Vm-dev May 2017

vm-dev@lists.squeakfoundation.org

39 participants
158 discussions

VM Maker: VMMaker.oscog-eem.2217.mcz
by commits＠source.squeak.org 19 May '17

19 May '17

Eliot Miranda uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-eem.2217.mcz ==================== Summary ==================== Name: VMMaker.oscog-eem.2217 Author: eem Time: 19 May 2017, 12:08:08.01384 pm UUID: 1b21a9ae-6dc4-40cb-9bb5-3da7e6b7c468 Ancestors: VMMaker.oscog-nice.2216 Add the Terf SoundPlugin extensions (to deal with multiple devices, their names, volumes, etc). =============== Diff against VMMaker.oscog-nice.2216 =============== Item was added: + ----- Method: SoundPlugin>>primitiveGetDefaultSoundPlayer (in category 'primitives') ----- + primitiveGetDefaultSoundPlayer + "Answer a String with the operating system name of the default output device, or nil" + "no arguments" + | cDeviceName sz newString newStringPtr | + <export: true> + <var: #cDeviceName type: 'char*'> + <var: #newStringPtr type: 'char*'> + + "Parse arguments" + interpreterProxy methodArgumentCount = 0 + ifFalse:[^interpreterProxy primitiveFail]. + + "Get the answer." + cDeviceName := self cCode: 'getDefaultSoundPlayer()'. + cDeviceName == 0 ifTrue: [ + ^interpreterProxy pop: 1 thenPush: interpreterProxy nilObject + ]. + + "Copy the answer to a Squeak String." + sz := self cCode: 'strlen(cDeviceName)'. + newString := interpreterProxy + instantiateClass: interpreterProxy classString + indexableSize: sz. + newStringPtr := interpreterProxy firstIndexableField: newString. + self cCode: 'strncpy(newStringPtr, cDeviceName, sz)'. + + self touch: newStringPtr. + self touch: cDeviceName. + "Pop the receiver, and answer the new string." + ^interpreterProxy pop: 1 thenPush: newString! Item was added: + ----- Method: SoundPlugin>>primitiveGetDefaultSoundRecorder (in category 'primitives') ----- + primitiveGetDefaultSoundRecorder + "Answer a String with the operating system name of the default input device, or nil" + "no arguments" + | cDeviceName sz newString newStringPtr | + <export: true> + <var: #cDeviceName type: 'char*'> + <var: #newStringPtr type: 'char*'> + + "Parse arguments" + interpreterProxy methodArgumentCount = 0 + ifFalse:[^interpreterProxy primitiveFail]. + + "Get the answer." + cDeviceName := self cCode: 'getDefaultSoundRecorder()'. + cDeviceName == 0 ifTrue: [ + ^interpreterProxy pop: 1 thenPush: interpreterProxy nilObject + ]. + + "Copy the answer to a Squeak String." + sz := self cCode: 'strlen(cDeviceName)'. + newString := interpreterProxy + instantiateClass: interpreterProxy classString + indexableSize: sz. + newStringPtr := interpreterProxy firstIndexableField: newString. + self cCode: 'strncpy(newStringPtr, cDeviceName, sz)'. + + self touch: newStringPtr. + self touch: cDeviceName. + "Pop the receiver, and answer the new string." + ^interpreterProxy pop: 1 thenPush: newString! Item was added: + ----- Method: SoundPlugin>>primitiveGetNumberOfSoundPlayerDevices (in category 'primitives') ----- + primitiveGetNumberOfSoundPlayerDevices + "arguments: name(type, stack offset) + dialString(String, 0)" + "answers an Integer" + | result | + <export: true> + + "Parse arguments" + interpreterProxy methodArgumentCount = 0 + ifFalse:[^interpreterProxy primitiveFail]. + + "get result" + result := self cCode: 'getNumberOfSoundPlayerDevices()'. + + "answer it" + result := interpreterProxy signed32BitIntegerFor: result. + ^interpreterProxy pop: 1 thenPush: result. "pop receiver, return result"! Item was added: + ----- Method: SoundPlugin>>primitiveGetNumberOfSoundRecorderDevices (in category 'primitives') ----- + primitiveGetNumberOfSoundRecorderDevices + "arguments: name(type, stack offset) + dialString(String, 0)" + "answers an Integer" + | result | + <export: