Before this commit, fixed point multiplication was implemented using an
assembly routine in a separate translation unit. This commit implements
this routine directly using inline assembly. By doing so, these
operations can be inlined when called from ARM code. Fixed point
division is implemented as well, along with various documentation and
style improvements.
option
This commit refactors string_stream and string_streamx into a common
basic_string_stream template class. When the EXT template == true, the
string_streamx formatting options are enabled, they default to disabled.
These options are enabled at compile time, and do not affect performance
when they are disabled. By implementing the two streams in this manner,
duplicated code is removed.
This commit also adds the ENDL template paramter. When ENDL is set to
zero, no endline character is printed when piping mtl::endl. Otherwise,
the character is printed. Defaults to '\n'. This allows logging on the
GBA to handle mtl::endl correctly and not print two newlines on the MGBA
emulator.
This implementation also writes the digits from left to right instead of
right to left. Using this method we can write the string to the
beginning of the buffer and still avoid reversing the string. It also
has the benefit of being slightly slower than the previous
implementation. The function's signature changed as well because there
is no longer a reason to pass the buffer size or a pointer to output the
start of the string.
Now uses a specialized implementation instead of the append multiple
characters implementation. Useful for appending single characters to
string streams (ex. newline).
CONFIGURE_DEPENDS option was added in CMake v3.12. This option allows the
build system to automatically re-run CMake if the glob changes, solving the
major issue with globbing. May have a performance impact, but it should be
negligible compared to the time spent building.
The istring and string_view operators have identical implementations. By
changing the istring operators to cast to string_view and use that
implementation instead, the number of redundant implementations is
reduced. This does incurr a small performance penalty, around 15 cycles
when tested on the MGBA Gameboy Advance emulator (uses an armv7tdmi).
When compared to the time operations take, the performance difference is
negligible. Ex. An insertion with two 8 character strings takes around
450 cycles.
