--- /dev/null
+From b7b73cd5d74694ed59abcdb4974dacb4ff8b2a2a Mon Sep 17 00:00:00 2001
+From: Eric Biggers <ebiggers@google.com>
+Date: Sat, 26 May 2018 00:08:58 -0700
+Subject: crypto: x86/salsa20 - remove x86 salsa20 implementations
+
+From: Eric Biggers <ebiggers@google.com>
+
+commit b7b73cd5d74694ed59abcdb4974dacb4ff8b2a2a upstream.
+
+The x86 assembly implementations of Salsa20 use the frame base pointer
+register (%ebp or %rbp), which breaks frame pointer convention and
+breaks stack traces when unwinding from an interrupt in the crypto code.
+Recent (v4.10+) kernels will warn about this, e.g.
+
+WARNING: kernel stack regs at 00000000a8291e69 in syzkaller047086:4677 has bad 'bp' value 000000001077994c
+[...]
+
+But after looking into it, I believe there's very little reason to still
+retain the x86 Salsa20 code. First, these are *not* vectorized
+(SSE2/SSSE3/AVX2) implementations, which would be needed to get anywhere
+close to the best Salsa20 performance on any remotely modern x86
+processor; they're just regular x86 assembly. Second, it's still
+unclear that anyone is actually using the kernel's Salsa20 at all,
+especially given that now ChaCha20 is supported too, and with much more
+efficient SSSE3 and AVX2 implementations. Finally, in benchmarks I did
+on both Intel and AMD processors with both gcc 8.1.0 and gcc 4.9.4, the
+x86_64 salsa20-asm is actually slightly *slower* than salsa20-generic
+(~3% slower on Skylake, ~10% slower on Zen), while the i686 salsa20-asm
+is only slightly faster than salsa20-generic (~15% faster on Skylake,
+~20% faster on Zen). The gcc version made little difference.
+
+So, the x86_64 salsa20-asm is pretty clearly useless. That leaves just
+the i686 salsa20-asm, which based on my tests provides a 15-20% speed
+boost. But that's without updating the code to not use %ebp. And given
+the maintenance cost, the small speed difference vs. salsa20-generic,
+the fact that few people still use i686 kernels, the doubt that anyone
+is even using the kernel's Salsa20 at all, and the fact that a SSE2
+implementation would almost certainly be much faster on any remotely
+modern x86 processor yet no one has cared enough to add one yet, I don't
+think it's worthwhile to keep.
+
+Thus, just remove both the x86_64 and i686 salsa20-asm implementations.
+
+Reported-by: syzbot+ffa3a158337bbc01ff09@syzkaller.appspotmail.com
+Signed-off-by: Eric Biggers <ebiggers@google.com>
+Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ arch/x86/crypto/Makefile | 4
+ arch/x86/crypto/salsa20-i586-asm_32.S | 1114 --------------------------------
+ arch/x86/crypto/salsa20-x86_64-asm_64.S | 919 --------------------------
+ arch/x86/crypto/salsa20_glue.c | 116 ---
+ crypto/Kconfig | 26
+ 5 files changed, 2179 deletions(-)
+
+--- a/arch/x86/crypto/Makefile
++++ b/arch/x86/crypto/Makefile
+@@ -15,7 +15,6 @@ obj-$(CONFIG_CRYPTO_GLUE_HELPER_X86) +=
+
+ obj-$(CONFIG_CRYPTO_AES_586) += aes-i586.o
+ obj-$(CONFIG_CRYPTO_TWOFISH_586) += twofish-i586.o
+-obj-$(CONFIG_CRYPTO_SALSA20_586) += salsa20-i586.o
+ obj-$(CONFIG_CRYPTO_SERPENT_SSE2_586) += serpent-sse2-i586.o
+
+ obj-$(CONFIG_CRYPTO_AES_X86_64) += aes-x86_64.o
+@@ -24,7 +23,6 @@ obj-$(CONFIG_CRYPTO_CAMELLIA_X86_64) +=
+ obj-$(CONFIG_CRYPTO_BLOWFISH_X86_64) += blowfish-x86_64.o
+ obj-$(CONFIG_CRYPTO_TWOFISH_X86_64) += twofish-x86_64.o
+ obj-$(CONFIG_CRYPTO_TWOFISH_X86_64_3WAY) += twofish-x86_64-3way.o
+-obj-$(CONFIG_CRYPTO_SALSA20_X86_64) += salsa20-x86_64.o
+ obj-$(CONFIG_CRYPTO_CHACHA20_X86_64) += chacha20-x86_64.o
+ obj-$(CONFIG_CRYPTO_SERPENT_SSE2_X86_64) += serpent-sse2-x86_64.o
+ obj-$(CONFIG_CRYPTO_AES_NI_INTEL) += aesni-intel.o
+@@ -59,7 +57,6 @@ endif
+
+ aes-i586-y := aes-i586-asm_32.o aes_glue.o
+ twofish-i586-y := twofish-i586-asm_32.o twofish_glue.o
+-salsa20-i586-y := salsa20-i586-asm_32.o salsa20_glue.o
+ serpent-sse2-i586-y := serpent-sse2-i586-asm_32.o serpent_sse2_glue.o
+
+ aes-x86_64-y := aes-x86_64-asm_64.o aes_glue.o
+@@ -68,7 +65,6 @@ camellia-x86_64-y := camellia-x86_64-asm
+ blowfish-x86_64-y := blowfish-x86_64-asm_64.o blowfish_glue.o
+ twofish-x86_64-y := twofish-x86_64-asm_64.o twofish_glue.o
+ twofish-x86_64-3way-y := twofish-x86_64-asm_64-3way.o twofish_glue_3way.o
+-salsa20-x86_64-y := salsa20-x86_64-asm_64.o salsa20_glue.o
+ chacha20-x86_64-y := chacha20-ssse3-x86_64.o chacha20_glue.o
+ serpent-sse2-x86_64-y := serpent-sse2-x86_64-asm_64.o serpent_sse2_glue.o
+
+--- a/arch/x86/crypto/salsa20-i586-asm_32.S
++++ /dev/null
+@@ -1,1114 +0,0 @@
+-# salsa20_pm.s version 20051229
+-# D. J. Bernstein
+-# Public domain.
+-
+-#include <linux/linkage.h>
+-
+-.text
+-
+-# enter salsa20_encrypt_bytes
+-ENTRY(salsa20_encrypt_bytes)
+- mov %esp,%eax
+- and $31,%eax
+- add $256,%eax
+- sub %eax,%esp
+- # eax_stack = eax
+- movl %eax,80(%esp)
+- # ebx_stack = ebx
+- movl %ebx,84(%esp)
+- # esi_stack = esi
+- movl %esi,88(%esp)
+- # edi_stack = edi
+- movl %edi,92(%esp)
+- # ebp_stack = ebp
+- movl %ebp,96(%esp)
+- # x = arg1
+- movl 4(%esp,%eax),%edx
+- # m = arg2
+- movl 8(%esp,%eax),%esi
+- # out = arg3
+- movl 12(%esp,%eax),%edi
+- # bytes = arg4
+- movl 16(%esp,%eax),%ebx
+- # bytes -= 0
+- sub $0,%ebx
+- # goto done if unsigned<=
+- jbe ._done
+-._start:
+- # in0 = *(uint32 *) (x + 0)
+- movl 0(%edx),%eax
+- # in1 = *(uint32 *) (x + 4)
+- movl 4(%edx),%ecx
+- # in2 = *(uint32 *) (x + 8)
+- movl 8(%edx),%ebp
+- # j0 = in0
+- movl %eax,164(%esp)
+- # in3 = *(uint32 *) (x + 12)
+- movl 12(%edx),%eax
+- # j1 = in1
+- movl %ecx,168(%esp)
+- # in4 = *(uint32 *) (x + 16)
+- movl 16(%edx),%ecx
+- # j2 = in2
+- movl %ebp,172(%esp)
+- # in5 = *(uint32 *) (x + 20)
+- movl 20(%edx),%ebp
+- # j3 = in3
+- movl %eax,176(%esp)
+- # in6 = *(uint32 *) (x + 24)
+- movl 24(%edx),%eax
+- # j4 = in4
+- movl %ecx,180(%esp)
+- # in7 = *(uint32 *) (x + 28)
+- movl 28(%edx),%ecx
+- # j5 = in5
+- movl %ebp,184(%esp)
+- # in8 = *(uint32 *) (x + 32)
+- movl 32(%edx),%ebp
+- # j6 = in6
+- movl %eax,188(%esp)
+- # in9 = *(uint32 *) (x + 36)
+- movl 36(%edx),%eax
+- # j7 = in7
+- movl %ecx,192(%esp)
+- # in10 = *(uint32 *) (x + 40)
+- movl 40(%edx),%ecx
+- # j8 = in8
+- movl %ebp,196(%esp)
+- # in11 = *(uint32 *) (x + 44)
+- movl 44(%edx),%ebp
+- # j9 = in9
+- movl %eax,200(%esp)
+- # in12 = *(uint32 *) (x + 48)
+- movl 48(%edx),%eax
+- # j10 = in10
+- movl %ecx,204(%esp)
+- # in13 = *(uint32 *) (x + 52)
+- movl 52(%edx),%ecx
+- # j11 = in11
+- movl %ebp,208(%esp)
+- # in14 = *(uint32 *) (x + 56)
+- movl 56(%edx),%ebp
+- # j12 = in12
+- movl %eax,212(%esp)
+- # in15 = *(uint32 *) (x + 60)
+- movl 60(%edx),%eax
+- # j13 = in13
+- movl %ecx,216(%esp)
+- # j14 = in14
+- movl %ebp,220(%esp)
+- # j15 = in15
+- movl %eax,224(%esp)
+- # x_backup = x
+- movl %edx,64(%esp)
+-._bytesatleast1:
+- # bytes - 64
+- cmp $64,%ebx
+- # goto nocopy if unsigned>=
+- jae ._nocopy
+- # ctarget = out
+- movl %edi,228(%esp)
+- # out = &tmp
+- leal 0(%esp),%edi
+- # i = bytes
+- mov %ebx,%ecx
+- # while (i) { *out++ = *m++; --i }
+- rep movsb
+- # out = &tmp
+- leal 0(%esp),%edi
+- # m = &tmp
+- leal 0(%esp),%esi
+-._nocopy:
+- # out_backup = out
+- movl %edi,72(%esp)
+- # m_backup = m
+- movl %esi,68(%esp)
+- # bytes_backup = bytes
+- movl %ebx,76(%esp)
+- # in0 = j0
+- movl 164(%esp),%eax
+- # in1 = j1
+- movl 168(%esp),%ecx
+- # in2 = j2
+- movl 172(%esp),%edx
+- # in3 = j3
+- movl 176(%esp),%ebx
+- # x0 = in0
+- movl %eax,100(%esp)
+- # x1 = in1
+- movl %ecx,104(%esp)
+- # x2 = in2
+- movl %edx,108(%esp)
+- # x3 = in3
+- movl %ebx,112(%esp)
+- # in4 = j4
+- movl 180(%esp),%eax
+- # in5 = j5
+- movl 184(%esp),%ecx
+- # in6 = j6
+- movl 188(%esp),%edx
+- # in7 = j7
+- movl 192(%esp),%ebx
+- # x4 = in4
+- movl %eax,116(%esp)
+- # x5 = in5
+- movl %ecx,120(%esp)
+- # x6 = in6
+- movl %edx,124(%esp)
+- # x7 = in7
+- movl %ebx,128(%esp)
+- # in8 = j8
+- movl 196(%esp),%eax
+- # in9 = j9
+- movl 200(%esp),%ecx
+- # in10 = j10
+- movl 204(%esp),%edx
+- # in11 = j11
+- movl 208(%esp),%ebx
+- # x8 = in8
+- movl %eax,132(%esp)
+- # x9 = in9
+- movl %ecx,136(%esp)
+- # x10 = in10
+- movl %edx,140(%esp)
+- # x11 = in11
+- movl %ebx,144(%esp)
+- # in12 = j12
+- movl 212(%esp),%eax
+- # in13 = j13
+- movl 216(%esp),%ecx
+- # in14 = j14
+- movl 220(%esp),%edx
+- # in15 = j15
+- movl 224(%esp),%ebx
+- # x12 = in12
+- movl %eax,148(%esp)
+- # x13 = in13
+- movl %ecx,152(%esp)
+- # x14 = in14
+- movl %edx,156(%esp)
+- # x15 = in15
+- movl %ebx,160(%esp)
+- # i = 20
+- mov $20,%ebp
+- # p = x0
+- movl 100(%esp),%eax
+- # s = x5
+- movl 120(%esp),%ecx
+- # t = x10
+- movl 140(%esp),%edx
+- # w = x15
+- movl 160(%esp),%ebx
+-._mainloop:
+- # x0 = p
+- movl %eax,100(%esp)
+- # x10 = t
+- movl %edx,140(%esp)
+- # p += x12
+- addl 148(%esp),%eax
+- # x5 = s
+- movl %ecx,120(%esp)
+- # t += x6
+- addl 124(%esp),%edx
+- # x15 = w
+- movl %ebx,160(%esp)
+- # r = x1
+- movl 104(%esp),%esi
+- # r += s
+- add %ecx,%esi
+- # v = x11
+- movl 144(%esp),%edi
+- # v += w
+- add %ebx,%edi
+- # p <<<= 7
+- rol $7,%eax
+- # p ^= x4
+- xorl 116(%esp),%eax
+- # t <<<= 7
+- rol $7,%edx
+- # t ^= x14
+- xorl 156(%esp),%edx
+- # r <<<= 7
+- rol $7,%esi
+- # r ^= x9
+- xorl 136(%esp),%esi
+- # v <<<= 7
+- rol $7,%edi
+- # v ^= x3
+- xorl 112(%esp),%edi
+- # x4 = p
+- movl %eax,116(%esp)
+- # x14 = t
+- movl %edx,156(%esp)
+- # p += x0
+- addl 100(%esp),%eax
+- # x9 = r
+- movl %esi,136(%esp)
+- # t += x10
+- addl 140(%esp),%edx
+- # x3 = v
+- movl %edi,112(%esp)
+- # p <<<= 9
+- rol $9,%eax
+- # p ^= x8
+- xorl 132(%esp),%eax
+- # t <<<= 9
+- rol $9,%edx
+- # t ^= x2
+- xorl 108(%esp),%edx
+- # s += r
+- add %esi,%ecx
+- # s <<<= 9
+- rol $9,%ecx
+- # s ^= x13
+- xorl 152(%esp),%ecx
+- # w += v
+- add %edi,%ebx
+- # w <<<= 9
+- rol $9,%ebx
+- # w ^= x7
+- xorl 128(%esp),%ebx
+- # x8 = p
+- movl %eax,132(%esp)
+- # x2 = t
+- movl %edx,108(%esp)
+- # p += x4
+- addl 116(%esp),%eax
+- # x13 = s
+- movl %ecx,152(%esp)
+- # t += x14
+- addl 156(%esp),%edx
+- # x7 = w
+- movl %ebx,128(%esp)
+- # p <<<= 13
+- rol $13,%eax
+- # p ^= x12
+- xorl 148(%esp),%eax
+- # t <<<= 13
+- rol $13,%edx
+- # t ^= x6
+- xorl 124(%esp),%edx
+- # r += s
+- add %ecx,%esi
+- # r <<<= 13
+- rol $13,%esi
+- # r ^= x1
+- xorl 104(%esp),%esi
