After debugging testautomation_surface, I have concluded that SW blending doesn’t work at least for me (big endian AmigaOS platform).
For example, if sR is 0 and dR is 255, we will get -255*sA casted to an unsigned value. Basically results are quite large numbers instead of the expected 0-255 range. Am I missing something here?
I think that was just trying to promote Uint8 to a larger type. Does it
work if you replace (unsigned) with (int)?On Tue, Jan 3, 2017 at 12:44 PM, capehill <juha.niemimaki at gmail.com> wrote:
After debugging testautomation_surface, I have concluded that SW blending
doesn’t work at least for me (big endian AmigaOS platform).
For example, if sR is 0 and dR is 255, we will get -255*sA casted to an
unsigned value. Basically results are quite large numbers instead of the
expected 0-255 range. Am I missing something here?
Unfortunately it doesn’t work for me. You probably have to cast sA to int, at least that helps here. Perhaps compiler promotes result to unsigned because sA’s type is unsigned? (SDL_blit_A.c)
Can you provide a patch on the bug?On Fri, Jan 6, 2017 at 11:26 AM, capehill <juha.niemimaki at gmail.com> wrote:
Unfortunately it doesn’t work for me. You probably have to cast sA to int,
at least that helps here. Perhaps compiler promotes result to unsigned
because sA’s type is unsigned? (SDL_blit_A.c)
Sam Lantinga wrote:> Can you provide a patch on the bug?
On Fri, Jan 6, 2017 at 11:26 AM, capehill <@capehill (@capehill)> wrote:
Unfortunately it doesn't work for me. You probably have to cast sA to int, at least that helps here. Perhaps compiler promotes result to unsigned because sA's type is unsigned? (SDL_blit_A.c)
Sam Lantinga wrote:
Okay, can you try this?https://hg.libsdl.org/SDL/rev/6285a56f6729 (https://hg.libsdl.org/SDL/rev/6285a56f6729)
On Fri, Jan 6, 2017 at 11:26 AM, capehill <> wrote:
Quote:
Unfortunately it doesn’t work for me. You probably have to cast sA to int,
at least that helps here. Perhaps compiler promotes result to unsigned
because sA’s type is unsigned? (SDL_blit_A.c)