true> + + "Parse arguments" + interpreterProxy methodArgumentCount = 0 + ifFalse:[^interpreterProxy primitiveFail]. + + "get result" + result := self cCode: 'getNumberOfSoundRecorderDevices()'. + + "answer it" + result := interpreterProxy signed32BitIntegerFor: result. + ^interpreterProxy pop: 1 thenPush: result. "pop receiver, return result"! Item was added: + ----- Method: SoundPlugin>>primitiveGetSoundPlayerDeviceName (in category 'primitives') ----- + primitiveGetSoundPlayerDeviceName + "arguments: name(type, stack offset) + deviceNumber(Integer, 0)" + "answers a string or nil" + | deviceNumber sz cDeviceName newString newStringPtr | + <export: true> + <var: #cDeviceName type: 'char*'> + <var: #newStringPtr type: 'char*'> + + "Parse arguments" + interpreterProxy methodArgumentCount = 1 + ifFalse:[^interpreterProxy primitiveFail]. + + deviceNumber := interpreterProxy positive32BitValueOf: (interpreterProxy stackValue: 0). + interpreterProxy failed ifTrue: [^nil]. + + "Get the answer." + cDeviceName := self cCode: 'getSoundPlayerDeviceName(deviceNumber - 1)'. + cDeviceName == 0 ifTrue: [ + ^interpreterProxy pop: 2 thenPush: interpreterProxy nilObject + ]. + + "Copy the answer to a Squeak String." + sz := self cCode: 'strlen(cDeviceName)'. + newString := interpreterProxy + instantiateClass: interpreterProxy classString + indexableSize: sz. + newStringPtr := interpreterProxy firstIndexableField: newString. + self cCode: 'strncpy(newStringPtr, cDeviceName, sz)'. + + self touch: deviceNumber. + self touch: newStringPtr. + self touch: cDeviceName. + "Pop the receiver and arg, and answer the new string." + ^interpreterProxy pop: 2 thenPush: newString! Item was added: + ----- Method: SoundPlugin>>primitiveGetSoundRecorderDeviceName (in category 'primitives') ----- + primitiveGetSoundRecorderDeviceName + "arguments: name(type, stack offset) + deviceNumber(Integer, 0)" + "answers a string or nil" + | deviceNumber sz cDeviceName newString newStringPtr | + <export: true> + <var: #cDeviceName type: 'char*'> + <var: #newStringPtr type: 'char*'> + + "Parse arguments" + interpreterProxy methodArgumentCount = 1 + ifFalse:[^interpreterProxy primitiveFail]. + + + deviceNumber := interpreterProxy positive32BitValueOf: (interpreterProxy stackValue: 0). + interpreterProxy failed ifTrue: [^nil]. + + "Get the answer." + cDeviceName := self cCode: 'getSoundRecorderDeviceName(deviceNumber - 1)'. + cDeviceName == 0 ifTrue: [ + ^interpreterProxy pop: 2 thenPush: interpreterProxy nilObject + ]. + + "Copy the answer to a Squeak String." + sz := self cCode: 'strlen(cDeviceName)'. + newString := interpreterProxy + instantiateClass: interpreterProxy classString + indexableSize: sz. + newStringPtr := interpreterProxy firstIndexableField: newString. + self cCode: 'strncpy(newStringPtr, cDeviceName, sz)'. + + self touch: deviceNumber. + self touch: newStringPtr. + self touch: cDeviceName. + "Pop the receiver and arg, and answer the new string." + ^interpreterProxy pop: 2 thenPush: newString! Item was added: + ----- Method: SoundPlugin>>primitiveSetDefaultSoundPlayer (in category 'primitives') ----- + primitiveSetDefaultSoundPlayer + "Tell the operating system to use the specified device name as the output device for sound." + "arg at top of stack is the String" + | deviceName obj srcPtr sz | + <export: true> + <var: 'deviceName' declareC: 'char deviceName[257]'> + <var: 'srcPtr' type: #'char *'> + + "Parse arguments" + interpreterProxy methodArgumentCount = 1 ifFalse: + [^interpreterProxy primitiveFail]. + obj := interpreterProxy stackValue: 0. + (interpreterProxy isBytes: obj) ifFalse: + [^interpreterProxy primitiveFail]. + (sz := interpreterProxy byteSizeOf: obj) <= 256 ifFalse: + [^interpreterProxy primitiveFail]. + srcPtr := interpreterProxy