openpower/package/hostboot/p8Patches/hostboot-0011-Convert-cmpi-to-cmpwi-to-fix-build-for-modern-binuti.patch

From 3e9a825d1a217ab02a24d14e80cad993a864079e Mon Sep 17 00:00:00 2001
From: Joel Stanley <joel@jms.id.au>
Date: Thu, 22 Jun 2017 13:43:47 +0930
Subject: [PATCH] Convert cmpi to cmpwi to fix build for modern binutils

From Linux 80f23935cadb ("powerpc: Convert cmp to cmpd in idle enter sequence"):

    PowerPC's "cmp" instruction has four operands. Normally people write
    "cmpw" or "cmpd" for the second cmp operand 0 or 1. But, frequently
    people forget, and write "cmp" with just three operands.

    With older binutils this is silently accepted as if this was "cmpw",
    while often "cmpd" is wanted. With newer binutils GAS will complain
    about this for 64-bit code. For 32-bit code it still silently assumes
    "cmpw" is what is meant.

    In this instance the code comes directly from ISA v2.07, including the
    cmp, but cmpd is correct. Backport to stable so that new toolchains can
    build old kernels.

This is change is a noop with the existing toolchain. We change from
implicitly generating the cmpwi to explicitly stating in for
compatibility with newer toolchains.

With gcc 4.9.3, binutils 2.25.2:

  $ cat asm-test.S
  .text

  .global test
          cmpi 0, 8, 1
  $ powerpc64-linux-gcc -c asm-test.S
  $ objdump -d asm-test.o
  0000000000000000 <.text>:
     	2c 08 00 01 	cmpwi   r8,1

Old compiler, updated instruction:

  $ cat asm-test.S
  .text

  .global test
          cmpwi 0, 8, 1
  $ powerpc64-linux-gcc -c asm-test.S
  $ objdump -d asm-test.o
  0000000000000000 <.text>:
     	2c 08 00 01 	cmpwi   r8,1

And then the new toolchain (gcc 6.3.0, binutils 2.28) with the updated
asm:

  $ cat asm-test.S
  .text

  .global test
          cmpwi 0, 8, 1
  $ powerpc64-linux-gnu-gcc -c asm-test.S
  $ objdump -d asm-test.o
  0000000000000000 <.text>:
     	2c 08 00 01 	cmpwi   r8,1

Change-Id: I878ab3596f54c221879945944f2dfbf053427026
Signed-off-by: Joel Stanley <joel@jms.id.au>
Signed-off-by: Joel Stanley <joel.stanley@au1.ibm.com>
---
 src/kernel/start.S     | 34 +++++++++++++++++-----------------
 src/runtime/rt_start.S |  2 +-
 2 files changed, 18 insertions(+), 18 deletions(-)

diff --git a/src/kernel/start.S b/src/kernel/start.S
index 6474a2ced794..5015b652e407 100644
--- a/src/kernel/start.S
+++ b/src/kernel/start.S
@@ -199,7 +199,7 @@ UNIMPL_INTERRUPT(hype_decrementer, 0x980)
 ;//
 .org _start + 0xC00
 intvect_system_call_fast:
-    cmpi cr0, r3, 0x0800
+    cmpwi cr0, r3, 0x0800
     bge  cr0, system_call_fast_path
 STD_INTERRUPT(system_call, 0xC08)
 
@@ -292,7 +292,7 @@ _other_thread_spinlock:
 1:
     ld r3, 0(r2)
     ;// Loop until value is 1...
-    cmpi cr0, r3, 1
+    cmpwi cr0, r3, 1
     beq _other_thread_spinlock_complete
     or 1,1,1 ;// Lower thread priority.
     b 1b
@@ -307,12 +307,12 @@ _other_thread_spinlock_complete:
     extrwi r1, r1, 3, 19
     sldi r1, r1, 3
     ldx r2, r1, r2              ;// Dereference to get on-node CPUs array.
-    cmpi cr0, r2, 0             ;// Check for NULL node array.
+    cmpwi cr0, r2, 0            ;// Check for NULL node array.
     beq- 1f
     mfspr r1, PIR               ;// Extract on-node CPU id.
     clrlslwi r1, r1, 22, 3
     ldx r3, r1, r2              ;// Load CPU object.
-    cmpi cr0, r3, 0             ;// Check for NULL CPU object.
+    cmpwi cr0, r3, 0            ;// Check for NULL CPU object.
     beq- 1f
     ld r1, CPU_KERNEL_STACK(r3) ;// Load initial stack.
 
