On Wed, Nov 29, 2017 at 02:13:43AM +0000, Ken Moffat wrote:
>
> Over the last 24 hours I've been updating qtwebengine and I see
> similar variation in the number of SBUs. I wonder if it is worth
> noting that the timings, particularly for large slow C++ programs,
> can vary a lot between different machines ?
>
This does seem to be very machine-specific: little difference
between an old 4-core phenom and a haswell i7 using just 4 cores,
(and SATA2 spinning rust on the phenom, SATA3 SSD on the haswell),
although obviously the clock time is more on the phenom.
But on an A10 Kaveri from last year (slower, but with 8GB DRAM and a
SATA3 SSD) the SBU for qtwebengine is about 25% greater. And on a
4GB i3 (swaps, but only marginally) I'm seeing 100+ SBU instead of
60.
So in 'SBU values in BLFS' (introduction/conventions) I'm proposing
to add a new paragraph after "These values indicate testing was done
using multiple cores. Note that while this speeds up the build on
systems with the appropriate hardware, the speedup is not linear and
to some extent depends on the individual package and specific
hardware used." -
"Where even a parallel build takes more than 15 SBU, on certain
machines the actual time may be considerably greater, even when the
build does not use swap."
ĸen
--
Truth, in front of her huge walk-in wardrobe, selected black leather
boots with stiletto heels for such a barefaced truth.
- Unseen Academicals
--
http://lists.linuxfromscratch.org/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
