11,%zmm11
+ vpaddq %zmm25,%zmm12,%zmm12
+
+ vpsrlq $26,%zmm14,%zmm28
+ vpandq %zmm5,%zmm14,%zmm14
+ vpaddq %zmm28,%zmm15,%zmm15
+
+ vpunpcklqdq %zmm6,%zmm10,%zmm7
+ vpunpckhqdq %zmm6,%zmm10,%zmm6
+
+ vmovdqa32 128(%rcx),%zmm25
+ movl $0x7777,%eax
+ kmovw %eax,%k1
+
+ vpermd %zmm16,%zmm25,%zmm16
+ vpermd %zmm17,%zmm25,%zmm17
+ vpermd %zmm18,%zmm25,%zmm18
+ vpermd %zmm19,%zmm25,%zmm19
+ vpermd %zmm20,%zmm25,%zmm20
+
+ vpermd %zmm11,%zmm25,%zmm16{%k1}
+ vpermd %zmm12,%zmm25,%zmm17{%k1}
+ vpermd %zmm13,%zmm25,%zmm18{%k1}
+ vpermd %zmm14,%zmm25,%zmm19{%k1}
+ vpermd %zmm15,%zmm25,%zmm20{%k1}
+
+ vpslld $2,%zmm17,%zmm21
+ vpslld $2,%zmm18,%zmm22
+ vpslld $2,%zmm19,%zmm23
+ vpslld $2,%zmm20,%zmm24
+ vpaddd %zmm17,%zmm21,%zmm21
+ vpaddd %zmm18,%zmm22,%zmm22
+ vpaddd %zmm19,%zmm23,%zmm23
+ vpaddd %zmm20,%zmm24,%zmm24
+
+ vpbroadcastq 32(%rcx),%zmm30
+
+ vpsrlq $52,%zmm7,%zmm9
+ vpsllq $12,%zmm6,%zmm10
+ vporq %zmm10,%zmm9,%zmm9
+ vpsrlq $26,%zmm7,%zmm8
+ vpsrlq $14,%zmm6,%zmm10
+ vpsrlq $40,%zmm6,%zmm6
+ vpandq %zmm5,%zmm9,%zmm9
+ vpandq %zmm5,%zmm7,%zmm7
+
+ vpaddq %zmm2,%zmm9,%zmm2
+ subq $192,%rdx
+ jbe .Ltail_avx512
+ jmp .Loop_avx512
+
+.align 32
+.Loop_avx512:
+
+ vpmuludq %zmm2,%zmm17,%zmm14
+ vpaddq %zmm0,%zmm7,%zmm0
+ vpmuludq %zmm2,%zmm18,%zmm15
+ vpandq %zmm5,%zmm8,%zmm8
+ vpmuludq %zmm2,%zmm23,%zmm11
+ vpandq %zmm5,%zmm10,%zmm10
+ vpmuludq %zmm2,%zmm24,%zmm12
+ vporq %zmm30,%zmm6,%zmm6
+ vpmuludq %zmm2,%zmm16,%zmm13
+ vpaddq %zmm1,%zmm8,%zmm1
+ vpaddq %zmm3,%zmm10,%zmm3
+ vpaddq %zmm4,%zmm6,%zmm4
+
+ vmovdqu64 0(%rsi),%zmm10
+ vmovdqu64 64(%rsi),%zmm6
+ leaq 128(%rsi),%rsi
+ vpmuludq %zmm0,%zmm19,%zmm28
+ vpmuludq %zmm0,%zmm20,%zmm29
+ vpmuludq %zmm0,%zmm16,%zmm25
+ vpmuludq %zmm0,%zmm17,%zmm26
+ vpaddq %zmm28,%zmm14,%zmm14
+ vpaddq %zmm29,%zmm15,%zmm15
+ vpaddq %zmm25,%zmm11,%zmm11
+ vpaddq %zmm26,%zmm12,%zmm12
+
+ vpmuludq %zmm1,%zmm18,%zmm28
+ vpmuludq %zmm1,%zmm19,%zmm29
+ vpmuludq %zmm1,%zmm24,%zmm25
+ vpmuludq %zmm0,%zmm18,%zmm27
+ vpaddq %zmm28,%zmm14,%zmm14
+ vpaddq %zmm29,%zmm15,%zmm15
+ vpaddq %zmm25,%zmm11,%zmm11
+ vpaddq %zmm27,%zmm13,%zmm13
+
+ vpunpcklqdq %zmm6,%zmm10,%zmm7
+ vpunpckhqdq %zmm6,%zmm10,%zmm6
+
+ vpmuludq %zmm3,%zmm16,%zmm28
+ vpmuludq %zmm3,%zmm17,%zmm29
+ vpmuludq %zmm1,%zmm16,%zmm26
+ vpmuludq %zmm1,%zmm17,%zmm27
+ vpaddq %zmm28,%zmm14,%zmm14
+ vpaddq %zmm29,%zmm15,%zmm15
+ vpaddq %zmm26,%zmm12,%zmm12
+ vpaddq %zmm27,%zmm13,%zmm13
+
+ vpmuludq %zmm4,%zmm24,%zmm28
+ vpmuludq %zmm4,%zmm16,%zmm29
+ vpmuludq %zmm3,%zmm22,%zmm25
+ vpmuludq %zmm3,%zmm23,%zmm26
+ vpaddq %zmm28,%zmm14,%zmm14
+ vpmuludq %zmm3,%zmm24,%zmm27
+ vpaddq %zmm29,%zmm15,%zmm15
+ vpaddq %zmm25,%zmm11,%zmm11
+ vpaddq %zmm26,%zmm12,%zmm12
+ vpaddq %zmm27,%zmm13,%zmm13
+
+ vpmuludq %zmm4,%zmm21,%zmm25
+ vpmuludq %zmm4,%zmm22,%zmm26
+ vpmuludq %zmm4,%zmm23,%zmm27
+ vpaddq %zmm25,%zmm11,%zmm0
+ vpaddq %zmm26,%zmm12,%zmm1
+ vpaddq %zmm27,%zmm13,%zmm2
+
+ vpsrlq $52,%zmm7,%zmm9
+ vpsllq $12,%zmm6,%zmm10
+
+ vpsrlq $26,%zmm14,%zmm3
+ vpandq %zmm5,%zmm14,%zmm14
+ vpaddq %zmm3,%zmm15,%zmm4
+
+ vporq %zmm10,%zmm9,%zmm9
+
+ vpsrlq $26,%zmm0,%zmm11
+ vpandq %zmm5,%zmm0,%zmm0
+ vpaddq %zmm11,%zmm1,%zmm1
+
+ vpandq %zmm5,%zmm9,%zmm9
+
+ vpsrlq $26,%zmm4,%zmm15
+ vpandq %zmm5,%zmm4,%zmm4
+
+ vpsrlq $26,%zmm1,%zmm12
+ vpandq %zmm5,%zmm1,%zmm1
+ vpaddq %zmm12,%zmm2,%zmm2
+
+ vpaddq %zmm15,%zmm0,%zmm0
+ vpsllq $2,%zmm15,%zmm15
+ vpaddq %zmm15,%zmm0,%zmm0
+
+ vpaddq %zmm9,%zmm2,%zmm2
+ vpsrlq $26,%zmm7,%zmm8
+
+ vpsrlq $26,%zmm2,%zmm13
+ vpandq %zmm5,%zmm2,%zmm2
+ vpaddq %zmm13,%zmm14,%zmm3
+
+ vpsrlq $14,%zmm6,%zmm10
+
+ vpsrlq $26,%zmm0,%zmm11
+ vpandq %zmm5,%zmm0,%zmm0
+ vpaddq %zmm11,%zmm1,%zmm1
+
+ vpsrlq $40,%zmm6,%zmm6
+
+ vpsrlq $26,%zmm3,%zmm14
+ vpandq %zmm5,%zmm3,%zmm3
+ vpaddq %zmm14,%zmm4,%zmm4
+
+ vpandq %zmm5,%zmm7,%zmm7
+
+ subq $128,%rdx
+ ja .Loop_avx512
+
+.Ltail_avx512:
+
+ vpsrlq $32,%zmm16,%zmm16
+ vpsrlq $32,%zmm17,%zmm17
+ vpsrlq $32,%zmm18,%zmm18
+ vpsrlq $32,%zmm23,%zmm23
+ vpsrlq $32,%zmm24,%zmm24
+ vpsrlq $32,%zmm19,%zmm19
+ vpsrlq $32,%zmm20,%zmm20
+ vpsrlq $32,%zmm21,%zmm21
+ vpsrlq $32,%zmm22,%zmm22
+
+ leaq (%rsi,%rdx,1),%rsi
+
+ vpaddq %zmm0,%zmm7,%zmm0
+
+ vpmuludq %zmm2,%zmm17,%zmm14
+ vpmuludq %zmm2,%zmm18,%zmm15
+ vpmuludq %zmm2,%zmm23,%zmm11
+ vpandq %zmm5,%zmm8,%zmm8
+ vpmuludq %zmm2,%zmm24,%zmm12
+ vpandq %zmm5,%zmm10,%zmm10
+ vpmuludq %zmm2,%zmm16,%zmm13
+ vporq %zmm30,%zmm6,%zmm6
+ vpaddq %zmm1,%zmm8,%zmm1
+ vpaddq %zmm3,%zmm10,%zmm3
+ vpaddq %zmm4,%zmm6,%zmm4
+
+ vmovdqu 0(%rsi),%xmm7
+ vpmuludq %zmm0,%zmm19,%zmm28
+ vpmuludq %zmm0,%zmm20,%zmm29
+ vpmuludq %zmm0,%zmm16,%zmm25
+ vpmuludq %zmm0,%zmm17,%zmm26
+ vpaddq %zmm28,%zmm14,%zmm14
+ vpaddq %zmm29,%zmm15,%zmm15
+ vpaddq %zmm25,%zmm11,%zmm11
+ vpaddq %zmm26,%zmm12,%zmm12
+
+ vmovdqu 16(%rsi),%xmm8
+ vpmuludq %zmm1,%zmm18,%zmm28
+ vpmuludq %zmm1,%zmm19,%zmm29
+ vpmuludq %zmm1,%zmm24,%zmm25
+ vpmuludq %zmm0,%zmm18,%zmm27
+ vpaddq %zmm28,%zmm14,%zmm14
+ vpaddq %zmm29,%zmm15,%zmm15
+ vpaddq %zmm25,%zmm11,%zmm11
+ vpaddq %zmm27,%zmm13,%zmm13
+
+ vinserti128 $1,32(%rsi),%ymm7,%ymm7
+ vpmuludq %zmm3,%zmm16,%zmm28
+ vpmuludq %zmm3,%zmm17,%zmm29
+ vpmuludq %zmm1,%zmm16,%zmm26
+ vpmuludq %zmm1,%zmm17,%zmm27
+ vpaddq %zmm28,%zmm14,%zmm14
+ vpaddq %zmm29,%zmm15,%zmm15
+ vpaddq %zmm26,%zmm12,%zmm12
+ vpaddq %zmm27,%zmm13,%zmm13
+
+ vinserti128 $1,48(%rsi),%ymm8,%ymm8
+ vpmuludq %zmm4,%zmm24,%zmm28
+ vpmuludq %zmm4,%zmm16,%zmm29
+ vpmuludq %zmm3,%zmm22,%zmm25
+ vpmuludq %zmm3,%zmm23,%zmm26
+ vpmuludq %zmm3,%zmm24,%zmm27
+ vpaddq %zmm28,%zmm14,%zmm3
+ vpaddq %zmm29,%zmm15,%zmm15
+ vpaddq %zmm25,%zmm11,%zmm11
+ vpaddq %zmm26,%zmm12,%zmm12
+ vpaddq %zmm27,%zmm13,%zmm13
+
+ vpmuludq %zmm4,%zmm21,%zmm25
+ vpmuludq %zmm4,%zmm22,%zmm26
+ vpmuludq %zmm4,%zmm23,%zmm27
+ vpaddq %zmm25,%zmm11,%zmm0
+ vpaddq %zmm26,%zmm12,%zmm1
+ vpaddq %zmm27,%zmm13,%zmm2
+
+ movl $1,%eax
+ vpermq $0xb1,%zmm3,%zmm14
+ vpermq $0xb1,%zmm15,%zmm4
+ vpermq $0xb1,%zmm0,%zmm11
+ vpermq $0xb1,%zmm1,%zmm12
+ vpermq $0xb1,%zmm2,%zmm13
+ vpaddq %zmm14,%zmm3,%zmm3
+ vpaddq %zmm15,%zmm4,%zmm4
+ vpaddq %zmm11,%zmm0,%zmm0
+ vpaddq %zmm12,%zmm1,%zmm1
+ vpaddq %zmm13,%zmm2,%zmm2
+
+ kmovw %eax,%k3
+ vpermq $0x2,%zmm3,%zmm14
+ vpermq $0x2,%zmm4,%zmm15
+ vpermq $0x2,%zmm0,%zmm11
+ vpermq $0x2,%zmm1,%zmm12
+ vpermq $0x2,%zmm2,%zmm13
+ vpaddq %zmm14,%zmm3,%zmm3
+ vpaddq %zmm15,%zmm4,%zmm4
+ vpaddq %zmm11,%zmm0,%zmm0
+ vpaddq %zmm12,%zmm1,%zmm1
+ vpaddq %zmm13,%zmm2,%zmm2
+
+ vextracti64x4 $0x1,%zmm3,%ymm14
+ vextracti64x4 $0x1,%zmm4,%ymm15
+ vextracti64x4 $0x1,%zmm0,%ymm11
+ vextracti64x4 $0x1,%zmm1,%ymm12
+ vextracti64x4 $0x1,%zmm2,%ymm13
+ vpaddq %zmm14,%zmm3,%zmm3{%k3}{z}
+ vpaddq %zmm15,%zmm4,%zmm4{%k3}{z}
+ vpaddq %zmm11,%zmm0,%zmm0{%k3}{z}
+ vpaddq %zmm12,%zmm1,%zmm1{%k3}{z}
+ vpaddq %zmm13,%zmm2,%zmm2{%k3}{z}
+
+ vpsrlq $26,%ymm3,%ymm14
+ vpand %ymm5,%ymm3,%ymm3
+ vpsrldq $6,%ymm7,%ymm9
+ vpsrldq $6,%ymm8,%ymm10
+ vpunpckhqdq %ymm8,%ymm7,%ymm6
+ vpaddq %ymm14,%ymm4,%ymm4
+
+ vpsrlq $26,%ymm0,%ymm11
+ vpand %ymm5,%ymm0,%ymm0
+ vpunpcklqdq %ymm10,%ymm9,%ymm9
+ vpunpcklqdq %ymm8,%ymm7,%ymm7
+ vpaddq %ymm11,%ymm1,%ymm1
+
+ vpsrlq $26,%ymm4,%ymm15
+ vpand %ymm5,%ymm4,%ymm4
+
+ vpsrlq $26,%ymm1,%ymm12
+ vpand %ymm5,%ymm1,%ymm1
+ vpsrlq $30,%ymm9,%ymm10
+ vpsrlq $4,%ymm9,%ymm9
+ vpaddq %ymm12,%ymm2,%ymm2
+
+ vpaddq %ymm15,%ymm0,%ymm0
+ vpsllq $2,%ymm15,%ymm15
+ vpsrlq $26,%ymm7,%ymm8
+ vpsrlq $40,%ymm6,%ymm6
+ vpaddq %ymm15,%ymm0,%ymm0
+
+ vpsrlq $26,%ymm2,%ymm13
+ vpand %ymm5,%ymm2,%ymm2
+ vpand %ymm5,%ymm9,%ymm9
+ vpand %ymm5,%ymm7,%ymm7
+ vpaddq %ymm13,%ymm3,%ymm3
+
+ vpsrlq $26,%ymm0,%ymm11
+ vpand %ymm5,%ymm0,%ymm0
+ vpaddq %ymm2,%ymm9,%ymm2
+ vpand %ymm5,%ymm8,%ymm8
+ vpaddq %ymm11,%ymm1,%ymm1
+
+ vpsrlq $26,%ymm3,%ymm14
+ vpand %ymm5,%ymm3,%ymm3
+ vpand %ymm5,%ymm10,%ymm10
+ vpor 32(%rcx),%ymm6,%ymm6
+ vpaddq %ymm14,%ymm4,%ymm4
+
+ leaq 144(%rsp),%rax
+ addq $64,%rdx
+ jnz .Ltail_avx2_512
+
+ vpsubq %ymm9,%ymm2,%ymm2
+ vmovd %xmm0,-112(%rdi)
+ vmovd %xmm1,-108(%rdi)
+ vmovd %xmm2,-104(%rdi)
+ vmovd %xmm3,-100(%rdi)
+ vmovd %xmm4,-96(%rdi)
+ vzeroall
+ leaq -8(%r10),%rsp
+
+ ret
+
+ENDPROC(poly1305_blocks_avx512)
+#endif /* CONFIG_AS_AVX512 */
--
2.18.0
^ permalink raw reply related
* [PATCH v2 08/17] zinc: Poly1305 ARM and ARM64 implementations
From: Jason A. Donenfeld @ 2018-08-24 21:38 UTC (permalink / raw)
To: linux-kernel, netdev, davem
Cc: Jason A. Donenfeld, Andy Lutomirski, Greg KH, Samuel Neves,
Jean-Philippe Aumasson, Andy Polyakov, Russell King,
linux-arm-kernel, linux-crypto
In-Reply-To: <20180824213849.23647-1-Jason@zx2c4.com>
These NEON and non-NEON implementations come from Andy Polyakov's
implementation, with several modifications for being friendly to kernel
space.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Greg KH <gregkh@linuxfoundation.org>
Cc: Samuel Neves <sneves@dei.uc.pt>
Cc: Jean-Philippe Aumasson <jeanphilippe.aumasson@gmail.com>
Cc: Andy Polyakov <appro@openssl.org>
Cc: Russell King <linux@armlinux.org.uk>
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-crypto@vger.kernel.org
---
lib/zinc/Makefile | 8 +
lib/zinc/poly1305/poly1305-arm-glue.h | 58 ++
lib/zinc/poly1305/poly1305-arm.S | 1115 +++++++++++++++++++++++++
lib/zinc/poly1305/poly1305-arm64.S | 820 ++++++++++++++++++
4 files changed, 2001 insertions(+)
create mode 100644 lib/zinc/poly1305/poly1305-arm-glue.h
create mode 100644 lib/zinc/poly1305/poly1305-arm.S
create mode 100644 lib/zinc/poly1305/poly1305-arm64.S
diff --git a/lib/zinc/Makefile b/lib/zinc/Makefile
index bf7b3a75f4c2..db70ffedf7dd 100644
--- a/lib/zinc/Makefile
+++ b/lib/zinc/Makefile
@@ -24,6 +24,14 @@ endif
ifeq ($(CONFIG_ZINC_POLY1305),y)
zinc-y += poly1305/poly1305.o
+ifeq ($(CONFIG_ARM),y)
+zinc-y += poly1305/poly1305-arm.o
+CFLAGS_poly1305.o += -include $(srctree)/$(src)/poly1305/poly1305-arm-glue.h
+endif
+ifeq ($(CONFIG_ARM64),y)
+zinc-y += poly1305/poly1305-arm64.o
+CFLAGS_poly1305.o += -include $(srctree)/$(src)/poly1305/poly1305-arm-glue.h
+endif
endif
zinc-y += main.o
diff --git a/lib/zinc/poly1305/poly1305-arm-glue.h b/lib/zinc/poly1305/poly1305-arm-glue.h
new file mode 100644
index 000000000000..38fa40f26393
--- /dev/null
+++ b/lib/zinc/poly1305/poly1305-arm-glue.h
@@ -0,0 +1,58 @@
+/* SPDX-License-Identifier: GPL-2.0
+ *
+ * Copyright (C) 2015-2018 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include <zinc/poly1305.h>
+
+asmlinkage void poly1305_init_arm(void *ctx, const u8 key[16]);
+asmlinkage void poly1305_blocks_arm(void *ctx, const u8 *inp, const size_t len, const u32 padbit);
+asmlinkage void poly1305_emit_arm(void *ctx, u8 mac[16], const u32 nonce[4]);
+#if IS_ENABLED(CONFIG_KERNEL_MODE_NEON) && (!defined(__LINUX_ARM_ARCH__) || __LINUX_ARM_ARCH__ >= 7)
+#define ARM_USE_NEON
+#include <asm/hwcap.h>
+#include <asm/neon.h>
+asmlinkage void poly1305_blocks_neon(void *ctx, const u8 *inp, const size_t len, const u32 padbit);
+asmlinkage void poly1305_emit_neon(void *ctx, u8 mac[16], const u32 nonce[4]);
+#endif
+
+static bool poly1305_use_neon __ro_after_init;
+
+void __init poly1305_fpu_init(void)
+{
+#if defined(CONFIG_ARM64)
+ poly1305_use_neon = elf_hwcap & HWCAP_ASIMD;
+#elif defined(CONFIG_ARM)
+ poly1305_use_neon = elf_hwcap & HWCAP_NEON;
+#endif
+}
+
+static inline bool poly1305_init_arch(void *ctx, const u8 key[POLY1305_KEY_SIZE], simd_context_t simd_context)
+{
+ poly1305_init_arm(ctx, key);
+ return true;
+}
+
+static inline bool poly1305_blocks_arch(void *ctx, const u8 *inp, const size_t len, const u32 padbit, simd_context_t simd_context)
+{
+#if defined(ARM_USE_NEON)
+ if (poly1305_use_neon && simd_context == HAVE_FULL_SIMD)
+ poly1305_blocks_neon(ctx, inp, len, padbit);
+ else
+#endif
+ poly1305_blocks_arm(ctx, inp, len, padbit);
+ return true;
+}
+
+static inline bool poly1305_emit_arch(void *ctx, u8 mac[POLY1305_MAC_SIZE], const u32 nonce[4], simd_context_t simd_context)
+{
+#if defined(ARM_USE_NEON)
+ if (poly1305_use_neon && simd_context == HAVE_FULL_SIMD)
+ poly1305_emit_neon(ctx, mac, nonce);
+ else
+#endif
+ poly1305_emit_arm(ctx, mac, nonce);
+ return true;
+}
+
+#define HAVE_POLY1305_ARCH_IMPLEMENTATION
diff --git a/lib/zinc/poly1305/poly1305-arm.S b/lib/zinc/poly1305/poly1305-arm.S
new file mode 100644
index 000000000000..d6b9a80dcc8b
--- /dev/null
+++ b/lib/zinc/poly1305/poly1305-arm.S
@@ -0,0 +1,1115 @@
+/* SPDX-License-Identifier: OpenSSL OR (BSD-3-Clause OR GPL-2.0)
+ *
+ * Copyright (C) 2015-2018 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ * Copyright 2016 The OpenSSL Project Authors. All Rights Reserved.
