[llvm] 3cc3960 - [X86] AMD Zen 3: cap LoopMicroOpBufferSize to workaround PR50384 (quadratic IndVars runtime)

[Roman Lebedev via llvm-commits]

Author: Roman Lebedev
Date: 2021-05-18T15:56:57+03:00
New Revision: 3cc39607668d6f942458a37e7a9111294d9373bd

URL: https://github.com/llvm/llvm-project/commit/3cc39607668d6f942458a37e7a9111294d9373bd
DIFF: https://github.com/llvm/llvm-project/commit/3cc39607668d6f942458a37e7a9111294d9373bd.diff

LOG: [X86] AMD Zen 3: cap LoopMicroOpBufferSize to workaround PR50384 (quadratic IndVars runtime)

While i would like to keep the right value here,
i would also like to be able to actually compile
e.g. vanilla test-suite.

256 is a pretty random guess, it should be pretty good enough
for serious loops, but small enough to result in tolerant
compile times for certain edge cases.

https://bugs.llvm.org/show_bug.cgi?id=50384

Added: 
    

Modified: 
    llvm/lib/Target/X86/X86ScheduleZnver3.td

Removed: 
    


################################################################################
diff  --git a/llvm/lib/Target/X86/X86ScheduleZnver3.td b/llvm/lib/Target/X86/X86ScheduleZnver3.td
index 84aee73bad63..cc0a67c2c287 100644
--- a/llvm/lib/Target/X86/X86ScheduleZnver3.td
+++ b/llvm/lib/Target/X86/X86ScheduleZnver3.td
@@ -35,7 +35,10 @@ def Znver3Model : SchedMachineModel {
   // The maximum capacity of the op cache is 4K ops.
   // Agner, 22.5 µop cache
   // The size of the µop cache is big enough for holding most critical loops.
-  let LoopMicroOpBufferSize = 4096;
+  // FIXME: PR50384: IndVars has quadradic complexity, with large values here
+  //        the compilation of certain loops ends up taking way too long.
+  // let LoopMicroOpBufferSize = 4096;
+  let LoopMicroOpBufferSize = 256;
   // AMD SOG 19h, 2.6.2 L1 Data Cache
   // The L1 data cache has a 4- or 5- cycle integer load-to-use latency.
   // AMD SOG 19h, 2.12 L1 Data Cache