Jump to content
Joel Bodenmann

Release Beta v0.14

Recommended Posts

The µGFX-Studio Beta v0.14 has just been released and is now available from our downloads section. Project files created with the µGFX-Studio v0.13 are fully compatible.

Please leave all your questions, comments, feedback, feature request and similar below.

The changelog can be found at the bottom of the download page:

Share this post


Link to post
Share on other sites

Glad to hear that you got it working!
Note that the X11 driver that is used for the preview in the uGFX-Studio on Linux suffers from serious performance issues. The application will run a lot smoother and faster on your actual hardware!

Share this post


Link to post
Share on other sites

Hii

       I also have some more doubts regarding porting the application generated in studio onto an embedded board.

The board that i am using is atmel-SAMAD5 Xplained board with a display integrated to it.

Is the generated code enough to port onto board??

Can u help me with any procedure to follow while porting ???

Thanks & regards,

mano

 

Share this post


Link to post
Share on other sites
5 minutes ago, Joel Bodenmann said:

Glad to hear that you got it working!
Note that the X11 driver that is used for the preview in the uGFX-Studio on Linux suffers from serious performance issues. The application will run a lot smoother and faster on your actual hardware!

Ok Tectu. thank u

Share this post


Link to post
Share on other sites

The code generated by the uGFX-Studio is completely hardware independent. It will run on any hardware on which the actual uGFX library is running on. The µGFX library itself is can run on virtually any platform. Running it on your Atmel SAM is possible without any problems after porting.

Regarding the porting: It is not as hard as it might look at first glance. The porting interface is well documented and the countless examples help to have an idea how things can be implemented.
You can always post your questions on the forum. Most people here are very happy to help wherever they can. However, please note that we (the actual µGFX developers) only maintain this forum in our spare time. You might want to consider purchasing a support plan to get access to our support system where all your questions will be answered as fast as possible and as best as possible.

 

Share this post


Link to post
Share on other sites

To solve this problem in preview generator:

.build/obj/GFXLIB/src/gdisp/gdisp.o: In function `gdispGDrawArc':
ugfx_library/src/gdisp/gdisp.c:1728: undefined reference to `sin'
ugfx_library/src/gdisp/gdisp.c:1728: undefined reference to `cos'
ugfx_library/src/gdisp/gdisp.c:1728: undefined reference to `round'

You need to add '-lm' parameter for LDFLAGS. For example at line 172 in 'ugfx_studio_0.14_linux/ugfx_library/tools/gmake_scripts/compiler_gcc.mk' 

LDFLAGS  += $(patsubst %,-L%,$(LIBPATH)) $(patsubst %,-l%,$(patsubst -l%,%,$(LIBS))) -lm

 

 

Looks like you forget about  'clean' option for preview generator. 

Edited by Bezet

Share this post


Link to post
Share on other sites

Hello Bezet and welcome to the community!

Thank you for your feedback. We will add the corresponding 'clean' option and the ability to specify additional compiler and linker flags for the next release.

Share this post


Link to post
Share on other sites

compiler_gcc.mk is not the correct place to add -lm. That file is platform independent but -lm is only needed on some platforms and will create link errors if included on other platforms.

The correct place is in your own makefile in the LIBS define (I think that is what it is called).

Another way around it is to add support in your gfxconf.h file for the GMISC module fast fixed integer or fast float trig. The gdisp routines will then use those routines instead of the C library routines (from -lm). The GMISC fast float and fixed routines are many many times faster than the C library routines as they are table based. The fast fixed versions even provide trig support on cpus with no floating point support and can save considerable code size on those machines because floating point emulation layers may then not need to be loaded.

Share this post


Link to post
Share on other sites

Please do not hesitate to inform us about any other missing features you're facing. The µGFX-Studio is currently in beta and lacks many many many features. However, we are working constantly on improving it. Any kind of feedback is highly welcomed!

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.

×
×
  • Create New...