Improve symbol_extractor.CreateNameToSymbolInfo handling of duplicate symbols.

Improve symbol_extractor.CreateNameToSymbolInfo handling of duplicate symbols.

If a symbol appears more than once in the binary, return the lowest offset for
that symbol, rather than the last one.  Also warn about duplicated symbols.
There are a couple hundred in libchrome.so.  Further investigation is needed to
determine why this is even possible, but in the mean time this should fix the
orderfile bot redness.

Committed: https://crrev.com/70f5220a69a64377fb9d27a72941941d7ea7fa2c
Cr-Commit-Position: refs/heads/master@{#317592}

[azarchs, 2015-02-23 13:43:41]

azarchs@chromium.org changed reviewers:
+ lizeb@chromium.org, pasko@chromium.org

[azarchs, 2015-02-23 13:45:13]

In case one is curious, here's a sample of some of the duplicated symbols. 
They're an eclectic mix with no clear pattern:

HFilter16i
GetForProfile
_ZN5blink12_GLOBAL__N_118ExtraDataContainerD2Ev
_ZNSaIPvE11_M_allocateEjRj.isra.162
TransformAC3
simpleNext
porterOpen
copy_and_extend_plane
_ZNSaIjE11_M_allocateEjRj.isra.41
_ZN7content12_GLOBAL__N_123EnableSuddenTerminationEiPNS_13FrameTreeNodeE
_ZNSaIPvE11_M_allocateEjRj.isra.106
_ZNSaIPvE11_M_allocateEjRj.isra.101
_ZNSaIPvE11_M_allocateEjRj.isra.102
vp8_set_reference
_ZNSaIPvE11_M_allocateEjRj.isra.108
SSE16x8
_ZN12_GLOBAL__N_123OnEventDispatchCompleteEN7content28PersistentNotificationStatusE
_ZN3gpu12_GLOBAL__N_123PerformNotifyCompletionENS_17AsyncMemoryParamsE13scoped_refptrINS_36AsyncPixelTransferCompletionObserverEE
sha1_final
Initialize
_ZN7content12_GLOBAL__N_120ReleaseOriginalFrameERK13scoped_refptrIN5media10VideoFrameEE
_ZL18initial_reorderingPK18hb_ot_shape_plan_tP9hb_font_tP11hb_buffer_t
HFilter16
Reset
process_data_simple_main
SSE16x16
_ZL13dline_xy_at_tPK7SkPointd
Launch
simpleClose
IsEnabled
_ZN6icu_54L30initializeSystemDefaultCenturyEv
_ZL13fline_xy_at_tPK7SkPointd
isConsonant
_ZN7content12_GLOBAL__N_119DataSourceExtraDataD1Ev
_ZN5blinkL16filterPropertiesEbRKN3WTF6VectorINS_11CSSPropertyELj256ENS0_16DefaultAllocatorEEERS4_RjRNS0_8BitArrayILj400EEE
hashDestroy
star_oh
_ZN3rtcL10FromStringIiEEbRKSsPT_
SimpleVFilter16
_ZN7content12_GLOBAL__N_122StreamTextureProxyImplD2Ev
TransformTwo
_enumPropertyStartsRange
VFilter16i
sha1_update
_ZN6icu_54L10getServiceEv
HFilter8
_ZN2cc12_GLOBAL__N_116RasterBufferImplD0Ev
UpsampleBgrLinePair
_ZN4base8internal9BindStateINS0_15RunnableAdapterIMN7content12_GLOBAL__N_122StreamTextureProxyImplEFviEEEFvPS5_iENS0_8TypeListIJNS0_17UnretainedWrapperIS5_EEiEEEED0Ev
_ZN12_GLOBAL__N_122AddForcedURLOnUIThreadE13scoped_refptrIN7history8TopSitesEERK4GURL

[Benoit L, 2015-02-23 14:09:49]

https://codereview.chromium.org/947073003/diff/1/tools/cygprofile/symbol_extr...
File tools/cygprofile/symbol_extractor.py (right):

https://codereview.chromium.org/947073003/diff/1/tools/cygprofile/symbol_extr...
tools/cygprofile/symbol_extractor.py:120: symbol_infos_by_name = {}
This function no longer does what it says in its documentation. Can you please
update it?

[pasko, 2015-02-23 14:28:13]

please put BUG=460906 to the description

please also write a test, it will be OK to have a test in a followup CL, while
this one would make the bot green.

https://codereview.chromium.org/947073003/diff/1/tools/cygprofile/symbol_extr...
File tools/cygprofile/symbol_extractor.py (right):

https://codereview.chromium.org/947073003/diff/1/tools/cygprofile/symbol_extr...
tools/cygprofile/symbol_extractor.py:121: duplicated_symbols = set()
They rather 'collided' than were 'duplicated'. So let's refer to it as 'symbol
collissions'?

https://codereview.chromium.org/947073003/diff/1/tools/cygprofile/symbol_extr...
tools/cygprofile/symbol_extractor.py:134: logging.warning('First %d duplicated
symbols:\n%s' % (
I would prefer warning level on the level of check_orderfile.py, where we have
_MAX_WARNINGS_TO_PRINT. Exploding logging limits is not very nice.

