Re-enable two flaky tests that were disabled due to timeouts.

chrome/browser/image_decoder_browsertest.cc

Index: chrome/browser/image_decoder_browsertest.cc
diff --git a/chrome/browser/image_decoder_browsertest.cc b/chrome/browser/image_decoder_browsertest.cc
index 314e1b88fb117016b8de4c4f999c6988b9d364e5..f2fdf537da8e9ec4ca5b5070632a994b822b34d9 100644
--- a/chrome/browser/image_decoder_browsertest.cc
+++ b/chrome/browser/image_decoder_browsertest.cc
@@ -243,16 +243,7 @@ IN_PROC_BROWSER_TEST_F(ImageDecoderBrowserTest, StartAndDestroy) {
 // Note: This test is inherently racy because KillProcessObserver lives on the
 // UI thread but ImageDecoder does its work mainly on the IO thread. So the test
 // checks for both possible valid outcomes.
-//
-// Flaky timeouts on Linux ASan / LSan, and on Mac. crbug.com/618206
-#if (defined(ADDRESS_SANITIZER) || defined(LEAK_SANITIZER)) && defined(OS_LINUX)
-#define MAYBE_StartAndKillProcess DISABLED_StartAndKillProcess
-#elif defined(OS_MACOSX)
-#define MAYBE_StartAndKillProcess DISABLED_StartAndKillProcess
-#else
-#define MAYBE_StartAndKillProcess StartAndKillProcess
-#endif
-IN_PROC_BROWSER_TEST_F(ImageDecoderBrowserTest, MAYBE_StartAndKillProcess) {
+IN_PROC_BROWSER_TEST_F(ImageDecoderBrowserTest, StartAndKillProcess) {
   KillProcessObserver observer;
   scoped_refptr<content::MessageLoopRunner> runner =
       new content::MessageLoopRunner;