firstIndexableField: obj. + self touch: srcPtr. + self touch: deviceName. + self touch: sz. + self cCode: 'strncpy(deviceName, srcPtr, sz)'. + self cCode: 'deviceName[sz] = NULL'. + + "do the work" + self cCode: 'setDefaultSoundPlayer(deviceName)'. + self success ifTrue: "pop arg, leave receiver" + [interpreterProxy pop: 1]! Item was added: + ----- Method: SoundPlugin>>primitiveSetDefaultSoundRecorder (in category 'primitives') ----- + primitiveSetDefaultSoundRecorder + "Tell the operating system to use the specified device name as the input device for sound." + "arg at top of stack is the String" + | deviceName obj srcPtr sz | + <export: true> + <var: 'deviceName' declareC: 'char deviceName[257]'> + <var: 'srcPtr' type: #'char *'> + + "Parse arguments" + interpreterProxy methodArgumentCount = 1 ifFalse: + [^interpreterProxy primitiveFail]. + obj := interpreterProxy stackValue: 0. + (interpreterProxy isBytes: obj) ifFalse: + [^interpreterProxy primitiveFail]. + (sz := interpreterProxy byteSizeOf: obj) <= 256 ifFalse: + [^interpreterProxy primitiveFail]. + srcPtr := interpreterProxy firstIndexableField: obj. + self touch: srcPtr. + self touch: deviceName. + self touch: sz. + self cCode: 'strncpy(deviceName, srcPtr, sz)'. + self cCode: 'deviceName[sz] = NULL'. + + "do the work" + self cCode: 'setDefaultSoundRecorder(deviceName)'. + self success ifTrue: "pop arg, leave receiver" + [interpreterProxy pop: 1]! Item was added: + ----- Method: SoundPlugin>>primitiveSoundEnableAEC: (in category 'primitives') ----- + primitiveSoundEnableAEC: trueOrFalse + "Enable or disable acoustic echo-cancellation (AEC). trueOrFalse should be 0 for false, and 1 for true." + | result | + self primitive: 'primitiveSoundEnableAEC' parameters: #(SmallInteger ). + interpreterProxy failed ifFalse: [ + result := self cCode: 'snd_EnableAEC(trueOrFalse)'. + result == 0 ifFalse: [interpreterProxy primitiveFailFor: result]. + ].! Item was added: + ----- Method: SoundPlugin>>primitiveSoundGetRecordLevel (in category 'primitives') ----- + primitiveSoundGetRecordLevel + "Get the sound input recording level in the range 0-1000." + | level | + <var: 'level' type: #int> + self primitive: 'primitiveSoundGetRecordLevel'. + level := self cCode: 'snd_GetRecordLevel()'. + ^level asPositiveIntegerObj + ! Item was changed: ----- Method: SoundPlugin>>primitiveSoundGetVolume (in category 'primitives') ----- primitiveSoundGetVolume + "Get the sound input recording level." - "Set the sound input recording level." | left right results | <var: #left type: #double> <var: #right type: #double> self primitive: 'primitiveSoundGetVolume' parameters: #( ). left := 0. right := 0. self cCode: 'snd_Volume((double *) &left,(double *) &right)'. interpreterProxy pushRemappableOop: (right asOop: Float). interpreterProxy pushRemappableOop: (left asOop: Float). + results := interpreterProxy instantiateClass: (interpreterProxy classArray) indexableSize: 2. - interpreterProxy pushRemappableOop: (interpreterProxy instantiateClass: (interpreterProxy classArray) indexableSize: 2). - results := interpreterProxy popRemappableOop. interpreterProxy storePointer: 0 ofObject: results withValue: interpreterProxy popRemappableOop. interpreterProxy storePointer: 1 ofObject: results withValue: interpreterProxy popRemappableOop. + ^results! - ^ results! Item was added: + ----- Method: SoundPlugin>>primitiveSoundSupportsAEC (in category 'primitives') ----- + primitiveSoundSupportsAEC + "Answer true if the OS/hardware supports echo-cancellation, and false otherwise." + | result | + self primitive: 'primitiveSoundSupportsAEC'. + interpreterProxy failed ifFalse: [ + result := self cCode: 'snd_SupportsAEC()'. + result == 0 ifTrue: [^interpreterProxy falseObject] ifFalse: [^interpreterProxy trueObject] + ]. + !

