Assembler atomics #10147

kjbracey · 2019-03-18T15:37:28Z

Description

Reimplement atomic code in inline assembly. This can improve optimisation, and avoids potential architectural problems with using LDREX/STREX intrinsics.

API further extended:

Bitwise operations (fetch_and/fetch_or/fetch_xor)
fetch_add and fetch_sub (like incr/decr, but returning old value - aligning with C++11)
compare_exchange_weak
Explicit memory order specification
Basic freestanding template overloads for C++

This gives our existing C implementation essentially all the functionality needed by C++11.

An actual Atomic<T> template based upon these C functions could follow.

Preliminary work for #10104.

Pull request type

[ ] Fix
[ ] Refactor
[ ] Target update
[X] Functionality change
[ ] Docs update
[ ] Test update
[ ] Breaking change

Release Notes

Atomic APIs separated into mbed_atomic.h from mbed_critical.h. Code not including mbed.h may need to to check their includes.
Atomic fetch_add, fetch_sub, fetch_and, fetch_or, fetch_xor and compare_exchange_weak operations added.
Explicit memory ordering specification added (don't use unless familiar with C++11 memory model)
Freestanding atomic function templates added.

ciarmcom · 2019-03-18T16:01:05Z

@kjbracey-arm, thank you for your changes.
@ARMmbed/mbed-os-core @ARMmbed/mbed-os-maintainers please review.

kjbracey · 2019-03-20T08:16:43Z

I'm now looking at a full rework here, which will reduce the source size and turn it into assembler, with the possibility of better code generation.

kjbracey · 2019-03-20T08:17:07Z

Will close temporarily.

cmonr · 2019-03-25T19:37:31Z

Hmm, this got nicked with the Astyle update bork.

Incoming fix from: #10221

cmonr · 2019-03-26T01:41:14Z

Hmm... The astyle fix is missing this fix: #10222

NirSonnenschein · 2019-03-26T08:27:45Z

@kjbracey-arm please re-base when you have time, it should resolve the astyle issue

mbed-ci · 2019-04-25T17:01:47Z

Test run: FAILED

Summary: 3 of 7 test jobs failed
Build number : 6
Build artifacts

Failed test jobs:

jenkins-ci/mbed-os-ci_build-GCC_ARM
jenkins-ci/mbed-os-ci_build-IAR
jenkins-ci/mbed-os-ci_build-ARM

adbridge · 2019-04-26T10:07:20Z

@kjbracey-arm Please look at test failures

Reimplement atomic code in inline assembly. This can improve optimisation, and avoids potential architectural problems with using LDREX/STREX intrinsics. API further extended: * Bitwise operations (fetch_and/fetch_or/fetch_xor) * fetch_add and fetch_sub (like incr/decr, but returning old value - aligning with C++11) * compare_exchange_weak * Explicit memory order specification * Basic freestanding template overloads for C++ This gives our existing C implementation essentially all the functionality needed by C++11. An actual Atomic<T> template based upon these C functions could follow.

These are platform tests, but rely on the RTOS to run multiple threads to exercise it. (The atomics are still useful in non-RTOS, to protect against interrupt handlers, but testing versus other threads is easier. The implementation is the same either way, so doesn't seem worth testing non-RTOS specifically).

Get rid of a volatile, and use atomics to synchronise with the interrupt routine instead. Useful as a non-RTOS basic compilation check for the atomics - the fuller atomic test relies on the RTOS.

Volatile makes no real difference when we're using assembler, or locked functions, but leaving it off could be more efficient for the basic loads and stores. So add non-volatile overloads in C++ for them.

kjbracey · 2019-04-26T10:27:11Z

Thought I'd done enough fork tests to find all the glitches, but apparently not.

Fixed the flash_api.c build failure there for LPC55S69_NS. Didn't see anything else.

adbridge · 2019-04-29T09:39:34Z

@ARMmbed/mbed-os-core Could somebody please review this ?

0xc0170 · 2019-05-02T14:02:23Z

@ARMmbed/mbed-os-core Could somebody please review this ?

@donatieng @bulislaw Can you help reviewing these?

bulislaw

Looks good, but to be fair I didn't dive into the implementation file.

adbridge · 2019-05-07T15:06:48Z

CI started

mbed-ci · 2019-05-07T20:59:04Z

Test run: FAILED

Summary: 2 of 11 test jobs failed
Build number : 7
Build artifacts

Failed test jobs:

jenkins-ci/mbed-os-ci_dynamic-memory-usage
jenkins-ci/mbed-os-ci_greentea-test

0xc0170 · 2019-05-10T07:56:16Z

CI restarted (py3 problem should be gone now)

mbed-ci · 2019-05-10T13:05:39Z

Test run: FAILED

Summary: 1 of 11 test jobs failed
Build number : 8
Build artifacts

Failed test jobs:

jenkins-ci/mbed-os-ci_exporter

0xc0170 · 2019-05-12T19:10:20Z

Exporters restarted

SanaMasood · 2019-05-16T09:33:53Z

Build errors in mbed-os tests with ARMC5 toolchain started to come from 13th of May.
Seems like this pull request(#10147) is breaking everything. Build error started to appear in this file 'mbed_atomic_impl.h' with ARMC5 toolchain.

Can you please check why this pull request is effecting the build? and they seem related too!

