Vm-dev February 2014

vm-dev@lists.squeakfoundation.org

33 participants
83 discussions

Re: [Pharo-dev] Simulator in Pharo 3
by Eliot Miranda 15 Feb '14

15 Feb '14

Hi Stefan! On Wed, Feb 12, 2014 at 12:35 PM, Stefan Marr <smalltalk(a)stefan-marr.de>wrote: > Hi: > > Below you see the result of today, the simulator for the StackInterpreter > running in a Pharo 3 image. > Fab! This is great. Over the last years, people have been tinkering with the simulator from > time to time, but I have the feeling everyone gave up before the thing was > fully functioning, and not even the trivial things have found their way > back into the VM code base. > How can anyone do anything significant with the VM without getting the simulator to work? [ ;-) ] You know about the first attempt at making a 64-bit VM for HP don't you? A nightmare tale :-). > I am aware that Phil was busy with it recently, but I think there was also > someone else posting notes here. > > Would be great if those people could send me their change sets so that I > can try getting everything working again. > > As you can see in the screenshot, the simulator loads up properly, and > executes the first 10,000,000 bytecodes. > Which is enough to make it notice that a primitive fails, and to pop up > the error dialog. > However, it is a Pharo 1.2 image. Later images trigger some bug, I haven't > identified yet. They execute code, but don't manage to bring up the display. > > > Thanks to having a unified code repository, everything I got so far (minus > a few small changes) is here: > https://github.com/smarr/pharo-vm/tree/fixes/simulator-bit-rot > Hmm. What's the easiest way to map this to a Monticello mcz so I can integrate your changes back into my packages? > More to come soon. > Best regards > Stefan > > > -- > Stefan Marr > INRIA Lille - Nord Europe > http://stefan-marr.de/research/ > > > > -- best, Eliot

10 30

Adapting CMakeVMMaker to build BochsIA32Plugin
by Stefan Marr 14 Feb '14

14 Feb '14

Hi: I am slowly getting acquainted with CMake and the CMakeVMMaker infrastructure. In order to build the BochsIA32Plugin for the Cog simulator, I need to pass a number of build flags. Would be great if someone could tell me how to realize the following things: 1. remove -std=gnu99 for only this plugin BochsIA32Plugin is written in C++ and Clang doesn’t support the `-std=gnu99` switch in that case. 2. add header include paths The plugin needs the following two includes to access Bochs’ headers: -I${root}/processors/IA32/bochs/ -I${root}/processors/IA32/bochs/instrument/stubs/ 3. the final binary needs the following linker flags -lcpu -ldisasm -lfpu -L${root}/processors/IA32/macbochs/disasm/ -L${root}/processors/IA32/macbochs/cpu/ -L${root}/processors/IA32/macbochs/fpu/ In addition to that I would also need to add a script to build the bochs library before linking the binary. Any hints how to do that are very appreciate. Thanks Stefan -- Stefan Marr INRIA Lille - Nord Europe http://stefan-marr.de/research/

1 0

Re: [Pharo-dev] [Pharo-users] [gsoc-mentors] GSoC: call for ideas
by Eliot Miranda 14 Feb '14

14 Feb '14

Hi Ronie, On Thu, Feb 13, 2014 at 11:47 AM, Ronie Salgado <roniesalg(a)gmail.com> wrote: > I am proposing the following as a student: > > Project idea > > Name: Unified Foreign Function Interface > Skill level: Advanced > Possible Mentors: Igor Stasenko / Esteban Lorenzano > Name of the Student: Ronie Salgado > > Description: > > Because NativeBoost has problems with portability and cannot be used > anywhere(iOS), one proposal is making a modular unified foreign function > interface. This UFFI should support multiples backends, a portable function > calls specification, portable easy to use callbacks support, and portable > foreign resources management. > > Full Description Document: > http://users.dcc.uchile.cl/~rsalgado/uffi-fourth-draft.pdf > This is interesting but I take issue with a couple of things. First of all "The old Squeak FFI has problems with portability, callbacks and speed." I disagree that there are problems with callbacks. In fact, IIRC the only callback facility we have is that which I implemented which is used both by Alien and FFI and works well enough that the Newspeak system used it to implement a native GUI on windows where callbacks are used to receive Windows events. That counts as working in my book. Also, the architecture I implemented is portable to other ABIs than x86. But my major objection is to this: "One interesting idea is to make a backend that uses LLVM compiler infrastruc- ture, because it is designed for compiling optimized C function and it already has multiples JIT backends. This small project is not to make a full Pharo VM JIT based in LLVM, because LLVM is not designed for that and it would require a lot of patching of LLVM." IMO that's a very poor choice. LLVM is large, having a footprint of about a couple of megabytes. The Cog JIT is in contrast about 330k at its largest (Newspeak, support for two instruction sets, plus primitives, plus GC interface, etc, etc). If we as a community put effort into extending this we'll have a smaller footprint, a better JIT, and less dependencies. There's insignificant opportunity for using LLVM's optimizer when engineering call-outs and call-backs; they're too simple. So there's nothing to be gained other than bloat and dependency. And not putting effort into our own infrastructure is IMO a very bad idea. Note the synergy with Clément and my work on speculative inlining which will extend the Cog JIT. 2014-02-13 12:21 GMT-03:00 tesonep(a)gmail.com <tesonep(a)gmail.com>: > > Project idea >> >> Description: Enhance refactorings with type information >> Skill level: Advance >> Possible Mentors: Guillermo Polito / Esteban Lorenzano / Santiago >> Bragagnolo >> Name of the Student: Pablo Tesone >> Keywords: tools, refactoring, code completion, type inference. >> >> Description: >> >> During the development of a program in any programming language, the IDE >> has a very important role. It will never replace the knowledge of the >> programmer but it can make his life easier. >> As Smalltalk doesn't have explicit type information, the IDE can not >> perform the same mechanisms that are present in IDE like Eclipse or >> Intellij IDEA. These mechanisms can be very useful. >> The idea of this project is to add type inference mechanisms to try to >> provide more information to the refactoring and code completion tools. >> It is important to say that the idea is not to check types or correct the >> program, but to allow the programmer to have more information. >> >> >> >> On Thu, Feb 13, 2014 at 12:06 PM, Usman Bhatti <usman.bhatti(a)gmail.com>wrote: >> >>> [Anne Etien could not post this msg on the mentors list I m forwarding >>> it on her behalf]. >>> >>> Title: FAST JavaScript model >>> >>> Level: advanced >>> >>> Possible mentor: Anne Etien >>> >>> Possible second mentor: Nicolas Anquetil or Yuriy Tymchuk >>> >>> Description: >>> For in depth source code analysis a support of abstract syntax trees is >>> required. FAST is an abstract syntax tree extension for FAMIX meta-model >>> that is used by Moose technology. The goal of this project is to create >>> a JavaScript version of FAST. >>> >>> Technical Details: >>> As programming languages are different, their AST representations are >>> different too. FAST model aims for creation of as generic as possible >>> core that can be extended for different languages. Also the structure of >>> a model allows creation of generic algorithms like symbol resolution, >>> metrics calculation and rule checking that will work for any language. >>> Prototypes of FAST for Smalltalk and Java are already implemented. >>> During the project a student will implement the JavaScript model. More >>> detailed information about FAST is provided in the Moose day >>> presentation: http://youtu.be/dRr3WHOD3x4 >>> >>> Benefits to the Student: >>> The student will gain a deep understanding of a JavaScript syntax and >>> abstract syntax tree model. He will also learn about PetitParser >>> framework, and gain knowledge about software modeling and analysis. >>> >>> Benefits to the Community: >>> Community will get a FAST model for JavaScript that can be used for >>> software assessment with Moose. Also this model can be used later in PhD >>> projects such as automation of source code translation. With this third >>> model, the community will have raw material to real think and improve >>> generic algorithms based on AST. >>> >>> >>> On Tue, Feb 11, 2014 at 10:42 AM, Damien Cassou <damien.cassou(a)gmail.com >>> > wrote: >>> >>>> Hi fellow Pharo hackers, >>>> >>>> ESUG, the European Smalltalk User Group, is applying for this >>>> year's Google Summer of Code. As you probably know, the Summer >>>> of Code provides the opportunity to fund students to work during >>>> the summer on Pharo. Please reply to this >>>> email (be sure to use "Reply to all") if you have ideas you >>>> would like to propose. >>>> >>>> Please include a summary of the project and links to web pages >>>> that can help prospective students to write their application. >>>> Please also include the following information: >>>> >>>> - if applicable, other dialects that you would be willing to >>>> mentor this project for >>>> >>>> - the skill level >>>> >>>> - name of the mentor(s), email addresses, and possibly any IRC >>>> network/channel/nickname where they can be found. >>>> >>>> Thanks for contributing to ESUG's Summer of Code application! >>>> >>>> -- >>>> Damien Cassou >>>> http://damiencassou.seasidehosting.st >>>> >>>> "Success is the ability to go from one failure to another without >>>> losing enthusiasm." >>>> Winston Churchill >>>> >>>> -- >>>> You received this message because you are subscribed to the Google >>>> Groups "Smalltalk GSoC mentors" group. >>>> To unsubscribe from this group and stop receiving emails from it, send >>>> an email to smalltalk-gsoc-mentors+unsubscribe(a)googlegroups.com. >>>> For more options, visit https://groups.google.com/groups/opt_out. >>>> >>> >>> -- >>> You received this message because you are subscribed to the Google >>> Groups "Smalltalk GSoC mentors" group. >>> To unsubscribe from this group and stop receiving emails from it, send >>> an email to smalltalk-gsoc-mentors+unsubscribe(a)googlegroups.com. >>> For more options, visit https://groups.google.com/groups/opt_out. >>> >> >> >> >> -- >> Pablo Tesone. >> tesonep(a)gmail.com >> > > -- best, Eliot