1 0

Re: [Vm-dev] OSProcess fork issue with Debian built VM
by Max Leske 19 May '17

19 May '17

> On 19 May 2017, at 03:36, vm-dev-request(a)lists.squeakfoundation.org wrote: > > On Fri, May 19, 2017 at 2:58 AM, Eliot Miranda <eliot.miranda(a)gmail.com <mailto:eliot.miranda@gmail.com>> > wrote: >> >> BTW, for all people installing and building their own please note this: >> >> On Tue, May 16, 2017 at 5:57 AM, Max Leske <maxleske(a)gmail.com <mailto:maxleske@gmail.com>> wrote: >> [snip] >>> >>> /usr/local/lib/phcogspurlinux/lib/pharo/5.0-/pharo >>> Pharo VM version: 5.0-$URL$ Tue May 16 12:22:16 UTC 2017 gcc 4.9.2 > [Production Spur ITHB VM] >>> Built from: CoInterpreter VMMaker.oscog-eem.2207 uuid: > 8da5de9b-33d7-478b-9081-58591f7da69a May 16 2017 >>> With: StackToRegisterMappingCogit VMMaker.oscog-eem.2208 uuid: > 4877be7d-941d-4e15-b6df-4f1b8c7072a8 May 16 2017 >>> Revision: VM: $URL$ $Date$ Date: $Rev$ Plugins: $Rev$ >>> Build host: Linux nuc 4.10.0-20-generic #22-Ubuntu SMP Thu Apr 20 > 09:22:42 UTC 2017 i686 GNU/Linux >>> plugin path: /usr/local/lib/phcogspurlinux/lib/pharo/5.0-/ [default: > /usr/local/lib/phcogspurlinux/lib/pharo/5.0-/] >> >> >> The way to get rid of the " Revision: VM: $URL$ $Date$ Date: $Rev$ > Plugins: $Rev$" nonsense in the version report is to make sure to run >> scripts/updateSCCSVersions > > Even though I know this, I usually forget for the first run on a new > machine. > Is there some reason this can't be run as part of make? > > >> exactly once before you build. From then on the checkin script >> scrips/gitci >> makes sure that this is up-to-date. You'll get something more useful > such as >> >> VM: 201705161708 eliot@Sisyphus:oscogvm $ Date: Tue May 16 10:08:12 2017 > -0700 $ >> >> which allows me to identify this VM as one I build myself on my Mac Mini > Sisyphus in directory oscogvm, which contrasts with the official binary VMs. >> >> Ideally the script would be run automatically after the repository was > cloned. Do git mavens know if a post-clone hook is available to make this > happen? If so, can someone make it so? > > The blocker to this is that clone does not install the hooks. > Maybe the hooks be installed by make? > Of course, everyone should look at the README, but unfortunate human nature > its a bit TL;DR. > The alternative could be a SETUP or QUICKSTART file in the root directory, > to make it more obvious it should be run. > > > btw, could someone actually explain how this works... > git config --local include.path ../.gitconfig tell git to respect a specific .gitconfig file > git stash -q || true stash all changes, i.e. don't throw them away but make the repository clean > echo "//" >> platforms/Cross/vm/sqSCCSVersion.h > echo "//" >> platforms/Cross/plugins/sqPluginsSCCSVersion.h this appends a comment to both files, don't know why though, it doesn't seem to have any effect other than making those versions dirty > git checkout HEAD -- platforms/Cross/vm/sqSCCSVersion.h > git checkout HEAD -- platforms/Cross/plugins/sqPluginsSCCSVersion.h replace those two files just modified with the versions from HEAD (makes sense if the repository hasn't been pulled, but I don't see why that would be the case...) > git stash pop -q || true again, clean the repository be without throwing away the changes (so all the modifications we made before are now gone... wtf...) > I can't grok its magic. +1 :) > Hopefully that highlights why its the same script for each of > post-checkout, post-commit & post-merge > > > Also, at https://github.com/OpenSmalltalk/opensmalltalk-vm <https://github.com/OpenSmalltalk/opensmalltalk-vm> > if you click on the green <Clone or download> button and then on "Download > ZIP" > I found (from memory) that running mvm seems to successfully build a VM on > a Mac > (but I didn't validate the version info). However on Linux the build fails > in a way that seems similar to when > updateSCCSVersions hasn't been run. But that script depends on git, > whereas > I presume it would be good to be able to build from the ZIP download > without git - particularly > for Linux distributions that source compile their packages. > Indeed, I think Alistair bumped into this packing the VM for nixos. > > So I notice that this file... > https://github.com/OpenSmalltalk/opensmalltalk-vm/blob/Cog/platforms/Cross/… <https://github.com/OpenSmalltalk/opensmalltalk-vm/blob/Cog/platforms/Cross/…> > is dated 16 Dec 2016. Why isn't this file kept synchronised to identify > the checked in build? > So that compiling from the ZIP download would have correct version info? > > cheers -ben

