From: "Johannes Schindelin via GitGitGadget" <gitgitgadget@gmail.com>
To: git@vger.kernel.org
Cc: Johannes Schindelin <johannes.schindelin@gmx.de>
Subject: [PATCH v3 0/3] Optionally skip linking/copying the built-ins
Date: Wed, 26 Aug 2020 11:56:49 +0000 [thread overview]
Message-ID: <pull.411.v3.git.1598443012.gitgitgadget@gmail.com> (raw)
In-Reply-To: <pull.411.v2.git.1598283480.gitgitgadget@gmail.com>
The dashed form of the built-ins is so passé.
Incidentally, this also handles the .pdb issue in MSVC's install Makefile
target that Peff pointed out in the context of the "slimming down" patch
series
[https://lore.kernel.org/git/20200813145719.GA891370@coredump.intra.peff.net/]
.
This addresses https://github.com/gitgitgadget/git/issues/406
Changes since v2:
* Reworded and clarified the commit messages of the second and third patch.
Changes since v1:
* Fixed check-docs under SKIP_DASHED_BUILT_INS
* Renamed ALL_PROGRAMS_AND_BUILT_INS to ALL_COMMANDS_TO_INSTALL to reflect
its purpose better.
* Revamped the commit message of patch 2/3 and 3/3.
Johannes Schindelin (3):
msvc: copy the correct `.pdb` files in the Makefile target `install`
Optionally skip linking/copying the built-ins
ci: stop linking built-ins to the dashed versions
Makefile | 71 +++++++++++++++++++++------------------
ci/run-build-and-tests.sh | 2 +-
2 files changed, 40 insertions(+), 33 deletions(-)
base-commit: 878e727637ec5815ccb3301eb994a54df95b21b8
Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-411%2Fdscho%2Foptionally-skip-dashed-built-ins-v3
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-411/dscho/optionally-skip-dashed-built-ins-v3
Pull-Request: https://github.com/gitgitgadget/git/pull/411
Range-diff vs v2:
1: 1880a0e4bf = 1: 1880a0e4bf msvc: copy the correct `.pdb` files in the Makefile target `install`
2: 166bd0d8fb ! 2: 52deafded5 Optionally skip linking/copying the built-ins
@@ Commit message
executable to the dashed name in `libexec/git-core/`), in practice, it
is probably almost irrelevant.
- In fact, some platforms (such as Windows) only started gaining
- meaningful Git support _after_ the dashed form was deprecated, and
- therefore one would expect that all this hard-linking is unnecessary on
- those platforms.
+ However, we *do* care about keeping people's scripts working (even if
+ they were written before the non-dashed form started to be recommended).
+
+ Keeping this backwards-compatibility is not necessarily cheap, though:
+ even so much as amending the tip commit in a git.git checkout will
+ require re-linking all of those dashed commands. On this developer's
+ laptop, this makes a noticeable difference:
+
+ $ touch version.c && time make
+ CC version.o
+ AR libgit.a
+ LINK git-bugreport.exe
+ [... 11 similar lines ...]
+ LN/CP git-remote-https.exe
+ LN/CP git-remote-ftp.exe
+ LN/CP git-remote-ftps.exe
+ LINK git.exe
+ BUILTIN git-add.exe
+ [... 123 similar lines ...]
+ BUILTIN all
+ SUBDIR git-gui
+ SUBDIR gitk-git
+ SUBDIR templates
+ LINK t/helper/test-fake-ssh.exe
+ LINK t/helper/test-line-buffer.exe
+ LINK t/helper/test-svn-fe.exe
+ LINK t/helper/test-tool.exe
+
+ real 0m36.633s
+ user 0m3.794s
+ sys 0m14.141s
+
+ $ touch version.c && time make SKIP_DASHED_BUILT_INS=1
+ CC version.o
+ AR libgit.a
+ LINK git-bugreport.exe
+ [... 11 similar lines ...]
+ LN/CP git-remote-https.exe
+ LN/CP git-remote-ftp.exe
+ LN/CP git-remote-ftps.exe
+ LINK git.exe
+ BUILTIN git-receive-pack.exe
+ BUILTIN git-upload-archive.exe
+ BUILTIN git-upload-pack.exe
+ BUILTIN all
+ SUBDIR git-gui
+ SUBDIR gitk-git
+ SUBDIR templates
+ LINK t/helper/test-fake-ssh.exe
+ LINK t/helper/test-line-buffer.exe
+ LINK t/helper/test-svn-fe.exe
+ LINK t/helper/test-tool.exe
+
+ real 0m23.717s
+ user 0m1.562s
+ sys 0m5.210s
+
+ Also, `.zip` files do not have any standardized support for hard-links,
+ therefore "zipping up" the executables will result in inflated disk
+ usage. (To keep down the size of the "MinGit" variant of Git for
+ Windows, which is distributed as a `.zip` file, the hard-links are
+ excluded specifically.)
In addition to that, some programs that are regularly used to assess
disk usage fail to realize that those are hard-links, and heavily
overcount disk usage. Most notably, this was the case with Windows
- Explorer up until the last couple of Windows 10 versions.
+ Explorer up until the last couple of Windows 10 versions. See e.g.
+ https://github.com/msysgit/msysgit/issues/58.
- To save on the time needed to hard-link these dashed commands, and to
- eventually stop shipping with those hard-links on Windows, let's
+ To save on the time needed to hard-link these dashed commands, with the
+ plan to eventually stop shipping with those hard-links on Windows, let's
introduce a Makefile knob to skip generating them.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
3: ea23ba5e26 ! 3: 99a5328492 ci: stop linking built-ins to the dashed versions
@@ Commit message
the dashed form, even if we deprecated that a _long_ time ago.