4 3

Re: [Vm-dev] Simulator in Pharo 3
by Craig Latta 13 Feb '14

13 Feb '14

(followups to vm-dev, please) Hi Phil-- > ...having a clean interpreter and a basic image (Maybe > PharoKernel/Hazelnut/Boostrapping/Spoon will give us a very simple > image to simulate - the whole image isn't really needed). Yeah, the simulator is crucial to the Spoon[1] work, and I've kept it working for its development memories at all times. None of those memories have GUIs in them, but I am simulating things that were never simulated before, like live network access (so remote messaging between a simulated memory and a normal one works). thanks, -C [1] http://netjam.org/spoon -- Craig Latta www.netjam.org/resume +31 6 2757 7177 (SMS ok) + 1 415 287 3547 (no SMS)

1 0

Re: [Pharo-dev] need some help with NBCallback related crashes
by Eliot Miranda 13 Feb '14

13 Feb '14

cc'ing vm-dev cuz /this is a vm-dev issue/. On Wed, Feb 12, 2014 at 4:40 AM, Danil Osipchuk <danil.osipchuk(a)gmail.com>wrote: > Hello all > > I'm currently playing with NativeBoost (a great thing to have) and pharo > consistently crashes after running for a while. My impression is that > crashes happen during GCs as if something moves under the installed > callback. This may be also related to the libpcap itself, but how can I > know for sure. I've tried everything newest (vm+image+libpcap) and I also > migrated into a much more rigid scheme of storing and passing callback. Now > I've run of ideas where to look further. > > So, it is like this. NBPCapHandlerCallback is a procedure which is called > by libpcap for every packet it has in its internal buffer after the > application calls pcap_dispatch(). pcap_dispatch receives the callback as a > parameter. I have to repeatedly call pcap_dispatch in so called > nonblocking mode (and pass the callback each time) every 5 milliseconds to > collect packets - we are single threaded. This amounts to awful whole 200 > callback installations per second, may be libpcap was not well tested in > this scenario (I suppose pcap_loop runs in separated thread in wireshark > and callback installation happens several times per session at most) - > hence my doubts. But also - could it be something with NB? > > Any hints how to proceed with it are greatly appreciated. This is amazing > how one can quickly assemble something that emulates thousands of > dhcp-clients (this is what the app does) with current Pharo. But the thing > dies off after 10-15 minutes. > There is a leak checker in the VM that you can use. Use -leakcheck 3 to run the leak checker before each full GC (1) and incremental GC (2, 1 + 2 = 3). Use -leakcheck 7 to run it before each become as well. This should at least print an error before the crash. If you then use gdb to stop in the warning function you can use facilities such as printOop to find out which object is crapping out. HTH and please bring issues like this up on vm-dev. > > > best wishes, > Danil > > http://www.tcpdump.org/manpages/pcap_loop.3pcap.html > > ========= > Generic failure > NBNativeCodeGen class>>signalError: > NBNativeCodeGen class>>handleFailureIn:nativeCode: > NBNativeCodeGen class>>methodAssembly: > > LargePositiveInteger(NBFFICallback)>>primLeave:stackPtr:contextOop:returnValue:primitiveMethod: > NBPCapHandlerCallback(NBFFICallback)>>pvtEnter:stackPointer:primitiveMethod: > in Block: > Segmentation fault Wed Feb 12 16:17:25 2014 > > > pharo VM version: 3.9-7 #1 Fri Feb 7 16:55:52 CET 2014 gcc 4.6.3 > [Production ITHB VM] > Built from: NBCoInterpreter NativeBoost-CogPlugin-GuillermoPolito.19 uuid: > acc98e51-2fba-4841-a965-2975997bba66 Feb 7 2014 > With: NBCogit NativeBoost-CogPlugin-GuillermoPolito.19 uuid: > acc98e51-2fba-4841-a965-2975997bba66 Feb 7 2014 > Revision: https://github.com/pharo-project/pharo-vm.git Commit: > ef5832e6f70e5b24e8b9b1f4b8509a62b6c88040 Date: 2014-01-26 15:34:28 +0100 > By: Esteban Lorenzano <estebanlm(a)gmail.com> Jenkins build #14797 > Build host: Linux pharo-linux 3.2.0-31-generic-pae #50-Ubuntu SMP Fri Sep > 7 16:39:45 UTC 2012 i686 i686 i386 GNU/Linux > plugin path: /home/Pharo/ [default: /home/Pharo/] > > > C stack backtrace: > ./pharo[0x809bc4c] > ./pharo[0x809bf66] > linux-gate.so.1(__kernel_rt_sigreturn+0x0)[0xb76e640c] > ./pharo(incrementalGC+0x18a)[0x80859da] > ./pharo[0x808655a] > ./pharo[0x808cbb7] > ./pharo[0x808ccba] > ./pharo(ceStackOverflow+0x59)[0x808f5f9] > [0x770f5260] > [0x771017e0] > [0x770fb11e] > [0x770fa939] > [0x770f9813] > [0x770f5700] > [0x770f55c0] > > > Smalltalk stack dump: > 0xbf8b6144 M BlockClosure>ensure: 0x7c429d44: a(n) BlockClosure > 0xbf8b6164 M Semaphore>critical: 0x772901d8: a(n) Semaphore > 0xbf8b6180 M SmallInteger(Integer)>atRandom 0x13=9 > 0xbf8b61a8 M NLDHCPClientInit>enter 0x7c3be76c: a(n) NLDHCPClientInit > 0xbf8b61c0 M [] in NLDHCPHost>changeStateTo: 0x793e816c: a(n) NLDHCPHost > 0xbf8b61e0 I [] in BlockClosure>newProcess 0x7c3be828: a(n) BlockClosure > > Most recent primitives > at:put: > at:put: > at:put: > at:put: > basicNew: > basicNew > class > class > replaceFrom:to:with:startingAt: > basicNew: > basicNew > class > class > replaceFrom:to:with:startingAt: > species > species > at:put: > at:put: > class > class > replaceFrom:to:with:startingAt: > species > species > at:put: > at:put: > class > class > replaceFrom:to:with:startingAt: > at:put: > species > basicNew: > replaceFrom:to:with:startingAt: > class > class > species > basicNew: > replaceFrom:to:with:startingAt: > replaceFrom:to:with:startingAt: > basicNew: > atAllPut: > class > class > replaceFrom:to:with:startingAt: > species > basicNew: > replaceFrom:to:with:startingAt: > at:put: > at:put: > class > at:put: > at:put: > at:put: > species > basicNew: > replaceFrom:to:with:startingAt: > species > basicNew: > replaceFrom:to:with:startingAt: > basicNew: > basicNew > class > class > replaceFrom:to:with:startingAt: > class > class > species > basicNew: > replaceFrom:to:with:startingAt: > replaceFrom:to:with:startingAt: > species > basicNew: > replaceFrom:to:with:startingAt: > basicNew: > basicNew > class > class > replaceFrom:to:with:startingAt: > class > class > replaceFrom:to:with:startingAt: > class > class > replaceFrom:to:with:startingAt: > basicNew: > at:put: > at:put: > class > class > replaceFrom:to:with:startingAt: > class > class > replaceFrom:to:with:startingAt: > class > class > replaceFrom:to:with:startingAt: > species > basicNew: > replaceFrom:to:with:startingAt: > new > at:put: > at:put: > class > stringHash:initialHash: > class > stringHash:initialHash: > class > stringHash:initialHash: > basicNew > objectAt: > basicNew: > stackp: > basicNew > primitiveResume > basicNew > basicNew > new: > basicNew > new: > objectAt: > basicNew: > stackp: > basicNew > primitiveResume > findNextUnwindContextUpTo: > terminateTo: > suspend > species > basicNew: > replaceFrom:to:with:startingAt: > species > species > class > class > basicNew > shallowCopy > shallowCopy > basicNew: > basicNew > species > basicNew: > replaceFrom:to:with:startingAt: > class > class > replaceFrom:to:with:startingAt: > species > basicNew: > replaceFrom:to:with:startingAt: > species > basicNew: > replaceFrom:to:with:startingAt: > replaceFrom:to:with:startingAt: > basicNew: > basicNew > species > basicNew: > replaceFrom:to:with:startingAt: > class > class > replaceFrom:to:with:startingAt: > species > basicNew: > replaceFrom:to:with:startingAt: > species > basicNew: > replaceFrom:to:with:startingAt: > replaceFrom:to:with:startingAt: > replaceFrom:to:with:startingAt: > wait > truncated > truncated > signal > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > new > at:put: > at:put: > class > at:put: > at:put: > class > stringHash:initialHash: > class > stringHash:initialHash: > class > stringHash:initialHash: > class > stringHash:initialHash: > class > stringHash:initialHash: > class > stringHash:initialHash: > species > basicNew: > replaceFrom:to:with:startingAt: > species > basicNew: > replaceFrom:to:with:startingAt: > species > basicNew: > replaceFrom:to:with:startingAt: > species > basicNew: > replaceFrom:to:with:startingAt: > class > species > hashBytes:startingWith: > species > species > class > class > class > stringHash:initialHash: > class > stringHash:initialHash: > class > stringHash:initialHash: > basicNew: > basicNew > class > class > replaceFrom:to:with:startingAt: > basicNew: > basicNew > class > class > replaceFrom:to:with:startingAt: > class > class > species > basicNew: > replaceFrom:to:with:startingAt: > replaceFrom:to:with:startingAt: > species > basicNew: > replaceFrom:to:with:startingAt: > class > class > species > basicNew: > replaceFrom:to:with:startingAt: > replaceFrom:to:with:startingAt: > species > basicNew: > replaceFrom:to:with:startingAt: > pcapSendPacket:size: > findNextUnwindContextUpTo: > terminateTo: > suspend > basicNew > basicNew > wait > **IncrementalGC** > > stack page bytes 4096 available headroom 3300 minimum unused headroom 3480 > > (Segmentation fault) > > -- best, Eliot

