vBulletin Search Engine Optimization
| |||||||
| Register | FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
| ||||
| Hi All, I have a Broadcom encryption accelerator in my motherboard (BCM 5823). I have modified the ubsec source and recompiled the kernel to support the 5823 chip. I now want to test it but don't really know how to go about testing whether the chip is actually accelerating encryption functions. I have checked the /var/run/dmesg.boot file and the 5823 is now being identified properly (after my changes). The files I changed were ubsec.c pcidevs pcidevs.h and pcidevs_data.h in the /usr/src/sys/dev/pci directory. Any ideas how I could do a quick and easy test to see if the chip is working? Thanks, Jim |
| |||
| On Fri, 29 Aug 2003, Jim Lambert wrote: > I have a Broadcom encryption accelerator in my motherboard (BCM 5823). I > have modified the ubsec source and recompiled the kernel to support the 5823 > chip. I now want to test it but don't really know how to go about testing > whether the chip is actually accelerating encryption functions. openssl speed des with sysctl kern.usercrypto on and off should be a good test. -- "The contagious people of Washington have stood firm against diversity during this long period of increment weather." - M. Barry, Mayor of Washington, DC |
| ||||
| "Ted Unangst" <tedu@stanford.edu> wrote in message news:Pine.GSO.4.44.0308291517010.29526-100000@saga7.Stanford.EDU... > On Fri, 29 Aug 2003, Jim Lambert wrote: > > > I have a Broadcom encryption accelerator in my motherboard (BCM 5823). I > > have modified the ubsec source and recompiled the kernel to support the 5823 > > chip. I now want to test it but don't really know how to go about testing > > whether the chip is actually accelerating encryption functions. > > openssl speed des with sysctl kern.usercrypto on and off should be a good > test. Thanks. I forgot to post a follow-up last Friday, but did figure out I could do an openssl 3des encryption of a big (140MB) file and see the differences. CPU time went from 62 seconds to 0.20 seconds after compiling in the support so I am assuming it is working. :-) Jim |