Refactor the story around switches (#10137)

This re-implements the code generation for case expressions at the Stg →
Cmm level, both for data type cases as well as for integral literal
cases. (Cases on float are still treated as before).

The goal is to allow for fancier strategies in implementing them, for a
cleaner separation of the strategy from the gritty details of Cmm, and
to run this later than the Common Block Optimization, allowing for one
way to attack #10124. The new module CmmSwitch contains a number of
notes explaining this changes. For example, it creates larger
consecutive jump tables than the previous code, if possible.

nofib shows little significant overall improvement of runtime. The
rather large wobbling comes from changes in the code block order
(see #8082, not much we can do about it). But the decrease in code size
alone makes this worthwhile.

```
        Program           Size    Allocs   Runtime   Elapsed  TotalMem
            Min          -1.8%      0.0%     -6.1%     -6.1%     -2.9%
            Max          -0.7%     +0.0%     +5.6%     +5.7%     +7.8%
 Geometric Mean          -1.4%     -0.0%     -0.3%     -0.3%     +0.0%
```

Compilation time increases slightly:
```
        -1 s.d.                -----            -2.0%
        +1 s.d.                -----            +2.5%
        Average                -----            +0.3%
```

The test case T783 regresses a lot, but it is the only one exhibiting
any regression. The cause is the changed order of branches in an
if-then-else tree, which makes the hoople data flow analysis traverse
the blocks in a suboptimal order. Reverting that gets rid of this
regression, but has a consistent, if only very small (+0.2%), negative
effect on runtime. So I conclude that this test is an extreme outlier
and no reason to change the code.

Differential Revision: https://phabricator.haskell.org/D720

1 files changed, 7 insertions, 5 deletions

diff --git a/compiler/nativeGen/SPARC/CodeGen.hs b/compiler/nativeGen/SPARC/CodeGen.hs
index bba849da61..a9d861946e 100644
--- a/compiler/nativeGen/SPARC/CodeGen.hs
+++ b/compiler/nativeGen/SPARC/CodeGen.hs
@@ -43,6 +43,7 @@ import NCGMonad
 import BlockId
 import Cmm
 import CmmUtils
+import CmmSwitch
 import Hoopl
 import PIC
 import Reg
@@ -150,8 +151,8 @@ stmtToInstrs stmt = do
     CmmCondBranch arg true false -> do b1 <- genCondJump true arg
                                        b2 <- genBranch false
                                        return (b1 `appOL` b2)
-    CmmSwitch   arg ids         -> do dflags <- getDynFlags
-                                      genSwitch dflags arg ids
+    CmmSwitch arg ids   -> do dflags <- getDynFlags
+                              genSwitch dflags arg ids
     CmmCall { cml_target = arg } -> genJump arg
 
     _
@@ -308,13 +309,13 @@ genCondJump bid bool = do
 -- -----------------------------------------------------------------------------
 -- Generating a table-branch
 
-genSwitch :: DynFlags -> CmmExpr -> [Maybe BlockId] -> NatM InstrBlock
-genSwitch dflags expr ids
+genSwitch :: DynFlags -> CmmExpr -> SwitchTargets -> NatM InstrBlock
+genSwitch dflags expr targets
         | gopt Opt_PIC dflags
         = error "MachCodeGen: sparc genSwitch PIC not finished\n"
 
         | otherwise
-        = do    (e_reg, e_code) <- getSomeReg expr
+        = do    (e_reg, e_code) <- getSomeReg (cmmOffset dflags expr offset)
 
                 base_reg        <- getNewRegNat II32
                 offset_reg      <- getNewRegNat II32
@@ -335,6 +336,7 @@ genSwitch dflags expr ids
                         , LD      II32 (AddrRegReg base_reg offset_reg) dst
                         , JMP_TBL (AddrRegImm dst (ImmInt 0)) ids label
                         , NOP ]
+  where (offset, ids) = switchTargetsToTable targets
 
 generateJumpTableForInstr :: DynFlags -> Instr
                           -> Maybe (NatCmmDecl CmmStatics Instr)