summaryrefslogtreecommitdiff
path: root/src/lib/libcrypto/camellia
diff options
context:
space:
mode:
authorderaadt <>2017-12-11 05:48:01 +0000
committerderaadt <>2017-12-11 05:48:01 +0000
commit739b09b33140a7cde6da09372fc371d59f4751c7 (patch)
treef45f495aceb89570179a0b9f1d0c3150b6da956e /src/lib/libcrypto/camellia
parentf470834311fadde98f690758b95be8f61963fddb (diff)
downloadopenbsd-739b09b33140a7cde6da09372fc371d59f4751c7.tar.gz
openbsd-739b09b33140a7cde6da09372fc371d59f4751c7.tar.bz2
openbsd-739b09b33140a7cde6da09372fc371d59f4751c7.zip
http://repzret.org/p/repzret/
My read of this: Long time ago (Think Conan, not dinasaurs) during the race to make speedier processors, a cpu vendor built a pipeline with a bad stall, and proposed a tremendously hasky workaround. A wizard adopted this into his perl scroll, and failed to reflect later when no compiler adopted the practice. This relic remains at the tail end of some functions in OpenSSL as ".byte 0xf3,0xc3". Banish it straight to hell. ok mlarkin, others also stared blankly
Diffstat (limited to 'src/lib/libcrypto/camellia')
-rw-r--r--src/lib/libcrypto/camellia/asm/cmll-x86_64.pl4
1 files changed, 2 insertions, 2 deletions
diff --git a/src/lib/libcrypto/camellia/asm/cmll-x86_64.pl b/src/lib/libcrypto/camellia/asm/cmll-x86_64.pl
index a171c654b2..df6bf11a28 100644
--- a/src/lib/libcrypto/camellia/asm/cmll-x86_64.pl
+++ b/src/lib/libcrypto/camellia/asm/cmll-x86_64.pl
@@ -218,7 +218,7 @@ $code.=<<___;
218 mov $t2,@S[2] 218 mov $t2,@S[2]
219 mov $t3,@S[3] 219 mov $t3,@S[3]
220 220
221 .byte 0xf3,0xc3 # rep ret 221 retq
222.size _x86_64_Camellia_encrypt,.-_x86_64_Camellia_encrypt 222.size _x86_64_Camellia_encrypt,.-_x86_64_Camellia_encrypt
223 223
224# V1.x API 224# V1.x API
@@ -329,7 +329,7 @@ $code.=<<___;
329 mov $t0,@S[2] 329 mov $t0,@S[2]
330 mov $t1,@S[3] 330 mov $t1,@S[3]
331 331
332 .byte 0xf3,0xc3 # rep ret 332 retq
333.size _x86_64_Camellia_decrypt,.-_x86_64_Camellia_decrypt 333.size _x86_64_Camellia_decrypt,.-_x86_64_Camellia_decrypt
334___ 334___
335 335