On Sun, Dec 15, 2019 at 12:30:52PM -0600, Bruce Dubbs via lfs-dev wrote:
> https://www.phoronix.com/scan.php?page=article&item=10980xe-intel-linux
>
> I find the benchmarks interesting. On page 5 is the time for kernel
> comparison. On my Haswell I get for linux-5.4.2:
>
> $ time make defconfig
> real 0m2.883s
>
> $ time make
> real 15m28.691s
>
> And after 'make mrproper; make defconfig'
>
> $ time make -j12
> real 2m24.326s
>
> Clearly it's time for me to consider upgrading my development system, but
> I'm waiting to see what Intel's 10nm processors can do and what they will
> cost. My current SBU time is 101s and some of the larger packages like
> libreoffice, thunderbird, qt, qtwebengine. etc are a bit tedious.
>
I don't ever make defconfig (apart from my laptop, my configs are
adequately smaller although some unwanted things invariably get
pulled in). But I see that on my i7 haswell (DDR3 1600!) the SBU is
around 101s - latest binutils adds a second or two - and on my
Picasso (DDR4 3000 but with binutils-2.32) the latest build was also
101s.
For kernel compiles, make -j{N+1} is the old recommendation, or from
people like Greg K-H make -j{Nx2}. Unlike C++ and rust the kernel
doesn't use a lot of RAM when compiling.
But I don't see that I'll buy intel again in the foreseeable future.
> On the other hand on page 6, Systemd Total Boot Time
>
> The best time was about 25 seconds. On my Haswell, my LFS System V boot
> time is 8 seconds.
>
> -- Bruce
My haswell often spends several seconds in boot with things like
unbound (for me, that is used on all systems, not as a server
package). But I haven't timed a boot recently. For systemd
desktops, the time will include both X (or Wayland) and the
display manager.
ĸen
--
We've all got both light and dark inside of us.
What matters is the part we choose to act on.
-- Sirius Black
--
http://lists.linuxfromscratch.org/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page