+- # v += w
+- add %ebx,%edi
+- # v <<<= 13
+- rol $13,%edi
+- # v ^= x11
+- xorl 144(%esp),%edi
+- # x12 = p
+- movl %eax,148(%esp)
+- # x6 = t
+- movl %edx,124(%esp)
+- # p += x8
+- addl 132(%esp),%eax
+- # x1 = r
+- movl %esi,104(%esp)
+- # t += x2
+- addl 108(%esp),%edx
+- # x11 = v
+- movl %edi,144(%esp)
+- # p <<<= 18
+- rol $18,%eax
+- # p ^= x0
+- xorl 100(%esp),%eax
+- # t <<<= 18
+- rol $18,%edx
+- # t ^= x10
+- xorl 140(%esp),%edx
+- # s += r
+- add %esi,%ecx
+- # s <<<= 18
+- rol $18,%ecx
+- # s ^= x5
+- xorl 120(%esp),%ecx
+- # w += v
+- add %edi,%ebx
+- # w <<<= 18
+- rol $18,%ebx
+- # w ^= x15
+- xorl 160(%esp),%ebx
+- # x0 = p
+- movl %eax,100(%esp)
+- # x10 = t
+- movl %edx,140(%esp)
+- # p += x3
+- addl 112(%esp),%eax
+- # p <<<= 7
+- rol $7,%eax
+- # x5 = s
+- movl %ecx,120(%esp)
+- # t += x9
+- addl 136(%esp),%edx
+- # x15 = w
+- movl %ebx,160(%esp)
+- # r = x4
+- movl 116(%esp),%esi
+- # r += s
+- add %ecx,%esi
+- # v = x14
+- movl 156(%esp),%edi
+- # v += w
+- add %ebx,%edi
+- # p ^= x1
+- xorl 104(%esp),%eax
+- # t <<<= 7
+- rol $7,%edx
+- # t ^= x11
+- xorl 144(%esp),%edx
+- # r <<<= 7
+- rol $7,%esi
+- # r ^= x6
+- xorl 124(%esp),%esi
+- # v <<<= 7
+- rol $7,%edi
+- # v ^= x12
+- xorl 148(%esp),%edi
+- # x1 = p
+- movl %eax,104(%esp)
+- # x11 = t
+- movl %edx,144(%esp)
+- # p += x0
+- addl 100(%esp),%eax
+- # x6 = r
+- movl %esi,124(%esp)
+- # t += x10
+- addl 140(%esp),%edx
+- # x12 = v
+- movl %edi,148(%esp)
+- # p <<<= 9
+- rol $9,%eax
+- # p ^= x2
+- xorl 108(%esp),%eax
+- # t <<<= 9
+- rol $9,%edx
+- # t ^= x8
+- xorl 132(%esp),%edx
+- # s += r
+- add %esi,%ecx
+- # s <<<= 9
+- rol $9,%ecx
+- # s ^= x7
+- xorl 128(%esp),%ecx
+- # w += v
+- add %edi,%ebx
+- # w <<<= 9
+- rol $9,%ebx
+- # w ^= x13
+- xorl 152(%esp),%ebx
+- # x2 = p
+- movl %eax,108(%esp)
+- # x8 = t
+- movl %edx,132(%esp)
+- # p += x1
+- addl 104(%esp),%eax
+- # x7 = s
+- movl %ecx,128(%esp)
+- # t += x11
+- addl 144(%esp),%edx
+- # x13 = w
+- movl %ebx,152(%esp)
+- # p <<<= 13
+- rol $13,%eax
+- # p ^= x3
+- xorl 112(%esp),%eax
+- # t <<<= 13
+- rol $13,%edx
+- # t ^= x9
+- xorl 136(%esp),%edx
+- # r += s
+- add %ecx,%esi
+- # r <<<= 13
+- rol $13,%esi
+- # r ^= x4
+- xorl 116(%esp),%esi
+- # v += w
+- add %ebx,%edi
+- # v <<<= 13
+- rol $13,%edi
+- # v ^= x14
+- xorl 156(%esp),%edi
+- # x3 = p
+- movl %eax,112(%esp)
+- # x9 = t
+- movl %edx,136(%esp)
+- # p += x2
+- addl 108(%esp),%eax
+- # x4 = r
+- movl %esi,116(%esp)
+- # t += x8
+- addl 132(%esp),%edx
+- # x14 = v
+- movl %edi,156(%esp)
+- # p <<<= 18
+- rol $18,%eax
+- # p ^= x0
+- xorl 100(%esp),%eax
+- # t <<<= 18
+- rol $18,%edx
+- # t ^= x10
+- xorl 140(%esp),%edx
+- # s += r
+- add %esi,%ecx
+- # s <<<= 18
+- rol $18,%ecx
+- # s ^= x5
+- xorl 120(%esp),%ecx
+- # w += v
+- add %edi,%ebx
+- # w <<<= 18
+- rol $18,%ebx
+- # w ^= x15
+- xorl 160(%esp),%ebx
+- # x0 = p
+- movl %eax,100(%esp)
+- # x10 = t
+- movl %edx,140(%esp)
+- # p += x12
+- addl 148(%esp),%eax
+- # x5 = s
+- movl %ecx,120(%esp)
+- # t += x6
+- addl 124(%esp),%edx
+- # x15 = w
+- movl %ebx,160(%esp)
+- # r = x1
+- movl 104(%esp),%esi
+- # r += s
+- add %ecx,%esi
+- # v = x11
+- movl 144(%esp),%edi
+- # v += w
+- add %ebx,%edi
+- # p <<<= 7
+- rol $7,%eax
+- # p ^= x4
+- xorl 116(%esp),%eax
+- # t <<<= 7
+- rol $7,%edx
+- # t ^= x14
+- xorl 156(%esp),%edx
+- # r <<<= 7
+- rol $7,%esi
+- # r ^= x9
+- xorl 136(%esp),%esi
+- # v <<<= 7
+- rol $7,%edi
+- # v ^= x3
+- xorl 112(%esp),%edi
+- # x4 = p
+- movl %eax,116(%esp)
+- # x14 = t
+- movl %edx,156(%esp)
+- # p += x0
+- addl 100(%esp),%eax
+- # x9 = r
+- movl %esi,136(%esp)
+- # t += x10
+- addl 140(%esp),%edx
+- # x3 = v
+- movl %edi,112(%esp)
+- # p <<<= 9
+- rol $9,%eax
+- # p ^= x8
+- xorl 132(%esp),%eax
+- # t <<<= 9
+- rol $9,%edx
+- # t ^= x2
+- xorl 108(%esp),%edx
+- # s += r
+- add %esi,%ecx
+- # s <<<= 9
+- rol $9,%ecx
+- # s ^= x13
+- xorl 152(%esp),%ecx
+- # w += v
+- add %edi,%ebx
+- # w <<<= 9
+- rol $9,%ebx
+- # w ^= x7
+- xorl 128(%esp),%ebx
+- # x8 = p
+- movl %eax,132(%esp)
+- # x2 = t
+- movl %edx,108(%esp)
+- # p += x4
+- addl 116(%esp),%eax
+- # x13 = s
+- movl %ecx,152(%esp)
+- # t += x14
+- addl 156(%esp),%edx
+- # x7 = w
+- movl %ebx,128(%esp)
+- # p <<<= 13
+- rol $13,%eax
+- # p ^= x12
+- xorl 148(%esp),%eax
+- # t <<<= 13
+- rol $13,%edx
+- # t ^= x6
+- xorl 124(%esp),%edx
+- # r += s
+- add %ecx,%esi
+- # r <<<= 13
+- rol $13,%esi
+- # r ^= x1
+- xorl 104(%esp),%esi
+- # v += w
+- add %ebx,%edi
+- # v <<<= 13
+- rol $13,%edi
+- # v ^= x11
+- xorl 144(%esp),%edi
+- # x12 = p
+- movl %eax,148(%esp)
+- # x6 = t
+- movl %edx,124(%esp)
+- # p += x8
+- addl 132(%esp),%eax
+- # x1 = r
+- movl %esi,104(%esp)
+- # t += x2
+- addl 108(%esp),%edx
+- # x11 = v
+- movl %edi,144(%esp)
+- # p <<<= 18
+- rol $18,%eax
+- # p ^= x0
+- xorl 100(%esp),%eax
+- # t <<<= 18
+- rol $18,%edx
+- # t ^= x10
+- xorl 140(%esp),%edx
+- # s += r
+- add %esi,%ecx
+- # s <<<= 18
+- rol $18,%ecx
+- # s ^= x5
+- xorl 120(%esp),%ecx
+- # w += v
+- add %edi,%ebx
+- # w <<<= 18
+- rol $18,%ebx
+- # w ^= x15
+- xorl 160(%esp),%ebx
+- # x0 = p
+- movl %eax,100(%esp)
+- # x10 = t
+- movl %edx,140(%esp)
+- # p += x3
+- addl 112(%esp),%eax
+- # p <<<= 7
+- rol $7,%eax
+- # x5 = s
+- movl %ecx,120(%esp)
+- # t += x9
+- addl 136(%esp),%edx
+- # x15 = w
+- movl %ebx,160(%esp)
+- # r = x4
+- movl 116(%esp),%esi
+- # r += s
+- add %ecx,%esi
+- # v = x14
+- movl 156(%esp),%edi
+- # v += w
+- add %ebx,%edi
+- # p ^= x1
+- xorl 104(%esp),%eax
+- # t <<<= 7
+- rol $7,%edx
+- # t ^= x11
+- xorl 144(%esp),%edx
+- # r <<<= 7
+- rol $7,%esi
+- # r ^= x6
+- xorl 124(%esp),%esi
+- # v <<<= 7
+- rol $7,%edi
+- # v ^= x12
+- xorl 148(%esp),%edi
+- # x1 = p
+- movl %eax,104(%esp)
+- # x11 = t
+- movl %edx,144(%esp)
+- # p += x0
+- addl 100(%esp),%eax
+- # x6 = r
+- movl %esi,124(%esp)
+- # t += x10
+- addl 140(%esp),%edx
+- # x12 = v
+- movl %edi,148(%esp)
+- # p <<<= 9
+- rol $9,%eax
+- # p ^= x2
+- xorl 108(%esp),%eax
+- # t <<<= 9
+- rol $9,%edx
+- # t ^= x8
+- xorl 132(%esp),%edx
+- # s += r
+- add %esi,%ecx
+- # s <<<= 9
+- rol $9,%ecx
+- # s ^= x7
+- xorl 128(%esp),%ecx
+- # w += v
+- add %edi,%ebx
+- # w <<<= 9
+- rol $9,%ebx
+- # w ^= x13
+- xorl 152(%esp),%ebx
+- # x2 = p
+- movl %eax,108(%esp)
+- # x8 = t
+- movl %edx,132(%esp)
+- # p += x1
+- addl 104(%esp),%eax
+- # x7 = s
+- movl %ecx,128(%esp)
+- # t += x11
+- addl 144(%esp),%edx
+- # x13 = w
+- movl %ebx,152(%esp)
+- # p <<<= 13
+- rol $13,%eax
+- # p ^= x3
+- xorl 112(%esp),%eax
+- # t <<<= 13
+- rol $13,%edx
+- # t ^= x9
+- xorl 136(%esp),%edx
+- # r += s
+- add %ecx,%esi
+- # r <<<= 13
+- rol $13,%esi
+- # r ^= x4
+- xorl 116(%esp),%esi
+- # v += w
+- add %ebx,%edi
+- # v <<<= 13
+- rol $13,%edi
+- # v ^= x14
+- xorl 156(%esp),%edi
+- # x3 = p
+- movl %eax,112(%esp)
+- # x9 = t
+- movl %edx,136(%esp)
+- # p += x2
+- addl 108(%esp),%eax
+- # x4 = r
+- movl %esi,116(%esp)
+- # t += x8
+- addl 132(%esp),%edx
+- # x14 = v
+- movl %edi,156(%esp)
+- # p <<<= 18
+- rol $18,%eax
+- # p ^= x0
+- xorl 100(%esp),%eax
+- # t <<<= 18
+- rol $18,%edx
+- # t ^= x10
+- xorl 140(%esp),%edx
+- # s += r
+- add %esi,%ecx
+- # s <<<= 18
+- rol $18,%ecx
+- # s ^= x5
+- xorl 120(%esp),%ecx
+- # w += v
+- add %edi,%ebx
+- # w <<<= 18
+- rol $18,%ebx
+- # w ^= x15
+- xorl 160(%esp),%ebx
+- # i -= 4
+- sub $4,%ebp
+- # goto mainloop if unsigned >
+- ja ._mainloop
+- # x0 = p
+- movl %eax,100(%esp)
+- # x5 = s
+- movl %ecx,120(%esp)
+- # x10 = t
+- movl %edx,140(%esp)
+- # x15 = w
+- movl %ebx,160(%esp)
+- # out = out_backup
+- movl 72(%esp),%edi
+- # m = m_backup
+- movl 68(%esp),%esi
+- # in0 = x0
+- movl 100(%esp),%eax
+- # in1 = x1
+- movl 104(%esp),%ecx
+- # in0 += j0
+- addl 164(%esp),%eax
+- # in1 += j1
+- addl 168(%esp),%ecx
+- # in0 ^= *(uint32 *) (m + 0)
+- xorl 0(%esi),%eax
+- # in1 ^= *(uint32 *) (m + 4)
+- xorl 4(%esi),%ecx
+- # *(uint32 *) (out + 0) = in0
+- movl %eax,0(%edi)
+- # *(uint32 *) (out + 4) = in1
+- movl %ecx,4(%edi)
+- # in2 = x2
+- movl 108(%esp),%eax
+- # in3 = x3
+- movl 112(%esp),%ecx
+- # in2 += j2
+- addl 172(%esp),%eax
+- # in3 += j3
+- addl 176(%esp),%ecx
+- # in2 ^= *(uint32 *) (m + 8)
+- xorl 8(%esi),%eax
+- # in3 ^= *(uint32 *) (m + 12)
+- xorl 12(%esi),%ecx
+- # *(uint32 *) (out + 8) = in2
+- movl %eax,8(%edi)
+- # *(uint32 *) (out + 12) = in3
+- movl %ecx,12(%edi)
+- # in4 = x4
+- movl 116(%esp),%eax
+- # in5 = x5
+- movl 120(%esp),%ecx
+- # in4 += j4
+- addl 180(%esp),%eax
+- # in5 += j5
+- addl 184(%esp),%ecx
+- # in4 ^= *(uint32 *) (m + 16)
+- xorl 16(%esi),%eax
+- # in5 ^= *(uint32 *) (m + 20)
+- xorl 20(%esi),%ecx
+- # *(uint32 *) (out + 16) = in4
+- movl %eax,16(%edi)
+- # *(uint32 *) (out + 20) = in5
+- movl %ecx,20(%edi)
+- # in6 = x6
+- movl 124(%esp),%eax
+- # in7 = x7
+- movl 128(%esp),%ecx
+- # in6 += j6
+- addl 188(%esp),%eax
+- # in7 += j7
+- addl 192(%esp),%ecx
+- # in6 ^= *(uint32 *) (m + 24)
+- xorl 24(%esi),%eax
+- # in7 ^= *(uint32 *) (m + 28)
+- xorl 28(%esi),%ecx
+- # *(uint32 *) (out + 24) = in6
+- movl %eax,24(%edi)
+- # *(uint32 *) (out + 28) = in7
+- movl %ecx,28(%edi)
+- # in8 = x8
+- movl 132(%esp),%eax
+- # in9 = x9
+- movl 136(%esp),%ecx
+- # in8 += j8
+- addl 196(%esp),%eax
+- # in9 += j9
+- addl 200(%esp),%ecx
+- # in8 ^= *(uint32 *) (m + 32)
+- xorl 32(%esi),%eax
+- # in9 ^= *(uint32 *) (m + 36)
+- xorl 36(%esi),%ecx
+- # *(uint32 *) (out + 32) = in8
+- movl %eax,32(%edi)
+- # *(uint32 *) (out + 36) = in9
+- movl %ecx,36(%edi)
+- # in10 = x10
+- movl 140(%esp),%eax
+- # in11 = x11
+- movl 144(%esp),%ecx
+- # in10 += j10
+- addl 204(%esp),%eax
+- # in11 += j11
+- addl 208(%esp),%ecx
+- # in10 ^= *(uint32 *) (m + 40)
+- xorl 40(%esi),%eax
+- # in11 ^= *(uint32 *) (m + 44)
+- xorl 44(%esi),%ecx
+- # *(uint32 *) (out + 40) = in10
+- movl %eax,40(%edi)
+- # *(uint32 *) (out + 44) = in11
+- movl %ecx,44(%edi)
+- # in12 = x12
+- movl 148(%esp),%eax
+- # in13 = x13
+- movl 152(%esp),%ecx
+- # in12 += j12
+- addl 212(%esp),%eax
+- # in13 += j13
+- addl 216(%esp),%ecx
+- # in12 ^= *(uint32 *) (m + 48)
+- xorl 48(%esi),%eax
+- # in13 ^= *(uint32 *) (m + 52)
+- xorl 52(%esi),%ecx
+- # *(uint32 *) (out + 48) = in12
+- movl %eax,48(%edi)
