<html>
<head>
<meta content="text/html; charset=utf-8" http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<p><br>
</p>
<blockquote
cite="mid:mailman.1289.1482443544.674.tor-relays@lists.torproject.org"
type="cite">
<pre wrap="">Please don't mix multiple questions into one thread.
</pre>
</blockquote>
Sorry, my bad.<br>
<br>
<blockquote
cite="mid:mailman.1289.1482443544.674.tor-relays@lists.torproject.org"
type="cite">
<pre wrap="">Tor does not implement crypto itself (mostly) and relies on a
cryptolibrary (which is OpenSSL/LibreSSL/etc) instead. Thus you should
check if AES-NI is enabled in your cryptolibrary.
An excerpt from StackOverflow answer [1] about it:
$ openssl speed -elapsed -evp aes-128-cbc
$ OPENSSL_ia32cap="~0x200000200000000" openssl speed -elapsed -evp
aes-128-cbc
"Output of the first line should be significantly faster than the
second." If there is no AES-NI enabled in "OpenSSL" these two should
give similar results.</pre>
</blockquote>
I couldn't do that test. OpenSSL was not installed. <br>
After I installed it I could perform that test and it was positive.<br>
Here is the output:<br>
<br>
$ openssl speed -elapsed -evp aes-128-cbc<br>
You have chosen to measure elapsed time instead of user CPU time.<br>
Doing aes-128-cbc for 3s on 16 size blocks: 33370007 aes-128-cbc's
in 3.00s<br>
Doing aes-128-cbc for 3s on 64 size blocks: 13118341 aes-128-cbc's
in 3.00s<br>
Doing aes-128-cbc for 3s on 256 size blocks: 3915543 aes-128-cbc's
in 3.00s<br>
Doing aes-128-cbc for 3s on 1024 size blocks: 1029134 aes-128-cbc's
in 3.00s<br>
Doing aes-128-cbc for 3s on 8192 size blocks: 130438 aes-128-cbc's
in 3.00s<br>
OpenSSL 1.0.1t 3 May 2016<br>
built on: Fri Sep 23 17:53:23 2016<br>
options:bn(64,64) rc4(16x,int) des(idx,cisc,16,int) aes(partial)
blowfish(idx) <br>
compiler: gcc -I. -I.. -I../include -fPIC -DOPENSSL_PIC
-DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -m64
-DL_ENDIAN -DTERMIO -g -O2 -fstack-protector-strong -Wformat
-Werror=format-security -D_FORTIFY_SOURCE=2 -Wl,-z,relro
-Wa,--noexecstack -Wall -DMD32_REG_T=int -DOPENSSL_IA32_SSE2
-DOPENSSL_BN_ASM_MONT -DOPENSSL_BN_ASM_MONT5 -DOPENSSL_BN_ASM_GF2m
-DSHA1_ASM -DSHA256_ASM -DSHA512_ASM -DMD5_ASM -DAES_ASM -DVPAES_ASM
-DBSAES_ASM -DWHIRLPOOL_ASM -DGHASH_ASM<br>
The 'numbers' are in 1000s of bytes per second processed.<br>
type 16 bytes 64 bytes 256 bytes 1024 bytes
8192 bytes<br>
aes-128-cbc 177973.37k 279857.94k 334126.34k 351277.74k
356182.70k<br>
<br>
<br>
$ OPENSSL_ia32cap="~0x200000200000000" openssl speed -elapsed -evp
aes-128-cbc<br>
You have chosen to measure elapsed time instead of user CPU time.<br>
Doing aes-128-cbc for 3s on 16 size blocks: 6232419 aes-128-cbc's in
3.00s<br>
Doing aes-128-cbc for 3s on 64 size blocks: 1776077 aes-128-cbc's in
3.00s<br>
Doing aes-128-cbc for 3s on 256 size blocks: 454887 aes-128-cbc's in
3.00s<br>
Doing aes-128-cbc for 3s on 1024 size blocks: 114409 aes-128-cbc's
in 3.00s<br>
Doing aes-128-cbc for 3s on 8192 size blocks: 14327 aes-128-cbc's in
3.00s<br>
OpenSSL 1.0.1t 3 May 2016<br>
built on: Fri Sep 23 17:53:23 2016<br>
options:bn(64,64) rc4(16x,int) des(idx,cisc,16,int) aes(partial)
blowfish(idx) <br>
compiler: gcc -I. -I.. -I../include -fPIC -DOPENSSL_PIC
-DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -m64
-DL_ENDIAN -DTERMIO -g -O2 -fstack-protector-strong -Wformat
-Werror=format-security -D_FORTIFY_SOURCE=2 -Wl,-z,relro
-Wa,--noexecstack -Wall -DMD32_REG_T=int -DOPENSSL_IA32_SSE2
-DOPENSSL_BN_ASM_MONT -DOPENSSL_BN_ASM_MONT5 -DOPENSSL_BN_ASM_GF2m
-DSHA1_ASM -DSHA256_ASM -DSHA512_ASM -DMD5_ASM -DAES_ASM -DVPAES_ASM
-DBSAES_ASM -DWHIRLPOOL_ASM -DGHASH_ASM<br>
The 'numbers' are in 1000s of bytes per second processed.<br>
type 16 bytes 64 bytes 256 bytes 1024 bytes
8192 bytes<br>
aes-128-cbc 33239.57k 37889.64k 38817.02k 39051.61k
39122.26k<br>
<br>
<br>
But it is a little confusing for me because there is this line in
the logs:<br>
<br>
Tor 0.2.9.8 (git-a0df013ea241b026) running on Linux with Libevent
2.0.21-stable, OpenSSL 1.0.1t and Zlib 1.2.8.<br>
<br>
From that I thought Tor used already OpenSSL but it wasn't
installed. :S<br>
<br>
I bought this board with this CPU (incl. AES-NI support) because I
thought it would give a benefit.<br>
<br>
<blockquote
cite="mid:mailman.1289.1482443544.674.tor-relays@lists.torproject.org"
type="cite">
<pre wrap="">N.B. AES-NI is not a feature of <b class="moz-txt-star"><span class="moz-txt-tag">*</span>motherboard<span class="moz-txt-tag">*</span></b> - it's CPU instructions
(NI stands for "New Instructions").</pre>
</blockquote>
I simply forgot that. ;)<br>
<br>
<br>
Cheers,<br>
Patrice<br>
</body>
</html>