PyAuto test for GPU memory consumption

chrome/test/functional/gpu_mem.py

+#!/usr/bin/env python
+# Copyright (c) 2011 The Chromium Authors. All rights reserved.
+# Use of this source code is governed by a BSD-style license that can be
+# found in the LICENSE file.
+
+import pyauto_functional
+import pyauto
+
+class GpuMemTest(pyauto.PyUITest):
+  # Return the number of bytes of GPU memory being used
+  def _getGpuMemUsage(self):
+    gpuMemStats = self.GetGpuMemStats()
+    if gpuMemStats['gpu_active'] == False:
+      self.fail(msg='GPU process not active.')
+    return gpuMemStats['gpu_memory_bytes_used']
+
+  def testSanity(self):

[nduca, 2012/07/17 22:40:25]

In general, prefer test name that tell someone who'se in some random chrome team
(e.g. not gpu) to understand what broadly is broken in case it fails. That way
they can route the bug failure to someone easily, give it a decent bug title,
and give the person who gets the bug a good first guess at whats going wrong.

E.g. "testThatGpuMemoryAPIReturnsAValue"

+    """Report GPU memory usage when test page with 3D CSS is open."""
+    url_3dcss = self.GetFileURLForDataPath('gpu/mem_3dcss.html')
+    self.NavigateToURL(url_3dcss, 0)
+    print 'Using ', self._getGpuMemUsage(), ' bytes of GPU memory'

[nduca, 2012/07/17 22:40:25]

I think the intent of this is that it consumes a lot of memory, right? SO you
should have a part of the test that verifies that a single page consumed at
least some <smoke test amount> of memory. So, for example, if we suddenly
optimize the drawing of solid color tiles to use shaders instead of texture, we
get an error that points out that the test has stopped being relevant.

+
+  def testOverSubscribe(self):
+    """Over-subscribe GPU memory."""
+    url_3dcss = self.GetFileURLForDataPath('gpu/mem_3dcss.html')
+    # Let tabs_to_open be the number of instances of mem_3dcss.html that
+    # will oversubscribe all available GPU memory by a factor of 4
+    # TODO - compute this reliably -- this is a wild guess

[nduca, 2012/07/17 22:40:25]

Hmmm so how are you going to compute this? Doesn't seem like we should land this
until you're computing this.

What if you add a command line option to chrome that boots up chrome with a
fixed gpu memory budget? E.g. memory manager uses a fake amount of memory.

+    tabs_to_open = 2
+    # Record memory usage statistics as each new tab is opened
+    for i in xrange(tabs_to_open):
+      self.AppendTab(url_3dcss, windex=0)
+    # TODO - verify that we never exceeded the GPU memory budget

[nduca, 2012/07/17 22:40:25]

How?

+    # TODO - verify that we actually used 'most' of the GPU memory (say,
+    #        half of it), and that the GPU memory manager did kick
+    #        around some resources

[nduca, 2012/07/17 22:40:25]

Maybe using a fixed amount of memory is good here

+    print 'Using ', self._getGpuMemUsage(), ' bytes of GPU memory'

[nduca, 2012/07/17 22:40:25]

Should we have a test that verifies that we get a sane value for memory stats?
E.g. is the code that figures out the gpu memory stats working?

[mmocny, 2012/07/18 15:20:18]

GpuMemoryManager will have unittests here, but those use fake stubs that will
always provide "valid" numbers for memory allocation hints etc.

We certainly should have a test to make sure we are getting valid/correct
numbers with real pages.  This is basically the same case as with tab
visibility, so perhaps there should first be a simple, separate, sanity test for
that?  E.g. create a tab, make sure its visible and non-0 alloc, create a second
tab make sure first is non visible, create many tabs make sure some have 0
alloc.

On 2012/07/17 22:40:25, nduca wrote:

+
+if __name__ == '__main__':
+  pyauto_functional.Main()