[ACTIVITY] Monday 14th - Friday 18th March

Port to microinstance - TCWG-432 [5/10] * Non-lab side of minimal trust benchmarking * More investigation of runtime anomalies * Reordered builder phases to do useful work while waiting for targets * Updated everything to work with benchmarking LAVA user (rather than running as me) Automated backp

[ACTIVITY] 29th Februrary - 11th March

Port to microinstance - TCWG-432 [17/10] * Investigating difference between LAVA and 'desktop Juno' runtimes ** Some of this was down to piles of /dev/console output - redirecting to file improved SPEC build time by 75%! ** Some cases make sense, others remain unexplained ** Might just go away if w

[ACTIVITY] 22nd - 26th February

Port to microinstance - TCWG-432 [7/10] * Merged last few months of development back to benchmarking branch * Restored support for multiple targets per builder * Updated builder landed, altered jobs to work with it ** Removed assumption that host filesystem is non-persistent ** Stacked up test runs

[ACTIVITY] 15th - 19th Februrary

Bug with compiler flag handling - (no ticket) [2/10] * Coremark-Pro was ignoring compiler flags * Fixed that, made flag handling consistent across all benchmarks Release benchmarking via Jenkins - TCWG-348 [1/10] * Seems to work with test workload Port to microinstance - TCG-432 [3/10] * Looked a

[ACTIVITY] 8th - 12th February

Port to microinstance - TCWG-432 [2/10] * Better reporting/error trapping in dispatch scripts * Some back and forth with Lab about builder implementation * Started assessing target stability * Fixed a few bugs in reporting/bundle generation Document benchmarking infrastructure - TCWG-496 [2/10] *

[ACTIVITY] 1st - 5th February

Automated release benchmark Jenkins job - TCWG-348 [2/10] * Drafted a job, it succeeds in dispatching to the uinstance * Needs testing Port to microinstance - TCWG-432 [5/10] * Almost entirely fixing up the CPU2006 -> LAVA reporting * Should now be able to report CPU2000 results as well Misc [3/1

[ACTIVITY] 25th - 29th January

Port to microinstance - TCWG-432 [4/10] * Went a few more rounds with the race condition, finally nailed it * Tidied up transfer of Coremark results to LAVA * Added metadata to LAVA results * Sketched some charts using LAVA reporting * Improved runtime benchmark logging Benchmarking infrastructure

[ACTIVITY] 18th - 22nd January

Port to microinstance - TCWG-432 [2/10] * More fallout from attempts to fix race condition * Began cleaning job submission YAML Controlled image builds - TCWG-360 [1/10] * (Re)learned how to generate images from hwpack + filesystem Struck down by flu [7/10] ___

[ACTIVITY] 11th-15th January

Port to microinstance - TCWG-432 [3/10] * Fallout from attempts to fix race condition * Various minor fixes - simplifications, better reporting Backport benchmarking - TCWG-352 [1/10] * Decoupled 'target triple' from 'toolchain' name ** Immediately, to let me benchmark with Juno-built native gcc *

[ACTIVITY] 4th-8th January

Centralized benchmark source - TCWG-354 [1/10] * Fixed some bugs in Coremark reporting * Experimented some more with clang builds, set this aside for now Port to microinstance - TCWG-432 [3/10] * Fixed some false assumptions exposed by change to uinstance * Worked on LAVA side of coremark reportin

Re: gcc-linaro-5.1 vs gcc-linaro-4.8

Hello, I'm not sure from the information below whether you have observed a performance gap, or are expecting to observe one. Have you seen a performance gap? Regards, Bernie On 5 January 2016 at 10:29, Xiaofeng Ren wrote: > Hello All, > > I found one difference between gcc-linaro-5.1 vs gcc-li

[ACTIVITY] 14th - 21st December

NB Last _6_ working days. Centralised benchmark source - TCWG-354 [6/10] * Abe integration * Wrote up some notes on collaborate * Enabled clang build, which flushed out some bugs, now fixed * LAVA reporting script Port to microinstance - TCWG-432 [1/10] * Investigating issues with Juno boot ** On

Re: [ACTIVITY] 7th-11th December

Oh, and [1/10] misc. ___ linaro-toolchain mailing list linaro-toolchain@lists.linaro.org https://lists.linaro.org/mailman/listinfo/linaro-toolchain

