February 2018 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] [Pharo 7.0-dev] Build #553: 20941-deprecate-theMetaclass-and-the-theNonMeta

by Alistair Grant

Hi Cyril, On 16 February 2018 at 15:57, Cyril Ferlicot <cyril.ferlicot(a)gmail.com> wrote: > > On what VM is this happening? > > For linux 64bits vm I get a lot a random crash during configuration > loading. But these crash vanish when using the latest Pharo 70 vm for > linux 64bits. Would you mind posting the output of "pharo --version" so I can try with exactly the same 64bit VM? Thanks! Alistair

Feb. 16, 2018

Re: [Pharo-dev] [Pharo 7.0-dev] Build #553: 20941-deprecate-theMetaclass-and-the-theNonMeta

by tesonep＠gmail.com

Hi Cyril, The problem in the bootstrap is happening in Linux 32bits. I cannot say nothing about problems in 64bits. Try creating a bigger Eden, you should run Smalltalk vm parameterAt: 45 put: (Smalltalk vm parameterAt: 44) * 4 The changes are only effected in the next loading of the image, so you should do something like: ./pharo "Pharo.image" eval --save "Smalltalk vm parameterAt: 45 put: (Smalltalk vm parameterAt: 44) * 4" And then try to load the configuration. Marcus, I am opening an issue for that. Cheers, On Fri, Feb 16, 2018 at 3:57 PM, Cyril Ferlicot <cyril.ferlicot(a)gmail.com> wrote: > On Fri, Feb 16, 2018 at 2:07 PM, tesonep(a)gmail.com <tesonep(a)gmail.com> > wrote: > > Hi, I have already checked this error. > > > > The error is random in the development branch, although in my branch is > > happening always. > > I have detected that the problem is that the GC is corrupting the the > > objects when it has to enlarge the Eden. > > > > This problem also was detected some months ago by Pavel when he tried to > > load a big baseline (Seaside in that moment). > > I have solved it in a similar way, increasing the Eden size before > loading > > BaselineOfIDE. > > > > Since this "hack" my branch is running normally (well... it has other > > problems with some tests... but that is expected). > > Maybe Clement can through more light to the matter. > > > > Hi, > > On what VM is this happening? > > For linux 64bits vm I get a lot a random crash during configuration > loading. But these crash vanish when using the latest Pharo 70 vm for > linux 64bits. > > > Cheers, > > > > > > > > -- > > Pablo Tesone. > > tesonep(a)gmail.com > > > > -- > Cyril Ferlicot > https://ferlicot.fr > > http://www.synectique.eu > 2 rue Jacques PrÃ©vert 01, > 59650 Villeneuve d'ascq France > > -- Pablo Tesone. tesonep(a)gmail.com

Feb. 16, 2018

Re: [Pharo-dev] [Pharo 7.0-dev] Build #556: 21353 Cleanup assertions in CompiledMethodTest (testCopy and testCopyWithTrailerBytes)

by Marcus Denker

> On 16 Feb 2018, at 15:40, ci-pharo-ci-jenkins2(a)inria.fr wrote: > > There is a new Pharo build available! > > The status of the build #556 was: FAILURE. > > The Pull Request #896 was integrated: "21353 Cleanup assertions in CompiledMethodTest (testCopy and testCopyWithTrailerBytes)" > Pull request url: https://github.com/pharo-project/pharo/pull/896 > > Issue Url: https://pharo.fogbugz.com/f/cases/21353 > Build Url: https://ci.inria.fr/pharo-ci-jenkins2/job/Test%20pending%20pull%20request%2… Reason: I think timeout, build took too long. Marcus

