Age | Commit message (Collapse) | Author |
|
Change-Id: Icfdf8dddc4da20785e0881a95f50ee498e8405cd
|
|
* this and Ib4762f5a260035f00b5e68cf45b687fdf02e9c02 reduces a default
build on my machine from 2min25sec to 2min12sec
* without unitchecks, its down to 1min50sec now
* it reduces the build time on a i7-4770 Windows tinderbox from 99min to
89min
* by now it also takes care of avoiding most string copying, although
this can certainly be considered overeager given that the file
creation on Windows took ~250ms per file before and still will take a
lot longer than any string operation with this change
Change-Id: I515432bdefe2b055c78b6ba97868adbde65d9165
|
|
Change-Id: Ieb653adbd0cb4371ec5db57e70bcc551872f647f
|
|
Change-Id: I3c43fcfe7216a833f1dbe298098b72de52f0f155
|
|
Change-Id: Ifaa1637122d6f9cae1e29b77ac36ca5d1f220aed
|
|
Change-Id: I4e4d10c1c82834608b14aae62f02e5859d10114d
|
|
Change-Id: I6404472040f38c14ec7ca4a2c51be0d2f7f8199a
Reviewed-on: https://gerrit.libreoffice.org/2659
Reviewed-by: Fridrich Strba <fridrich@documentfoundation.org>
Tested-by: Fridrich Strba <fridrich@documentfoundation.org>
|
|
As suggested by Norbert
Change-Id: I1b23f2e0de8524db9e1c903863163a82a3965e41
|
|
Change-Id: I29de91f2eeb5c9317271aecf861f64a3c8eff73f
Reviewed-on: https://gerrit.libreoffice.org/2521
Reviewed-by: Michael Meeks <michael.meeks@suse.com>
Reviewed-by: Andras Timar <atimar@suse.com>
Tested-by: Andras Timar <atimar@suse.com>
|
|
Change-Id: I0e79f3e330fe1535c473d49010da34b04b9c5875
|
|
- do not use gb_UnpackedTarball_copy_header_files for boost
- adapt the optimization in concat-deps.c for new path
- use boost_headers in all LinkTargets that require it
- add explicit include paths to mysqlc, mysqlcppconn, libvisio, liborcus
Change-Id: I0c43e73ed43cc9d2e6bce8faf55e992d655a0bb9
|
|
...as it will do in one of the following commits.
Change-Id: Ic13c0d398943d34dffbc1a87efcac213d6699ca9
|
|
Change-Id: Iae93f7a81959a5d9c23e9bddb65e6154655374e6
|
|
Change-Id: I5e5d9b39cb0440644bf3ffb8dca4913dc123f672
|
|
Change-Id: I8aedf930bc007bf83f8111db9a5565668142321e
|
|
Should fix cases where for weirdly formatted input files the target was
mangled erroneously.
Change-Id: I28d94a6c714b5f893ac873f84fe2022e96cf1327
|
|
Change-Id: I541fb875d4d0bc46eb1bbb21793343c0283fd412
|
|
Mangle all included files in the UnpackedTarball directory in
the generated dependency files to point to the target of the
UnpackedTarball instead. This will cause all files that include them to
be rebuilt on every unpack, which is sound and much easier to maintain
than manually enumerating all internal header files somewhere.
Change-Id: Iaaa9b27450549d26b40d32ae0f740c2d5ef0edc5
|
|
In final (NDEBUG) builds assert expands to (more or less) nothing.
Change-Id: Ie6d43cac381975880c9db2b4a01bd9ed30e2714b
|
|
add error handling for fread, fwrite, and fscanf calls
use "fputs" instead of "fprintf" to avoid warning that it couldn't varify
the format string because it wasn't a string literal.
Change-Id: I1137160d28461fbac69c8e05d32ba010da491c57
Reviewed-on: https://gerrit.libreoffice.org/903
Reviewed-by: Petr Mladek <pmladek@suse.cz>
Tested-by: Petr Mladek <pmladek@suse.cz>
|
|
The call to elide_dependency in _process was broken because it checked
the full line, which ends with " : " usually, breaking the build with LO
customised GNU make on cygwin.
(regression from 140281898f21f707582f395918e5234bdb67e435)
Also handle empty lines following '\' better while at it.
Change-Id: Iadc6349e5b6930945e46e244bcbee37a632e0d71
|
|
Change-Id: I4b001e6fd963df2fad97aaf5885b0b38189e59d6
|
|
Change-Id: I18281f9cdf4c20ab4f939b2f7d3b0a6c3bf5387b
|
|
Also - verified that codemaker touches both .hdl and .hpp regardless
of whether the contents changed in anyway under gbuild.
Change-Id: I391e9cf952c61a6c59087cd7cd0ddc8c54a14df3
|
|
Change-Id: Ibd80b49eba4b259027d57845b8ebba3e35908e82
|
|
Change-Id: Ie321ada6d2c078881fdedc61d72c94034297fea3
|
|
Change-Id: I80bd0329e15cb365091ea6f5c7e1ef3559d681fc
|
|
After a string has been passed as key parameter to hash_store, it must
not be modified, as happens in the print_nodotdot and print_fullpaths
functions; if it doesn't crash then the dep files are twice as large
in a MSVC build because duplicates are not eliminated.
(regression from c2b467b84a81bd45ca9df1f7f07e2700fd6e396a)
Change-Id: I0ae96a7fbcefa20b118717d923f60b4f2255a642
|
|
Change-Id: If1d43f8f5e61f52c554bcdcf7f86639f1ea60046
|
|
* this helps getting nonbroken deps if using some ccache/CCACHE_BASEDIR combos
Change-Id: I401e55cc10b16aa5901348909237e97d05ab2761
|
|
Change-Id: Ie8e8f0b42b05a7d99fc81ad3a57e661758c12533
|
|
* make concat-deps use absolute paths whereever possible
* collapse /../ _after_ making the path absolute otherwise it will break for
relative paths starting with ../
* substitude SRCDIR variable, so trees can be moved
* this fixes subsequentcheck as it does run gbuild from a different work
directory in a few non-tail_build modules, which are then complaining about
missing/unresolvable deps
* this should fix troubles from module rebuild as relative paths were different
between tail_build and the module dir
Change-Id: I5a25e1f55bdc2b475df2af04b711fd808d95cdaf
|
|
|
|
|
|
|
|
|
|
|
|
|
|
This touch configure.in with a trivial space removal to force a
./bootstrap on the tinderboxes
This make ./bootstrap build a 'build-side' executable concat-deps
and modify LinkTarget.mk to use it instead of concat-deps.pl
The impact is a conservative x10 times improvement in both elapsed and
cpu time in overhead due to DEP generations.
time make -sr build in tail_build after removing
workdir/$INPATH/Dep/LinkTarget/Library/*
(all on Intel X3360, 2.83GHz, 4-core)
before:
elapsed: 116.5 cpu: 115.9
after
elapsed: 24.3 cpu: 23.9
base (no dep processing, i.e no rm of Dep/...)
elapsed: 17.5 cpu: 17.4
overhead ratio:
elapsed: 14.5x cpu: 15x
same with -j6
before:
elapsed: 47 cpu: 116
after:
elapsed: 20.8 cpu: 24
base:
elapsed: 17.5 cpu: 17.5
overhead ratio:
elapsed: 9x cpu: 15x
Note: for now the executable is generated directly in the source tree
(in solevn/bin ). That is not ideal. It and other similar polution
should be moved, to workdir most likely..
For now we just hide the mess under the .gitignore carpet
|
|
|
|
|
|
|