Fix Int8#, Int16#, Word8#, Word16# tests under LLVM code generator
Unfortunately it looks like LLVM doesn't like that we use bitcast to narrow integers. E.g. CmpWord8 currently fails with:
=====> CmpWord8(llvm) 18 of 23 [0, 6, 0]
cd "primops/should_run/CmpWord8.run" && "/mnt/work/ghc/ghc-compare-1/inplace/test spaces/ghc-stage2" -o CmpWord8 CmpWord8.hs -dcore-lint -dcmm-lint -no-user-package-db -rtsopts -fno-warn-missed-specialisations -fshow-warning-groups -fdiagnostics-color=never -fno-diagnostics-show-caret -Werror=compat -dno-debug-output -fllvm
Compile failed (exit code 1) errors were:
[1 of 1] Compiling Main ( CmpInt8.hs, CmpInt8.o )
opt: /run/user/1000/ghc32287_0/ghc_2.ll:14549:20: error: invalid cast opcode for cast from 'i64' to 'i8'
%ln6uE = bitcast i64 %ln6uD to i8
^
`opt' failed in phase `LLVM Optimiser'. (Exit code: 1)Trac metadata
| Trac field | Value |
|---|---|
| Version | 8.6.1 |
| Type | Bug |
| TypeOfFailure | OtherFailure |
| Priority | normal |
| Resolution | Unresolved |
| Component | Compiler (LLVM) |
| Test case | |
| Differential revisions | |
| BlockedBy | |
| Related | |
| Blocking | |
| CC | michalt |
| Operating system | |
| Architecture |