SkOnce: 2 bytes -> 1 byte

SkOnce: 2 bytes -> 1 byte

This uses the same logic we worked out for SkOncePtr to reduce
the memory footprint of SkOnce from a done byte and lock byte
to a single 3-state byte:

  - NotStarted: no thread has tried to run fn() yet
  - Active:     a thread is running fn()
  - Done:       fn() is complete

Threads which see Done return immediately.
Threads which see NotStarted try to move to Active, run fn(), then move to Done.
Threads which see Active spin until the active thread moves to Done.

This additionally fixes a too-weak memory order bug in SkOncePtr,
and adds a big note to explain.

BUG=skia:
GOLD_TRYBOT_URL= https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&issue=1904483003

Committed: https://skia.googlesource.com/skia/+/df02d338be8e3c1c50b48a3a9faa582703a39c07

Committed: https://skia.googlesource.com/skia/+/650f9e9a2630026d578970c6d031d7d4644f63b4

[mtklein_C, 2016-04-19 19:37:49]

Description was changed from

==========
SkOnce: 2 bytes -> 1 byte

This uses the same 3-state logic we worked out for SkOncePtr to
reduce the memory footprint of SkOnce from a done byte and lock
byte to a single byte:

  - NotStarted: no thread has tried to run fn() yet
  - Active:     a thread is running fn()
  - Done:       fn() is complete

Threads which see Done return immediately.
Threads which see NotStarted try to move to Active, run fn(), then move to Done.
Threads which see Active spin until the active thread moves to Done.

BUG=skia:
==========

to

==========
SkOnce: 2 bytes -> 1 byte

This uses the same 3-state logic we worked out for SkOncePtr to
reduce the memory footprint of SkOnce from a done byte and lock
byte to a single byte:

  - NotStarted: no thread has tried to run fn() yet
  - Active:     a thread is running fn()
  - Done:       fn() is complete

Threads which see Done return immediately.
Threads which see NotStarted try to move to Active, run fn(), then move to Done.
Threads which see Active spin until the active thread moves to Done.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[mtklein_C, 2016-04-19 19:38:53]

Description was changed from

==========
SkOnce: 2 bytes -> 1 byte

This uses the same 3-state logic we worked out for SkOncePtr to
reduce the memory footprint of SkOnce from a done byte and lock
byte to a single byte:

  - NotStarted: no thread has tried to run fn() yet
  - Active:     a thread is running fn()
  - Done:       fn() is complete

Threads which see Done return immediately.
Threads which see NotStarted try to move to Active, run fn(), then move to Done.
Threads which see Active spin until the active thread moves to Done.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
SkOnce: 2 bytes -> 1 byte

This uses the same logic we worked out for SkOncePtr to
reduce the memory footprint of SkOnce from a done byte and lock
byte to a single 3-state byte:

  - NotStarted: no thread has tried to run fn() yet
  - Active:     a thread is running fn()
  - Done:       fn() is complete

Threads which see Done return immediately.
Threads which see NotStarted try to move to Active, run fn(), then move to Done.
Threads which see Active spin until the active thread moves to Done.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[mtklein_C, 2016-04-19 19:39:08]

Description was changed from

==========
SkOnce: 2 bytes -> 1 byte

This uses the same logic we worked out for SkOncePtr to
reduce the memory footprint of SkOnce from a done byte and lock
byte to a single 3-state byte:

  - NotStarted: no thread has tried to run fn() yet
  - Active:     a thread is running fn()
  - Done:       fn() is complete

Threads which see Done return immediately.
Threads which see NotStarted try to move to Active, run fn(), then move to Done.
Threads which see Active spin until the active thread moves to Done.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
SkOnce: 2 bytes -> 1 byte

This uses the same logic we worked out for SkOncePtr to reduce
the memory footprint of SkOnce from a done byte and lock byte
to a single 3-state byte:

  - NotStarted: no thread has tried to run fn() yet
  - Active:     a thread is running fn()
  - Done:       fn() is complete

Threads which see Done return immediately.
Threads which see NotStarted try to move to Active, run fn(), then move to Done.
Threads which see Active spin until the active thread moves to Done.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[mtklein_C, 2016-04-19 19:40:05]

The CQ bit was checked by mtklein@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-04-19 19:40:08]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1904483003/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1904483003/1

[commit-bot: I haz the power, 2016-04-19 19:53:38]

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

[commit-bot: I haz the power, 2016-04-19 19:53:38]

Dry run: This issue passed the CQ dry run.