[ACTIVITY] 7th-11th December

Centralised benchmark source - TCWG-354 [6/10] * Understood CoremarkPro run rules/behaviour better * Experimented with shortening runs while still giving meaningful results * Cleaned up build/run scaffolding Port to microinstance - TCWG-432 [1/10] * Got access to microinstance, learned a bit about

[ACTIVITY] 30th November - 4th December

Holiday [2/10] Port to microinstance - TCWG-432 [1/10] * Set up reporting for CPU2006 * Learned how to generate metadata, but not how to use it Trigger benchmarks on backports - TCWG-352 [2/10] * Figured out the rough shape * Created, didn't test, rough implementation TCWG-354 [3/10] * Build/run

[ACTIVITY] 23rd - 27th November

Controlled image builds - TCWG-360 [1/10] * Tried, failed to generate bootable images Jenkins benchmarking job - TCWG-348 [5/10] * Jenkins job functional on kvms in main instance * Wrote job-dispatch script for non-Jenkins use cases Juno crashdump [1/10] * Struggled with, worked around network pr

[ACTIVITY] 16th - 20th November

Controlled image builds - TCWG-360 [2/10] * A few more test/debug cycles with ci-loop-built image Jenkins benchmarking job - TCWG-348 [3/10] * YAML-ised Jenkins job, more test/debug cycles Juno crashdump [1/10] * Got a usable dump (via alt-sysrq-c) with latest patches plus some fiddling SPEC-on-

[ACTIVITY] 9th-13th November

Implement LAVA jobs for microinstance - TCWG-432 [6/10] * Refactoring to permit sharing of code between uinstance & main instance, as far as possible * Further refactoring for sane submission of bundles without inserting LAVA assumptions in the wrong places * Tested as far as possible in main insta

[ACTIVITY] 2nd - 6th November

Holiday [2/10] Juno crash analysis [2/10] * Spent some time fiddling with kexec on AArch64 * Worked in one very specific case * Another patch series is (apparently) coming, will look out for it and try again SPEC-on-Android [2/10] * Supporting Qian on getting this working * Wrote a readme for the

[ACTIVITY] 26th - 30th October

Noise control experiments - TCWG-358 [3/10] * Some analysis of data to date Debian filesystem - TCWG-360 [3/10] * Got stuck on LAVA interactions * Now booting-to-LAVA-usability, needs some cleanup and testing with real benchmark runs Benchmarking-via-Jenkins - TCWG-348 [1/10] * Picked back up on

[ACTIVITY] 19th - 23rd October

Mustang benchmarking bringup - (no ticket) [2/10] Controlled Debian build - TCWG-360 [1/10] * Constructed filesystem, works OK as chroot * Next step is to boot it Investigate Workload Automation framework - TCWG-361 [1/10] * Seems to basically work * SPEC support dubious Misc * LAVA uinstance ba

[ACTIVITY] 12th - 16th October

Investigate effectiveness of noise-control measures - TCWG-358 [4/10] * Recovered from a couple more crashes, existing logging sheds no light * Explored the existing data a bit, need more to draw any conclusions Controlled image builds - TCWG-360 [2/10] * Looked into controlled, repeatable debian

[ACTIVITY] 4th - 9th October

LAVA uinstance for benchmarking - TCWG-396 [3/10] * Long meeting with Renato + LAVA/lab people * Much thinking about the security side * Wrote up rough draft of what I understand the design to be Investigate effectiveness of noise control measures - TCWG-358 [3/10] * Set up new host node * Dealt w

[ACTIVITY] 28th August - 2nd October

Connect recovery - [2/10] Investigate effectiveness of noise-control measures - TCWG-358 [3/10] * Host node crashed, tried to recover it, failed, started building a new one Jenkins automation - TCWG-348 [3/10] * Everything in place up to first interaction with LAVA kvm * Considered LAVA team's mi

[ACTIVITY] 14th - 18th September

Investigate effectiveness of noise-control measures - TCWG-358 [6/10] * Finished setup, started running experiments * Initial data pretty noisy, more runs needed * Kicked off more runs for week of Connect Ensure that all critical data is logged - TCWG-349 [1/10] * Logged a few more variables, test