+ */
+
+#include <linux/linkage.h>
+
+.text
+#if defined(__thumb2__)
+.syntax unified
+.thumb
+#else
+.code 32
+#endif
+
+.align 5
+ENTRY(poly1305_init_arm)
+ stmdb sp!,{r4-r11}
+
+ eor r3,r3,r3
+ cmp r1,#0
+ str r3,[r0,#0] @ zero hash value
+ str r3,[r0,#4]
+ str r3,[r0,#8]
+ str r3,[r0,#12]
+ str r3,[r0,#16]
+ str r3,[r0,#36] @ is_base2_26
+ add r0,r0,#20
+
+#ifdef __thumb2__
+ it eq
+#endif
+ moveq r0,#0
+ beq .Lno_key
+
+ ldrb r4,[r1,#0]
+ mov r10,#0x0fffffff
+ ldrb r5,[r1,#1]
+ and r3,r10,#-4 @ 0x0ffffffc
+ ldrb r6,[r1,#2]
+ ldrb r7,[r1,#3]
+ orr r4,r4,r5,lsl#8
+ ldrb r5,[r1,#4]
+ orr r4,r4,r6,lsl#16
+ ldrb r6,[r1,#5]
+ orr r4,r4,r7,lsl#24
+ ldrb r7,[r1,#6]
+ and r4,r4,r10
+
+ ldrb r8,[r1,#7]
+ orr r5,r5,r6,lsl#8
+ ldrb r6,[r1,#8]
+ orr r5,r5,r7,lsl#16
+ ldrb r7,[r1,#9]
+ orr r5,r5,r8,lsl#24
+ ldrb r8,[r1,#10]
+ and r5,r5,r3
+
+ ldrb r9,[r1,#11]
+ orr r6,r6,r7,lsl#8
+ ldrb r7,[r1,#12]
+ orr r6,r6,r8,lsl#16
+ ldrb r8,[r1,#13]
+ orr r6,r6,r9,lsl#24
+ ldrb r9,[r1,#14]
+ and r6,r6,r3
+
+ ldrb r10,[r1,#15]
+ orr r7,r7,r8,lsl#8
+ str r4,[r0,#0]
+ orr r7,r7,r9,lsl#16
+ str r5,[r0,#4]
+ orr r7,r7,r10,lsl#24
+ str r6,[r0,#8]
+ and r7,r7,r3
+ str r7,[r0,#12]
+.Lno_key:
+ ldmia sp!,{r4-r11}
+#if __LINUX_ARM_ARCH__ >= 5
+ bx lr @ bx lr
+#else
+ tst lr,#1
+ moveq pc,lr @ be binary compatible with V4, yet
+ .word 0xe12fff1e @ interoperable with Thumb ISA:-)
+#endif
+ENDPROC(poly1305_init_arm)
+
+.align 5
+ENTRY(poly1305_blocks_arm)
+.Lpoly1305_blocks_arm:
+ stmdb sp!,{r3-r11,lr}
+
+ ands r2,r2,#-16
+ beq .Lno_data
+
+ cmp r3,#0
+ add r2,r2,r1 @ end pointer
+ sub sp,sp,#32
+
+ ldmia r0,{r4-r12} @ load context
+
+ str r0,[sp,#12] @ offload stuff
+ mov lr,r1
+ str r2,[sp,#16]
+ str r10,[sp,#20]
+ str r11,[sp,#24]
+ str r12,[sp,#28]
+ b .Loop
+
+.Loop:
+#if __LINUX_ARM_ARCH__ < 7
+ ldrb r0,[lr],#16 @ load input
+#ifdef __thumb2__
+ it hi
+#endif
+ addhi r8,r8,#1 @ 1<<128
+ ldrb r1,[lr,#-15]
+ ldrb r2,[lr,#-14]
+ ldrb r3,[lr,#-13]
+ orr r1,r0,r1,lsl#8
+ ldrb r0,[lr,#-12]
+ orr r2,r1,r2,lsl#16
+ ldrb r1,[lr,#-11]
+ orr r3,r2,r3,lsl#24
+ ldrb r2,[lr,#-10]
+ adds r4,r4,r3 @ accumulate input
+
+ ldrb r3,[lr,#-9]
+ orr r1,r0,r1,lsl#8
+ ldrb r0,[lr,#-8]
+ orr r2,r1,r2,lsl#16
+ ldrb r1,[lr,#-7]
+ orr r3,r2,r3,lsl#24
+ ldrb r2,[lr,#-6]
+ adcs r5,r5,r3
+
+ ldrb r3,[lr,#-5]
+ orr r1,r0,r1,lsl#8
+ ldrb r0,[lr,#-4]
+ orr r2,r1,r2,lsl#16
+ ldrb r1,[lr,#-3]
+ orr r3,r2,r3,lsl#24
+ ldrb r2,[lr,#-2]
+ adcs r6,r6,r3
+
+ ldrb r3,[lr,#-1]
+ orr r1,r0,r1,lsl#8
+ str lr,[sp,#8] @ offload input pointer
+ orr r2,r1,r2,lsl#16
+ add r10,r10,r10,lsr#2
+ orr r3,r2,r3,lsl#24
+#else
+ ldr r0,[lr],#16 @ load input
+#ifdef __thumb2__
+ it hi
+#endif
+ addhi r8,r8,#1 @ padbit
+ ldr r1,[lr,#-12]
+ ldr r2,[lr,#-8]
+ ldr r3,[lr,#-4]
+#ifdef __ARMEB__
+ rev r0,r0
+ rev r1,r1
+ rev r2,r2
+ rev r3,r3
+#endif
+ adds r4,r4,r0 @ accumulate input
+ str lr,[sp,#8] @ offload input pointer
+ adcs r5,r5,r1
+ add r10,r10,r10,lsr#2
+ adcs r6,r6,r2
+#endif
+ add r11,r11,r11,lsr#2
+ adcs r7,r7,r3
+ add r12,r12,r12,lsr#2
+
+ umull r2,r3,r5,r9
+ adc r8,r8,#0
+ umull r0,r1,r4,r9
+ umlal r2,r3,r8,r10
+ umlal r0,r1,r7,r10
+ ldr r10,[sp,#20] @ reload r10
+ umlal r2,r3,r6,r12
+ umlal r0,r1,r5,r12
+ umlal r2,r3,r7,r11
+ umlal r0,r1,r6,r11
+ umlal r2,r3,r4,r10
+ str r0,[sp,#0] @ future r4
+ mul r0,r11,r8
+ ldr r11,[sp,#24] @ reload r11
+ adds r2,r2,r1 @ d1+=d0>>32
+ eor r1,r1,r1
+ adc lr,r3,#0 @ future r6
+ str r2,[sp,#4] @ future r5
+
+ mul r2,r12,r8
+ eor r3,r3,r3
+ umlal r0,r1,r7,r12
+ ldr r12,[sp,#28] @ reload r12
+ umlal r2,r3,r7,r9
+ umlal r0,r1,r6,r9
+ umlal r2,r3,r6,r10
+ umlal r0,r1,r5,r10
+ umlal r2,r3,r5,r11
+ umlal r0,r1,r4,r11
+ umlal r2,r3,r4,r12
+ ldr r4,[sp,#0]
+ mul r8,r9,r8
+ ldr r5,[sp,#4]
+
+ adds r6,lr,r0 @ d2+=d1>>32
+ ldr lr,[sp,#8] @ reload input pointer
+ adc r1,r1,#0
+ adds r7,r2,r1 @ d3+=d2>>32
+ ldr r0,[sp,#16] @ reload end pointer
+ adc r3,r3,#0
+ add r8,r8,r3 @ h4+=d3>>32
+
+ and r1,r8,#-4
+ and r8,r8,#3
+ add r1,r1,r1,lsr#2 @ *=5
+ adds r4,r4,r1
+ adcs r5,r5,#0
+ adcs r6,r6,#0
+ adcs r7,r7,#0
+ adc r8,r8,#0
+
+ cmp r0,lr @ done yet?
+ bhi .Loop
+
+ ldr r0,[sp,#12]
+ add sp,sp,#32
+ stmia r0,{r4-r8} @ store the result
+
+.Lno_data:
+#if __LINUX_ARM_ARCH__ >= 5
+ ldmia sp!,{r3-r11,pc}
+#else
+ ldmia sp!,{r3-r11,lr}
+ tst lr,#1
+ moveq pc,lr @ be binary compatible with V4, yet
+ .word 0xe12fff1e @ interoperable with Thumb ISA:-)
+#endif
+ENDPROC(poly1305_blocks_arm)
+
+.align 5
+ENTRY(poly1305_emit_arm)
+ stmdb sp!,{r4-r11}
+.Lpoly1305_emit_enter:
+ ldmia r0,{r3-r7}
+ adds r8,r3,#5 @ compare to modulus
+ adcs r9,r4,#0
+ adcs r10,r5,#0
+ adcs r11,r6,#0
+ adc r7,r7,#0
+ tst r7,#4 @ did it carry/borrow?
+
+#ifdef __thumb2__
+ it ne
+#endif
+ movne r3,r8
+ ldr r8,[r2,#0]
+#ifdef __thumb2__
+ it ne
+#endif
+ movne r4,r9
+ ldr r9,[r2,#4]
+#ifdef __thumb2__
+ it ne
+#endif
+ movne r5,r10
+ ldr r10,[r2,#8]
+#ifdef __thumb2__
+ it ne
+#endif
+ movne r6,r11
+ ldr r11,[r2,#12]
+
+ adds r3,r3,r8
+ adcs r4,r4,r9
+ adcs r5,r5,r10
+ adc r6,r6,r11
+
+#if __LINUX_ARM_ARCH__ >= 7
+#ifdef __ARMEB__
+ rev r3,r3
+ rev r4,r4
+ rev r5,r5
+ rev r6,r6
+#endif
+ str r3,[r1,#0]
+ str r4,[r1,#4]
+ str r5,[r1,#8]
+ str r6,[r1,#12]
+#else
+ strb r3,[r1,#0]
+ mov r3,r3,lsr#8
+ strb r4,[r1,#4]
+ mov r4,r4,lsr#8
+ strb r5,[r1,#8]
+ mov r5,r5,lsr#8
+ strb r6,[r1,#12]
+ mov r6,r6,lsr#8
+
+ strb r3,[r1,#1]
+ mov r3,r3,lsr#8
+ strb r4,[r1,#5]
+ mov r4,r4,lsr#8
+ strb r5,[r1,#9]
+ mov r5,r5,lsr#8
+ strb r6,[r1,#13]
+ mov r6,r6,lsr#8
+
+ strb r3,[r1,#2]
+ mov r3,r3,lsr#8
+ strb r4,[r1,#6]
+ mov r4,r4,lsr#8
+ strb r5,[r1,#10]
+ mov r5,r5,lsr#8
+ strb r6,[r1,#14]
+ mov r6,r6,lsr#8
+
+ strb r3,[r1,#3]
+ strb r4,[r1,#7]
+ strb r5,[r1,#11]
+ strb r6,[r1,#15]
+#endif
+ ldmia sp!,{r4-r11}
+#if __LINUX_ARM_ARCH__ >= 5
+ bx lr @ bx lr
+#else
+ tst lr,#1
+ moveq pc,lr @ be binary compatible with V4, yet
+ .word 0xe12fff1e @ interoperable with Thumb ISA:-)
+#endif
+ENDPROC(poly1305_emit_arm)
+
+
+#if __LINUX_ARM_ARCH__ >= 7
+.fpu neon
+
+.align 5
+ENTRY(poly1305_init_neon)
+.Lpoly1305_init_neon:
+ ldr r4,[r0,#20] @ load key base 2^32
+ ldr r5,[r0,#24]
+ ldr r6,[r0,#28]
+ ldr r7,[r0,#32]
+
+ and r2,r4,#0x03ffffff @ base 2^32 -> base 2^26
+ mov r3,r4,lsr#26
+ mov r4,r5,lsr#20
+ orr r3,r3,r5,lsl#6
+ mov r5,r6,lsr#14
+ orr r4,r4,r6,lsl#12
+ mov r6,r7,lsr#8
+ orr r5,r5,r7,lsl#18
+ and r3,r3,#0x03ffffff
+ and r4,r4,#0x03ffffff
+ and r5,r5,#0x03ffffff
+
+ vdup.32 d0,r2 @ r^1 in both lanes
+ add r2,r3,r3,lsl#2 @ *5
+ vdup.32 d1,r3
+ add r3,r4,r4,lsl#2
+ vdup.32 d2,r2
+ vdup.32 d3,r4
+ add r4,r5,r5,lsl#2
+ vdup.32 d4,r3
+ vdup.32 d5,r5
+ add r5,r6,r6,lsl#2
+ vdup.32 d6,r4
+ vdup.32 d7,r6
+ vdup.32 d8,r5
+
+ mov r5,#2 @ counter
+
+.Lsquare_neon:
+ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+ @ d0 = h0*r0 + h4*5*r1 + h3*5*r2 + h2*5*r3 + h1*5*r4
+ @ d1 = h1*r0 + h0*r1 + h4*5*r2 + h3*5*r3 + h2*5*r4
+ @ d2 = h2*r0 + h1*r1 + h0*r2 + h4*5*r3 + h3*5*r4
+ @ d3 = h3*r0 + h2*r1 + h1*r2 + h0*r3 + h4*5*r4
+ @ d4 = h4*r0 + h3*r1 + h2*r2 + h1*r3 + h0*r4
+
+ vmull.u32 q5,d0,d0[1]
+ vmull.u32 q6,d1,d0[1]
+ vmull.u32 q7,d3,d0[1]
+ vmull.u32 q8,d5,d0[1]
+ vmull.u32 q9,d7,d0[1]
+
+ vmlal.u32 q5,d7,d2[1]
+ vmlal.u32 q6,d0,d1[1]
+ vmlal.u32 q7,d1,d1[1]
+ vmlal.u32 q8,d3,d1[1]
+ vmlal.u32 q9,d5,d1[1]
+
+ vmlal.u32 q5,d5,d4[1]
+ vmlal.u32 q6,d7,d4[1]
+ vmlal.u32 q8,d1,d3[1]
+ vmlal.u32 q7,d0,d3[1]
+ vmlal.u32 q9,d3,d3[1]
+
+ vmlal.u32 q5,d3,d6[1]
+ vmlal.u32 q8,d0,d5[1]
+ vmlal.u32 q6,d5,d6[1]
+ vmlal.u32 q7,d7,d6[1]
+ vmlal.u32 q9,d1,d5[1]
+
+ vmlal.u32 q8,d7,d8[1]
+ vmlal.u32 q5,d1,d8[1]
+ vmlal.u32 q6,d3,d8[1]
+ vmlal.u32 q7,d5,d8[1]
+ vmlal.u32 q9,d0,d7[1]
+
+ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+ @ lazy reduction as discussed in "NEON crypto" by D.J. Bernstein
+ @ and P. Schwabe
+ @
+ @ H0>>+H1>>+H2>>+H3>>+H4
+ @ H3>>+H4>>*5+H0>>+H1
+ @
+ @ Trivia.
+ @
+ @ Result of multiplication of n-bit number by m-bit number is
+ @ n+m bits wide. However! Even though 2^n is a n+1-bit number,
+ @ m-bit number multiplied by 2^n is still n+m bits wide.
+ @
+ @ Sum of two n-bit numbers is n+1 bits wide, sum of three - n+2,
+ @ and so is sum of four. Sum of 2^m n-m-bit numbers and n-bit
+ @ one is n+1 bits wide.
+ @
+ @ >>+ denotes Hnext += Hn>>26, Hn &= 0x3ffffff. This means that
+ @ H0, H2, H3 are guaranteed to be 26 bits wide, while H1 and H4
+ @ can be 27. However! In cases when their width exceeds 26 bits
+ @ they are limited by 2^26+2^6. This in turn means that *sum*
+ @ of the products with these values can still be viewed as sum
+ @ of 52-bit numbers as long as the amount of addends is not a
+ @ power of 2. For example,
+ @
+ @ H4 = H4*R0 + H3*R1 + H2*R2 + H1*R3 + H0 * R4,
+ @
+ @ which can't be larger than 5 * (2^26 + 2^6) * (2^26 + 2^6), or
+ @ 5 * (2^52 + 2*2^32 + 2^12), which in turn is smaller than
+ @ 8 * (2^52) or 2^55. However, the value is then multiplied by
+ @ by 5, so we should be looking at 5 * 5 * (2^52 + 2^33 + 2^12),
+ @ which is less than 32 * (2^52) or 2^57. And when processing
+ @ data we are looking at triple as many addends...
+ @
+ @ In key setup procedure pre-reduced H0 is limited by 5*4+1 and
+ @ 5*H4 - by 5*5 52-bit addends, or 57 bits. But when hashing the
+ @ input H0 is limited by (5*4+1)*3 addends, or 58 bits, while
+ @ 5*H4 by 5*5*3, or 59[!] bits. How is this relevant? vmlal.u32
+ @ instruction accepts 2x32-bit input and writes 2x64-bit result.
+ @ This means that result of reduction have to be compressed upon
+ @ loop wrap-around. This can be done in the process of reduction
+ @ to minimize amount of instructions [as well as amount of
+ @ 128-bit instructions, which benefits low-end processors], but
+ @ one has to watch for H2 (which is narrower than H0) and 5*H4
+ @ not being wider than 58 bits, so that result of right shift
+ @ by 26 bits fits in 32 bits. This is also useful on x86,
+ @ because it allows to use paddd in place for paddq, which
+ @ benefits Atom, where paddq is ridiculously slow.