1 0

VM Maker: ..VMMaker.oscog-eem.618.mcz
by commits＠source.squeak.org 13 Feb '14

13 Feb '14

Eliot Miranda uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/..VMMaker.oscog-eem.618.mcz ==================== Summary ==================== Name: ..VMMaker.oscog-eem.618 Author: eem Time: 13 February 2014, 11:51:10.104 am UUID: 3756532a-fdbf-4e96-8cc9-be0f35157864 Ancestors: ..VMMaker.oscog-eem.617 Some musings on Spur compaction. =============== Diff against ..VMMaker.oscog-eem.617 =============== Item was added: + ----- Method: SpurMemoryManager>>abstractCompaction (in category 'compaction - analysis') ----- + abstractCompaction + "This method answers a rough estimate of compactibility." + <doNotGenerate> + | lowestFree freeChunks used movable | + lowestFree := SmallInteger maxVal. + freeChunks := Set new. + used := Set new. + movable := Set new. + self allObjectsInFreeTreeDo: + [:f| + (self addressAfter: f) < endOfMemory ifTrue: + [freeChunks add: f. + f < lowestFree ifTrue: [lowestFree := f]]]. + self allOldSpaceObjectsFrom: lowestFree do: + [:o| | size delta best | + size := self bytesInObject: o. + delta := SmallInteger maxVal. + freeChunks do: [:f| | fs | + ((fs := self bytesInObject: f) >= size) ifTrue: + [delta > (fs - size) ifTrue: + [delta := fs - size. best := f]]]. + best ifNotNil: + [movable add: o. + used add: (freeChunks remove: best)]]. + ^{ totalFreeOldSpace. movable inject: 0 into: [:s :o| s + (self bytesInObject: o)]. used inject: 0 into: [:s :o| s + (self bytesInObject: o)] }! Item was added: + ----- Method: SpurMemoryManager>>abstractPigCompaction (in category 'compaction - analysis') ----- + abstractPigCompaction + "This method answers a rough estimate of compactibility using a pig (a large free chunk)." + <doNotGenerate> + | pig pork moved unmoved nmoved nunmoved | + pig := self findAPig. + pork := self bytesInObject: pig. + moved := unmoved := nmoved := nunmoved := 0. + self allOldSpaceObjectsFrom: pig do: + [:o| | bytes | + bytes := self bytesInObject: o. + bytes <= pork + ifTrue: + [moved := moved + bytes. + nmoved := nmoved + 1. + pork := pork - bytes] + ifFalse: + [unmoved := unmoved + bytes. + nunmoved := nunmoved + 1]]. + ^{ self bytesInObject: pig. pork. moved. nmoved. unmoved. nunmoved }! Item was added: + ----- Method: SpurMemoryManager>>biggies (in category 'compaction - analysis') ----- + biggies + "This method answers a sorted collection of the objects >= 1,000,000 bytes long, + above the lowest large free chunk, sandwiched between nilObj and the end of memory." + <doNotGenerate> + | lowestFree biggies | + lowestFree := SmallInteger maxVal. + self allObjectsInFreeTreeDo: + [:f| (self addressAfter: f) < endOfMemory ifTrue: [f < lowestFree ifTrue: [lowestFree := f]]]. + biggies := SortedCollection new. + self allOldSpaceObjectsFrom: lowestFree do: + [:f| + (self bytesInObject: f) >= 1000000 ifTrue: + [biggies add: f]]. + ^{{nilObj hex. #nil}}, (biggies collect: [:f| {f hex. self bytesInObject: f}]), {{endOfMemory hex. #endOfMemory}}! Item was added: + ----- Method: SpurMemoryManager>>compactionIssues (in category 'compaction - analysis') ----- + compactionIssues + <doNotGenerate> + "Compaction isn't working well. It rarely moves more than a few tens of kilobytes. Why? + Load an image and before you run it, or just before a GC, run these anaylsis routines. + e.g. + self abstractCompaction #(63230272 75456 63210648) + shows we can move 75456 bytes of objects, but that would use 63210648 of free space. + i.e. there are lots of big free chunks in play, not many small ones that fit the bill. + + self largeFreeChunkDistribution + #( #('16r31C788' #nil) + #('16r1423AC0' 2061864) + #('16r1B705E8' 1515200) + #('16r1D31D20' 2011152) + #('16r1F37818' 1491480) + #('16r2225968' 1450512) + #('16r24C92C8' 48575672) (16r24C92C8 + 48575672) hex '16r531C780' a free chunk + #('16r531C788' #endOfMemory)) + shows there's plenty of large free chunks. And the trailing 16-byte free chunk shows coallescing is not working properly. + + self biggies #(#('16r31C788' #nil) #('16r531C788' #endOfMemory)) + shows there are no large objects to be moved. + + So... looks like compaction should hold onto the lowest large chunk and preferentially move objects into that. + Let's call it a pig. Compaction needs to whittle away at the pig. + + e.g. + self abstractPigCompaction #(2061864 0 2061864 18759 2018224 34757) + shows we can move 18759 objects that will occupy 2018224 bytes into that + low pig of 2061864 bytes."