1 0

VM Maker: VMMaker.oscog-nice.2216.mcz
by commits＠source.squeak.org 19 May '17

19 May '17

Nicolas Cellier uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-nice.2216.mcz ==================== Summary ==================== Name: VMMaker.oscog-nice.2216 Author: nice Time: 19 May 2017, 2:42:50.200889 pm UUID: e7100d55-b865-d843-8232-f14273653125 Ancestors: VMMaker.oscog-eem.2215 The code for restoring saved registers after a CallRT: did a benign iteration in excess. The register save part does store the highBit of the mask in reg. For example, for callerSavedRegsToBeSaved mask = 2r100, we will have reg = 3 after the save loop. The restore did start by inspecting bit 1<<reg, that is 2r1000, and that's one bit to far. This is benign as long as no overflow occurs, but let's be safer. =============== Diff against VMMaker.oscog-eem.2215 =============== Item was changed: ----- Method: Cogit>>CallFullRT:registersToBeSavedMask: (in category 'compile abstract instructions') ----- CallFullRT: callTarget registersToBeSavedMask: registersToBeSaved <returnTypeC: #'AbstractInstruction *'> | callerSavedRegsToBeSaved lastInst reg registersToBePushed | <var: 'lastInst' type: #'AbstractInstruction *'> callerSavedRegsToBeSaved := CallerSavedRegisterMask bitAnd: registersToBeSaved. backEnd canPushPopMultipleRegisters ifTrue: [backEnd genPushRegisterMask: callerSavedRegsToBeSaved] ifFalse: [registersToBePushed := callerSavedRegsToBeSaved. reg := 0. [registersToBePushed ~= 0] whileTrue: [(registersToBePushed anyMask: 1) ifTrue: [self PushR: reg]. reg := reg + 1. registersToBePushed := registersToBePushed >>> 1]]. lastInst := self CallFullRT: callTarget. backEnd canPushPopMultipleRegisters ifTrue: [^backEnd genPopRegisterMask: callerSavedRegsToBeSaved] ifFalse: + [[reg > 0] whileTrue: + [reg := reg - 1. + (callerSavedRegsToBeSaved anyMask: 1 << reg) ifTrue: + [lastInst := self PopR: reg]]. - [[reg >= 0] whileTrue: - [(callerSavedRegsToBeSaved anyMask: 1 << reg) ifTrue: - [lastInst := self PopR: reg]. - reg := reg - 1]. ^lastInst]! Item was changed: ----- Method: Cogit>>CallRT:registersToBeSavedMask: (in category 'compile abstract instructions') ----- CallRT: callTarget registersToBeSavedMask: registersToBeSaved <returnTypeC: #'AbstractInstruction *'> | callerSavedRegsToBeSaved lastInst reg registersToBePushed | <var: 'lastInst' type: #'AbstractInstruction *'> callerSavedRegsToBeSaved := CallerSavedRegisterMask bitAnd: registersToBeSaved. backEnd canPushPopMultipleRegisters ifTrue: [backEnd genPushRegisterMask: callerSavedRegsToBeSaved] ifFalse: [registersToBePushed := callerSavedRegsToBeSaved. reg := 0. [registersToBePushed ~= 0] whileTrue: [(registersToBePushed anyMask: 1) ifTrue: [self PushR: reg]. reg := reg + 1. registersToBePushed := registersToBePushed >>> 1]]. lastInst := self CallRT: callTarget. backEnd canPushPopMultipleRegisters ifTrue: [^backEnd genPopRegisterMask: callerSavedRegsToBeSaved] ifFalse: + [[reg > 0] whileTrue: + [reg := reg - 1. + (callerSavedRegsToBeSaved anyMask: 1 << reg) ifTrue: + [lastInst := self PopR: reg]]. - [[reg >= 0] whileTrue: - [(callerSavedRegsToBeSaved anyMask: 1 << reg) ifTrue: - [lastInst := self PopR: reg]. - reg := reg - 1]. ^lastInst]!

