That was the gcc manual:
https://gcc.gnu.org/onlinedocs/gcc-4.9.0/gcc/Optimize-Options.html#Optimize-Options
quote:
"Link-time optimization does not work well with generation of debugging
information. Combining -flto with -g is currently experimental and expected
to produce unexpected results. "
Yes, the x64 target.
--
Precog is a next-generation analytics platform capable of advanced
analytics on semi-structured data. The platform includes APIs for building
apps and a phenomenal toolset for data science. Developer
Substituting _wstat with _wstati64, works too of course.
--
Precog is a next-generation analytics platform capable of advanced
analytics on semi-structured data. The platform includes APIs for building
apps and a phenomenal
Hi, I have identified a pretty nasty bug in the MinGW64 headers.
The case that follows happens for _USE_32BIT_TIME_T undefined
_wstat is defined as _wstat64i32 in _mingw_stat64.h
This function expects a stat argument, and judging from the name the
argument should be of type _stat64i32 instead, s
May be related to openmp hang that I have reported earlier on this list. I
can repost the example if needed.
--
Precog is a next-generation analytics platform capable of advanced
analytics on semi-structured data. The platf
Hmmm, OK, it looks like 'ended' is set in the code after all in
pthread_create_wrapper, after the thread function exits. I'll keep
investigating this.
--
Monitor your physical, virtual and cloud infrastructure from a singl
Hi, time to look into this issue again.
I built a debug version of winpthreads but as you can understand
debugging this is a PITA.
Hang seems to happen around line 1539 of src/thread.c of winpthreads.
This reads like:
if (tv->ended == 0)
WaitForSingleObject (tv->h, INFINITE);
Trying to fig
Usually __MINGW64__ is defined so you can check for that in the source.
--
Don't let slow site performance ruin your business. Deploy New Relic APM
Deploy New Relic app performance management and know exactly
what is happe
Hi niXman, no it won't cause a link error because the #pragma just
gets ignored so essentially there is no parallelism, hence, no hang.
It is interesting though that when using -fopenmp, the -lpthread flag
is unnecessary. Maybe when using openmp links some other version of
pthread somehow?
Just fo
gcc -v ?
--
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security, mobi
Hi, any news on this issue? Can I help in some way?
--
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discu
All these libraries are not 'built in' since MinGW-w64 is a compiler,
not a library collection, but they are compatible with MinGW-w64, you
just have to use MSYS to compile them. You can find quite a few of
them precompiled for MinGW-w64 in the blender lib folder
(https://svn.blender.org/svnroot/bf
It is reproducable with just one thread too. This should be the simpler case:
#include
#include
#include
void *do_thread(void *) {
int niterations = 200;
int i;
#pragma omp parallel for
for (i = 0; i < niterations; i++)
{
printf (
Hi, I think I have figured it out:
The issue may be that internally openmp may be allocating thread
identifiers that may already be used by pthread so pthread_join
somehow ends up waiting for the parent thread to finish, deadlocking
itself.
Try this:
#include
#include
#include
void *do_threa
Indeed. I do not have much experience with thread programmng unfortunately. :/
In any case trying to debug this from within blender is quite hard
because I can't get a stacktrace. A disassembly shows that the problem
occurs within pthread_create_wrapper. Here is a small disassembly
around the cras
Just as a note, removing the 'private' openmp directive also crashes.
--
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers
Hi, I am looking into openmp-related crashes that we've had into
blender. I see that the issue is known (
http://www.mail-archive.com/mingw-w64-public@lists.sourceforge.net/msg03982.html
) and that people from this mailing list have requested test files to
debug so I am sending a small program that
This looks exactly like what we need, thanks a lot :)
--
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Disc
Hi, I have found traces of a discussion in the MinGW mailing lists about
including w_char support in fstream. Since we do support wide (and,
consequently special unicode) characters in blender and the lack of such
functionality is evident on our MinGW64 builds, I'd like to know the state
of affairs
We also experience an issue with boost::filesystem::exists in blender. If
this fixes that it will be a blessing :).
--
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
thre
On 22 May 2012 20:10, Antony Riakiotakis wrote:
> I mean in terms of processors supported by the compiler itself. As far
> as operating systems go we support up to windows 2000 I think? Not
> quite sure.
Bad phrasing, compiler supports all processors: I meant processors
that the compil
I mean in terms of processors supported by the compiler itself. As far
as operating systems go we support up to windows 2000 I think? Not
quite sure.
