Merge git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6

Pull crypto fixes from Herbert Xu:
 "This fixes a build problem with sha-mb with old toolchains and an
  implementation bug in the ctr(aes)/by8 branch of aesni-intel that's
  enabled when AVX is available"

* git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
  crypto: sha-mb - Add avx2_supported check.
  crypto: aesni - fix "by8" variant for 128 bit keys

2 files changed, 36 insertions, 12 deletions

diff --git a/arch/x86/crypto/Makefile b/arch/x86/crypto/Makefile
index fd0f848938cc..5a4a089e8b1f 100644
--- a/arch/x86/crypto/Makefile
+++ b/arch/x86/crypto/Makefile
@@ -26,7 +26,6 @@ obj-$(CONFIG_CRYPTO_GHASH_CLMUL_NI_INTEL) += ghash-clmulni-intel.o
 
 obj-$(CONFIG_CRYPTO_CRC32C_INTEL) += crc32c-intel.o
 obj-$(CONFIG_CRYPTO_SHA1_SSSE3) += sha1-ssse3.o
-obj-$(CONFIG_CRYPTO_SHA1_MB) += sha-mb/
 obj-$(CONFIG_CRYPTO_CRC32_PCLMUL) += crc32-pclmul.o
 obj-$(CONFIG_CRYPTO_SHA256_SSSE3) += sha256-ssse3.o
 obj-$(CONFIG_CRYPTO_SHA512_SSSE3) += sha512-ssse3.o
@@ -46,6 +45,7 @@ endif
 ifeq ($(avx2_supported),yes)
         obj-$(CONFIG_CRYPTO_CAMELLIA_AESNI_AVX2_X86_64) += camellia-aesni-avx2.o
         obj-$(CONFIG_CRYPTO_SERPENT_AVX2_X86_64) += serpent-avx2.o
+        obj-$(CONFIG_CRYPTO_SHA1_MB) += sha-mb/
 endif
 
 aes-i586-y := aes-i586-asm_32.o aes_glue.o
diff --git a/arch/x86/crypto/aes_ctrby8_avx-x86_64.S b/arch/x86/crypto/aes_ctrby8_avx-x86_64.S
index 2df2a0298f5a..a916c4a61165 100644
--- a/arch/x86/crypto/aes_ctrby8_avx-x86_64.S
+++ b/arch/x86/crypto/aes_ctrby8_avx-x86_64.S
@@ -208,7 +208,7 @@ ddq_add_8:
 
         .if (klen == KEY_128)
                 .if (load_keys)
-                        vmovdqa 3*16(p_keys), xkeyA
+                        vmovdqa 3*16(p_keys), xkey4
                 .endif
         .else
                 vmovdqa 3*16(p_keys), xkeyA
@@ -224,7 +224,7 @@ ddq_add_8:
         add     $(16*by), p_in
 
         .if (klen == KEY_128)
-                vmovdqa 4*16(p_keys), xkey4
+                vmovdqa 4*16(p_keys), xkeyB
         .else
                 .if (load_keys)
                         vmovdqa 4*16(p_keys), xkey4
@@ -234,7 +234,12 @@ ddq_add_8:
         .set i, 0
         .rept by
                 club XDATA, i
-                vaesenc xkeyA, var_xdata, var_xdata             /* key 3 */
+                /* key 3 */
+                .if (klen == KEY_128)
+                        vaesenc xkey4, var_xdata, var_xdata
+                .else
+                        vaesenc xkeyA, var_xdata, var_xdata
+                .endif
                 .set i, (i +1)
         .endr
 
@@ -243,13 +248,18 @@ ddq_add_8:
         .set i, 0
         .rept by
                 club XDATA, i
-                vaesenc xkey4, var_xdata, var_xdata             /* key 4 */
+                /* key 4 */
+                .if (klen == KEY_128)
+                        vaesenc xkeyB, var_xdata, var_xdata
+                .else
+                        vaesenc xkey4, var_xdata, var_xdata
+                .endif
                 .set i, (i +1)
         .endr
 
         .if (klen == KEY_128)
                 .if (load_keys)
-                        vmovdqa 6*16(p_keys), xkeyB
+                        vmovdqa 6*16(p_keys), xkey8
                 .endif
         .else
                 vmovdqa 6*16(p_keys), xkeyB
@@ -267,12 +277,17 @@ ddq_add_8:
         .set i, 0
         .rept by
                 club XDATA, i
-                vaesenc xkeyB, var_xdata, var_xdata             /* key 6 */
+                /* key 6 */
+                .if (klen == KEY_128)
+                        vaesenc xkey8, var_xdata, var_xdata
+                .else
+                        vaesenc xkeyB, var_xdata, var_xdata
+                .endif
                 .set i, (i +1)
         .endr
 
         .if (klen == KEY_128)
-                vmovdqa 8*16(p_keys), xkey8
+                vmovdqa 8*16(p_keys), xkeyB
         .else
                 .if (load_keys)
                         vmovdqa 8*16(p_keys), xkey8
@@ -288,7 +303,7 @@ ddq_add_8:
 
         .if (klen == KEY_128)
                 .if (load_keys)
-                        vmovdqa 9*16(p_keys), xkeyA
+                        vmovdqa 9*16(p_keys), xkey12
                 .endif
         .else
                 vmovdqa 9*16(p_keys), xkeyA
@@ -297,7 +312,12 @@ ddq_add_8:
         .set i, 0
         .rept by
                 club XDATA, i
-                vaesenc xkey8, var_xdata, var_xdata             /* key 8 */
+                /* key 8 */
+                .if (klen == KEY_128)
+                        vaesenc xkeyB, var_xdata, var_xdata
+                .else
+                        vaesenc xkey8, var_xdata, var_xdata
+                .endif
                 .set i, (i +1)
         .endr
 
@@ -306,7 +326,12 @@ ddq_add_8:
         .set i, 0
         .rept by
                 club XDATA, i
-                vaesenc xkeyA, var_xdata, var_xdata             /* key 9 */
+                /* key 9 */
+                .if (klen == KEY_128)
+                        vaesenc xkey12, var_xdata, var_xdata
+                .else
+                        vaesenc xkeyA, var_xdata, var_xdata
+                .endif
                 .set i, (i +1)
         .endr
 
@@ -412,7 +437,6 @@ ddq_add_8:
 /* main body of aes ctr load */
 
 .macro do_aes_ctrmain key_len
-
         cmp     $16, num_bytes
         jb      .Ldo_return2\key_len