! Item was added: + ----- Method: SpurMemoryManager>>findAPig (in category 'compaction - analysis') ----- + findAPig + "Answer a large low free chuink." + <doNotGenerate> + | pig | + self allObjectsInFreeTreeDo: + [:f| + (self bytesInObject: f) >= 1000000 ifTrue: + [(pig isNil or: [pig > f]) ifTrue: + [pig := f]]]. + ^pig! Item was added: + ----- Method: SpurMemoryManager>>largeFreeChunkDistribution (in category 'compaction - analysis') ----- + largeFreeChunkDistribution + "This method answers a sorted collection of the free chunks >= 1,000,000 bytes long, + sandwiched between nilObj and the end of memory (ignoring the large chunk often found at the end of the heap)." + <doNotGenerate> + | freeChunks | + freeChunks := SortedCollection new. + self allObjectsInFreeTreeDo: + [:f| + ((self addressAfter: f) < endOfMemory + and: [(self bytesInObject: f) >= 1000000]) ifTrue: + [freeChunks add: f]]. + ^{{nilObj hex. #nil}}, (freeChunks collect: [:f| {f hex. self bytesInObject: f}]), {{endOfMemory hex. #endOfMemory}}!

1 0

VM Maker: ..VMMaker.oscog-eem.617.mcz
by commits＠source.squeak.org 13 Feb '14

13 Feb '14

Eliot Miranda uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/..VMMaker.oscog-eem.617.mcz ==================== Summary ==================== Name: ..VMMaker.oscog-eem.617 Author: eem Time: 13 February 2014, 9:38:02.186 am UUID: 522aeee6-194d-4777-a4a5-19e312108e86 Ancestors: ..VMMaker.oscog-eem.616 Use Smalltalk vm getSystemAttribute: instead of plain old Smalltalk getSystemAttribute: =============== Diff against ..VMMaker.oscog-eem.616 =============== Item was changed: ----- Method: CogVMSimulator>>primitiveGetAttribute (in category 'other primitives') ----- primitiveGetAttribute "Fetch the system attribute with the given integer ID. The result is a string, which will be empty if the attribute is not defined." | index s attribute | index := self stackIntegerValue: 0. self successful ifTrue: [ + attribute := systemAttributes at: index ifAbsent: [Smalltalk vm getSystemAttribute: index]. - attribute := systemAttributes at: index ifAbsent: [Smalltalk getSystemAttribute: index]. attribute ifNil: [ ^self primitiveFail ]. s := objectMemory instantiateClass: (objectMemory splObj: ClassByteString) indexableSize: attribute size. 1 to: attribute size do: [ :i | objectMemory storeByte: i-1 ofObject: s withValue: (attribute at: i) asciiValue]. + self pop: 2 "rcvr, attr" thenPush: s]! - self pop: 2. "rcvr, attr" - self push: s]! Item was changed: ----- Method: InterpreterSimulator>>primitiveGetAttribute (in category 'other primitives') ----- primitiveGetAttribute "Fetch the system attribute with the given integer ID. The result is a string, which will be empty if the attribute is not defined." | attr s attribute | attr := self stackIntegerValue: 0. successFlag ifTrue: [ + attribute := Smalltalk vm getSystemAttribute: attr. - attribute := Smalltalk getSystemAttribute: attr. attribute ifNil: [ ^self primitiveFail ]. s := self instantiateClass: (self splObj: ClassByteString) indexableSize: attribute size. 1 to: attribute size do: [ :i | self storeByte: i-1 ofObject: s withValue: (attribute at: i) asciiValue]. + self pop: 2 "rcvr, attr" thenPush: s]. - self pop: 2. "rcvr, attr" - self push: s]. ! Item was changed: ----- Method: NewspeakInterpreterSimulator>>primitiveGetAttribute (in category 'other primitives') ----- primitiveGetAttribute "Fetch the system attribute with the given integer ID. The result is a string, which will be empty if the attribute is not defined." | attr s attribute | attr := self stackIntegerValue: 0. self successful ifTrue: [ + attribute := Smalltalk vm getSystemAttribute: attr. - attribute := Smalltalk getSystemAttribute: attr. attribute ifNil: [ ^self primitiveFail ]. s := self instantiateClass: (self splObj: ClassByteString) indexableSize: attribute size. 1 to: attribute size do: [ :i | self storeByte: i-1 ofObject: s withValue: (attribute at: i) asciiValue]. + self pop: 2 "rcvr, attr" thenPush: s]. - self pop: 2. "rcvr, attr" - self push: s]. ! Item was changed: ----- Method: StackInterpreterSimulator>>primitiveGetAttribute (in category 'other primitives') ----- primitiveGetAttribute "Fetch the system attribute with the given integer ID. The result is a string, which will be empty if the attribute is not defined." | index s attribute | index := self stackIntegerValue: 0. self successful ifTrue: [ + attribute := systemAttributes at: index ifAbsent: [Smalltalk vm getSystemAttribute: index]. - attribute := systemAttributes at: index ifAbsent: [Smalltalk getSystemAttribute: index]. attribute ifNil: [ ^self primitiveFail ]. s := objectMemory instantiateClass: (objectMemory splObj: ClassByteString) indexableSize: attribute size. 1 to: attribute size do: [ :i | objectMemory storeByte: i-1 ofObject: s withValue: (attribute at: i) asciiValue]. + self pop: 2 "rcvr, attr" thenPush: s]! - self pop: 2. "rcvr, attr" - self push: s]!