+
+ vshr.u64 q15,q8,#26
+ vmovn.i64 d16,q8
+ vshr.u64 q4,q5,#26
+ vmovn.i64 d10,q5
+ vadd.i64 q9,q9,q15 @ h3 -> h4
+ vbic.i32 d16,#0xfc000000 @ &=0x03ffffff
+ vadd.i64 q6,q6,q4 @ h0 -> h1
+ vbic.i32 d10,#0xfc000000
+
+ vshrn.u64 d30,q9,#26
+ vmovn.i64 d18,q9
+ vshr.u64 q4,q6,#26
+ vmovn.i64 d12,q6
+ vadd.i64 q7,q7,q4 @ h1 -> h2
+ vbic.i32 d18,#0xfc000000
+ vbic.i32 d12,#0xfc000000
+
+ vadd.i32 d10,d10,d30
+ vshl.u32 d30,d30,#2
+ vshrn.u64 d8,q7,#26
+ vmovn.i64 d14,q7
+ vadd.i32 d10,d10,d30 @ h4 -> h0
+ vadd.i32 d16,d16,d8 @ h2 -> h3
+ vbic.i32 d14,#0xfc000000
+
+ vshr.u32 d30,d10,#26
+ vbic.i32 d10,#0xfc000000
+ vshr.u32 d8,d16,#26
+ vbic.i32 d16,#0xfc000000
+ vadd.i32 d12,d12,d30 @ h0 -> h1
+ vadd.i32 d18,d18,d8 @ h3 -> h4
+
+ subs r5,r5,#1
+ beq .Lsquare_break_neon
+
+ add r6,r0,#(48+0*9*4)
+ add r7,r0,#(48+1*9*4)
+
+ vtrn.32 d0,d10 @ r^2:r^1
+ vtrn.32 d3,d14
+ vtrn.32 d5,d16
+ vtrn.32 d1,d12
+ vtrn.32 d7,d18
+
+ vshl.u32 d4,d3,#2 @ *5
+ vshl.u32 d6,d5,#2
+ vshl.u32 d2,d1,#2
+ vshl.u32 d8,d7,#2
+ vadd.i32 d4,d4,d3
+ vadd.i32 d2,d2,d1
+ vadd.i32 d6,d6,d5
+ vadd.i32 d8,d8,d7
+
+ vst4.32 {d0[0],d1[0],d2[0],d3[0]},[r6]!
+ vst4.32 {d0[1],d1[1],d2[1],d3[1]},[r7]!
+ vst4.32 {d4[0],d5[0],d6[0],d7[0]},[r6]!
+ vst4.32 {d4[1],d5[1],d6[1],d7[1]},[r7]!
+ vst1.32 {d8[0]},[r6,:32]
+ vst1.32 {d8[1]},[r7,:32]
+
+ b .Lsquare_neon
+
+.align 4
+.Lsquare_break_neon:
+ add r6,r0,#(48+2*4*9)
+ add r7,r0,#(48+3*4*9)
+
+ vmov d0,d10 @ r^4:r^3
+ vshl.u32 d2,d12,#2 @ *5
+ vmov d1,d12
+ vshl.u32 d4,d14,#2
+ vmov d3,d14
+ vshl.u32 d6,d16,#2
+ vmov d5,d16
+ vshl.u32 d8,d18,#2
+ vmov d7,d18
+ vadd.i32 d2,d2,d12
+ vadd.i32 d4,d4,d14
+ vadd.i32 d6,d6,d16
+ vadd.i32 d8,d8,d18
+
+ vst4.32 {d0[0],d1[0],d2[0],d3[0]},[r6]!
+ vst4.32 {d0[1],d1[1],d2[1],d3[1]},[r7]!
+ vst4.32 {d4[0],d5[0],d6[0],d7[0]},[r6]!
+ vst4.32 {d4[1],d5[1],d6[1],d7[1]},[r7]!
+ vst1.32 {d8[0]},[r6]
+ vst1.32 {d8[1]},[r7]
+
+ bx lr @ bx lr
+ENDPROC(poly1305_init_neon)
+
+.align 5
+ENTRY(poly1305_blocks_neon)
+ ldr ip,[r0,#36] @ is_base2_26
+ ands r2,r2,#-16
+ beq .Lno_data_neon
+
+ cmp r2,#64
+ bhs .Lenter_neon
+ tst ip,ip @ is_base2_26?
+ beq .Lpoly1305_blocks_arm
+
+.Lenter_neon:
+ stmdb sp!,{r4-r7}
+ vstmdb sp!,{d8-d15} @ ABI specification says so
+
+ tst ip,ip @ is_base2_26?
+ bne .Lbase2_26_neon
+
+ stmdb sp!,{r1-r3,lr}
+ bl .Lpoly1305_init_neon
+
+ ldr r4,[r0,#0] @ load hash value base 2^32
+ ldr r5,[r0,#4]
+ ldr r6,[r0,#8]
+ ldr r7,[r0,#12]
+ ldr ip,[r0,#16]
+
+ and r2,r4,#0x03ffffff @ base 2^32 -> base 2^26
+ mov r3,r4,lsr#26
+ veor d10,d10,d10
+ mov r4,r5,lsr#20
+ orr r3,r3,r5,lsl#6
+ veor d12,d12,d12
+ mov r5,r6,lsr#14
+ orr r4,r4,r6,lsl#12
+ veor d14,d14,d14
+ mov r6,r7,lsr#8
+ orr r5,r5,r7,lsl#18
+ veor d16,d16,d16
+ and r3,r3,#0x03ffffff
+ orr r6,r6,ip,lsl#24
+ veor d18,d18,d18
+ and r4,r4,#0x03ffffff
+ mov r1,#1
+ and r5,r5,#0x03ffffff
+ str r1,[r0,#36] @ is_base2_26
+
+ vmov.32 d10[0],r2
+ vmov.32 d12[0],r3
+ vmov.32 d14[0],r4
+ vmov.32 d16[0],r5
+ vmov.32 d18[0],r6
+ adr r5,.Lzeros
+
+ ldmia sp!,{r1-r3,lr}
+ b .Lbase2_32_neon
+
+.align 4
+.Lbase2_26_neon:
+ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+ @ load hash value
+
+ veor d10,d10,d10
+ veor d12,d12,d12
+ veor d14,d14,d14
+ veor d16,d16,d16
+ veor d18,d18,d18
+ vld4.32 {d10[0],d12[0],d14[0],d16[0]},[r0]!
+ adr r5,.Lzeros
+ vld1.32 {d18[0]},[r0]
+ sub r0,r0,#16 @ rewind
+
+.Lbase2_32_neon:
+ add r4,r1,#32
+ mov r3,r3,lsl#24
+ tst r2,#31
+ beq .Leven
+
+ vld4.32 {d20[0],d22[0],d24[0],d26[0]},[r1]!
+ vmov.32 d28[0],r3
+ sub r2,r2,#16
+ add r4,r1,#32
+
+#ifdef __ARMEB__
+ vrev32.8 q10,q10
+ vrev32.8 q13,q13
+ vrev32.8 q11,q11
+ vrev32.8 q12,q12
+#endif
+ vsri.u32 d28,d26,#8 @ base 2^32 -> base 2^26
+ vshl.u32 d26,d26,#18
+
+ vsri.u32 d26,d24,#14
+ vshl.u32 d24,d24,#12
+ vadd.i32 d29,d28,d18 @ add hash value and move to #hi
+
+ vbic.i32 d26,#0xfc000000
+ vsri.u32 d24,d22,#20
+ vshl.u32 d22,d22,#6
+
+ vbic.i32 d24,#0xfc000000
+ vsri.u32 d22,d20,#26
+ vadd.i32 d27,d26,d16
+
+ vbic.i32 d20,#0xfc000000
+ vbic.i32 d22,#0xfc000000
+ vadd.i32 d25,d24,d14
+
+ vadd.i32 d21,d20,d10
+ vadd.i32 d23,d22,d12
+
+ mov r7,r5
+ add r6,r0,#48
+
+ cmp r2,r2
+ b .Long_tail
+
+.align 4
+.Leven:
+ subs r2,r2,#64
+ it lo
+ movlo r4,r5
+
+ vmov.i32 q14,#1<<24 @ padbit, yes, always
+ vld4.32 {d20,d22,d24,d26},[r1] @ inp[0:1]
+ add r1,r1,#64
+ vld4.32 {d21,d23,d25,d27},[r4] @ inp[2:3] (or 0)
+ add r4,r4,#64
+ itt hi
+ addhi r7,r0,#(48+1*9*4)
+ addhi r6,r0,#(48+3*9*4)
+
+#ifdef __ARMEB__
+ vrev32.8 q10,q10
+ vrev32.8 q13,q13
+ vrev32.8 q11,q11
+ vrev32.8 q12,q12
+#endif
+ vsri.u32 q14,q13,#8 @ base 2^32 -> base 2^26
+ vshl.u32 q13,q13,#18
+
+ vsri.u32 q13,q12,#14
+ vshl.u32 q12,q12,#12
+
+ vbic.i32 q13,#0xfc000000
+ vsri.u32 q12,q11,#20
+ vshl.u32 q11,q11,#6
+
+ vbic.i32 q12,#0xfc000000
+ vsri.u32 q11,q10,#26
+
+ vbic.i32 q10,#0xfc000000
+ vbic.i32 q11,#0xfc000000
+
+ bls .Lskip_loop
+
+ vld4.32 {d0[1],d1[1],d2[1],d3[1]},[r7]! @ load r^2
+ vld4.32 {d0[0],d1[0],d2[0],d3[0]},[r6]! @ load r^4
+ vld4.32 {d4[1],d5[1],d6[1],d7[1]},[r7]!
+ vld4.32 {d4[0],d5[0],d6[0],d7[0]},[r6]!
+ b .Loop_neon
+
+.align 5
+.Loop_neon:
+ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+ @ ((inp[0]*r^4+inp[2]*r^2+inp[4])*r^4+inp[6]*r^2
+ @ ((inp[1]*r^4+inp[3]*r^2+inp[5])*r^3+inp[7]*r
+ @ ___________________/
+ @ ((inp[0]*r^4+inp[2]*r^2+inp[4])*r^4+inp[6]*r^2+inp[8])*r^2
+ @ ((inp[1]*r^4+inp[3]*r^2+inp[5])*r^4+inp[7]*r^2+inp[9])*r
+ @ ___________________/ ____________________/
+ @
+ @ Note that we start with inp[2:3]*r^2. This is because it
+ @ doesn't depend on reduction in previous iteration.
+ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+ @ d4 = h4*r0 + h3*r1 + h2*r2 + h1*r3 + h0*r4
+ @ d3 = h3*r0 + h2*r1 + h1*r2 + h0*r3 + h4*5*r4
+ @ d2 = h2*r0 + h1*r1 + h0*r2 + h4*5*r3 + h3*5*r4
+ @ d1 = h1*r0 + h0*r1 + h4*5*r2 + h3*5*r3 + h2*5*r4
+ @ d0 = h0*r0 + h4*5*r1 + h3*5*r2 + h2*5*r3 + h1*5*r4
+
+ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+ @ inp[2:3]*r^2
+
+ vadd.i32 d24,d24,d14 @ accumulate inp[0:1]
+ vmull.u32 q7,d25,d0[1]
+ vadd.i32 d20,d20,d10
+ vmull.u32 q5,d21,d0[1]
+ vadd.i32 d26,d26,d16
+ vmull.u32 q8,d27,d0[1]
+ vmlal.u32 q7,d23,d1[1]
+ vadd.i32 d22,d22,d12
+ vmull.u32 q6,d23,d0[1]
+
+ vadd.i32 d28,d28,d18
+ vmull.u32 q9,d29,d0[1]
+ subs r2,r2,#64
+ vmlal.u32 q5,d29,d2[1]
+ it lo
+ movlo r4,r5
+ vmlal.u32 q8,d25,d1[1]
+ vld1.32 d8[1],[r7,:32]
+ vmlal.u32 q6,d21,d1[1]
+ vmlal.u32 q9,d27,d1[1]
+
+ vmlal.u32 q5,d27,d4[1]
+ vmlal.u32 q8,d23,d3[1]
+ vmlal.u32 q9,d25,d3[1]
+ vmlal.u32 q6,d29,d4[1]
+ vmlal.u32 q7,d21,d3[1]
+
+ vmlal.u32 q8,d21,d5[1]
+ vmlal.u32 q5,d25,d6[1]
+ vmlal.u32 q9,d23,d5[1]
+ vmlal.u32 q6,d27,d6[1]
+ vmlal.u32 q7,d29,d6[1]
+
+ vmlal.u32 q8,d29,d8[1]
+ vmlal.u32 q5,d23,d8[1]
+ vmlal.u32 q9,d21,d7[1]
+ vmlal.u32 q6,d25,d8[1]
+ vmlal.u32 q7,d27,d8[1]
+
+ vld4.32 {d21,d23,d25,d27},[r4] @ inp[2:3] (or 0)
+ add r4,r4,#64
+
+ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+ @ (hash+inp[0:1])*r^4 and accumulate
+
+ vmlal.u32 q8,d26,d0[0]
+ vmlal.u32 q5,d20,d0[0]
+ vmlal.u32 q9,d28,d0[0]
+ vmlal.u32 q6,d22,d0[0]
+ vmlal.u32 q7,d24,d0[0]
+ vld1.32 d8[0],[r6,:32]
+
+ vmlal.u32 q8,d24,d1[0]
+ vmlal.u32 q5,d28,d2[0]
+ vmlal.u32 q9,d26,d1[0]
+ vmlal.u32 q6,d20,d1[0]
+ vmlal.u32 q7,d22,d1[0]
+
+ vmlal.u32 q8,d22,d3[0]
+ vmlal.u32 q5,d26,d4[0]
+ vmlal.u32 q9,d24,d3[0]
+ vmlal.u32 q6,d28,d4[0]
+ vmlal.u32 q7,d20,d3[0]
+
+ vmlal.u32 q8,d20,d5[0]
+ vmlal.u32 q5,d24,d6[0]
+ vmlal.u32 q9,d22,d5[0]
+ vmlal.u32 q6,d26,d6[0]
+ vmlal.u32 q8,d28,d8[0]
+
+ vmlal.u32 q7,d28,d6[0]
+ vmlal.u32 q5,d22,d8[0]
+ vmlal.u32 q9,d20,d7[0]
+ vmov.i32 q14,#1<<24 @ padbit, yes, always
+ vmlal.u32 q6,d24,d8[0]
+ vmlal.u32 q7,d26,d8[0]
+
+ vld4.32 {d20,d22,d24,d26},[r1] @ inp[0:1]
+ add r1,r1,#64
+#ifdef __ARMEB__
+ vrev32.8 q10,q10
+ vrev32.8 q11,q11
+ vrev32.8 q12,q12
+ vrev32.8 q13,q13
+#endif
+
+ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+ @ lazy reduction interleaved with base 2^32 -> base 2^26 of
+ @ inp[0:3] previously loaded to q10-q13 and smashed to q10-q14.
+
+ vshr.u64 q15,q8,#26
+ vmovn.i64 d16,q8
+ vshr.u64 q4,q5,#26
+ vmovn.i64 d10,q5
+ vadd.i64 q9,q9,q15 @ h3 -> h4
+ vbic.i32 d16,#0xfc000000
+ vsri.u32 q14,q13,#8 @ base 2^32 -> base 2^26
+ vadd.i64 q6,q6,q4 @ h0 -> h1
+ vshl.u32 q13,q13,#18
+ vbic.i32 d10,#0xfc000000
+
+ vshrn.u64 d30,q9,#26
+ vmovn.i64 d18,q9
+ vshr.u64 q4,q6,#26
+ vmovn.i64 d12,q6
+ vadd.i64 q7,q7,q4 @ h1 -> h2
+ vsri.u32 q13,q12,#14
+ vbic.i32 d18,#0xfc000000
+ vshl.u32 q12,q12,#12
+ vbic.i32 d12,#0xfc000000
+
+ vadd.i32 d10,d10,d30
+ vshl.u32 d30,d30,#2
+ vbic.i32 q13,#0xfc000000
+ vshrn.u64 d8,q7,#26
+ vmovn.i64 d14,q7
+ vaddl.u32 q5,d10,d30 @ h4 -> h0 [widen for a sec]
+ vsri.u32 q12,q11,#20
+ vadd.i32 d16,d16,d8 @ h2 -> h3
+ vshl.u32 q11,q11,#6
+ vbic.i32 d14,#0xfc000000
+ vbic.i32 q12,#0xfc000000
+
+ vshrn.u64 d30,q5,#26 @ re-narrow
+ vmovn.i64 d10,q5
+ vsri.u32 q11,q10,#26
+ vbic.i32 q10,#0xfc000000
+ vshr.u32 d8,d16,#26
+ vbic.i32 d16,#0xfc000000
+ vbic.i32 d10,#0xfc000000
+ vadd.i32 d12,d12,d30 @ h0 -> h1
+ vadd.i32 d18,d18,d8 @ h3 -> h4
+ vbic.i32 q11,#0xfc000000
+
+ bhi .Loop_neon
+
+.Lskip_loop:
+ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+ @ multiply (inp[0:1]+hash) or inp[2:3] by r^2:r^1
+
+ add r7,r0,#(48+0*9*4)
+ add r6,r0,#(48+1*9*4)
+ adds r2,r2,#32
+ it ne
+ movne r2,#0
+ bne .Long_tail
+
+ vadd.i32 d25,d24,d14 @ add hash value and move to #hi
+ vadd.i32 d21,d20,d10
+ vadd.i32 d27,d26,d16
+ vadd.i32 d23,d22,d12
+ vadd.i32 d29,d28,d18
+
+.Long_tail:
+ vld4.32 {d0[1],d1[1],d2[1],d3[1]},[r7]! @ load r^1
+ vld4.32 {d0[0],d1[0],d2[0],d3[0]},[r6]! @ load r^2
+
+ vadd.i32 d24,d24,d14 @ can be redundant
+ vmull.u32 q7,d25,d0
+ vadd.i32 d20,d20,d10
+ vmull.u32 q5,d21,d0
+ vadd.i32 d26,d26,d16
+ vmull.u32 q8,d27,d0
+ vadd.i32 d22,d22,d12
+ vmull.u32 q6,d23,d0
+ vadd.i32 d28,d28,d18
+ vmull.u32 q9,d29,d0
+
+ vmlal.u32 q5,d29,d2
+ vld4.32 {d4[1],d5[1],d6[1],d7[1]},[r7]!
+ vmlal.u32 q8,d25,d1
+ vld4.32 {d4[0],d5[0],d6[0],d7[0]},[r6]!
+ vmlal.u32 q6,d21,d1
+ vmlal.u32 q9,d27,d1
+ vmlal.u32 q7,d23,d1
+
+ vmlal.u32 q8,d23,d3
+ vld1.32 d8[1],[r7,:32]
+ vmlal.u32 q5,d27,d4
+ vld1.32 d8[0],[r6,:32]
+ vmlal.u32 q9,d25,d3
+ vmlal.u32 q6,d29,d4
+ vmlal.u32 q7,d21,d3
+
+ vmlal.u32 q8,d21,d5
+ it ne
+ addne r7,r0,#(48+2*9*4)
+ vmlal.u32 q5,d25,d6
+ it ne
+ addne r6,r0,#(48+3*9*4)
+ vmlal.u32 q9,d23,d5
+ vmlal.u32 q6,d27,d6
+ vmlal.u32 q7,d29,d6
+
+ vmlal.u32 q8,d29,d8
+ vorn q0,q0,q0 @ all-ones, can be redundant
+ vmlal.u32 q5,d23,d8
+ vshr.u64 q0,q0,#38
+ vmlal.u32 q9,d21,d7
+ vmlal.u32 q6,d25,d8
+ vmlal.u32 q7,d27,d8
+
+ beq .Lshort_tail
+
+ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+ @ (hash+inp[0:1])*r^4:r^3 and accumulate
+
+ vld4.32 {d0[1],d1[1],d2[1],d3[1]},[r7]! @ load r^3
+ vld4.32 {d0[0],d1[0],d2[0],d3[0]},[r6]! @ load r^4
+
+ vmlal.u32 q7,d24,d0
+ vmlal.u32 q5,d20,d0
+ vmlal.u32 q8,d26,d0
+ vmlal.u32 q6,d22,d0
+ vmlal.u32 q9,d28,d0
+
+ vmlal.u32 q5,d28,d2
+ vld4.32 {d4[1],d5[1],d6[1],d7[1]},[r7]!
