fuchsia: More fiddling with base_paths

Move FILE_EXE/_MODULE back to being in /system. That's where the test
data currently is (rather than the aspirational location of
/pkg/bin/...) so tell base about it, so that tests are able to load data
file that are currently loaded relative to DIR_EXE.

Bug: 706592
Change-Id: I5c13af84b1c90fdf4357756fe4d821f21e2cd68b
Reviewed-on: https://chromium-review.googlesource.com/523851
Reviewed-by: Lei Zhang <thestig@chromium.org>
Commit-Queue: Scott Graham <scottmg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#477370}

base/base_paths_fuchsia.cc

@@ -14,10 +14,17 @@ bool PathProviderFuchsia(int key, FilePath* result) {
   // correct as fixed paths like this. See https://crbug.com/726124.
   switch (key) {
     case FILE_EXE:
-      *result = FilePath("/pkg/bin/chrome");
-      return true;
     case FILE_MODULE:
-      *result = FilePath("/pkg/lib/chrome");
+      // TODO(fuchsia): This is incorrect per
+      // https://fuchsia.googlesource.com/docs/+/master/namespaces.md, and
+      // should be /pkg/{bin,lib}/something. However, binaries are currently run
+      // by packing them into the system bootfs rather than running a "real"
+      // installer (which doesn't currently exist). Additionally, to the
+      // installer not existing, mmap() currently only works on bootfs file
+      // systems (like /system) but won't for files installed dynamically in
+      // other locations on other types of file systems. So, for now, we use
+      // /system/ as the location for everything.
+      *result = FilePath("/system/chrome");
       return true;
     case DIR_SOURCE_ROOT:
       // This is only used for tests, so we return the binary location for now.