This function can be implemented as GroupSymbolInfosByName, where for each name
symbol_infos would be sorted by offset. (see about 'Slicing' here:
http://stackoverflow.com/questions/8024571/insert-an-item-into-sorted-list-py...)

Then it's a slight modification in _CountMisorderedSymbols to take the first
item, and limit warnings).

[azarchs, 2015-02-23 15:39:38]

https://codereview.chromium.org/947073003/diff/1/tools/cygprofile/symbol_extr...
File tools/cygprofile/symbol_extractor.py (right):

https://codereview.chromium.org/947073003/diff/1/tools/cygprofile/symbol_extr...
tools/cygprofile/symbol_extractor.py:120: symbol_infos_by_name = {}
On 2015/02/23 14:09:49, lizeb wrote:
> This function no longer does what it says in its documentation. Can you please
> update it?

I'd say it's functionality relative the documentation is unchanged. It was
perhaps underspecified before, so I've improved that.

https://codereview.chromium.org/947073003/diff/1/tools/cygprofile/symbol_extr...
tools/cygprofile/symbol_extractor.py:121: duplicated_symbols = set()
On 2015/02/23 14:28:13, pasko wrote:
> They rather 'collided' than were 'duplicated'. So let's refer to it as 'symbol
> collissions'?

Done.

https://codereview.chromium.org/947073003/diff/1/tools/cygprofile/symbol_extr...
tools/cygprofile/symbol_extractor.py:134: logging.warning('First %d duplicated
symbols:\n%s' % (
On 2015/02/23 14:28:13, pasko wrote:
> I would prefer warning level on the level of check_orderfile.py, where we have
> _MAX_WARNINGS_TO_PRINT. Exploding logging limits is not very nice.
> 
> This function can be implemented as GroupSymbolInfosByName, where for each
name
> symbol_infos would be sorted by offset. (see about 'Slicing' here:
>
http://stackoverflow.com/questions/8024571/insert-an-item-into-sorted-list-py...)
> 
> Then it's a slight modification in _CountMisorderedSymbols to take the first
> item, and limit warnings).

Done.

[Benoit L, 2015-02-23 16:02:48]

lgtm

https://codereview.chromium.org/947073003/diff/40001/tools/cygprofile/symbol_...
File tools/cygprofile/symbol_extractor.py (right):

https://codereview.chromium.org/947073003/diff/40001/tools/cygprofile/symbol_...
tools/cygprofile/symbol_extractor.py:114: A a symbol can have several offsets,
this is a 1-to-many relationship.
Typo: A -> As

[pasko, 2015-02-23 16:08:04]

lgtm

[azarchs, 2015-02-23 16:09:23]

New patchsets have been uploaded after l-g-t-m from
lizeb@chromium.org,pasko@chromium.org

[azarchs, 2015-02-23 16:10:08]

https://codereview.chromium.org/947073003/diff/40001/tools/cygprofile/symbol_...
File tools/cygprofile/symbol_extractor.py (right):

https://codereview.chromium.org/947073003/diff/40001/tools/cygprofile/symbol_...
tools/cygprofile/symbol_extractor.py:114: A a symbol can have several offsets,
this is a 1-to-many relationship.
On 2015/02/23 16:02:48, lizeb wrote:
> Typo: A -> As

Done.

[azarchs, 2015-02-23 16:10:14]

The CQ bit was checked by azarchs@chromium.org

[commit-bot: I haz the power, 2015-02-23 16:10:46]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/947073003/60001

[commit-bot: I haz the power, 2015-02-23 16:13:23]

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

[commit-bot: I haz the power, 2015-02-23 16:13:24]

Try jobs failed on following builders:
  ios_dbg_simulator on tryserver.chromium.mac (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.mac/builders/ios_dbg_simulator...)

[azarchs, 2015-02-23 16:31:43]

The CQ bit was checked by azarchs@chromium.org

[commit-bot: I haz the power, 2015-02-23 16:32:48]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/947073003/60001

[commit-bot: I haz the power, 2015-02-23 16:43:21]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2015-02-23 16:44:13]

Patchset 4 (id:??) landed as
https://crrev.com/70f5220a69a64377fb9d27a72941941d7ea7fa2c
Cr-Commit-Position: refs/heads/master@{#317592}

[pasko, 2015-02-24 10:26:22]

nobody likes bugs :)
nevermind, I'll continue my reminders...