Feb. 16, 2018

Re: [Pharo-dev] [Pharo 7.0-dev] Build #553: 20941-deprecate-theMetaclass-and-the-theNonMeta

by Cyril Ferlicot

On Fri, Feb 16, 2018 at 2:07 PM, tesonep(a)gmail.com <tesonep(a)gmail.com> wrote: > Hi, I have already checked this error. > > The error is random in the development branch, although in my branch is > happening always. > I have detected that the problem is that the GC is corrupting the the > objects when it has to enlarge the Eden. > > This problem also was detected some months ago by Pavel when he tried to > load a big baseline (Seaside in that moment). > I have solved it in a similar way, increasing the Eden size before loading > BaselineOfIDE. > > Since this "hack" my branch is running normally (well... it has other > problems with some tests... but that is expected). > Maybe Clement can through more light to the matter. > Hi, On what VM is this happening? For linux 64bits vm I get a lot a random crash during configuration loading. But these crash vanish when using the latest Pharo 70 vm for linux 64bits. > Cheers, > > > > -- > Pablo Tesone. > tesonep(a)gmail.com -- Cyril Ferlicot https://ferlicot.fr http://www.synectique.eu 2 rue Jacques PrÃ©vert 01, 59650 Villeneuve d'ascq France

Feb. 16, 2018

[Pharo 7.0-dev] Build #556: 21353 Cleanup assertions in CompiledMethodTest (testCopy and testCopyWithTrailerBytes)

by ci-pharo-ci-jenkins2＠inria.fr

There is a new Pharo build available! The status of the build #556 was: FAILURE. The Pull Request #896 was integrated: "21353 Cleanup assertions in CompiledMethodTest (testCopy and testCopyWithTrailerBytes)" Pull request url: https://github.com/pharo-project/pharo/pull/896 Issue Url: https://pharo.fogbugz.com/f/cases/21353 Build Url: https://ci.inria.fr/pharo-ci-jenkins2/job/Test%20pending%20pull%20request%2…

Feb. 16, 2018

Re: [Pharo-dev] [Pharo 7.0-dev] Build #553: 20941-deprecate-theMetaclass-and-the-theNonMeta

by Marcus Denker

Could we make that the default? > On 16 Feb 2018, at 14:07, tesonep(a)gmail.com wrote: > > Hi, I have already checked this error. > > The error is random in the development branch, although in my branch is happening always. > I have detected that the problem is that the GC is corrupting the the objects when it has to enlarge the Eden. > > This problem also was detected some months ago by Pavel when he tried to load a big baseline (Seaside in that moment). > I have solved it in a similar way, increasing the Eden size before loading BaselineOfIDE. > > Since this "hack" my branch is running normally (well... it has other problems with some tests... but that is expected). > Maybe Clement can through more light to the matter. > > Cheers, > > > On Fri, Feb 16, 2018 at 12:50 PM, Marcus Denker <marcus.denker(a)inria.fr <mailto:marcus.denker@inria.fr>> wrote: > > >> On 16 Feb 2018, at 12:22, ci-pharo-ci-jenkins2(a)inria.fr <mailto:ci-pharo-ci-jenkins2@inria.fr> wrote: >> >> There is a new Pharo build available! >> >> The status of the build #553 was: FAILURE. >> >> The Pull Request #888 was integrated: "20941-deprecate-theMetaclass-and-the-theNonMeta" >> Pull request url: https://github.com/pharo-project/pharo/pull/888 <https://github.com/pharo-project/pharo/pull/888> >> >> Issue Url: https://pharo.fogbugz.com/f/cases/20941 <https://pharo.fogbugz.com/f/cases/20941> >> Build Url: https://ci.inria.fr/pharo-ci-jenkins2/job/Test%20pending%20pull%20request%2… <https://ci.inria.fr/pharo-ci-jenkins2/job/Test%20pending%20pull%20request%2…> > > FAILURE due to VM crash: > > [32] replaceFrom:to:with:startingAt: > [32] replaceFrom:to:with:startingAt: > [32] replaceFrom:to:with:startingAt: > [32] basicIdentityHash > [32] replaceFrom:to:with:startingAt: > [32] replaceFrom:to:with:startingAt: > [32] new: > [32] new: > [32] > [32] stack page bytes 4096 available headroom 2788 minimum unused headroom 156 > [32] > [32] (Segmentation fault) > [32] ./vm/pharo: line 11: 13443 Aborted "$DIR"/"pharo-vm/pharo" --nodisplay "$@" > > > > > -- > Pablo Tesone. > tesonep(a)gmail.com <mailto:tesonep@gmail.com>