1 0

Build Version info (was Re: OSProcess fork issue with Debian built VM)
by Ben Coman 19 May '17

19 May '17

On Fri, May 19, 2017 at 2:58 AM, Eliot Miranda <eliot.miranda(a)gmail.com> wrote: > > BTW, for all people installing and building their own please note this: > > On Tue, May 16, 2017 at 5:57 AM, Max Leske <maxleske(a)gmail.com> wrote: > [snip] >> >> /usr/local/lib/phcogspurlinux/lib/pharo/5.0-/pharo >> Pharo VM version: 5.0-$URL$ Tue May 16 12:22:16 UTC 2017 gcc 4.9.2 [Production Spur ITHB VM] >> Built from: CoInterpreter VMMaker.oscog-eem.2207 uuid: 8da5de9b-33d7-478b-9081-58591f7da69a May 16 2017 >> With: StackToRegisterMappingCogit VMMaker.oscog-eem.2208 uuid: 4877be7d-941d-4e15-b6df-4f1b8c7072a8 May 16 2017 >> Revision: VM: $URL$ $Date$ Date: $Rev$ Plugins: $Rev$ >> Build host: Linux nuc 4.10.0-20-generic #22-Ubuntu SMP Thu Apr 20 09:22:42 UTC 2017 i686 GNU/Linux >> plugin path: /usr/local/lib/phcogspurlinux/lib/pharo/5.0-/ [default: /usr/local/lib/phcogspurlinux/lib/pharo/5.0-/] > > > The way to get rid of the " Revision: VM: $URL$ $Date$ Date: $Rev$ Plugins: $Rev$" nonsense in the version report is to make sure to run > scripts/updateSCCSVersions Even though I know this, I usually forget for the first run on a new machine. Is there some reason this can't be run as part of make? > exactly once before you build. From then on the checkin script > scrips/gitci > makes sure that this is up-to-date. You'll get something more useful such as > > VM: 201705161708 eliot@Sisyphus:oscogvm $ Date: Tue May 16 10:08:12 2017 -0700 $ > > which allows me to identify this VM as one I build myself on my Mac Mini Sisyphus in directory oscogvm, which contrasts with the official binary VMs. > > Ideally the script would be run automatically after the repository was cloned. Do git mavens know if a post-clone hook is available to make this happen? If so, can someone make it so? The blocker to this is that clone does not install the hooks. Maybe the hooks be installed by make? Of course, everyone should look at the README, but unfortunate human nature its a bit TL;DR. The alternative could be a SETUP or QUICKSTART file in the root directory, to make it more obvious it should be run. btw, could someone actually explain how this works... git config --local include.path ../.gitconfig git stash -q || true echo "//" >> platforms/Cross/vm/sqSCCSVersion.h echo "//" >> platforms/Cross/plugins/sqPluginsSCCSVersion.h git checkout HEAD -- platforms/Cross/vm/sqSCCSVersion.h git checkout HEAD -- platforms/Cross/plugins/sqPluginsSCCSVersion.h git stash pop -q || true I can't grok its magic. Hopefully that highlights why its the same script for each of post-checkout, post-commit & post-merge Also, at https://github.com/OpenSmalltalk/opensmalltalk-vm if you click on the green <Clone or download> button and then on "Download ZIP" I found (from memory) that running mvm seems to successfully build a VM on a Mac (but I didn't validate the version info). However on Linux the build fails in a way that seems similar to when updateSCCSVersions hasn't been run. But that script depends on git, whereas I presume it would be good to be able to build from the ZIP download without git - particularly for Linux distributions that source compile their packages. Indeed, I think Alistair bumped into this packing the VM for nixos. So I notice that this file... https://github.com/OpenSmalltalk/opensmalltalk-vm/blob/Cog/platforms/Cross/… is dated 16 Dec 2016. Why isn't this file kept synchronised to identify the checked in build? So that compiling from the ZIP download would have correct version info? cheers -ben

1 0

64bits Pharo VM for windows
by Nicolas Cellier 19 May '17

19 May '17

Hi, I've built a 64bits pharo.stack.spur VM for windows on my machine, and I'm uploading the changes to opensmalltalk-vm in branch build_pharo_win32_with_cygwin If the appveyor job correctly succeed, I will emit a pull request. The VM does not have the SqueakSSL plugin yet. The 64bits squeak/pharo.cog.spur JIT for windows is still to come, but I did not work on it for a few months... One thing at a time. Let's cross finger Nicolas

