Vm-dev December 2019

vm-dev@lists.squeakfoundation.org

31 participants
265 discussions

[OpenSmalltalk/opensmalltalk-vm] Fix checkSCCSversion invocation in mvm scripts of linux build subdirectories (#405)
by Tim Johnson 20 Oct '20

20 Oct '20

Currently the `mvm` scripts within Linux build directories are failing to properly invoke `checkSCCSversion` because their those directories are nested three levels deep rather than two (as in other platforms). To fix, update all `mvm` scripts within three-level nested linux build directories to invoke `../../../scripts/checkSCCSversion` rather than `../../scripts/checkSCCSversion` I believe this can be accomplished by executing the following one-liner from the repo root: ``` for f in $(find . -name mvm -exec grep -q checkSCCS {} \; -print | awk -F/ 'NF == 5 { print $0 }') ; do printf "g/checkSCCSversion/s/\.\.\//&\.\.\//\nw\nq" | ed ${f} ; done ``` -- 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/405

2 1

[OpenSmalltalk/opensmalltalk-vm] primitiveFileSize fails on Stdio handles (#430)
by Fabio Niephaus 20 Oct '20

20 Oct '20

`FileStream stdout closed` returns `true` in a currently Squeak trunk image. The reason for this is that the `primitiveFileSize` fails. To be fair, it makes sense as `stdout` does not really have a size. Nonetheless, it does not make sense that Squeak thinks `stdout` is closed, especially when it really isn't. On SqueakJS, for example, the primitive returns `0` which lets `FileStream stdout closed` return `false`. I'm not sure where to fix this, probably in the `FilePlugin` code of VMMaker? Could someone please take a look? Thanks! See also: http://forum.world.st/FileStream-stdout-closed-td5105395.html -- 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/430

2 2

[OpenSmalltalk/opensmalltalk-vm] Customization for Pharo & About Dialog (#388)
by Pablo Tesone 19 Oct '20

19 Oct '20

- We added customization for Pharo to allow a different menu to be used. - We modified the default action of help menu to open an URL (only for Pharo) - We added a Info.plist property with the string value of the URL (only for Pharo) - And also the about dialog was not receiving the close event, we have fixed that. It was related with the filtering of events for SDL. @demarey You can view, comment on, or merge this pull request online at: https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/388 -- Commit Summary -- * Adding a custom Menu for Pharo VMs * Adding a optional URL for the help, only fr Pharo * Handling correctly the events of the about window * Making the about window property as weak. -- File Changes -- M build.macos32x86/common/Makefile.app (8) M build.macos32x86/pharo.cog.spur.lowcode/Makefile (3) M build.macos32x86/pharo.cog.spur.minheadless/Makefile (2) M build.macos32x86/pharo.cog.spur/Makefile (1) M build.macos32x86/pharo.cog.spur/plugins.ext (2) M build.macos32x86/pharo.cog.v3/Makefile (2) M build.macos32x86/pharo.sista.spur/Makefile (2) M build.macos32x86/pharo.stack.spur.lowcode/Makefile (2) M build.macos32x86/pharo.stack.spur/Makefile (2) M build.macos64x64/common/Makefile.app (8) M build.macos64x64/pharo.cog.spur.lowcode/Makefile (1) M build.macos64x64/pharo.cog.spur/Makefile (1) M build.macos64x64/pharo.sista.spur/Makefile (1) M build.macos64x64/pharo.stack.spur.lowcode/Makefile (1) M build.macos64x64/pharo.stack.spur/Makefile (1) A platforms/iOS/vm/English.lproj/Pharo-MainMenu-opengl.xib (1261) A platforms/iOS/vm/English.lproj/Pharo-MainMenu.xib (1261) M platforms/iOS/vm/OSX/Pharo-Info.plist (2) M platforms/iOS/vm/OSX/SqueakOSXApplication.m (30) M platforms/iOS/vm/OSX/sqSqueakOSXApplication+events.m (16) M platforms/iOS/vm/OSX/sqSqueakOSXApplication.h (5) M platforms/iOS/vm/OSX/sqSqueakOSXApplication.m (1) -- Patch Links -- https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/388.patch https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/388.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/388

7 12

[OpenSmalltalk/opensmalltalk-vm] DO NOT INTEGRATE - FOR DISCUSSION ONLY Minimalistic headless x64 msvc2017 (#312)
by Ben Coman 11 Sep '20

11 Sep '20

Ronie submitted a pull request that fixes the build with cygwin for win32 ( https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/311 ) and I'm curious if any of my changes here are still applicable. @ronsaldo could you comment on these... You can view, comment on, or merge this pull request online at: https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/312 -- Commit Summary -- * Fix function call declarations and return types (x64-msvc2017) * Fix unresolved external symbol 'alloca' (x64-msvc2017) * Workaround _controlfp() does not support _MCW_PC or _MCW_IC on x64. Expert review required. * Add architecture define _M_IX86 to ABI (x86-msvc2017) * Workaround hang allocating initial memory (x64-msvc2017). Expert review required. * Better 32bit CMakeLists.txt for SqueakFFIPrims (x86-msvc2017) -- File Changes -- M CMakeLists.txt (43) M platforms/Cross/plugins/IA32ABI/ia32abicc.c (2) M platforms/Cross/plugins/IA32ABI/x64win64abicc.c (3) M platforms/Cross/plugins/IA32ABI/xabicc.c (2) M platforms/minheadless/common/sqMain.c (1) M platforms/minheadless/windows/sqPlatformSpecific-Win32.c (9) M platforms/minheadless/windows/sqPlatformSpecific-Win32.h (2) M platforms/minheadless/windows/sqWin32Main.c (1) M platforms/minheadless/windows/sqWin32SpurAlloc.c (9) -- Patch Links -- https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/312.patch https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/312.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/312