+- # *(uint32 *) (out + 52) = in13
+- movl %ecx,52(%edi)
+- # in14 = x14
+- movl 156(%esp),%eax
+- # in15 = x15
+- movl 160(%esp),%ecx
+- # in14 += j14
+- addl 220(%esp),%eax
+- # in15 += j15
+- addl 224(%esp),%ecx
+- # in14 ^= *(uint32 *) (m + 56)
+- xorl 56(%esi),%eax
+- # in15 ^= *(uint32 *) (m + 60)
+- xorl 60(%esi),%ecx
+- # *(uint32 *) (out + 56) = in14
+- movl %eax,56(%edi)
+- # *(uint32 *) (out + 60) = in15
+- movl %ecx,60(%edi)
+- # bytes = bytes_backup
+- movl 76(%esp),%ebx
+- # in8 = j8
+- movl 196(%esp),%eax
+- # in9 = j9
+- movl 200(%esp),%ecx
+- # in8 += 1
+- add $1,%eax
+- # in9 += 0 + carry
+- adc $0,%ecx
+- # j8 = in8
+- movl %eax,196(%esp)
+- # j9 = in9
+- movl %ecx,200(%esp)
+- # bytes - 64
+- cmp $64,%ebx
+- # goto bytesatleast65 if unsigned>
+- ja ._bytesatleast65
+- # goto bytesatleast64 if unsigned>=
+- jae ._bytesatleast64
+- # m = out
+- mov %edi,%esi
+- # out = ctarget
+- movl 228(%esp),%edi
+- # i = bytes
+- mov %ebx,%ecx
+- # while (i) { *out++ = *m++; --i }
+- rep movsb
+-._bytesatleast64:
+- # x = x_backup
+- movl 64(%esp),%eax
+- # in8 = j8
+- movl 196(%esp),%ecx
+- # in9 = j9
+- movl 200(%esp),%edx
+- # *(uint32 *) (x + 32) = in8
+- movl %ecx,32(%eax)
+- # *(uint32 *) (x + 36) = in9
+- movl %edx,36(%eax)
+-._done:
+- # eax = eax_stack
+- movl 80(%esp),%eax
+- # ebx = ebx_stack
+- movl 84(%esp),%ebx
+- # esi = esi_stack
+- movl 88(%esp),%esi
+- # edi = edi_stack
+- movl 92(%esp),%edi
+- # ebp = ebp_stack
+- movl 96(%esp),%ebp
+- # leave
+- add %eax,%esp
+- ret
+-._bytesatleast65:
+- # bytes -= 64
+- sub $64,%ebx
+- # out += 64
+- add $64,%edi
+- # m += 64
+- add $64,%esi
+- # goto bytesatleast1
+- jmp ._bytesatleast1
+-ENDPROC(salsa20_encrypt_bytes)
+-
+-# enter salsa20_keysetup
+-ENTRY(salsa20_keysetup)
+- mov %esp,%eax
+- and $31,%eax
+- add $256,%eax
+- sub %eax,%esp
+- # eax_stack = eax
+- movl %eax,64(%esp)
+- # ebx_stack = ebx
+- movl %ebx,68(%esp)
+- # esi_stack = esi
+- movl %esi,72(%esp)
+- # edi_stack = edi
+- movl %edi,76(%esp)
+- # ebp_stack = ebp
+- movl %ebp,80(%esp)
+- # k = arg2
+- movl 8(%esp,%eax),%ecx
+- # kbits = arg3
+- movl 12(%esp,%eax),%edx
+- # x = arg1
+- movl 4(%esp,%eax),%eax
+- # in1 = *(uint32 *) (k + 0)
+- movl 0(%ecx),%ebx
+- # in2 = *(uint32 *) (k + 4)
+- movl 4(%ecx),%esi
+- # in3 = *(uint32 *) (k + 8)
+- movl 8(%ecx),%edi
+- # in4 = *(uint32 *) (k + 12)
+- movl 12(%ecx),%ebp
+- # *(uint32 *) (x + 4) = in1
+- movl %ebx,4(%eax)
+- # *(uint32 *) (x + 8) = in2
+- movl %esi,8(%eax)
+- # *(uint32 *) (x + 12) = in3
+- movl %edi,12(%eax)
+- # *(uint32 *) (x + 16) = in4
+- movl %ebp,16(%eax)
+- # kbits - 256
+- cmp $256,%edx
+- # goto kbits128 if unsigned<
+- jb ._kbits128
+-._kbits256:
+- # in11 = *(uint32 *) (k + 16)
+- movl 16(%ecx),%edx
+- # in12 = *(uint32 *) (k + 20)
+- movl 20(%ecx),%ebx
+- # in13 = *(uint32 *) (k + 24)
+- movl 24(%ecx),%esi
+- # in14 = *(uint32 *) (k + 28)
+- movl 28(%ecx),%ecx
+- # *(uint32 *) (x + 44) = in11
+- movl %edx,44(%eax)
+- # *(uint32 *) (x + 48) = in12
+- movl %ebx,48(%eax)
+- # *(uint32 *) (x + 52) = in13
+- movl %esi,52(%eax)
+- # *(uint32 *) (x + 56) = in14
+- movl %ecx,56(%eax)
+- # in0 = 1634760805
+- mov $1634760805,%ecx
+- # in5 = 857760878
+- mov $857760878,%edx
+- # in10 = 2036477234
+- mov $2036477234,%ebx
+- # in15 = 1797285236
+- mov $1797285236,%esi
+- # *(uint32 *) (x + 0) = in0
+- movl %ecx,0(%eax)
+- # *(uint32 *) (x + 20) = in5
+- movl %edx,20(%eax)
+- # *(uint32 *) (x + 40) = in10
+- movl %ebx,40(%eax)
+- # *(uint32 *) (x + 60) = in15
+- movl %esi,60(%eax)
+- # goto keysetupdone
+- jmp ._keysetupdone
+-._kbits128:
+- # in11 = *(uint32 *) (k + 0)
+- movl 0(%ecx),%edx
+- # in12 = *(uint32 *) (k + 4)
+- movl 4(%ecx),%ebx
+- # in13 = *(uint32 *) (k + 8)
+- movl 8(%ecx),%esi
+- # in14 = *(uint32 *) (k + 12)
+- movl 12(%ecx),%ecx
+- # *(uint32 *) (x + 44) = in11
+- movl %edx,44(%eax)
+- # *(uint32 *) (x + 48) = in12
+- movl %ebx,48(%eax)
+- # *(uint32 *) (x + 52) = in13
+- movl %esi,52(%eax)
+- # *(uint32 *) (x + 56) = in14
+- movl %ecx,56(%eax)
+- # in0 = 1634760805
+- mov $1634760805,%ecx
+- # in5 = 824206446
+- mov $824206446,%edx
+- # in10 = 2036477238
+- mov $2036477238,%ebx
+- # in15 = 1797285236
+- mov $1797285236,%esi
+- # *(uint32 *) (x + 0) = in0
+- movl %ecx,0(%eax)
+- # *(uint32 *) (x + 20) = in5
+- movl %edx,20(%eax)
+- # *(uint32 *) (x + 40) = in10
+- movl %ebx,40(%eax)
+- # *(uint32 *) (x + 60) = in15
+- movl %esi,60(%eax)
+-._keysetupdone:
+- # eax = eax_stack
+- movl 64(%esp),%eax
+- # ebx = ebx_stack
+- movl 68(%esp),%ebx
+- # esi = esi_stack
+- movl 72(%esp),%esi
+- # edi = edi_stack
+- movl 76(%esp),%edi
+- # ebp = ebp_stack
+- movl 80(%esp),%ebp
+- # leave
+- add %eax,%esp
+- ret
+-ENDPROC(salsa20_keysetup)
+-
+-# enter salsa20_ivsetup
+-ENTRY(salsa20_ivsetup)
+- mov %esp,%eax
+- and $31,%eax
+- add $256,%eax
+- sub %eax,%esp
+- # eax_stack = eax
+- movl %eax,64(%esp)
+- # ebx_stack = ebx
+- movl %ebx,68(%esp)
+- # esi_stack = esi
+- movl %esi,72(%esp)
+- # edi_stack = edi
+- movl %edi,76(%esp)
+- # ebp_stack = ebp
+- movl %ebp,80(%esp)
+- # iv = arg2
+- movl 8(%esp,%eax),%ecx
+- # x = arg1
+- movl 4(%esp,%eax),%eax
+- # in6 = *(uint32 *) (iv + 0)
+- movl 0(%ecx),%edx
+- # in7 = *(uint32 *) (iv + 4)
+- movl 4(%ecx),%ecx
+- # in8 = 0
+- mov $0,%ebx
+- # in9 = 0
+- mov $0,%esi
+- # *(uint32 *) (x + 24) = in6
+- movl %edx,24(%eax)
+- # *(uint32 *) (x + 28) = in7
+- movl %ecx,28(%eax)
+- # *(uint32 *) (x + 32) = in8
+- movl %ebx,32(%eax)
+- # *(uint32 *) (x + 36) = in9
+- movl %esi,36(%eax)
+- # eax = eax_stack
+- movl 64(%esp),%eax
+- # ebx = ebx_stack
+- movl 68(%esp),%ebx
+- # esi = esi_stack
+- movl 72(%esp),%esi
+- # edi = edi_stack
+- movl 76(%esp),%edi
+- # ebp = ebp_stack
+- movl 80(%esp),%ebp
+- # leave
+- add %eax,%esp
+- ret
+-ENDPROC(salsa20_ivsetup)
+--- a/arch/x86/crypto/salsa20-x86_64-asm_64.S
++++ /dev/null
+@@ -1,919 +0,0 @@
+-/* SPDX-License-Identifier: GPL-2.0 */
+-#include <linux/linkage.h>
+-
+-# enter salsa20_encrypt_bytes
+-ENTRY(salsa20_encrypt_bytes)
+- mov %rsp,%r11
+- and $31,%r11
+- add $256,%r11
+- sub %r11,%rsp
+- # x = arg1
+- mov %rdi,%r8
+- # m = arg2
+- mov %rsi,%rsi
+- # out = arg3
+- mov %rdx,%rdi
+- # bytes = arg4
+- mov %rcx,%rdx
+- # unsigned>? bytes - 0
+- cmp $0,%rdx
+- # comment:fp stack unchanged by jump
+- # goto done if !unsigned>
+- jbe ._done
+- # comment:fp stack unchanged by fallthrough
+-# start:
+-._start:
+- # r11_stack = r11
+- movq %r11,0(%rsp)
+- # r12_stack = r12
+- movq %r12,8(%rsp)
+- # r13_stack = r13
+- movq %r13,16(%rsp)
+- # r14_stack = r14
+- movq %r14,24(%rsp)
+- # r15_stack = r15
+- movq %r15,32(%rsp)
+- # rbx_stack = rbx
+- movq %rbx,40(%rsp)
+- # rbp_stack = rbp
+- movq %rbp,48(%rsp)
+- # in0 = *(uint64 *) (x + 0)
+- movq 0(%r8),%rcx
+- # in2 = *(uint64 *) (x + 8)
+- movq 8(%r8),%r9
+- # in4 = *(uint64 *) (x + 16)
+- movq 16(%r8),%rax
+- # in6 = *(uint64 *) (x + 24)
+- movq 24(%r8),%r10
+- # in8 = *(uint64 *) (x + 32)
+- movq 32(%r8),%r11
+- # in10 = *(uint64 *) (x + 40)
+- movq 40(%r8),%r12
+- # in12 = *(uint64 *) (x + 48)
+- movq 48(%r8),%r13
+- # in14 = *(uint64 *) (x + 56)
+- movq 56(%r8),%r14
+- # j0 = in0
+- movq %rcx,56(%rsp)
+- # j2 = in2
+- movq %r9,64(%rsp)
+- # j4 = in4
+- movq %rax,72(%rsp)
+- # j6 = in6
+- movq %r10,80(%rsp)
+- # j8 = in8
+- movq %r11,88(%rsp)
+- # j10 = in10
+- movq %r12,96(%rsp)
+- # j12 = in12
+- movq %r13,104(%rsp)
+- # j14 = in14
+- movq %r14,112(%rsp)
+- # x_backup = x
+- movq %r8,120(%rsp)
+-# bytesatleast1:
+-._bytesatleast1:
+- # unsigned<? bytes - 64
+- cmp $64,%rdx
+- # comment:fp stack unchanged by jump
+- # goto nocopy if !unsigned<
+- jae ._nocopy
+- # ctarget = out
+- movq %rdi,128(%rsp)
+- # out = &tmp
+- leaq 192(%rsp),%rdi
+- # i = bytes
+- mov %rdx,%rcx
+- # while (i) { *out++ = *m++; --i }
+- rep movsb
+- # out = &tmp
+- leaq 192(%rsp),%rdi
+- # m = &tmp
+- leaq 192(%rsp),%rsi
+- # comment:fp stack unchanged by fallthrough
+-# nocopy:
+-._nocopy:
+- # out_backup = out
+- movq %rdi,136(%rsp)
+- # m_backup = m
+- movq %rsi,144(%rsp)
+- # bytes_backup = bytes
+- movq %rdx,152(%rsp)
+- # x1 = j0
+- movq 56(%rsp),%rdi
+- # x0 = x1
+- mov %rdi,%rdx
+- # (uint64) x1 >>= 32
+- shr $32,%rdi
+- # x3 = j2
+- movq 64(%rsp),%rsi
+- # x2 = x3
+- mov %rsi,%rcx
+- # (uint64) x3 >>= 32
+- shr $32,%rsi
+- # x5 = j4
+- movq 72(%rsp),%r8
+- # x4 = x5
+- mov %r8,%r9
+- # (uint64) x5 >>= 32
+- shr $32,%r8
+- # x5_stack = x5
+- movq %r8,160(%rsp)
+- # x7 = j6
+- movq 80(%rsp),%r8
+- # x6 = x7
+- mov %r8,%rax
+- # (uint64) x7 >>= 32
+- shr $32,%r8
+- # x9 = j8
+- movq 88(%rsp),%r10
+- # x8 = x9
+- mov %r10,%r11
+- # (uint64) x9 >>= 32
+- shr $32,%r10
+- # x11 = j10
+- movq 96(%rsp),%r12
+- # x10 = x11
+- mov %r12,%r13
+- # x10_stack = x10
+- movq %r13,168(%rsp)
+- # (uint64) x11 >>= 32
+- shr $32,%r12
+- # x13 = j12
+- movq 104(%rsp),%r13
+- # x12 = x13
+- mov %r13,%r14
+- # (uint64) x13 >>= 32
+- shr $32,%r13
+- # x15 = j14
+- movq 112(%rsp),%r15
+- # x14 = x15
+- mov %r15,%rbx
+- # (uint64) x15 >>= 32
+- shr $32,%r15
+- # x15_stack = x15
+- movq %r15,176(%rsp)
+- # i = 20
+- mov $20,%r15
+-# mainloop:
+-._mainloop:
+- # i_backup = i
+- movq %r15,184(%rsp)
+- # x5 = x5_stack
+- movq 160(%rsp),%r15
+- # a = x12 + x0
+- lea (%r14,%rdx),%rbp
+- # (uint32) a <<<= 7
+- rol $7,%ebp
+- # x4 ^= a
+- xor %rbp,%r9
+- # b = x1 + x5
+- lea (%rdi,%r15),%rbp
+- # (uint32) b <<<= 7
+- rol $7,%ebp
+- # x9 ^= b
+- xor %rbp,%r10
+- # a = x0 + x4
+- lea (%rdx,%r9),%rbp
+- # (uint32) a <<<= 9
+- rol $9,%ebp
+- # x8 ^= a
+- xor %rbp,%r11
+- # b = x5 + x9
+- lea (%r15,%r10),%rbp
+- # (uint32) b <<<= 9
+- rol $9,%ebp
+- # x13 ^= b
+- xor %rbp,%r13
+- # a = x4 + x8
+- lea (%r9,%r11),%rbp
+- # (uint32) a <<<= 13
+- rol $13,%ebp
+- # x12 ^= a
+- xor %rbp,%r14
+- # b = x9 + x13
+- lea (%r10,%r13),%rbp
+- # (uint32) b <<<= 13
+- rol $13,%ebp
+- # x1 ^= b
+- xor %rbp,%rdi
+- # a = x8 + x12
+- lea (%r11,%r14),%rbp
+- # (uint32) a <<<= 18
+- rol $18,%ebp
+- # x0 ^= a
+- xor %rbp,%rdx
+- # b = x13 + x1
+- lea (%r13,%rdi),%rbp
+- # (uint32) b <<<= 18
+- rol $18,%ebp
+- # x5 ^= b
+- xor %rbp,%r15
+- # x10 = x10_stack
+- movq 168(%rsp),%rbp
+- # x5_stack = x5
+- movq %r15,160(%rsp)
+- # c = x6 + x10
+- lea (%rax,%rbp),%r15