1 0

Re: [Pharo-dev] [Vm-dev] Simulator in Pharo 3
by Eliot Miranda 13 Feb '14

13 Feb '14

On Thu, Feb 13, 2014 at 6:19 AM, <btc(a)openinworld.com> wrote: > phil(a)highoctane.be wrote: > >> Ron, >> >> Sure but at this time, not being able to run the simulator in a full >> Pharo environment is a severe issue. >> >> The key advantage is that the Smalltalk VM is written in itself. >> >> Now, what do I experience is that it is hard to embark on VM work. >> >> Why? Even if the PharoVMBuilders help in generating and compiling the VM >> for several platforms (I can build for Windows 8.1, OSX, iOS, Debian 7 etc) >> this is only one part of the puzzle. >> >> The next step, is to understand how things do work for real, the >> interpret() C loop isn't gonna help me one bit, I need to be able to >> simulate this one in a Simulator. And I want to do that on the Pharo >> platform. Currently on 2.0 - this will (again) be fun to get to work on 3.0. >> >> I am not in the league of the VM maintainers obviously. >> >> But Clement provided me with Eliot's Cog Simulation image, which works >> for his own work. But only his own. We do not have something like a >> Configuration to get the same thing. And that's a very acute pain. There >> are only so many hours in a day, true. So, in order to bring in more >> people/resources for the VM work, the barrier to entry should be moved a >> tad down. >> >> We tried to have the StackSimulator working for a while (Stephan I'll >> definitely get your version, thanks for it). This is also because >> InterpreterSimulator doesn't work anymore and we have to use the >> StackInterpreter instead (which is complicated already vs pure interpreter). >> >> All of this rant to say that having a clean interpreter and a basic image >> (Maybe PharoKernel/Hazelnut/Boostrapping/Spoon will give us a very >> simple image to simulate - the whole image isn't really needed. Hopefully >> Oz will also move us forward on that front). >> >> A real Pharo image is overkill for this as a ton of plugins get involved >> etc. >> >> The core system should be the same for Pharo/Squeak/... : >> >> VMMaker-MemoryManager >> VMMaker-Interpreter >> VMMaker-InterpreterSimulation >> VMMaker-MemoryManagerSimulation >> >> and of course the Slang things (VMMaker-Support and VMMaker-Translation >> to C). >> >> What I'd love to end up with is an embeddable VM that we could hook into >> other environments, languages etc (a bit like TCL for example). This >> requires work on how the VM core runs. And to explore this, simulation is >> needed. >> >> Sorry for the long post. I wish I was a millionaire to pour in some cash >> into the project to speed some areas up. Working on it :-) Maybe should we >> work in that direction for funding. http://www.gv.com/ where are you? >> >> Phil >> >> Getting the simulator working in Pharo could be a GSoC project? or is it > too advanced? > If you guys would read and contribute to vm-dev you'd already know that someone with no VM experience (tty) just implemented event processing for the simulator window. The simulator only supported the old polling event architecture. So no it's not too advanced. But more importantly why aren't you all discussing vm stuff on the vm forum? > cheers -ben > -- best, Eliot

1 0

VM Maker: ..VMMaker.oscog-eem.616.mcz
by commits＠source.squeak.org 13 Feb '14

13 Feb '14

Eliot Miranda uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/..VMMaker.oscog-eem.616.mcz ==================== Summary ==================== Name: ..VMMaker.oscog-eem.616 Author: eem Time: 12 February 2014, 8:33:36.811 pm UUID: 37fac92f-2bc3-4414-9fd6-4c5ac1378e46 Ancestors: .VMMaker.oscog-eem.615 Fix a Slang snafu in the last commit =============== Diff against .VMMaker.oscog-eem.615 =============== Item was changed: ----- Method: SpurSegmentManager>>postSnapshot (in category 'snapshot') ----- postSnapshot "Restore all shortened segments to their proper size, re-freeing the trailing space." numSegments - 1 to: 0 by: -1 do: [:i| (segments at: i) lastFreeObject ifNotNil: [:freeChunk| | address | address := (segments at: i) segLimit - manager bridgeSize. (segments at: i) segSize: (segments at: i) savedSegSize. + self bridgeFrom: (self addressOf: (segments at: i)) + to: (i < (numSegments - 1) ifTrue: [self addressOf: (segments at: i + 1)]). - self bridgeFrom: (segments at: i) - to: (i < (numSegments - 1) ifTrue: [segments at: i + 1]). manager addFreeChunkWithBytes: (segments at: i) segLimit - address - manager bridgeSize at: address]]. "perhaps this should read manager setEndOfMemory: 0; assimilateNewSegment: (segments at: numSegments - 1)" manager setEndOfMemory: (segments at: numSegments - 1) segLimit - manager bridgeSize! Item was changed: ----- Method: SpurSegmentManager>>prepareForSnapshot (in category 'snapshot') ----- prepareForSnapshot "shorten all segments by any trailing free space." <var: #seg type: #'SpurSegmentInfo *'> 0 to: numSegments - 1 do: [:i| (segments at: i) savedSegSize: (segments at: i) segSize; lastFreeObject: nil]. "Ideally finding the lastFreeObject of each segment would be done in some linear pass through the heap. But for now KISS." manager freeTreeNodesDo: [:freeChunk| | next seg | + next := manager objectAfter: freeChunk limit: manager endOfMemory. + (manager isSegmentBridge: next) ifTrue: - next := manager objectAfter: freeChunk limit: manager endOfMemory. - (manager isSegmentBridge: next) ifTrue: [seg := self segmentContainingObj: freeChunk. seg lastFreeObject: freeChunk]. + freeChunk]. - freeChunk]. 0 to: numSegments - 1 do: [:i| (segments at: i) lastFreeObject ifNotNil: [:freeChunk| manager detachFreeObject: freeChunk. (segments at: i) segSize: (manager startOfObject: freeChunk) + manager bridgeSize - (segments at: i) segStart. + self bridgeFrom: (self addressOf: (segments at: i)) + to: (i < (numSegments - 1) ifTrue: [self addressOf: (segments at: i + 1)])]]. - self bridgeFrom: (segments at: i) - to: (i < (numSegments - 1) ifTrue: [segments at: i + 1])]]. "perhaps this should read manager setEndOfMemory: 0; assimilateNewSegment: (segments at: numSegments - 1)" manager setEndOfMemory: (segments at: numSegments - 1) segLimit - manager bridgeSize!

