https://bugs.documentfoundation.org/show_bug.cgi?id=39179
Buovjaga changed:
What|Removed |Added
CC||todven...@suomi24.fi
--- Comment #48
https://bugs.documentfoundation.org/show_bug.cgi?id=39179
Telesto changed:
What|Removed |Added
Keywords||filter:docx
--
You are receiving this
https://bugs.documentfoundation.org/show_bug.cgi?id=39179
kompilainenn <79045_79...@mail.ru> changed:
What|Removed |Added
CC|
https://bugs.documentfoundation.org/show_bug.cgi?id=39179
--- Comment #46 from Xisco Faulí ---
(In reply to Telesto from comment #45)
> Still a repro. However not at the same extend as the initial report (+/-60
> seconds)
> Version: 5.5.0.0.alpha0+
> Build ID:
https://bugs.documentfoundation.org/show_bug.cgi?id=39179
Xisco Faulí changed:
What|Removed |Added
Keywords||perf
https://bugs.documentfoundation.org/show_bug.cgi?id=39179
--- Comment #45 from Telesto ---
Still a repro. However not at the same extend as the initial report (+/-60
seconds)
Version: 5.5.0.0.alpha0+
Build ID: ec79f3453471ee9b6ae32e71ff16ea99d9b7751c
CPU threads: 4; OS:
https://bugs.documentfoundation.org/show_bug.cgi?id=39179
Telesto changed:
What|Removed |Added
Blocks|88173 |104450
Referenced
https://bugs.documentfoundation.org/show_bug.cgi?id=39179
Telesto changed:
What|Removed |Added
Blocks||88173
Referenced
https://bugs.documentfoundation.org/show_bug.cgi?id=39179
--- Comment #44 from Michael Meeks ---
Sure - performance measurements on dbgutil builds are -totally- unhelpful =)
only a product build should ever be used to measure or assess performance; so
I'm happy to
https://bugs.documentfoundation.org/show_bug.cgi?id=39179
--- Comment #43 from Julien Nabet ---
I tested with LO Debian package 5.1.3.2, I confirm what Jean-Baptiste
indicated, it takes less than 20s to load the 50 pages version.
Also, when I kept on going up the calling
https://bugs.documentfoundation.org/show_bug.cgi?id=39179
--- Comment #42 from Julien Nabet ---
Trying to find the different paths, I noticed this part
5240 void DomainMapper_Impl::substream(Id rName,
5241 ::writerfilter::Reference::Pointer_t const& ref)
https://bugs.documentfoundation.org/show_bug.cgi?id=39179
--- Comment #41 from Michael Meeks ---
Interesting: so this:
123 for (sal_Int32 i : maAttributeTokens)
124 if (i == Token)
125 return true;
is I'm afraid reasonably
https://bugs.documentfoundation.org/show_bug.cgi?id=39179
--- Comment #40 from Julien Nabet ---
Created attachment 125095
--> https://bugs.documentfoundation.org/attachment.cgi?id=125095=edit
kcachegrind screenshot
I use dbg-util to try to retrieve the maximum of
https://bugs.documentfoundation.org/show_bug.cgi?id=39179
--- Comment #39 from Michael Meeks ---
Hi Julien.
Thanks so much for running that; but I guess:
> I retrieved it with master sources updated yesterday (enable-dbgutil)
dbgutil adds -so- much more debugging
https://bugs.documentfoundation.org/show_bug.cgi?id=39179
--- Comment #38 from Julien Nabet ---
I put a new callgrind trace here:
http://serval2412.free.fr/tdf_39179_callgrind.out.7206.bz2
I retrieved it with master sources updated yesterday (enable-dbgutil)
--
You are
https://bugs.documentfoundation.org/show_bug.cgi?id=39179
Jean-Baptiste Faure changed:
What|Removed |Added
CC|
https://bugs.documentfoundation.org/show_bug.cgi?id=39179
--- Comment #36 from tommy27 ba...@quipo.it ---
(In reply to tommy27 from comment #34)
(In reply to Astrix from comment #0)
Created attachment 49025 [details]
shorter 50 page docx document
still takes 1 minute and 40 seconds to
https://bugs.documentfoundation.org/show_bug.cgi?id=39179
Zangune zang...@users.sourceforge.net changed:
What|Removed |Added
CC|
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #34 from tommy27 ba...@quipo.it ---
(In reply to Astrix from comment #0)
Created attachment 49025 [details]
shorter 50 page docx document
still takes 1 minute and 40 seconds to load this file with LibO 4.3.4.1 and
recent 4.5.0.0
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #35 from Michael Meeks michael.me...@collabora.com ---
Might be worth building a new profile of a build with debug symbols on vs.
master - a ton of stuff changed in the XFastParser recently - and that has a
very helpful effect wrt.