[ACTIVITY] 7th-11th September

Ensure all critical (benchmarking) data is logged - TCWG-349 [1/10] * Added logging of several factors * Documented what we do and don't log Noise control experiments on Juno - TCWG-349 [4/10] * Rediscovered that my Juno is an r0, not an r1 * Rebuilt target image in a carefully scripted way * Fidd

[ACTIVITY] 31st August - 4th September

Holiday [4/10] Multinode wrapper - TCWG-350 [2/10] * Merged to benchmarking branch, roughly documented * Tested & added job definition templates Setting up VPN [2/10] * Much struggling with mutt, pgp, VM vs real system, Mac vs Linux * Still not working, but wrote up what I've learned on the Colla

[ACTIVITY] 24th - 28th August

(Reporting with new Jira numbers) Holiday [6/10] Investigate effectiveness of noise control - TCWG-358 [2/10] * Learned to build OE filesystems * Got Juno running with more or less provenance-tracked firmware, kernel, filesystem Misc - [2/10] * Pushed through some updates to benchmark sources *

[ACTIVITY] 17th - 21st August

Upcoming Absences * Away from this Wednesday until next Tuesday, inclusive Benchmark infrastructure - TCWG-360 [1/10] * Finished filling out cards/bugs for benchmarking work * Which makes the rest of this report more fine-grained Multinode wrapper - TCWG-888 [2/10] * Completed oversubscription wo

[ACTIVITY] 10th - 14th August

Benchmark infrastructure - TCWG-360 [6/10] * Fixed remaining known critical issues in multinode ** Testing stymied by LAVA oversubscription ** Attempted to implement some workarounds for oversubscription * Some discussion/investigation on generating filesystems for benchmarking * Started brain-dump

[ACTIVITY] 27th - 31st July

Benchmark infrastructure - TCWG-360 [8/10] * Testing found many problems in multinode * Iterating to solutions Misc [2/10] =Plan= Holiday next week. Then back to fixing multinode, incorporating into jenkins, noise control experiments ___ linaro-toolc

[ACTIVITY] 20th - 24th July

Benchmark infrastructure - TCWG-360 [6/10] * Some user support/bugfixing/bugraising * Multinode job more or less working (not fully tested) * Additional restructuring got rid of some more complexity ** Though if my simplifying assumption doesn't hold, I'll have to put it back Benchmarking 101 pres

[ACTIVITY] 13th - 17th June

Benchmark infrastructure - TCWG-360 [5/10] * Worked through my Jenkins issues with Fathi, raised some tickets at him * Converting LAVA end into multinode job ** Having some trouble with multinode API Benchmarking 101 presentation [3/10] * 1/2 day of discussions/reading, full day of redrafting * Lo

[ACTIVITY] 6th - 10th June

Benchmark infrastructure - TCWG-360 [5/10] * More thinking/prototyping sufficiently-secure Jenkins benchmarking * Converting LAVA end into multinode job Benchmarking presentation [2/10] * A couple of helpful discussions * Read a couple of helpful docs Misc [3/10] =Plan= * Complete multinode jo

[ACTIVITY] 29th - 3rd June

Benchmarking presentation [7/10] * More reading * Ran through a couple more drafts Misc [3/10] * Featuring a bug in my backup scripts that took ~1/10 to fix =Plan= Back to benchmark automation as main activity Presentation in the background ___ linaro

[ACTIVITY] 22nd - 26th June

Benchmark automation - TCWG-360 [3/10] * Created a partial Jenkins prototype * Considered some security issues Benchmarking presentation [5/10] * Drafted some slides, did some reading Misc [2/10] =Plan= More of the above ___ linaro-toolchain mailing

[ACTIVITY] 15th-19th June

Benchmark automation - TCWG-360 [7/10] * Arndales stopped booting ** Package servers for elderly filesystem had gone ** Investigated some approaches to creating more stable filesystems ** Realized I could just updated image to point at old-releases, so did that for now * _More_ time thinking about

[ACTIVITY] 1st-5th June

Holiday [2/10] Benchmark Automation - TCWG-360 [4/10] * Learned about central Jenkins instance * Considered ways of integrating benchmark scripts with said instance * Considered alternative way for benchmark scripts to interoperate with LAVA Misc [4/10] * Large mail backlog * Travel prep =Plan=