For that reason, we just introduced a Makefile knob to skip linking
- them. TO make sure that this keeps working, teach the CI
+ them. To make sure that this keeps working, teach the CI
(and PR) builds to skip generating those hard-links.
+ This is actually not such a big change: e4597aae6590 (run test suite
+ without dashed git-commands in PATH, 2009-12-02) made sure that our test
+ suite does not require dashed commands. With this Makefile knob, the
+ commitment is just a little stronger (running tests with `--with-dashes`
+ would _still_ not see the dashed form of the built-ins).
+
+ There is a subtle change in behavior with this patch, though: as we no
+ longer even _build_ the dashed executables, running the test suite would
+ fail if any of Git's scripted commands (e.g. `git-request-pull`) still
+ This would have succeeded previously (and would have been unintentional,
+ of course) because `bin-wrappers/git` sets `GIT_EXEC_PATH` to the
+ top-level directory (which would still have contained, say,
+ `git-rev-parse`).
+
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
## ci/run-build-and-tests.sh ##
--
gitgitgadget
next prev parent reply other threads:[~2020-08-26 11:57 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-08-17 9:07 [PATCH 0/3] Optionally skip linking/copying the built-ins Johannes Schindelin
2020-08-17 4:55 ` Johannes Schindelin
2020-08-17 18:02 ` Junio C Hamano
2020-08-24 12:47 ` Johannes Schindelin
2020-08-24 18:42 ` Junio C Hamano
2020-08-25 8:07 ` Johannes Schindelin
2020-08-25 16:03 ` Junio C Hamano
2020-08-17 9:07 ` [PATCH 1/3] msvc: copy the correct `.pdb` files in the Makefile target `install` Johannes Schindelin
2020-08-17 9:24 ` Jeff King
2020-08-17 5:51 ` Johannes Schindelin
2020-08-17 21:37 ` Jeff King
2020-08-18 6:17 ` Johannes Schindelin
2020-08-17 9:07 ` [PATCH 2/3] Optionally skip linking/copying the built-ins Johannes Schindelin
2020-08-17 18:19 ` Junio C Hamano
2020-08-24 14:58 ` Johannes Schindelin
2020-08-17 9:07 ` [PATCH 3/3] ci: stop linking built-ins to the dashed versions Johannes Schindelin
2020-08-17 18:26 ` Junio C Hamano
2020-08-24 15:37 ` [PATCH v2 0/3] Optionally skip linking/copying the built-ins Johannes Schindelin via GitGitGadget
2020-08-24 15:37 ` [PATCH v2 1/3] msvc: copy the correct `.pdb` files in the Makefile target `install` Johannes Schindelin via GitGitGadget
2020-08-24 15:37 ` [PATCH v2 2/3] Optionally skip linking/copying the built-ins Johannes Schindelin via GitGitGadget
2020-08-24 19:02 ` Junio C Hamano
2020-08-25 8:20 ` Johannes Schindelin
2020-08-24 15:38 ` [PATCH v2 3/3] ci: stop linking built-ins to the dashed versions Johannes Schindelin via GitGitGadget
2020-08-24 19:06 ` Junio C Hamano
2020-08-25 8:30 ` Johannes Schindelin
2020-08-25 13:47 ` SZEDER Gábor
2020-08-25 15:42 ` Junio C Hamano
2020-08-26 4:19 ` Johannes Schindelin
2020-08-26 16:13 ` Junio C Hamano
2020-08-26 16:24 ` Junio C Hamano
2020-09-02 7:06 ` Johannes Schindelin
2020-09-02 20:50 ` Junio C Hamano
2020-08-27 8:30 ` SZEDER Gábor
2020-08-24 18:55 ` [PATCH v2 0/3] Optionally skip linking/copying the built-ins Junio C Hamano
2020-08-24 19:03 ` Jeff King
2020-08-24 19:51 ` Junio C Hamano
2020-08-26 11:56 ` Johannes Schindelin via GitGitGadget [this message]
2020-08-26 11:56 ` [PATCH v3 1/3] msvc: copy the correct `.pdb` files in the Makefile target `install` Johannes Schindelin via GitGitGadget
2020-08-26 11:56 ` [PATCH v3 2/3] Optionally skip linking/copying the built-ins Johannes Schindelin via GitGitGadget
2020-08-26 16:20 ` Junio C Hamano
2020-08-26 11:56 ` [PATCH v3 3/3] ci: stop linking built-ins to the dashed versions Johannes Schindelin via GitGitGadget
2020-09-03 10:45 ` SZEDER Gábor
2020-09-08 11:32 ` Johannes Schindelin
2020-09-08 11:48 ` SZEDER Gábor
2020-09-08 17:18 ` Junio C Hamano
2020-09-21 22:28 ` [PATCH v4 0/3] Optionally skip linking/copying the built-ins Johannes Schindelin via GitGitGadget
2020-09-21 22:28 ` [PATCH v4 1/3] msvc: copy the correct `.pdb` files in the Makefile target `install` Johannes Schindelin via GitGitGadget
2020-09-21 22:28 ` [PATCH v4 2/3] Optionally skip linking/copying the built-ins Johannes Schindelin via GitGitGadget
2020-09-21 22:28 ` [PATCH v4 3/3] ci: stop linking built-ins to the dashed versions Johannes Schindelin via GitGitGadget
2020-09-21 22:53 ` Junio C Hamano
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=pull.411.v3.git.1598443012.gitgitgadget@gmail.com \
--to=gitgitgadget@gmail.com \
--cc=git@vger.kernel.org \
--cc=johannes.schindelin@gmx.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.