1. Forum
  2. DOVE-Net
  3. Synchronet Programming

  • src/conio/scale.c

    From Deuc¿@VERT to Git commit to main/sbbs/master on Tue Oct 22 13:51:54 2024
    https://gitlab.synchro.net/main/sbbs/-/commit/19e20fd9e27799f0c9bc77bd
    Modified Files:
    src/conio/scale.c
    Log Message:
    Decorate some arguments to allow compiler to optimize harder

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Deuc¿@VERT to Git commit to main/sbbs/master on Tue Oct 22 14:35:00 2024
    https://gitlab.synchro.net/main/sbbs/-/commit/30a163fae2b0e9fe34f08acb
    Modified Files:
    src/conio/scale.c
    Log Message:
    Don't use the C99 restrict keyword because MSVC is apparently still
    more than 25 years behind the C standard.

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Deuc¿@VERT to Git commit to main/sbbs/master on Mon Nov 4 19:49:46 2024
    https://gitlab.synchro.net/main/sbbs/-/commit/80b9fccc77dbfe64da1ff062
    Modified Files:
    src/conio/scale.c
    Log Message:
    Add most const decorations.

    Not sure what macOS is doing, but it looks like something weird is
    happening in interpolate_width... maybe some unexpected integer
    promotion?

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Deuc¿@VERT to Git commit to main/sbbs/master on Tue Nov 5 09:13:24 2024
    https://gitlab.synchro.net/main/sbbs/-/commit/76d03c231fcf2d6bd76a2bef
    Modified Files:
    src/conio/scale.c
    Log Message:
    Use uint64_t instead of double for interpolation.

    More effort to fix the macOS issue. This could potentially result
    in slight problens with the rightmost pixel, but will protect
    against any weird FP issues and -ffast-math concerns.

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Deuc¿@VERT to Git commit to main/sbbs/master on Sat Jan 4 21:51:35 2025
    https://gitlab.synchro.net/main/sbbs/-/commit/433e4147a08f1e6898b8006b
    Modified Files:
    src/conio/scale.c
    Log Message:
    Don't scale when width or height is zero.

    That's just silly.
    Found by scan-build

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Deuc¿@VERT to Git commit to main/sbbs/master on Sat Jan 4 22:30:16 2025
    https://gitlab.synchro.net/main/sbbs/-/commit/528bb7779e33ddafe1222730
    Modified Files:
    src/conio/scale.c
    Log Message:
    Return before getting scaling buffers.

    Found by scan-build

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Deuc¿@VERT to Git commit to main/sbbs/master on Sun Mar 15 16:01:53 2026
    https://gitlab.synchro.net/main/sbbs/-/commit/ff2e724b522779f6f00f1e27
    Modified Files:
    src/conio/scale.c
    Log Message:
    Fix aspect_reverse() using *x instead of *y for height calculations

    Two places in aspect_reverse() computed the height dimension from the
    x input instead of y: the no-aspect-ratio early return and the
    height-controls branch. Both should use *y to match the parallel
    structure of the width-controls branch.

    Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net
  • From Deuc¿@VERT to Git commit to main/sbbs/master on Sun Mar 15 16:01:53 2026
    https://gitlab.synchro.net/main/sbbs/-/commit/6c02f7d0b797a45eefde6016
    Modified Files:
    src/conio/scale.c
    Log Message:
    Optimize multiply_scale() with row memcpy for duplicate rows

    Instead of re-reading source pixels and rewriting each output row
    ymult times, write the first row then memcpy it for the remaining
    duplicates. For common scaling factors (e.g. 4x), this eliminates
    75% of the inner loop work and leverages SIMD-optimized memcpy.

    Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

    ---
    þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net