+ vmlal.u32 q8,d24,d1
+ vld4.32 {d4[0],d5[0],d6[0],d7[0]},[r6]!
+ vmlal.u32 q6,d20,d1
+ vmlal.u32 q9,d26,d1
+ vmlal.u32 q7,d22,d1
+
+ vmlal.u32 q8,d22,d3
+ vld1.32 d8[1],[r7,:32]
+ vmlal.u32 q5,d26,d4
+ vld1.32 d8[0],[r6,:32]
+ vmlal.u32 q9,d24,d3
+ vmlal.u32 q6,d28,d4
+ vmlal.u32 q7,d20,d3
+
+ vmlal.u32 q8,d20,d5
+ vmlal.u32 q5,d24,d6
+ vmlal.u32 q9,d22,d5
+ vmlal.u32 q6,d26,d6
+ vmlal.u32 q7,d28,d6
+
+ vmlal.u32 q8,d28,d8
+ vorn q0,q0,q0 @ all-ones
+ vmlal.u32 q5,d22,d8
+ vshr.u64 q0,q0,#38
+ vmlal.u32 q9,d20,d7
+ vmlal.u32 q6,d24,d8
+ vmlal.u32 q7,d26,d8
+
+.Lshort_tail:
+ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+ @ horizontal addition
+
+ vadd.i64 d16,d16,d17
+ vadd.i64 d10,d10,d11
+ vadd.i64 d18,d18,d19
+ vadd.i64 d12,d12,d13
+ vadd.i64 d14,d14,d15
+
+ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+ @ lazy reduction, but without narrowing
+
+ vshr.u64 q15,q8,#26
+ vand.i64 q8,q8,q0
+ vshr.u64 q4,q5,#26
+ vand.i64 q5,q5,q0
+ vadd.i64 q9,q9,q15 @ h3 -> h4
+ vadd.i64 q6,q6,q4 @ h0 -> h1
+
+ vshr.u64 q15,q9,#26
+ vand.i64 q9,q9,q0
+ vshr.u64 q4,q6,#26
+ vand.i64 q6,q6,q0
+ vadd.i64 q7,q7,q4 @ h1 -> h2
+
+ vadd.i64 q5,q5,q15
+ vshl.u64 q15,q15,#2
+ vshr.u64 q4,q7,#26
+ vand.i64 q7,q7,q0
+ vadd.i64 q5,q5,q15 @ h4 -> h0
+ vadd.i64 q8,q8,q4 @ h2 -> h3
+
+ vshr.u64 q15,q5,#26
+ vand.i64 q5,q5,q0
+ vshr.u64 q4,q8,#26
+ vand.i64 q8,q8,q0
+ vadd.i64 q6,q6,q15 @ h0 -> h1
+ vadd.i64 q9,q9,q4 @ h3 -> h4
+
+ cmp r2,#0
+ bne .Leven
+
+ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
+ @ store hash value
+
+ vst4.32 {d10[0],d12[0],d14[0],d16[0]},[r0]!
+ vst1.32 {d18[0]},[r0]
+
+ vldmia sp!,{d8-d15} @ epilogue
+ ldmia sp!,{r4-r7}
+.Lno_data_neon:
+ bx lr @ bx lr
+ENDPROC(poly1305_blocks_neon)
+
+.align 5
+ENTRY(poly1305_emit_neon)
+ ldr ip,[r0,#36] @ is_base2_26
+
+ stmdb sp!,{r4-r11}
+
+ tst ip,ip
+ beq .Lpoly1305_emit_enter
+
+ ldmia r0,{r3-r7}
+ eor r8,r8,r8
+
+ adds r3,r3,r4,lsl#26 @ base 2^26 -> base 2^32
+ mov r4,r4,lsr#6
+ adcs r4,r4,r5,lsl#20
+ mov r5,r5,lsr#12
+ adcs r5,r5,r6,lsl#14
+ mov r6,r6,lsr#18
+ adcs r6,r6,r7,lsl#8
+ adc r7,r8,r7,lsr#24 @ can be partially reduced ...
+
+ and r8,r7,#-4 @ ... so reduce
+ and r7,r6,#3
+ add r8,r8,r8,lsr#2 @ *= 5
+ adds r3,r3,r8
+ adcs r4,r4,#0
+ adcs r5,r5,#0
+ adcs r6,r6,#0
+ adc r7,r7,#0
+
+ adds r8,r3,#5 @ compare to modulus
+ adcs r9,r4,#0
+ adcs r10,r5,#0
+ adcs r11,r6,#0
+ adc r7,r7,#0
+ tst r7,#4 @ did it carry/borrow?
+
+ it ne
+ movne r3,r8
+ ldr r8,[r2,#0]
+ it ne
+ movne r4,r9
+ ldr r9,[r2,#4]
+ it ne
+ movne r5,r10
+ ldr r10,[r2,#8]
+ it ne
+ movne r6,r11
+ ldr r11,[r2,#12]
+
+ adds r3,r3,r8 @ accumulate nonce
+ adcs r4,r4,r9
+ adcs r5,r5,r10
+ adc r6,r6,r11
+
+#ifdef __ARMEB__
+ rev r3,r3
+ rev r4,r4
+ rev r5,r5
+ rev r6,r6
+#endif
+ str r3,[r1,#0] @ store the result
+ str r4,[r1,#4]
+ str r5,[r1,#8]
+ str r6,[r1,#12]
+
+ ldmia sp!,{r4-r11}
+ bx lr @ bx lr
+ENDPROC(poly1305_emit_neon)
+
+.align 5
+.Lzeros:
+.long 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0
+#endif
diff --git a/lib/zinc/poly1305/poly1305-arm64.S b/lib/zinc/poly1305/poly1305-arm64.S
new file mode 100644
index 000000000000..911b57edeb20
--- /dev/null
+++ b/lib/zinc/poly1305/poly1305-arm64.S
@@ -0,0 +1,820 @@
+/* SPDX-License-Identifier: OpenSSL OR (BSD-3-Clause OR GPL-2.0)
+ *
+ * Copyright (C) 2015-2018 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ * Copyright 2016 The OpenSSL Project Authors. All Rights Reserved.
+ */
+
+#include <linux/linkage.h>
+.text
+
+.align 5
+ENTRY(poly1305_init_arm)
+ cmp x1,xzr
+ stp xzr,xzr,[x0] // zero hash value
+ stp xzr,xzr,[x0,#16] // [along with is_base2_26]
+
+ csel x0,xzr,x0,eq
+ b.eq .Lno_key
+
+ ldp x7,x8,[x1] // load key
+ mov x9,#0xfffffffc0fffffff
+ movk x9,#0x0fff,lsl#48
+#ifdef __ARMEB__
+ rev x7,x7 // flip bytes
+ rev x8,x8
+#endif
+ and x7,x7,x9 // &=0ffffffc0fffffff
+ and x9,x9,#-4
+ and x8,x8,x9 // &=0ffffffc0ffffffc
+ stp x7,x8,[x0,#32] // save key value
+
+.Lno_key:
+ ret
+ENDPROC(poly1305_init_arm)
+
+.align 5
+ENTRY(poly1305_blocks_arm)
+ ands x2,x2,#-16
+ b.eq .Lno_data
+
+ ldp x4,x5,[x0] // load hash value
+ ldp x7,x8,[x0,#32] // load key value
+ ldr x6,[x0,#16]
+ add x9,x8,x8,lsr#2 // s1 = r1 + (r1 >> 2)
+ b .Loop
+
+.align 5
+.Loop:
+ ldp x10,x11,[x1],#16 // load input
+ sub x2,x2,#16
+#ifdef __ARMEB__
+ rev x10,x10
+ rev x11,x11
+#endif
+ adds x4,x4,x10 // accumulate input
+ adcs x5,x5,x11
+
+ mul x12,x4,x7 // h0*r0
+ adc x6,x6,x3
+ umulh x13,x4,x7
+
+ mul x10,x5,x9 // h1*5*r1
+ umulh x11,x5,x9
+
+ adds x12,x12,x10
+ mul x10,x4,x8 // h0*r1
+ adc x13,x13,x11
+ umulh x14,x4,x8
+
+ adds x13,x13,x10
+ mul x10,x5,x7 // h1*r0
+ adc x14,x14,xzr
+ umulh x11,x5,x7
+
+ adds x13,x13,x10
+ mul x10,x6,x9 // h2*5*r1
+ adc x14,x14,x11
+ mul x11,x6,x7 // h2*r0
+
+ adds x13,x13,x10
+ adc x14,x14,x11
+
+ and x10,x14,#-4 // final reduction
+ and x6,x14,#3
+ add x10,x10,x14,lsr#2
+ adds x4,x12,x10
+ adcs x5,x13,xzr
+ adc x6,x6,xzr
+
+ cbnz x2,.Loop
+
+ stp x4,x5,[x0] // store hash value
+ str x6,[x0,#16]
+
+.Lno_data:
+ ret
+ENDPROC(poly1305_blocks_arm)
+
+.align 5
+ENTRY(poly1305_emit_arm)
+ ldp x4,x5,[x0] // load hash base 2^64
+ ldr x6,[x0,#16]
+ ldp x10,x11,[x2] // load nonce
+
+ adds x12,x4,#5 // compare to modulus
+ adcs x13,x5,xzr
+ adc x14,x6,xzr
+
+ tst x14,#-4 // see if it's carried/borrowed
+
+ csel x4,x4,x12,eq
+ csel x5,x5,x13,eq
+
+#ifdef __ARMEB__
+ ror x10,x10,#32 // flip nonce words
+ ror x11,x11,#32
+#endif
+ adds x4,x4,x10 // accumulate nonce
+ adc x5,x5,x11
+#ifdef __ARMEB__
+ rev x4,x4 // flip output bytes
+ rev x5,x5
+#endif
+ stp x4,x5,[x1] // write result
+
+ ret
+ENDPROC(poly1305_emit_arm)
+
+.align 5
+__poly1305_mult:
+ mul x12,x4,x7 // h0*r0
+ umulh x13,x4,x7
+
+ mul x10,x5,x9 // h1*5*r1
+ umulh x11,x5,x9
+
+ adds x12,x12,x10
+ mul x10,x4,x8 // h0*r1
+ adc x13,x13,x11
+ umulh x14,x4,x8
+
+ adds x13,x13,x10
+ mul x10,x5,x7 // h1*r0
+ adc x14,x14,xzr
+ umulh x11,x5,x7
+
+ adds x13,x13,x10
+ mul x10,x6,x9 // h2*5*r1
+ adc x14,x14,x11
+ mul x11,x6,x7 // h2*r0
+
+ adds x13,x13,x10
+ adc x14,x14,x11
+
+ and x10,x14,#-4 // final reduction
+ and x6,x14,#3
+ add x10,x10,x14,lsr#2
+ adds x4,x12,x10
+ adcs x5,x13,xzr
+ adc x6,x6,xzr
+
+ ret
+
+__poly1305_splat:
+ and x12,x4,#0x03ffffff // base 2^64 -> base 2^26
+ ubfx x13,x4,#26,#26
+ extr x14,x5,x4,#52
+ and x14,x14,#0x03ffffff
+ ubfx x15,x5,#14,#26
+ extr x16,x6,x5,#40
+
+ str w12,[x0,#16*0] // r0
+ add w12,w13,w13,lsl#2 // r1*5
+ str w13,[x0,#16*1] // r1
+ add w13,w14,w14,lsl#2 // r2*5
+ str w12,[x0,#16*2] // s1
+ str w14,[x0,#16*3] // r2
+ add w14,w15,w15,lsl#2 // r3*5
+ str w13,[x0,#16*4] // s2
+ str w15,[x0,#16*5] // r3
+ add w15,w16,w16,lsl#2 // r4*5
+ str w14,[x0,#16*6] // s3
+ str w16,[x0,#16*7] // r4
+ str w15,[x0,#16*8] // s4
+
+ ret
+
+.align 5
+ENTRY(poly1305_blocks_neon)
+ ldr x17,[x0,#24]
+ cmp x2,#128
+ b.hs .Lblocks_neon
+ cbz x17,poly1305_blocks_arm
+
+.Lblocks_neon:
+ stp x29,x30,[sp,#-80]!
+ add x29,sp,#0
+
+ ands x2,x2,#-16
+ b.eq .Lno_data_neon
+
+ cbz x17,.Lbase2_64_neon
+
+ ldp w10,w11,[x0] // load hash value base 2^26
+ ldp w12,w13,[x0,#8]
+ ldr w14,[x0,#16]
+
+ tst x2,#31
+ b.eq .Leven_neon
+
+ ldp x7,x8,[x0,#32] // load key value
+
+ add x4,x10,x11,lsl#26 // base 2^26 -> base 2^64
+ lsr x5,x12,#12
+ adds x4,x4,x12,lsl#52
+ add x5,x5,x13,lsl#14
+ adc x5,x5,xzr
+ lsr x6,x14,#24
+ adds x5,x5,x14,lsl#40
+ adc x14,x6,xzr // can be partially reduced...
+
+ ldp x12,x13,[x1],#16 // load input
+ sub x2,x2,#16
+ add x9,x8,x8,lsr#2 // s1 = r1 + (r1 >> 2)
+
+ and x10,x14,#-4 // ... so reduce
+ and x6,x14,#3
+ add x10,x10,x14,lsr#2
+ adds x4,x4,x10
+ adcs x5,x5,xzr
+ adc x6,x6,xzr
+
+#ifdef __ARMEB__
+ rev x12,x12
+ rev x13,x13
+#endif
+ adds x4,x4,x12 // accumulate input
+ adcs x5,x5,x13
+ adc x6,x6,x3
+
+ bl __poly1305_mult
+ ldr x30,[sp,#8]
+
+ cbz x3,.Lstore_base2_64_neon
+
+ and x10,x4,#0x03ffffff // base 2^64 -> base 2^26
+ ubfx x11,x4,#26,#26
+ extr x12,x5,x4,#52
+ and x12,x12,#0x03ffffff
+ ubfx x13,x5,#14,#26
+ extr x14,x6,x5,#40
+
+ cbnz x2,.Leven_neon
+
+ stp w10,w11,[x0] // store hash value base 2^26
+ stp w12,w13,[x0,#8]
+ str w14,[x0,#16]
+ b .Lno_data_neon
+
+.align 4
+.Lstore_base2_64_neon:
+ stp x4,x5,[x0] // store hash value base 2^64
+ stp x6,xzr,[x0,#16] // note that is_base2_26 is zeroed
+ b .Lno_data_neon
+
+.align 4
+.Lbase2_64_neon:
+ ldp x7,x8,[x0,#32] // load key value
+
+ ldp x4,x5,[x0] // load hash value base 2^64
+ ldr x6,[x0,#16]
+
+ tst x2,#31
+ b.eq .Linit_neon
+
+ ldp x12,x13,[x1],#16 // load input
+ sub x2,x2,#16
+ add x9,x8,x8,lsr#2 // s1 = r1 + (r1 >> 2)
+#ifdef __ARMEB__
+ rev x12,x12
+ rev x13,x13
+#endif
+ adds x4,x4,x12 // accumulate input
+ adcs x5,x5,x13
+ adc x6,x6,x3
+
+ bl __poly1305_mult
+
+.Linit_neon:
+ and x10,x4,#0x03ffffff // base 2^64 -> base 2^26
+ ubfx x11,x4,#26,#26
+ extr x12,x5,x4,#52
+ and x12,x12,#0x03ffffff
+ ubfx x13,x5,#14,#26
+ extr x14,x6,x5,#40
+
+ stp d8,d9,[sp,#16] // meet ABI requirements
+ stp d10,d11,[sp,#32]
+ stp d12,d13,[sp,#48]
+ stp d14,d15,[sp,#64]
+
+ fmov d24,x10
+ fmov d25,x11
+ fmov d26,x12
+ fmov d27,x13
+ fmov d28,x14
+
+ ////////////////////////////////// initialize r^n table
+ mov x4,x7 // r^1
+ add x9,x8,x8,lsr#2 // s1 = r1 + (r1 >> 2)
+ mov x5,x8
+ mov x6,xzr
+ add x0,x0,#48+12
+ bl __poly1305_splat
+
+ bl __poly1305_mult // r^2
+ sub x0,x0,#4
+ bl __poly1305_splat
+
+ bl __poly1305_mult // r^3
+ sub x0,x0,#4
+ bl __poly1305_splat
+
+ bl __poly1305_mult // r^4
+ sub x0,x0,#4
+ bl __poly1305_splat
+ ldr x30,[sp,#8]
+
+ add x16,x1,#32
+ adr x17,.Lzeros
+ subs x2,x2,#64
+ csel x16,x17,x16,lo
+
+ mov x4,#1
+ str x4,[x0,#-24] // set is_base2_26
+ sub x0,x0,#48 // restore original x0
+ b .Ldo_neon
+
+.align 4
+.Leven_neon:
+ add x16,x1,#32
+ adr x17,.Lzeros
+ subs x2,x2,#64
+ csel x16,x17,x16,lo
+
+ stp d8,d9,[sp,#16] // meet ABI requirements
+ stp d10,d11,[sp,#32]
+ stp d12,d13,[sp,#48]
+ stp d14,d15,[sp,#64]
+
+ fmov d24,x10
+ fmov d25,x11
+ fmov d26,x12
+ fmov d27,x13
+ fmov d28,x14
+
+.Ldo_neon:
+ ldp x8,x12,[x16],#16 // inp[2:3] (or zero)
+ ldp x9,x13,[x16],#48
+
+ lsl x3,x3,#24
+ add x15,x0,#48
+
+#ifdef __ARMEB__
+ rev x8,x8
+ rev x12,x12
+ rev x9,x9
+ rev x13,x13
+#endif
+ and x4,x8,#0x03ffffff // base 2^64 -> base 2^26
+ and x5,x9,#0x03ffffff
+ ubfx x6,x8,#26,#26
+ ubfx x7,x9,#26,#26
+ add x4,x4,x5,lsl#32 // bfi x4,x5,#32,#32
+ extr x8,x12,x8,#52
+ extr x9,x13,x9,#52
+ add x6,x6,x7,lsl#32 // bfi x6,x7,#32,#32
+ fmov d14,x4
+ and x8,x8,#0x03ffffff
+ and x9,x9,#0x03ffffff
+ ubfx x10,x12,#14,#26
+ ubfx x11,x13,#14,#26
+ add x12,x3,x12,lsr#40
+ add x13,x3,x13,lsr#40
+ add x8,x8,x9,lsl#32 // bfi x8,x9,#32,#32
+ fmov d15,x6
+ add x10,x10,x11,lsl#32 // bfi x10,x11,#32,#32
+ add x12,x12,x13,lsl#32 // bfi x12,x13,#32,#32
+ fmov d16,x8
+ fmov d17,x10
+ fmov d18,x12
+
+ ldp x8,x12,[x1],#16 // inp[0:1]
+ ldp x9,x13,[x1],#48
+
+ ld1 {v0.4s,v1.4s,v2.4s,v3.4s},[x15],#64
+ ld1 {v4.4s,v5.4s,v6.4s,v7.4s},[x15],#64
+ ld1 {v8.4s},[x15]
+
+#ifdef __ARMEB__
+ rev x8,x8
+ rev x12,x12
+ rev x9,x9
+ rev x13,x13
+#endif
+ and x4,x8,#0x03ffffff // base 2^64 -> base 2^26
+ and x5,x9,#0x03ffffff
+ ubfx x6,x8,#26,#26
+ ubfx x7,x9,#26,#26
+ add x4,x4,x5,lsl#32 // bfi x4,x5,#32,#32
+ extr x8,x12,x8,#52
+ extr x9,x13,x9,#52
+ add x6,x6,x7,lsl#32 // bfi x6,x7,#32,#32
+ fmov d9,x4
+ and x8,x8,#0x03ffffff
+ and x9,x9,#0x03ffffff
+ ubfx x10,x12,#14,#26
+ ubfx x11,x13,#14,#26
+ add x12,x3,x12,lsr#40
+ add x13,x3,x13,lsr#40
+ add x8,x8,x9,lsl#32 // bfi x8,x9,#32,#32
+ fmov d10,x6
+ add x10,x10,x11,lsl#32 // bfi x10,x11,#32,#32
+ add x12,x12,x13,lsl#32 // bfi x12,x13,#32,#32
+ movi v31.2d,#-1
+ fmov d11,x8
+ fmov d12,x10
+ fmov d13,x12
+ ushr v31.2d,v31.2d,#38
+
+ b.ls .Lskip_loop
+
+.align 4
+.Loop_neon:
+ ////////////////////////////////////////////////////////////////
+ // ((inp[0]*r^4+inp[2]*r^2+inp[4])*r^4+inp[6]*r^2
+ // ((inp[1]*r^4+inp[3]*r^2+inp[5])*r^3+inp[7]*r
+ // ___________________/
+ // ((inp[0]*r^4+inp[2]*r^2+inp[4])*r^4+inp[6]*r^2+inp[8])*r^2
+ // ((inp[1]*r^4+inp[3]*r^2+inp[5])*r^4+inp[7]*r^2+inp[9])*r
+ // ___________________/ ____________________/
+ //
+ // Note that we start with inp[2:3]*r^2. This is because it
+ // doesn't depend on reduction in previous iteration.