5 12

[OpenSmalltalk/opensmalltalk-vm] MUSL deltas (mostly harmless) (#450)
by Ken Dickey 10 Sep '20

10 Sep '20

I have built and tested aarch64 builds on Raspbery Pi 3B / Arch Linux / MUSL+Busybox Chromebook Two / Linux / libc The mvm "-O1" seems to work in both cases -- no need to go to "-O0" Please ignore the spurious sqSCCSVersion.h -- I don't know how to elide this. The getwd() -> getcwd() in sqUnixSecurityPlugin.c should likewise be harmless [no getwd() in MUSL] Not a time concern. I don't know of anyone asking for this. You can view, comment on, or merge this pull request online at: https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/450 -- Commit Summary -- * MUSL for AlpineLinux * MUSL for AlpineLinux -- File Changes -- M build.linux64ARMv8/squeak.stack.spur/build.debug/mvm (7) M build.linux64ARMv8/squeak.stack.spur/build/mvm (10) M platforms/Cross/plugins/sqPluginsSCCSVersion.h (8) M platforms/Cross/vm/sqSCCSVersion.h (14) M platforms/Cross/vm/sqVirtualMachine.c (5) M platforms/unix/plugins/SecurityPlugin/sqUnixSecurity.c (3) M platforms/unix/vm/sqUnixMain.c (5) -- Patch Links -- https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/450.patch https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/450.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/450

6 28

[OpenSmalltalk/opensmalltalk-vm] Be explicit about which test image version - move to CI config (#331)
by Ben Coman 10 Sep '20

10 Sep '20

First try to see what Travis thinks about these changes. You can view, comment on, or merge this pull request online at: https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/331 -- Commit Summary -- * Be explicit about which test image version - move to CI config -- File Changes -- M .travis.yml (91) M scripts/ci/travis_test.sh (6) M tests/smalltalkCI.sh (25) -- Patch Links -- https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/331.patch https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/331.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/331

3 4

[OpenSmalltalk/opensmalltalk-vm] Simplify gitignore build directories [skip ci] (#351)
by Ben Coman 10 Sep '20

10 Sep '20

I noticed that modified Makefiles weren't being pickup up by `git status`. It was too tedious trying to fix the existing definitions, so I've condensed them down to 4 lines from 34. I'm reasonably sure the result is the same, but if I've missed something its low impact, so maybe just integrate first to let everyone suck-n-see. Doing a `git status` before each `git commit` will let you confirm that its working. You can view, comment on, or merge this pull request online at: https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/351 -- Commit Summary -- * Simplify gitignore build directories -- File Changes -- M .gitignore (43) -- Patch Links -- https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/351.patch https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/351.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/351

2 2

[OpenSmalltalk/opensmalltalk-vm] third-party: Stop building/using vulnerable software (#386)
by Holger Freyther 10 Sep '20

10 Sep '20

You can view, comment on, or merge this pull request online at: https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/386 -- Commit Summary -- * third-party: Stop building/using vulnerable software -- File Changes -- M third-party/libgit2.spec (10) M third-party/libpng.spec (6) M third-party/libssh2.spec (6) M third-party/openssl.spec (6) -- Patch Links -- https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/386.patch https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/386.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/386

7 13

[OpenSmalltalk/opensmalltalk-vm] fix libgit2 makefile (#390)
by maenu 10 Sep '20

10 Sep '20

I am using MacPort's CMake which works fine, except that it needs a path argument for the libgit2 makefile. Can somebody check if this is compatible with their CMakes? You can view, comment on, or merge this pull request online at: https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/390 -- Commit Summary -- * fix libgit2 makefile -- File Changes -- M build.macos64x64/third-party/Makefile.libgit2 (2) -- Patch Links -- https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/390.patch https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/390.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/390

2 2

[OpenSmalltalk/opensmalltalk-vm] primitives 54 and 554 for timesTwoPower: answer incorrect result on 64bits macOS VM (#383)
by Nicolas Cellier 20 Jul '20

20 Jul '20

This is due to a macOS bug in 64bits libm! (Hard to believe, isn't it?) I've opened a ticket https://bugreport.apple.com/web/?problemID=48021471 > Area: > Something not on this list > > Summary: ldexp incorrectly rounds gradual underflow for some specific values. > > Steps to Reproduce: > #include <math.h> > #include <stdio.h> > int main() { > int exp=-54; double y=ldexp(11.0,exp); /* y is binary 1.011*2^-51 */ > double u=ldexp(1.0,-1074); /* u is the minimal denormalized IEEE754 double */ > double v=ldexp(y,-1023); /* v is binary 1.011*2^-1074 and should round to u */ > printf("u=%g v=%g\n",u,v); > return 0; > } > > Expected Results: > v should be rounded to 1.0*2^-1074 (round to nearest, tie to even default rule) > Thus we should have u == v. > > Actual Results: > v is rounded upward to binary 10.0*2^-1074 = 1.0*2^-1073 > output is u=4.94066e-324 v=9.88131e-324 > > Note 1: this fails for 4 different values of exp -54,-53,+968,+969 > (replace exponent -1023 by -1074-3-exp, that is -1023,-1024,-2045,-2046) > Note 2: this did not happen previously with 32bits libm version. > Note 3: this sounds a bit like this bug https://stackoverflow.com/questions/32150888/should-ldexp-round-correctly > > Version/Build: > macOS HighSierra 10.13.16 > > Configuration: > compiled with > clang --version > Apple LLVM version 10.0.0 (clang-1000.11.45.5) > Target: x86_64-apple-darwin17.7.0 > Thread model: posix > InstalledDir: /Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin > Xcode Version 10.1 (10B61) > > -- 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/383

5 6

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 December 2019