Vm-dev July 2016

vm-dev@lists.squeakfoundation.org

34 participants
214 discussions

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

Amazing ARM simulator experience
by Eliot Miranda 03 May '17

03 May '17

Hi All, just had to tell people about this morning's experience using the ARM simulator. I've been building Smalltalk VMs since 1983, so 33 years. My first on the Three Rivers PERQ was dog slow. My undergraduate project was done the the National Semiconductor 32016 based Whitechapel Computer Works workstation and its and 32032-based successor. This morning I was revamping register management in Cog's ARMv5/v6 back end, making more registers available by using two of the C argument registers for two of registers the Cogit assigns to various fixed tasks, and using store and load multiple instructions to save and restore concisely the registers around calls into the runtime. Remember the architecture here. The simulator generates ARM machine code into the ByteArray that re[resents the address space, holding all of generated machine code, a small C stack, and the Smalltalk heap. A plugin, derived from Gdb's ARM simulator written in C, interprets that machine code, running for a couple of milliseconds at a time in a loop, applying breakpoint calculations, and asserts on every call into the run-time, and that calls into the run-time and accesses of variables in the simulator is done by using illegal addresses in the generated machine code. Each illegal access causes the Gdb-derived machine code interpreter to return with an error, this error is turned into an exception, the handler for which maps the specific illegal address into a variable or message selector, accesses the variable or activates the message selector, providing the result, and allowing the execution to continue. In changing the register management I had a test case that worked, an image that prompted for an expression ands evaluated it, which worked both in the simulator and with the generated VM. But the real VM, crashed when used on a proper image. So to debug I started launching the real interactive image in the simulator. Well, the amazing experience is that that image, whose machine code is being _interpreted by a C program_ feels /faster/ than my 32016 based implementation back in 1984-ish. Quite amazing. I can open windows, type text, access source code (was playing with Message Names) etc. It's sluggish, but usable. Amazing how fast modern machines are. All on my 2012 vintage 2.2 Ghz Core i7 MacBook Pro. I'm blown away :-) _,,,^..^,,,_ best, Eliot

5 7

error "please insert disk" (windows vm)
by Nicolai Hess 14 Jan '17

14 Jan '17

we have the following bug report 16928 <https://pharo.fogbugz.com/f/cases/16928/no-disk-in-the-drive-reply-to-World…> "no disk in the drive" reply to World | Save As (and other places) "On a clean Pharo4.0 #40618 image, on Windows 7 World Menu | Save As brings up a modal alert window, "There is no disk in the drive. Please insert a disk into drive <etc>" On my system, it does this 4 times. I'm assuming that each instance refers to one of the card-reader drives which are attached, but without a card inserted. This is a pain to deal with every single time a File Browser or World | Save As dialog is opened. Could the alerts either be - only raised when that specific drive is requested by a user action or, less preferably - given by a non-modal window, and fading away after a period" I was able to reproduce this error on squeaks and pharos window vm. (insert a usb card reader, with card, wait some time , remove only the card, every access on this drive, for example open a FileList and scroll to the drive letter). The solution I found is to call SetErrorMode( *SEM_FAILCRITICALERRORS)* I would like to propose the following change to sqwin32directory.c wrap the call to FindFirstFileW/FindNextFileW with UINT prevMode = SetErrorMode(SEM_FAILCRITICALERRORS); call FindFirstFileW/FindNextFileW SetErrorMode(prevMode); ... and call this functions in at least dir_lookup() (the other file/directory methods are save). Alternative solution, we could call the SetErrorMode function once at program start but I don't know if there are other "useful" errors message that we don't want to disable. I tested both solution, both seems to work fine. What do you think?

1 3

CogSpur crash on OS X when resizing with Moom
by Bernhard Pieber 09 Dec '16

09 Dec '16

Dear VM developers, I can reliably crash the latest stable CogSpur.r3732.app using the Squeak5.1-15113.image when I resize it with Moom (http://manytricks.com/moom/) right after starting the image. I tried to reproduce it by resizing it manually without Moom. I *think* it happened occasionally when resizing manually, but I could not reproduce it today. Versions up to r3427 do not crash. The crash happens starting with version r3602. With the latest unstable version 201606301459 built by Travis the crash still happens. I attach the crash dump for version r3732. Maybe it is helpful. Thanks for all your work! Cheers, Bernhard

3 4

New case to crash Spur VM
by Denis Kudriashov 09 Dec '16

09 Dec '16

Hi In debugger (Squeak or Pharo) try "step over" for expression "Object basicSize": image will crash. We have opened issue for this in Pharo tracker 17962 <https://pharo.fogbugz.com/f/cases/17962/Image-is-crashed-when-step-over-Obj…>

5 12

Speedcenter setup
by Tim Felgentreff 10 Oct '16

10 Oct '16

Hi, I finally got around to enabling automatic nightly benchmarking for all branches for both opensmalltalk-vm and rsqueak. Before I had to manually trigger benchmarks for opensmalltalk-vm. Now a nightly cron job checks for new passed builds on all branches and then tries to run benchmarks for the resulting binaries. Note, however, that we're only uploading binaries for the master and Cog branch right now on opensmalltalk-vm, so all other branches are schedules but then skipped, because no binaries are available. The benchmarking machine is still running at this point, a full set of benchmarks takes about 4-5 hours per binary. So I expect the results for yesterdays VMs to come in this afternoon GMT+1. Not all of the benchmarks are well scaled right now, so that might need some tweaking, but since there has been talk about various performance optimizations recently, I figured we should be more vigilant in tracking the results of that. For example, Clement mentioned recent optimizations that should have results in binarytrees. It's nice if we can see that in the timeline. This page in particular may be interesting to check once in a while: http://speed.squeak.org/timeline/#/?exe=1,5&ben=grid&env=2&revs=50&equid=on If anyone wants to take a look at the code and benchmarks, the about page links to all relevant repositories. I'm happy to give people access so they can add or change or remove benchmarks. cheers, Tim

