fix-32-bit-unreg

commit 9194c9c066a31cbb7a49830e4b5e2454fd4af6ba
Author: Cheng Shao <terrorjack@type.dance>
Date:   Mon Jan 23 13:17:07 2023 +0000

    CmmToC: fix CmmRegOff for 64-bit register on a 32-bit target

    We used to print the offset value to a platform word sized integer.
    This is incorrect when the offset is negative (e.g. output of cmm
    constant folding) and the register is 64-bit but on a 32-bit target,
    and may lead to incorrect runtime result (e.g. #22607).

    The fix is simple: just treat it as a proper MO_Add, with the correct
    width info inferred from the register itself.

    Metric Increase:
        T12707
        T13379
        T4801
        T5321FD
        T5321Fun

    (cherry picked from commit d151546e59a50158f25c3df6728b00d3c27bb4b9)

Gbp-Pq: Name fix-32-bit-unreg

diff --git a/compiler/GHC/CmmToC.hs b/compiler/GHC/CmmToC.hs
index c36d8e69638743e0eba2c9ad2f1a0e37e66428ac..328d5714424f4346501e7562443704eb4b2b0945 100644 (file)
--- a/compiler/GHC/CmmToC.hs
+++ b/compiler/GHC/CmmToC.hs
@@ -382,8 +382,8 @@ pprExpr platform e = case e of
     CmmRegOff reg 0    -> pprCastReg reg
 
     -- CmmRegOff is an alias of MO_Add
-    CmmRegOff reg i -> pprCastReg reg <> char '+' <>
-                       pprHexVal platform (fromIntegral i) (wordWidth platform)
+    CmmRegOff reg i    -> pprExpr platform $ CmmMachOp (MO_Add w) [CmmReg reg, CmmLit $ CmmInt (toInteger i) w]
+      where w = cmmRegWidth platform reg
 
     CmmMachOp mop args -> pprMachOpApp platform mop args