[ACTIVITY] 18th - 22nd May

Holiday Thursday afternoon and Friday [3/10] Benchmark automation - TCWG-360 [4/10] * CPU2006 patches cleaned up, submitted * A few other small fixes Misc [3/10] * ~1/10 getting a Juno running =Plan= Holiday next week, then: Pick a direction on streaming vs listeners, write down the reasoning

[ACTIVITY] 11th - 15th May

Benchmark automation - TCWG-360 [7/10] * Convinced benchmark scripts to run entirely within LAVA * Did a couple of benchmark runs for Kugan * Cleaned up my CPU2000 patches and committed to central repository Misc [3/10] =Plan= Clean up, commit CPU2006 patches Pick a direction on streaming vs li

[ACTIVITY] 4th - 8th May

(Public) Holiday [2/10] Benchmark automation - TCWG-360 [6/10] * More unit tests * Refactoring - test writing is an effective spotlight * Experimented with LAVA log streaming as a 'listener' replacement ** Reduces complexity, for (possibly insignificant) costs in generality and scalability * Gave

[ACTIVITY] 27th April - 1st May

Holiday [2/10] Benchmark automation - TCWG-360 [5/10] * Finished rolling backbench patches into review * Submitted some test suite patches to reduce the number of untested cases * Began writing tests for benchmarking in general Misc [3/10] ___ linaro-to

[ACTIVITY] 20th - 24th April

benchmark automation - TCWG-360 [3/10] * Shifted more generic patches from backport branch to review * One complete run of backport benchmarking before lab went down Misc [7/10] * Most of this (4 or 5 tenths) is reaction to lab downtime =Plan= Benchmarking test suite Fish out, finish off though

Re: Linaro Toolchain Product Release Process Changes (Effective 2015.04)

> TCWG will introduce the concept of a 'release candidate source archive' and > 'release candidate binary toolchain archive' six weeks before the projected > release date of the Quarterly Binary Toolchain Release. For example, the > 2015.05 Quarterly Binary Toolchain and Quarterly Source Archive w

[ACTIVITY] 13th - 17th April

Benchmark automation - TCWG-360 [5/10] * Flushed the rest of the easy stuff from my 'small fixups' to my 'staging' branch * Started rolling generic patches from my backport-benchmark branch into gerrit * Prodded at the prototype backport-benchmark job until the pieces worked ** Ran out of time to t

[ACTIVITY] 5th - 10th April

catomics - TCWG-436 [4/10] * Fiddled with SPEC runs with unstripped binaries on tiny Juno discs * Got a run for the interesting subset of SPEC, non-atomic catomics still show no effect * Tried a few catomics variations with the glibc malloc microbenchmark, here non-atomic catomics show a significan

[ACTIVITY] 30th March - 3rd April

catomics - TCWG-436 [4/10] * Fiddled catomics to not use atomics at all, set off a new SPEC run * Not using atomics seems wrong in general, but OK for malloc/free Benchmark automation - TCWG-360 [1/10] * Quick look at Tyler's LAVA console streamer * Ported benchmark scripts to work on Debian ** Tu

[ACTIVITY] 23rd - 27th March

catomics - TCWG-436 [5/10] * Got pointed at a suitable set of benchmarks, results still underwhelming * However, patches were using relaxed atomics rather than no atomics at all * Fiddled abe into building sysroots for me (I get libstdc++ that way) Misc - [5/10] * Tidied up some 'perf shotgun' scr

[ACTIVITY] 16th - 20th March

catomics - TCWG-436 [6/10] * Started a series of runs on a local board I'd borrowed ** Then had to give it back before they'd really got anywhere * Got some, possibly dubious, results back from A15 from previous week ** If the results are worth anything, they suggest that catomics don't achieve any

Re: [ACTIVITY] 9th - 13th March

Urgh - I filled in the tenths from the wrong week. For project management, the correct numbers are: LDTS-1238 [4/10] TCWG-436 [1/10] TCWG-360 [1/10] Misc [4/10] (about 1/2 of this was juno-01 fixing) On 23 March 2015 at 08:57, Bernie Ogden wrote: > Juno cache effects - LDTS-1238 [6/10] > *