+ ////////////////////////////////////////////////////////////////
+ // d4 = h0*r4 + h1*r3 + h2*r2 + h3*r1 + h4*r0
+ // d3 = h0*r3 + h1*r2 + h2*r1 + h3*r0 + h4*5*r4
+ // d2 = h0*r2 + h1*r1 + h2*r0 + h3*5*r4 + h4*5*r3
+ // d1 = h0*r1 + h1*r0 + h2*5*r4 + h3*5*r3 + h4*5*r2
+ // d0 = h0*r0 + h1*5*r4 + h2*5*r3 + h3*5*r2 + h4*5*r1
+
+ subs x2,x2,#64
+ umull v23.2d,v14.2s,v7.s[2]
+ csel x16,x17,x16,lo
+ umull v22.2d,v14.2s,v5.s[2]
+ umull v21.2d,v14.2s,v3.s[2]
+ ldp x8,x12,[x16],#16 // inp[2:3] (or zero)
+ umull v20.2d,v14.2s,v1.s[2]
+ ldp x9,x13,[x16],#48
+ umull v19.2d,v14.2s,v0.s[2]
+#ifdef __ARMEB__
+ rev x8,x8
+ rev x12,x12
+ rev x9,x9
+ rev x13,x13
+#endif
+
+ umlal v23.2d,v15.2s,v5.s[2]
+ and x4,x8,#0x03ffffff // base 2^64 -> base 2^26
+ umlal v22.2d,v15.2s,v3.s[2]
+ and x5,x9,#0x03ffffff
+ umlal v21.2d,v15.2s,v1.s[2]
+ ubfx x6,x8,#26,#26
+ umlal v20.2d,v15.2s,v0.s[2]
+ ubfx x7,x9,#26,#26
+ umlal v19.2d,v15.2s,v8.s[2]
+ add x4,x4,x5,lsl#32 // bfi x4,x5,#32,#32
+
+ umlal v23.2d,v16.2s,v3.s[2]
+ extr x8,x12,x8,#52
+ umlal v22.2d,v16.2s,v1.s[2]
+ extr x9,x13,x9,#52
+ umlal v21.2d,v16.2s,v0.s[2]
+ add x6,x6,x7,lsl#32 // bfi x6,x7,#32,#32
+ umlal v20.2d,v16.2s,v8.s[2]
+ fmov d14,x4
+ umlal v19.2d,v16.2s,v6.s[2]
+ and x8,x8,#0x03ffffff
+
+ umlal v23.2d,v17.2s,v1.s[2]
+ and x9,x9,#0x03ffffff
+ umlal v22.2d,v17.2s,v0.s[2]
+ ubfx x10,x12,#14,#26
+ umlal v21.2d,v17.2s,v8.s[2]
+ ubfx x11,x13,#14,#26
+ umlal v20.2d,v17.2s,v6.s[2]
+ add x8,x8,x9,lsl#32 // bfi x8,x9,#32,#32
+ umlal v19.2d,v17.2s,v4.s[2]
+ fmov d15,x6
+
+ add v11.2s,v11.2s,v26.2s
+ add x12,x3,x12,lsr#40
+ umlal v23.2d,v18.2s,v0.s[2]
+ add x13,x3,x13,lsr#40
+ umlal v22.2d,v18.2s,v8.s[2]
+ add x10,x10,x11,lsl#32 // bfi x10,x11,#32,#32
+ umlal v21.2d,v18.2s,v6.s[2]
+ add x12,x12,x13,lsl#32 // bfi x12,x13,#32,#32
+ umlal v20.2d,v18.2s,v4.s[2]
+ fmov d16,x8
+ umlal v19.2d,v18.2s,v2.s[2]
+ fmov d17,x10
+
+ ////////////////////////////////////////////////////////////////
+ // (hash+inp[0:1])*r^4 and accumulate
+
+ add v9.2s,v9.2s,v24.2s
+ fmov d18,x12
+ umlal v22.2d,v11.2s,v1.s[0]
+ ldp x8,x12,[x1],#16 // inp[0:1]
+ umlal v19.2d,v11.2s,v6.s[0]
+ ldp x9,x13,[x1],#48
+ umlal v23.2d,v11.2s,v3.s[0]
+ umlal v20.2d,v11.2s,v8.s[0]
+ umlal v21.2d,v11.2s,v0.s[0]
+#ifdef __ARMEB__
+ rev x8,x8
+ rev x12,x12
+ rev x9,x9
+ rev x13,x13
+#endif
+
+ add v10.2s,v10.2s,v25.2s
+ umlal v22.2d,v9.2s,v5.s[0]
+ umlal v23.2d,v9.2s,v7.s[0]
+ and x4,x8,#0x03ffffff // base 2^64 -> base 2^26
+ umlal v21.2d,v9.2s,v3.s[0]
+ and x5,x9,#0x03ffffff
+ umlal v19.2d,v9.2s,v0.s[0]
+ ubfx x6,x8,#26,#26
+ umlal v20.2d,v9.2s,v1.s[0]
+ ubfx x7,x9,#26,#26
+
+ add v12.2s,v12.2s,v27.2s
+ add x4,x4,x5,lsl#32 // bfi x4,x5,#32,#32
+ umlal v22.2d,v10.2s,v3.s[0]
+ extr x8,x12,x8,#52
+ umlal v23.2d,v10.2s,v5.s[0]
+ extr x9,x13,x9,#52
+ umlal v19.2d,v10.2s,v8.s[0]
+ add x6,x6,x7,lsl#32 // bfi x6,x7,#32,#32
+ umlal v21.2d,v10.2s,v1.s[0]
+ fmov d9,x4
+ umlal v20.2d,v10.2s,v0.s[0]
+ and x8,x8,#0x03ffffff
+
+ add v13.2s,v13.2s,v28.2s
+ and x9,x9,#0x03ffffff
+ umlal v22.2d,v12.2s,v0.s[0]
+ ubfx x10,x12,#14,#26
+ umlal v19.2d,v12.2s,v4.s[0]
+ ubfx x11,x13,#14,#26
+ umlal v23.2d,v12.2s,v1.s[0]
+ add x8,x8,x9,lsl#32 // bfi x8,x9,#32,#32
+ umlal v20.2d,v12.2s,v6.s[0]
+ fmov d10,x6
+ umlal v21.2d,v12.2s,v8.s[0]
+ add x12,x3,x12,lsr#40
+
+ umlal v22.2d,v13.2s,v8.s[0]
+ add x13,x3,x13,lsr#40
+ umlal v19.2d,v13.2s,v2.s[0]
+ add x10,x10,x11,lsl#32 // bfi x10,x11,#32,#32
+ umlal v23.2d,v13.2s,v0.s[0]
+ add x12,x12,x13,lsl#32 // bfi x12,x13,#32,#32
+ umlal v20.2d,v13.2s,v4.s[0]
+ fmov d11,x8
+ umlal v21.2d,v13.2s,v6.s[0]
+ fmov d12,x10
+ fmov d13,x12
+
+ /////////////////////////////////////////////////////////////////
+ // lazy reduction as discussed in "NEON crypto" by D.J. Bernstein
+ // and P. Schwabe
+ //
+ // [see discussion in poly1305-armv4 module]
+
+ ushr v29.2d,v22.2d,#26
+ xtn v27.2s,v22.2d
+ ushr v30.2d,v19.2d,#26
+ and v19.16b,v19.16b,v31.16b
+ add v23.2d,v23.2d,v29.2d // h3 -> h4
+ bic v27.2s,#0xfc,lsl#24 // &=0x03ffffff
+ add v20.2d,v20.2d,v30.2d // h0 -> h1
+
+ ushr v29.2d,v23.2d,#26
+ xtn v28.2s,v23.2d
+ ushr v30.2d,v20.2d,#26
+ xtn v25.2s,v20.2d
+ bic v28.2s,#0xfc,lsl#24
+ add v21.2d,v21.2d,v30.2d // h1 -> h2
+
+ add v19.2d,v19.2d,v29.2d
+ shl v29.2d,v29.2d,#2
+ shrn v30.2s,v21.2d,#26
+ xtn v26.2s,v21.2d
+ add v19.2d,v19.2d,v29.2d // h4 -> h0
+ bic v25.2s,#0xfc,lsl#24
+ add v27.2s,v27.2s,v30.2s // h2 -> h3
+ bic v26.2s,#0xfc,lsl#24
+
+ shrn v29.2s,v19.2d,#26
+ xtn v24.2s,v19.2d
+ ushr v30.2s,v27.2s,#26
+ bic v27.2s,#0xfc,lsl#24
+ bic v24.2s,#0xfc,lsl#24
+ add v25.2s,v25.2s,v29.2s // h0 -> h1
+ add v28.2s,v28.2s,v30.2s // h3 -> h4
+
+ b.hi .Loop_neon
+
+.Lskip_loop:
+ dup v16.2d,v16.d[0]
+ add v11.2s,v11.2s,v26.2s
+
+ ////////////////////////////////////////////////////////////////
+ // multiply (inp[0:1]+hash) or inp[2:3] by r^2:r^1
+
+ adds x2,x2,#32
+ b.ne .Long_tail
+
+ dup v16.2d,v11.d[0]
+ add v14.2s,v9.2s,v24.2s
+ add v17.2s,v12.2s,v27.2s
+ add v15.2s,v10.2s,v25.2s
+ add v18.2s,v13.2s,v28.2s
+
+.Long_tail:
+ dup v14.2d,v14.d[0]
+ umull2 v19.2d,v16.4s,v6.4s
+ umull2 v22.2d,v16.4s,v1.4s
+ umull2 v23.2d,v16.4s,v3.4s
+ umull2 v21.2d,v16.4s,v0.4s
+ umull2 v20.2d,v16.4s,v8.4s
+
+ dup v15.2d,v15.d[0]
+ umlal2 v19.2d,v14.4s,v0.4s
+ umlal2 v21.2d,v14.4s,v3.4s
+ umlal2 v22.2d,v14.4s,v5.4s
+ umlal2 v23.2d,v14.4s,v7.4s
+ umlal2 v20.2d,v14.4s,v1.4s
+
+ dup v17.2d,v17.d[0]
+ umlal2 v19.2d,v15.4s,v8.4s
+ umlal2 v22.2d,v15.4s,v3.4s
+ umlal2 v21.2d,v15.4s,v1.4s
+ umlal2 v23.2d,v15.4s,v5.4s
+ umlal2 v20.2d,v15.4s,v0.4s
+
+ dup v18.2d,v18.d[0]
+ umlal2 v22.2d,v17.4s,v0.4s
+ umlal2 v23.2d,v17.4s,v1.4s
+ umlal2 v19.2d,v17.4s,v4.4s
+ umlal2 v20.2d,v17.4s,v6.4s
+ umlal2 v21.2d,v17.4s,v8.4s
+
+ umlal2 v22.2d,v18.4s,v8.4s
+ umlal2 v19.2d,v18.4s,v2.4s
+ umlal2 v23.2d,v18.4s,v0.4s
+ umlal2 v20.2d,v18.4s,v4.4s
+ umlal2 v21.2d,v18.4s,v6.4s
+
+ b.eq .Lshort_tail
+
+ ////////////////////////////////////////////////////////////////
+ // (hash+inp[0:1])*r^4:r^3 and accumulate
+
+ add v9.2s,v9.2s,v24.2s
+ umlal v22.2d,v11.2s,v1.2s
+ umlal v19.2d,v11.2s,v6.2s
+ umlal v23.2d,v11.2s,v3.2s
+ umlal v20.2d,v11.2s,v8.2s
+ umlal v21.2d,v11.2s,v0.2s
+
+ add v10.2s,v10.2s,v25.2s
+ umlal v22.2d,v9.2s,v5.2s
+ umlal v19.2d,v9.2s,v0.2s
+ umlal v23.2d,v9.2s,v7.2s
+ umlal v20.2d,v9.2s,v1.2s
+ umlal v21.2d,v9.2s,v3.2s
+
+ add v12.2s,v12.2s,v27.2s
+ umlal v22.2d,v10.2s,v3.2s
+ umlal v19.2d,v10.2s,v8.2s
+ umlal v23.2d,v10.2s,v5.2s
+ umlal v20.2d,v10.2s,v0.2s
+ umlal v21.2d,v10.2s,v1.2s
+
+ add v13.2s,v13.2s,v28.2s
+ umlal v22.2d,v12.2s,v0.2s
+ umlal v19.2d,v12.2s,v4.2s
+ umlal v23.2d,v12.2s,v1.2s
+ umlal v20.2d,v12.2s,v6.2s
+ umlal v21.2d,v12.2s,v8.2s
+
+ umlal v22.2d,v13.2s,v8.2s
+ umlal v19.2d,v13.2s,v2.2s
+ umlal v23.2d,v13.2s,v0.2s
+ umlal v20.2d,v13.2s,v4.2s
+ umlal v21.2d,v13.2s,v6.2s
+
+.Lshort_tail:
+ ////////////////////////////////////////////////////////////////
+ // horizontal add
+
+ addp v22.2d,v22.2d,v22.2d
+ ldp d8,d9,[sp,#16] // meet ABI requirements
+ addp v19.2d,v19.2d,v19.2d
+ ldp d10,d11,[sp,#32]
+ addp v23.2d,v23.2d,v23.2d
+ ldp d12,d13,[sp,#48]
+ addp v20.2d,v20.2d,v20.2d
+ ldp d14,d15,[sp,#64]
+ addp v21.2d,v21.2d,v21.2d
+
+ ////////////////////////////////////////////////////////////////
+ // lazy reduction, but without narrowing
+
+ ushr v29.2d,v22.2d,#26
+ and v22.16b,v22.16b,v31.16b
+ ushr v30.2d,v19.2d,#26
+ and v19.16b,v19.16b,v31.16b
+
+ add v23.2d,v23.2d,v29.2d // h3 -> h4
+ add v20.2d,v20.2d,v30.2d // h0 -> h1
+
+ ushr v29.2d,v23.2d,#26
+ and v23.16b,v23.16b,v31.16b
+ ushr v30.2d,v20.2d,#26
+ and v20.16b,v20.16b,v31.16b
+ add v21.2d,v21.2d,v30.2d // h1 -> h2
+
+ add v19.2d,v19.2d,v29.2d
+ shl v29.2d,v29.2d,#2
+ ushr v30.2d,v21.2d,#26
+ and v21.16b,v21.16b,v31.16b
+ add v19.2d,v19.2d,v29.2d // h4 -> h0
+ add v22.2d,v22.2d,v30.2d // h2 -> h3
+
+ ushr v29.2d,v19.2d,#26
+ and v19.16b,v19.16b,v31.16b
+ ushr v30.2d,v22.2d,#26
+ and v22.16b,v22.16b,v31.16b
+ add v20.2d,v20.2d,v29.2d // h0 -> h1
+ add v23.2d,v23.2d,v30.2d // h3 -> h4
+
+ ////////////////////////////////////////////////////////////////
+ // write the result, can be partially reduced
+
+ st4 {v19.s,v20.s,v21.s,v22.s}[0],[x0],#16
+ st1 {v23.s}[0],[x0]
+
+.Lno_data_neon:
+ ldr x29,[sp],#80
+ ret
+ENDPROC(poly1305_blocks_neon)
+
+.align 5
+ENTRY(poly1305_emit_neon)
+ ldr x17,[x0,#24]
+ cbz x17,poly1305_emit_arm
+
+ ldp w10,w11,[x0] // load hash value base 2^26
+ ldp w12,w13,[x0,#8]
+ ldr w14,[x0,#16]
+
+ add x4,x10,x11,lsl#26 // base 2^26 -> base 2^64
+ lsr x5,x12,#12
+ adds x4,x4,x12,lsl#52
+ add x5,x5,x13,lsl#14
+ adc x5,x5,xzr
+ lsr x6,x14,#24
+ adds x5,x5,x14,lsl#40
+ adc x6,x6,xzr // can be partially reduced...
+
+ ldp x10,x11,[x2] // load nonce
+
+ and x12,x6,#-4 // ... so reduce
+ add x12,x12,x6,lsr#2
+ and x6,x6,#3
+ adds x4,x4,x12
+ adcs x5,x5,xzr
+ adc x6,x6,xzr
+
+ adds x12,x4,#5 // compare to modulus
+ adcs x13,x5,xzr
+ adc x14,x6,xzr
+
+ tst x14,#-4 // see if it's carried/borrowed
+
+ csel x4,x4,x12,eq
+ csel x5,x5,x13,eq
+
+#ifdef __ARMEB__
+ ror x10,x10,#32 // flip nonce words
+ ror x11,x11,#32
+#endif
+ adds x4,x4,x10 // accumulate nonce
+ adc x5,x5,x11
+#ifdef __ARMEB__
+ rev x4,x4 // flip output bytes
+ rev x5,x5
+#endif
+ stp x4,x5,[x1] // write result
+
+ ret
+ENDPROC(poly1305_emit_neon)
+
+.align 5
+.Lzeros:
+.long 0,0,0,0,0,0,0,0
--
2.18.0
^ permalink raw reply related
* [PATCH v2 03/17] zinc: ChaCha20 generic C implementation
From: Jason A. Donenfeld @ 2018-08-24 21:38 UTC (permalink / raw)
To: linux-kernel, netdev, davem
Cc: Jason A. Donenfeld, Andy Lutomirski, Greg KH, Samuel Neves,
Jean-Philippe Aumasson, linux-crypto
In-Reply-To: <20180824213849.23647-1-Jason@zx2c4.com>
This implements the ChaCha20 permutation as a single C statement, by way
of the comma operator, which the compiler is able to simplify
terrifically.