1 0

VM Maker: .VMMaker.oscog-eem.615.mcz
by commits＠source.squeak.org 13 Feb '14

13 Feb '14

Eliot Miranda uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/.VMMaker.oscog-eem.615.mcz ==================== Summary ==================== Name: .VMMaker.oscog-eem.615 Author: eem Time: 12 February 2014, 8:02:02.982 pm UUID: f518acf8-ff68-42b9-9e4e-aec19abc4d41 Ancestors: VMMaker.oscog-eem.614 Make Spur snapshot avoid writing trailing free space in each segment to the image file Fix initialization of primitiveAccessorDepthTable during simulation. =============== Diff against VMMaker.oscog-eem.614 =============== Item was changed: ----- Method: CCodeGenerator>>shouldIncludeMethodFor:selector: (in category 'utilities') ----- shouldIncludeMethodFor: aClass selector: selector "process optional methods by interpreting the argument to the option: pragma as either a Cogit class name or a class variable name or a variable name in VMBasicConstants." (aClass >> selector pragmaAt: #option:) ifNotNil: [:pragma| | key | key := pragma argumentAt: 1. vmMaker ifNotNil: [(Cogit withAllSubclasses anySatisfy: [:c| c name = key]) ifTrue: [| cogitClass optionClass | cogitClass := Smalltalk classNamed: vmMaker cogitClassName. optionClass := Smalltalk classNamed: key. ^cogitClass includesBehavior: optionClass]. + ((vmClass + ifNotNil: [vmClass initializationOptions] + ifNil: [vmMaker options]) at: key ifAbsent: [false]) ifNotNil: - (vmMaker options at: key ifAbsent: [false]) ifNotNil: [:option| option ~~ false ifTrue: [^true]]. (aClass bindingOf: key) ifNotNil: [:binding| binding value ~~ false ifTrue: [^true]]. (VMBasicConstants bindingOf: key) ifNotNil: [:binding| binding value ~~ false ifTrue: [^true]]]. ^false]. ^true! Item was added: + ----- Method: CogVMSimulator>>codeGeneratorToComputeAccessorDepth (in category 'primitive support') ----- + codeGeneratorToComputeAccessorDepth + ^VMMaker new + cogitClass: (Smalltalk classNamed: (self class initializationOptions + at: #Cogit + ifAbsent: [self class cogitClass name])); + buildCodeGeneratorForInterpreter: CoInterpreterPrimitives + includeAPIMethods: false + initializeClasses: false! Item was changed: ----- Method: CogVMSimulator>>initialize (in category 'initialization') ----- initialize "Initialize the CogVMSimulator when running the interpreter inside Smalltalk. The primary responsibility of this method is to allocate Smalltalk Arrays for variables that will be declared as statically-allocated global arrays in the translated code." + transcript := Transcript. + objectMemory ifNil: [objectMemory := self class objectMemoryClass simulatorClass new]. cogit ifNil: [cogit := self class cogitClass new setInterpreter: self]. objectMemory coInterpreter: self cogit: cogit. cogit numRegArgs > 0 ifTrue: [debugStackDepthDictionary := Dictionary new]. cogThreadManager ifNotNil: [super initialize]. "Note: we must initialize ConstMinusOne & HasBeenReturnedFromMCPC differently for simulation, due to the fact that the simulator works only with +ve 32-bit values" ConstMinusOne := objectMemory integerObjectOf: -1. HasBeenReturnedFromMCPC := objectMemory integerObjectOf: -1. cogMethodZone := cogit methodZone. "Because Slang can't remove intermediate implicit receivers (cogit methodZone foo doesn't reduce to foo())" enableCog := true. methodCache := Array new: MethodCacheSize. atCache := Array new: AtCacheTotalSize. self flushMethodCache. self flushAtCache. cogCompiledCodeCompactionCalledFor := false. gcSemaphoreIndex := 0. externalSemaphoreSignalRequests := externalSemaphoreSignalResponses := #(). externalPrimitiveTable := CArrayAccessor on: (Array new: MaxExternalPrimitiveTableSize). externalPrimitiveTableFirstFreeIndex := 0. primitiveTable := self class primitiveTable copy. - primitiveAccessorDepthTable := objectMemory hasSpurMemoryManagerAPI ifTrue: - [self class primitiveAccessorDepthTable]. - pluginList := {'' -> self }. mappedPluginEntries := OrderedCollection new. + objectMemory hasSpurMemoryManagerAPI + ifTrue: + [primitiveAccessorDepthTable := Array new: primitiveTable size. + pluginList := {}. + self loadNewPlugin: ''] + ifFalse: + [pluginList := {'' -> self }]. desiredNumStackPages := desiredEdenBytes := desiredCogCodeSize := 0. "This is initialized on loading the image, but convenient for testing stack page values..." numStackPages := self defaultNumStackPages. startMicroseconds := Time totalSeconds * 1000000. maxLiteralCountForCompile := MaxLiteralCountForCompile. minBackwardJumpCountForCompile := MinBackwardJumpCountForCompile. flagInterpretedMethods := false. "initialize InterpreterSimulator variables used for debugging" byteCount := lastPollCount := sendCount := 0. quitBlock := [^ self]. traceOn := true. printSends := printFrameAtEachStep := printBytecodeAtEachStep := false. myBitBlt := BitBltSimulator new setInterpreter: self. - transcript := Transcript. displayForm := 'Display has not yet been installed' asDisplayText form. suppressHeartbeatFlag := deferSmash := deferredSmash := false. systemAttributes := Dictionary new. primTraceLog := CArrayAccessor on: (Array new: 256 withAll: 0). primTraceLogIndex := 0. traceLog := CArrayAccessor on: (Array new: TraceBufferSize withAll: 0). traceLogIndex := 0. traceSources := TraceSources. statCodeCompactionCount := 0. statCodeCompactionUsecs := 0. extSemTabSize := 256! Item was changed: ----- Method: FilePlugin>>primitiveDirectorySetMacTypeAndCreator (in category 'directory primitives') ----- primitiveDirectorySetMacTypeAndCreator + | creatorString typeString fileName creatorStringIndex typeStringIndex fileNameIndex fileNameSize okToSet | - | creatorString typeString fileName creatorStringIndex typeStringIndex fileNameIndex fileNameSize okToSet | <var: 'creatorStringIndex' type: 'char *'> <var: 'typeStringIndex' type: 'char *'> <var: 'fileNameIndex' type: 'char *'> <export: true> creatorString := interpreterProxy stackValue: 0. typeString := interpreterProxy stackValue: 1. fileName := interpreterProxy stackValue: 2. ((interpreterProxy isBytes: creatorString) + and: [(interpreterProxy isBytes: typeString) + and: [(interpreterProxy isBytes: fileName) + and: [(interpreterProxy byteSizeOf: creatorString) = 4 + and: [(interpreterProxy byteSizeOf: typeString) = 4]]]]) ifFalse: + [^interpreterProxy primitiveFail]. - and: [(interpreterProxy byteSizeOf: creatorString) - = 4]) - ifFalse: [^interpreterProxy primitiveFail]. - ((interpreterProxy isBytes: typeString) - and: [(interpreterProxy byteSizeOf: typeString) - = 4]) - ifFalse: [^interpreterProxy primitiveFail]. - (interpreterProxy