[ACTIVITY] 9th - 13th March

Juno cache effects - LDTS-1238 [6/10] * Seems to be mainly due to (expected) instruction scheduling limitations, and prefetcher effects * Reported back, hopefully this will wrap up now catomics - TCWG-436 [1/10] * Shepherding benchmark runs in LAVA, usual problems with ssh-agent, juno contention a

[ACTIVITY] 2nd - 6th March

Juno cache effects - LDTS-1238 [6/10] * Ran more experiments * Cobbled together some gdb/python script to run perf stat within an address range * Becoming more confident in my hypotheses catomics - TCWG-436 [1/10] * Found and fixed some more sysroot benchmark bugs * Kicked off a bunch of spec runs

[ACTIVITY] 23th - 27th Februrary

Holiday Friday [2/10] catomics - TCWG-436 [2/10] * Fixed broken benchmark-against-sysroot code in scripts * Built a lot of sysroots Juno cache effects - LDTS-1238 [4/10] * Quite a bit of reading and learning some more perf and other tools * Formed a few hypothesis, started testing * Looking promi

Re: [ACTIVITY] 16th - 20th February

Oh -- and I'm on holiday this Friday (27th). On 23 February 2015 at 10:12, Bernie Ogden wrote: > Connect detox [2/10] > > ABE benchmarking automation - TCWG-360 [5/10] > * Backport benchmarking > ** Harder than expected, have to build the backport into a binary, > the

[ACTIVITY] 16th - 20th February

Connect detox [2/10] ABE benchmarking automation - TCWG-360 [5/10] * Backport benchmarking ** Harder than expected, have to build the backport into a binary, then benchmark that binary, passing information between jobs. ** AFAIK I'm blazing the Jenkins-chainging trail for us. * Release benchmarkin

[ACTIVITY] 2nd - 13th February

Holiday [3/10] ABE benchmarking automation - TCWG-360 [3/10] * Initial Jenkins implementation Investigating cache effects on Juno - LDTS-1238 [2/10] * One effect due to write streaming, others still to look at Misc [2/10] * Unpacking, some background catomics Connect [10/10] =Plan= Carry on with

Re: [ACTIVITY] 19th - 23rd January

Correction - that was 26th - 30th January. On 2 February 2015 at 14:34, Bernie Ogden wrote: > ABE benchmarking automation - TCWG-360 [8/10] > * Pursued sneaky bugs through twisty mazes of bash > * Learned that ControlMaster does not play nice with my scripts > * Learned more about

[ACTIVITY] 19th - 23rd January

ABE benchmarking automation - TCWG-360 [8/10] * Pursued sneaky bugs through twisty mazes of bash * Learned that ControlMaster does not play nice with my scripts * Learned more about IPC pitfalls * Cleaned up build behaviour of scripts somewhat * Implemented support for benchmarking against sysroot

[ACTIVITY] 19th - 23rd Jan

ABE benchmarking automation - TCWG-360 [6/10] * Fixed an awkward escaped process * Struggled with LAVA timeouts & discovered that some Junos accidentally have a baked-in timeout * More discussion/thinking about where to store benchmark source * Some trouble with intermittent flakiness when copying

[ACTIVITY] 12th - 16th January

ABE benchmarking automation - TCWG-360 [5/10] * Converted Juno runs from OE to Ubuntu * Fixed cross-compilation in SPEC2000 * Picked up a couple of users, spent some time supporting them * Discussion about storage catomics - TCWG-436 [3/10] * Took a remarkably long time to make the last few patche

[ACTIVITY] 4th - 9th January

ABE benchmarking automation - TCWG-360 [8/10] * Added some sanity checks, fixed some bugs in LAVA URL handling * Some struggling with Jenkins, with support from Rob * Compress builds before scp-ing * Resurrected SPEC, fixed some problems I'd forgotten about Misc [2/10] * Email catchup, start-of-y

[ACTIVITY] 15th - 19th December

ABE benchmarking automation - TCWG-360 [6/10] * TCWG/LAVA lab interactions working with just a couple hacks * Jenkins job created (thanks Rob) and beginning to work Misc - [4/10] * Included a quick look at Will's single-thread performance patches - unfortunately, no time to actually do anything w

