Age | Commit message (Collapse) | Author |
|
sc, scaddins, sccomp, scripting
Change-Id: Ia99fec9e238033821cb784810edd4762c09bd5db
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/112049
Tested-by: Jenkins
Reviewed-by: Ilmari Lauhakangas <ilmari.lauhakangas@libreoffice.org>
|
|
Change-Id: If621bb2585fda4afef4be9a189c7f4c00008d358
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/104029
Tested-by: Noel Grandin <noel.grandin@collabora.co.uk>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Calc's dependency check done before parallel formula cell group
calculation tries to ensure valid cell values for all the dependencies
of the group's cell, and if it detects a problem such as a cycle
it bails out. But since ScFormulaCell::Interpret() simply bailed out
without doing anything, other cells could use that cell's possibly
incorrect value for their calculation and get their dirty flag reset.
This fix adds a flag to mark that bailing out is in progress, which
ensures the bail-out is short-circuited and no cell values are set.
Change-Id: Ia93c70d456682e19ce533abd2cf65ce35ffed9ca
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/96838
Reviewed-by: Dennis Francis <dennis.francis@collabora.com>
Tested-by: Jenkins
|
|
Change-Id: I335e0c5cf7944efa487e4535a9e6a5baab2f36dd
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/95140
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
This reverts commit fb1d3b580763a333bbbfe115d09e1b5cd8849675.
Now that we know that making fields has negative side effects
like disabling assignment operator generation.
Change-Id: Ib48334ffbeb2c768896dd8ced6818aa0b9910b0b
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/90333
Tested-by: Noel Grandin <noel.grandin@collabora.co.uk>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
which is much better for CPU cache, since the representation is more
compact, and since we almost always do insert() in pointer order,
there is surprisingly little sorting
Also add a count() method for compatibility with std::set and the
proposed std::flat_set
Change-Id: I2a3211dc59919cfec5cac1497530a4c3600d50ca
Reviewed-on: https://gerrit.libreoffice.org/c/core/+/87793
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I627e094a8c8565f29fa20cca288adb81b5a4fbd0
Reviewed-on: https://gerrit.libreoffice.org/82189
Reviewed-by: Julien Nabet <serval2412@yahoo.fr>
Tested-by: Julien Nabet <serval2412@yahoo.fr>
|
|
When a cycle of formula-groups is detected, do not conclude
that there is a circular dependency of cells. Only mark the
cells with Err:522 when all formula-groups in the cycle have
cleanly backed off from the dependency evaluation mode.
This commit also fixes places where we overlooked to back-off from
dependency evaluation mode on detection of a cycle of formula-groups.
Additionally mark formula-groups with self references as
"part of cycle" by setting mbPartOfCycle.
Unit tests for all these fixes are in a follow-up commit.
Change-Id: I57a88bbc88adf177d49768a5d4585b3e53729aea
Reviewed-on: https://gerrit.libreoffice.org/82074
Reviewed-by: Dennis Francis <dennis.francis@collabora.com>
Tested-by: Dennis Francis <dennis.francis@collabora.com>
|
|
Just before about to thread a FG, look to left and right for
"mutually" independent FG's with some restrictions and thread
this group of FG's together treating it as a single but longer
computation load.
For now the restrictions are :-
All formula-groups in a FG "group" must have :-
1. Same length
2. Same relative position.
3. Same weight.
This is very helpful in cases similar to the below :
There are lots of (say 32) consecutive formula-groups
all with same "small" length (say 8) and same weight.
By conventional formula-group-threading the speed-up is
limited to 8x even if we have a 256 core processor, but
with this threading-multiple-formula-groups patch
(in this case) we can get a speed-up of 256x provided
we have a >= 256 core machine. So effectively with this
patch the speed-up is now only limited to the number of
cells in a range consisting of mutually indepdendent
formula-groups rather than number of cells in each
formula-group.
Change-Id: Ib25b5abbb583fa207e8befff9a908d14313f3d51
Reviewed-on: https://gerrit.libreoffice.org/79485
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Change-Id: I78c4fb4acf21756f91582caee5e30e3ad1fc2ae4
Reviewed-on: https://gerrit.libreoffice.org/79543
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
So that e.g. unit tests can be easily run with OpenCL forced. This forces
even single cells to be evaluated using the forced method (many correctness
tests are just a single cell which normally would not be used for grouped
calculation).
Change-Id: If5c6e77a6e0d8696d5416d760cf5e47b8acf3d27
Reviewed-on: https://gerrit.libreoffice.org/63188
Tested-by: Jenkins
Reviewed-by: Luboš Luňák <l.lunak@collabora.com>
|
|
Change-Id: If326175d571d15752efd1b63df45b2bc785f7541
Reviewed-on: https://gerrit.libreoffice.org/61653
Tested-by: Jenkins
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Found with bin/find-unneeded-includes
Only removal proposals are dealt with here.
With this patch, find-unneeded-includes gives no more problems
Change-Id: I24e62be7a387d7fd97a36f54eed0ba9d83ddaf0c
Reviewed-on: https://gerrit.libreoffice.org/58667
Tested-by: Jenkins
Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk>
|
|
...cycles involving a mix of formula-groups and plain
formula-cells which are not grouped.
This fixes a crash in tdf#104213/1
Change-Id: I63b8b7379e4e5eaf322a3318f061a9e5fbc931ef
Reviewed-on: https://gerrit.libreoffice.org/57031
Tested-by: Jenkins
Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
|
|
software interpreter like in CPU threading.
This patch also reworks the cycle detection
to make it more robust.
