mirror of
https://github.com/llvm/llvm-project.git
synced 2025-05-19 10:26:06 +00:00
dd5d65adb6
This patch adds the CodeGen changes needed for enabling HIP parallel algorithm offload on AMDGPU targets. This change relaxes restrictions on what gets emitted on the device path, when compiling in `hipstdpar` mode:
1. Unless a function is explicitly marked `__host__`, it will get emitted, whereas before only `__device__` and `__global__` functions would be emitted;
2. Unsupported builtins are ignored as opposed to being marked as an error, as the decision on their validity is deferred to the `hipstdpar` specific code selection pass;
3. We add a `hipstdpar` specific pass to the opt pipeline, independent of optimisation level:
- When compiling for the host, iff the user requested it via the `--hipstdpar-interpose-alloc` flag, we add a pass which replaces canonical allocation / deallocation functions with accelerator aware equivalents.
A test to validate that unannotated functions get correctly emitted is added as well.
Reviewed by: yaxunl, efriedma
Differential Revision: https://reviews.llvm.org/D155850
IRgen optimization opportunities. //===---------------------------------------------------------------------===// The common pattern of -- short x; // or char, etc (x == 10) -- generates an zext/sext of x which can easily be avoided. //===---------------------------------------------------------------------===// Bitfields accesses can be shifted to simplify masking and sign extension. For example, if the bitfield width is 8 and it is appropriately aligned then is is a lot shorter to just load the char directly. //===---------------------------------------------------------------------===// It may be worth avoiding creation of alloca's for formal arguments for the common situation where the argument is never written to or has its address taken. The idea would be to begin generating code by using the argument directly and if its address is taken or it is stored to then generate the alloca and patch up the existing code. In theory, the same optimization could be a win for block local variables as long as the declaration dominates all statements in the block. NOTE: The main case we care about this for is for -O0 -g compile time performance, and in that scenario we will need to emit the alloca anyway currently to emit proper debug info. So this is blocked by being able to emit debug information which refers to an LLVM temporary, not an alloca. //===---------------------------------------------------------------------===// We should try and avoid generating basic blocks which only contain jumps. At -O0, this penalizes us all the way from IRgen (malloc & instruction overhead), all the way down through code generation and assembly time. On 176.gcc:expr.ll, it looks like over 12% of basic blocks are just direct branches! //===---------------------------------------------------------------------===//