kernel: move GAP_KERNEL_{MINOR,MAJOR}_VERSION to version.h resp. gap_version.h #4674
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
fingolfin
added
topic: kernel
release notes: not needed
fingolfin
commented
Oct 18, 2021
fingolfin
commented
Oct 18, 2021
fingolfin
commented
Oct 18, 2021
fingolfin
force-pushed
the
mh/reduce-autoconf-usage-in-headers
branch
2 times, most recently
from
9329663
to
ac21b7d
Compare
|
Weird failure in the "Wrap release" CI scripts when building HPC-GAP out of tree. E.g. in this build log, we see: However, I cannot reproduce this on locally (yet). |
fingolfin
force-pushed
the
mh/reduce-autoconf-usage-in-headers
branch
from
dcc8847
to
58e7079
Compare
fingolfin
changed the title
config.h) in our header files
fingolfin
force-pushed
the
mh/reduce-autoconf-usage-in-headers
branch
from
58e7079
to
0fa0b0c
Compare
fingolfin
commented
Oct 25, 2021
fingolfin
force-pushed
the
mh/reduce-autoconf-usage-in-headers
branch
2 times, most recently
from
5fa5bd4
to
b28c8a0
Compare
The `vpath` directives are what makes out-of-tree builds possible. With
them, if a source file is not present in `${builddir}`, then GNU make
automatically looks for it in `${srcdir}`.
For files in `src`, this is generally what we want. But for generated
files in `build`, we do *not* want that. As a result, it could happen
that instead of `${builddir}/build/gap_version.c` we would access
`${srcdir}/build/gap_version.c`, and so on.
This patch fixes this by restricting the `vpath` directives to files
inside `src` subdirectory.
fingolfin
force-pushed
the
mh/reduce-autoconf-usage-in-headers
branch
from
1934161
to
5e2e37b
Compare
Also rename gap_version.c to version.c and likewise for the header file.
fingolfin
force-pushed
the
mh/reduce-autoconf-usage-in-headers
branch
from
5e2e37b
to
807ff98
Compare
7 tasks
wilfwilson
approved these changes
Oct 26, 2021
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The goal is to eventually move
#include "config.h"fromcommon.hinto all the.cand.ccfiles. This greatly simplifies the installation of headers as part ofmake install. It also means that the same set of headers can be shared by multiple binary builds of GAP; i.e. one could then install plain GAP, HPC-GAP and Julia-GAP in parallel.This is just a step towards this goal (not the last one, nor the first one), but we are getting pretty close now.
This is a subset of PR #4492.
UPDATE: most of this PR has now been merged via PR #4678. What remains is the code which moves
GAP_KERNEL_{MINOR,MAJOR}_VERSIONtoversion.hresp.gap_version.h. The reasons why I give two names is because I've switched around between the two: I prefer the former one as simpler, but I thought for a while that this was perhaps the reason behind the strange failures in the CI tests for the release process, which break down when building HPC-GAP, becauseversion.hresp.gap_version.hare not found (i.e., they are not generated). But apparently the name is not the cause for the breakage after all... very weirdCloses #4679.