* Use the commonly used function names for SHA-256 (Sigma, etc.)
* Use the same xxx_write for all, and avoid unwarranted memcpy
* Remove the bitcount
* Use nested ROR() speedup
* That's more like it! When compiled in 64-bit we get about the same speed as 7-zip
(slower when 32-bit, but shhh, or people will ask for a 64-bit version of Rufus...)
* Looks like what was holding us back was the sha256_write() from Brad Conte's
* Also fix WDK compilation and harmonize the BE->LE conversions
* SetChecksumAffinity() now spreads the affinity evenly between cores
* Also increase the read buffer size to help with performance
* At this stage, the only limiting factor regarding performance seems
to be the speed of the SHA-256 algorithm...
* Also update Bled to latest, as well as build scripts
* Note: Considering that Visual Studio 2015 is both freely and legally
available for anyone who wants to use it to compile Rufus, starting
with this commit, I will NOT be supporting any other version of Visual
Studio but 2015.
