Vm-dev February 2018

vm-dev@lists.squeakfoundation.org

27 participants
131 discussions

[OpenSmalltalk/opensmalltalk-vm] libgit build fails because of libssl and libcrypto (#210)
by Javier Pimas 03 Dec '19

03 Dec '19

on /build.linux32x86/pharo.cog.spur/build I run ./mvm and get make[3]: *** No rule to make target '/usr/lib/i386-linux-gnu/libssl.so', needed by 'libgit2.so.0.25.1'. Stop. I worked around it by running sudo ln -s /lib/i386-linux-gnu/libssl.so.1.0.0 /usr/lib/i386-linux-gnu/libssl.so but shouldn't be necessary as libssl is downloaded and built by the script. The same happens with libcrypto -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/OpenSmalltalk/opensmalltalk-vm/issues/210

3 4

[OpenSmalltalk/opensmalltalk-vm] Make updateSCCSVersions work in more situations (#213)
by Jakob Reschke 29 Dec '18

29 Dec '18

- allow spaces in paths (added quotes) - support cases where the repository is not in a .git directory at the top of the working tree - do not hardcode the directory level of this script - avoid stashing the whole working tree when only two well-known files need to be updated - touch should suffice to make a file different from its index version (i. e., eligible for checkout) Please review for compatibility and pick those parts of the changes that you like. You can view, comment on, or merge this pull request online at: https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/213 -- Commit Summary -- * make updateSCCSVersions work in more situations -- File Changes -- M scripts/updateSCCSVersions (26) -- Patch Links -- https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/213.patch https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/213.diff -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/213

5 9

[OpenSmalltalk/opensmalltalk-vm] HowToBuild doesn't have to mention missing libssl (#186)
by Nicolas Cellier 18 Oct '18

18 Oct '18

Because we don't need to link against libssl. Keep the hint from Eliot for an hypothetical libfoobar.dylib missing. You can view, comment on, or merge this pull request online at: https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/186 -- Commit Summary -- * HowToBuild doesn't have to mention missing libssl -- File Changes -- M build.macos32x86/HowToBuild (31) -- Patch Links -- https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/186.patch https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/186.diff -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/186

3 5

[OpenSmalltalk/opensmalltalk-vm] Boot images produced by 201801252341 VM don't bring up a window on launch (#196)
by Ryan Macnak 13 May '18

13 May '18

If I produce Newspeak [boot images](https://bitbucket.org/newspeaklanguage/nsboot) from the [201801252341 Mac VMs](https://github.com/OpenSmalltalk/opensmalltalk-vm/releases/tag/2018012…, these images do not bring up the UI when subsequently opened, either by the new VMs or old VMs. The new VMs are able to properly bring up images produced by old VMs. (Maybe there's some new headlessness property in the image header?) -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/OpenSmalltalk/opensmalltalk-vm/issues/196

3 6

StackInterp/Cog>>transferTo: - assertValidExecutionPointe:r:s differences
by Ben Coman 06 May '18

06 May '18

Just curious about the difference in instructionPointer between these two... StackInterpreter>>transferTo: self assertValidExecutionPointe: instructionPointer + 1 r: framePointer s: stackPointer. CoIntepreter>>transferTo:from: self assertValidExecutionPointe: instructionPointer r: framePointer s: stackPointer. cheers -ben

