From 892faf25eff4691860da127aab8b71c0bcdfa3b9 Mon Sep 17 00:00:00 2001 From: mcgrathr Date: Fri, 2 Oct 2015 14:54:42 -0700 Subject: PPAPI regression test for NaCl x86-32 IRT stack alignment sensitivity The bug with IRT stack alignment sensitivity can manifest in two ways. The issue of the stack being aligned properly at callbacks from the IRT is tested on the NaCl side (irt_stack_alignment test). The other issue is alignment-sensitive SSE instructions being used to copy 16-byte arguments from the stack, assuming they will be properly aligned on the stack. That situation only comes up with the PPAPI interfaces that use a 16-byte argument type such as PP_Var. So test that case here. BUG= https://code.google.com/p/nativeclient/issues/detail?id=3935 TEST= browser_tests --gtest_filter=PPAPINaClNewlibTest.NaClIRTStackAlignment on x86-32 R=bbudge@chromium.org, dschuff@chromium.org Review URL: https://codereview.chromium.org/1384583006 Cr-Commit-Position: refs/heads/master@{#352146} --- ppapi/ppapi_sources.gypi | 2 ++ 1 file changed, 2 insertions(+) (limited to 'ppapi/ppapi_sources.gypi') diff --git a/ppapi/ppapi_sources.gypi b/ppapi/ppapi_sources.gypi index 55e1ff2..ea7ccbb 100644 --- a/ppapi/ppapi_sources.gypi +++ b/ppapi/ppapi_sources.gypi @@ -513,6 +513,8 @@ # 'test_nacl_source_files': [ # Test cases (PLEASE KEEP THIS SECTION IN ALPHABETICAL ORDER) + 'tests/test_nacl_irt_stack_alignment.cc', + 'tests/test_nacl_irt_stack_alignment.h', 'tests/test_tcp_server_socket_private_disallowed.cc', 'tests/test_tcp_socket_private_disallowed.cc', 'tests/test_udp_socket_private_disallowed.cc', -- cgit v1.1