On Wed, 27 Jul 2011, Carsten Haitzler (The Rasterman) wrote:

> On Tue, 12 Jul 2011 10:01:29 +0900 cnook <kimci...@gmail.com> said:
>
>> Dear Mr. Vincent Torri,
>>
>> Hello,
>>
>> 1. The elementary_test (based on Rev 61152) shows with black screen. but,
>> the elementary_test of early of June (Rev 59900) shows properly.
>> 2. Would you please let me know why the elementary_test so slow. The
>> elementary_test gives log as bellow.
>>
>>   "WARNING: stucklock on image ######## (C:\MinGw\.../default.edj)"
>>   "WARNING: lock still there after 1 usec"
>
> #1 vtorri fixed, #2 - dont enable threads in evas - there is some unknown
> thread deadlock problem that shows up on windows. --disable-pthreads when on
> windows for evas's configure options. it means no async preload, so evas
> blocks, but we can't have everything. :) eventually this is going to get all
> revamps and rewritten/designed as part of a core rendering/imaging engine
> refactor.

#2 : we've discussed about that on IRC. The problem is in evas_common.h : 
when thrads are disabled :

# define LKT(x) 0

and in evas_image_scalecache.c, in 
evas_common_rgba_image_scalecache_prepare() :

    ret = LKT(im->cache.lock);

So ret is always 0, hence all the warnings.

raster, you proposed to define LKT(x) to 1. I don't know if it's a good 
idea or not. If it does not hurt, it will simplify the fix of the problem.

Of course, the best thing is to fix scalecache on Windows. But when I 
implement the code, and try expedite, there is a seg fault in 
evas_object_text.c, which seems to me completely unrelated to the 
scalecache stuff. I can paste the gdb log if you want. Unfortunately, 
there is no valgrind on Windows :/

Vincent

------------------------------------------------------------------------------
Got Input?   Slashdot Needs You.
Take our quick survey online.  Come on, we don't ask for help often.
Plus, you'll get a chance to win $100 to spend on ThinkGeek.
http://p.sf.net/sfu/slashdot-survey
_______________________________________________
enlightenment-devel mailing list
enlightenment-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/enlightenment-devel

Reply via email to