3 8

VM Maker: VMMaker.oscog-eem.2338.mcz
by commits＠source.squeak.org 14 Apr '18

14 Apr '18

Eliot Miranda uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-eem.2338.mcz ==================== Summary ==================== Name: VMMaker.oscog-eem.2338 Author: eem Time: 22 February 2018, 10:25:17.794166 am UUID: 1672c678-41b4-41f2-a061-e201562fc75c Ancestors: VMMaker.oscog-eem.2337 Fix a slip in primitiveDecompressFromByteArray caught by IncludedMethodsTests. Thank you David! =============== Diff against VMMaker.oscog-eem.2337 =============== Item was changed: ----- Method: MiscPrimitivePlugin>>primitiveDecompressFromByteArray (in category 'primitives') ----- primitiveDecompressFromByteArray "Bitmap decompress: bm fromByteArray: ba at: index" <export: true> | bm ba index i anInt code data end k n pastEnd | <var: 'ba' type: #'unsigned char *'> <var: 'bm' type: #'int *'> <var: 'anInt' type: #'unsigned int'> <var: 'code' type: #'unsigned int'> <var: 'data' type: #'unsigned int'> <var: 'n' type: #'unsigned int'> bm := self cCode: [interpreterProxy arrayValueOf: (interpreterProxy stackValue: 2)] inSmalltalk: [interpreterProxy cCoerce: (interpreterProxy arrayValueOf: (interpreterProxy stackValue: 2)) to: #'int *']. (interpreterProxy isOopImmutable: (interpreterProxy stackValue: 2)) ifTrue: [^interpreterProxy primitiveFailFor: PrimErrNoModification]. (interpreterProxy isBytes: (interpreterProxy stackValue: 1)) ifFalse: [^interpreterProxy primitiveFail]. ba := interpreterProxy firstIndexableField: (interpreterProxy stackValue: 1). index := interpreterProxy stackIntegerValue: 0. interpreterProxy failed ifTrue: [^nil]. i := index - 1. k := 0. end := interpreterProxy sizeOfSTArrayFromCPrimitive: ba. pastEnd := interpreterProxy sizeOfSTArrayFromCPrimitive: bm. [i < end] whileTrue: [anInt := ba at: i. i := i + 1. anInt <= 223 ifFalse: [anInt <= 254 ifTrue: [anInt := anInt - 224 * 256 + (ba at: i). i := i + 1] ifFalse: [anInt := 0. 1 to: 4 by: 1 do: [ :j | anInt := (anInt bitShift: 8) + (ba at: i). i := i + 1]]]. n := anInt >> 2. + k + n > pastEnd ifTrue: [^interpreterProxy primitiveFail]. - k + n >= pastEnd ifTrue: [^interpreterProxy primitiveFail]. code := anInt bitAnd: 3. "code = 0 ifTrue: [nil]." code = 1 ifTrue: [data := ba at: i. i := i + 1. data := data bitOr: (data bitShift: 8). data := data bitOr: (data bitShift: 16). 1 to: n do: [ :j | bm at: k put: data. k := k + 1]]. code = 2 ifTrue: [data := 0. 1 to: 4 do: [ :j | data := (data bitShift: 8) bitOr: (ba at: i). i := i + 1]. 1 to: n do: [ :j | bm at: k put: data. k := k + 1]]. code = 3 ifTrue: [1 to: n do: [ :m | data := 0. 1 to: 4 do: [ :j | data := (data bitShift: 8) bitOr: (ba at: i). i := i + 1]. bm at: k put: data. k := k + 1]]]. interpreterProxy pop: interpreterProxy methodArgumentCount!

3 6

[OpenSmalltalk/opensmalltalk-vm] Readonly primitives are missing in ARM VM (#216)
by Denis Kudriashov 06 Apr '18

06 Apr '18

-- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/OpenSmalltalk/opensmalltalk-vm/issues/216

5 13

Alien: separate core, tests and win32 into own packages
by Torsten Bergmann 11 Mar '18