Information: https://cr.yp.to/chacha.html
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Greg KH <gregkh@linuxfoundation.org>
Cc: Samuel Neves <sneves@dei.uc.pt>
Cc: Jean-Philippe Aumasson <jeanphilippe.aumasson@gmail.com>
Cc: linux-crypto@vger.kernel.org
---
include/zinc/chacha20.h | 47 ++++++++++++
lib/zinc/Kconfig | 4 +
lib/zinc/Makefile | 4 +
lib/zinc/chacha20/chacha20.c | 137 +++++++++++++++++++++++++++++++++++
lib/zinc/main.c | 5 ++
5 files changed, 197 insertions(+)
create mode 100644 include/zinc/chacha20.h
create mode 100644 lib/zinc/chacha20/chacha20.c
diff --git a/include/zinc/chacha20.h b/include/zinc/chacha20.h
new file mode 100644
index 000000000000..ba50a2424ad0
--- /dev/null
+++ b/include/zinc/chacha20.h
@@ -0,0 +1,47 @@
+/* SPDX-License-Identifier: GPL-2.0
+ *
+ * Copyright (C) 2015-2018 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#ifndef _ZINC_CHACHA20_H
+#define _ZINC_CHACHA20_H
+
+#include <linux/simd.h>
+#include <linux/kernel.h>
+#include <linux/types.h>
+
+enum {
+ CHACHA20_IV_SIZE = 16,
+ CHACHA20_KEY_SIZE = 32,
+ CHACHA20_BLOCK_SIZE = 64,
+ HCHACHA20_KEY_SIZE = 32,
+ HCHACHA20_NONCE_SIZE = 16
+};
+
+struct chacha20_ctx {
+ u32 key[8];
+ u32 counter[4];
+} __aligned(32);
+
+void chacha20_fpu_init(void);
+
+static inline void chacha20_init(struct chacha20_ctx *state, const u8 key[CHACHA20_KEY_SIZE], const u64 nonce)
+{
+ __le32 *le_key = (__le32 *)key;
+ state->key[0] = le32_to_cpu(le_key[0]);
+ state->key[1] = le32_to_cpu(le_key[1]);
+ state->key[2] = le32_to_cpu(le_key[2]);
+ state->key[3] = le32_to_cpu(le_key[3]);
+ state->key[4] = le32_to_cpu(le_key[4]);
+ state->key[5] = le32_to_cpu(le_key[5]);
+ state->key[6] = le32_to_cpu(le_key[6]);
+ state->key[7] = le32_to_cpu(le_key[7]);
+ state->counter[0] = state->counter[1] = 0;
+ state->counter[2] = nonce & U32_MAX;
+ state->counter[3] = nonce >> 32;
+}
+void chacha20(struct chacha20_ctx *state, u8 *dst, const u8 *src, u32 len, simd_context_t simd_context);
+
+void hchacha20(u8 derived_key[CHACHA20_KEY_SIZE], const u8 nonce[HCHACHA20_NONCE_SIZE], const u8 key[HCHACHA20_KEY_SIZE], simd_context_t simd_context);
+
+#endif /* _ZINC_CHACHA20_H */
diff --git a/lib/zinc/Kconfig b/lib/zinc/Kconfig
index aa4f8d449d6b..5311a0d6ba8b 100644
--- a/lib/zinc/Kconfig
+++ b/lib/zinc/Kconfig
@@ -6,6 +6,10 @@ config ZINC
select NEON
select KERNEL_MODE_NEON
+config ZINC_CHACHA20
+ bool
+ select ZINC
+
config ZINC_DEBUG
bool "Zinc cryptography library debugging and self-tests"
depends on ZINC
diff --git a/lib/zinc/Makefile b/lib/zinc/Makefile
index 8e30115217db..6ec77feb4ab7 100644
--- a/lib/zinc/Makefile
+++ b/lib/zinc/Makefile
@@ -2,6 +2,10 @@ ccflags-y := -O3
ccflags-y += -Wframe-larger-than=8192
ccflags-y += -D'pr_fmt(fmt)=KBUILD_MODNAME ": " fmt'
+ifeq ($(CONFIG_ZINC_CHACHA20),y)
+zinc-y += chacha20/chacha20.o
+endif
+
zinc-y += main.o
obj-$(CONFIG_ZINC) := zinc.o
diff --git a/lib/zinc/chacha20/chacha20.c b/lib/zinc/chacha20/chacha20.c
new file mode 100644
index 000000000000..29c487375ccd
--- /dev/null
+++ b/lib/zinc/chacha20/chacha20.c
@@ -0,0 +1,137 @@
+/* SPDX-License-Identifier: GPL-2.0
+ *
+ * Copyright (C) 2015-2018 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ */
+
+#include <zinc/chacha20.h>
+
+#include <linux/kernel.h>
+#include <crypto/algapi.h>
+
+#ifndef HAVE_CHACHA20_ARCH_IMPLEMENTATION
+void __init chacha20_fpu_init(void) { }
+static inline bool chacha20_arch(u8 *out, const u8 *in, const size_t len, const u32 key[8], const u32 counter[4], simd_context_t simd_context) { return false; }
+static inline bool hchacha20_arch(u8 *derived_key, const u8 *nonce, const u8 *key, simd_context_t simd_context) { return false; }
+#endif
+
+#define EXPAND_32_BYTE_K 0x61707865U, 0x3320646eU, 0x79622d32U, 0x6b206574U
+
+#define QUARTER_ROUND(x, a, b, c, d) ( \
+ x[a] += x[b], \
+ x[d] = rol32((x[d] ^ x[a]), 16), \
+ x[c] += x[d], \
+ x[b] = rol32((x[b] ^ x[c]), 12), \
+ x[a] += x[b], \
+ x[d] = rol32((x[d] ^ x[a]), 8), \
+ x[c] += x[d], \
+ x[b] = rol32((x[b] ^ x[c]), 7) \
+)
+
+#define C(i, j) (i * 4 + j)
+
+#define DOUBLE_ROUND(x) ( \
+ /* Column Round */ \
+ QUARTER_ROUND(x, C(0, 0), C(1, 0), C(2, 0), C(3, 0)), \
+ QUARTER_ROUND(x, C(0, 1), C(1, 1), C(2, 1), C(3, 1)), \
+ QUARTER_ROUND(x, C(0, 2), C(1, 2), C(2, 2), C(3, 2)), \
+ QUARTER_ROUND(x, C(0, 3), C(1, 3), C(2, 3), C(3, 3)), \
+ /* Diagonal Round */ \
+ QUARTER_ROUND(x, C(0, 0), C(1, 1), C(2, 2), C(3, 3)), \
+ QUARTER_ROUND(x, C(0, 1), C(1, 2), C(2, 3), C(3, 0)), \
+ QUARTER_ROUND(x, C(0, 2), C(1, 3), C(2, 0), C(3, 1)), \
+ QUARTER_ROUND(x, C(0, 3), C(1, 0), C(2, 1), C(3, 2)) \
+)
+
+#define TWENTY_ROUNDS(x) ( \
+ DOUBLE_ROUND(x), \
+ DOUBLE_ROUND(x), \
+ DOUBLE_ROUND(x), \
+ DOUBLE_ROUND(x), \
+ DOUBLE_ROUND(x), \
+ DOUBLE_ROUND(x), \
+ DOUBLE_ROUND(x), \
+ DOUBLE_ROUND(x), \
+ DOUBLE_ROUND(x), \
+ DOUBLE_ROUND(x) \
+)
+
+static void chacha20_block_generic(__le32 *stream, u32 *state)
+{
+ u32 x[CHACHA20_BLOCK_SIZE / sizeof(u32)];
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(x); ++i)
+ x[i] = state[i];
+
+ TWENTY_ROUNDS(x);
+
+ for (i = 0; i < ARRAY_SIZE(x); ++i)
+ stream[i] = cpu_to_le32(x[i] + state[i]);
+
+ ++state[12];
+}
+
+static void chacha20_generic(u8 *out, const u8 *in, u32 len, const u32 key[8], const u32 counter[4])
+{
+ __le32 buf[CHACHA20_BLOCK_SIZE / sizeof(__le32)];
+ u32 x[] = {
+ EXPAND_32_BYTE_K,
+ key[0], key[1], key[2], key[3],
+ key[4], key[5], key[6], key[7],
+ counter[0], counter[1], counter[2], counter[3]
+ };
+
+ if (out != in)
+ memmove(out, in, len);
+
+ while (len >= CHACHA20_BLOCK_SIZE) {
+ chacha20_block_generic(buf, x);
+ crypto_xor(out, (u8 *)buf, CHACHA20_BLOCK_SIZE);
+ len -= CHACHA20_BLOCK_SIZE;
+ out += CHACHA20_BLOCK_SIZE;
+ }
+ if (len) {
+ chacha20_block_generic(buf, x);
+ crypto_xor(out, (u8 *)buf, len);
+ }
+}
+
+void chacha20(struct chacha20_ctx *state, u8 *dst, const u8 *src, u32 len, simd_context_t simd_context)
+{
+ if (!chacha20_arch(dst, src, len, state->key, state->counter, simd_context))
+ chacha20_generic(dst, src, len, state->key, state->counter);
+ state->counter[0] += (len + 63) / 64;
+}
+EXPORT_SYMBOL(chacha20);
+
+static void hchacha20_generic(u8 derived_key[CHACHA20_KEY_SIZE], const u8 nonce[HCHACHA20_NONCE_SIZE], const u8 key[HCHACHA20_KEY_SIZE])
+{
+ __le32 *out = (__force __le32 *)derived_key;
+ u32 x[] = {
+ EXPAND_32_BYTE_K,
+ le32_to_cpup((__le32 *)(key + 0)), le32_to_cpup((__le32 *)(key + 4)), le32_to_cpup((__le32 *)(key + 8)), le32_to_cpup((__le32 *)(key + 12)),
+ le32_to_cpup((__le32 *)(key + 16)), le32_to_cpup((__le32 *)(key + 20)), le32_to_cpup((__le32 *)(key + 24)), le32_to_cpup((__le32 *)(key + 28)),
+ le32_to_cpup((__le32 *)(nonce + 0)), le32_to_cpup((__le32 *)(nonce + 4)), le32_to_cpup((__le32 *)(nonce + 8)), le32_to_cpup((__le32 *)(nonce + 12))
+ };
+
+ TWENTY_ROUNDS(x);
+
+ out[0] = cpu_to_le32(x[0]);
+ out[1] = cpu_to_le32(x[1]);
+ out[2] = cpu_to_le32(x[2]);
+ out[3] = cpu_to_le32(x[3]);
+ out[4] = cpu_to_le32(x[12]);
+ out[5] = cpu_to_le32(x[13]);
+ out[6] = cpu_to_le32(x[14]);
+ out[7] = cpu_to_le32(x[15]);
+}
+
+void hchacha20(u8 derived_key[CHACHA20_KEY_SIZE], const u8 nonce[HCHACHA20_NONCE_SIZE], const u8 key[HCHACHA20_KEY_SIZE], simd_context_t simd_context)
+{
+ if (!hchacha20_arch(derived_key, nonce, key, simd_context))
+ hchacha20_generic(derived_key, nonce, key);
+}
+/* Deliberately not EXPORT_SYMBOL'd, since there are few reasons why somebody
+ * should be using this directly, rather than via xchacha20. Revisit only in
+ * the unlikely event that somebody has a good reason to export this.
+ */
diff --git a/lib/zinc/main.c b/lib/zinc/main.c
index 590872563955..79b85045f22f 100644
--- a/lib/zinc/main.c
+++ b/lib/zinc/main.c
@@ -3,6 +3,8 @@
* Copyright (C) 2015-2018 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
*/
+#include <zinc/chacha20.h>
+
#include <linux/init.h>
#include <linux/module.h>
@@ -17,6 +19,9 @@
static int __init mod_init(void)
{
+#ifdef CONFIG_ZINC_CHACHA20
+ chacha20_fpu_init();
+#endif
return 0;
}
--
2.18.0
^ permalink raw reply related
* [PATCH v2 00/17] WireGuard: Secure Network Tunnel
From: Jason A. Donenfeld @ 2018-08-24 21:38 UTC (permalink / raw)
To: linux-kernel, netdev, davem; +Cc: Jason A. Donenfeld
This patchset is available on git.kernel.org in this branch, where it may be
pulled directly for inclusion into net-next:
* https://git.kernel.org/pub/scm/linux/kernel/git/zx2c4/linux.git/log/?h=jd/wireguard
WireGuard is a secure network tunnel written especially for Linux, which
has faced around three years of serious development, deployment, and
scrutiny. It delivers excellent performance and is extremely easy to
use and configure. It has been designed with the primary goal of being
both easy to audit by virtue of being small and highly secure from a
cryptography and systems security perspective. WireGuard is used by some
massive companies pushing enormous amounts of traffic, and likely
already today you've consumed bytes that at some point transited through
a WireGuard tunnel. Even as an out-of-tree module, WireGuard has been
integrated into various userspace tools, Linux distributions, mobile
phones, and data centers. There are ports in several languages to
several operating systems, and even commercial hardware and services
sold integrating WireGuard. It is time, therefore, for WireGuard to be
properly integrated into Linux.
Ample information, including documentation, installation instructions,
and project details, is available at:
* https://www.wireguard.com/
* https://www.wireguard.com/papers/wireguard.pdf
As it is currently an out-of-tree module, it lives in its own git repo
and has its own mailing list, and every commit for the module is tested
against every stable kernel since 3.10 on a variety of architectures
using an extensive test suite:
* https://git.zx2c4.com/WireGuard
https://git.kernel.org/pub/scm/linux/kernel/git/zx2c4/WireGuard.git/
* https://lists.zx2c4.com/mailman/listinfo/wireguard
* https://www.wireguard.com/build-status/
The project has been broadly discussed at conferences, and was presented
to the Netdev developers in Seoul last November, where a paper was
released detailing some interesting aspects of the project. Dave asked
me after the talk if I would consider sending in a v1 "sooner rather
than later", hence this patchset. A decision is still waiting from the
LPC network track committee, but an update on these topics may be
presented in Vancouver in a few months. Presentations:
* https://www.wireguard.com/presentations/
* https://www.wireguard.com/papers/wireguard-netdev22.pdf
The cryptography in the protocol itself has been formally verified by
several independent academic teams with positive results, and I know of
two additional efforts on their way to further corroborate those
findings. The version 1 protocol is "complete", and so the purpose of
this review is to assess the implementation of the protocol. However, it
still may be of interest to know that the thing you're reviewing uses a
protocol with various nice security properties:
* https://www.wireguard.com/formal-verification/
This patchset is divided into three segments. The first introduces a very
simple helper for working with the FPU state for the purposes of amortizing
SIMD operations. The second segment is a small collection of cryptographic
primitives, split up into several commits by primitive and by hardware. The
third is WireGuard itself, presented as an unintrusive and self-contained
virtual network driver.
It is intended that this entire patch series enter the kernel through
DaveM's net-next tree.
Changes v1->v2:
- Zinc has been split into multiple patches.
- Zinc has been split into multiple config options.
- WireGuard has been wrapped at 80 characters.
- Numerous cleanups of inside WireGuard.
- Splitting out the simd helper into linux/.
- Simplification of bitwise arithmetic inside some Zinc primitives.
- Better handling of unaligned buffers inside Zinc.
- Expanded CC list.
Enjoy,
Jason
^ permalink raw reply
* [PATCH bpf 2/2] bpf, sockmap: fix psock refcount leak in bpf_tcp_recvmsg
From: Daniel Borkmann @ 2018-08-24 20:08 UTC (permalink / raw)
To: alexei.starovoitov; +Cc: john.fastabend, netdev, Daniel Borkmann
In-Reply-To: <20180824200851.12308-1-daniel@iogearbox.net>
In bpf_tcp_recvmsg() we first took a reference on the psock, however
once we find that there are skbs in the normal socket's receive queue
we return with processing them through tcp_recvmsg(). Problem is that
we leak the taken reference on the psock in that path. Given we don't
really do anything with the psock at this point, move the skb_queue_empty()
test before we fetch the psock to fix this case.
Fixes: 8934ce2fd081 ("bpf: sockmap redirect ingress support")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
---
kernel/bpf/sockmap.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/kernel/bpf/sockmap.c b/kernel/bpf/sockmap.c
index 01879e4..26d8a30 100644
--- a/kernel/bpf/sockmap.c
+++ b/kernel/bpf/sockmap.c
@@ -912,6 +912,8 @@ static int bpf_tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
if (unlikely(flags & MSG_ERRQUEUE))
return inet_recv_error(sk, msg, len, addr_len);
+ if (!skb_queue_empty(&sk->sk_receive_queue))
+ return tcp_recvmsg(sk, msg, len, nonblock, flags, addr_len);
rcu_read_lock();
psock = smap_psock_sk(sk);
@@ -922,9 +924,6 @@ static int bpf_tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
goto out;
rcu_read_unlock();
- if (!skb_queue_empty(&sk->sk_receive_queue))
- return tcp_recvmsg(sk, msg, len, nonblock, flags, addr_len);
-
lock_sock(sk);
bytes_ready:
while (copied != len) {
--
2.9.5
^ permalink raw reply related
* [PATCH bpf 1/2] bpf, sockmap: fix potential use after free in bpf_tcp_close
From: Daniel Borkmann @ 2018-08-24 20:08 UTC (permalink / raw)
To: alexei.starovoitov; +Cc: john.fastabend, netdev, Daniel Borkmann
In-Reply-To: <20180824200851.12308-1-daniel@iogearbox.net>
bpf_tcp_close() we pop the psock linkage to a map via psock_map_pop().
A parallel update on the sock hash map can happen between psock_map_pop()
and lookup_elem_raw() where we override the element under link->hash /
link->key. In bpf_tcp_close()'s lookup_elem_raw() we subsequently only
test whether an element is present, but we do not test whether the
element is infact the element we were looking for.
We lock the sock in bpf_tcp_close() during that time, so do we hold
the lock in sock_hash_update_elem(). However, the latter locks the
sock which is newly updated, not the one we're purging from the hash
table. This means that while one CPU is doing the lookup from bpf_tcp_close(),
another CPU is doing the map update in parallel, dropped our sock from
the hlist and released the psock.
Subsequently the first CPU will find the new sock and attempts to drop
and release the old sock yet another time. Fix is that we need to check
the elements for a match after lookup, similar as we do in the sock map.
Note that the hash tab elems are freed via RCU, so access to their
link->hash / link->key is fine since we're under RCU read side there.
Fixes: e9db4ef6bf4c ("bpf: sockhash fix omitted bucket lock in sock_close")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
---
kernel/bpf/sockmap.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kernel/bpf/sockmap.c b/kernel/bpf/sockmap.c
index cf5195c..01879e4 100644
--- a/kernel/bpf/sockmap.c
+++ b/kernel/bpf/sockmap.c
@@ -369,7 +369,7 @@ static void bpf_tcp_close(struct sock *sk, long timeout)
/* If another thread deleted this object skip deletion.
* The refcnt on psock may or may not be zero.
*/
- if (l) {
+ if (l && l == link) {
hlist_del_rcu(&link->hash_node);
smap_release_sock(psock, link->sk);
free_htab_elem(htab, link);
--
2.9.5
^ permalink raw reply related
* [PATCH bpf 0/2] Two BPF sockmap fixes
From: Daniel Borkmann @ 2018-08-24 20:08 UTC (permalink / raw)
To: alexei.starovoitov; +Cc: john.fastabend, netdev, Daniel Borkmann
This series contains two fixes found while working with the BPF
sockmap code. One refcount leakage and one case for a small window
of use after free. See patches for more info. Thanks!
Daniel Borkmann (2):
bpf, sockmap: fix potential use after free in bpf_tcp_close
bpf, sockmap: fix psock refcount leak in bpf_tcp_recvmsg
kernel/bpf/sockmap.c | 7 +++----
1 file changed, 3 insertions(+), 4 deletions(-)
--
2.9.5
^ permalink raw reply
* Re: KASAN: use-after-free Read in __rhashtable_lookup (2)
From: Sowmini Varadhan @ 2018-08-24 23:37 UTC (permalink / raw)
To: Dmitry Vyukov
Cc: syzbot, David Miller, LKML, linux-rdma, netdev, rds-devel,
Santosh Shilimkar, syzkaller-bugs
In-Reply-To: <CACT4Y+a6P+q8jtRrd=cAf5hWbqQO4CQdWUaCCX-ygRbGsWHnGQ@mail.gmail.com>
On (08/24/18 16:10), Dmitry Vyukov wrote:
> You need a colon after dup.
I see.
> But this can't possibly be a dup of that bug. That bug was fixed in February:
Apologies, I unknowingly made a mistake in my
syzbot-tiquette in that case,
I did not intend to *close* the bug as a dup, I was
merely trying to indicate that this is yet another
manifestation of the original report with subject
"KASAN: use-after-free Read in rds_find_bound"
All of these problems are happening because we are missing
a synchronize_net() in rds_release.
A full description of the race conditon can be found in
https://www.spinics.net/lists/netdev/msg475074.html
If rds_release is going to nuke the rds_socket
it needs to sychronize_net() with other threads that may be
looking up the bind_hash_table. If we dont do that,
all of these "use-after-free" races can happen.
As long as we dont have the synchronize_net() we are
missing a big (and standard) bit of RCU synchronization.
--Sowmini
^ permalink raw reply
* Re: Re: KASAN: use-after-free Read in __rhashtable_lookup (2)
From: Dmitry Vyukov @ 2018-08-24 23:10 UTC (permalink / raw)
To: syzbot
Cc: Sowmini Varadhan, David Miller, LKML, linux-rdma, netdev,
rds-devel, Santosh Shilimkar, syzkaller-bugs
In-Reply-To: <000000000000d8879a05743667a2@google.com>
On Fri, Aug 24, 2018 at 4:05 PM, syzbot
<syzbot+8967084bcac563795dc6@syzkaller.appspotmail.com> wrote:
>
>
>> #syz dup KASAN: use-after-free Read in rds_find_bound
>
>
> unknown command "dup"
You need a colon after dup.
