Date: Sat, 10 May 2014 11:07:38 -0700 From: Bakul Shah <bakul@bitblocks.com> To: Mark R V Murray <markm@FreeBSD.org> Cc: freebsd-arm <freebsd-arm@freebsd.org> Subject: Re: Fast cycle counter for ARM chips with SCC - patch for review. Message-ID: <7D5EE653-922D-4AF8-92A9-56344148FC7F@bitblocks.com> In-Reply-To: <22E12094-E6B2-42F9-94AB-014A702D17F2@FreeBSD.org>
index | next in thread | previous in thread | raw e-mail
FYI: BCM2835 has a built in h/w RNG (using a reverse biased transistor as an entropy source). Access to it was enabled in the RPi firmware around Jan 30, 2013. FWIW, it passes rngtest as well as FreeBSD's /dev/random (on x86/amd64) does. For the spec. see http://pastehtml.com/view/crkxyohmp.rtxt Referenced from this thread: http://www.raspberrypi.org/forums/viewtopic.php?f=29&t=19334&p=273944#p273944 > On May 10, 2014, at 10:39 AM, Mark R V Murray <markm@FreeBSD.org> wrote: > > Hi * > > This patch makes the ARM6 kernels that have an SCC coprocessor (RPI and WANDBOARD have them) get a MUCH better implementation of get_cyclecount(9), but not a perfect one. The incrementing rate is good (+- 1 per instruction), but its only 32 bits. Later, if there is interest, I may wish to fix that with an overflow interrupt, but for now its easily good enough for the kernel entropy harvesting service. Also, its MUUUCH more efficient; a simple read rather can calling the internal kernel binuptime(9) clock. > > Comments, please? I’m keen to commit. > > M > -- > Mark R V Murray > <arm_ccnt_counter.diff> >home | help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?7D5EE653-922D-4AF8-92A9-56344148FC7F>