+- # (uint32) c <<<= 7
+- rol $7,%r15d
+- # x14 ^= c
+- xor %r15,%rbx
+- # c = x10 + x14
+- lea (%rbp,%rbx),%r15
+- # (uint32) c <<<= 9
+- rol $9,%r15d
+- # x2 ^= c
+- xor %r15,%rcx
+- # c = x14 + x2
+- lea (%rbx,%rcx),%r15
+- # (uint32) c <<<= 13
+- rol $13,%r15d
+- # x6 ^= c
+- xor %r15,%rax
+- # c = x2 + x6
+- lea (%rcx,%rax),%r15
+- # (uint32) c <<<= 18
+- rol $18,%r15d
+- # x10 ^= c
+- xor %r15,%rbp
+- # x15 = x15_stack
+- movq 176(%rsp),%r15
+- # x10_stack = x10
+- movq %rbp,168(%rsp)
+- # d = x11 + x15
+- lea (%r12,%r15),%rbp
+- # (uint32) d <<<= 7
+- rol $7,%ebp
+- # x3 ^= d
+- xor %rbp,%rsi
+- # d = x15 + x3
+- lea (%r15,%rsi),%rbp
+- # (uint32) d <<<= 9
+- rol $9,%ebp
+- # x7 ^= d
+- xor %rbp,%r8
+- # d = x3 + x7
+- lea (%rsi,%r8),%rbp
+- # (uint32) d <<<= 13
+- rol $13,%ebp
+- # x11 ^= d
+- xor %rbp,%r12
+- # d = x7 + x11
+- lea (%r8,%r12),%rbp
+- # (uint32) d <<<= 18
+- rol $18,%ebp
+- # x15 ^= d
+- xor %rbp,%r15
+- # x15_stack = x15
+- movq %r15,176(%rsp)
+- # x5 = x5_stack
+- movq 160(%rsp),%r15
+- # a = x3 + x0
+- lea (%rsi,%rdx),%rbp
+- # (uint32) a <<<= 7
+- rol $7,%ebp
+- # x1 ^= a
+- xor %rbp,%rdi
+- # b = x4 + x5
+- lea (%r9,%r15),%rbp
+- # (uint32) b <<<= 7
+- rol $7,%ebp
+- # x6 ^= b
+- xor %rbp,%rax
+- # a = x0 + x1
+- lea (%rdx,%rdi),%rbp
+- # (uint32) a <<<= 9
+- rol $9,%ebp
+- # x2 ^= a
+- xor %rbp,%rcx
+- # b = x5 + x6
+- lea (%r15,%rax),%rbp
+- # (uint32) b <<<= 9
+- rol $9,%ebp
+- # x7 ^= b
+- xor %rbp,%r8
+- # a = x1 + x2
+- lea (%rdi,%rcx),%rbp
+- # (uint32) a <<<= 13
+- rol $13,%ebp
+- # x3 ^= a
+- xor %rbp,%rsi
+- # b = x6 + x7
+- lea (%rax,%r8),%rbp
+- # (uint32) b <<<= 13
+- rol $13,%ebp
+- # x4 ^= b
+- xor %rbp,%r9
+- # a = x2 + x3
+- lea (%rcx,%rsi),%rbp
+- # (uint32) a <<<= 18
+- rol $18,%ebp
+- # x0 ^= a
+- xor %rbp,%rdx
+- # b = x7 + x4
+- lea (%r8,%r9),%rbp
+- # (uint32) b <<<= 18
+- rol $18,%ebp
+- # x5 ^= b
+- xor %rbp,%r15
+- # x10 = x10_stack
+- movq 168(%rsp),%rbp
+- # x5_stack = x5
+- movq %r15,160(%rsp)
+- # c = x9 + x10
+- lea (%r10,%rbp),%r15
+- # (uint32) c <<<= 7
+- rol $7,%r15d
+- # x11 ^= c
+- xor %r15,%r12
+- # c = x10 + x11
+- lea (%rbp,%r12),%r15
+- # (uint32) c <<<= 9
+- rol $9,%r15d
+- # x8 ^= c
+- xor %r15,%r11
+- # c = x11 + x8
+- lea (%r12,%r11),%r15
+- # (uint32) c <<<= 13
+- rol $13,%r15d
+- # x9 ^= c
+- xor %r15,%r10
+- # c = x8 + x9
+- lea (%r11,%r10),%r15
+- # (uint32) c <<<= 18
+- rol $18,%r15d
+- # x10 ^= c
+- xor %r15,%rbp
+- # x15 = x15_stack
+- movq 176(%rsp),%r15
+- # x10_stack = x10
+- movq %rbp,168(%rsp)
+- # d = x14 + x15
+- lea (%rbx,%r15),%rbp
+- # (uint32) d <<<= 7
+- rol $7,%ebp
+- # x12 ^= d
+- xor %rbp,%r14
+- # d = x15 + x12
+- lea (%r15,%r14),%rbp
+- # (uint32) d <<<= 9
+- rol $9,%ebp
+- # x13 ^= d
+- xor %rbp,%r13
+- # d = x12 + x13
+- lea (%r14,%r13),%rbp
+- # (uint32) d <<<= 13
+- rol $13,%ebp
+- # x14 ^= d
+- xor %rbp,%rbx
+- # d = x13 + x14
+- lea (%r13,%rbx),%rbp
+- # (uint32) d <<<= 18
+- rol $18,%ebp
+- # x15 ^= d
+- xor %rbp,%r15
+- # x15_stack = x15
+- movq %r15,176(%rsp)
+- # x5 = x5_stack
+- movq 160(%rsp),%r15
+- # a = x12 + x0
+- lea (%r14,%rdx),%rbp
+- # (uint32) a <<<= 7
+- rol $7,%ebp
+- # x4 ^= a
+- xor %rbp,%r9
+- # b = x1 + x5
+- lea (%rdi,%r15),%rbp
+- # (uint32) b <<<= 7
+- rol $7,%ebp
+- # x9 ^= b
+- xor %rbp,%r10
+- # a = x0 + x4
+- lea (%rdx,%r9),%rbp
+- # (uint32) a <<<= 9
+- rol $9,%ebp
+- # x8 ^= a
+- xor %rbp,%r11
+- # b = x5 + x9
+- lea (%r15,%r10),%rbp
+- # (uint32) b <<<= 9
+- rol $9,%ebp
+- # x13 ^= b
+- xor %rbp,%r13
+- # a = x4 + x8
+- lea (%r9,%r11),%rbp
+- # (uint32) a <<<= 13
+- rol $13,%ebp
+- # x12 ^= a
+- xor %rbp,%r14
+- # b = x9 + x13
+- lea (%r10,%r13),%rbp
+- # (uint32) b <<<= 13
+- rol $13,%ebp
+- # x1 ^= b
+- xor %rbp,%rdi
+- # a = x8 + x12
+- lea (%r11,%r14),%rbp
+- # (uint32) a <<<= 18
+- rol $18,%ebp
+- # x0 ^= a
+- xor %rbp,%rdx
+- # b = x13 + x1
+- lea (%r13,%rdi),%rbp
+- # (uint32) b <<<= 18
+- rol $18,%ebp
+- # x5 ^= b
+- xor %rbp,%r15
+- # x10 = x10_stack
+- movq 168(%rsp),%rbp
+- # x5_stack = x5
+- movq %r15,160(%rsp)
+- # c = x6 + x10
+- lea (%rax,%rbp),%r15
+- # (uint32) c <<<= 7
+- rol $7,%r15d
+- # x14 ^= c
+- xor %r15,%rbx
+- # c = x10 + x14
+- lea (%rbp,%rbx),%r15
+- # (uint32) c <<<= 9
+- rol $9,%r15d
+- # x2 ^= c
+- xor %r15,%rcx
+- # c = x14 + x2
+- lea (%rbx,%rcx),%r15
+- # (uint32) c <<<= 13
+- rol $13,%r15d
+- # x6 ^= c
+- xor %r15,%rax
+- # c = x2 + x6
+- lea (%rcx,%rax),%r15
+- # (uint32) c <<<= 18
+- rol $18,%r15d
+- # x10 ^= c
+- xor %r15,%rbp
+- # x15 = x15_stack
+- movq 176(%rsp),%r15
+- # x10_stack = x10
+- movq %rbp,168(%rsp)
+- # d = x11 + x15
+- lea (%r12,%r15),%rbp
+- # (uint32) d <<<= 7
+- rol $7,%ebp
+- # x3 ^= d
+- xor %rbp,%rsi
+- # d = x15 + x3
+- lea (%r15,%rsi),%rbp
+- # (uint32) d <<<= 9
+- rol $9,%ebp
+- # x7 ^= d
+- xor %rbp,%r8
+- # d = x3 + x7
+- lea (%rsi,%r8),%rbp
+- # (uint32) d <<<= 13
+- rol $13,%ebp
+- # x11 ^= d
+- xor %rbp,%r12
+- # d = x7 + x11
+- lea (%r8,%r12),%rbp
+- # (uint32) d <<<= 18
+- rol $18,%ebp
+- # x15 ^= d
+- xor %rbp,%r15
+- # x15_stack = x15
+- movq %r15,176(%rsp)
+- # x5 = x5_stack
+- movq 160(%rsp),%r15
+- # a = x3 + x0
+- lea (%rsi,%rdx),%rbp
+- # (uint32) a <<<= 7
+- rol $7,%ebp
+- # x1 ^= a
+- xor %rbp,%rdi
+- # b = x4 + x5
+- lea (%r9,%r15),%rbp
+- # (uint32) b <<<= 7
+- rol $7,%ebp
+- # x6 ^= b
+- xor %rbp,%rax
+- # a = x0 + x1
+- lea (%rdx,%rdi),%rbp
+- # (uint32) a <<<= 9
+- rol $9,%ebp
+- # x2 ^= a
+- xor %rbp,%rcx
+- # b = x5 + x6
+- lea (%r15,%rax),%rbp
+- # (uint32) b <<<= 9
+- rol $9,%ebp
+- # x7 ^= b
+- xor %rbp,%r8
+- # a = x1 + x2
+- lea (%rdi,%rcx),%rbp
+- # (uint32) a <<<= 13
+- rol $13,%ebp
+- # x3 ^= a
+- xor %rbp,%rsi
+- # b = x6 + x7
+- lea (%rax,%r8),%rbp
+- # (uint32) b <<<= 13
+- rol $13,%ebp
+- # x4 ^= b
+- xor %rbp,%r9
+- # a = x2 + x3
+- lea (%rcx,%rsi),%rbp
+- # (uint32) a <<<= 18
+- rol $18,%ebp
+- # x0 ^= a
+- xor %rbp,%rdx
+- # b = x7 + x4
+- lea (%r8,%r9),%rbp
+- # (uint32) b <<<= 18
+- rol $18,%ebp
+- # x5 ^= b
+- xor %rbp,%r15
+- # x10 = x10_stack
+- movq 168(%rsp),%rbp
+- # x5_stack = x5
+- movq %r15,160(%rsp)
+- # c = x9 + x10
+- lea (%r10,%rbp),%r15
+- # (uint32) c <<<= 7
+- rol $7,%r15d
+- # x11 ^= c
+- xor %r15,%r12
+- # c = x10 + x11
+- lea (%rbp,%r12),%r15
+- # (uint32) c <<<= 9
+- rol $9,%r15d
+- # x8 ^= c
+- xor %r15,%r11
+- # c = x11 + x8
+- lea (%r12,%r11),%r15
+- # (uint32) c <<<= 13
+- rol $13,%r15d
+- # x9 ^= c
+- xor %r15,%r10
+- # c = x8 + x9
+- lea (%r11,%r10),%r15
+- # (uint32) c <<<= 18
+- rol $18,%r15d
+- # x10 ^= c
+- xor %r15,%rbp
+- # x15 = x15_stack
+- movq 176(%rsp),%r15
+- # x10_stack = x10
+- movq %rbp,168(%rsp)
+- # d = x14 + x15
+- lea (%rbx,%r15),%rbp
+- # (uint32) d <<<= 7
+- rol $7,%ebp
+- # x12 ^= d
+- xor %rbp,%r14
+- # d = x15 + x12
+- lea (%r15,%r14),%rbp
+- # (uint32) d <<<= 9
+- rol $9,%ebp
+- # x13 ^= d
+- xor %rbp,%r13
+- # d = x12 + x13
+- lea (%r14,%r13),%rbp
+- # (uint32) d <<<= 13
+- rol $13,%ebp
+- # x14 ^= d
+- xor %rbp,%rbx
+- # d = x13 + x14
+- lea (%r13,%rbx),%rbp
+- # (uint32) d <<<= 18
+- rol $18,%ebp
+- # x15 ^= d
+- xor %rbp,%r15
+- # x15_stack = x15
+- movq %r15,176(%rsp)
+- # i = i_backup
+- movq 184(%rsp),%r15
+- # unsigned>? i -= 4
+- sub $4,%r15
+- # comment:fp stack unchanged by jump
+- # goto mainloop if unsigned>
+- ja ._mainloop
+- # (uint32) x2 += j2
+- addl 64(%rsp),%ecx
+- # x3 <<= 32
+- shl $32,%rsi
+- # x3 += j2
+- addq 64(%rsp),%rsi
+- # (uint64) x3 >>= 32
+- shr $32,%rsi
+- # x3 <<= 32
+- shl $32,%rsi
+- # x2 += x3
+- add %rsi,%rcx
+- # (uint32) x6 += j6
+- addl 80(%rsp),%eax
+- # x7 <<= 32
+- shl $32,%r8
+- # x7 += j6
+- addq 80(%rsp),%r8
+- # (uint64) x7 >>= 32
+- shr $32,%r8
+- # x7 <<= 32
+- shl $32,%r8
+- # x6 += x7
+- add %r8,%rax
+- # (uint32) x8 += j8
+- addl 88(%rsp),%r11d
+- # x9 <<= 32
+- shl $32,%r10
+- # x9 += j8
+- addq 88(%rsp),%r10
+- # (uint64) x9 >>= 32
+- shr $32,%r10
+- # x9 <<= 32
+- shl $32,%r10
+- # x8 += x9
+- add %r10,%r11
+- # (uint32) x12 += j12
+- addl 104(%rsp),%r14d
+- # x13 <<= 32
+- shl $32,%r13
+- # x13 += j12
+- addq 104(%rsp),%r13
+- # (uint64) x13 >>= 32
+- shr $32,%r13
+- # x13 <<= 32
+- shl $32,%r13
+- # x12 += x13
+- add %r13,%r14
+- # (uint32) x0 += j0
+- addl 56(%rsp),%edx
+- # x1 <<= 32
+- shl $32,%rdi
+- # x1 += j0
+- addq 56(%rsp),%rdi
+- # (uint64) x1 >>= 32
+- shr $32,%rdi
+- # x1 <<= 32
+- shl $32,%rdi
+- # x0 += x1
+- add %rdi,%rdx
+- # x5 = x5_stack
+- movq 160(%rsp),%rdi
+- # (uint32) x4 += j4
+- addl 72(%rsp),%r9d
+- # x5 <<= 32
+- shl $32,%rdi
+- # x5 += j4
+- addq 72(%rsp),%rdi
+- # (uint64) x5 >>= 32
+- shr $32,%rdi
+- # x5 <<= 32
+- shl $32,%rdi
+- # x4 += x5
+- add %rdi,%r9
+- # x10 = x10_stack
+- movq 168(%rsp),%r8
+- # (uint32) x10 += j10
+- addl 96(%rsp),%r8d
+- # x11 <<= 32
+- shl $32,%r12
+- # x11 += j10
+- addq 96(%rsp),%r12
+- # (uint64) x11 >>= 32
+- shr $32,%r12
+- # x11 <<= 32
+- shl $32,%r12
+- # x10 += x11
+- add %r12,%r8
+- # x15 = x15_stack
+- movq 176(%rsp),%rdi
+- # (uint32) x14 += j14
+- addl 112(%rsp),%ebx
+- # x15 <<= 32
+- shl $32,%rdi
+- # x15 += j14
+- addq 112(%rsp),%rdi
+- # (uint64) x15 >>= 32
+- shr $32,%rdi
+- # x15 <<= 32
+- shl $32,%rdi
+- # x14 += x15
+- add %rdi,%rbx
+- # out = out_backup
+- movq 136(%rsp),%rdi
+- # m = m_backup
+- movq 144(%rsp),%rsi
+- # x0 ^= *(uint64 *) (m + 0)
+- xorq 0(%rsi),%rdx
+- # *(uint64 *) (out + 0) = x0
+- movq %rdx,0(%rdi)
+- # x2 ^= *(uint64 *) (m + 8)
+- xorq 8(%rsi),%rcx
+- # *(uint64 *) (out + 8) = x2
+- movq %rcx,8(%rdi)
+- # x4 ^= *(uint64 *) (m + 16)
+- xorq 16(%rsi),%r9
+- # *(uint64 *) (out + 16) = x4
+- movq %r9,16(%rdi)
+- # x6 ^= *(uint64 *) (m + 24)
+- xorq 24(%rsi),%rax
+- # *(uint64 *) (out + 24) = x6
+- movq %rax,24(%rdi)
+- # x8 ^= *(uint64 *) (m + 32)
+- xorq 32(%rsi),%r11
+- # *(uint64 *) (out + 32) = x8
+- movq %r11,32(%rdi)
+- # x10 ^= *(uint64 *) (m + 40)
+- xorq 40(%rsi),%r8
+- # *(uint64 *) (out + 40) = x10
+- movq %r8,40(%rdi)
+- # x12 ^= *(uint64 *) (m + 48)
+- xorq 48(%rsi),%r14
+- # *(uint64 *) (out + 48) = x12
+- movq %r14,48(%rdi)
+- # x14 ^= *(uint64 *) (m + 56)
+- xorq 56(%rsi),%rbx
+- # *(uint64 *) (out + 56) = x14
+- movq %rbx,56(%rdi)
+- # bytes = bytes_backup
+- movq 152(%rsp),%rdx
+- # in8 = j8
+- movq 88(%rsp),%rcx