4 18

VM Maker: VMMaker.oscog-eem.2215.mcz
by commits＠source.squeak.org 18 May '17

18 May '17

Eliot Miranda uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-eem.2215.mcz ==================== Summary ==================== Name: VMMaker.oscog-eem.2215 Author: eem Time: 18 May 2017, 9:20:11.128923 am UUID: 832c8202-ffb4-43ed-9f41-9c6d2e2bf80c Ancestors: VMMaker.oscog-nice.2214 Slang/FilePlugin: Avoid remapping oops on creating directory entries in the FilePlugin on Spur by using remapOop:in:. Change Slang to avoid creating nested #if-#else-#endif structures for nested remapOop:in:'s. Nuke an unused configuration. Spur: Never inline objectsReachableFromRoots:. Fix a speeling rorre. =============== Diff against VMMaker.oscog-nice.2214 =============== Item was changed: ----- Method: FilePlugin>>makeDirEntryName:size:createDate:modDate:isDir:fileSize: (in category 'directory primitives') ----- makeDirEntryName: entryName size: entryNameSize createDate: createDate modDate: modifiedDate + isDir: dirFlag fileSize: fileSize + <var: 'entryName' type: #'char *'> + <var: 'fileSize' type: #squeakFileOffsetType> - isDir: dirFlag fileSize: fileSize - | modDateOop createDateOop nameString results stringPtr fileSizeOop | + <var: 'stringPtr' type: #'char *'> - <var: 'entryName' type: 'char *'> - <var: 'stringPtr' type:'char *'> - <var: 'fileSize' type:'squeakFileOffsetType '> "allocate storage for results, remapping newly allocated + oops if required in case GC happens during allocation" + results := interpreterProxy instantiateClass: interpreterProxy classArray indexableSize: 5. + self remapOop: results in: + [nameString := interpreterProxy instantiateClass: interpreterProxy classString indexableSize: entryNameSize. + self remapOop: nameString in: + [createDateOop := interpreterProxy positive32BitIntegerFor: createDate. + self remapOop: createDateOop in: + [modDateOop := interpreterProxy positive32BitIntegerFor: modifiedDate. + self remapOop: modDateOop in: + [fileSizeOop := interpreterProxy positive64BitIntegerFor: fileSize]]]]. - oops in case GC happens during allocation" - interpreterProxy pushRemappableOop: - (interpreterProxy instantiateClass: (interpreterProxy classString) indexableSize: entryNameSize). - interpreterProxy pushRemappableOop: - (interpreterProxy positive32BitIntegerFor: createDate). - interpreterProxy pushRemappableOop: - (interpreterProxy positive32BitIntegerFor: modifiedDate). - interpreterProxy pushRemappableOop: - (interpreterProxy positive64BitIntegerFor: fileSize). - results := interpreterProxy instantiateClass: (interpreterProxy classArray) indexableSize: 5. - - fileSizeOop := interpreterProxy popRemappableOop. - modDateOop := interpreterProxy popRemappableOop. - createDateOop := interpreterProxy popRemappableOop. - nameString := interpreterProxy popRemappableOop. - "copy name into Smalltalk string" stringPtr := interpreterProxy firstIndexableField: nameString. 0 to: entryNameSize - 1 do: [ :i | self cCode: [stringPtr at: i put: (entryName at: i)] inSmalltalk: [interpreterProxy storeByte: i ofObject: nameString withValue: (entryName at: i+1) asciiValue]]. interpreterProxy storePointer: 0 ofObject: results withValue: nameString; storePointer: 1 ofObject: results withValue: createDateOop; storePointer: 2 ofObject: results withValue: modDateOop; storePointer: 3 ofObject: results withValue: (dirFlag ifTrue: [interpreterProxy trueObject] ifFalse: [interpreterProxy falseObject]); storePointer: 4 ofObject: results withValue: fileSizeOop. ^results! Item was changed: ----- Method: FilePlugin>>makeDirEntryName:size:createDate:modDate:isDir:fileSize:posixPermissions:isSymlink: (in category 'directory primitives') ----- makeDirEntryName: entryName size: entryNameSize createDate: createDate modDate: modifiedDate isDir: dirFlag fileSize: fileSize posixPermissions: posixPermissions isSymlink: symlinkFlag + <var: 'entryName' type: #'char *'> + <var: 'fileSize' type: #squeakFileOffsetType> - <var: 'entryName' type: 'char *'> - <var: 'fileSize' type: 'squeakFileOffsetType '> <option: #PharoVM> - | modDateOop createDateOop nameString results stringPtr posixPermissionsOop fileSizeOop | + <var: 'stringPtr' type: #'char *'> - <var: 'stringPtr' type: 'char *'> "allocate storage for results, remapping newly allocated + oops if required in case GC happens during allocation" + results := interpreterProxy instantiateClass: interpreterProxy classArray indexableSize: 