Strange results from OpenSSL 1.0.2h built on an older sparc server with
Oracle Studio 12.4 and with ALL testsuite tests passed :
mimas$ openssl version
OpenSSL 1.0.2h 3 May 2016
mimas$ openssl speed
Doing mdc2 for 3s on 16 size blocks: 30887 mdc2's in 0.00s
Doing mdc2 for 3s on 64 size blocks: 8500 mdc2's in 0.00s
Doing mdc2 for 3s on 256 size blocks: 1858 mdc2's in 0.00s
Doing mdc2 for 3s on 1024 size blocks: 549 mdc2's in 0.00s
Doing mdc2 for 3s on 8192 size blocks: 69 mdc2's in 0.00s
Doing md4 for 3s on 16 size blocks: 127674 md4's in 0.00s
Doing md4 for 3s on 64 size blocks: 99595 md4's in 0.00s
Doing md4 for 3s on 256 size blocks: 59892 md4's in 0.00s
.
. etc etc
.
Doing 163 bit ecdh's for 10s: 193 163-bit ECDH ops in 0.00s
Doing 233 bit ecdh's for 10s: 94 233-bit ECDH ops in 0.00s
Doing 283 bit ecdh's for 10s: 52 283-bit ECDH ops in 0.00s
Doing 409 bit ecdh's for 10s: 22 409-bit ECDH ops in 0.00s
Doing 571 bit ecdh's for 10s: 9 571-bit ECDH ops in 0.00s
OpenSSL 1.0.2h 3 May 2016
built on: reproducible build, date unspecified
options:bn(64,32) rc4(ptr,char) des(ptr,risc1,16,int) aes(partial)
idea(int) blowfish(ptr)
compiler: /opt/solarisstudio12.4/bin/cc -I. -I.. -I../include -KPIC
-DOPENSSL_PIC -DZLIB -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN
-DHAVE_DLFCN_H -m64 -xtarget=ultra2e -xarch=sparcvis -xchip=ultra2e
-xcache=generic -errfmt=error -erroff=%none -errshort=full -xstrconst
-xildoff -m64 -xmemalign=8s -xnolibmil -Xa -xcode=pic32 -xregs=no%appl
-xlibmieee -mc -g -xs -ftrap=%none -Qy -xbuiltin=%none
-xdebugformat=dwarf -xunroll=1 -D_TS_ERRNO -D_POSIX_PTHREAD_SEMANTICS
-D_LARGEFILE64_SOURCE -D_REENTRANT -xdepend -DB_ENDIAN
The 'numbers' are in 1000s of bytes per second processed.
type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes
md2 0.00 0.00 0.00 0.00 0.00
mdc2 Infk Infk Infk Infk Infk
md4 Infk Infk Infk Infk Infk
md5 Infk Infk Infk Infk Infk
.
.
.
ghash Infk Infk Infk Infk Infk
sign verify sign/s verify/s
rsa 512 bits 0.000000s 0.000000s Inf Inf
rsa 1024 bits 0.000000s 0.000000s Inf Inf
rsa 2048 bits 0.000000s 0.000000s Inf Inf
rsa 4096 bits 0.000000s 0.000000s Inf Inf
sign verify sign/s verify/s
dsa 512 bits 0.000000s 0.000000s Inf Inf
dsa 1024 bits 0.000000s 0.000000s Inf Inf
dsa 2048 bits 0.000000s 0.000000s Inf Inf
sign verify sign/s verify/s
160 bit ecdsa (secp160r1) 0.0000s 0.0000s Inf Inf
192 bit ecdsa (nistp192) 0.0000s 0.0000s Inf Inf
224 bit ecdsa (nistp224) 0.0000s 0.0000s Inf Inf
256 bit ecdsa (nistp256) 0.0000s 0.0000s Inf Inf
384 bit ecdsa (nistp384) 0.0000s 0.0000s Inf Inf
521 bit ecdsa (nistp521) 0.0000s 0.0000s Inf Inf
163 bit ecdsa (nistk163) 0.0000s 0.0000s Inf Inf
233 bit ecdsa (nistk233) 0.0000s 0.0000s Inf Inf
283 bit ecdsa (nistk283) 0.0000s 0.0000s Inf Inf
409 bit ecdsa (nistk409) 0.0000s 0.0000s Inf Inf
571 bit ecdsa (nistk571) 0.0000s 0.0000s Inf Inf
163 bit ecdsa (nistb163) 0.0000s 0.0000s Inf Inf
233 bit ecdsa (nistb233) 0.0000s 0.0000s Inf Inf
283 bit ecdsa (nistb283) 0.0000s 0.0000s Inf Inf
409 bit ecdsa (nistb409) 0.0000s 0.0000s Inf Inf
571 bit ecdsa (nistb571) 0.0000s 0.0000s Inf Inf
op op/s
160 bit ecdh (secp160r1) 0.0000s Inf
192 bit ecdh (nistp192) 0.0000s Inf
224 bit ecdh (nistp224) 0.0000s Inf
256 bit ecdh (nistp256) 0.0000s Inf
384 bit ecdh (nistp384) 0.0000s Inf
521 bit ecdh (nistp521) 0.0000s Inf
163 bit ecdh (nistk163) 0.0000s Inf
233 bit ecdh (nistk233) 0.0000s Inf
283 bit ecdh (nistk283) 0.0000s Inf
409 bit ecdh (nistk409) 0.0000s Inf
571 bit ecdh (nistk571) 0.0000s Inf
163 bit ecdh (nistb163) 0.0000s Inf
233 bit ecdh (nistb233) 0.0000s Inf
283 bit ecdh (nistb283) 0.0000s Inf
409 bit ecdh (nistb409) 0.0000s Inf
571 bit ecdh (nistb571) 0.0000s Inf
mimas$
While I appreciate infinite performance in zero time the wall clock
disagrees wildly. :-\
Some system info :
mimas$ psrinfo -pv
The physical processor has 1 virtual processor (0)
UltraSPARC-IIe (portid 0 impl 0x13 ver 0x14 clock 500 MHz)
mimas$ uname -a
SunOS mimas 5.10 Generic_150400-35 sun4u sparc SUNW,UltraAX-i2
mimas$ cat /etc/release
Oracle Solaris 10 1/13 s10s_u11wos_24a SPARC
Copyright (c) 1983, 2013, Oracle and/or its affiliates. All rights
reserved.
Assembled 17 January 2013
mimas$ which cc
/usr/bin/cc
mimas$ cc -V
cc: Sun C 5.13 SunOS_sparc 2014/10/20
Is this a known bug on older slower hardware ?
Dennis Clarke
