Vm-dev April 2021

vm-dev@lists.squeakfoundation.org

15 participants
95 discussions

[OpenSmalltalk/opensmalltalk-vm] For epoll aio, close and reopen the epoll fd in forked child process (#552)
by David T Lewis 24 Jul '21

24 Jul '21

Factor the epoll initiation out of aioInit into new epollInit function and invoke it via pthread_atfork in the child process after a fork. Prevents misdelivery of events from the common kernel epoll structures. You can view, comment on, or merge this pull request online at: https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/552 -- Commit Summary -- * For epoll aio, close and reopen the epoll fd in forked child process -- File Changes -- M platforms/unix/vm/aio.c (23) -- Patch Links -- https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/552.patch https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/552.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/552

2 3

[OpenSmalltalk/opensmalltalk-vm] How to build on Win32 using WSL/Ubuntu | Missing headers (#510)
by Christoph Thiede 09 Jul '21

09 Jul '21

`build.win32x86/HowToBuild` describes how we can build the VM using cygwin. However, I have installed WSL instead. When running `mvm`, it gives me: ``` ../../platforms/win32/vm/sqWin32Utils.c:11:10: fatal error: Windows.h: No such file or directory ``` What do I have to do to enable mingw to find these header files? Any help is appreciated. -- 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/510

7 20

VM Maker Inbox: VMMaker.oscog-eem.2765.mcz
by commits＠source.squeak.org 09 Jul '21