+- # in8 += 1
+- add $1,%rcx
+- # j8 = in8
+- movq %rcx,88(%rsp)
+- # unsigned>? unsigned<? bytes - 64
+- cmp $64,%rdx
+- # comment:fp stack unchanged by jump
+- # goto bytesatleast65 if unsigned>
+- ja ._bytesatleast65
+- # comment:fp stack unchanged by jump
+- # goto bytesatleast64 if !unsigned<
+- jae ._bytesatleast64
+- # m = out
+- mov %rdi,%rsi
+- # out = ctarget
+- movq 128(%rsp),%rdi
+- # i = bytes
+- mov %rdx,%rcx
+- # while (i) { *out++ = *m++; --i }
+- rep movsb
+- # comment:fp stack unchanged by fallthrough
+-# bytesatleast64:
+-._bytesatleast64:
+- # x = x_backup
+- movq 120(%rsp),%rdi
+- # in8 = j8
+- movq 88(%rsp),%rsi
+- # *(uint64 *) (x + 32) = in8
+- movq %rsi,32(%rdi)
+- # r11 = r11_stack
+- movq 0(%rsp),%r11
+- # r12 = r12_stack
+- movq 8(%rsp),%r12
+- # r13 = r13_stack
+- movq 16(%rsp),%r13
+- # r14 = r14_stack
+- movq 24(%rsp),%r14
+- # r15 = r15_stack
+- movq 32(%rsp),%r15
+- # rbx = rbx_stack
+- movq 40(%rsp),%rbx
+- # rbp = rbp_stack
+- movq 48(%rsp),%rbp
+- # comment:fp stack unchanged by fallthrough
+-# done:
+-._done:
+- # leave
+- add %r11,%rsp
+- mov %rdi,%rax
+- mov %rsi,%rdx
+- ret
+-# bytesatleast65:
+-._bytesatleast65:
+- # bytes -= 64
+- sub $64,%rdx
+- # out += 64
+- add $64,%rdi
+- # m += 64
+- add $64,%rsi
+- # comment:fp stack unchanged by jump
+- # goto bytesatleast1
+- jmp ._bytesatleast1
+-ENDPROC(salsa20_encrypt_bytes)
+-
+-# enter salsa20_keysetup
+-ENTRY(salsa20_keysetup)
+- mov %rsp,%r11
+- and $31,%r11
+- add $256,%r11
+- sub %r11,%rsp
+- # k = arg2
+- mov %rsi,%rsi
+- # kbits = arg3
+- mov %rdx,%rdx
+- # x = arg1
+- mov %rdi,%rdi
+- # in0 = *(uint64 *) (k + 0)
+- movq 0(%rsi),%r8
+- # in2 = *(uint64 *) (k + 8)
+- movq 8(%rsi),%r9
+- # *(uint64 *) (x + 4) = in0
+- movq %r8,4(%rdi)
+- # *(uint64 *) (x + 12) = in2
+- movq %r9,12(%rdi)
+- # unsigned<? kbits - 256
+- cmp $256,%rdx
+- # comment:fp stack unchanged by jump
+- # goto kbits128 if unsigned<
+- jb ._kbits128
+-# kbits256:
+-._kbits256:
+- # in10 = *(uint64 *) (k + 16)
+- movq 16(%rsi),%rdx
+- # in12 = *(uint64 *) (k + 24)
+- movq 24(%rsi),%rsi
+- # *(uint64 *) (x + 44) = in10
+- movq %rdx,44(%rdi)
+- # *(uint64 *) (x + 52) = in12
+- movq %rsi,52(%rdi)
+- # in0 = 1634760805
+- mov $1634760805,%rsi
+- # in4 = 857760878
+- mov $857760878,%rdx
+- # in10 = 2036477234
+- mov $2036477234,%rcx
+- # in14 = 1797285236
+- mov $1797285236,%r8
+- # *(uint32 *) (x + 0) = in0
+- movl %esi,0(%rdi)
+- # *(uint32 *) (x + 20) = in4
+- movl %edx,20(%rdi)
+- # *(uint32 *) (x + 40) = in10
+- movl %ecx,40(%rdi)
+- # *(uint32 *) (x + 60) = in14
+- movl %r8d,60(%rdi)
+- # comment:fp stack unchanged by jump
+- # goto keysetupdone
+- jmp ._keysetupdone
+-# kbits128:
+-._kbits128:
+- # in10 = *(uint64 *) (k + 0)
+- movq 0(%rsi),%rdx
+- # in12 = *(uint64 *) (k + 8)
+- movq 8(%rsi),%rsi
+- # *(uint64 *) (x + 44) = in10
+- movq %rdx,44(%rdi)
+- # *(uint64 *) (x + 52) = in12
+- movq %rsi,52(%rdi)
+- # in0 = 1634760805
+- mov $1634760805,%rsi
+- # in4 = 824206446
+- mov $824206446,%rdx
+- # in10 = 2036477238
+- mov $2036477238,%rcx
+- # in14 = 1797285236
+- mov $1797285236,%r8
+- # *(uint32 *) (x + 0) = in0
+- movl %esi,0(%rdi)
+- # *(uint32 *) (x + 20) = in4
+- movl %edx,20(%rdi)
+- # *(uint32 *) (x + 40) = in10
+- movl %ecx,40(%rdi)
+- # *(uint32 *) (x + 60) = in14
+- movl %r8d,60(%rdi)
+-# keysetupdone:
+-._keysetupdone:
+- # leave
+- add %r11,%rsp
+- mov %rdi,%rax
+- mov %rsi,%rdx
+- ret
+-ENDPROC(salsa20_keysetup)
+-
+-# enter salsa20_ivsetup
+-ENTRY(salsa20_ivsetup)
+- mov %rsp,%r11
+- and $31,%r11
+- add $256,%r11
+- sub %r11,%rsp
+- # iv = arg2
+- mov %rsi,%rsi
+- # x = arg1
+- mov %rdi,%rdi
+- # in6 = *(uint64 *) (iv + 0)
+- movq 0(%rsi),%rsi
+- # in8 = 0
+- mov $0,%r8
+- # *(uint64 *) (x + 24) = in6
+- movq %rsi,24(%rdi)
+- # *(uint64 *) (x + 32) = in8
+- movq %r8,32(%rdi)
+- # leave
+- add %r11,%rsp
+- mov %rdi,%rax
+- mov %rsi,%rdx
+- ret
+-ENDPROC(salsa20_ivsetup)
+--- a/arch/x86/crypto/salsa20_glue.c
++++ /dev/null
+@@ -1,116 +0,0 @@
+-/*
+- * Glue code for optimized assembly version of Salsa20.
+- *
+- * Copyright (c) 2007 Tan Swee Heng <thesweeheng@gmail.com>
+- *
+- * The assembly codes are public domain assembly codes written by Daniel. J.
+- * Bernstein <djb@cr.yp.to>. The codes are modified to include indentation
+- * and to remove extraneous comments and functions that are not needed.
+- * - i586 version, renamed as salsa20-i586-asm_32.S
+- * available from <http://cr.yp.to/snuffle/salsa20/x86-pm/salsa20.s>
+- * - x86-64 version, renamed as salsa20-x86_64-asm_64.S
+- * available from <http://cr.yp.to/snuffle/salsa20/amd64-3/salsa20.s>
+- *
+- * This program is free software; you can redistribute it and/or modify it
+- * under the terms of the GNU General Public License as published by the Free
+- * Software Foundation; either version 2 of the License, or (at your option)
+- * any later version.
+- *
+- */
+-
+-#include <crypto/algapi.h>
+-#include <linux/module.h>
+-#include <linux/crypto.h>
+-
+-#define SALSA20_IV_SIZE 8U
+-#define SALSA20_MIN_KEY_SIZE 16U
+-#define SALSA20_MAX_KEY_SIZE 32U
+-
+-struct salsa20_ctx
+-{
+- u32 input[16];
+-};
+-
+-asmlinkage void salsa20_keysetup(struct salsa20_ctx *ctx, const u8 *k,
+- u32 keysize, u32 ivsize);
+-asmlinkage void salsa20_ivsetup(struct salsa20_ctx *ctx, const u8 *iv);
+-asmlinkage void salsa20_encrypt_bytes(struct salsa20_ctx *ctx,
+- const u8 *src, u8 *dst, u32 bytes);
+-
+-static int setkey(struct crypto_tfm *tfm, const u8 *key,
+- unsigned int keysize)
+-{
+- struct salsa20_ctx *ctx = crypto_tfm_ctx(tfm);
+- salsa20_keysetup(ctx, key, keysize*8, SALSA20_IV_SIZE*8);
+- return 0;
+-}
+-
+-static int encrypt(struct blkcipher_desc *desc,
+- struct scatterlist *dst, struct scatterlist *src,
+- unsigned int nbytes)
+-{
+- struct blkcipher_walk walk;
+- struct crypto_blkcipher *tfm = desc->tfm;
+- struct salsa20_ctx *ctx = crypto_blkcipher_ctx(tfm);
+- int err;
+-
+- blkcipher_walk_init(&walk, dst, src, nbytes);
+- err = blkcipher_walk_virt_block(desc, &walk, 64);
+-
+- salsa20_ivsetup(ctx, walk.iv);
+-
+- while (walk.nbytes >= 64) {
+- salsa20_encrypt_bytes(ctx, walk.src.virt.addr,
+- walk.dst.virt.addr,
+- walk.nbytes - (walk.nbytes % 64));
+- err = blkcipher_walk_done(desc, &walk, walk.nbytes % 64);
+- }
+-
+- if (walk.nbytes) {
+- salsa20_encrypt_bytes(ctx, walk.src.virt.addr,
+- walk.dst.virt.addr, walk.nbytes);
+- err = blkcipher_walk_done(desc, &walk, 0);
+- }
+-
+- return err;
+-}
+-
+-static struct crypto_alg alg = {
+- .cra_name = "salsa20",
+- .cra_driver_name = "salsa20-asm",
+- .cra_priority = 200,
+- .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER,
+- .cra_type = &crypto_blkcipher_type,
+- .cra_blocksize = 1,
+- .cra_ctxsize = sizeof(struct salsa20_ctx),
+- .cra_alignmask = 3,
+- .cra_module = THIS_MODULE,
+- .cra_u = {
+- .blkcipher = {
+- .setkey = setkey,
+- .encrypt = encrypt,
+- .decrypt = encrypt,
+- .min_keysize = SALSA20_MIN_KEY_SIZE,
+- .max_keysize = SALSA20_MAX_KEY_SIZE,
+- .ivsize = SALSA20_IV_SIZE,
+- }
+- }
+-};
+-
+-static int __init init(void)
+-{
+- return crypto_register_alg(&alg);
+-}
+-
+-static void __exit fini(void)
+-{
+- crypto_unregister_alg(&alg);
+-}
+-
+-module_init(init);
+-module_exit(fini);
+-
+-MODULE_LICENSE("GPL");
+-MODULE_DESCRIPTION ("Salsa20 stream cipher algorithm (optimized assembly version)");
+-MODULE_ALIAS_CRYPTO("salsa20");
+-MODULE_ALIAS_CRYPTO("salsa20-asm");
+--- a/crypto/Kconfig
++++ b/crypto/Kconfig
+@@ -1324,32 +1324,6 @@ config CRYPTO_SALSA20
+ The Salsa20 stream cipher algorithm is designed by Daniel J.
+ Bernstein <djb@cr.yp.to>. See <http://cr.yp.to/snuffle.html>
+
+-config CRYPTO_SALSA20_586
+- tristate "Salsa20 stream cipher algorithm (i586)"
+- depends on (X86 || UML_X86) && !64BIT
+- select CRYPTO_BLKCIPHER
+- help
+- Salsa20 stream cipher algorithm.
+-
+- Salsa20 is a stream cipher submitted to eSTREAM, the ECRYPT
+- Stream Cipher Project. See <http://www.ecrypt.eu.org/stream/>
+-
+- The Salsa20 stream cipher algorithm is designed by Daniel J.
+- Bernstein <djb@cr.yp.to>. See <http://cr.yp.to/snuffle.html>
+-
+-config CRYPTO_SALSA20_X86_64
+- tristate "Salsa20 stream cipher algorithm (x86_64)"
+- depends on (X86 || UML_X86) && 64BIT
+- select CRYPTO_BLKCIPHER
+- help
+- Salsa20 stream cipher algorithm.
+-
+- Salsa20 is a stream cipher submitted to eSTREAM, the ECRYPT
+- Stream Cipher Project. See <http://www.ecrypt.eu.org/stream/>
+-
+- The Salsa20 stream cipher algorithm is designed by Daniel J.
+- Bernstein <djb@cr.yp.to>. See <http://cr.yp.to/snuffle.html>
+-
+ config CRYPTO_CHACHA20
+ tristate "ChaCha20 cipher algorithm"
+ select CRYPTO_BLKCIPHER
--- /dev/null
+From a4f843bd004d775cbb360cd375969b8a479568a9 Mon Sep 17 00:00:00 2001
+From: Jaegeuk Kim <jaegeuk@kernel.org>
+Date: Mon, 23 Apr 2018 23:02:31 -0600
+Subject: f2fs: give message and set need_fsck given broken node id
+
+From: Jaegeuk Kim <jaegeuk@kernel.org>
+
+commit a4f843bd004d775cbb360cd375969b8a479568a9 upstream.
+
+syzbot hit the following crash on upstream commit
+83beed7b2b26f232d782127792dd0cd4362fdc41 (Fri Apr 20 17:56:32 2018 +0000)
+Merge branch 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/evalenti/linux-soc-thermal
+syzbot dashboard link: https://syzkaller.appspot.com/bug?extid=d154ec99402c6f628887
+
+C reproducer: https://syzkaller.appspot.com/x/repro.c?id=5414336294027264
+syzkaller reproducer: https://syzkaller.appspot.com/x/repro.syz?id=5471683234234368
+Raw console output: https://syzkaller.appspot.com/x/log.txt?id=5436660795834368
+Kernel config: https://syzkaller.appspot.com/x/.config?id=1808800213120130118
+compiler: gcc (GCC) 8.0.1 20180413 (experimental)
+
+IMPORTANT: if you fix the bug, please add the following tag to the commit:
+Reported-by: syzbot+d154ec99402c6f628887@syzkaller.appspotmail.com
+It will help syzbot understand when the bug is fixed. See footer for details.
+If you forward the report, please keep this part and the footer.
+
+F2FS-fs (loop0): Magic Mismatch, valid(0xf2f52010) - read(0x0)
+F2FS-fs (loop0): Can't find valid F2FS filesystem in 1th superblock
+F2FS-fs (loop0): invalid crc value
+------------[ cut here ]------------
+kernel BUG at fs/f2fs/node.c:1185!
