emulators/qemu: Enable user networking by default The current build system doesn't build slirp/user networking by default, but a LOT of guides on the internet to get started with qemu uses it. This modifies the configure script to add a new option to explicitly enable it. Reviewed_by: koobs (ports) Approved by: ??? Differential_Revision: D27778 MFH: <branch|no> <reason>
Details
- portlint: ?? (<summary output)
- testport: ?? (poudriere: <versions>, <archs> tested)
Diff Detail
- Repository
- rP FreeBSD ports repository
- Lint
No Lint Coverage - Unit
No Test Coverage - Build Status
Buildable 35699 Build 32588: arc lint + arc unit
Event Timeline
@adrian Can you confirm QA passes (portlint/poudriere, filling out TEST PLAN section) and add MFH: value <branch|no> <reason> in commit log message
Also Question: do any of the other qemu-* ports need to have this change made too, or do they all inherit/child port off the main one (I haven't looked)
For future reference sake: Ports compliance/recommendations require ports to 'unbundle' any bundled libraries (in this case slirp) except in cases where that is creates an unnecessary or inappropriate support or maintenance burden (such as go / node modules). We'll land this change in the short term until a slirp port can be produced
emulators/qemu/files/patch-configure | ||
---|---|---|
30–31 | Probably best to set =internal in CONFIGURE_ARGS (so its explicitly defined/set in Makefile), rather than here, unless upstreams configure is missing a default and this is going to go upstream? |
hi!
Can I just try to land this one first, and worry about the others later? I'd at least like to see a working qemu port with user networking available for other freebsd users to test with.
emulators/qemu/files/patch-configure | ||
---|---|---|
30–31 | I'm not sure what to send upstream yet; so I'm just making --enable-slirp do a default thing. Otherwise I also have to add a bunch of other sanity checking :-) |
Portlint output:
adrian@gertrude:~/work/freebsd/ports/head/emulators/qemu % env PORTSDIR=/home/adrian/work/freebsd/ports/head portlint
WARN: Makefile: [75]: NOPORTDOCS found. Consider using PORT_OPTIONS:MDOCS.
WARN: Makefile: Consider adding support for a NLS knob to conditionally disable gettext support.
WARN: Makefile: unless this is a master port, MAINTAINER has to be set by "=", not by "?=".
WARN: Makefile: unless this is a master port, COMMENT has to be set by "=", not by "?=".
WARN: Makefile: you may want directory for dependency sphinx-build to be textproc/py-sphinx@${PY_FLAVOR}
WARN: Makefile: "ONLY_FOR_ARCHS" has to appear earlier.
WARN: Makefile: "USES" has to appear earlier.
0 fatal errors and 7 warnings found.
Sure, let's just follow this changeset up in a timely manner so all ports are consistent
Re PORTLINT output, these ones at least should be addressed:
WARN: Makefile: you may want directory for dependency sphinx-build to be textproc/py-sphinx@${PY_FLAVOR} WARN: Makefile: "ONLY_FOR_ARCHS" has to appear earlier. WARN: Makefile: "USES" has to appear earlier.
Does poudriere testport pass?
Shouldn't that be done via another commit? I'm not trying to take ownership of the port, I'm just trying to fix a glaringly missing feature. :-)
Does poudriere testport pass?
I dunno! I don't have a full poudriere setup here! I build them one at a time.
In the absence of poudriere (one should set it up), you can run the set of commands outlined here:
https://www.freebsd.org/doc/en_US.ISO8859-1/books/porters-handbook/porting-testing.html
The delta between this set of commands and poudriere is the isolation of the build environment from the local system (which can pickup several additional classes of issues)