Move entry spills determination to JNI compiler.

The calling convention no longer describes entry spills as
spilling is the JNI compiler's responsibility. This allows
future improvements, such as spilling registers directly to
the HandleScope or outgoing stack args.

Remove the notion of interprocedural scratch register from
calling conventions and let assemblers deal with all scratch
register uses. The remaining JNI assembler APIs that take
scratch registers are currently unused and can be removed.

Also fix a bug in disassembly comparison for tests; the
contents of two files were considered identical if the
second one just contained additional data.

This change fully preserves the generated code and adds TODO
comments where doing so results in weird or suboptimal code.

Test: m test-art-host-gtest
Test: testrunner.py --host --optimizing
Test: Prebuilt boot image is unchanged.
Test: aosp_taimen-userdebug boots.
Bug: 12189621
Change-Id: Ic26a670276920313cd907a6eda8d982cf0abfd81

1 files changed, 7 insertions, 8 deletions

diff --git a/compiler/utils/assembler_test_base.h b/compiler/utils/assembler_test_base.h
index 5fa0b3cd39..0a7cf11ba2 100644
--- a/compiler/utils/assembler_test_base.h
+++ b/compiler/utils/assembler_test_base.h
@@ -364,14 +364,13 @@ class AssemblerTestInfrastructure {
     std::ifstream f1_in(f1);
     std::ifstream f2_in(f2);
 
-    bool result = std::equal(std::istreambuf_iterator<char>(f1_in),
-                             std::istreambuf_iterator<char>(),
-                             std::istreambuf_iterator<char>(f2_in));
-
-    f1_in.close();
-    f2_in.close();
-
-    return result;
+    bool read1_ok, read2_ok;
+    char c1, c2;
+    do {
+      read1_ok = static_cast<bool>(f1_in >> c1);
+      read2_ok = static_cast<bool>(f2_in >> c2);
+    } while (read1_ok && read2_ok && c1 == c2);
+    return !read1_ok && !read2_ok;  // Did we reach the end of both streams?
   }
 
   // Compile the given assembly code and extract the binary, if possible. Put result into res.