[DEBUG] Output: "./mbed-os/platform/internal/mbed_atomic_impl.h", line 306: Error: #52: expected a macro parameter name
[DEBUG] Output: "./mbed-os/platform/internal/mbed_atomic_impl.h", line 498: Warning: #3732-D: use of this instruction in inline assembler is deprecated
[DEBUG] Output: "./mbed-os/platform/internal/mbed_atomic_impl.h", line 498: Error: #3081: expected end of line or a ";"

The build was passing on May 13, 2019 2:06:40 AM but after that it failed on May 13, 2019 2:43 PM (UK time)

kjbracey · 2019-05-16T10:16:57Z

@SanaMasood - thanks for notification, I've made a PR for the fix.

ciarmcom requested review from a team March 18, 2019 16:01

ciarmcom added the needs: review label Mar 18, 2019

cmonr added the release-version: 5.13.0-rc1 label Mar 19, 2019

kjbracey closed this Mar 20, 2019

kjbracey removed the needs: review label Mar 20, 2019

kjbracey reopened this Mar 25, 2019

kjbracey force-pushed the atomic_bitwise branch from b8bf0f7 to cc1747a Compare March 25, 2019 15:50

kjbracey changed the title ~~Atomic bitwise operations~~ Assembler atomics Mar 25, 2019

kjbracey force-pushed the atomic_bitwise branch from cc1747a to a2a31ad Compare March 25, 2019 15:55

cmonr added the needs: review label Mar 25, 2019

cmonr added the needs: preceding PR label Mar 26, 2019

kjbracey force-pushed the atomic_bitwise branch 3 times, most recently from 827cc51 to 16eee15 Compare March 26, 2019 12:11

cmonr added needs: work and removed needs: preceding PR needs: review labels Mar 26, 2019

kjbracey force-pushed the atomic_bitwise branch 6 times, most recently from 53a3682 to 3c36dfe Compare March 27, 2019 14:27

kjbracey added 4 commits April 26, 2019 13:12

Critical tests: use atomics

4fe06e0

Get rid of a volatile, and use atomics to synchronise with the interrupt routine instead. Useful as a non-RTOS basic compilation check for the atomics - the fuller atomic test relies on the RTOS.

Add non-volatile atomic loads and stores

2bb4045

Volatile makes no real difference when we're using assembler, or locked functions, but leaving it off could be more efficient for the basic loads and stores. So add non-volatile overloads in C++ for them.

kjbracey force-pushed the atomic_bitwise branch from 1632263 to 2bb4045 Compare April 26, 2019 10:22

adbridge added needs: review and removed needs: CI labels Apr 29, 2019

0xc0170 requested review from donatieng and bulislaw May 2, 2019 14:02

bulislaw approved these changes May 3, 2019

View reviewed changes

0xc0170 added needs: CI and removed needs: review labels May 6, 2019

0xc0170 added ready for merge and removed needs: CI labels May 13, 2019

0xc0170 merged commit 3ea1c56 into ARMmbed:master May 13, 2019

0xc0170 removed the ready for merge label May 13, 2019

kjbracey deleted the atomic_bitwise branch May 14, 2019 08:38

kjbracey mentioned this pull request May 16, 2019

ARMC5: fix mbed_atomic_impl.h assembly #10594

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Assembler atomics #10147

Assembler atomics #10147

kjbracey commented Mar 18, 2019 •

edited

Loading

ciarmcom commented Mar 18, 2019

kjbracey commented Mar 20, 2019

kjbracey commented Mar 20, 2019

cmonr commented Mar 25, 2019 •

edited

Loading

cmonr commented Mar 26, 2019

NirSonnenschein commented Mar 26, 2019

mbed-ci commented Apr 25, 2019

adbridge commented Apr 26, 2019

kjbracey commented Apr 26, 2019

adbridge commented Apr 29, 2019

0xc0170 commented May 2, 2019

bulislaw left a comment

adbridge commented May 7, 2019

mbed-ci commented May 7, 2019

0xc0170 commented May 10, 2019

mbed-ci commented May 10, 2019

0xc0170 commented May 12, 2019

SanaMasood commented May 16, 2019 •

edited

Loading

kjbracey commented May 16, 2019

Assembler atomics #10147

Assembler atomics #10147

Conversation

kjbracey commented Mar 18, 2019 • edited Loading

Description

Pull request type

Release Notes

ciarmcom commented Mar 18, 2019

kjbracey commented Mar 20, 2019

kjbracey commented Mar 20, 2019

cmonr commented Mar 25, 2019 • edited Loading

cmonr commented Mar 26, 2019

NirSonnenschein commented Mar 26, 2019

mbed-ci commented Apr 25, 2019

Test run: FAILED

adbridge commented Apr 26, 2019

kjbracey commented Apr 26, 2019

adbridge commented Apr 29, 2019

0xc0170 commented May 2, 2019

bulislaw left a comment

Choose a reason for hiding this comment

adbridge commented May 7, 2019

mbed-ci commented May 7, 2019

Test run: FAILED

0xc0170 commented May 10, 2019

mbed-ci commented May 10, 2019

Test run: FAILED

0xc0170 commented May 12, 2019

SanaMasood commented May 16, 2019 • edited Loading

kjbracey commented May 16, 2019

kjbracey commented Mar 18, 2019 •

edited

Loading

cmonr commented Mar 25, 2019 •

edited

Loading

SanaMasood commented May 16, 2019 •

edited

Loading