> *An:* "darktable-devel@lists.sourceforge.net"
>
> *Betreff:* Re: [darktable-devel] dt only shows black screen at start
> make install isn't supposed to delete anything, so it's not its fault.
> You have to remove old files before installing new ones if you compi
etreff: Re: [darktable-devel] dt only shows black screen at start
make install isn't supposed to delete anything, so it's not its fault. You have to remove old files before installing new ones if you compile software yourself, like any package manager does. Otherwise you might get left with o
make install isn't supposed to delete anything, so it's not its fault. You have
to remove old files before installing new ones if you compile software
yourself, like any package manager does. Otherwise you might get left with old
stuff not present in new install which may screw things up. In the
On 01/05/15 20:09, Markus Jung wrote:
> Am 05.01.2015 um 22:54 schrieb Francois Connetable:
>> ...Last check: it really has to do with some sort of
>> sizeof(darktable_t) call - if I reintroduce two int32_t variable
>> with *any* name, the fix still works. So, somewhere a piece of code
>> assume th
Am 05.01.2015 um 22:54 schrieb Francois Connetable:
> ...Last check: it really has to do with some sort of sizeof(darktable_t)
> call - if I reintroduce two int32_t variable with *any* name, the fix
> still works. So, somewhere a piece of code assume that the size of
> darktable_t is different than
...Last check: it really has to do with some sort of sizeof(darktable_t)
call - if I reintroduce two int32_t variable with *any* name, the fix
still works. So, somewhere a piece of code assume that the size of
darktable_t is different than what it is once these two variables are
removed...
-Franco
Sorry for not reporting this earlier - I'm not familiar with darktable
code and certainly not an expert in C code... I did replay all changes
induced by the "faulty" commit and found out that the removal of
thumbnail_width and thumbnail_height variables declaration in
src/common/darktable.h is caus
Hello
With strace the program is filling terminal with endless lines such as
...
lseek(3, 154642432, SEEK_SET) = 154642432
read(3,
"\r\0\0\0\2\1E\0\2\234\1E\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"...,
1024) = 1024
lseek(3, 154643456, SEEK_SET) = 154643456
read(3,
"\r\0\0\0\
Latest lines from strace darktable:
fcntl(3, F_SETLK, {type=F_UNLCK, whence=SEEK_SET, start=1073741824,
len=1}) = 0
access("/home/urs/.config/darktable/library.db-journal", F_OK) = -1
ENOENT (No such file or directory)
fstat(3, {st_mode=S_IFREG|0644, st_size=1310720, ...}) = 0
lseek(3, 24, SEEK_
I've tried gdb, it shows no errors:
Starting program: /opt/darktable/bin/darktable
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
[init] moving database into new XDG directory structure
[New Thread 0x7fffeb2f3700 (LWP 2591)]
The json id!=NULL error I see is often facebook related.
I first noticed it when trying to export to facebook from darktable
Is the black, hung display perhaps related to thsi issue?
Anyone else happen to be exporting to facebook immediately before the
display issue began?
How do I generate more d
Hello
My screen is still black. AMD Phenom II X6 1075T, no OpenCl. Linux Mint
17, Mate 64x.
With respect,
Alexander Rabtchevich
johannes hanika wrote:
> thanks for bisecting, darktable-cli definitely had a bug writing to
> non-alloced memory there (fixed in master). still very much unsure
> a
thanks for bisecting, darktable-cli definitely had a bug writing to
non-alloced memory there (fixed in master). still very much unsure about
the main darktable program though.
-jo
On Sun, Jan 4, 2015 at 1:13 PM, thokster wrote:
>
> Remove thumbnail_width/height settings
> 0a00f6126a95984b959c19
No obvious difference with both.
Am 04.01.2015, 01:43 Uhr, schrieb Pedro Côrte-Real :
> Could you try changing the 1280 * 720 numbers in
> src/develop/pixelpipe_hb.c to higher (say 1920 * 1080) and lower (say
> 640 * 480) and see if that makes a difference?
>
> Thanks,
>
> Pedro
>
> On Sun, Ja
Could you try changing the 1280 * 720 numbers in
src/develop/pixelpipe_hb.c to higher (say 1920 * 1080) and lower (say
640 * 480) and see if that makes a difference?