isBytes: fileName) - ifFalse: [^interpreterProxy primitiveFail]. creatorStringIndex := interpreterProxy firstIndexableField: creatorString. typeStringIndex := interpreterProxy firstIndexableField: typeString. fileNameIndex := interpreterProxy firstIndexableField: fileName. fileNameSize := interpreterProxy byteSizeOf: fileName. "If the security plugin can be loaded, use it to check for permission. If not, assume it's ok" + sCSFTfn ~= 0 ifTrue: + [okToSet := self + cCode: '((sqInt (*)(char *, sqInt))sCSFTfn)(fileNameIndex, fileNameSize)' + inSmalltalk: [true]. + okToSet ifFalse: + [^interpreterProxy primitiveFail]]. - sCSFTfn ~= 0 - ifTrue: [okToSet := self cCode: ' ((sqInt (*)(char *, sqInt))sCSFTfn)(fileNameIndex, fileNameSize)'. - okToSet - ifFalse: [^interpreterProxy primitiveFail]]. (self + cCode: 'dir_SetMacFileTypeAndCreator(fileNameIndex, fileNameSize, typeStringIndex, creatorStringIndex)' + inSmalltalk: [true]) ifFalse: + [^interpreterProxy primitiveFail]. - cCode: 'dir_SetMacFileTypeAndCreator(fileNameIndex, fileNameSize,typeStringIndex, creatorStringIndex)' - inSmalltalk: [true]) - ifFalse: [^interpreterProxy primitiveFail]. interpreterProxy pop: 3! Item was added: + ----- Method: ObjectMemory>>postSnapshot (in category 'image save/restore') ----- + postSnapshot + "No op for Spur compatibility."! Item was changed: ----- Method: SpurMemoryManager>>garbageCollectForSnapshot (in category 'snapshot') ----- garbageCollectForSnapshot self flushNewSpace. "There is no place to put newSpace in the snapshot file." + self fullGC. + segmentManager prepareForSnapshot! - self fullGC! Item was added: + ----- Method: SpurMemoryManager>>postSnapshot (in category 'snapshot') ----- + postSnapshot + <doNotGenerate> + segmentManager postSnapshot! Item was changed: VMStructType subclass: #SpurSegmentInfo + instanceVariableNames: 'segStart segSize swizzle containsPinned savedSegSize lastFreeObject' - instanceVariableNames: 'segStart segSize swizzle containsPinned' classVariableNames: '' poolDictionaries: '' category: 'VMMaker-SpurMemoryManager'! Item was added: + ----- Method: SpurSegmentInfo>>lastFreeObject (in category 'accessing') ----- + lastFreeObject + "Answer the value of lastFreeObject" + + ^ lastFreeObject! Item was added: + ----- Method: SpurSegmentInfo>>lastFreeObject: (in category 'accessing') ----- + lastFreeObject: anObject + "Set the value of lastFreeObject" + + ^lastFreeObject := anObject! Item was added: + ----- Method: SpurSegmentInfo>>savedSegSize (in category 'accessing') ----- + savedSegSize + "Answer the value of savedSegSize" + + ^ savedSegSize! Item was added: + ----- Method: SpurSegmentInfo>>savedSegSize: (in category 'accessing') ----- + savedSegSize: anObject + "Set the value of savedSegSize" + + ^savedSegSize := anObject! Item was added: + ----- Method: SpurSegmentManager>>postSnapshot (in category 'snapshot') ----- + postSnapshot + "Restore all shortened segments to their proper size, + re-freeing the trailing space." + numSegments - 1 to: 0 by: -1 do: + [:i| + (segments at: i) lastFreeObject ifNotNil: + [:freeChunk| | address | + address := (segments at: i) segLimit - manager bridgeSize. + (segments at: i) segSize: (segments at: i) savedSegSize. + self bridgeFrom: (segments at: i) + to: (i < (numSegments - 1) ifTrue: [segments at: i + 1]). + manager + addFreeChunkWithBytes: (segments at: i) segLimit - address - manager bridgeSize + at: address]]. + + "perhaps this should read + manager setEndOfMemory: 0; assimilateNewSegment: (segments at: numSegments - 1)" + manager setEndOfMemory: (segments at: numSegments - 1) segLimit - manager bridgeSize! Item was added: + ----- Method: SpurSegmentManager>>prepareForSnapshot (in category 'snapshot') ----- + prepareForSnapshot + "shorten all segments by any trailing free space." + <var: #seg type: #'SpurSegmentInfo *'> + 0 to: numSegments - 1 do: + [:i| + (segments at: i) + savedSegSize: (segments at: i) segSize; + lastFreeObject: nil]. + + "Ideally finding the lastFreeObject of each segment would be + done in some linear pass through the heap. But for now KISS." + manager freeTreeNodesDo: + [:freeChunk| | next seg | + next := manager objectAfter: freeChunk limit: manager endOfMemory. + (manager isSegmentBridge: next) ifTrue: + [seg := self segmentContainingObj: freeChunk. + seg lastFreeObject: freeChunk]. + freeChunk]. + + 0 to: numSegments - 1 do: + [:i| + (segments at: i) lastFreeObject ifNotNil: + [:freeChunk| + manager detachFreeObject: freeChunk. + (segments at: i) + segSize: (manager startOfObject: freeChunk) + + manager bridgeSize + - (segments at: i) segStart. + self bridgeFrom: (segments at: i) + to: (i < (numSegments - 1) ifTrue: [segments at: i + 1])]]. + + "perhaps this should read + manager setEndOfMemory: 0; assimilateNewSegment: (segments at: numSegments - 1)" + manager setEndOfMemory: (segments at: numSegments - 1) segLimit - manager bridgeSize! Item was added: + ----- Method: SpurSegmentManager>>segmentContainingObj: (in category 'accessing') ----- + segmentContainingObj: objOop + <returnTypeC: #'SpurSegmentInfo *'> + numSegments - 1 to: 0 by: -1 do: + [:i| + objOop >= (segments at: i) segStart ifTrue: + [^self addressOf: (segments at: i)]]. + ^nil! Item was added: + ----- Method: StackInterpreter>>codeGeneratorToComputeAccessorDepth (in category 'primitive support') ----- + codeGeneratorToComputeAccessorDepth + ^VMMaker new + buildCodeGeneratorForInterpreter: StackInterpreterPrimitives + includeAPIMethods: false + initializeClasses: false! Item was changed: ----- Method: StackInterpreter>>snapshot: (in category 'image save/restore') ----- snapshot: embedded "update state of active context" | activeContext activeProc rcvr setMacType stackIndex | <var: #setMacType type: #'void *'> "For now the stack munging below doesn't deal with more than one argument. It can, and should." argumentCount ~= 0 ifTrue: [^self primitiveFailFor: PrimErrBadNumArgs]. "Need to convert all frames into contexts since the snapshot file only holds objects." self push: instructionPointer. activeContext := self voidVMStateForSnapshotFlushingExternalPrimitivesIf: true. "update state of active process" activeProc := self activeProcess. objectMemory storePointer: SuspendedContextIndex ofObject: activeProc withValue: activeContext. tempOop := activeContext. objectMemory garbageCollectForSnapshot. "Nothing moves from here on so it is safe to grab the activeContext again." activeContext := tempOop. tempOop := 0. self successful ifTrue: ["Without contexts or stacks simulate rcvr := self popStack. ''pop rcvr'' self