https://bugs.freedesktop.org/show_bug.cgi?id=39179
Michael Stahl mst...@redhat.com changed:
What|Removed |Added
Version|3.4.3 release |Inherited From OOo
https://bugs.freedesktop.org/show_bug.cgi?id=39179
Michael Stahl mst...@redhat.com changed:
What|Removed |Added
Attachment #49033|0 |1
is obsolete|
https://bugs.freedesktop.org/show_bug.cgi?id=39179
Foss f...@openmailbox.org changed:
What|Removed |Added
Whiteboard||Confirmed:4.2.0.4:OSX
---
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #29 from bugquestcon...@online.de ---
I opened in LibO 4.1.4.2 on XP/SP3:
- 50 page file (123kB): in 46.9 sec; bug report: 2 min
- 119 page file (312kB): in 147 sec, bug report: 6 min
- 17 page file with 1.5 MB: in 4 sec; the bug
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #30 from Michael Meeks michael.me...@collabora.com ---
Julien your callgrind trace is 404 now (urgh) =) in general running this stuff
in callgrind (with symbols) and then looking at the results carefully, adding
printf's to the slow
https://bugs.freedesktop.org/show_bug.cgi?id=39179
Michael Stahl mst...@redhat.com changed:
What|Removed |Added
Attachment #49026|text/plain
https://bugs.freedesktop.org/show_bug.cgi?id=39179
Michael Stahl mst...@redhat.com changed:
What|Removed |Added
Attachment #49027|text/plain
https://bugs.freedesktop.org/show_bug.cgi?id=39179
the_lette...@outlook.com changed:
What|Removed |Added
Version|Inherited From OOo |3.4.3 release
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #26 from the_lette...@outlook.com ---
Regression sometime before the most recent Ubuntu LTS in 2012 -- the slow
documents are still slow as ever, but the 'fast' document is *also* now very
slow to open. Sigh. LibreOffice 3.5.7 on
https://bugs.freedesktop.org/show_bug.cgi?id=39179
tommy27 ba...@quipo.it changed:
What|Removed |Added
Summary|FileOpen: Writer takes very |FILEOPEN: Writer takes very
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #25 from Miklos Vajna vmik...@suse.cz ---
That is about unhandled tokens, not related to performance. If we want to
improve performance, we should first profile.
--
You are receiving this mail because:
You are the assignee for the
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #24 from Julien Nabet serval2...@yahoo.fr ---
I updated my local master repo and had the same.
I noticed lots of logs like this:
warn:writerfilter:18467:1:writerfilter/source/dmapper/DomainMapper.cxx:1479:
DomainMapper::attribute() -
https://bugs.freedesktop.org/show_bug.cgi?id=39179
tommy27 ba...@quipo.it changed:
What|Removed |Added
CC||ba...@quipo.it
--- Comment #23
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #22 from Michael Meeks michael.me...@novell.com ---
Thanks - firing it up in kcachegrind doesn't show any staggeringly obvious
single bottleneck. I suspect that the 2+ million calls to:
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #18 from Julien Nabet serval2...@yahoo.fr ---
Created attachment 69376
-- https://bugs.freedesktop.org/attachment.cgi?id=69376action=edit
some bts at random
I attached some bts taken at random
--
You are receiving this mail
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #19 from Michael Meeks michael.me...@novell.com ---
Hi Julien,
valgrind --tool=callgrind --simulate-cache=yes --dump-instr=yes ./soffice.bin
--splash-pipe=0 ~/compile-libreoffice/bugs/39179/callgrind.txt
So - the command-line you're
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #20 from Julien Nabet serval2...@yahoo.fr ---
(In reply to comment #19)
Hi Julien,
valgrind --tool=callgrind --simulate-cache=yes --dump-instr=yes
./soffice.bin --splash-pipe=0 ~/compile-libreoffice/bugs/39179/callgrind.txt
So
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #21 from Julien Nabet serval2...@yahoo.fr ---
I put the file available here:
http://serval2412.free.fr/callgrind.out.fdo39179.bz2
(I tried with http://wiki.documentfoundation.org/QA/Bugzilla-Attachments but it
doesn't seem to accept
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #14 from Miklos Vajna vmik...@suse.cz ---
Julien,
It makes little sense to reproduce a performance problem with a debug-enabled
build, which has optimization disabled. :-)
Anyway, those warnings are just about some ignored
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #15 from Julien Nabet serval2...@yahoo.fr ---
(In reply to comment #14)
It makes little sense to reproduce a performance problem with a
debug-enabled build, which has optimization disabled. :-)
If the goal is to compare time, I
https://bugs.freedesktop.org/show_bug.cgi?id=39179
Michael Meeks michael.me...@novell.com changed:
What|Removed |Added
CC|
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #17 from Julien Nabet serval2...@yahoo.fr ---
Michael: tried to follow your comment but after the 2 exports you quoted I
runned this:
valgrind --tool=callgrind --simulate-cache=yes --dump-instr=yes ./soffice.bin
--splash-pipe=0
https://bugs.freedesktop.org/show_bug.cgi?id=39179
Julien Nabet serval2...@yahoo.fr changed:
What|Removed |Added
CC||serval2...@yahoo.fr,
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #12 from Roman Eisele b...@eikota.de 2012-01-16 00:05:59 PST ---
Acknowledgement: Problem is still present in LibreOffice 3.4.4 and 3.5beta2
running on MacOS X 10.6.8 German.
--
Configure bugmail:
https://bugs.freedesktop.org/show_bug.cgi?id=39179
sasha.libreoff...@gmail.com changed:
What|Removed |Added
Platform|x86-64 (AMD64) |All
https://bugs.freedesktop.org/show_bug.cgi?id=39179
Björn Michaelsen bjoern.michael...@canonical.com changed:
What|Removed |Added
Status|NEW
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #7 from Astrix marais.han...@gmail.com 2011-09-21 01:40:17 PDT ---
Writer only takes so long to open a docx file, if the file has footnotes in it.
--
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
---
https://bugs.freedesktop.org/show_bug.cgi?id=39179
--- Comment #9 from Astrix marais.han...@gmail.com 2011-09-21 13:46:55 PDT ---
(In reply to comment #8)
Problem tested and confirmed with LibreOffice 3.4.3 (OOO340m1 (Build:302))
running on MacOS X 10.6.8, German, on a MacBook Pro with 2.53 GHz
48 matches
Mail list logo