11 Mar '18

Hi Eliot, (CC'ing vm-dev) hope you had a good start into 2018 and sorry again if the "Alien commit into Alien" [1] created some trouble. But while we are at discussing about Alien I would like to ask one question: can we split the Alien package into two packages instead of just one. One with the core functionality and and have the unit tests separated into an own separate package? Also the Win32 struct wrappers should be in an own package as this is platform dependent. Because with the new possibility to bootstrap smaller images one can build minimal images with just a foreign interface/Alien but where SUnit and tests are not loaded. Without having the tests in an own separately loadable package such a thing is not possible to realize because loading Alien would give automatically a dependency to SUnit. Also Alien-Win32 do not make sense if one builds a platform specific image for Mac, Unix or the Pi. So can we split it to be more clean? What do you think about this? I can spend some time (possibly next week) on separating Alien into two packages and also do/help/check in changing loading configs for VMMaker or other. For the naming: while in Pharo the packages could already be named - "Alien-Core" - "Alien-Core-Test" - "Alien-Win32" as Pharo already supports explicit packages. But I'm pretty sure this would be problematic for Squeak 6.0 which still relies on the old category solution. So I would suggest to follow Seaside naming convention and name them - "Alien-Core" - "Alien-Tests-Core" - "Alien-Win32" so it is loadable and compatible in all dialects. If you would prefer other names we can align easily. Thanks for an answer in advance! Bye T. [1] http://lists.squeakfoundation.org/pipermail/vm-dev/2018-January/026350.html

3 5

[OpenSmalltalk/opensmalltalk-vm] isReadOnlyObject should return false when writebarrier is not supported (#217)
by Denis Kudriashov 08 Mar '18

08 Mar '18

Object is definitely not readonly when VM do not support write barrier. But right now it will be primitive failure. For example it makes impossible to modify classes in Pharo 7 on ARM VM. Pharo 7 checks instances for readonly before migrating them -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/OpenSmalltalk/opensmalltalk-vm/issues/217

2 3

[OpenSmalltalk/opensmalltalk-vm] Added pharo.sista.spur builds (#219)
by Clement Bera 08 Mar '18

08 Mar '18

I checked the Mac 32 builds were working fine. I did not check Linux & win. Is there something to change on the CI to make them build ? You can view, comment on, or merge this pull request online at: https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/219 -- Commit Summary -- * added pharo.sista.spur builds -- File Changes -- A build.linux32x86/pharo.sista.spur/build.assert.itimerheartbeat/mvm (46) A build.linux32x86/pharo.sista.spur/build.assert/mvm (46) A build.linux32x86/pharo.sista.spur/build.debug.itimerheartbeat/mvm (49) A build.linux32x86/pharo.sista.spur/build.debug/mvm (49) A build.linux32x86/pharo.sista.spur/build.itimerheartbeat/mvm (54) A build.linux32x86/pharo.sista.spur/build/mvm (54) A build.linux32x86/pharo.sista.spur/makeallclean (15) A build.linux32x86/pharo.sista.spur/makealldirty (15) A build.linux32x86/pharo.sista.spur/plugins.ext (15) A build.linux32x86/pharo.sista.spur/plugins.int (36) A build.linux64x64/pharo.sista.spur/NotYetImplemented (0) A build.linux64x64/pharo.sista.spur/makeallclean (15) A build.linux64x64/pharo.sista.spur/makealldirty (15) A build.macos32x86/pharo.sista.spur/Makefile (31) A build.macos32x86/pharo.sista.spur/mvm (32) A build.macos32x86/pharo.sista.spur/plugins.ext (28) A build.macos32x86/pharo.sista.spur/plugins.int (31) A build.macos64x64/pharo.sista.spur/Makefile (29) A build.macos64x64/pharo.sista.spur/mvm (32) A build.macos64x64/pharo.sista.spur/plugins.ext (27) A build.macos64x64/pharo.sista.spur/plugins.int (30) A build.win32x86/pharo.sista.spur/Makefile (38) A build.win32x86/pharo.sista.spur/Pharo.def.in (3) A build.win32x86/pharo.sista.spur/Pharo.exe.manifest (30) A build.win32x86/pharo.sista.spur/Pharo.ico (0) A build.win32x86/pharo.sista.spur/Pharo.rc (32) A build.win32x86/pharo.sista.spur/mvm (33) A build.win32x86/pharo.sista.spur/plugins.ext (9) A build.win32x86/pharo.sista.spur/plugins.int (40) -- Patch Links -- https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/219.patch https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/219.diff -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/219

2 10

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Vm-dev February 2018