09 Jul '21

Eliot Miranda uploaded a new version of VMMaker to project VM Maker Inbox: http://source.squeak.org/VMMakerInbox/VMMaker.oscog-eem.2765.mcz ==================== Summary ==================== Name: VMMaker.oscog-eem.2765 Author: eem Time: 23 June 2020, 6:41:19.556484 pm UUID: 91b9976e-60b9-4b5c-a2f8-b621dd454a07 Ancestors: VMMaker.oscog-eem.2764 Spur: Rewrite the revised followForwardedObjectFields:toDepth: soi it can be correctly inlined. Slang: Change the order of application of ensureConditionalAssignmentsAreTransformedIn: so it is always the last thing done in tryToInlineMethodsIn:. Straight-forward optimization of bindVariablesIn: in the common case of methods being inlined swith unchanged parameters. =============== Diff against VMMaker.oscog-eem.2764 =============== Item was changed: ----- Method: Spur32BitCoMemoryManager>>followForwardedObjectFields:toDepth: (in category 'as yet unclassified') ----- followForwardedObjectFields: objOop toDepth: depth "Follow pointers in the object to depth. Answer if any forwarders were found. How to avoid cyclic structures?? A temporary mark bit? eem 6/22/2020 no need since depth is always finite." <api> <inline: false> + | found fmt limit | - | oop found fmt | found := false. self assert: ((self isPointers: objOop) or: [self isOopCompiledMethod: objOop]). fmt := self formatOf: objOop. + limit := (self numPointerSlotsOf: objOop format: fmt) - 1. "It is essential to skip the first field of a method because it may be a reference to a Cog method in the method zone, not a real object at all." ((self isCompiledMethodFormat: fmt) ifTrue: [1] ifFalse: [0]) + to: limit + do: [:i| | oop | - to: (self numPointerSlotsOf: objOop format: fmt) - 1 - do: [:i| oop := self fetchPointer: i ofObject: objOop. (self isNonImmediate: oop) ifTrue: [(self isForwarded: oop) ifTrue: [found := true. oop := self followForwarded: oop. self storePointer: i ofObject: objOop withValue: oop]. (depth > 0 and: [(self hasPointerFields: oop) and: [self followForwardedObjectFields: oop toDepth: depth - 1]]) ifTrue: [found := true]]]. ^found! Item was changed: ----- Method: Spur64BitCoMemoryManager>>followForwardedObjectFields:toDepth: (in category 'forwarding') ----- followForwardedObjectFields: objOop toDepth: depth "Follow pointers in the object to depth. Answer if any forwarders were found. How to avoid cyclic structures?? A temporary mark bit? eem 6/22/2020 no need since depth is always finite." <api> <inline: false> + | found fmt limit | - | oop found fmt | found := false. self assert: ((self isPointers: objOop) or: [self isOopCompiledMethod: objOop]). fmt := self formatOf: objOop. + limit := (self numPointerSlotsOf: objOop format: fmt) - 1. "It is essential to skip the first field of a method because it may be a reference to a Cog method in the method zone, not a real object at all." ((self isCompiledMethodFormat: fmt) ifTrue: [1] ifFalse: [0]) + to: limit + do: [:i| | oop | - to: (self numPointerSlotsOf: objOop format: fmt) - 1 - do: [:i| oop := self fetchPointer: i ofObject: objOop. (self isNonImmediate: oop) ifTrue: [(self isForwarded: oop) ifTrue: [found := true. oop := self followForwarded: oop. self storePointer: i ofObject: objOop withValue: oop]. (depth > 0 and: [(self hasPointerFields: oop) and: [self followForwardedObjectFields: oop toDepth: depth - 1]]) ifTrue: [found := true]]]. ^found! Item was changed: ----- Method: TMethod>>exitVar:label:in: (in category 'inlining') ----- exitVar: exitVar label: exitLabel in: aCodeGen "Replace each return statement in this method with an assignment to the exit variable followed by either a return or a goto to the given label. Answer if a goto was generated." "Optimization: If exitVar is nil, the return value of the inlined method is not being used, so don't add the assignment statement." | labelUsed map elisions eliminateReturnSelfs | labelUsed := false. map := Dictionary new. elisions := Set new. "Conceivably one might ^self from a struct class and mean it. In most cases though ^self means `get me outta here, fast'. So unless this method is from a VMStruct class, elide any ^self's" eliminateReturnSelfs := ((definingClass inheritsFrom: VMClass) and: [definingClass isStructClass]) not and: [returnType = #void or: [returnType = #sqInt]]. parseTree nodesDo: [:node | | replacement | node isReturn ifTrue: [self transformReturnSubExpression: node toAssignmentOf: exitVar andGoto: exitLabel unless: eliminateReturnSelfs into: [:rep :labelWasUsed| replacement := rep. labelWasUsed ifTrue: [labelUsed := true]] in: aCodeGen. "replaceNodesIn: is strictly top-down, so any replacement for ^expr ifTrue: [...^fu...] ifFalse: [...^bar...] will prevent replacement of either ^fu or ^bar. The corollary is that ^expr ifTrue: [foo] ifFalse: [^bar] must be transformed into expr ifTrue: [^foo] ifFalse: [^bar]" (node expression isConditionalSend and: [node expression hasExplicitReturn]) ifTrue: [elisions add: node. (node expression args reject: [:arg| arg endsWithReturn]) do: [:nodeNeedingReturn| self transformReturnSubExpression: nodeNeedingReturn statements last toAssignmentOf: exitVar andGoto: exitLabel unless: eliminateReturnSelfs into: [:rep :labelWasUsed| replacement := rep. + labelWasUsed ifTrue: [labelUsed := true]] + in: aCodeGen. - labelWasUsed ifTrue: [labelUsed := true]]. map at: nodeNeedingReturn statements last put: replacement]] ifFalse: [map at: node put: (replacement ifNil: [TLabeledCommentNode new setComment: 'return ', node expression printString])]]]. map isEmpty ifTrue: [self deny: labelUsed. ^false]. "Now do a top-down replacement for all returns that should be mapped to assignments and gotos" parseTree replaceNodesIn: map. "Now it is safe to eliminate the returning ifs..." elisions isEmpty ifFalse: [| elisionMap | elisionMap := Dictionary new. elisions do: [:returnNode| elisionMap at: returnNode put: returnNode expression]. parseTree replaceNodesIn: elisionMap]. "Now flatten any new statement lists..." parseTree nodesDo: [:node| | list | (node isStmtList and: [node statements notEmpty and: [node statements last isStmtList]]) ifTrue: [list := node statements last statements. node statements removeLast; addAllLast: list]]. ^labelUsed! Item was changed: ----- Method: TMethod>>inlineFunctionCall:in: (in category 'inlining') ----- inlineFunctionCall: aSendNode in: aCodeGen "Answer the body of the called function, substituting the actual parameters for the formal argument variables in the method body. Assume caller has established that: 1. the method arguments are all substitutable nodes, and 2. the method to be inlined contains no additional embedded returns." | sel meth doNotRename argsForInlining substitutionDict | + aCodeGen maybeBreakForInlineOf: aSendNode in: self. sel := aSendNode selector. meth := (aCodeGen methodNamed: sel) copy. meth ifNil: [^self inlineBuiltin: aSendNode in: aCodeGen]. doNotRename := Set withAll: args. argsForInlining := aSendNode argumentsForInliningCodeGenerator: aCodeGen. meth args with: argsForInlining do: [ :argName :exprNode | exprNode isLeaf ifTrue: [doNotRename add: argName]]. (meth statements size = 2 and: [meth statements first isSend and: [meth statements first selector == #flag:]]) ifTrue: [meth statements removeFirst]. meth renameVarsForInliningInto: self except: doNotRename in: aCodeGen. meth renameLabelsForInliningInto: self. self addVarsDeclarationsAndLabelsOf: meth except: doNotRename. substitutionDict := Dictionary new: meth args size * 2. meth args with: argsForInlining do: [ :argName :exprNode | + (exprNode isVariable and: [exprNode name = argName]) ifFalse: + [substitutionDict at: argName put: exprNode]. - substitutionDict at: argName put: exprNode. (doNotRename includes: argName) ifFalse: [locals remove: argName]]. meth parseTree bindVariablesIn: substitutionDict. ^meth parseTree endsWithReturn ifTrue: [meth parseTree copyWithoutReturn] ifFalse: [meth parseTree]! Item was changed: ----- Method: TMethod>>tryToInlineMethodsIn: (in category 'inlining') ----- tryToInlineMethodsIn: aCodeGen "Expand any (complete) inline methods sent by this method. Set the complete flag when all inlining has been done. Answer if something was inlined." | didSomething statementLists | "complete ifTrue: [^false]." self definedAsMacro ifTrue: [complete ifTrue: [^false]. ^complete := true]. - self ensureConditionalAssignmentsAreTransformedIn: aCodeGen. didSomething := self tryToInlineMethodStatementsIn: aCodeGen statementListsInto: [:stmtLists| statementLists := stmtLists]. didSomething := (self tryToInlineMethodExpressionsIn: aCodeGen) or: [didSomething]. + self ensureConditionalAssignmentsAreTransformedIn: aCodeGen. didSomething ifTrue: [writtenToGlobalVarsCache := nil]. complete ifFalse: [self checkForCompletenessIn: aCodeGen. complete ifTrue: [didSomething := true]]. "marking a method complete is progress" ^didSomething! Item was changed: ----- Method: TStmtListNode>>bindVariablesIn: (in category 'transformations') ----- bindVariablesIn: aDictionary + aDictionary notEmpty ifTrue: + [statements := statements collect: [:s| s bindVariablesIn: aDictionary]]! - statements := statements collect: [ :s | s bindVariablesIn: aDictionary ].!