6 15

Script needed; git mavens can you help?
by Eliot Miranda 27 Sep '16

27 Sep '16

Hi git mavens, I've used a script called svnrevert in the svn version that reverts any changed file in its argument(s), which may be directories, /and/ sets the modification date to the date at which the file was last checked in. I'd like the same functionality in git. Her'es the subversion version: #!/bin/bash # Revert file(s) or all modified files in a directory # and touch them back to the checkin date for f in "$@" do if [ -d "$f" ]; then $0 `svn st "$f" | grep "^M" | sed 's/^M *//'` else if svn revert "$f"; then changed="`svn info \"$f\" | grep 'Last Changed Date:' | sed 's/ *(.*//'`" touch -t "`date -j -f 'Last Changed Date: %Y-%m-%d %H:%M:%S %z' \"$changed\" '+%Y%m%d%H%M.%S'`" "$f" fi fi done _,,,^..^,,,_ best, Eliot

5 5

Cog + Pi + OSProcess
by Henrik Johansen 19 Aug '16

19 Aug '16

On the Cog VM shipped with latest Raspbian (3553, plugins 3543) running on a Pi3, after loading OSProcess, evaluating (PipeableOSProcess command: '/bin/echo this is a test') output segfaults the VM, crash.dmp indicating UnixOSProcessAccessor >> #forkAndExec: <snip> The shipped interpreter vm (4.10.2 platform rev2614) works with OSProcess, so I can use that instead, but has anyone encountered this already and happen to have a plugin built for Cog/Pi that allows command: use?. Cheers, Henry PS. Doing the layman's hack (copying over and renaming so.AioPlugin/so.UnixOSProcessPlugin from interpreter vm) does not work very well, no more crashes, but the PipeableOSProcess created has pid nil, status notYetRunning and an empty string output, so I assume a Cog-specific variant is needed.

8 28

Object basicNew vs object shallowCopy
by Denis Kudriashov 17 Aug '16

17 Aug '16

Hi. I compared performance between object instantiation and object cloning. I was wondering that instantiation almost twice faster than clone (primitive 70 vs 148). Could you explain why it like that and could it be improved? I was think that new object construction is much complex because it requires to fill all object fields (header structure and etc). And I was think that copy is just simple function like memcpy which just copy bytes without any logic. Here is my code: object := Object new. 3 timesRepeat: [ Smalltalk garbageCollect ]. result1 := [ Object basicNew ] benchFor: 10 seconds. 3 timesRepeat: [ Smalltalk garbageCollect ]. result2 := [ object shallowCopy ] benchFor: 10 seconds. {result1. result2}. "an Array(a BenchmarkResult(518,021,045 iterations in 10 seconds 2 milliseconds. 51,791,746 per second) a BenchmarkResult(302,807,253 iterations in 10 seconds 4 milliseconds. 30,268,618 per second))" (I run it on latest Pharo on Mac SpurVM) Best regards, Denis

5 9

ffi-test-main.c fails on FreeBSD
by Gary Dunn 12 Aug '16

12 Aug '16

In conjunction with getting Open Cobalt running on FreeBSD I have been looking into the problems with SqueakFFIPrims on FreeBSD and have reached a point that is beyond my understanding. I am not sure if the fix should come from the VM side or the FreeBSD port maintainer. I needed to get source code back into the ports tree so I did a make in /usr/ports/lang/squeak. I already had the 3.9 tarball, so that just repopulated the source and did the initial build. In the "files" folder I found patch files typical for a FreeBSD port. These patch source files from a generic tarball to work with FreeBSD. The one of interest was patch-platforms__unix__plugins__SqueakFFIPrims__ffi-config which contains ----x-----x----- --- platforms/unix/plugins/SqueakFFIPrims/ffi-config.org Wed Apr 26 20:27:53 2006 +++ platforms/unix/plugins/SqueakFFIPrims/ffi-config Wed Apr 26 20:29:00 2006 @@ -39,6 +39,7 @@ case ${abi} in linux) abi=sysv;; + freebsd*) abi=sysv;; darwin*) abi=darwin;; *) abi=libffi; lib="-lffi";; esac -----x-----x----- I then had the folder work/Squeak-3.9-7/platforms/unix/plugins/SqueakFFIPrims In it is a file 00README which explains how to test FFI. It says to run ffi-test-config but I have no such file. It says to type "make" to build a test suite but the build failed. It says to try with make CPU=any ABI=libffi LIB=-lffi which also failed. I traced the problem to the gcc search paths for includes and libraries -- turns out FreeBSD only searches /usr/include and /usr/lib when the required files are in /usr/local/include and /usr/local/lib. After some fiddling I finally had a "main" to test. It failed: ffi-test-main.c failed at line 361. Here is a bit from that file, ending with line 361: ffiInitialize(); for (ul= 0; ul < 15; ++ul) ffiPushSingleFloat(fa[ul]); GO(FFITypeSingleFloat, many); f= ffiReturnFloatValue(); ffiCleanup(); CHECK(f - ff < FLT_EPSILON); I worked out the CHECK macro but could not find what FLT_EPSILON is. At that point I decided to stop and come to you for help. Where do I go from here? -- Gary Dunn, Honolulu osp(a)aloha.com http://openslate.net/ http://e9erust.blogspot.com/ Sent from Slate001

5 8

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 July 2016