But this can't possibly be a dup of that bug. That bug was fixed in February:
https://syzkaller.appspot.com/bug?extid=93a5839deb355537440f
And this crash happened a day ago. This is a different bug. If you dup
it to a closed one, this one will also be considered closed, and then
a new crash will immediately create a new bug.
>> The problem is a missing synchronize_net() in rds_release.
>> References:
>> https://www.spinics.net/lists/netdev/msg475074.html
>> https://www.spinics.net/lists/netdev/msg507908.html
>
>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "syzkaller-bugs" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to syzkaller-bugs+unsubscribe@googlegroups.com.
>> To view this discussion on the web visit
>> https://groups.google.com/d/msgid/syzkaller-bugs/20180824230446.GA16891%40oracle.com.
>> For more options, visit https://groups.google.com/d/optout.
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "syzkaller-bugs" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to syzkaller-bugs+unsubscribe@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/syzkaller-bugs/000000000000d8879a05743667a2%40google.com.
>
> For more options, visit https://groups.google.com/d/optout.
^ permalink raw reply
* Re: Re: KASAN: use-after-free Read in __rhashtable_lookup (2)
From: syzbot @ 2018-08-24 23:05 UTC (permalink / raw)
To: Sowmini Varadhan
Cc: davem, linux-kernel, linux-rdma, netdev, rds-devel,
santosh.shilimkar, sowmini.varadhan, syzkaller-bugs
In-Reply-To: <20180824230446.GA16891@oracle.com>
> #syz dup KASAN: use-after-free Read in rds_find_bound
unknown command "dup"
> The problem is a missing synchronize_net() in rds_release.
> References:
> https://www.spinics.net/lists/netdev/msg475074.html
> https://www.spinics.net/lists/netdev/msg507908.html
> --
> You received this message because you are subscribed to the Google
> Groups "syzkaller-bugs" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to syzkaller-bugs+unsubscribe@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/syzkaller-bugs/20180824230446.GA16891%40oracle.com.
> For more options, visit https://groups.google.com/d/optout.
^ permalink raw reply
* Re: Re: KASAN: use-after-free Read in __rhashtable_lookup (2)
From: syzbot @ 2018-08-24 23:05 UTC (permalink / raw)
To: Sowmini Varadhan
Cc: davem, linux-kernel, linux-rdma, netdev, rds-devel,
santosh.shilimkar, sowmini.varadhan, syzkaller-bugs
In-Reply-To: <20180824230446.GA16891@oracle.com>
> #syz dup KASAN: use-after-free Read in rds_find_bound
unknown command "dup"
> The problem is a missing synchronize_net() in rds_release.
> References:
> https://www.spinics.net/lists/netdev/msg475074.html
> https://www.spinics.net/lists/netdev/msg507908.html
^ permalink raw reply
* [Patch net] tipc: switch to rhashtable iterator
From: Cong Wang @ 2018-08-24 19:28 UTC (permalink / raw)
To: netdev; +Cc: tipc-discussion, Cong Wang, Jon Maloy, Ying Xue
syzbot reported a use-after-free in tipc_group_fill_sock_diag(),
where tipc_group_fill_sock_diag() still reads tsk->group meanwhile
tipc_group_delete() just deletes it in tipc_release().
tipc_nl_sk_walk() aims to lock this sock when walking each sock
in the hash table to close race conditions with sock changes like
this one, by acquiring tsk->sk.sk_lock.slock spinlock, unfortunately
this doesn't work at all. All non-BH call path should take
lock_sock() instead to make it work.
tipc_nl_sk_walk() brutally iterates with raw rht_for_each_entry_rcu()
where RCU read lock is required, this is the reason why lock_sock()
can't be taken on this path. This could be resolved by switching to
rhashtable iterator API's, where taking a sleepable lock is possible.
Also, the iterator API's are friendly for restartable calls like
diag dump, the last position is remembered behind the scence,
all we need to do here is saving the iterator into cb->args[].
I tested this with parallel tipc diag dump and thousands of tipc
socket creation and release, no crash or memory leak.
Reported-by: syzbot+b9c8f3ab2994b7cd1625@syzkaller.appspotmail.com
Cc: Jon Maloy <jon.maloy@ericsson.com>
Cc: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
---
net/tipc/diag.c | 2 ++
net/tipc/netlink.c | 2 ++
net/tipc/socket.c | 76 +++++++++++++++++++++++++++++++++++-------------------
net/tipc/socket.h | 2 ++
4 files changed, 56 insertions(+), 26 deletions(-)
diff --git a/net/tipc/diag.c b/net/tipc/diag.c
index aaabb0b776dd..73137f4aeb68 100644
--- a/net/tipc/diag.c
+++ b/net/tipc/diag.c
@@ -84,7 +84,9 @@ static int tipc_sock_diag_handler_dump(struct sk_buff *skb,
if (h->nlmsg_flags & NLM_F_DUMP) {
struct netlink_dump_control c = {
+ .start = tipc_dump_start,
.dump = tipc_diag_dump,
+ .done = tipc_dump_done,
};
netlink_dump_start(net->diag_nlsk, skb, h, &c);
return 0;
diff --git a/net/tipc/netlink.c b/net/tipc/netlink.c
index 6ff2254088f6..99ee419210ba 100644
--- a/net/tipc/netlink.c
+++ b/net/tipc/netlink.c
@@ -167,7 +167,9 @@ static const struct genl_ops tipc_genl_v2_ops[] = {
},
{
.cmd = TIPC_NL_SOCK_GET,
+ .start = tipc_dump_start,
.dumpit = tipc_nl_sk_dump,
+ .done = tipc_dump_done,
.policy = tipc_nl_policy,
},
{
diff --git a/net/tipc/socket.c b/net/tipc/socket.c
index c9a50b62c738..ab7a2a7178f7 100644
--- a/net/tipc/socket.c
+++ b/net/tipc/socket.c
@@ -3229,45 +3229,69 @@ int tipc_nl_sk_walk(struct sk_buff *skb, struct netlink_callback *cb,
struct netlink_callback *cb,
struct tipc_sock *tsk))
{
- struct net *net = sock_net(skb->sk);
- struct tipc_net *tn = tipc_net(net);
- const struct bucket_table *tbl;
- u32 prev_portid = cb->args[1];
- u32 tbl_id = cb->args[0];
- struct rhash_head *pos;
+ struct rhashtable_iter *iter = (void *)cb->args[0];
struct tipc_sock *tsk;
int err;
- rcu_read_lock();
- tbl = rht_dereference_rcu((&tn->sk_rht)->tbl, &tn->sk_rht);
- for (; tbl_id < tbl->size; tbl_id++) {
- rht_for_each_entry_rcu(tsk, pos, tbl, tbl_id, node) {
- spin_lock_bh(&tsk->sk.sk_lock.slock);
- if (prev_portid && prev_portid != tsk->portid) {
- spin_unlock_bh(&tsk->sk.sk_lock.slock);
+ rhashtable_walk_start(iter);
+ while ((tsk = rhashtable_walk_next(iter)) != NULL) {
+ if (IS_ERR(tsk)) {
+ err = PTR_ERR(tsk);
+ if (err == -EAGAIN) {
+ err = 0;
continue;
}
+ break;
+ }
- err = skb_handler(skb, cb, tsk);
- if (err) {
- prev_portid = tsk->portid;
- spin_unlock_bh(&tsk->sk.sk_lock.slock);
- goto out;
- }
-
- prev_portid = 0;
- spin_unlock_bh(&tsk->sk.sk_lock.slock);
+ sock_hold(&tsk->sk);
+ rhashtable_walk_stop(iter);
+ lock_sock(&tsk->sk);
+ err = skb_handler(skb, cb, tsk);
+ if (err) {
+ release_sock(&tsk->sk);
+ sock_put(&tsk->sk);
+ goto out;
}
+ release_sock(&tsk->sk);
+ rhashtable_walk_start(iter);
+ sock_put(&tsk->sk);
}
+ rhashtable_walk_stop(iter);
out:
- rcu_read_unlock();
- cb->args[0] = tbl_id;
- cb->args[1] = prev_portid;
-
return skb->len;
}
EXPORT_SYMBOL(tipc_nl_sk_walk);
+int tipc_dump_start(struct netlink_callback *cb)
+{
+ struct rhashtable_iter *iter = (void *)cb->args[0];
+ struct net *net = sock_net(cb->skb->sk);
+ struct tipc_net *tn = tipc_net(net);
+
+ if (!iter) {
+ iter = kmalloc(sizeof(*iter), GFP_KERNEL);
+ if (!iter)
+ return -ENOMEM;
+
+ cb->args[0] = (long)iter;
+ }
+
+ rhashtable_walk_enter(&tn->sk_rht, iter);
+ return 0;
+}
+EXPORT_SYMBOL(tipc_dump_start);
+
+int tipc_dump_done(struct netlink_callback *cb)
+{
+ struct rhashtable_iter *hti = (void *)cb->args[0];
+
+ rhashtable_walk_exit(hti);
+ kfree(hti);
+ return 0;
+}
+EXPORT_SYMBOL(tipc_dump_done);
+
int tipc_sk_fill_sock_diag(struct sk_buff *skb, struct netlink_callback *cb,
struct tipc_sock *tsk, u32 sk_filter_state,
u64 (*tipc_diag_gen_cookie)(struct sock *sk))
diff --git a/net/tipc/socket.h b/net/tipc/socket.h
index aff9b2ae5a1f..d43032e26532 100644
--- a/net/tipc/socket.h
+++ b/net/tipc/socket.h
@@ -68,4 +68,6 @@ int tipc_nl_sk_walk(struct sk_buff *skb, struct netlink_callback *cb,
int (*skb_handler)(struct sk_buff *skb,
struct netlink_callback *cb,
struct tipc_sock *tsk));
+int tipc_dump_start(struct netlink_callback *cb);
+int tipc_dump_done(struct netlink_callback *cb);
#endif
--
2.14.4
^ permalink raw reply related
* Re: [PATCH v2 17/17] net: WireGuard secure network tunnel
From: Andrew Lunn @ 2018-08-24 23:00 UTC (permalink / raw)
To: Jason A. Donenfeld; +Cc: linux-kernel, netdev, davem, Greg KH
In-Reply-To: <20180824213849.23647-18-Jason@zx2c4.com>
Hi Jason
It is normal to include after the --- what you have changed since the
previous version.
I ran checkpatch on this again.
Last time, we had:
total: 6 errors, 763 warnings, 6514 lines checked
This time, we have:
total: 8 errors, 196 warnings, 7470 lines checked
So much better with the warnings. Thanks.
The errors should be simple to fix:
ERROR: that open brace { should be on the previous line
ERROR: that open brace { should be on the previous line
ERROR: spaces required around that '=' (ctx:VxW)
ERROR: spaces required around that '=' (ctx:VxW)
ERROR: spaces required around that '=' (ctx:VxW)
ERROR: that open brace { should be on the previous line
ERROR: Use of __initconst requires a separate use of const
ERROR: trailing whitespace
One appears to be a false positive:
+static const struct {
+ bool result;
+ unsigned int msec_to_sleep_before;
+} expected_results[] __initconst = {
+ [0 ... PACKETS_BURSTABLE - 1] = { true, 0 },
+ [PACKETS_BURSTABLE] = { false, 0 },
+ [PACKETS_BURSTABLE + 1] = { true, MSEC_PER_SEC / PACKETS_PER_SECOND },
+ [PACKETS_BURSTABLE + 2] = { false, 0 },
+ [PACKETS_BURSTABLE + 3] = { true, (MSEC_PER_SEC / PACKETS_PER_SECOND) * 2 },
+ [PACKETS_BURSTABLE + 4] = { true, 0 },
+ [PACKETS_BURSTABLE + 5] = { false, 0 }
+};
Looking at some of the warnings:
WARNING: Avoid crashing the kernel - try using WARN_ON & recovery code rather than BUG() or BUG_ON()
#428: FILE: drivers/net/wireguard/allowedips.c:112:
+ BUG_ON(len >= 128);
WARNING: Macros with flow control statements should be avoided
#5905: FILE: drivers/net/wireguard/selftest/ratelimiter.h:87:
+#define ensure_time do { \
+ if (time_is_before_jiffies(loop_start_time + \
+ maximum_jiffies_at_index(i))) { \
+ if (++tries >= 5000) \
+ goto err; \
+ gc_entries(NULL); \
+ rcu_barrier(); \
+ msleep(500); \
+ goto restart; \
+ } \
+ } while (0)
WARNING: Macros with flow control statements should be avoided
#6948: FILE: drivers/net/wireguard/timers.c:29:
+#define peer_get_from_timer(timer_name) \
+ struct wireguard_peer *peer; \
+ rcu_read_lock_bh(); \
+ peer = peer_get_maybe_zero(from_timer(peer, timer, timer_name)); \
+ rcu_read_unlock_bh(); \
+ if (unlikely(!peer)) \
+ return;
are not very nice.
WARNING: Possible unnecessary 'out of memory' message
#5306: FILE: drivers/net/wireguard/selftest/allowedips.h:261:
+ if (!peers) {
+ pr_info("allowedips random self-test: out of memory\n");
kcalloc and friends are pretty noisy when they fails. No need to add
your own print.
I see we still have a lot of __always_inline in C files :-(
Andrew
^ permalink raw reply
* [net 11/11] i40e: fix condition of WARN_ONCE for stat strings
From: Jeff Kirsher @ 2018-08-24 18:47 UTC (permalink / raw)
To: davem; +Cc: Jacob Keller, netdev, nhorman, sassmann, jogreene, Jeff Kirsher
In-Reply-To: <20180824184735.32175-1-jeffrey.t.kirsher@intel.com>
From: Jacob Keller <jacob.e.keller@intel.com>
Commit 9b10df596bd4 ("i40e: use WARN_ONCE to replace the commented
BUG_ON size check") introduced a warning check to make sure
that the size of the stat strings was always the expected value. This
code accidentally inverted the check of the data pointer. Fix this so
that we accurately count the size of the stats we copied in.
This fixes an erroneous WARN kernel splat that occurs when requesting
ethtool statistics.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Tested-by: Mauro S M Rodrigues <maurosr@linux.vnet.ibm.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ethernet/intel/i40e/i40e_ethtool.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index abcd096ede14..5ff6caa83948 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -2013,7 +2013,7 @@ static void i40e_get_stat_strings(struct net_device *netdev, u8 *data)
for (i = 0; i < I40E_MAX_USER_PRIORITY; i++)
i40e_add_stat_strings(&data, i40e_gstrings_pfc_stats, i);
- WARN_ONCE(p - data != i40e_get_stats_count(netdev) * ETH_GSTRING_LEN,
+ WARN_ONCE(data - p != i40e_get_stats_count(netdev) * ETH_GSTRING_LEN,
"stat strings count mismatch!");
}
--
2.17.1
^ permalink raw reply related
* [net 06/11] igb: Replace mdelay() with msleep() in igb_integrated_phy_loopback()
From: Jeff Kirsher @ 2018-08-24 18:47 UTC (permalink / raw)
To: davem; +Cc: Jia-Ju Bai, netdev, nhorman, sassmann, jogreene, Jeff Kirsher
In-Reply-To: <20180824184735.32175-1-jeffrey.t.kirsher@intel.com>
From: Jia-Ju Bai <baijiaju1990@gmail.com>
igb_integrated_phy_loopback() is never called in atomic context.
It calls mdelay() to busily wait, which is not necessary.
mdelay() can be replaced with msleep().
This is found by a static analysis tool named DCNS written by myself.
Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ethernet/intel/igb/igb_ethtool.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c b/drivers/net/ethernet/intel/igb/igb_ethtool.c
index f92f7918112d..5acf3b743876 100644
--- a/drivers/net/ethernet/intel/igb/igb_ethtool.c
+++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c
@@ -1649,7 +1649,7 @@ static int igb_integrated_phy_loopback(struct igb_adapter *adapter)
if (hw->phy.type == e1000_phy_m88)
igb_phy_disable_receiver(adapter);
- mdelay(500);
+ msleep(500);
return 0;
}
--
2.17.1
^ permalink raw reply related
* [net 00/11][pull request] Intel Wired LAN Driver Updates 2018-08-24
From: Jeff Kirsher @ 2018-08-24 18:47 UTC (permalink / raw)
To: davem; +Cc: Jeff Kirsher, netdev, nhorman, sassmann, jogreene
This series contains fixes to e1000, igb, ixgb, ixgbe and i40e.
YueHaibing from Huawei provides a change to use dma_zalloc_coherent()
instead of calls to allocator followed by a memset for ixgb.
Bo Chen provides a couple of fixes for e1000, first by adding a check to
prevent a NULL pointer dereference. The second change is to clean up a
possible resource leak on old transmit and receive rings when the device
is not up.
Jesus fixes an issue in the usage of an advanced transmit context
descriptor for retrieving the timestamp of a packet for AF_PACKET if the
IGB_TX_FLAGS_VLAN is not set in igb.
Jia-Ju Bai provides several patches which replace GFP_ATOMIC with
GFP_KERNEL, when using kzalloc() and kcalloc() which is not necessary.
Also found an instance of mdelay() call which could be replaced with
msleep().
Tony fixes ixgbe to allow MTU changes with XDP, by adding checks to
ensure only supported values and return -EINVAL for when it is not
supported.
Sebastian fixed an issue that was not clearing VF mailbox memory and
ensure queues are re-enabled correctly.
Martyna fixes a transmit timeout when DCB is configured when bringing up
an interface.
Jake fixes a previous commit which accidentally reversed the check of
the data pointer, so we can accurately count the size of the stats.
The following are changes since commit ff0fadfffe681203bfe134e1041ab6ccb4aa3dff:
Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf
and are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-queue 1GbE
Bo Chen (2):
e1000: check on netif_running() before calling e1000_up()
e1000: ensure to free old tx/rx rings in set_ringparam()
Jacob Keller (1):
i40e: fix condition of WARN_ONCE for stat strings
Jesus Sanchez-Palencia (1):
igb: Use an advanced ctx descriptor for launchtime
Jia-Ju Bai (3):
igb: Replace GFP_ATOMIC with GFP_KERNEL in igb_sw_init()
igb: Replace mdelay() with msleep() in igb_integrated_phy_loopback()
ixgbe: Replace GFP_ATOMIC with GFP_KERNEL
Martyna Szapar (1):
i40e: Fix for Tx timeouts when interface is brought up if DCB is
enabled
Sebastian Basierski (1):
ixgbe: fix driver behaviour after issuing VFLR
Tony Nguyen (1):
ixgbe: Prevent unsupported configurations with XDP
YueHaibing (1):
ixgb: use dma_zalloc_coherent instead of allocator/memset
.../net/ethernet/intel/e1000/e1000_ethtool.c | 7 +++--
.../net/ethernet/intel/i40e/i40e_ethtool.c | 2 +-
drivers/net/ethernet/intel/i40e/i40e_main.c | 15 ++++-----
drivers/net/ethernet/intel/igb/igb_ethtool.c | 2 +-
drivers/net/ethernet/intel/igb/igb_main.c | 7 +++--
drivers/net/ethernet/intel/ixgb/ixgb_main.c | 5 ++-
drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c | 4 +--
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 30 ++++++++++++++++--
.../net/ethernet/intel/ixgbe/ixgbe_sriov.c | 31 +++++++++++++++++++
drivers/net/ethernet/intel/ixgbe/ixgbe_type.h | 1 +
10 files changed, 81 insertions(+), 23 deletions(-)
--
2.17.1
^ permalink raw reply
* [net 10/11] i40e: Fix for Tx timeouts when interface is brought up if DCB is enabled
From: Jeff Kirsher @ 2018-08-24 18:47 UTC (permalink / raw)
To: davem; +Cc: Martyna Szapar, netdev, nhorman, sassmann, jogreene, Jeff Kirsher
In-Reply-To: <20180824184735.32175-1-jeffrey.t.kirsher@intel.com>
From: Martyna Szapar <martyna.szapar@intel.com>
If interface is connected to switch port configured for DCB there are
TX timeouts when bringing up interface. Problem started appearing after
adding in i40e driver code mqprio hardware offload mode. In function
i40e_vsi_configure_bw_alloc was added resetting BW rate which should
be executing when mqprio qdisc is removed but was also when there was
no mqprio qdisc added and DCB was enabled. In this patch was added
additional check for DCB flag so now when DCB is enabled the correct
DCB configs from before mqprio patch are restored.
