Precise and IEEE Strictness in HLSL

Last week on HLSL there was relatively little activity, given the Memorial Day weekend and the awesome weather we had in Seattle. However, there were a few changes related to precise that I thought were worth commenting on. First, ‘precise’ doesn’t do anything to the numerical precision of any one operation. Instead, it indicates that…

0

Last Week on DirectX Shader Compiler (2017-05-24)

Last week we had bug fixes as codegen cleanup as pretty much every week, including a bunch of work on ViewID. We also improved support for external libraries, which you can now configure in dndxc via the Tools | Options menu, as well as specify via command-line arguments to dxopt. Two other interesting changes are…

0

Last Week on DirectX Shader Compiler (2017-05-16)

In addition to a whole bunch of fixes and improvements, last week was interesting because we saw the dxil-v1.1 branch come and go. We used the branch to build up some changes related to SV_ViewID that required some massaging to work with the validator made available in the Windows SDK, which is the set of…

0

Last Week on DirectX Shader Compiler (2017-05-09)

As per usual, lots of good stuff went in. Improved versioning; more components are starting to become aware (version-aware, that is, not the other kind), as we’d expect while we continue making progress while maintaining back compat. SPIR-V work continues. Fixes and codegen improvements. These are all very specific so they don’t always have a…

0

Last Week on DirectX Shader Compiler (2017-05-02)

“A Day Late Edition” Support for the next revision of DXIL is starting to trickle in, along with code generation improvements and a change in how htan(-inf) and htan(inf) work. DXBC didn’t have a built-in intrinsic for this, and the expansion produced NaNs. Two new SM 6.1 specifics were added last week. Barycentrics ViewID The…

0