Re: renaming completed

Oh, it's as easy as: git push /home/bernie.ogden/src/abe bernie_benchmarking:bernie/benchmarking Benchmarking branch is now in the abe repository at bernie/benchmarking. I'll pester Rob about automation tomorrow ;) On 20 November 2014 at 10:44, Bernie Ogden wrote: > I'

[ACTIVITY] 8th - 12th December

ABE benchmarking - TCWG-360 [7/10] * Finished post-move LAVA lab workarounds * Should be flexible enough for TCWG lab too, but we'll see * Refactored sources - now more robust and maintainable lowlevellock.h comments - CARD-341 [1/10] * A further round of changes, got an OK Raised a bug on GCC re

[ACTIVITY] 1st - 5th December

Holiday [6/10] Misc [3/10] * Mail backlog * Moved all current AArch64 work off 'my' Juno, as ARM needed it back * A little bit of a look at another possible memcpy performance issue libm exercising - TCWG-558 [1/10] * Reduced 'needless calls to pow' to a simple test case ** Found that this is act

[ACTIVITY] 24th - 28th November

ABE benchmarking - TCWG-360 [4/10] * Implemented most of a solution to the 'must be in same network' restriction libm exercising - TCWG-558 [4/10] * lulesh generates needless calls to pow on AArch64 (as opposed to 'pow is slow') ** Working on a reduced test case * Ran a chunk of benchfft, left a p

[ACTIVITY] 17th - 21st November

cbuild2/ABE benchmarking - TCWG-360 [1/10] * Attempted to use LAVA for benchmarks ** Fell over on lack of TCWG machines in same network libm exercising - TCWG-558 [6/10] * Much fiddling with chroots * Some fiddling with benchfft * Little actual progress Meetings/mail/etc - [3/10] =Plan= libm e

Re: renaming completed

I'll need to move my benchmarking branch. How long is 'a while'? On 20 November 2014 07:40, Maxim Kuvyrkov wrote: > On Nov 19, 2014, at 2:01 PM, Rob Savoye wrote: > >> I've completed renaming Cbuildv2 to Abe, as well as modifying all the >> board files for remote testing to match what the new D

[ACTIVITY] 10th - 14th November

cbuild2 benchmarking - TCWG-360 [5/10] * Fixed some bugs and did some general tidying up * Pulled SPEC2000 into my framework * Did some test runs on local machines, looks promising libm exercising - TCWG-558 [3/10] * Much fiddling with one benchmark (MCB) * Experimented, thought about methodology

[ACTIVITY] 3rd - 7th November

cbuild2 benchmarking - TCWG-360 [4/10] * SPEC 2006 cross-built binaries now running. Easy when you know how. libm exercising - CARD-1693 [4/10] * Understood improved libm usage on 1 benchmark. * Tried 3 more. 1 shows significantly more time in libm on AArch64 over AArch32. Meetings/mail/etc - [2/

[ACTIVITY] 27th - 31st October

cbuild2 benchmarking - TCWG-360 [8/10] * Scripts build SPEC 2006 tools on x86_64 and arm, not yet on AArch64 * Scripts cross-build for x86 to arm and aarch64 * Cross-built binaries refuse to run Meetings/mail/etc - [2/10] =Plan= cbuild2 benchmarking * Make cross-built binaries run, collect repo

[ACTIVITY] 20th - 24th October

cbuild2 benchmarking - TCWG-360 [2/10] * Figured out how to cook my own OE images * Started remembering how to build spec libm exercising - CARD-1693 [2/10] * Borrowed a usable Juno * Found that lapack tests segfault on AArch64 * Ran linpack hpl, didn't observe it exercising libm much ** Haven't r

[ACTIVITY] 13th - 17th October

cbuild2 benchmarking - TCWG-360 [3/10] * A few small enhancements and bug-fixes * Tried to run spec2006 on Juno, looks tricky libm profiling - CARD-1693 [3/10] * Pulled together lapack + blas * Looked a bit at how it exercises libm on x86 * Tried to look at how it exercises libm on Juno, looks tri

[ACTIVITY] 6th - 10th October

cbuild2 benchmarking - TCWG-360 [7/10] * Fixed/worked around some especially resilient bugs ** Mostly relate to running benchmarks through LAVA, which may be less important in the near future * Wrote a doc * Upstreamed some code that works about as I want it to Meetings/mail/etc [3/10] =Plan= c

