Use fast ELF symbolizer for symbols.py and tombstones

Use fast ELF symbolizer for symbols.py and tombstones

Try to reduce stack_tool_with_logcat_dump time by using fast ELF
symbolizer. The origin CL (https://codereview.chromium.org/164113003/)
was created 1.5 yrs ago. This is an effort to check it in.

BUG=463971
Committed: https://crrev.com/9c8f78855ca845ddd86d2daaec1fd1092eddd264
Cr-Commit-Position: refs/heads/master@{#346944}

[Menglin, 2015-09-01 16:06:12]

mlliu@chromium.org changed reviewers:
+ primiano@chromium.org

[Menglin, 2015-09-01 16:06:12]

Hi Primiano,

Please review this CL. 

Unfortunately, this doesn't reduce the stack_tool_with_logcat_dump time. I tried
on the trybot:
http://build.chromium.org/p/tryserver.chromium.linux/builders/linux_android_r....
It's still 4+ minutes. We will need to figure out some other way to reduce the
time.

Thank you!

Menglin

[Primiano Tucci (use gerrit), 2015-09-01 16:15:53]

LGTM thanks.

A tip on timing: I think most of the time is spent doing I/O. You could probably
make it faster as follows: way before the actual symbolization step is called,
ideally while you are running tests on the device and just waiting on the host,
you could start a background process which does "cat libchrome.so > /dev/null".
This should have the side effect of pulling the libchrome.so in the pagecache
and the next random I/O accesses performed in the symbolization step should be
hopefully much faster.
It only gets trickier to deal with that properly in component builds where you
need to warm-up multiple .so's.
You could probably try something like: 

starting in background
find . -name '*.sp' -exec dd if={} of=/dev/null bs=1M \;

before you run the actual tests steps (Which will take several minutes to
complete) and before you reach the actual symbolization step.

[Menglin, 2015-09-01 16:57:51]

mlliu@chromium.org changed reviewers:
+ rmcilroy@chromium.org

[Menglin, 2015-09-01 16:57:51]

+rmcilroy

Please review this CL. It is primiano's CL
(https://codereview.chromium.org/164113003/) 1.5 yrs ago with one slight
difference.

Thanks

Menglin

[rmcilroy, 2015-09-02 10:14:39]

lgtm

[Menglin, 2015-09-02 15:52:55]

The CQ bit was checked by mlliu@chromium.org

[commit-bot: I haz the power, 2015-09-02 15:53:12]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1304873010/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1304873010/40001

[commit-bot: I haz the power, 2015-09-02 16:03:57]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-09-02 16:03:58]

Try jobs failed on following builders:
  chromium_presubmit on tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[Menglin, 2015-09-02 16:53:04]

The CQ bit was checked by mlliu@chromium.org

[Menglin, 2015-09-02 16:53:04]

The patchset sent to the CQ was uploaded after l-g-t-m from
primiano@chromium.org, rmcilroy@chromium.org
Link to the patchset: https://codereview.chromium.org/1304873010/#ps60001
(title: "remove the empty line added in patchset2")

[commit-bot: I haz the power, 2015-09-02 16:53:27]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1304873010/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1304873010/60001

[commit-bot: I haz the power, 2015-09-02 17:03:53]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2015-09-02 17:04:36]

Patchset 4 (id:??) landed as
https://crrev.com/9c8f78855ca845ddd86d2daaec1fd1092eddd264
Cr-Commit-Position: refs/heads/master@{#346944}