@@ -388,7 +388,7 @@ kernel_save_task:
     std r31, TASK_GPR_31(r1)    ;// Save GPR31
 
     ld r2, TASK_FP_CONTEXT(r1)  ;// Load FP Context pointer.
-    cmpi cr0, r2, 0
+    cmpwi cr0, r2, 0
     bne- cr0, 1f                ;// Jump to FP-save if != NULL.
 2:
 
@@ -470,7 +470,7 @@ kernel_dispatch_task:
 
                         ;// Check if FP enabled, load context.
     ld r2, TASK_FP_CONTEXT(r1)
-    cmpi cr0, r2, 0
+    cmpwi cr0, r2, 0
     bne- 1f
 2:
                         ;// Restore GPRs from context.
@@ -587,7 +587,7 @@ intvect_system_reset:
     lis r2, kernel_other_thread_spinlock@h
     ori r2, r2, kernel_other_thread_spinlock@l
     ld r2, 0(r2)
-    cmpi cr0, r2, 0
+    cmpwi cr0, r2, 0
     beq- _start
 
     ;// Get CPU object from thread ID, check for NULL which implies not
@@ -598,12 +598,12 @@ intvect_system_reset:
     extrwi r1, r1, 3, 19
     sldi r1, r1, 3
     ldx r2, r1, r2              ;// Dereference to get on-node CPUs array.
-    cmpi cr0, r2, 0             ;// Check for NULL node array.
+    cmpwi cr0, r2, 0            ;// Check for NULL node array.
     beq- _start
     mfspr r1, PIR               ;// Extract on-node CPU id.
     clrlslwi r1, r1, 22, 3
     ldx r2, r1, r2              ;// Load CPU object.
-    cmpi cr0, r2, 0             ;// Check for NULL CPU object.
+    cmpwi cr0, r2, 0            ;// Check for NULL CPU object.
     beq- _start
 
     ;// Check for inactive CPU.
@@ -617,13 +617,13 @@ intvect_system_reset:
     mfsrr1 r2
     extrdi r2, r2, 3, 42
     ;// Check for decrementer (bits = 011).
-    cmpi cr0, r2, 0x3
+    cmpwi cr0, r2, 0x3
     beq+ intvect_system_reset_decrementer
     ;// Check for external interrupt (bits = 010).
-    cmpi cr0, r2, 0x4
+    cmpwi cr0, r2, 0x4
     beq+ intvect_system_reset_external
     ;// Check for HMI (bits = 101).
-    cmpi cr0, r2, 0x5
+    cmpwi cr0, r2, 0x5
     beq+ 1f ;// Unable to handle HMI, jump to 'unknown reason'.
 
 1:  ;// Unknown reason, call as unhandled_exception.
@@ -721,14 +721,14 @@ system_call_fast_path:
     b 1f                        ;// Jump to exit point.
         ;// Check if this is HMER write (0x801).
 2:
-    cmpi cr0, r3, 0x801
+    cmpwi cr0, r3, 0x801
     bne cr0, 3f
     mtspr HMER, r4
     li r3, 0
     b 1f                        ;// Jump to exit point.
         ;// Check if this is SCRATCH read (0x802).
 3:
-    cmpi cr0, r3, 0x802
+    cmpwi cr0, r3, 0x802
     bne cr0, 4f
         ;// Check for being on master processor.
     mfsprg3 r6          ;// Get task structure.
@@ -755,7 +755,7 @@ system_call_fast_path:
     b intvect_system_call
         ;// Check if this is SCRATCH write (0x803).
 4:
-    cmpi cr0, r3, 0x803
+    cmpwi cr0, r3, 0x803
     bne cr0, 5f
         ;// Check for master processor.
     mfsprg3 r6          ;// Get task structure.
@@ -770,7 +770,7 @@ system_call_fast_path:
     b 1f                        ;// Jump to exit point.
         ;// Check if this is PVR read (0x804).
 5:
-    cmpi cr0, r3, 0x804
+    cmpwi cr0, r3, 0x804
     bne cr0, 6f
     mfspr r3, PVR
     b 1f                        ;// Jump to exit point.
@@ -800,7 +800,7 @@ system_call_fast_path:
 .global userspace_task_entry
 userspace_task_entry:
         ;// Skip stack frame if GPR1 == NULL.
-    cmpi cr0, r1, 0
+    cmpwi cr0, r1, 0
     beq- 1f
         ;// Create frame.
         ;//     NULL back-chain + 48 bytes + quad-word alignment.  See ABI.
diff --git a/src/runtime/rt_start.S b/src/runtime/rt_start.S
index d69184aef29f..821e9f956b8e 100644
--- a/src/runtime/rt_start.S
+++ b/src/runtime/rt_start.S
@@ -47,7 +47,7 @@ _init:
 
     ld r8, 0(r10)             # Get count of relocations.
 
-    cmpi cr0, r8, 0           # Perform relocations (if any).
+    cmpwi cr0, r8, 0          # Perform relocations (if any).
     beq 2f
     mtctr r8
 1:
-- 
2.13.1