[mtklein_C, 2016-04-19 19:58:09]

The CQ bit was checked by mtklein@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-04-19 19:58:18]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1904483003/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1904483003/20001

[mtklein_C, 2016-04-19 19:58:42]

Description was changed from

==========
SkOnce: 2 bytes -> 1 byte

This uses the same logic we worked out for SkOncePtr to reduce
the memory footprint of SkOnce from a done byte and lock byte
to a single 3-state byte:

  - NotStarted: no thread has tried to run fn() yet
  - Active:     a thread is running fn()
  - Done:       fn() is complete

Threads which see Done return immediately.
Threads which see NotStarted try to move to Active, run fn(), then move to Done.
Threads which see Active spin until the active thread moves to Done.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
SkOnce: 2 bytes -> 1 byte

This uses the same logic we worked out for SkOncePtr to reduce
the memory footprint of SkOnce from a done byte and lock byte
to a single 3-state byte:

  - NotStarted: no thread has tried to run fn() yet
  - Active:     a thread is running fn()
  - Done:       fn() is complete

Threads which see Done return immediately.
Threads which see NotStarted try to move to Active, run fn(), then move to Done.
Threads which see Active spin until the active thread moves to Done.

This additionally fixes a too-weak memory order bug in SkOncePtr,
and adds a big note to explain.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

[mtklein_C, 2016-04-19 19:59:50]

The CQ bit was checked by mtklein@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-04-19 19:59:58]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1904483003/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1904483003/40001

[commit-bot: I haz the power, 2016-04-19 20:14:32]

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

[commit-bot: I haz the power, 2016-04-19 20:14:33]

Dry run: This issue passed the CQ dry run.

[mtklein_C, 2016-04-19 20:32:45]

mtklein@chromium.org changed reviewers:
+ herb@google.com

[herb_g, 2016-04-20 16:38:37]

https://codereview.chromium.org/1904483003/diff/40001/include/private/SkOnce.h
File include/private/SkOnce.h (right):

https://codereview.chromium.org/1904483003/diff/40001/include/private/SkOnce....
include/private/SkOnce.h:26: if (state == NotStarted) {
Instead of having:
if (state == NotStarted) {
...
}
while (state == Active) {
...
}

could you wrap the whole thing in while.

while (state != Done) {
  if (state == NotStarted) {
    ...
  }
}

Making this change will reduce this code to using a single branch in the common
case, and while (state != Done) is a clearer invariant of the state needed to
proceed.

https://codereview.chromium.org/1904483003/diff/40001/include/private/SkOnce....
include/private/SkOnce.h:44: enum State : uint8_t { NotStarted, Active, Done};
To me, calling the state Initializing is clearer than Active.

[mtklein_C, 2016-04-20 17:44:36]

The CQ bit was checked by mtklein@chromium.org

[mtklein_C, 2016-04-20 17:44:37]

https://codereview.chromium.org/1904483003/diff/40001/include/private/SkOnce.h
File include/private/SkOnce.h (right):

https://codereview.chromium.org/1904483003/diff/40001/include/private/SkOnce....
include/private/SkOnce.h:26: if (state == NotStarted) {
On 2016/04/20 at 16:38:37, herb_g wrote:
> Instead of having:
> if (state == NotStarted) {
> ...
> }
> while (state == Active) {
> ...
> }
> 
> could you wrap the whole thing in while.
> 
> while (state != Done) {
>   if (state == NotStarted) {
>     ...
>   }
> }
> 
> Making this change will reduce this code to using a single branch in the
common case, and while (state != Done) is a clearer invariant of the state
needed to proceed.

Done.

https://codereview.chromium.org/1904483003/diff/40001/include/private/SkOnce....
include/private/SkOnce.h:44: enum State : uint8_t { NotStarted, Active, Done};
On 2016/04/20 at 16:38:37, herb_g wrote:
> To me, calling the state Initializing is clearer than Active.

I've changed it to Calling.