--
Live Security Virtual Conference
Exclusive live event will cover all
Hi,
Trying your latest 4.7 build here without problems (on an i5 :)
).Looks like indeed I had -march in mind when criticizing the use of
-mtune. The two are similar but -march impiles -mtune rather than the
opposite.
Personally I don't mind using a toolchain with sse3 support
requirements but we
Hi, thanks for the quick response. I will try to get a backtrace
though last time I tried I remember I got some assembly mumbo-jumbo,
related to openmp for certain(If memory serves right, something like
omp_get_thread_name hit a null pointer or something similar).
Unfortunately I don't have a unix
Hi Ruben, I have been looking into some openmp crashes we've had with
blender and i have read the info in your first email on this thread.
It looks like we may have been experiencing the same problem
unfortunately. We are using the ray-linn build so if the default
behaviour is this for every MinGW6
Ah, looks like the reason for the strange behaviour was my using
release with debug info build type for cmake. Optimizations play funky
with the debugger it seems. Now all is in order...phew.
So, summarizing again for reference:
* Make sure python 2.7 is installed
* Make sure you don't enable any
Hmmm...I really can't say.. unfortunately, openmp destabilizes the
build (crash when rendering subdivision modified objects) so it may be
causing the error with threading on your build. Also we bundle the
pthread library that came from a different MinGW64 build, which may
cause the issues with thre
Looks like I managed to make it work with QT too, sorry for the fuss.
Still when tracing with gdb the cursor seems to hop randomly in the
source...I don't know why that may be unfortunately :/
--
Live Security Virtual Conf
Hi,
First, good news:
MinGW-w64 support for blender is getting more complete by the day, to
the point where users download and test (and may even prefer, due to
the speed increase ;) ) mingw-w64 builds. There have been a few
compiler-related issues but more like gcc-related than MinGW-64
related I
OK, thanks!
--
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and
threat landscape has changed and how IT managers can respond. Discussions
will include endpoint security,
I forgot some essential information:
build instructions:
http://wiki.blender.org/index.php/Dev:2.5/Doc/Building_Blender/Windows/MinGW/CMake
http://wiki.blender.org/index.php/Dev:2.5/Doc/Building_Blender/Windows/MinGW/Scons
homepage of the project:
www.blender.org
trunk build r45947 (almost vers
Hi, for 32-bit versions we use the 'official' MinGW pojects for the
reasons outlined above (latest mingw-get-inst gets the necessary
packages and is expected to run well). There are still no serious
performance comparisons apart from one build here:
http://graphicall.org/918
where downloaders hav
Whew...that was an ultra intensive few days to get everything compiled
and running.
For anyone interested this is the commit log:
http://lists.blender.org/pipermail/bf-blender-cvs/2012-April/045077.html
Since this initial commit, support for Scons has also been added.
Blender needs a lot of ext
Forgot to mention the separate build that I used:
http://sourceforge.net/projects/mingw-w64/files/Toolchains%20targetting%20Win64/Personal%20Builds/ray_linn/GCC-4.7.0-with-ada/
Sorry for polluting the list.
--
For Develo
Hi, trying the
https://sourceforge.net/projects/mingwbuilds/files/windows-host/4.7.0/
build on ilmbase-1.0.1 seems to produce errors when using make that
the recent builds do not produce (maybe string limit related?).
When I tried using a separate build with the libraries produced by the
aforemen
This looks extremely interesting. I will check this out for certain :)
--
Better than sec? Nothing is better than sec when it comes to
monitoring Big Data applications. Try Boundary one-second
resolution app monitoring to
I am thinking this may be problematic if we have to maintain a
compiler that doesn't have a specific tested build that blender will
compile against. I was thinking maybe include an official version
build in our servers (with all the credentials of course) if others
agree. I will let you know how th
Hi, thanks for the reply!
I know there are nightly and personal builds but was refering to
official builds of official releases.
Maybe I oversaw them but couldn't locate these in the binary list.
About libraries I meant clashes in 32bit vs 64bit libraries when
linking with the compilers as the li
Hi, I am interested in trying to add support for MinGW64 in blender3D
but I feel there are some issues that will better be answered by
asking here.
1) What is the state of the MinGW64 compiler? Is it considered stable
enough to support in a big project such as blender?
2) Currently it's quite a pr
39 matches
Mail list logo