[ACTIVITY] 29th September - 3rd October

cbuild2 benchmarking - TCWG-360 [7/10] * Cleaned up and upstreamed some code that works * Fixed a couple of issues raised by LAVA people ** No longer scraping the logs ** No longer assuming stable IP addresses lowlevellock.h comments - CARD-341 [1/10] * Respun based on Carlos' comments Meetings/m

[ACTIVITY] 8th - 26th September

cbuild2 benchmarking - TCWG-360 [9/10] * Knocked off a lot of rough edges * Now working fairly robustly lowlevellock.h comments - CARD-341 [4/10] * Got a bit stuck trying to follow condvar locking * But wasn't really needed to describe the code in question LCA [10/10] LCA recovery day [2/10] Me

[ACTIVITY] 1st - 4th September

=Progress= cbuild2 benchmarking - TCWG-360 [8/10] * Got an end-to-end run in my test environment * Patched tests to run more selectively (in review) Meetings/mail/etc - [2/10] =Plan= cbuild2 benchmarking: * Clean up some lose ends * Sort out source/results storage glibc: * Finish comments in l

[ACTIVITY] 18th - 22nd August 2014

=Progress= cbuild2 benchmarking - TCWG-360 [5/10] * Proposed a plan for benchmark storage * A bit more thought on cross-running * Prototyped script to launch/run/release on LAVA targets (e.g. wg) ** The 'run' part is pretty much independent of the 'LAVA' part lowlevellock.h commentary - [2/1

[ACTIVITY] 8th - 14th August 2014

=Progress= cbuild2 benchmarking - TCWG-360 [4/10] * Some disagreements with gerrit * One of my patches broke bare metal builds (now fixed) * More review of the way we store benchmark sources Other - [5/10] * Meetings, mail, etc (featuring newlib build options, defaults, cbuild vs benchmark script

[ACTIVITY] 4th - 8th August 2014

=Progress= cbuild2 benchmarking- TCWG-360 [6/10] * Existed patches tidied up and gerritified * A few drive-by bugfixes * Cross-building support added and also gerritified lowlevellock patch [1/10] * Resubmitted bugfix, reacted to review * Some wrestling with space-mangling evil in mail clients M

[ACTIVITY] 28th July - 1st August 2014

=Progress= cbuild2 benchmarking - TCWG-360 [7/10] * Integrated benchmarking more into the core of cbuild2 * 'Audited' the benchmarks lying around on toolchain64 * Persuaded cbuild2 to use git-over-ssh Meetings/mail/etc [3/10] =Plan= cbuild2 benchmarking * Convert eembc integration into reviewab

[ACTIVITY] July 19th - 25th

=Progress= GNU Cauldron TCWG Sprint [8/10] * Really helpful to meet (almost) everybody * Some useful discussion, too memcpy on A15 - TCWG-390 [1/10] * Should have let this lie but I had an odd 1/2 day and some data begging to be looked at * Turned out I'd fat-fingered the wrong data *

[ACTIVITY] July 14th - 18th

=Progress= cbuild2 benchmarking - TCWG-360 [7/10] * Much fighting with benchmarking branch on a panda * Ported eembc benchmarking over to my way of doing things * A bit of cleanup around the 'reduce noise by shutting down services' stuff * Slew my ailing Ubuntu VM, started bringing up a beagle as

[ACTIVITY] July 7th - 11th

=Progress= memcpy regression on A9 - TCWG-390 [6/10] * Not down to the branch predictor after all * Can be fixed for A9 with explicit pld, but this isn't free cbuild2 benchmarking - TCWG-360 [1/10] * Dug out my branch, merged in trunk, refreshed memory Meetings/mail/etc [3/10] * Including a litt

[ACTIVITY] June 30th - July 4th

=Progress= memcpy regression on A9 - TCWG-390 [7/10] * Tweaked bench.py some more * Cobbled together a bare metal version of cortex-strings benchmark * Was surprised to find that the problem does not relate to Linux * Have some evidence that points at the branch predictor Meetings/mail/etc [3/10]

[ACTIVITY] June 23 - 27

=Progress= memset performance improvement - TCWG-156 [2/10] * Bugfixed/improved cortex-strings-bench-on-lava memcpy regression on A9 - TCWG-390 [6/10] * Scanned through lots of data * Learned some perf and some streamline * Still don't know what's going on here lowlevellock performance bugs - TC