Feb. 16, 2018

[Pharo 7.0-dev] Build #555: 21351 Remove unnecessary "self run:" in BagTest (last remaining)

by ci-pharo-ci-jenkins2＠inria.fr

There is a new Pharo build available! The status of the build #555 was: SUCCESS. The Pull Request #895 was integrated: "21351 Remove unnecessary "self run:" in BagTest (last remaining)" Pull request url: https://github.com/pharo-project/pharo/pull/895 Issue Url: https://pharo.fogbugz.com/f/cases/21351 Build Url: https://ci.inria.fr/pharo-ci-jenkins2/job/Test%20pending%20pull%20request%2…

Feb. 16, 2018

[Pharo7 ci] reduced # executors

by Marcus Denker

Hi, For Pharo Builds (Both CI Pull Request checks as well as builds after merges) things will be a bit slow as I reduces for all slaves the number of executors from 4 to 1. Reason: Experiment to see if this leads to more stable performance / less problems Marcus

Feb. 16, 2018

Re: [Pharo-dev] [Pharo 7.0-dev] Build #553: 20941-deprecate-theMetaclass-and-the-theNonMeta

by tesonep＠gmail.com

Hi, I have already checked this error. The error is random in the development branch, although in my branch is happening always. I have detected that the problem is that the GC is corrupting the the objects when it has to enlarge the Eden. This problem also was detected some months ago by Pavel when he tried to load a big baseline (Seaside in that moment). I have solved it in a similar way, increasing the Eden size before loading BaselineOfIDE. Since this "hack" my branch is running normally (well... it has other problems with some tests... but that is expected). Maybe Clement can through more light to the matter. Cheers, On Fri, Feb 16, 2018 at 12:50 PM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > > > On 16 Feb 2018, at 12:22, ci-pharo-ci-jenkins2(a)inria.fr wrote: > > There is a new Pharo build available! > > The status of the build #553 was: FAILURE. > > The Pull Request #888 was integrated: "20941-deprecate-theMetaclass- > and-the-theNonMeta" > Pull request url: https://github.com/pharo-project/pharo/pull/888 > > Issue Url: https://pharo.fogbugz.com/f/cases/20941 > Build Url: https://ci.inria.fr/pharo-ci-jenkins2/job/Test%20pending% > 20pull%20request%20and%20branch%20Pipeline/job/development/553/ > > > FAILURE due to VM crash: > > [32] replaceFrom:to:with:startingAt: > [32] replaceFrom:to:with:startingAt: > [32] replaceFrom:to:with:startingAt: > [32] basicIdentityHash > [32] replaceFrom:to:with:startingAt: > [32] replaceFrom:to:with:startingAt: > [32] new: > [32] new: > [32] > [32] stack page bytes 4096 available headroom 2788 minimum unused headroom 156 > [32] > [32] (Segmentation fault) > [32] ./vm/pharo: line 11: 13443 Aborted "$DIR"/"pharo-vm/pharo" --nodisplay "$@" > > > -- Pablo Tesone. tesonep(a)gmail.com

Feb. 16, 2018

Re: [Pharo-dev] VirtualMachine vs. Platform and Bootstrap

by Henrik Sperre Johansen

It's a question of responsibilities. Right now, it's used to determine which dll's to use on linux (on linux, see ffiModuleName), which means the distinction between 32/64 platforms is which ABI to use when doing UFFI, or more accurately, where to find libraries for the ABI supported by the vm. To make things less confusing, it would be better to remove the 32/64 subclasses from platform, (since they are really a separation only relevant to UFFI) and have the ffiModuleName: extension branch on Smalltalk vm is32/64 as appropriate for UnixPlatform. Then one would be free to implement either 32/64 subclasses as you suggest if there's prim support, or a #getDetailedPlatformInfo dependent on FFI to fetch more specific details like 32/64 bit, actual windows version, etc.. -- Sent from: http://forum.world.st/Pharo-Smalltalk-Developers-f1294837.html

Feb. 16, 2018