diff options
author | Mathias Krause <minipli@googlemail.com> | 2014-03-24 12:10:37 -0400 |
---|---|---|
committer | Herbert Xu <herbert@gondor.apana.org.au> | 2014-03-25 08:25:42 -0400 |
commit | 6ca5afb8c26991cf4f13a8bcca870ec2a9522bf7 (patch) | |
tree | 095efa7d609b3c8e1c7d85e56480a22f5354ea46 | |
parent | 7c1da8d0d046174a4188b5729d7579abf3d29427 (diff) |
crypto: x86/sha1 - re-enable the AVX variant
Commit 7c1da8d0d0 "crypto: sha - SHA1 transform x86_64 AVX2"
accidentally disabled the AVX variant by making the avx_usable() test
not only fail in case the CPU doesn't support AVX or OSXSAVE but also
if it doesn't support AVX2.
Fix that regression by splitting up the AVX/AVX2 test into two
functions. Also test for the BMI1 extension in the avx2_usable() test
as the AVX2 implementation not only makes use of BMI2 but also BMI1
instructions.
Cc: Chandramouli Narayanan <mouli@linux.intel.com>
Signed-off-by: Mathias Krause <minipli@googlemail.com>
Reviewed-by: H. Peter Anvin <hpa@linux.intel.com>
Reviewed-by: Marek Vasut <marex@denx.de>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
-rw-r--r-- | arch/x86/crypto/sha1_ssse3_glue.c | 26 |
1 files changed, 16 insertions, 10 deletions
diff --git a/arch/x86/crypto/sha1_ssse3_glue.c b/arch/x86/crypto/sha1_ssse3_glue.c index 139a55c04d82..74d16ef707c7 100644 --- a/arch/x86/crypto/sha1_ssse3_glue.c +++ b/arch/x86/crypto/sha1_ssse3_glue.c | |||
@@ -208,11 +208,7 @@ static bool __init avx_usable(void) | |||
208 | { | 208 | { |
209 | u64 xcr0; | 209 | u64 xcr0; |
210 | 210 | ||
211 | #if defined(CONFIG_AS_AVX2) | ||
212 | if (!cpu_has_avx || !cpu_has_avx2 || !cpu_has_osxsave) | ||
213 | #else | ||
214 | if (!cpu_has_avx || !cpu_has_osxsave) | 211 | if (!cpu_has_avx || !cpu_has_osxsave) |
215 | #endif | ||
216 | return false; | 212 | return false; |
217 | 213 | ||
218 | xcr0 = xgetbv(XCR_XFEATURE_ENABLED_MASK); | 214 | xcr0 = xgetbv(XCR_XFEATURE_ENABLED_MASK); |
@@ -224,11 +220,23 @@ static bool __init avx_usable(void) | |||
224 | 220 | ||
225 | return true; | 221 | return true; |
226 | } | 222 | } |
223 | |||
224 | #ifdef CONFIG_AS_AVX2 | ||
225 | static bool __init avx2_usable(void) | ||
226 | { | ||
227 | if (avx_usable() && cpu_has_avx2 && boot_cpu_has(X86_FEATURE_BMI1) && | ||
228 | boot_cpu_has(X86_FEATURE_BMI2)) | ||
229 | return true; | ||
230 | |||
231 | return false; | ||
232 | } | ||
233 | #endif | ||
227 | #endif | 234 | #endif |
228 | 235 | ||
229 | static int __init sha1_ssse3_mod_init(void) | 236 | static int __init sha1_ssse3_mod_init(void) |
230 | { | 237 | { |
231 | char *algo_name; | 238 | char *algo_name; |
239 | |||
232 | /* test for SSSE3 first */ | 240 | /* test for SSSE3 first */ |
233 | if (cpu_has_ssse3) { | 241 | if (cpu_has_ssse3) { |
234 | sha1_transform_asm = sha1_transform_ssse3; | 242 | sha1_transform_asm = sha1_transform_ssse3; |
@@ -238,13 +246,11 @@ static int __init sha1_ssse3_mod_init(void) | |||
238 | #ifdef CONFIG_AS_AVX | 246 | #ifdef CONFIG_AS_AVX |
239 | /* allow AVX to override SSSE3, it's a little faster */ | 247 | /* allow AVX to override SSSE3, it's a little faster */ |
240 | if (avx_usable()) { | 248 | if (avx_usable()) { |
241 | if (cpu_has_avx) { | 249 | sha1_transform_asm = sha1_transform_avx; |
242 | sha1_transform_asm = sha1_transform_avx; | 250 | algo_name = "AVX"; |
243 | algo_name = "AVX"; | ||
244 | } | ||
245 | #ifdef CONFIG_AS_AVX2 | 251 | #ifdef CONFIG_AS_AVX2 |
246 | if (cpu_has_avx2 && boot_cpu_has(X86_FEATURE_BMI2)) { | 252 | /* allow AVX2 to override AVX, it's a little faster */ |
247 | /* allow AVX2 to override AVX, it's a little faster */ | 253 | if (avx2_usable()) { |
248 | sha1_transform_asm = sha1_apply_transform_avx2; | 254 | sha1_transform_asm = sha1_apply_transform_avx2; |
249 | algo_name = "AVX2"; | 255 | algo_name = "AVX2"; |
250 | } | 256 | } |