Date: Sat, 8 May 2021 12:40:05 -0700 From: Mark Millard <marklmi@yahoo.com> To: Adriaan de Groot <adridg@freebsd.org> Cc: "freebsd-arm@freebsd.org" <freebsd-arm@freebsd.org> Subject: Re: Rock64 flaky ethernet? Message-ID: <380C949F-A02E-42B9-BBFE-C265047BDE76@yahoo.com> In-Reply-To: <78C848CC-08E2-44C7-9C09-51A32C232691@yahoo.com> References: <3976804.O2WMGSuNBG@beastie.bionicmutton.org> <78C848CC-08E2-44C7-9C09-51A32C232691@yahoo.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 2021-May-8, at 12:19, Mark Millard <marklmi at yahoo.com> wrote: > On 2021-May-8, at 10:42, Adriaan de Groot <adridg at freebsd.org> = wrote: >=20 >> I've got an old-ish (2018 .. v2) Rock64 that hasn't been used much. I = have=20 >> fetched and booted it into the 13-RELEASE image: >>=20 >> FreeBSD-13.0-RELEASE-arm64-aarch64-ROCK64.img.xz 452690956 = 2021-Apr-09 06:46 >>=20 >> I have written this to an 8GB generic micro-SD card. For good = measure, I have=20 >> written a recent Armbian image to the same brand and age of micro-SD = card, so=20 >> I have comparison material. (I could also write everything to the = 32GB eMMC=20 >> chip that's in the Rock64, if needed, but I'm holding off on that = until I have=20 >> something that works). >>=20 >> After boot, dwc0 gets an address from DHCP, and I start pinging = 10.0.0.2 in my=20 >> home network. I get ~3% packet loss if I'm lucky, more often around = 7%. If I'm=20 >> pinging a machine in the network and also pinging my Rock64 from that = machine,=20 >> I hit around 22% packet loss on both sides. >=20 > For my mid-March non-debug build of main [so: 14] I get, > for example, (.149 is the Rock64 V2.0; .120 is a > ThreadRipper 1950X, both going at the same time > here): >=20 > --- 192.168.1.149 ping statistics --- > 72 packets transmitted, 72 packets received, 0.0% packet loss > round-trip min/avg/max/stddev =3D 0.136/0.230/1.903/0.279 ms >=20 > --- 192.168.1.120 ping statistics --- > 81 packets transmitted, 81 packets received, 0.0% packet loss > round-trip min/avg/max/stddev =3D 0.211/0.241/0.286/0.013 ms >=20 > So, no problem observed. >=20 > The root file system is on a USB3 SSD. A removable eMMC is > involved in the booting but no microsd card is present. >=20 > For reference: >=20 > # ~/fbsd-based-on-what-freebsd-main.sh=20 > FreeBSD Rock64_RPi_4_3_2v1p2 14.0-CURRENT FreeBSD 14.0-CURRENT = mm-src-n245445-def0058cc690 GENERIC-NODBG arm64 aarch64 1400005 1400005 > def0058cc690 (HEAD -> mm-src) mm-src snapshot for mm's patched build = in git context. > merge-base: 7381bbee29df959e88ec59866cf2878263e7f3b2 > merge-base: CommitDate: 2021-03-12 20:29:42 +0000 > 7381bbee29df (freebsd/main, freebsd/HEAD, pure-src, main) cam: Run all = XPT_ASYNC ccbs in a dedicated thread > n245444 (--first-parent --count for merge-base) >=20 >> Armbian works fine - poweroff, swap SD card, power on - in the same = setup. >>=20 >> With this kind of packet loss I can't really do anything with the = board in=20 >> FreeBSD; `pkg bootstrap` for instance seems to sit there forever. >>=20 >> .. ok, digging in the archives gets me lots of threads from june/july = 2020=20 >> about dwc timing, I'll dig through those first before carrying on. >>=20 >> [ade] >>=20 >> PS. Pine H6 never did catch on in FreeBSD-land, did it. Pity, I've = got a=20 >> couple that I'd be happy enough to use instead of the Rock64 if they = worked.=20 >>=20 >=20 For reference as far as U-Boot goes: U-Boot 2021.04 (Apr 09 2021 - 17:05:14 +0000) Model: Pine64 Rock64 DRAM: 4 GiB PMIC: RK8050 (on=3D0x40, off=3D0x00) MMC: mmc@ff500000: 1, mmc@ff520000: 0 It is from a sysutils/u-boot-rock64 build with no local patches. Previously 2020.10 was in use, also without problems. (But one of my non-debug builds of main was in use.) =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?380C949F-A02E-42B9-BBFE-C265047BDE76>