Since the dependency computation also does
cycle detection, there is no need to disable
group-calc(threaded/OpenCL/SW Interpreter)
for non-leaf nodes in recursive interpret.
The rework of cycle detection ensures that
it fixes tdf#95748 correctly.
Change-Id: I460addb768eedc0914491a3d24ae7220c3afbb20
Reviewed-on: https://gerrit.libreoffice.org/55665
Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
Tested-by: Jenkins
|
|
...on indirect dependencies too.
Here a self reference to any formula-group
means if there are any references in a formula
(of the formula-group itself or any of its dependencies)
that points to any element inside the formula-group.
If there are any self-references, then that formula-group
can't be computed in parallel.
For example, with this patch we can detect the following case:-
Suppose the formula-group that we want to check is:
"=(F2+G2-10)*10.0" spanning A2:A100. Let the formula-group
starting at F2 be "=A1*0.1-10". The indirect dependency
formula-group starting at F2, references back the elements of
our original formula-group at A2. This makes the F.G at
A2 unsafe for parallel computation.
Concretly, this patch fixes a recalc crash on tdf#63638/1
Change-Id: I7b999a34571b191d2f70da6a3831f78b24a6b0a7
Reviewed-on: https://gerrit.libreoffice.org/54433
Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
Tested-by: Jenkins <ci@libreoffice.org>
|
|
Change-Id: I75c483972db787ea9c941eaac0df023f1929b09c
|
|
sorry, huge one...
|
|
Change-Id: I160f7af515d0a42f20fbc694fff7f39567ebeec5
Reviewed-on: https://gerrit.libreoffice.org/8062
Tested-by: LibreOffice gerrit bot <gerrit@libreoffice.org>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
|
|
Patches contributed by Herbert Duerr
i#118735 prevent endless loop if vlookup/hlookup doesn't find anything
http://svn.apache.org/viewvc?view=revision&revision=1239673
Patches contributed by Andre Fischer
remove lp_solver
http://svn.apache.org/viewvc?view=revision&revision=1199180
i#118160: Added external CoinMP library.
http://svn.apache.org/viewvc?view=revision&revision=1233909
Patches contributed by Armin Le-Grand
i#118485 - Styles for OLEs are not saved.
http://svn.apache.org/viewvc?view=revision&revision=1182166
i#118524: apply patch, followup fixes to 118485
http://svn.apache.org/viewvc?view=revision&revision=1186077
Patches contributed by lihuiibm
i#108860 - Fix range validation.
http://svn.apache.org/viewvc?view=revision&revision=1242846
i#118954 Chart data will lost after copy to different file
http://svn.apache.org/viewvc?view=revision&revision=1301345
Patches contributed by Ariel Constenla-Haile
Fix Linux build breaker: extra qualification on member
http://svn.apache.org/viewvc?view=revision&revision=1301591
i#118696 - i#118697 - Fix some Sheet Tab Color API issues
http://svn.apache.org/viewvc?view=revision&revision=1225428
i#118697 - Fix uninitialized variable
http://svn.apache.org/viewvc?view=revision&revision=1225859
i#118771 - ScUndoImportTab should preserve tab background color
http://svn.apache.org/viewvc?view=revision&revision=1230356
i#118921 - Repaint linked sheet tab background color after updating link
http://svn.apache.org/viewvc?view=revision&revision=1245177
i#118927 - Undo/Redo "Update Link" does not reset sheet tab color
http://svn.apache.org/viewvc?view=revision&revision=1245241
i#118747 - Copy tab color when transferring sheets across documents
http://svn.apache.org/viewvc?view=revision&revision=1230355
Patch contributed by Oliver Rainer-Wittman
i#118012 - methods <ScBroadcastAreaSlot::AreaBroadcast(..)> and
<ScBroadcastAreaSlot::AreaBroadcastInRange(..)> adapt stl-container
iteration in order to avoid destroyed iterators during iteration.
http://svn.apache.org/viewvc?view=revision&revision=1297916
Patches contributed by Mathias Bauer
gnumake4 work variously
http://svn.apache.org/viewvc?view=revision&revision=1394707
http://svn.apache.org/viewvc?view=revision&revision=1394326
http://svn.apache.org/viewvc?view=revision&revision=1396797
http://svn.apache.org/viewvc?view=revision&revision=1397315
Patch contributed by Daniel Rentz
calc69: #i116936# fix VBA symbol Cells
http://svn.apache.org/viewvc?view=revision&revision=1172135
Patches contributed by leiw:
i#118546 CPU 100% on switched off AutoCalculate with Conditional
Formatting on date values
http://svn.apache.org/viewvc?view=revision&revision=1301380
Re-add new function documentation. Many various cleanups.
Add missing calc66: #o11817313# also look at formula result
number format, remove redundant binaries.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Microsystems to Oracle; remove CVS style keywords (RCSfile, Revision)
|
|
2008/04/01 12:36:00 thb 1.5.34.2: #i85898# Stripping all external header guards
2008/03/31 17:13:33 rt 1.5.34.1: #i87441# Change license header to LPGL v3.
|
|
2008/01/17 17:31:33 er 1.4.498.1: #i32340# variable cell result ScFormulaResult, first wave
|
|
2005/10/04 21:05:54 dr 1.2.118.2: RESYNC: (1.2-1.3); FILE MERGED
2005/09/07 11:11:50 er 1.2.118.1: #i54276# include string results in iteration convergence tests
|
|
2005/09/05 15:00:55 rt 1.2.130.1: #i54170# Change license header: remove SISSL
|
|
2005/02/18 18:52:24 er 1.1.2.1: #i20986# #i21124# #i4874# #i7309# new algorithm for recursions and iterations
|