About glyph error code

Question

Hi, EW team:

   When I was testing my multi-language project with PCB, the console showed error code as below:

    EMWI ERROR: code: 180, info1: "Fonts_GeneralHFMoBestSans_32", info2: 1602

   I searched the graphic engine source and find the code line 10134 of ewgfx.c. My platform package is RT1172 RGBA8888. I tried to deduce the cause of the error and found that a failure of finding font glyph resource in the function FindGlyph on Line 339 in ewextfnt.c can lead to it. After analyzing the function I guessed that the generated data structure containing font glyph resource should be the root cause.

  Please help see if it is like how I think and how to solve this problem.

 Thank you.

 

Best regards.

Stephen

Answer 1

Hi Stephen,

please let me know some more details:

• What version of Embedded Wizard Studio are you using?
• Which Platform Package are you using (according to your Profile setting)?
• Do you get the same error within the Prototyper?
• Can you reduce the issue to a small example (e.g. just one short text view using the mentioned glyph 1602)?

Best regards,

Manfred.

Comments

Hi, Manfred:

        The version of EW studio is 11. And the platform package is RT1172, VGLite.

      Because I take the responsibility of integration, I have little idea of how it is within the Prototyper.  I will give you more feed back when I get them.

 

Best regards.

Stephen

---

Hi, Manfred:
         Today I did some more debugging on this issue.  The console gave the error code as below.
EMWI ERROR: code: 182, info1: "Fonts_GeneralHFMoBest_Arab_32", info2: 65260
       After the error code occured, the related TextView did not show text. What was strange was that I could never reproduce the same error within the prototyper. So I guessed the issue might be related to resource configuration of embedded software.
      So I tried to enlarge the values below, especially the width.
#define EW_MAX_GLYPH_SURFACE_WIDTH       192
#define EW_MAX_GLYPH_SURFACE_HEIGHT      96
     And it worked.  
     The problematic  charactor  comes from Arabic word, such as مقفلة. I wonder what's the rationale under the issue. How should I tune this parameter to reach a optimized point for the embedded system?
     Thank you.

Best regards.
Stephen

---

Hi Stephen,

the glyph cache seems to be configured very small, so that not all glyphs (e.g. Arabic ligatures) can be loaded into the glyph cache. So increasing the both defines is the proper solution.

Do you have not enough memory on your system? Otherwise increasing the glyph cache will increase the performance, because it is not necessary to reload all glyphs while drawing text.

Within the Memory (RAM) usage window you can see the absolute minimum size of the glyph cache.

Best regards,

Manfred.