+invalid opcode: 0000 [#1] SMP KASAN
+Dumping ftrace buffer:
+ (ftrace buffer empty)
+Modules linked in:
+CPU: 1 PID: 4549 Comm: syzkaller704305 Not tainted 4.17.0-rc1+ #10
+Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
+RIP: 0010:__get_node_page+0xb68/0x16e0 fs/f2fs/node.c:1185
+RSP: 0018:ffff8801d960e820 EFLAGS: 00010293
+RAX: ffff8801d88205c0 RBX: 0000000000000003 RCX: ffffffff82f6cc06
+RDX: 0000000000000000 RSI: ffffffff82f6d5e8 RDI: 0000000000000004
+RBP: ffff8801d960ec30 R08: ffff8801d88205c0 R09: ffffed003b5e46c2
+R10: 0000000000000003 R11: 0000000000000003 R12: ffff8801a86e00c0
+R13: 0000000000000001 R14: ffff8801a86e0530 R15: ffff8801d9745240
+FS: 000000000072c880(0000) GS:ffff8801daf00000(0000) knlGS:0000000000000000
+CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
+CR2: 00007f3d403209b8 CR3: 00000001d8f3f000 CR4: 00000000001406e0
+DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
+DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
+Call Trace:
+ get_node_page fs/f2fs/node.c:1237 [inline]
+ truncate_xattr_node+0x152/0x2e0 fs/f2fs/node.c:1014
+ remove_inode_page+0x200/0xaf0 fs/f2fs/node.c:1039
+ f2fs_evict_inode+0xe86/0x1710 fs/f2fs/inode.c:547
+ evict+0x4a6/0x960 fs/inode.c:557
+ iput_final fs/inode.c:1519 [inline]
+ iput+0x62d/0xa80 fs/inode.c:1545
+ f2fs_fill_super+0x5f4e/0x7bf0 fs/f2fs/super.c:2849
+ mount_bdev+0x30c/0x3e0 fs/super.c:1164
+ f2fs_mount+0x34/0x40 fs/f2fs/super.c:3020
+ mount_fs+0xae/0x328 fs/super.c:1267
+ vfs_kern_mount.part.34+0xd4/0x4d0 fs/namespace.c:1037
+ vfs_kern_mount fs/namespace.c:1027 [inline]
+ do_new_mount fs/namespace.c:2518 [inline]
+ do_mount+0x564/0x3070 fs/namespace.c:2848
+ ksys_mount+0x12d/0x140 fs/namespace.c:3064
+ __do_sys_mount fs/namespace.c:3078 [inline]
+ __se_sys_mount fs/namespace.c:3075 [inline]
+ __x64_sys_mount+0xbe/0x150 fs/namespace.c:3075
+ do_syscall_64+0x1b1/0x800 arch/x86/entry/common.c:287
+ entry_SYSCALL_64_after_hwframe+0x49/0xbe
+RIP: 0033:0x443dea
+RSP: 002b:00007ffcc7882368 EFLAGS: 00000297 ORIG_RAX: 00000000000000a5
+RAX: ffffffffffffffda RBX: 0000000020000c00 RCX: 0000000000443dea
+RDX: 0000000020000000 RSI: 0000000020000100 RDI: 00007ffcc7882370
+RBP: 0000000000000003 R08: 0000000020016a00 R09: 000000000000000a
+R10: 0000000000000000 R11: 0000000000000297 R12: 0000000000000004
+R13: 0000000000402ce0 R14: 0000000000000000 R15: 0000000000000000
+RIP: __get_node_page+0xb68/0x16e0 fs/f2fs/node.c:1185 RSP: ffff8801d960e820
+---[ end trace 4edbeb71f002bb76 ]---
+
+Reported-and-tested-by: syzbot+d154ec99402c6f628887@syzkaller.appspotmail.com
+Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ fs/f2fs/f2fs.h | 13 +------------
+ fs/f2fs/inode.c | 13 ++++++-------
+ fs/f2fs/node.c | 21 +++++++++++++++++++--
+ 3 files changed, 26 insertions(+), 21 deletions(-)
+
+--- a/fs/f2fs/f2fs.h
++++ b/fs/f2fs/f2fs.h
+@@ -1471,18 +1471,6 @@ static inline bool __exist_node_summarie
+ }
+
+ /*
+- * Check whether the given nid is within node id range.
+- */
+-static inline int check_nid_range(struct f2fs_sb_info *sbi, nid_t nid)
+-{
+- if (unlikely(nid < F2FS_ROOT_INO(sbi)))
+- return -EINVAL;
+- if (unlikely(nid >= NM_I(sbi)->max_nid))
+- return -EINVAL;
+- return 0;
+-}
+-
+-/*
+ * Check whether the inode has blocks or not
+ */
+ static inline int F2FS_HAS_BLOCKS(struct inode *inode)
+@@ -2470,6 +2458,7 @@ f2fs_hash_t f2fs_dentry_hash(const struc
+ struct dnode_of_data;
+ struct node_info;
+
++int check_nid_range(struct f2fs_sb_info *sbi, nid_t nid);
+ bool available_free_memory(struct f2fs_sb_info *sbi, int type);
+ int need_dentry_mark(struct f2fs_sb_info *sbi, nid_t nid);
+ bool is_checkpointed_node(struct f2fs_sb_info *sbi, nid_t nid);
+--- a/fs/f2fs/inode.c
++++ b/fs/f2fs/inode.c
+@@ -188,12 +188,8 @@ static int do_read_inode(struct inode *i
+ projid_t i_projid;
+
+ /* Check if ino is within scope */
+- if (check_nid_range(sbi, inode->i_ino)) {
+- f2fs_msg(inode->i_sb, KERN_ERR, "bad inode number: %lu",
+- (unsigned long) inode->i_ino);
+- WARN_ON(1);
++ if (check_nid_range(sbi, inode->i_ino))
+ return -EINVAL;
+- }
+
+ node_page = get_node_page(sbi, inode->i_ino);
+ if (IS_ERR(node_page))
+@@ -538,8 +534,11 @@ no_delete:
+ alloc_nid_failed(sbi, inode->i_ino);
+ clear_inode_flag(inode, FI_FREE_NID);
+ } else {
+- f2fs_bug_on(sbi, err &&
+- !exist_written_data(sbi, inode->i_ino, ORPHAN_INO));
++ /*
++ * If xattr nid is corrupted, we can reach out error condition,
++ * err & !exist_written_data(sbi, inode->i_ino, ORPHAN_INO)).
++ * In that case, check_nid_range() is enough to give a clue.
++ */
+ }
+ out_clear:
+ fscrypt_put_encryption_info(inode, NULL);
+--- a/fs/f2fs/node.c
++++ b/fs/f2fs/node.c
+@@ -29,6 +29,21 @@ static struct kmem_cache *nat_entry_slab
+ static struct kmem_cache *free_nid_slab;
+ static struct kmem_cache *nat_entry_set_slab;
+
++/*
++ * Check whether the given nid is within node id range.
++ */
++int check_nid_range(struct f2fs_sb_info *sbi, nid_t nid)
++{
++ if (unlikely(nid < F2FS_ROOT_INO(sbi) || nid >= NM_I(sbi)->max_nid)) {
++ set_sbi_flag(sbi, SBI_NEED_FSCK);
++ f2fs_msg(sbi->sb, KERN_WARNING,
++ "%s: out-of-range nid=%x, run fsck to fix.",
++ __func__, nid);
++ return -EINVAL;
++ }
++ return 0;
++}
++
+ bool available_free_memory(struct f2fs_sb_info *sbi, int type)
+ {
+ struct f2fs_nm_info *nm_i = NM_I(sbi);
+@@ -1122,7 +1137,8 @@ void ra_node_page(struct f2fs_sb_info *s
+
+ if (!nid)
+ return;
+- f2fs_bug_on(sbi, check_nid_range(sbi, nid));
++ if (check_nid_range(sbi, nid))
++ return;
+
+ rcu_read_lock();
+ apage = radix_tree_lookup(&NODE_MAPPING(sbi)->page_tree, nid);
+@@ -1146,7 +1162,8 @@ static struct page *__get_node_page(stru
+
+ if (!nid)
+ return ERR_PTR(-ENOENT);
+- f2fs_bug_on(sbi, check_nid_range(sbi, nid));
++ if (check_nid_range(sbi, nid))
++ return ERR_PTR(-EINVAL);
+ repeat:
+ page = f2fs_grab_cache_page(NODE_MAPPING(sbi), nid, false);
+ if (!page)
--- /dev/null
+From b2ca374f33bd33fd822eb871876e4888cf79dc97 Mon Sep 17 00:00:00 2001
+From: Jaegeuk Kim <jaegeuk@kernel.org>
+Date: Tue, 24 Apr 2018 15:44:16 -0600
+Subject: f2fs: sanity check on sit entry
+
+From: Jaegeuk Kim <jaegeuk@kernel.org>
+
+commit b2ca374f33bd33fd822eb871876e4888cf79dc97 upstream.
+
+syzbot hit the following crash on upstream commit
+87ef12027b9b1dd0e0b12cf311fbcb19f9d92539 (Wed Apr 18 19:48:17 2018 +0000)
+Merge tag 'ceph-for-4.17-rc2' of git://github.com/ceph/ceph-client
+syzbot dashboard link: https://syzkaller.appspot.com/bug?extid=83699adeb2d13579c31e
+
+C reproducer: https://syzkaller.appspot.com/x/repro.c?id=5805208181407744
+syzkaller reproducer: https://syzkaller.appspot.com/x/repro.syz?id=6005073343676416
+Raw console output: https://syzkaller.appspot.com/x/log.txt?id=6555047731134464
+Kernel config: https://syzkaller.appspot.com/x/.config?id=1808800213120130118
+compiler: gcc (GCC) 8.0.1 20180413 (experimental)
+
+IMPORTANT: if you fix the bug, please add the following tag to the commit:
+Reported-by: syzbot+83699adeb2d13579c31e@syzkaller.appspotmail.com
+It will help syzbot understand when the bug is fixed. See footer for details.
+If you forward the report, please keep this part and the footer.
+
+F2FS-fs (loop0): Magic Mismatch, valid(0xf2f52010) - read(0x0)
+F2FS-fs (loop0): Can't find valid F2FS filesystem in 1th superblock
+F2FS-fs (loop0): invalid crc value
+BUG: unable to handle kernel paging request at ffffed006b2a50c0
+PGD 21ffee067 P4D 21ffee067 PUD 21fbeb067 PMD 0
+Oops: 0000 [#1] SMP KASAN
+Dumping ftrace buffer:
+ (ftrace buffer empty)
+Modules linked in:
+CPU: 0 PID: 4514 Comm: syzkaller989480 Not tainted 4.17.0-rc1+ #8
+Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
+RIP: 0010:build_sit_entries fs/f2fs/segment.c:3653 [inline]
+RIP: 0010:build_segment_manager+0x7ef7/0xbf70 fs/f2fs/segment.c:3852
+RSP: 0018:ffff8801b102e5b0 EFLAGS: 00010a06
+RAX: 1ffff1006b2a50c0 RBX: 0000000000000004 RCX: 0000000000000001
+RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffff8801ac74243e
+RBP: ffff8801b102f410 R08: ffff8801acbd46c0 R09: fffffbfff14d9af8
+R10: fffffbfff14d9af8 R11: ffff8801acbd46c0 R12: ffff8801ac742a80
+R13: ffff8801d9519100 R14: dffffc0000000000 R15: ffff880359528600
+FS: 0000000001e04880(0000) GS:ffff8801dae00000(0000) knlGS:0000000000000000
+CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
+CR2: ffffed006b2a50c0 CR3: 00000001ac6ac000 CR4: 00000000001406f0
+DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
+DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
+Call Trace:
+ f2fs_fill_super+0x4095/0x7bf0 fs/f2fs/super.c:2803
+ mount_bdev+0x30c/0x3e0 fs/super.c:1165
+ f2fs_mount+0x34/0x40 fs/f2fs/super.c:3020
+ mount_fs+0xae/0x328 fs/super.c:1268
+ vfs_kern_mount.part.34+0xd4/0x4d0 fs/namespace.c:1037
+ vfs_kern_mount fs/namespace.c:1027 [inline]
+ do_new_mount fs/namespace.c:2517 [inline]
+ do_mount+0x564/0x3070 fs/namespace.c:2847
+ ksys_mount+0x12d/0x140 fs/namespace.c:3063
+ __do_sys_mount fs/namespace.c:3077 [inline]
+ __se_sys_mount fs/namespace.c:3074 [inline]
+ __x64_sys_mount+0xbe/0x150 fs/namespace.c:3074
+ do_syscall_64+0x1b1/0x800 arch/x86/entry/common.c:287
+ entry_SYSCALL_64_after_hwframe+0x49/0xbe
+RIP: 0033:0x443d6a
+RSP: 002b:00007ffd312813c8 EFLAGS: 00000297 ORIG_RAX: 00000000000000a5
+RAX: ffffffffffffffda RBX: 0000000020000c00 RCX: 0000000000443d6a
+RDX: 0000000020000000 RSI: 0000000020000100 RDI: 00007ffd312813d0
+RBP: 0000000000000003 R08: 0000000020016a00 R09: 000000000000000a
+R10: 0000000000000000 R11: 0000000000000297 R12: 0000000000000004
+R13: 0000000000402c60 R14: 0000000000000000 R15: 0000000000000000
+RIP: build_sit_entries fs/f2fs/segment.c:3653 [inline] RSP: ffff8801b102e5b0
+RIP: build_segment_manager+0x7ef7/0xbf70 fs/f2fs/segment.c:3852 RSP: ffff8801b102e5b0
+CR2: ffffed006b2a50c0
+---[ end trace a2034989e196ff17 ]---
+
+Reported-and-tested-by: syzbot+83699adeb2d13579c31e@syzkaller.appspotmail.com
+Reviewed-by: Chao Yu <yuchao0@huawei.com>
+Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ fs/f2fs/segment.c | 9 +++++++++
+ 1 file changed, 9 insertions(+)
+
+--- a/fs/f2fs/segment.c
++++ b/fs/f2fs/segment.c
+@@ -3293,6 +3293,15 @@ static void build_sit_entries(struct f2f
+ unsigned int old_valid_blocks;
+
+ start = le32_to_cpu(segno_in_journal(journal, i));
++ if (start >= MAIN_SEGS(sbi)) {
++ f2fs_msg(sbi->sb, KERN_ERR,
++ "Wrong journal entry on segno %u",
++ start);
++ set_sbi_flag(sbi, SBI_NEED_FSCK);
++ err = -EINVAL;
++ break;
++ }
++
+ se = &sit_i->sentries[start];
+ sit = sit_in_journal(journal, i);
+
--- /dev/null
+From 0447378a4a793da008451fad50bc0f93e9675ae6 Mon Sep 17 00:00:00 2001
+From: Marc Orr <marcorr@google.com>
+Date: Wed, 20 Jun 2018 17:21:29 -0700
+Subject: kvm: vmx: Nested VM-entry prereqs for event inj.
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+From: Marc Orr <marcorr@google.com>
+
+commit 0447378a4a793da008451fad50bc0f93e9675ae6 upstream.
+
+This patch extends the checks done prior to a nested VM entry.
+Specifically, it extends the check_vmentry_prereqs function with checks
+for fields relevant to the VM-entry event injection information, as
+described in the Intel SDM, volume 3.
+
+This patch is motivated by a syzkaller bug, where a bad VM-entry
+interruption information field is generated in the VMCS02, which causes
+the nested VM launch to fail. Then, KVM fails to resume L1.
+
+While KVM should be improved to correctly resume L1 execution after a
+failed nested launch, this change is justified because the existing code
+to resume L1 is flaky/ad-hoc and the test coverage for resuming L1 is
+sparse.
+
+Reported-by: syzbot <syzkaller@googlegroups.com>
+Signed-off-by: Marc Orr <marcorr@google.com>
+[Removed comment whose parts were describing previous revisions and the
+ rest was obvious from function/variable naming. - Radim]
+Signed-off-by: Radim Krčmář <rkrcmar@redhat.com>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ arch/x86/include/asm/vmx.h | 3 ++
+ arch/x86/kvm/vmx.c | 67 +++++++++++++++++++++++++++++++++++++++++++++
+ arch/x86/kvm/x86.h | 9 ++++++
+ 3 files changed, 79 insertions(+)
+
+--- a/arch/x86/include/asm/vmx.h
++++ b/arch/x86/include/asm/vmx.h
+@@ -114,6 +114,7 @@
+ #define VMX_MISC_PREEMPTION_TIMER_RATE_MASK 0x0000001f
+ #define VMX_MISC_SAVE_EFER_LMA 0x00000020
+ #define VMX_MISC_ACTIVITY_HLT 0x00000040
++#define VMX_MISC_ZERO_LEN_INS 0x40000000
+
+ /* VMFUNC functions */
+ #define VMX_VMFUNC_EPTP_SWITCHING 0x00000001
+@@ -349,11 +350,13 @@ enum vmcs_field {
+ #define VECTORING_INFO_VALID_MASK INTR_INFO_VALID_MASK
+
+ #define INTR_TYPE_EXT_INTR (0 << 8) /* external interrupt */
++#define INTR_TYPE_RESERVED (1 << 8) /* reserved */
+ #define INTR_TYPE_NMI_INTR (2 << 8) /* NMI */
+ #define INTR_TYPE_HARD_EXCEPTION (3 << 8) /* processor exception */
+ #define INTR_TYPE_SOFT_INTR (4 << 8) /* software interrupt */
+ #define INTR_TYPE_PRIV_SW_EXCEPTION (5 << 8) /* ICE breakpoint - undocumented */
+ #define INTR_TYPE_SOFT_EXCEPTION (6 << 8) /* software exception */
++#define INTR_TYPE_OTHER_EVENT (7 << 8) /* other event */
+
+ /* GUEST_INTERRUPTIBILITY_INFO flags. */
+ #define GUEST_INTR_STATE_STI 0x00000001
+--- a/arch/x86/kvm/vmx.c
++++ b/arch/x86/kvm/vmx.c
+@@ -1350,6 +1350,17 @@ static inline unsigned nested_cpu_vmx_mi
+ return vmx_misc_cr3_count(to_vmx(vcpu)->nested.nested_vmx_misc_low);
+ }
+
++static inline bool nested_cpu_has_zero_length_injection(struct kvm_vcpu *vcpu)
++{
++ return to_vmx(vcpu)->nested.msrs.misc_low & VMX_MISC_ZERO_LEN_INS;
++}
++
++static inline bool nested_cpu_supports_monitor_trap_flag(struct kvm_vcpu *vcpu)
++{
++ return to_vmx(vcpu)->nested.msrs.procbased_ctls_high &
++ CPU_BASED_MONITOR_TRAP_FLAG;
++}
++
+ static inline bool nested_cpu_has(struct vmcs12 *vmcs12, u32 bit)
+ {
+ return vmcs12->cpu_based_vm_exec_control & bit;
+@@ -11024,6 +11035,62 @@ static int check_vmentry_prereqs(struct
+ !nested_cr3_valid(vcpu, vmcs12->host_cr3))
+ return VMXERR_ENTRY_INVALID_HOST_STATE_FIELD;
+
++ /*
++ * From the Intel SDM, volume 3:
++ * Fields relevant to VM-entry event injection must be set properly.
