Add support for large number of argumets in CCW call on Arm64 (#20670)

* Fix CCW with large numbers of arguments on arm64
- Correctly align the stack in COMToCLRDispatchHelper
- Add tests for naturally 16 byte aligned stack growth, and non-aligned growth
* New many arguments COM test
* Add support for the IL stub ETW diagnostic event

1 files changed, 15 insertions, 5 deletions

diff --git a/src/vm/arm64/asmhelpers.asm b/src/vm/arm64/asmhelpers.asm
index 0cbc0a82f4..07e78cbd56 100644
--- a/src/vm/arm64/asmhelpers.asm
+++ b/src/vm/arm64/asmhelpers.asm
@@ -742,12 +742,22 @@ GenericComCallStub_FirstStackAdjust     SETA GenericComCallStub_FirstStackAdjust
     cbz x0, COMToCLRDispatchHelper_RegSetup
 
     add x9, x1, #SIZEOF__ComMethodFrame
-    add x9, x9, x0, LSL #3
+
+    ; Compute number of 8 bytes slots to copy. This is done by rounding up the
+    ; dwStackSlots value to the nearest even value
+    add x0, x0, #1
+    bic x0, x0, #1
+
+    ; Compute how many slots to adjust the address to copy from. Since we
+    ; are copying 16 bytes at a time, adjust by -1 from the rounded value
+    sub x6, x0, #1
+    add x9, x9, x6, LSL #3
+
 COMToCLRDispatchHelper_StackLoop
-    ldr x8, [x9, #-8]!
-    str x8, [sp, #-8]!
-    sub x0, x0, #1
-    cbnz x0, COMToCLRDispatchHelper_StackLoop
+    ldp     x7, x8, [x9], #-16  ; post-index
+    stp     x7, x8, [sp, #-16]! ; pre-index
+    subs    x0, x0, #2
+    bne     COMToCLRDispatchHelper_StackLoop
     
 COMToCLRDispatchHelper_RegSetup