A bug in the STM32 LTDC Driver

[nathanwiebe]

Line 545 of gdisp_lld_STM32LTDC.c currently reads:

srcstart = LTDC_PIXELBYTES * ((gU32)g->p.x2 * g->p.y1 * + g->p.x1) + (gU32)g->p.ptr;

Given that the operator "* +" clearly compiles but is nonsense, the correct line of code should be:

srcstart = LTDC_PIXELBYTES * ((gU32)g->p.x2 * g->p.y1 + g->p.x1) + (gU32)g->p.ptr;

I think this is a plain and simple typo, but it has been there since uGFX 2.7 or earlier.  This means the srcstart (the starting address of the blit source) is not being calculated correctly.  I believe the reason this slipped passed the radar in v2.7 is that, for some unknown reason, this same calculation was repeated (without the typo) in line 578, where DMA2D->FGMAR was set to the source address.  So in 2.7, the only time the erroneous value was used was in cache flushing and invalidation.  In the latest (2.9 release and guthub) version, the erroneously-calculated srcstart is being used directly to set FGMAR (as it should), but due to the typo, the resulting blit regions are copying essentially from random areas as their source.

As for reproduction steps, any blit from a pixmap to the display will calculate incorrectly except for, I suppose, blits with a source of (0,0) and a few other lucky combinations.

I just thought I would report this to be helpful.

[Joel Bodenmann]

Thank you for bringing this to our attention. Any form of feedback & possible bug report is welcomed

I'll be working the STM32LTDC driver in an upcoming project (few days/weeks from now). I will look into it then.

[Joel Bodenmann]

Took a bit longer than expected to get to this but yeah: This is definitely a typo.
Fixed in commit 888c7e8640caf02082adc23c967c5c0ba49a5146

Thank you for reporting this!

There were some other improvements to the STM32LTDC driver in the meantime (which is why getting to this took a bit longer):