++ * These fields are the VM-entry interruption-information field, the
++ * VM-entry exception error code, and the VM-entry instruction length.
++ */
++ if (vmcs12->vm_entry_intr_info_field & INTR_INFO_VALID_MASK) {
++ u32 intr_info = vmcs12->vm_entry_intr_info_field;
++ u8 vector = intr_info & INTR_INFO_VECTOR_MASK;
++ u32 intr_type = intr_info & INTR_INFO_INTR_TYPE_MASK;
++ bool has_error_code = intr_info & INTR_INFO_DELIVER_CODE_MASK;
++ bool should_have_error_code;
++ bool urg = nested_cpu_has2(vmcs12,
++ SECONDARY_EXEC_UNRESTRICTED_GUEST);
++ bool prot_mode = !urg || vmcs12->guest_cr0 & X86_CR0_PE;
++
++ /* VM-entry interruption-info field: interruption type */
++ if (intr_type == INTR_TYPE_RESERVED ||
++ (intr_type == INTR_TYPE_OTHER_EVENT &&
++ !nested_cpu_supports_monitor_trap_flag(vcpu)))
++ return VMXERR_ENTRY_INVALID_CONTROL_FIELD;
++
++ /* VM-entry interruption-info field: vector */
++ if ((intr_type == INTR_TYPE_NMI_INTR && vector != NMI_VECTOR) ||
++ (intr_type == INTR_TYPE_HARD_EXCEPTION && vector > 31) ||
++ (intr_type == INTR_TYPE_OTHER_EVENT && vector != 0))
++ return VMXERR_ENTRY_INVALID_CONTROL_FIELD;
++
++ /* VM-entry interruption-info field: deliver error code */
++ should_have_error_code =
++ intr_type == INTR_TYPE_HARD_EXCEPTION && prot_mode &&
++ x86_exception_has_error_code(vector);
++ if (has_error_code != should_have_error_code)
++ return VMXERR_ENTRY_INVALID_CONTROL_FIELD;
++
++ /* VM-entry exception error code */
++ if (has_error_code &&
++ vmcs12->vm_entry_exception_error_code & GENMASK(31, 15))
++ return VMXERR_ENTRY_INVALID_CONTROL_FIELD;
++
++ /* VM-entry interruption-info field: reserved bits */
++ if (intr_info & INTR_INFO_RESVD_BITS_MASK)
++ return VMXERR_ENTRY_INVALID_CONTROL_FIELD;
++
++ /* VM-entry instruction length */
++ switch (intr_type) {
++ case INTR_TYPE_SOFT_EXCEPTION:
++ case INTR_TYPE_SOFT_INTR:
++ case INTR_TYPE_PRIV_SW_EXCEPTION:
++ if ((vmcs12->vm_entry_instruction_len > 15) ||
++ (vmcs12->vm_entry_instruction_len == 0 &&
++ !nested_cpu_has_zero_length_injection(vcpu)))
++ return VMXERR_ENTRY_INVALID_CONTROL_FIELD;
++ }
++ }
++
+ return 0;
+ }
+
+--- a/arch/x86/kvm/x86.h
++++ b/arch/x86/kvm/x86.h
+@@ -74,6 +74,15 @@ static inline bool is_la57_mode(struct k
+ #endif
+ }
+
++static inline bool x86_exception_has_error_code(unsigned int vector)
++{
++ static u32 exception_has_error_code = BIT(DF_VECTOR) | BIT(TS_VECTOR) |
++ BIT(NP_VECTOR) | BIT(SS_VECTOR) | BIT(GP_VECTOR) |
++ BIT(PF_VECTOR) | BIT(AC_VECTOR);
++
++ return (1U << vector) & exception_has_error_code;
++}
++
+ static inline bool mmu_is_nested(struct kvm_vcpu *vcpu)
+ {
+ return vcpu->arch.walk_mmu == &vcpu->arch.nested_mmu;
--- /dev/null
+From d2ac838e4cd7e5e9891ecc094d626734b0245c99 Mon Sep 17 00:00:00 2001
+From: Theodore Ts'o <tytso@mit.edu>
+Date: Mon, 7 May 2018 11:37:58 -0400
+Subject: loop: add recursion validation to LOOP_CHANGE_FD
+
+From: Theodore Ts'o <tytso@mit.edu>
+
+commit d2ac838e4cd7e5e9891ecc094d626734b0245c99 upstream.
+
+Refactor the validation code used in LOOP_SET_FD so it is also used in
+LOOP_CHANGE_FD. Otherwise it is possible to construct a set of loop
+devices that all refer to each other. This can lead to a infinite
+loop in starting with "while (is_loop_device(f)) .." in loop_set_fd().
+
+Fix this by refactoring out the validation code and using it for
+LOOP_CHANGE_FD as well as LOOP_SET_FD.
+
+Reported-by: syzbot+4349872271ece473a7c91190b68b4bac7c5dbc87@syzkaller.appspotmail.com
+Reported-by: syzbot+40bd32c4d9a3cc12a339@syzkaller.appspotmail.com
+Reported-by: syzbot+769c54e66f994b041be7@syzkaller.appspotmail.com
+Reported-by: syzbot+0a89a9ce473936c57065@syzkaller.appspotmail.com
+Signed-off-by: Theodore Ts'o <tytso@mit.edu>
+Signed-off-by: Jens Axboe <axboe@kernel.dk>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ drivers/block/loop.c | 68 ++++++++++++++++++++++++++++-----------------------
+ 1 file changed, 38 insertions(+), 30 deletions(-)
+
+--- a/drivers/block/loop.c
++++ b/drivers/block/loop.c
+@@ -617,6 +617,36 @@ static void loop_reread_partitions(struc
+ __func__, lo->lo_number, lo->lo_file_name, rc);
+ }
+
++static inline int is_loop_device(struct file *file)
++{
++ struct inode *i = file->f_mapping->host;
++
++ return i && S_ISBLK(i->i_mode) && MAJOR(i->i_rdev) == LOOP_MAJOR;
++}
++
++static int loop_validate_file(struct file *file, struct block_device *bdev)
++{
++ struct inode *inode = file->f_mapping->host;
++ struct file *f = file;
++
++ /* Avoid recursion */
++ while (is_loop_device(f)) {
++ struct loop_device *l;
++
++ if (f->f_mapping->host->i_bdev == bdev)
++ return -EBADF;
++
++ l = f->f_mapping->host->i_bdev->bd_disk->private_data;
++ if (l->lo_state == Lo_unbound) {
++ return -EINVAL;
++ }
++ f = l->lo_backing_file;
++ }
++ if (!S_ISREG(inode->i_mode) && !S_ISBLK(inode->i_mode))
++ return -EINVAL;
++ return 0;
++}
++
+ /*
+ * loop_change_fd switched the backing store of a loopback device to
+ * a new file. This is useful for operating system installers to free up
+@@ -646,14 +676,15 @@ static int loop_change_fd(struct loop_de
+ if (!file)
+ goto out;
+
++ error = loop_validate_file(file, bdev);
++ if (error)
++ goto out_putf;
++
+ inode = file->f_mapping->host;
+ old_file = lo->lo_backing_file;
+
+ error = -EINVAL;
+
+- if (!S_ISREG(inode->i_mode) && !S_ISBLK(inode->i_mode))
+- goto out_putf;
+-
+ /* size of the new backing store needs to be the same */
+ if (get_loop_size(lo, file) != get_loop_size(lo, old_file))
+ goto out_putf;
+@@ -679,13 +710,6 @@ static int loop_change_fd(struct loop_de
+ return error;
+ }
+
+-static inline int is_loop_device(struct file *file)
+-{
+- struct inode *i = file->f_mapping->host;
+-
+- return i && S_ISBLK(i->i_mode) && MAJOR(i->i_rdev) == LOOP_MAJOR;
+-}
+-
+ /* loop sysfs attributes */
+
+ static ssize_t loop_attr_show(struct device *dev, char *page,
+@@ -850,7 +874,7 @@ static int loop_prepare_queue(struct loo
+ static int loop_set_fd(struct loop_device *lo, fmode_t mode,
+ struct block_device *bdev, unsigned int arg)
+ {
+- struct file *file, *f;
++ struct file *file;
+ struct inode *inode;
+ struct address_space *mapping;
+ int lo_flags = 0;
+@@ -869,29 +893,13 @@ static int loop_set_fd(struct loop_devic
+ if (lo->lo_state != Lo_unbound)
+ goto out_putf;
+
+- /* Avoid recursion */
+- f = file;
+- while (is_loop_device(f)) {
+- struct loop_device *l;
+-
+- if (f->f_mapping->host->i_bdev == bdev)
+- goto out_putf;
+-
+- l = f->f_mapping->host->i_bdev->bd_disk->private_data;
+- if (l->lo_state == Lo_unbound) {
+- error = -EINVAL;
+- goto out_putf;
+- }
+- f = l->lo_backing_file;
+- }
++ error = loop_validate_file(file, bdev);
++ if (error)
++ goto out_putf;
+
+ mapping = file->f_mapping;
+ inode = mapping->host;
+
+- error = -EINVAL;
+- if (!S_ISREG(inode->i_mode) && !S_ISBLK(inode->i_mode))
+- goto out_putf;
+-
+ if (!(file->f_mode & FMODE_WRITE) || !(mode & FMODE_WRITE) ||
+ !file->f_op->write_iter)
+ lo_flags |= LO_FLAGS_READ_ONLY;
--- /dev/null
+From d3349b6b3c373ac1fbfb040b810fcee5e2adc7e0 Mon Sep 17 00:00:00 2001
+From: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
+Date: Fri, 4 May 2018 10:58:09 -0600
+Subject: loop: remember whether sysfs_create_group() was done
+
+From: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
+
+commit d3349b6b3c373ac1fbfb040b810fcee5e2adc7e0 upstream.
+
+syzbot is hitting WARN() triggered by memory allocation fault
+injection [1] because loop module is calling sysfs_remove_group()
+when sysfs_create_group() failed.
+Fix this by remembering whether sysfs_create_group() succeeded.
+
+[1] https://syzkaller.appspot.com/bug?id=3f86c0edf75c86d2633aeb9dd69eccc70bc7e90b
+
+Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
+Reported-by: syzbot <syzbot+9f03168400f56df89dbc6f1751f4458fe739ff29@syzkaller.appspotmail.com>
+Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+Renamed sysfs_ready -> sysfs_inited.
+
+Signed-off-by: Jens Axboe <axboe@kernel.dk>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ drivers/block/loop.c | 11 ++++++-----
+ drivers/block/loop.h | 1 +
+ 2 files changed, 7 insertions(+), 5 deletions(-)
+
+--- a/drivers/block/loop.c
++++ b/drivers/block/loop.c
+@@ -806,16 +806,17 @@ static struct attribute_group loop_attri
+ .attrs= loop_attrs,
+ };
+
+-static int loop_sysfs_init(struct loop_device *lo)
++static void loop_sysfs_init(struct loop_device *lo)
+ {
+- return sysfs_create_group(&disk_to_dev(lo->lo_disk)->kobj,
+- &loop_attribute_group);
++ lo->sysfs_inited = !sysfs_create_group(&disk_to_dev(lo->lo_disk)->kobj,
++ &loop_attribute_group);
+ }
+
+ static void loop_sysfs_exit(struct loop_device *lo)
+ {
+- sysfs_remove_group(&disk_to_dev(lo->lo_disk)->kobj,
+- &loop_attribute_group);
++ if (lo->sysfs_inited)
++ sysfs_remove_group(&disk_to_dev(lo->lo_disk)->kobj,
++ &loop_attribute_group);
+ }
+
+ static void loop_config_discard(struct loop_device *lo)
+--- a/drivers/block/loop.h
++++ b/drivers/block/loop.h
+@@ -58,6 +58,7 @@ struct loop_device {
+ struct kthread_worker worker;
+ struct task_struct *worker_task;
+ bool use_dio;
++ bool sysfs_inited;
+
+ struct request_queue *lo_queue;
+ struct blk_mq_tag_set tag_set;
--- /dev/null
+From ba062ebb2cd561d404e0fba8ee4b3f5ebce7cbfc Mon Sep 17 00:00:00 2001
+From: Eric Dumazet <edumazet@google.com>
+Date: Wed, 13 Jun 2018 09:13:39 -0700
+Subject: netfilter: nf_queue: augment nfqa_cfg_policy
+
+From: Eric Dumazet <edumazet@google.com>
+
+commit ba062ebb2cd561d404e0fba8ee4b3f5ebce7cbfc upstream.
+
+Three attributes are currently not verified, thus can trigger KMSAN
+warnings such as :
+
+BUG: KMSAN: uninit-value in __arch_swab32 arch/x86/include/uapi/asm/swab.h:10 [inline]
+BUG: KMSAN: uninit-value in __fswab32 include/uapi/linux/swab.h:59 [inline]
+BUG: KMSAN: uninit-value in nfqnl_recv_config+0x939/0x17d0 net/netfilter/nfnetlink_queue.c:1268
+CPU: 1 PID: 4521 Comm: syz-executor120 Not tainted 4.17.0+ #5
+Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
+Call Trace:
+ __dump_stack lib/dump_stack.c:77 [inline]
+ dump_stack+0x185/0x1d0 lib/dump_stack.c:113
+ kmsan_report+0x188/0x2a0 mm/kmsan/kmsan.c:1117
+ __msan_warning_32+0x70/0xc0 mm/kmsan/kmsan_instr.c:620
+ __arch_swab32 arch/x86/include/uapi/asm/swab.h:10 [inline]
+ __fswab32 include/uapi/linux/swab.h:59 [inline]
+ nfqnl_recv_config+0x939/0x17d0 net/netfilter/nfnetlink_queue.c:1268
+ nfnetlink_rcv_msg+0xb2e/0xc80 net/netfilter/nfnetlink.c:212
+ netlink_rcv_skb+0x37e/0x600 net/netlink/af_netlink.c:2448
+ nfnetlink_rcv+0x2fe/0x680 net/netfilter/nfnetlink.c:513
+ netlink_unicast_kernel net/netlink/af_netlink.c:1310 [inline]
+ netlink_unicast+0x1680/0x1750 net/netlink/af_netlink.c:1336
+ netlink_sendmsg+0x104f/0x1350 net/netlink/af_netlink.c:1901
+ sock_sendmsg_nosec net/socket.c:629 [inline]
+ sock_sendmsg net/socket.c:639 [inline]
+ ___sys_sendmsg+0xec8/0x1320 net/socket.c:2117
+ __sys_sendmsg net/socket.c:2155 [inline]
+ __do_sys_sendmsg net/socket.c:2164 [inline]
+ __se_sys_sendmsg net/socket.c:2162 [inline]
+ __x64_sys_sendmsg+0x331/0x460 net/socket.c:2162
+ do_syscall_64+0x15b/0x230 arch/x86/entry/common.c:287
+ entry_SYSCALL_64_after_hwframe+0x44/0xa9
+RIP: 0033:0x43fd59
+RSP: 002b:00007ffde0e30d28 EFLAGS: 00000213 ORIG_RAX: 000000000000002e
+RAX: ffffffffffffffda RBX: 00000000004002c8 RCX: 000000000043fd59
+RDX: 0000000000000000 RSI: 0000000020000080 RDI: 0000000000000003
+RBP: 00000000006ca018 R08: 00000000004002c8 R09: 00000000004002c8
+R10: 00000000004002c8 R11: 0000000000000213 R12: 0000000000401680
+R13: 0000000000401710 R14: 0000000000000000 R15: 0000000000000000
+
+Uninit was created at:
+ kmsan_save_stack_with_flags mm/kmsan/kmsan.c:279 [inline]
+ kmsan_internal_poison_shadow+0xb8/0x1b0 mm/kmsan/kmsan.c:189
+ kmsan_kmalloc+0x94/0x100 mm/kmsan/kmsan.c:315
+ kmsan_slab_alloc+0x10/0x20 mm/kmsan/kmsan.c:322
+ slab_post_alloc_hook mm/slab.h:446 [inline]
+ slab_alloc_node mm/slub.c:2753 [inline]
+ __kmalloc_node_track_caller+0xb35/0x11b0 mm/slub.c:4395
+ __kmalloc_reserve net/core/skbuff.c:138 [inline]
+ __alloc_skb+0x2cb/0x9e0 net/core/skbuff.c:206
+ alloc_skb include/linux/skbuff.h:988 [inline]
+ netlink_alloc_large_skb net/netlink/af_netlink.c:1182 [inline]
+ netlink_sendmsg+0x76e/0x1350 net/netlink/af_netlink.c:1876
+ sock_sendmsg_nosec net/socket.c:629 [inline]
+ sock_sendmsg net/socket.c:639 [inline]
+ ___sys_sendmsg+0xec8/0x1320 net/socket.c:2117
+ __sys_sendmsg net/socket.c:2155 [inline]
+ __do_sys_sendmsg net/socket.c:2164 [inline]
+ __se_sys_sendmsg net/socket.c:2162 [inline]
+ __x64_sys_sendmsg+0x331/0x460 net/socket.c:2162
+ do_syscall_64+0x15b/0x230 arch/x86/entry/common.c:287
+ entry_SYSCALL_64_after_hwframe+0x44/0xa9
+
+Fixes: fdb694a01f1f ("netfilter: Add fail-open support")
+Fixes: 829e17a1a602 ("[NETFILTER]: nfnetlink_queue: allow changing queue length through netlink")
+Signed-off-by: Eric Dumazet <edumazet@google.com>
+Reported-by: syzbot <syzkaller@googlegroups.com>
+Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ net/netfilter/nfnetlink_queue.c | 3 +++
+ 1 file changed, 3 insertions(+)
+
+--- a/net/netfilter/nfnetlink_queue.c
++++ b/net/netfilter/nfnetlink_queue.c
+@@ -1228,6 +1228,9 @@ static int nfqnl_recv_unsupp(struct net
+ static const struct nla_policy nfqa_cfg_policy[NFQA_CFG_MAX+1] = {
+ [NFQA_CFG_CMD] = { .len = sizeof(struct nfqnl_msg_config_cmd) },
+ [NFQA_CFG_PARAMS] = { .len = sizeof(struct nfqnl_msg_config_params) },
++ [NFQA_CFG_QUEUE_MAXLEN] = { .type = NLA_U32 },
++ [NFQA_CFG_MASK] = { .type = NLA_U32 },
++ [NFQA_CFG_FLAGS] = { .type = NLA_U32 },
+ };
+
+ static const struct nf_queue_handler nfqh = {
--- /dev/null
+From c568503ef02030f169c9e19204def610a3510918 Mon Sep 17 00:00:00 2001
+From: Florian Westphal <fw@strlen.de>
+Date: Thu, 7 Jun 2018 21:34:43 +0200
+Subject: netfilter: x_tables: initialise match/target check parameter struct
+
+From: Florian Westphal <fw@strlen.de>
+
+commit c568503ef02030f169c9e19204def610a3510918 upstream.
