Date: Sun, 11 Jul 2021 04:03:12 -0700 From: Mark Millard via freebsd-arm <freebsd-arm@freebsd.org> To: freebsd-arm <freebsd-arm@freebsd.org> Subject: Re: HoneyComb first-boot notes [a L3/L2/L1/RAM performance oddity] Message-ID: <C0426887-59D9-4524-8542-8DA6DBAFF744@yahoo.com> In-Reply-To: <0B577668-97AB-44B6-B1A7-C68F6CC299E5@yahoo.com> References: <8A6C415F-A57B-4F2F-861F-052B487166D6.ref@yahoo.com> <8A6C415F-A57B-4F2F-861F-052B487166D6@yahoo.com> <YNGT5hcHOBd6cU4T@x230.ds> <40AE6447-77AF-4D0E-864F-AD52D9F3346F@yahoo.com> <YNGf999RsaTfNhcp@x230.ds> <Rv9QGaKflpIjLPsxUFG3ht12loej__FxMBy7SQ1QzDTk1NLcFjGb4ScQuF32SakZi68wjgPQpIVp2dipMoYteJIAMhSrXbPM6-mRSeL_744=@a9development.com> <C4D3B585-63B6-4C2A-B8DA-264073C6E2C2@yahoo.com> <12A4EDD1-A2AB-4CE3-AB0E-A4B5D6FB4674@yahoo.com> <5B1B5E1A-8AE4-4889-ABE6-50C206F896FB@yahoo.com> <7DBDC8AB-C80B-4E26-B58F-251A3D29CE41@yahoo.com> <5BBF1B55-F02C-4817-B805-677EDDC5B809@yahoo.com> <0B577668-97AB-44B6-B1A7-C68F6CC299E5@yahoo.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 2021-Jul-10, at 22:09, Mark Millard <marklmi at yahoo.com> wrote: > On 2021-Jun-24, at 16:25, Mark Millard <marklmi at yahoo.com> wrote: >=20 >> On 2021-Jun-24, at 16:00, Mark Millard <marklmi at yahoo.com> wrote: >>=20 >>> On 2021-Jun-24, at 13:39, Mark Millard <marklmi at yahoo.com> wrote: >>>=20 >>>> Repeating here what I've reported on teh solidrun discord: >>>>=20 >>>> I decided to experiment with monitoring the temperatures reported >>>> as things are. For the default heat-sink/fan and the 2 other fans >>>> in the case, buildworld with load average 16.? for some time has >>>> stayed with tz0 through tz6 reporting between 61.0degC and = 66.0degC, >>>> say about 20degC for ambiant. (tz7 and tz8 report 0.1C.) During >>>> stages with lower load averages, the tz0..tz6 tempuratures back off >>>> some. So it looks like my default context keeps the system >>>> sufficiently cool for such use. >>>>=20 >>>> I'll note that the default heat-sink's fan is not operating at = rates >>>> that I hear it upstairs. I've heard the noisy mode from there = during >>>> early parts of booting for Fedora 34 server, for example. >>>=20 >>> So I updated my stable/13 source and built and installed >>> the update, then did a rm -fr of the build directory >>> tree context and started a from-scratch build. The >>> build had: >>>=20 >>> SYSTEM_COMPILER: Determined that CC=3Dcc matches the source tree. = Not bootstrapping a cross-compiler. >>> and: >>> SYSTEM_LINKER: Determined that LD=3Dld matches the source tree. Not = bootstrapping a cross-linker. >>>=20 >>> as is my standard context for doing such "how long does >>> it take" buildworld buildkernel testing. >>>=20 >>> On aarch64 I do not build for targeting non-arm architectures. >>> This does save some time on the builds. >>=20 >> I should have mentioned that my builds are based on tuning >> for the cortex-a72 via -mcpu=3Dcortex-a72 being used. This >> was also true of the live system that was running, kernel >> and world. >>=20 >>> The results for the HoneyComb configuration I'm using: >>>=20 >>> World build completed on Thu Jun 24 15:30:11 PDT 2021 >>> World built in 3173 seconds, ncpu: 16, make -j16 >>> Kernel build for GENERIC-NODBG-CA72 completed on Thu Jun 24 15:34:45 = PDT 2021 >>> Kernel(s) GENERIC-NODBG-CA72 built in 274 seconds, ncpu: 16, make = -j16 >>>=20 >>> So World+Kernel took a a little under 1 hr to build (-j16). >>>=20 >>>=20 >>>=20 >>> Comparison/contrast to prior aarch64 systems that I've used >>> for buildworld