2 1

[OpenSmalltalk/opensmalltalk.github.io] the opensmalltalk.org web site no longer works (#1)
by Alan Zimmerman 19 May '21

19 May '21

It goes to a github "site not found" page -- 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.github.io/issues/1

4 5

ARMv8 BitBLT speedups provided by RPi/Ben Avison
by tim Rowledge 12 May '21

12 May '21

Our friends in Cambridge (Real Cambridge, not Backup Cambridge) have spent a nice chunk of money to bring the ARM32 bitblt speedups they provided in '14/5/6 into the ARMv8 world. Quite a few microbenchmarks covering the tests Ben did (and he wrote a testing framework that ran some 10s of thousands of tests) are improved by an order of magnitude, some by 30X and one by 57X ! All the code to date is pushed to https://github.com/bavison/opensmalltalk-vm/tree/Ben with the latest version of the test harnesses at https://github.com/bavison/SqueakBitBltTest Look for items changed in the last few days ie 25th april or later. I haven't had a chance to do more than glance as yet. I *suspect* Ben has manually edited the generated BitBltPlugin.c file, which will mean doing some diffing and back-porting into Slang. Maybe. I, for one, welcome our new v8 blit overlords. tim tim Rowledge; tim(a)rowledge.org; http://www.rowledge.org/tim "#define QUESTION ((bb) || !(bb)) - Shakespeare."