+
+syzbot reports following splat:
+
+BUG: KMSAN: uninit-value in ebt_stp_mt_check+0x24b/0x450
+ net/bridge/netfilter/ebt_stp.c:162
+ ebt_stp_mt_check+0x24b/0x450 net/bridge/netfilter/ebt_stp.c:162
+ xt_check_match+0x1438/0x1650 net/netfilter/x_tables.c:506
+ ebt_check_match net/bridge/netfilter/ebtables.c:372 [inline]
+ ebt_check_entry net/bridge/netfilter/ebtables.c:702 [inline]
+
+The uninitialised access is
+ xt_mtchk_param->nft_compat
+
+... which should be set to 0.
+Fix it by zeroing the struct beforehand, same for tgchk.
+
+ip(6)tables targetinfo uses c99-style initialiser, so no change
+needed there.
+
+Reported-by: syzbot+da4494182233c23a5fcf@syzkaller.appspotmail.com
+Fixes: 55917a21d0cc0 ("netfilter: x_tables: add context to know if extension runs from nft_compat")
+Signed-off-by: Florian Westphal <fw@strlen.de>
+Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ net/bridge/netfilter/ebtables.c | 2 ++
+ net/ipv4/netfilter/ip_tables.c | 1 +
+ net/ipv6/netfilter/ip6_tables.c | 1 +
+ 3 files changed, 4 insertions(+)
+
+--- a/net/bridge/netfilter/ebtables.c
++++ b/net/bridge/netfilter/ebtables.c
+@@ -696,6 +696,8 @@ ebt_check_entry(struct ebt_entry *e, str
+ }
+ i = 0;
+
++ memset(&mtpar, 0, sizeof(mtpar));
++ memset(&tgpar, 0, sizeof(tgpar));
+ mtpar.net = tgpar.net = net;
+ mtpar.table = tgpar.table = name;
+ mtpar.entryinfo = tgpar.entryinfo = e;
+--- a/net/ipv4/netfilter/ip_tables.c
++++ b/net/ipv4/netfilter/ip_tables.c
+@@ -541,6 +541,7 @@ find_check_entry(struct ipt_entry *e, st
+ return -ENOMEM;
+
+ j = 0;
++ memset(&mtpar, 0, sizeof(mtpar));
+ mtpar.net = net;
+ mtpar.table = name;
+ mtpar.entryinfo = &e->ip;
+--- a/net/ipv6/netfilter/ip6_tables.c
++++ b/net/ipv6/netfilter/ip6_tables.c
+@@ -561,6 +561,7 @@ find_check_entry(struct ip6t_entry *e, s
+ return -ENOMEM;
+
+ j = 0;
++ memset(&mtpar, 0, sizeof(mtpar));
+ mtpar.net = net;
+ mtpar.table = name;
+ mtpar.entryinfo = &e->ipv6;
--- /dev/null
+From 815c6704bf9f1c59f3a6be380a4032b9c57b12f1 Mon Sep 17 00:00:00 2001
+From: Keith Busch <keith.busch@intel.com>
+Date: Tue, 13 Feb 2018 05:44:44 -0700
+Subject: nvme-pci: Remap CMB SQ entries on every controller reset
+
+From: Keith Busch <keith.busch@intel.com>
+
+commit 815c6704bf9f1c59f3a6be380a4032b9c57b12f1 upstream.
+
+The controller memory buffer is remapped into a kernel address on each
+reset, but the driver was setting the submission queue base address
+only on the very first queue creation. The remapped address is likely to
+change after a reset, so accessing the old address will hit a kernel bug.
+
+This patch fixes that by setting the queue's CMB base address each time
+the queue is created.
+
+Fixes: f63572dff1421 ("nvme: unmap CMB and remove sysfs file in reset path")
+Reported-by: Christian Black <christian.d.black@intel.com>
+Cc: Jon Derrick <jonathan.derrick@intel.com>
+Cc: <stable@vger.kernel.org> # 4.9+
+Signed-off-by: Keith Busch <keith.busch@intel.com>
+Reviewed-by: Christoph Hellwig <hch@lst.de>
+Signed-off-by: Scott Bauer <scott.bauer@intel.com>
+Reviewed-by: Jon Derrick <jonathan.derrick@intel.com>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ drivers/nvme/host/pci.c | 27 ++++++++++++++++-----------
+ 1 file changed, 16 insertions(+), 11 deletions(-)
+
+--- a/drivers/nvme/host/pci.c
++++ b/drivers/nvme/host/pci.c
+@@ -1233,17 +1233,15 @@ static int nvme_cmb_qdepth(struct nvme_d
+ static int nvme_alloc_sq_cmds(struct nvme_dev *dev, struct nvme_queue *nvmeq,
+ int qid, int depth)
+ {
+- if (qid && dev->cmb && use_cmb_sqes && NVME_CMB_SQS(dev->cmbsz)) {
+- unsigned offset = (qid - 1) * roundup(SQ_SIZE(depth),
+- dev->ctrl.page_size);
+- nvmeq->sq_dma_addr = dev->cmb_bus_addr + offset;
+- nvmeq->sq_cmds_io = dev->cmb + offset;
+- } else {
+- nvmeq->sq_cmds = dma_alloc_coherent(dev->dev, SQ_SIZE(depth),
+- &nvmeq->sq_dma_addr, GFP_KERNEL);
+- if (!nvmeq->sq_cmds)
+- return -ENOMEM;
+- }
++
++ /* CMB SQEs will be mapped before creation */
++ if (qid && dev->cmb && use_cmb_sqes && NVME_CMB_SQS(dev->cmbsz))
++ return 0;
++
++ nvmeq->sq_cmds = dma_alloc_coherent(dev->dev, SQ_SIZE(depth),
++ &nvmeq->sq_dma_addr, GFP_KERNEL);
++ if (!nvmeq->sq_cmds)
++ return -ENOMEM;
+
+ return 0;
+ }
+@@ -1320,6 +1318,13 @@ static int nvme_create_queue(struct nvme
+ struct nvme_dev *dev = nvmeq->dev;
+ int result;
+
++ if (qid && dev->cmb && use_cmb_sqes && NVME_CMB_SQS(dev->cmbsz)) {
++ unsigned offset = (qid - 1) * roundup(SQ_SIZE(nvmeq->q_depth),
++ dev->ctrl.page_size);
++ nvmeq->sq_dma_addr = dev->cmb_bus_addr + offset;
++ nvmeq->sq_cmds_io = dev->cmb + offset;
++ }
++
+ nvmeq->cq_vector = qid - 1;
+ result = adapter_alloc_cq(dev, qid, nvmeq);
+ if (result < 0)
--- /dev/null
+From fc14eebfc20854a38fd9f1d93a42b1783dad4d17 Mon Sep 17 00:00:00 2001
+From: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
+Date: Sat, 26 May 2018 09:59:36 +0900
+Subject: PM / hibernate: Fix oops at snapshot_write()
+
+From: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
+
+commit fc14eebfc20854a38fd9f1d93a42b1783dad4d17 upstream.
+
+syzbot is reporting NULL pointer dereference at snapshot_write() [1].
+This is because data->handle is zero-cleared by ioctl(SNAPSHOT_FREE).
+Fix this by checking data_of(data->handle) != NULL before using it.
+
+[1] https://syzkaller.appspot.com/bug?id=828a3c71bd344a6de8b6a31233d51a72099f27fd
+
+Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
+Reported-by: syzbot <syzbot+ae590932da6e45d6564d@syzkaller.appspotmail.com>
+Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ kernel/power/user.c | 5 +++++
+ 1 file changed, 5 insertions(+)
+
+--- a/kernel/power/user.c
++++ b/kernel/power/user.c
+@@ -186,6 +186,11 @@ static ssize_t snapshot_write(struct fil
+ res = PAGE_SIZE - pg_offp;
+ }
+
++ if (!data_of(data->handle)) {
++ res = -EINVAL;
++ goto unlock;
++ }
++
+ res = simple_write_to_buffer(data_of(data->handle), res, &pg_offp,
+ buf, count);
+ if (res > 0)
--- /dev/null
+From 7a8690ed6f5346f6738971892205e91d39b6b901 Mon Sep 17 00:00:00 2001
+From: Leon Romanovsky <leonro@mellanox.com>
+Date: Wed, 23 May 2018 08:22:11 +0300
+Subject: RDMA/ucm: Mark UCM interface as BROKEN
+
+From: Leon Romanovsky <leonro@mellanox.com>
+
+commit 7a8690ed6f5346f6738971892205e91d39b6b901 upstream.
+
+In commit 357d23c811a7 ("Remove the obsolete libibcm library")
+in rdma-core [1], we removed obsolete library which used the
+/dev/infiniband/ucmX interface.
+
+Following multiple syzkaller reports about non-sanitized
+user input in the UCMA module, the short audit reveals the same
+issues in UCM module too.
+
+It is better to disable this interface in the kernel,
+before syzkaller team invests time and energy to harden
+this unused interface.
+
+[1] https://github.com/linux-rdma/rdma-core/pull/279
+
+Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
+Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ drivers/infiniband/Kconfig | 12 ++++++++++++
+ drivers/infiniband/core/Makefile | 4 ++--
+ 2 files changed, 14 insertions(+), 2 deletions(-)
+
+--- a/drivers/infiniband/Kconfig
++++ b/drivers/infiniband/Kconfig
+@@ -34,6 +34,18 @@ config INFINIBAND_USER_ACCESS
+ libibverbs, libibcm and a hardware driver library from
+ <http://www.openfabrics.org/git/>.
+
++config INFINIBAND_USER_ACCESS_UCM
++ bool "Userspace CM (UCM, DEPRECATED)"
++ depends on BROKEN
++ depends on INFINIBAND_USER_ACCESS
++ help
++ The UCM module has known security flaws, which no one is
++ interested to fix. The user-space part of this code was
++ dropped from the upstream a long time ago.
++
++ This option is DEPRECATED and planned to be removed.
++
++
+ config INFINIBAND_EXP_USER_ACCESS
+ bool "Allow experimental support for Infiniband ABI"
+ depends on INFINIBAND_USER_ACCESS
+--- a/drivers/infiniband/core/Makefile
++++ b/drivers/infiniband/core/Makefile
+@@ -5,8 +5,8 @@ user_access-$(CONFIG_INFINIBAND_ADDR_TRA
+ obj-$(CONFIG_INFINIBAND) += ib_core.o ib_cm.o iw_cm.o \
+ $(infiniband-y)
+ obj-$(CONFIG_INFINIBAND_USER_MAD) += ib_umad.o
+-obj-$(CONFIG_INFINIBAND_USER_ACCESS) += ib_uverbs.o ib_ucm.o \
+- $(user_access-y)
++obj-$(CONFIG_INFINIBAND_USER_ACCESS) += ib_uverbs.o $(user_access-y)
++obj-$(CONFIG_INFINIBAND_USER_ACCESS_UCM) += ib_ucm.o $(user_access-y)
+
+ ib_core-y := packer.o ud_header.o verbs.o cq.o rw.o sysfs.o \
+ device.o fmr_pool.o cache.o netlink.o \
i2c-tegra-fix-nack-error-handling.patch
iw_cxgb4-correctly-enforce-the-max-reg_mr-depth.patch
xen-setup-pv-irq-ops-vector-earlier.patch
+nvme-pci-remap-cmb-sq-entries-on-every-controller-reset.patch
+crypto-x86-salsa20-remove-x86-salsa20-implementations.patch
+uprobes-x86-remove-incorrect-warn_on-in-uprobe_init_insn.patch
+netfilter-nf_queue-augment-nfqa_cfg_policy.patch
+netfilter-x_tables-initialise-match-target-check-parameter-struct.patch
+loop-add-recursion-validation-to-loop_change_fd.patch
+pm-hibernate-fix-oops-at-snapshot_write.patch
+rdma-ucm-mark-ucm-interface-as-broken.patch
+loop-remember-whether-sysfs_create_group-was-done.patch
+kvm-vmx-nested-vm-entry-prereqs-for-event-inj.patch
+f2fs-give-message-and-set-need_fsck-given-broken-node-id.patch
+f2fs-sanity-check-on-sit-entry.patch
--- /dev/null
+From 90718e32e1dcc2479acfa208ccfc6442850b594c Mon Sep 17 00:00:00 2001
+From: Oleg Nesterov <oleg@redhat.com>
+Date: Fri, 18 May 2018 18:27:39 +0200
+Subject: uprobes/x86: Remove incorrect WARN_ON() in uprobe_init_insn()
+
+From: Oleg Nesterov <oleg@redhat.com>
+
+commit 90718e32e1dcc2479acfa208ccfc6442850b594c upstream.
+
+insn_get_length() has the side-effect of processing the entire instruction
+but only if it was decoded successfully, otherwise insn_complete() can fail
+and in this case we need to just return an error without warning.
+
+Reported-by: syzbot+30d675e3ca03c1c351e7@syzkaller.appspotmail.com
+Signed-off-by: Oleg Nesterov <oleg@redhat.com>
+Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
+Cc: Linus Torvalds <torvalds@linux-foundation.org>
+Cc: Peter Zijlstra <peterz@infradead.org>
+Cc: Thomas Gleixner <tglx@linutronix.de>
+Cc: syzkaller-bugs@googlegroups.com
+Link: https://lkml.kernel.org/lkml/20180518162739.GA5559@redhat.com
+Signed-off-by: Ingo Molnar <mingo@kernel.org>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ arch/x86/kernel/uprobes.c | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+--- a/arch/x86/kernel/uprobes.c
++++ b/arch/x86/kernel/uprobes.c
+@@ -290,7 +290,7 @@ static int uprobe_init_insn(struct arch_
+ insn_init(insn, auprobe->insn, sizeof(auprobe->insn), x86_64);
+ /* has the side-effect of processing the entire instruction */
+ insn_get_length(insn);
+- if (WARN_ON_ONCE(!insn_complete(insn)))
++ if (!insn_complete(insn))
+ return -ENOEXEC;
+
+ if (is_prefix_bad(insn))
projects / thirdparty / kernel / stable-queue.git / commitdiff