[ACTIVITY] 16 - 20 June

=Progress= memcpy regression on A9 - TCWG-390 [6/10] * cortex-strings benchmarks suggest this isn't just an outlier, nor entirely about preload * Quick test on A15 suggests a similar (less severe) issue * Got cortex-strings benchmarks more-or-less running via lava memset performance improvement -

[ACTIVITY] 09 - 13 June

=Progress= memcpy regression on A9 - REQ-363 [3/10] * Verified regression on a couple of targets, some investigation around cause lowlevellock performance bugs - TCWG-435 [3/10] * Upstreamed, some reworking based on review, obsessed about odd code change & intermittent test failure memset improv

[ACTIVITY] Week 23

=Progress= lowlevellock performance bugs - TCWG-435 [1/10] * All ready to go cbuild benchmarking - TCWG-360 [4/10] * Completed a draft implementation for spec2k * Parked pending review Meetings/mail/etc [5/10] =Plan= Send lowlevellock patches to list Get back to benchmarking/improving cortex-

[ACTIVITY] Week 21

== Progress== lowlevellock performance bugs - TCWG-435 [5/10] * Tried various methods to build/test glibc for aarch64 * Eventually succeeded (tests passed) cbuild benchmarking - TCWG-360 [3/10] * cbuildized spec2xxx scripts working as far as 'run' Meetings/mail/etc [2/10] == Plan == Holiday f

Re: Using binfmt_misc for cross testing

If we add models as targets then that alleviates the sharing problem a bit - for test runs that aren't going to be too slow on models. And you'll need to watch out for nested '"' in $@, but I guess you know that. I guess this approach is similar to what 'make check test-wrapper="./glibc/script

[ACTIVITY] Week 20

== Progress== lowlevellock performance bugs - TCWG-435 [3/10] * Trying to build/test aarch64 on a foundation model cbuild benchmarking - TCWG-360 [4/10] * Integrating Maxim's spec scripts into Kugan's benchmarking branch * Began modifying the branch to use existing cbuild functions where possible

[ACTIVITY] Week 19

== Progress== Holiday [2/10] Fixing performance bugs in lowlevellock - TCWG-435 [4/10] * Initial patch for lowlevellock.h largely done, awaiting an aarch64 test run * About 1/2 of the lowlevellock.h's remain post-patch, need to ask some questions about them cortex-strings memset - TCWG-156 [1/10]

[ACTIVITY] Week 18

(Resending to correct address) == Progress== TCWG-435 needless busy-wait in lowlevellock.c (0/10) * Patches for lowlevellock.c sent to list * Patch for lowlevellock.h still to do TCWG-156 cortex-strings memset (5/10) * Dug through a bunch of docs and fiddled with the source Looking at the cbuil

[ACTIVITY] Week 17

== Progress== TCWG-435 needless busy-wait in lowlevellock.c (3/10) * Patches ready, testing a bit slow & fiddly TCWG-156 cortex-strings memset (3/10) * Worried about noisy benchmarks * Learned to use Lava * Found some helpful internal docs * Managed to tweak the code slightly == Misc == Meetin

[ACTIVITY] Week 16

== Progress == TCWG-156 cortex-strings memset (3/10) * Got a full set of benchmarks (for my 2 targets) * Cleaned up code * Sped up small memsets (for A9, A15 results pending) glibc performance bug in lowlevellock.c (1/10 - I'll make a card next week) * Learned to build and test glibc * Understo

[ACTIVITY] Week 15

== Progress == TCWG-156 cortex-strings memset (5/10) * Fixed a couple of bugs in the no-VFP case * Ran benchmarks, discarded broken benchmarks, ran more benchmarks * Explored benchmarking scripts * Took a hard look at the memset tests * Experimented with ARM-internal cortex-strings benchmark (not

[ACTIVITY] Week 14

TCWG-156 (5/10) * Hacked v7 memcpy into a memset * Much fiddling with builds, targets * Kicked off a benchmark run Misc * Meetings (1/10) * Finding hardware/setting up working environments/figuring out workflows (4/10) ___ linaro-toolchain mailing list l