3 4

[OpenSmalltalk/opensmalltalk-vm] [Win10] [Squeak] High CPU usage in different images for a few minutes after resume from sleep (#537)
by Christoph Thiede 08 May '21

08 May '21

I noticed this a few times, and today I observed it a bit: After resuming my Win10 message from sleep, my two open Squeak.exe processes both made up each ~14% CPU load. I suspect this performance gap is caused by the VM because, in one image, a nearly empty Morphic world was open (two non-stepping windows only) and in the other one, a completely empty MVC world; both images were responsive; and according to Squeak's CPU watcher, only ~20% of time were spent in the UI process but 80% in idle. After a few minutes, both VM instances fall back to ~0.1% of total CPU usage. Is there any chance to find an explanation for this behavior in the VM implementation, are there any hidden background operations (maybe certain WM_MESSAGES) that are triggered after resuming from sleep? Does anyone else notice similar behavior? For sake of completeness, I should mention that my RAM and disk usage are chronically near maximum (90% of RAM in use, 25 GB on SSD free. Squeak images are so large 😐). But this does not change after a few minutes so I don't think it explains the observed slowdown completely. VM build 202010232046; but I could already have experienced similar issues a few months ago. -- 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/537

2 2

[OpenSmalltalk/opensmalltalk-vm] Fix UDP binding in Windows sockets (#563)
by jdougan 05 May '21

05 May '21

Changed sqSocketListenOnPortBacklogSizeInterface() to match general behavior of Unix socket version. Previously was always incorrectly binding UDP sockets to INADDR_ANY instead of the specified IPv4 address. You can view, comment on, or merge this pull request online at: https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/563 -- Commit Summary -- * Changed sqSocketListenOnPortBacklogSizeInterface() to match general -- File Changes -- M platforms/win32/plugins/SocketPlugin/sqWin32NewNet.c (2) -- Patch Links -- https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/563.patch https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/563.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/563

4 5

classic VM: timezone correction in sqUnixUtcWithOffset()
by stes＠PANDORA.BE 27 Apr '21

27 Apr '21

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Hi, Here's another fix for sqUnixMain.c in the SVN (subversion) repository. This is for timezone correction in sqUnixUtcWithOffset(). Basically there is code for the Linux/glibc case which is correct: #if defined(HAVE_TM_GMTOFF) *offset= localtime(&seconds)->tm_gmtoff; #else The above code is correct but it depends on tm_gmtoff (glibc) as on Linux. The incorrect code is for platforms that use the plain standard C library, without the glibc extension (the #else clause). The timezone (HAVE_TIMEZONE) is itself also extension which is enabled by default for Solaris because it is _XOPEN_SOURCE but not _STRICT_STDC. - - { - - struct tm *local= localtime(&seconds); - - struct tm *gmt= gmtime(&seconds); - - int d= local->tm_yday - gmt->tm_yday; - - int h= ((d < -1 ? 24 : 1 < d ? -24 : d * 24) + local->tm_hour - gmt->tm_hour); - - int m= h * 60 + local->tm_min - gmt->tm_min; - - *offset= m * 60; - - } The above is incorrect because localtime() and gmtime() will overwrite the same buffer, so the gmtime() call erases the value of localtime(). I have a patch in attach that uses the same logic as plugins/LocalePlugin/sqLocalePlugin.c and as in convertToSqueakTime() which is doing it right. +#ifdef HAVE_TIMEZONE + *offset = ((daylight) * 60 * 60) - timezone; +#else +#error: cannot determine timezone correction #endif +#endif return 0; } Please apply the patch in attach to subversion ... I think it is safe to apply this patch because (1) it does not change the #if HAVE_TM_GMTOFF case (the Linux case is unchanged) and (2) it is consistent with plugins/LocalePlugin/sqLocalePlugin.c and convertToSqueakTime() and (3) the code with the difference between localtime() and gmtime() is certainly wrong. Thanks, David Stes -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQEcBAEBCAAGBQJgha+cAAoJENdFDkXGicizymEIAIqI8cicPesACVW3zXFzWWl9 m1L8BU8CAlgP4omO+scOtza+rj7Fy4I5tg+XFNX8u1cEtJIgYj2fvEpoOqUsYkbv /kzlQX850TGWKYFA5ya3RE7XRI3fx5hqqP2ntEd3Nb0Ve9yFsmlo3KxKRSkJMKfm 9Jwgrpu549AqwR+Hl+XkJxIiJ5xBotJyzipyB6ySXGqpCYs4qVJl0nM00ozVNY7n W7AhNh6mWX1pbvZLH9xPVO0qSRLNewQ9sGBbaGIAaQm7HScSHpM9DtDE3jY/FF0J CdNH84hKYdN7z8am1C1IDX95V4sILORyFaaa0F+QnOBFASMW1ltfkx+up6yHzXU= =4EbY -----END PGP SIGNATURE-----