buildkernel . . . >>>=20 >>>=20 >>> By contrast, the (now failed) OverDrive 1000's last timing >>> was (building releng/13 instead of stable/13): >>>=20 >>> World build completed on Tue Apr 27 02:50:52 PDT 2021 >>> World built in 12402 seconds, ncpu: 4, make -j4 >>> Kernel build for GENERIC-NODBG-CA72 completed on Tue Apr 27 03:08:04 = PDT 2021 >>> Kernel(s) GENERIC-NODBG-CA72 built in 1033 seconds, ncpu: 4, make = -j4 >>>=20 >>> So World+Kernel took a a little under 3.75 hrs to build (-j4). >>>=20 >>>=20 >>> The MACCHIATObin Double Shot's last timing was >>> (building a 13-CURRENT): >>>=20 >>> World build completed on Tue Jan 19 03:44:59 PST 2021 >>> World built in 14902 seconds, ncpu: 4, make -j4 >>> Kernel build for GENERIC-NODBG completed on Tue Jan 19 04:04:25 PST = 2021 >>> Kernel(s) GENERIC-NODBG built in 1166 seconds, ncpu: 4, make -j4 >>>=20 >>> So World+Kernel took a little under 4.5 hrs to build (-j4). >>>=20 >>>=20 >>> The RPi4B 8GiByte's last timing was >>> ( arm_freq=3D2000, sdram_freq_min=3D3200, force_turbo=3D1, USB3 SSD >>> building releng/13 ): >>>=20 >>> World build completed on Tue Apr 20 14:34:38 PDT 2021 >>> World built in 22104 seconds, ncpu: 4, make -j4 >>> Kernel build for GENERIC-NODBG completed on Tue Apr 20 15:03:24 PDT = 2021 >>> Kernel(s) GENERIC-NODBG built in 1726 seconds, ncpu: 4, make -j4 >>>=20 >>> So World+Kernel took somewhat under 6 hrs 40 min to build. >>=20 >> The -mcpu=3Dcortex-a72 use note also applies to the OverDrive 1000, >> MACCHIATObin Double Shot, and RPi4B 8 GiByte contexts. >>=20 >=20 > I've run into an issue where what FreeBSD calls cpu 0 has > significantly different L3/L2/L1/RAM subsystem performance > than all the other cores (cpu 0 being worse). Similarly for > compared/contrasted to all 4 MACCHIATObin Double Shot cores. >=20 > A plot with curves showing the issue is at: >=20 > = https://github.com/markmi/acpphint/blob/master/acpphint_example_data/Honey= CombFreeBSDcpu0RAMAccessPerformanceIsOdd.png >=20 > The dark red curves in the plot show the expected general > shape for such and are for cpu 0. The lighter colored > curves are the MACCHIATObin curves. The darker ones are > the HoneyComb curves, where the L3/L2/L1 is relatively > effective (other than cpu 0). >=20 > My notes on Discord (so far) are . . . >=20 > The curves are from my C++ variant of the old Hierarchical > INTegration benchmark (historically abbreviated HINT). You > can read the approximate size of a level of cache from=20 > the x-axis for where the curve drops faster. So, right > (most obvious) to left (least obvious): L3 8 MiByte, L2 1 > MiByte (per core pair, as it turns out), L1 32 KiByte. >=20 > The curves here are for single thread benchmark > configurations with cpuset used to control which CPU is > used. I first noticed this via odd performance variations > in multithreading with more cores allowed than in use (so > migrations to a variety of cpus over time). >=20 > I explored all the CPUs (cores), not just what I plotted. > Only the one gets the odd performing memory access > structure in its curve. >=20 > FYI: The FreeBSD boot is UEFI/ACPI based for both systems, > not U-Boot based. >=20 Jon Nettleton has replicated the memory access performance issue on the one cpu via a different HoneyComb, running some Linux kernel, using tinymembench as the benchmark. =3D=3D=3D Mark Millard marklmi at yahoo.com ( dsl-only.net went away in early 2018-Mar)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?C0426887-59D9-4524-8542-8DA6DBAFF744>