Thanks,
Pedro
On Sun, Jan 4, 2015 at 12:13 AM, thokster wrote:
>
> Remove thumbnail_width/height settings
> 0a00f6126a95984b959c1
Remove thumbnail_width/height settings
0a00f6126a95984b959c19f5e4a1ecc89c8e4fed
introduces the issue over here
Am 03.01.2015, 23:36 Uhr, schrieb Francois Connetable
:
> FYI, reverting to a "randomly selected" commit from DEC24:
> af6ffb25e9b75f4b8fb5bcc24510c374e4e8d59b
>
> ...Everything
Probably this one:
Merge remote-tracking branch 'origin/cache2'
56eef68faf91f324f26fdd720c9c04a39b999ac9
Am 03.01.2015, 23:36 Uhr, schrieb Francois Connetable
:
> FYI, reverting to a "randomly selected" commit from DEC24:
> af6ffb25e9b75f4b8fb5bcc24510c374e4e8d59b
>
> ...Everything is fine a
FYI, reverting to a "randomly selected" commit from DEC24:
af6ffb25e9b75f4b8fb5bcc24510c374e4e8d59b
...Everything is fine again. So I'll do some further try to see if I can
find out the commit id that introduces the issue..
-Francois
Le samedi 03 janvier 2015 à 23:30 +0100, Francois Connetable
Just to confirm, I upgraded my ubuntu to "vivid". I'm now running
libgtk-3.14.6. I did a full rebuild of darktable -> still the same
issue. Will try to revert to a previous commit and see if I can get it
to run again...
Not sure it can explain but my system went through an update to latest
nvidia
3.14.4-0ubuntu1
Am 03.01.2015, 18:05 Uhr, schrieb johannes hanika :
> which version of libgtk-3 do you guys have where it hangs?
>
> -jo
>
> On Sun, Jan 4, 2015 at 4:28 AM, Francois Connetable <
> francois.conneta...@free.fr> wrote:
>
>> Same thing here (running on GNOME ubuntu latest, NVIDIA o
I have now updated to gtk 3.14.
darktable is still presenting a black screen.
Ubuntu 14.10, latest git pull.
Running from the command line with -d flag produces only
(darktable:19827): Json-CRITICAL **: json_node_get_object: assertion 'node
!= NULL' failed
Regards
Dave
On 3 January 2015 at 19:
Mine hangs.
I also have 3.12.2-0ubun amd64 installed.
-dave
On 3 January 2015 at 19:22, Francois Connetable wrote:
> Hi Jo,
>
> Assuming this is what your are looking for, a "dpkg -l libgtk-3-0" reports:
> 3.12.2-0ubun amd64
>
> Now it makes me wonder - didn't you say that 3.14 was the minimum
we /think/ 3.10 should suffice, but of course can't be sure without further
testing. gtk changed the threading model, and since these backtraces all
show some gdk_threads_enter() related stack traces that's my best guess so
far.
running gtk-3.14 just fine here..
-jo
On Sun, Jan 4, 2015 at 6:22 A
Hi Jo,
Assuming this is what your are looking for, a "dpkg -l libgtk-3-0" reports:
3.12.2-0ubun amd64
Now it makes me wonder - didn't you say that 3.14 was the minimum version
supported?
Thanks for your help,
-FrancoisLe 3 janv. 2015 18:05, johannes hanika a écrit :
>
> which version of libgt
which version of libgtk-3 do you guys have where it hangs?
-jo
On Sun, Jan 4, 2015 at 4:28 AM, Francois Connetable <
francois.conneta...@free.fr> wrote:
> Same thing here (running on GNOME ubuntu latest, NVIDIA openCL).
>
> Hope it helps,
> -Francois
> Le 3 janv. 2015 11:25, Alexander Rabtchevic
Same thing here (running on GNOME ubuntu latest, NVIDIA openCL).
Hope it helps,
-Francois
Le 3 janv. 2015 11:25, Alexander Rabtchevich
a écrit :
>
> Hello
>
> I can only see a dark rectangle of the program instead of its controls.
> Pressing L, D, F11 changes nothing. gdb shows no problems and
Ditto me.
Thought I'd broken something :(.
On 3 Jan 2015 12:28, "Alexander Rabtchevich" <
alexander.v.rabtchev...@gmx.net> wrote:
> Hello
>
> I can only see a dark rectangle of the program instead of its controls.
> Pressing L, D, F11 changes nothing. gdb shows no problems and the
> program does
Hello
I can only see a dark rectangle of the program instead of its controls.
Pressing L, D, F11 changes nothing. gdb shows no problems and the
program does nothing at least it does not consume processor power.
Current git. Deletion of library file hasn't helped.
With respect,
Alexander Rabtch
28 matches
Mail list logo