push: trueObj. to arrange that the snapshot resumes with true. N.B. stackIndex is one-relative." stackIndex := self quickFetchInteger: StackPointerIndex ofObject: activeContext. rcvr := objectMemory fetchPointer: stackIndex + CtxtTempFrameStart - 1 ofObject: activeContext. objectMemory storePointerUnchecked: stackIndex + CtxtTempFrameStart - 1 ofObject: activeContext withValue: objectMemory trueObject. "now attempt to write the snapshot file" self writeImageFileIO. (self successful and: [embedded not]) ifTrue: ["set Mac file type and creator; this is a noop on other platforms" setMacType := self ioLoadFunction: 'setMacFileTypeAndCreator' From: 'FilePlugin'. setMacType = 0 ifFalse: [self cCode: '((sqInt (*)(char *, char *, char *))setMacType)(imageName, "STim", "FAST")']]. "Without contexts or stacks simulate self pop: 1" objectMemory storePointerUnchecked: StackPointerIndex ofObject: activeContext withValue: (objectMemory integerObjectOf: stackIndex - 1)]. + objectMemory postSnapshot. self marryContextInNewStackPageAndInitializeInterpreterRegisters: activeContext. self successful ifTrue: [self push: objectMemory falseObject] ifFalse: [self push: rcvr. self justActivateNewMethod]! Item was changed: ----- Method: StackInterpreter>>tryLoadNewPlugin:pluginEntries: (in category 'primitive support') ----- tryLoadNewPlugin: pluginString pluginEntries: pluginEntries "Load the plugin and if on Spur, populate pluginEntries with the prmitives in the plugin." <doNotGenerate> | plugin plugins simulatorClasses | self transcript cr; show: 'Looking for module ', pluginString. "Defeat loading of the FloatArrayPlugin & Matrix2x3Plugin since complications with 32-bit float support prevent simulation. If you feel up to tackling this start by implementing cCoerce: value to: cType ^cType = 'float' ifTrue: [value asIEEE32BitWord] ifFalse: [value] in FloatArrayPlugin & Matrix2x3Plugin and then address the issues in the BalloonEnginePlugin. See http://forum.world.st/Simulating-the-BalloonEnginePlugin-FloatArrayPlugin-a…" (#('FloatArrayPlugin' 'Matrix2x3Plugin') includes: pluginString) ifTrue: [self transcript show: ' ... defeated'. ^nil]. pluginString isEmpty ifTrue: [plugin := self] ifFalse: [plugins := InterpreterPlugin allSubclasses select: [:psc| psc moduleName asString = pluginString asString]. simulatorClasses := (plugins select: [:psc| psc simulatorClass notNil] thenCollect: [:psc| psc simulatorClass]) asSet. simulatorClasses isEmpty ifTrue: [self transcript show: ' ... not found'. ^nil]. simulatorClasses size > 1 ifTrue: [^self error: 'This won''t work...']. plugins size > 1 ifTrue: [self transcript show: '...multiple plugin classes; choosing ', plugins last name]. plugin := simulatorClasses anyOne newFor: plugins last. "hopefully lowest in the hierarchy..." plugin setInterpreter: objectMemory. "Ignore return value from setInterpreter" (plugin respondsTo: #initialiseModule) ifTrue: [plugin initialiseModule ifFalse: [self transcript show: ' ... initialiser failed'. ^nil]]]. "module initialiser failed" self transcript show: ' ... loaded'. objectMemory hasSpurMemoryManagerAPI ifTrue: [| realPlugin cg | self transcript show: '...computing accessor depths'. plugin class isPluginClass ifTrue: [realPlugin := plugin class withAllSuperclasses detect: [:class| class shouldBeTranslated]. cg := realPlugin buildCodeGeneratorUpTo: realPlugin] ifFalse: + [cg := self codeGeneratorToComputeAccessorDepth. + primitiveTable withIndexDo: + [:prim :index| | depth | + prim isSymbol ifTrue: + [depth := cg accessorDepthForSelector: prim. + self assert: depth isInteger. + primitiveAccessorDepthTable at: index - 1 put: depth]]]. - [cg := VMMaker new - buildCodeGeneratorForInterpreter: StackInterpreter - includeAPIMethods: false - initializeClasses: false]. cg exportedPrimitiveNames do: [:primName| | fnSymbol | fnSymbol := primName asSymbol. pluginEntries addLast: {plugin. fnSymbol. [plugin perform: fnSymbol. self]. cg accessorDepthForSelector: fnSymbol}]. self transcript show: '...done']. ^pluginString asString -> plugin! Item was changed: ----- Method: StackInterpreterSimulator>>initialize (in category 'initialization') ----- initialize "Initialize the StackInterpreterSimulator when running the interpreter inside Smalltalk. The primary responsibility of this method is to allocate Smalltalk Arrays for variables that will be declared as statically-allocated global arrays in the translated code." bootstrapping := false. + transcript := Transcript. objectMemory ifNil: [objectMemory := self class objectMemoryClass simulatorClass new]. objectMemory coInterpreter: self. "Note: we must initialize ConstMinusOne differently for simulation, due to the fact that the simulator works only with +ve 32-bit values" ConstMinusOne := objectMemory integerObjectOf: -1. methodCache := Array new: MethodCacheSize. atCache := Array new: AtCacheTotalSize. self flushMethodCache. gcSemaphoreIndex := 0. externalSemaphoreSignalRequests := externalSemaphoreSignalResponses := #(). externalPrimitiveTable := CArrayAccessor on: (Array new: MaxExternalPrimitiveTableSize). externalPrimitiveTableFirstFreeIndex := 0. primitiveTable := self class primitiveTable copy. + mappedPluginEntries := OrderedCollection new. objectMemory hasSpurMemoryManagerAPI ifTrue: + [primitiveAccessorDepthTable := Array new: primitiveTable size. + pluginList := {}. + self loadNewPlugin: ''] - [primitiveAccessorDepthTable := self class primitiveAccessorDepthTable. - pluginList := {}] ifFalse: [pluginList := {'' -> self }]. - mappedPluginEntries := OrderedCollection new. desiredNumStackPages := desiredEdenBytes := 0. "This is initialized on loading the image, but convenient for testing stack page values..." numStackPages := self defaultNumStackPages. startMicroseconds := Time totalSeconds * 1000000. "initialize InterpreterSimulator variables used for debugging" byteCount := 0. sendCount := 0. quitBlock := [^ self]. traceOn := true. printSends := printReturns := printFrameAtEachStep := printBytecodeAtEachStep := false. myBitBlt := BitBltSimulator new setInterpreter: self. - transcript := Transcript. displayForm := 'Display has not yet been installed' asDisplayText form. suppressHeartbeatFlag := false. systemAttributes := Dictionary new. extSemTabSize := 256. disableBooleanCheat := false! Item was added: + ----- Method: VMClass class>>initializationOptions (in category 'initialization') ----- + initializationOptions + ^initializationOptions!

1 0

← Newer
1
2
3
4
5
6
7
8
9
Older →

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Vm-dev February 2014