[commit-bot: I haz the power, 2016-04-20 17:44:42]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1904483003/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1904483003/60001

[commit-bot: I haz the power, 2016-04-20 17:44:43]

Note for Reviewers:
The CQ is waiting for an approval.
If you believe that the CL is not ready yet, or if you would like to L-G-T-M
with comments then please uncheck the CQ checkbox.

Waiting for LGTM from valid reviewer(s) till 2016-04-20 23:44 UTC

[herb_g, 2016-04-20 17:52:05]

lgtm

[commit-bot: I haz the power, 2016-04-20 17:54:56]

Description was changed from

==========
SkOnce: 2 bytes -> 1 byte

This uses the same logic we worked out for SkOncePtr to reduce
the memory footprint of SkOnce from a done byte and lock byte
to a single 3-state byte:

  - NotStarted: no thread has tried to run fn() yet
  - Active:     a thread is running fn()
  - Done:       fn() is complete

Threads which see Done return immediately.
Threads which see NotStarted try to move to Active, run fn(), then move to Done.
Threads which see Active spin until the active thread moves to Done.

This additionally fixes a too-weak memory order bug in SkOncePtr,
and adds a big note to explain.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...
==========

to

==========
SkOnce: 2 bytes -> 1 byte

This uses the same logic we worked out for SkOncePtr to reduce
the memory footprint of SkOnce from a done byte and lock byte
to a single 3-state byte:

  - NotStarted: no thread has tried to run fn() yet
  - Active:     a thread is running fn()
  - Done:       fn() is complete

Threads which see Done return immediately.
Threads which see NotStarted try to move to Active, run fn(), then move to Done.
Threads which see Active spin until the active thread moves to Done.

This additionally fixes a too-weak memory order bug in SkOncePtr,
and adds a big note to explain.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...

Committed:
https://skia.googlesource.com/skia/+/df02d338be8e3c1c50b48a3a9faa582703a39c07
==========

[commit-bot: I haz the power, 2016-04-20 17:54:57]

Committed patchset #4 (id:60001) as
https://skia.googlesource.com/skia/+/df02d338be8e3c1c50b48a3a9faa582703a39c07

[fmalita_google_do_not_use, 2016-04-20 19:47:32]

On 2016/04/20 17:54:57, commit-bot: I haz the power wrote:
> Committed patchset #4 (id:60001) as
> https://skia.googlesource.com/skia/+/df02d338be8e3c1c50b48a3a9faa582703a39c07

Das roll is geborked:
https://build.chromium.org/p/tryserver.blink/builders/linux_blink_rel/builds/...

clang: error: linker command failed with exit code 1 (use -v to see invocation)
FAILED: python
"/b/build/slave/linux_layout/build/src/build/toolchain/gcc_solink_wrapper.py"
--readelf="readelf" --nm="nm"  --sofile="./libmus_library.so"
--tocfile="./libmus_library.so.TOC" --output="./libmus_library.so" --
../../third_party/llvm-build/Release+Asserts/bin/clang++ -shared
-Wl,--fatal-warnings -fPIC -Wl,-z,noexecstack -Wl,-z,now -Wl,-z,relro
-Wl,-z,defs -B../../third_party/binutils/Linux_x64/Release/bin -fuse-ld=gold
-Wl,--icf=all -pthread -m64 -Wl,-O1 -Wl,--gc-sections -Wl,--as-needed
--sysroot=../../build/linux/debian_wheezy_amd64-sysroot  -Wl,--export-dynamic -o
"./libmus_library.so" -Wl,-soname="libmus_library.so" @"./libmus_library.so.rsp"
../../third_party/skia/include/private/SkOnce.h:24: error: undefined reference
to 'std::atomic<SkOnce::State>::load(std::memory_order) const'
../../third_party/skia/include/private/SkOnce.h:33: error: undefined reference
to 'std::atomic<SkOnce::State>::compare_exchange_strong(SkOnce::State&,
SkOnce::State, std::memory_order)'
../../third_party/skia/include/private/SkOnce.h:36: error: undefined reference
to 'std::atomic<SkOnce::State>::store(SkOnce::State, std::memory_order)'
../../third_party/skia/include/private/SkOnce.h:42: error: undefined reference
to 'std::atomic<SkOnce::State>::load(std::memory_order) const'
../../third_party/skia/include/gpu/../private/SkOnce.h:24: error: undefined
reference to 'std::atomic<SkOnce::State>::load(std::memory_order) const'
../../third_party/skia/include/gpu/../private/SkOnce.h:33: error: undefined
reference to
'std::atomic<SkOnce::State>::compare_exchange_strong(SkOnce::State&,
SkOnce::State, std::memory_order)'
../../third_party/skia/include/gpu/../private/SkOnce.h:36: error: undefined
reference to 'std::atomic<SkOnce::State>::store(SkOnce::State,
std::memory_order)'
../../third_party/skia/include/gpu/../private/SkOnce.h:42: error: undefined
reference to 'std::atomic<SkOnce::State>::load(std::memory_order) const'
../../third_party/skia/include/gpu/../private/SkOnce.h:33: error: undefined
reference to
'std::atomic<SkOnce::State>::compare_exchange_strong(SkOnce::State&,
SkOnce::State, std::memory_order)'
../../third_party/skia/include/gpu/../private/SkOnce.h:36: error: undefined
reference to 'std::atomic<SkOnce::State>::store(SkOnce::State,
std::memory_order)'
../../third_party/skia/include/gpu/../private/SkOnce.h:33: error: undefined
reference to
'std::atomic<SkOnce::State>::compare_exchange_strong(SkOnce::State&,
SkOnce::State, std::memory_order)'
../../third_party/skia/include/gpu/../private/SkOnce.h:36: error: undefined
reference to 'std::atomic<SkOnce::State>::store(SkOnce::State,
std::memory_order)'
clang: error: linker command failed with exit code 1 (use -v to see invocation)
ninja: build stopped: subcommand failed.