2 1

openindiana 2954 cog vm
by stes＠PANDORA.BE 25 Apr '21

25 Apr '21

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Hello, Update about the OpenSmalltalk VM and classical VM Squeak on "OpenIndiana". Current upgrades in the repository, http://pkg.openindiana.org/hipster/en/index.shtml squeak-4 : version 4.19.6 (32bit + 64bit) squeak-5 : version 5.0.2952 (Stack 32bit + 64bit) squeak-5c : version 5.0.2954 (COG 32bit + 64bit) There are packages for 'headless' (-nodisplay) operation and with GUI. Special thanks to everybody in the "Squeak" and "opensmalltalk-vm" team; the OpenSmalltalk VM configures and builds OK in both 32bit and 64bit. Thanks, David Stes -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQEcBAEBCAAGBQJghYpYAAoJENdFDkXGicizGqEH/0s5T1BMibKN4zQyGUltLo4z KxscFLSAD3AFL6naNdaxbnpPkefoa5OGcdIjCaWsihw//W32LVOqdThsVqdeS08F Kg6/60JdD644IGjXLZf56E/OXIitOWa9Cf1YYROgfagEVm9GimftksEITWSwUSDk mm69Ko5j6446H9a5sMYbZyLrMKG/9DzaRLq7/vOpcVPIQuoGRazbOWdgK3zZhu9j ihUMc3BJX7s21FjOMaxr+0RsJDycUny94S1Gq7nw7ZagcUpJEEOJg94mzuC3J2dW 0+uY5brgQwVPottp85iHnHABcNkBoKtvDaXVN/j8Q0KJtBhgjiX93YHwGJelQ8o= =jXQf -----END PGP SIGNATURE-----

1 0

Re: [Vm-dev] How to convert 32-bit images to 64-bit images?
by stes＠PANDORA.BE 24 Apr '21

24 Apr '21

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Hi, One document that I've found interesting is : http://wiki.squeak.org/squeak/769 "FAQ: How can I bootstrap a Squeak image?" It discusses technologies that I'm personally NOT familiar with such as KernelImage and SystemTracer. This is not really an answer to your question but the subject is similar, namely how to build or convert an image. As to Tim Rowledge answer to check out Spur32to64BitImageConverter, this is probably a much better answer. Regards, David Stes -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQEcBAEBCAAGBQJgdIC0AAoJENdFDkXGicizcuEIAKPKWKmOpfew7n25Kvz7My+Z ZbRXRiM8RxePA+mZg8r+CHqlgFL6D+PvCGyNkqmdJsvZyaiaiuu5uI8OOUTVsD0F xjsRBY8XcDpiQel5M/5qyn5pV1D69ev9phCN+mwl+UPTluU7FNXJJ2zeZ1qeTDpe gpwzSxErPN1PfhwJJPax55eJehIX6MVd6mPqT2Grqc7aldZ249N5rElRhgeWEfSi dyoa/EvI3r3Cwrfwz7YQPHwlIYTAw4pKtk91/nG0d8L2S6ptbyr60lwuNb8Ama9o XddZiQSnOid9beeiTWG+PGvTIZYGymVAlm98A7ARVjoksXJoP7JlGoy/XeNqrno= =nK9F -----END PGP SIGNATURE-----

3 5

← Newer
1
2
3
4
5
6
7
8
9
10
Older →

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Vm-dev April 2021