Issues building FreeBSD-current (head f9668e4)
Mark Millard
marklmi at yahoo.com
Wed Apr 28 21:13:20 UTC 2021
On 2021-Apr-28, at 13:57, Mark Millard <marklmi at yahoo.com> wrote:
> On 2021-Apr-28, at 04:39, jake h <jakehfreebsd at gmail.com> wrote:
>
>> Hey everyone,
>> I have been having some issues building FreeBSD-current from src, on
>> freebsd-13 (build 1300500). It appears to stem from several files compiling
>> with -Werror , despite my specification in make.conf of -Wno-error . As
>> these files then have warnings, the make kernel exits. A log of the error
>> is attached.
>> System info:
>> Current version: 13.0 - STABLE, build 1300500
>> make.conf: CFLAGS= -Wall -Wextra -Wno-error
>> Attempting to update to: 14-CURRENT (Git head f8668e4)
>>
>
> Your use of the number from sys/sys/param.h :
>
> #define __FreeBSD_version 1300500 /* Master, propagated to newvers */
>
> is a bit problematical and does not identify a specific
> build or a specific commit. It would be incremented
> for any addition to the kernel Application Binary
> Interface, for example, but is left unchanged by most
> development activity.
>
> The 1300500 number was established in:
>
> author Glen Barber <gjb at FreeBSD.org> 2021-02-05 16:38:27 +0000
> committer Glen Barber <gjb at FreeBSD.org> 2021-02-05 16:38:27 +0000
> commit 3c6a89748a01869c18955d5e3bfcdf35f6705d26 (patch)
>
> The next change to the figure was 2 months later, in:
>
> author Konstantin Belousov <kib at FreeBSD.org> 2021-04-05 03:05:44 +0000
> committer Konstantin Belousov <kib at FreeBSD.org> 2021-04-23 11:14:06 +0000
> commit c3f97dd75a1c294c4f60f42b604ee8bcda17be09 (patch)
>
> It was changed to be:
>
> #define __FreeBSD_version 1300501 /* Master, propagated to newvers */
>
> All builds of any commit on the stable/13 branch between
> those use the 1300500 number. (On 2021-04-26 it was updated
> again, this time to 1300502 .)
>
> As for disabling treating things as errors . . .
>
> "man src.conf" reports, in part,
>
> WITHOUT_WARNS
> Set this to not add warning flags to the compiler invocations.
> Useful as a temporary workaround when code enters the tree which
> triggers warnings in environments that differ from the original
> developer.
>
> WITHOUT_WERROR
> Set this to not treat compiler warnings as errors. Useful as a
> temporary workaround when working on fixing compiler warnings.
> When set, warnings are still printed in the build log but do not
> fail the build.
>
> Direct manipulation of CFLAGS for this is not recommended
> and various Makefile contents may override the attempt to
> control CFLAGS (or part of its content).
>
> Other difficulties with having context to provide help
> . . .
>
> You did not include the text of the specific commands in the
> build that generated the output that you provided.
>
> My recent build of main (so: 14) was as of commit 1b00608b264f .
> The buildworld buildkernel sequence worked but was based on
> building from an environment running based on releng/13.0's
> release/13.0.0commit instead of on a stable/13 commit:
>
> # uname -apKU
> FreeBSD CA72_4c8G_ZFS 13.0-RELEASE FreeBSD 13.0-RELEASE #1 releng/13.0-n244733-ea31abc261ff-dirty: Wed Apr 28 05:45:27 PDT 2021 root at CA72_4c8G_ZFS:/usr/obj/BUILDs/13_0R-CA72-nodbg-clang/usr/src/arm64.aarch64/sys/GENERIC-NODBG-CA72 arm64 aarch64 1300139 1300139
>
> But I can not even tell the relationship of 1b00608b264f to
> what you tried to build:
>
> https://cgit.freebsd.org/src/log/?qt=range&q=f8668e4
>
> reports no match for your "Git head f8668e4". Nor does:
>
> https://github.com/freebsd/freebsd-src/commit/f8668e4
>
> find a match.
Ahh, I see the subject line lists f9668e4 instead.
That is 8 commits more recent than what I built (as
reported in my initial reply and shown again later
below):
https://cgit.freebsd.org/src/log/?qt=range&q=f9668e4 reports
(in part) . . .
Commit message (Expand) Author Age Files Lines
* Add rib_walk_from() wrapper for selective rib tree traversal. Alexander V. Chernikov 13 hours 2 -0/+38
* cxgbe(4): Add support for NIC suspend/resume and live reset. Navdeep Parhar 15 hours 6 -306/+1393
* geom_uzip(4): fix a typo Ceri Davies 16 hours 1 -3/+3
* nfscl: add check for NULL clp and forced dismounts to nfscl_delegreturnvp() Rick Macklem 21 hours 2 -3/+12
* nfsd: fix a NFSv4.1 Linux client mount stuck in CLOSE_WAIT Rick Macklem 23 hours 1 -0/+5
* e1000: Rework em_msi_link interrupt filter Kevin Bowling 23 hours 2 -18/+24
* [fib algo] Delay algo init at fib growth to to allow to reliably use rib KPI. Alexander V. Chernikov 23 hours 3 -33/+76
* riscv: Remove old qemu compatibility code Brandon Bergren 24 hours 2 -113/+0
* Merge llvm commits for kernel address and memory sanitizer support Dimitry Andric 26 hours 2 -4/+14
None of that looks like it should make a difference
vs. my buildworld buildkernel .
> For reference, for the main (so: 14) that I built . . .
>
> # ~/fbsd-based-on-what-freebsd.sh
> branch: main
> merge-base: 1b00608b264f1fac846e571e23d8ef7134834638
> merge-base: CommitDate: 2021-04-27 19:18:13 +0000
> 1b00608b264f (HEAD -> main, freebsd/main, freebsd/HEAD) Merge llvm commits for kernel address and memory sanitizer support
> n246360 (--first-parent --count for merge-base)
>
> and for what produced that output:
>
> # more ~/fbsd-based-on-what-freebsd.sh
> #! /bin/sh
> branch="`git branch --show-current`" \
> && echo "branch: $branch" \
> && base="`git merge-base freebsd/$branch HEAD`" \
> && git log --oneline --no-color $base..HEAD \
> && base_date="`TZ=UTC git log --format=fuller --date=iso-local --no-color $base^..$base | grep CommitDate:`" \
> && echo "merge-base: $base" \
> && echo "merge-base: $base_date" \
> && git log --oneline --no-color $base^..$base \
> && echo "n`git rev-list --first-parent --count $base` (--first-parent --count for merge-base)"
>
> When booted uname -apKU reports for the installed main build:
>
> # uname -apKU
> FreeBSD CA72_4c8G_ZFS 14.0-CURRENT FreeBSD 14.0-CURRENT #0 main-n246360-1b00608b264f-dirty: Wed Apr 28 01:22:56 PDT 2021 root at CA72_4c8G_ZFS:/usr/obj/BUILDs/main-CA72-nodbg-clang/usr/src/arm64.aarch64/sys/GENERIC-NODBG-CA72 arm64 aarch64 1400012 1400012
It still appears that the basic information required
to identify the issue in your context is likely missing.
===
Mark Millard
marklmi at yahoo.com
( dsl-only.net went
away in early 2018-Mar)
More information about the freebsd-hackers
mailing list