Re: devel/electron25: Build failure on CURRENT with clang17 [build time]

From: Mark Millard <marklmi_at_yahoo.com>
Date: Wed, 20 Dec 2023 09:30:04 UTC
Dimitry Andric <dim_at_FreeBSD.org> wrote on
Date: Fri, 15 Dec 2023 11:00:48 UTC :

. . .
> I will take a look at the sources, because building
> these ports is pretty awful: [ 6% 2757/45534] after about an hour.


For the amd64 builder machine configuration that I now use:

[00:00:16] [01] [00:00:00] Builder starting
[00:00:16] [01] [00:00:00] Builder started
[00:00:16] [01] [00:00:00] Building devel/electron25 | electron25-25.9.8_1
[00:47:52] [01] [00:47:36] Finished devel/electron25 | electron25-25.9.8_1: Success
[00:47:53] Stopping 1 builders

In order to get that, I use the configuration:

A Ryzen 9 7950X3D (so 32 hardware threads via 16 cores) with 192 GiBytes of RAM
USE_TMPFS=all
ALLOW_MAKE_JOBS=yes
524288Mi in the swap partition (on PCIe Optane 905P media)
(The file system(s) are also on such media.)

(The large SWAP for RAM+SWAP is really for when the number of active
builders is more like the number of hardware threads. Combined with
use of ALLOW_MAKE_JOBS , that can make for large load averages,
such as maximum observed: 299.17, 236.59, 200.66 .)


For reference:

# uname -apKU
FreeBSD 7950X3D-ZFS 15.0-CURRENT FreeBSD 15.0-CURRENT #134 main-n266876-e183039f0882-dirty: Fri Dec  8 14:22:35 PST 2023     root@7950X3D-ZFS:/usr/obj/BUILDs/main-amd64-nodbg-clang/usr/main-src/amd64.amd64/sys/GENERIC-NODBG amd64 amd64 1500006 1500006

# ~/fbsd-based-on-what-commit.sh -C /usr/ports
a2aef21762c6 (HEAD -> main, freebsd/main, freebsd/HEAD) science/opensph: update 0.4.0 → 0.4.1
Author:     Yuri Victorovich <yuri@FreeBSD.org>
Commit:     Yuri Victorovich <yuri@FreeBSD.org>
CommitDate: 2023-12-18 18:02:39 +0000
branch: main
merge-base: a2aef21762c6fc498ec300b169a8ef8c87e7277a
merge-base: CommitDate: 2023-12-18 18:02:39 +0000
n643981 (--first-parent --count for merge-base)

===
Mark Millard
marklmi at yahoo.com