On Thu, Nov 7, 2013 at 9:54 AM, Jim Pingle <[email protected]> wrote: > The sheet could really use some more data, so anyone who has an AES-NI > capable system, feel free to run through the tests and help fill out the > sheet. :-) >
/usr/bin/openssl speed -evp aes-128-cbc -elapsed The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 99399.17k 108488.44k 111490.64k 113100.76k 114121.28k The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 97350.86k 108496.66k 111490.32k 112989.75k 114082.46k /usr/local/bin/openssl speed -evp aes-128-cbc -elapsed The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 397045.72k 422822.44k 430420.31k 432295.94k 432848.90k The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 396872.65k 423517.87k 430512.98k 432295.25k 432848.90k cryptotest -va aes cipher aes keylen 16 CIOCGSESSION: Invalid argument Now we load aesni.... /usr/bin/openssl speed -evp aes-128-cbc -elapsed The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 10898.07k 40515.02k 122446.32k 249924.25k 299990.00k The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 11021.06k 40772.30k 122881.76k 250294.41k 300729.52k /usr/local/bin/openssl speed -evp aes-128-cbc -elapsed The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 11031.72k 40436.22k 122958.08k 250292.91k 300296.87k The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 10894.13k 40242.52k 121988.61k 249543.34k 300127.57k /usr/bin/openssl speed -evp aes-128-cbc -elapsed -engine cryptodev The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 11049.62k 40428.86k 122644.60k 250343.91k 300689.88k The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 11058.16k 40830.55k 122402.15k 249873.61k 300421.51k /usr/local/bin/openssl speed -evp aes-128-cbc -elapsed -engine cryptodev The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 10948.26k 40138.15k 121933.23k 248702.14k 300463.45k The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes aes-128-cbc 11028.30k 40161.71k 122037.42k 248784.53k 300324.18k cryptotest -va aes session = 0x0 device = aesni0 count = 1, size = 16 iv: 0000: 6f 38 32 75 74 74 6a 34 62 62 62 21 69 74 6e 6f cleartext: 0000: 35 38 65 74 61 69 75 62 6e 31 6e 6a 6a 33 37 38 cleartext: 0000: 35 38 65 74 61 69 75 62 6e 31 6e 6a 6a 33 37 38 0.000 sec, 2 aes crypts, 16 bytes, 2461538 byte/sec, 18.8 Mb/sec If I run without -elapsed, I can see the CPU time used by the tests. The selection of engine cryptodev makes no difference: openssl always offloads from the CPU to the AES-NI engine once it is loaded, as the CPU time is fraction of a second over the 3 second tests. True for both versions of openssl. So it seems that the overall speed is lower with freebsd's AES-NI engine, but the offload from the CPU is significant. CPU: Intel(R) Xeon(R) CPU E31220L @ 2.20GHz (2195.02-MHz K8-class CPU) Origin = "GenuineIntel" Id = 0x206a7 Family = 6 Model = 2a Stepping = 7 Features=0xbfebfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE> Features2=0x17bae3ff<SSE3,PCLMULQDQ,DTES64,MON,DS_CPL,VMX,SMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,x2APIC,POPCNT,TSCDLT,AESNI,XSAVE,AVX> AMD Features=0x28100800<SYSCALL,NX,RDTSCP,LM> AMD Features2=0x1<LAHF> TSC: P-state invariant
_______________________________________________ List mailing list [email protected] http://lists.pfsense.org/mailman/listinfo/list