[mtklein_C, 2016-04-20 20:02:01]

A revert of this CL (patchset #4 id:60001) has been created in
https://codereview.chromium.org/1898413004/ by mtklein@chromium.org.

The reason for reverting is: bust the roll.

[mtklein_C, 2016-04-20 20:39:13]

The CQ bit was checked by mtklein@chromium.org

[mtklein_C, 2016-04-20 20:39:13]

The patchset sent to the CQ was uploaded after l-g-t-m from herb@google.com
Link to the patchset: https://codereview.chromium.org/1904483003/#ps80001
(title: "basic uint8_t type")

[commit-bot: I haz the power, 2016-04-20 20:39:20]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1904483003/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1904483003/80001

[commit-bot: I haz the power, 2016-04-20 20:49:18]

Description was changed from

==========
SkOnce: 2 bytes -> 1 byte

This uses the same logic we worked out for SkOncePtr to reduce
the memory footprint of SkOnce from a done byte and lock byte
to a single 3-state byte:

  - NotStarted: no thread has tried to run fn() yet
  - Active:     a thread is running fn()
  - Done:       fn() is complete

Threads which see Done return immediately.
Threads which see NotStarted try to move to Active, run fn(), then move to Done.
Threads which see Active spin until the active thread moves to Done.

This additionally fixes a too-weak memory order bug in SkOncePtr,
and adds a big note to explain.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...

Committed:
https://skia.googlesource.com/skia/+/df02d338be8e3c1c50b48a3a9faa582703a39c07
==========

to

==========
SkOnce: 2 bytes -> 1 byte

This uses the same logic we worked out for SkOncePtr to reduce
the memory footprint of SkOnce from a done byte and lock byte
to a single 3-state byte:

  - NotStarted: no thread has tried to run fn() yet
  - Active:     a thread is running fn()
  - Done:       fn() is complete

Threads which see Done return immediately.
Threads which see NotStarted try to move to Active, run fn(), then move to Done.
Threads which see Active spin until the active thread moves to Done.

This additionally fixes a too-weak memory order bug in SkOncePtr,
and adds a big note to explain.

BUG=skia:
GOLD_TRYBOT_URL=
https://gold.skia.org/search2?unt=true&query=source_type%3Dgm&master=false&is...

Committed:
https://skia.googlesource.com/skia/+/df02d338be8e3c1c50b48a3a9faa582703a39c07

Committed:
https://skia.googlesource.com/skia/+/650f9e9a2630026d578970c6d031d7d4644f63b4
==========

[commit-bot: I haz the power, 2016-04-20 20:49:19]

Committed patchset #5 (id:80001) as
https://skia.googlesource.com/skia/+/650f9e9a2630026d578970c6d031d7d4644f63b4