Like SSE4-8 and SSE4-16, these use 8-bit and 16-bit values for mask
elements, respectively, and thus should generate the best code when used
for computation with datatypes of those sizes.
1. builtins/target-nvptx64.ll to write, now it is just a copy of target-generic-1.ll
2. add __global__ & __device__ scope
2. make code work for a single cuda thread
3. use tasks to work as a block grid and programIndex as laneIdx, programCount as warpSize
4. ... and more...
Along the lines of sse4-8, this is an 8-wide target for SSE4, using
16-bit elements for the mask. It's thus (in principle) the best
target for SIMD computation with 16-bit datatypes.
This change adds a new 'sse4-8' target, where programCount is 16 and
the mask element size is 8-bits. (i.e. the most appropriate sizing of
the mask for SIMD computation with 8-bit datatypes.)
There were a number of places throughout the system that assumed that the
execution mask would only have either 32-bit or 1-bit elements. This
commit makes it possible to have a target with an 8- or 16-bit mask.
Initial support for ARM NEON on Cortex-A9 and A15 CPUs. All but ~10 tests
pass, and all examples compile and run correctly. Most of the examples
show a ~2x speedup on a single A15 core versus scalar code.
Current open issues/TODOs
- Code quality looks decent, but hasn't been carefully examined. Known
issues/opportunities for improvement include:
- fp32 vector divide is done as a series of scalar divides rather than
a vector divide (which I believe exists, but I may be mistaken.)
This is particularly harmful to examples/rt, which only runs ~1.5x
faster with ispc, likely due to long chains of scalar divides.
- The compiler isn't generating a vmin.f32 for e.g. the final scalar
min in reduce_min(); instead it's generating a compare and then a
select instruction (and similarly elsewhere).
- There are some additional FIXMEs in builtins/target-neon.ll that
include both a few pieces of missing functionality (e.g. rounding
doubles) as well as places that deserve attention for possible
code quality improvements.
- Currently only the "cortex-a9" and "cortex-15" CPU targets are
supported; LLVM supports many other ARM CPUs and ispc should provide
access to all of the ones that have NEON support (and aren't too
obscure.)
- ~5 of the reduce-* tests hit an assertion inside LLVM (unfortunately
only when the compiler runs on an ARM host, though).
- The Windows build hasn't been tested (though I've tried to update
ispc.vcxproj appropriately). It may just work, but will more likely
have various small issues.)
- Anything related to 64-bit ARM has seen no attention.
The stdilb code just calls the signed int{32,64} functions,
which gives the right result for the unsigned case anyway.
The various targets didn't consistently define the unsigned
variants in any case.
This ensures that they have static linkage, which in turn lets one
have multiple object files compiled to multiple targets without having
those cause link errors.
Issue #355.
Flag 32-bit vector types as only requiring 32-bit alignment (preemptive
bug fix for 32xi1 vectors).
Force module datalayouts to be the same before linking them to silence
an LLVM warning.
Finishes issue #309.
We now use InternalLinkage for the 'programIndex' symbol (and similar)
if we're not compiling with debugging symbols. This prevents those
symbol names/definitions from polluting the global namespace for
the common case.
Basically addresses Issue #274.
It's now possible to successfully print out the value of programIndex,
programCount, etc., in the debugger. The issue was that they were
defined as having InternalLinkage, which meant that DCE removed them
at the end of compilation. Now they're declared to have WeakODRLinkage,
which ensures that one copy survives (but there aren't multiply-defined
symbols when compiling multiple files.)
Previously, we uniqued AtomicTypes, so that they could be compared
by pointer equality, but with forthcoming SOA variability changes,
this would become too unwieldy (lacking a more general / ubiquitous
type uniquing implementation.)
(i.e., stop just reusing the ones for AVX1).
For now the only difference is that the int/uint min/max
functions call the new intrinsic for that. Once gather is
available from LLVM, that will go here as well.
When used, these targets end up with calls to undefined functions for all
of the various special vector stuff ispc needs to compile ispc programs
(masked store, gather, min/max, sqrt, etc.).
These targets are not yet useful for anything, but are a step toward
having an option to C++ code with calls out to intrinsics.
Reorganized the directory structure a bit and put the LLVM bitcode used
to define target-specific stuff (as well as some generic built-ins stuff)
into a builtins/ directory.
Note that for building on Windows, it's now necessary to set a LLVM_VERSION
environment variable (with values like LLVM_2_9, LLVM_3_0, LLVM_3_1svn, etc.)