Signed-off-by: Martyna Szapar <martyna.szapar@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ethernet/intel/i40e/i40e_main.c | 15 ++++++++-------
1 file changed, 8 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index f2c622e78802..ac685ad4d877 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -5122,15 +5122,17 @@ static int i40e_vsi_configure_bw_alloc(struct i40e_vsi *vsi, u8 enabled_tc,
u8 *bw_share)
{
struct i40e_aqc_configure_vsi_tc_bw_data bw_data;
+ struct i40e_pf *pf = vsi->back;
i40e_status ret;
int i;
- if (vsi->back->flags & I40E_FLAG_TC_MQPRIO)
+ /* There is no need to reset BW when mqprio mode is on. */
+ if (pf->flags & I40E_FLAG_TC_MQPRIO)
return 0;
- if (!vsi->mqprio_qopt.qopt.hw) {
+ if (!vsi->mqprio_qopt.qopt.hw && !(pf->flags & I40E_FLAG_DCB_ENABLED)) {
ret = i40e_set_bw_limit(vsi, vsi->seid, 0);
if (ret)
- dev_info(&vsi->back->pdev->dev,
+ dev_info(&pf->pdev->dev,
"Failed to reset tx rate for vsi->seid %u\n",
vsi->seid);
return ret;
@@ -5139,12 +5141,11 @@ static int i40e_vsi_configure_bw_alloc(struct i40e_vsi *vsi, u8 enabled_tc,
for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++)
bw_data.tc_bw_credits[i] = bw_share[i];
- ret = i40e_aq_config_vsi_tc_bw(&vsi->back->hw, vsi->seid, &bw_data,
- NULL);
+ ret = i40e_aq_config_vsi_tc_bw(&pf->hw, vsi->seid, &bw_data, NULL);
if (ret) {
- dev_info(&vsi->back->pdev->dev,
+ dev_info(&pf->pdev->dev,
"AQ command Config VSI BW allocation per TC failed = %d\n",
- vsi->back->hw.aq.asq_last_status);
+ pf->hw.aq.asq_last_status);
return -EINVAL;
}
--
2.17.1
^ permalink raw reply related
* [net 07/11] ixgbe: Replace GFP_ATOMIC with GFP_KERNEL
From: Jeff Kirsher @ 2018-08-24 18:47 UTC (permalink / raw)
To: davem; +Cc: Jia-Ju Bai, netdev, nhorman, sassmann, jogreene, Jeff Kirsher
In-Reply-To: <20180824184735.32175-1-jeffrey.t.kirsher@intel.com>
From: Jia-Ju Bai <baijiaju1990@gmail.com>
ixgbe_fcoe_ddp_setup(), ixgbe_setup_fcoe_ddp_resources() and
ixgbe_sw_init() are never called in atomic context.
They call kmalloc(), dma_pool_alloc() and kzalloc() with GFP_ATOMIC,
which is not necessary.
GFP_ATOMIC can be replaced with GFP_KERNEL.
This is found by a static analysis tool named DCNS written by myself.
Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
Acked-by: Sebastian Basierski <sebastianx.basierski@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c | 4 ++--
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 2 +-
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c
index 94b3165ff543..ccd852ad62a4 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_fcoe.c
@@ -192,7 +192,7 @@ static int ixgbe_fcoe_ddp_setup(struct net_device *netdev, u16 xid,
}
/* alloc the udl from per cpu ddp pool */
- ddp->udl = dma_pool_alloc(ddp_pool->pool, GFP_ATOMIC, &ddp->udp);
+ ddp->udl = dma_pool_alloc(ddp_pool->pool, GFP_KERNEL, &ddp->udp);
if (!ddp->udl) {
e_err(drv, "failed allocated ddp context\n");
goto out_noddp_unmap;
@@ -760,7 +760,7 @@ int ixgbe_setup_fcoe_ddp_resources(struct ixgbe_adapter *adapter)
return 0;
/* Extra buffer to be shared by all DDPs for HW work around */
- buffer = kmalloc(IXGBE_FCBUFF_MIN, GFP_ATOMIC);
+ buffer = kmalloc(IXGBE_FCBUFF_MIN, GFP_KERNEL);
if (!buffer)
return -ENOMEM;
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index af4c9ae7f432..663d59ba527a 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -6201,7 +6201,7 @@ static int ixgbe_sw_init(struct ixgbe_adapter *adapter,
adapter->mac_table = kcalloc(hw->mac.num_rar_entries,
sizeof(struct ixgbe_mac_addr),
- GFP_ATOMIC);
+ GFP_KERNEL);
if (!adapter->mac_table)
return -ENOMEM;
--
2.17.1
^ permalink raw reply related
* [net 09/11] ixgbe: fix driver behaviour after issuing VFLR
From: Jeff Kirsher @ 2018-08-24 18:47 UTC (permalink / raw)
To: davem
Cc: Sebastian Basierski, netdev, nhorman, sassmann, jogreene,
Jeff Kirsher
In-Reply-To: <20180824184735.32175-1-jeffrey.t.kirsher@intel.com>
From: Sebastian Basierski <sebastianx.basierski@intel.com>
Since VFLR doesn't clear VFMBMEM (VF Mailbox Memory)
and is not re-enabling queues correctly we should fix
this behavior.
Signed-off-by: Sebastian Basierski <sebastianx.basierski@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
.../net/ethernet/intel/ixgbe/ixgbe_sriov.c | 26 +++++++++++++++++++
drivers/net/ethernet/intel/ixgbe/ixgbe_type.h | 1 +
2 files changed, 27 insertions(+)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
index 9264a5f8a5d0..3c6f01c41b78 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
@@ -693,8 +693,13 @@ static int ixgbe_set_vf_macvlan(struct ixgbe_adapter *adapter,
static inline void ixgbe_vf_reset_event(struct ixgbe_adapter *adapter, u32 vf)
{
struct ixgbe_hw *hw = &adapter->hw;
+ struct ixgbe_ring_feature *vmdq = &adapter->ring_feature[RING_F_VMDQ];
struct vf_data_storage *vfinfo = &adapter->vfinfo[vf];
+ u32 q_per_pool = __ALIGN_MASK(1, ~vmdq->mask);
u8 num_tcs = adapter->hw_tcs;
+ u32 reg_val;
+ u32 queue;
+ u32 word;
/* remove VLAN filters beloning to this VF */
ixgbe_clear_vf_vlans(adapter, vf);
@@ -731,6 +736,27 @@ static inline void ixgbe_vf_reset_event(struct ixgbe_adapter *adapter, u32 vf)
/* reset VF api back to unknown */
adapter->vfinfo[vf].vf_api = ixgbe_mbox_api_10;
+
+ /* Restart each queue for given VF */
+ for (queue = 0; queue < q_per_pool; queue++) {
+ unsigned int reg_idx = (vf * q_per_pool) + queue;
+
+ reg_val = IXGBE_READ_REG(hw, IXGBE_PVFTXDCTL(reg_idx));
+
+ /* Re-enabling only configured queues */
+ if (reg_val) {
+ reg_val |= IXGBE_TXDCTL_ENABLE;
+ IXGBE_WRITE_REG(hw, IXGBE_PVFTXDCTL(reg_idx), reg_val);
+ reg_val &= ~IXGBE_TXDCTL_ENABLE;
+ IXGBE_WRITE_REG(hw, IXGBE_PVFTXDCTL(reg_idx), reg_val);
+ }
+ }
+
+ /* Clear VF's mailbox memory */
+ for (word = 0; word < IXGBE_VFMAILBOX_SIZE; word++)
+ IXGBE_WRITE_REG_ARRAY(hw, IXGBE_PFMBMEM(vf), word, 0);
+
+ IXGBE_WRITE_FLUSH(hw);
}
static int ixgbe_set_vf_mac(struct ixgbe_adapter *adapter,
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
index 44cfb2021145..41bcbb337e83 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_type.h
@@ -2518,6 +2518,7 @@ enum {
/* Translated register #defines */
#define IXGBE_PVFTDH(P) (0x06010 + (0x40 * (P)))
#define IXGBE_PVFTDT(P) (0x06018 + (0x40 * (P)))
+#define IXGBE_PVFTXDCTL(P) (0x06028 + (0x40 * (P)))
#define IXGBE_PVFTDWBAL(P) (0x06038 + (0x40 * (P)))
#define IXGBE_PVFTDWBAH(P) (0x0603C + (0x40 * (P)))
--
2.17.1
^ permalink raw reply related
* [net 08/11] ixgbe: Prevent unsupported configurations with XDP
From: Jeff Kirsher @ 2018-08-24 18:47 UTC (permalink / raw)
To: davem
Cc: Tony Nguyen, netdev, nhorman, sassmann, jogreene, Jakub Kicinski,
Jeff Kirsher
In-Reply-To: <20180824184735.32175-1-jeffrey.t.kirsher@intel.com>
From: Tony Nguyen <anthony.l.nguyen@intel.com>
These changes address comments by Jakub Kicinski on
commit 38b7e7f8ae82 ("ixgbe: Do not allow LRO or MTU change with XDP").
Change the MTU check with XDP to allow any supported value and only
reject those outside of the range as opposed to rejecting any change
when XDP is active. In situations where MTU size is not supported,
return -EINVAL instead of -EPERM.
Add checks when enabling SRIOV, DCB, or adding L2FW offloaded device
as they are not supported with XDP.
CC: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 28 +++++++++++++++++--
.../net/ethernet/intel/ixgbe/ixgbe_sriov.c | 5 ++++
2 files changed, 31 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 663d59ba527a..9a23d33a47ed 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -6620,8 +6620,18 @@ static int ixgbe_change_mtu(struct net_device *netdev, int new_mtu)
struct ixgbe_adapter *adapter = netdev_priv(netdev);
if (adapter->xdp_prog) {
- e_warn(probe, "MTU cannot be changed while XDP program is loaded\n");
- return -EPERM;
+ int new_frame_size = new_mtu + ETH_HLEN + ETH_FCS_LEN +
+ VLAN_HLEN;
+ int i;
+
+ for (i = 0; i < adapter->num_rx_queues; i++) {
+ struct ixgbe_ring *ring = adapter->rx_ring[i];
+
+ if (new_frame_size > ixgbe_rx_bufsz(ring)) {
+ e_warn(probe, "Requested MTU size is not supported with XDP\n");
+ return -EINVAL;
+ }
+ }
}
/*
@@ -8983,6 +8993,15 @@ int ixgbe_setup_tc(struct net_device *dev, u8 tc)
#ifdef CONFIG_IXGBE_DCB
if (tc) {
+ if (adapter->xdp_prog) {
+ e_warn(probe, "DCB is not supported with XDP\n");
+
+ ixgbe_init_interrupt_scheme(adapter);
+ if (netif_running(dev))
+ ixgbe_open(dev);
+ return -EINVAL;
+ }
+
netdev_set_num_tc(dev, tc);
ixgbe_set_prio_tc_map(adapter);
@@ -9934,6 +9953,11 @@ static void *ixgbe_fwd_add(struct net_device *pdev, struct net_device *vdev)
int tcs = adapter->hw_tcs ? : 1;
int pool, err;
+ if (adapter->xdp_prog) {
+ e_warn(probe, "L2FW offload is not supported with XDP\n");
+ return ERR_PTR(-EINVAL);
+ }
+
/* The hardware supported by ixgbe only filters on the destination MAC
* address. In order to avoid issues we only support offloading modes
* where the hardware can actually provide the functionality.
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
index 6f59933cdff7..9264a5f8a5d0 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c
@@ -53,6 +53,11 @@ static int __ixgbe_enable_sriov(struct ixgbe_adapter *adapter,
struct ixgbe_hw *hw = &adapter->hw;
int i;
+ if (adapter->xdp_prog) {
+ e_warn(probe, "SRIOV is not supported with XDP\n");
+ return -EINVAL;
+ }
+
/* Enable VMDq flag so device will be set in VM mode */
adapter->flags |= IXGBE_FLAG_SRIOV_ENABLED |
IXGBE_FLAG_VMDQ_ENABLED;
--
2.17.1
^ permalink raw reply related
* [net 03/11] e1000: ensure to free old tx/rx rings in set_ringparam()
From: Jeff Kirsher @ 2018-08-24 18:47 UTC (permalink / raw)
To: davem; +Cc: Bo Chen, netdev, nhorman, sassmann, jogreene, Jeff Kirsher
In-Reply-To: <20180824184735.32175-1-jeffrey.t.kirsher@intel.com>
From: Bo Chen <chenbo@pdx.edu>
In 'e1000_set_ringparam()', the tx_ring and rx_ring are updated with new value
and the old tx/rx rings are freed only when the device is up. There are resource
leaks on old tx/rx rings when the device is not up. This bug is reported by COD,
a tool for testing kernel module binaries I am building.
This patch fixes the bug by always calling 'kfree()' on old tx/rx rings in
'e1000_set_ringparam()'.
Signed-off-by: Bo Chen <chenbo@pdx.edu>
Reviewed-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ethernet/intel/e1000/e1000_ethtool.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
index c1e4e94f100f..2569a168334c 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
@@ -624,14 +624,14 @@ static int e1000_set_ringparam(struct net_device *netdev,
adapter->tx_ring = tx_old;
e1000_free_all_rx_resources(adapter);
e1000_free_all_tx_resources(adapter);
- kfree(tx_old);
- kfree(rx_old);
adapter->rx_ring = rxdr;
adapter->tx_ring = txdr;
err = e1000_up(adapter);
if (err)
goto err_setup;
}
+ kfree(tx_old);
+ kfree(rx_old);
clear_bit(__E1000_RESETTING, &adapter->flags);
return 0;
--
2.17.1
^ permalink raw reply related
* [net 04/11] igb: Use an advanced ctx descriptor for launchtime
From: Jeff Kirsher @ 2018-08-24 18:47 UTC (permalink / raw)
To: davem
Cc: Jesus Sanchez-Palencia, netdev, nhorman, sassmann, jogreene,
Jeff Kirsher
In-Reply-To: <20180824184735.32175-1-jeffrey.t.kirsher@intel.com>
From: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com>
On i210, Launchtime (TxTime) requires the usage of an "Advanced
Transmit Context Descriptor" for retrieving the timestamp of a packet.
The igb driver correctly builds such descriptor on the segmentation
flow (i.e. igb_tso()) or on the checksum one (i.e. igb_tx_csum()), but the
feature is broken for AF_PACKET if the IGB_TX_FLAGS_VLAN is not set,
which happens due to an early return on igb_tx_csum().
This flag is only set by the kernel when a VLAN interface is used,
thus we can't just rely on it. Here we are fixing this issue by checking
if launchtime is enabled for the current tx_ring before performing the
early return.
Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ethernet/intel/igb/igb_main.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index d03c2f0d7592..74416f8a9446 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -5816,7 +5816,8 @@ static void igb_tx_csum(struct igb_ring *tx_ring, struct igb_tx_buffer *first)
if (skb->ip_summed != CHECKSUM_PARTIAL) {
csum_failed:
- if (!(first->tx_flags & IGB_TX_FLAGS_VLAN))
+ if (!(first->tx_flags & IGB_TX_FLAGS_VLAN) &&
+ !tx_ring->launchtime_enable)
return;
goto no_csum;
}
--
2.17.1
^ permalink raw reply related
* [net 02/11] e1000: check on netif_running() before calling e1000_up()
From: Jeff Kirsher @ 2018-08-24 18:47 UTC (permalink / raw)
To: davem; +Cc: Bo Chen, netdev, nhorman, sassmann, jogreene, Jeff Kirsher
In-Reply-To: <20180824184735.32175-1-jeffrey.t.kirsher@intel.com>
From: Bo Chen <chenbo@pdx.edu>
When the device is not up, the call to 'e1000_up()' from the error handling path
of 'e1000_set_ringparam()' causes a kernel oops with a null-pointer
dereference. The null-pointer dereference is triggered in function
'e1000_alloc_rx_buffers()' at line 'buffer_info = &rx_ring->buffer_info[i]'.
This bug was reported by COD, a tool for testing kernel module binaries I am
building. This bug was also detected by KFI from Dr. Kai Cong.
This patch fixes the bug by checking on 'netif_running()' before calling
'e1000_up()' in 'e1000_set_ringparam()'.
Signed-off-by: Bo Chen <chenbo@pdx.edu>
Acked-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ethernet/intel/e1000/e1000_ethtool.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
index bdb3f8e65ed4..c1e4e94f100f 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
@@ -644,7 +644,8 @@ static int e1000_set_ringparam(struct net_device *netdev,
err_alloc_rx:
kfree(txdr);
err_alloc_tx:
- e1000_up(adapter);
+ if (netif_running(adapter->netdev))
+ e1000_up(adapter);
err_setup:
clear_bit(__E1000_RESETTING, &adapter->flags);
return err;
--
2.17.1
^ permalink raw reply related
* [net 05/11] igb: Replace GFP_ATOMIC with GFP_KERNEL in igb_sw_init()
From: Jeff Kirsher @ 2018-08-24 18:47 UTC (permalink / raw)
To: davem; +Cc: Jia-Ju Bai, netdev, nhorman, sassmann, jogreene, Jeff Kirsher
In-Reply-To: <20180824184735.32175-1-jeffrey.t.kirsher@intel.com>
From: Jia-Ju Bai <baijiaju1990@gmail.com>
igb_sw_init() is never called in atomic context.
It calls kzalloc() and kcalloc() with GFP_ATOMIC, which is not necessary.
GFP_ATOMIC can be replaced with GFP_KERNEL.
This is found by a static analysis tool named DCNS written by myself.
Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ethernet/intel/igb/igb_main.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 74416f8a9446..a32c576c1e65 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -3873,7 +3873,7 @@ static int igb_sw_init(struct igb_adapter *adapter)
adapter->mac_table = kcalloc(hw->mac.rar_entry_count,
sizeof(struct igb_mac_addr),
- GFP_ATOMIC);
+ GFP_KERNEL);
if (!adapter->mac_table)
return -ENOMEM;
@@ -3883,7 +3883,7 @@ static int igb_sw_init(struct igb_adapter *adapter)
/* Setup and initialize a copy of the hw vlan table array */
adapter->shadow_vfta = kcalloc(E1000_VLAN_FILTER_TBL_SIZE, sizeof(u32),
- GFP_ATOMIC);
+ GFP_KERNEL);
if (!adapter->shadow_vfta)
return -ENOMEM;
--
2.17.1
^ permalink raw reply related
* [net 01/11] ixgb: use dma_zalloc_coherent instead of allocator/memset
From: Jeff Kirsher @ 2018-08-24 18:47 UTC (permalink / raw)
To: davem; +Cc: YueHaibing, netdev, nhorman, sassmann, jogreene, Jeff Kirsher
In-Reply-To: <20180824184735.32175-1-jeffrey.t.kirsher@intel.com>
From: YueHaibing <yuehaibing@huawei.com>
Use dma_zalloc_coherent instead of dma_alloc_coherent
followed by memset 0.
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
drivers/net/ethernet/intel/ixgb/ixgb_main.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_main.c b/drivers/net/ethernet/intel/ixgb/ixgb_main.c
index 43664adf7a3c..d3e72d0f66ef 100644
--- a/drivers/net/ethernet/intel/ixgb/ixgb_main.c
+++ b/drivers/net/ethernet/intel/ixgb/ixgb_main.c
@@ -771,14 +771,13 @@ ixgb_setup_rx_resources(struct ixgb_adapter *adapter)
rxdr->size = rxdr->count * sizeof(struct ixgb_rx_desc);
rxdr->size = ALIGN(rxdr->size, 4096);
- rxdr->desc = dma_alloc_coherent(&pdev->dev, rxdr->size, &rxdr->dma,
- GFP_KERNEL);
+ rxdr->desc = dma_zalloc_coherent(&pdev->dev, rxdr->size, &rxdr->dma,
+ GFP_KERNEL);
if (!rxdr->desc) {
vfree(rxdr->buffer_info);
return -ENOMEM;
}
- memset(rxdr->desc, 0, rxdr->size);
rxdr->next_to_clean = 0;
rxdr->next_to_use = 0;
--
2.17.1
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox