| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
|
|
|
|
closured code
|
|
|
|
|
|
|
|
|
|
that is ta0, 1 and 2-compatible
|
|
|
|
|
|
|
|
were modified and if the program is the same as before
|
|
for skipping preparation altogether when possible
|
|
further optimizations
|
|
|
|
|
|
|
|
|
|
Egl
|
|
|
|
resize the <canvas> element on the web page in 'generic' fashion without having to depend on SDL or X11 headers in C code. This function is intended to be used in conjunction with EGL to specify a custom pixel resolution for the canvas (EGL API does not have expressiveness to do arbitrary pixel sizes).
|
|
|
|
of glBufferSubData on large buffers (even when we use a small part of them)
|
|
|
|
we can actually support
|
|
|
|
|
|
|
|
|
|
|
|
buffers are being used. allows use of vertex data from arbitrary parts of the vertex data range, not just the beginning
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
with mouse button staying down after leaving, until reentering and clicking
|
|
Adding LLVM intrinsic llvm_uadd_with_overflow_i16.
|
|
Add missing sdl functions
|
|
compiled with emulation
|
|
With -O1, sometimes LLVM generates code which calls its intrinsic
llvm_uadd_with_overflow_i16, which was previously missing in
emscripten's library.
This commit adds that intrinsic, and also adds a test which
causes LLVM to call that intrinsic (only in 'o1'). This test
previously failed in 'o1' but now passes.
|
|
|
|
strndump function implementation
|
