Re: AES-NI in 1.0.1

2012-03-18 Thread Andy Polyakov
So if you directly use the AES API you used to have a little better performance, but now you don't get the AES-NI support and so are a factor slower when using it. Is this the normal and expected behaviour? >>> I hope this isn't true. If it is, it means applications

Re: AES-NI in 1.0.1

2012-03-18 Thread Kurt Roeckx
On Sun, Mar 18, 2012 at 07:24:25PM +0100, Andy Polyakov wrote: > >> So if you directly use the AES API you used to have a little better > >> performance, > >> but now you don't get the AES-NI support and so are a factor slower when > >> using it. > >> > >> Is this the normal and expected behaviou

Re: AES-NI in 1.0.1

2012-03-18 Thread Andy Polyakov
>>> 1.0.1: >>> $ openssl speed aes-128-cbc: >>> type 16 bytes 64 bytes256 bytes 1024 bytes 8192 >>> bytes >>> aes-128 cbc 110450.10k 120831.36k 122216.11k 123465.05k >>> 123909.46k >>> >>> $ openssl speed -evp aes-128-cbc: >>> type 16 bytes 64 b

Re: AES-NI in 1.0.1

2012-03-18 Thread Kurt Roeckx
On Sun, Mar 18, 2012 at 07:16:27PM +0100, Andy Polyakov wrote: > Hi, > > > 1.0.1: > > $ openssl speed aes-128-cbc: > > type 16 bytes 64 bytes256 bytes 1024 bytes 8192 > > bytes > > aes-128 cbc 110450.10k 120831.36k 122216.11k 123465.05k > > 123909.46k > > > >

Re: AES-NI in 1.0.1

2012-03-18 Thread Andy Polyakov
>> So if you directly use the AES API you used to have a little better >> performance, >> but now you don't get the AES-NI support and so are a factor slower when >> using it. >> >> Is this the normal and expected behaviour? > > I hope this isn't true. If it is, it means applications like OpenS

Re: AES-NI in 1.0.1

2012-03-18 Thread Andy Polyakov
Hi, > 1.0.1: > $ openssl speed aes-128-cbc: > type 16 bytes 64 bytes256 bytes 1024 bytes 8192 bytes > aes-128 cbc 110450.10k 120831.36k 122216.11k 123465.05k 123909.46k > > $ openssl speed -evp aes-128-cbc: > type 16 bytes 64 bytes256 bytes

Re: AES-NI in 1.0.1

2012-03-18 Thread Thor Lancelot Simon
On Sun, Mar 18, 2012 at 05:57:24PM +0100, Kurt Roeckx wrote: > > So if you directly use the AES API you used to have a little better > performance, > but now you don't get the AES-NI support and so are a factor slower when > using it. > > Is this the normal and expected behaviour? I hope this

AES-NI in 1.0.1

2012-03-18 Thread Kurt Roeckx
Hi, I was looking at the AES-NI support in 1.0.1, and it seems that it now has been merged in EVP (but test/test_aesni still exists in HEAD). This has the following effect: 1.0.0h: $ openssl speed aes-128-cbc: type 16 bytes 64 bytes256 bytes 1024 bytes 8192 bytes aes-128