[llvm] fc22de8 - [AMDGPU] Clarify calling conv about inactive lanes

[Sebastian Neubauer via llvm-commits]

Author: Sebastian Neubauer
Date: 2021-02-02T10:15:09+01:00
New Revision: fc22de82182e05794123be70df5dfe15600c02b4

URL: https://github.com/llvm/llvm-project/commit/fc22de82182e05794123be70df5dfe15600c02b4
DIFF: https://github.com/llvm/llvm-project/commit/fc22de82182e05794123be70df5dfe15600c02b4.diff

LOG: [AMDGPU] Clarify calling conv about inactive lanes

So far, it was not specified what happens with the VGPRs of inactive
lanes when functions are called. This patch explicitely mentions that
the VGPR values of inactive lanes need to be preserved for all
registers.

This describes the current behavior, as only active lanes of registers
are saved to scratch. Also, as the multi-lane nature of VGPRs is not
properly modeled, we cannot determine the live VGPRs from inactive lanes
at calls. So we cannot save them, even if we intended to do so.

Differential Revision: https://reviews.llvm.org/D95610

Added: 
    

Modified: 
    llvm/docs/AMDGPUUsage.rst

Removed: 
    


################################################################################
diff  --git a/llvm/docs/AMDGPUUsage.rst b/llvm/docs/AMDGPUUsage.rst
index f16594db47ea..f9e1ca293d88 100644
--- a/llvm/docs/AMDGPUUsage.rst
+++ b/llvm/docs/AMDGPUUsage.rst
@@ -8240,23 +8240,27 @@ On exit from a function:
     * GFX6-GFX8: M0
     * All SGPR registers except the clobbered registers of SGPR4-31.
     * VGPR40-47
-      VGPR56-63
-      VGPR72-79
-      VGPR88-95
-      VGPR104-111
-      VGPR120-127
-      VGPR136-143
-      VGPR152-159
-      VGPR168-175
-      VGPR184-191
-      VGPR200-207
-      VGPR216-223
-      VGPR232-239
-      VGPR248-255
-
-        *Except the argument registers, the VGPR clobbered and the preserved
-        registers are intermixed at regular intervals in order to
-        get a better occupancy.*
+    * VGPR56-63
+    * VGPR72-79
+    * VGPR88-95
+    * VGPR104-111
+    * VGPR120-127
+    * VGPR136-143
+    * VGPR152-159
+    * VGPR168-175
+    * VGPR184-191
+    * VGPR200-207
+    * VGPR216-223
+    * VGPR232-239
+    * VGPR248-255
+
+        .. note::
+
+          Except the argument registers, the VGPRs clobbered and the preserved
+          registers are intermixed at regular intervals in order to keep a
+          similar ratio independent of the number of allocated VGPRs.
+
+    * Lanes of all VGPRs that are inactive at the call site.
 
       For the AMDGPU backend, an inter-procedural register allocation (IPRA)
       optimization may mark some of clobbered SGPR and VGPR registers as