7. + self remapOop: results in: + [nameString := interpreterProxy instantiateClass: interpreterProxy classString indexableSize: entryNameSize. + self remapOop: nameString in: + [createDateOop := interpreterProxy positive32BitIntegerFor: createDate. + self remapOop: createDateOop in: + [modDateOop := interpreterProxy positive32BitIntegerFor: modifiedDate. + self remapOop: modDateOop in: + [fileSizeOop := interpreterProxy positive64BitIntegerFor: fileSize. + self remapOop: fileSizeOop in: + [posixPermissionsOop := interpreterProxy positive32BitIntegerFor: posixPermissions]]]]]. - oops in case GC happens during allocation" - interpreterProxy pushRemappableOop: - (interpreterProxy instantiateClass: (interpreterProxy classArray) indexableSize: 7). - interpreterProxy pushRemappableOop: - (interpreterProxy instantiateClass: (interpreterProxy classString) indexableSize: entryNameSize). - interpreterProxy pushRemappableOop: - (interpreterProxy positive32BitIntegerFor: createDate). - interpreterProxy pushRemappableOop: - (interpreterProxy positive32BitIntegerFor: modifiedDate). - interpreterProxy pushRemappableOop: - (interpreterProxy positive64BitIntegerFor: fileSize). - interpreterProxy pushRemappableOop: - (interpreterProxy positive32BitIntegerFor: posixPermissions). - posixPermissionsOop := interpreterProxy popRemappableOop. - fileSizeOop := interpreterProxy popRemappableOop. - modDateOop := interpreterProxy popRemappableOop. - createDateOop := interpreterProxy popRemappableOop. - nameString := interpreterProxy popRemappableOop. - results := interpreterProxy popRemappableOop. - "copy name into Smalltalk string" stringPtr := interpreterProxy firstIndexableField: nameString. + 0 to: entryNameSize - 1 do: + [ :i | + self cCode: [stringPtr at: i put: (entryName at: i)] + inSmalltalk: [interpreterProxy storeByte: i ofObject: nameString withValue: (entryName at: i+1) asciiValue]]. - 0 to: entryNameSize - 1 do: [ :i | - stringPtr at: i put: (entryName at: i). - ]. + interpreterProxy + storePointer: 0 ofObject: results withValue: nameString; + storePointer: 1 ofObject: results withValue: createDateOop; + storePointer: 2 ofObject: results withValue: modDateOop; + storePointer: 3 ofObject: results withValue: (dirFlag + ifTrue: [interpreterProxy trueObject] + ifFalse: [interpreterProxy falseObject]); + storePointer: 4 ofObject: results withValue: fileSizeOop; + storePointer: 5 ofObject: results withValue: posixPermissionsOop; + storePointer: 6 ofObject: results withValue: (symlinkFlag + ifTrue: [interpreterProxy trueObject] + ifFalse: [interpreterProxy falseObject]). + ^results! - interpreterProxy storePointer: 0 ofObject: results withValue: nameString. - interpreterProxy storePointer: 1 ofObject: results withValue: createDateOop. - interpreterProxy storePointer: 2 ofObject: results withValue: modDateOop. - dirFlag - ifTrue: [ interpreterProxy storePointer: 3 ofObject: results withValue: interpreterProxy trueObject ] - ifFalse: [ interpreterProxy storePointer: 3 ofObject: results withValue: interpreterProxy falseObject ]. - interpreterProxy storePointer: 4 ofObject: results withValue: fileSizeOop. - interpreterProxy storePointer: 5 ofObject: results withValue: posixPermissionsOop. - symlinkFlag - ifTrue: [ interpreterProxy storePointer: 6 ofObject: results withValue: interpreterProxy trueObject ] - ifFalse: [ interpreterProxy storePointer: 6 ofObject: results withValue: interpreterProxy falseObject ]. - ^ results! Item was changed: ----- Method: SpurMemoryManager>>objectsReachableFromRoots: (in category 'image segment in/out') ----- objectsReachableFromRoots: arrayOfRoots "This is part of storeImageSegmentInto:outPointers:roots:. Answer an Array of all the objects only reachable from the argument, an Array of root objects, starting with arrayOfRoots. If there is no space, answer a SmallInteger whose value is the number of slots required. This is used to collect the objects to include in an image segment on Spur, separate from creating the segment, hence simplifying the implementation. Thanks to Igor Stasenko for this idea." | freeChunk ptr start limit count oop objOop | + <inline: #never> self assert: (self isArray: arrayOfRoots). "Mark all objects except those only reachable from the arrayOfRoots by marking each object in arrayOfRoots and then marking all reachable objects (from the system roots). This leaves unmarked only objects reachable from the arrayOfRoots. N.B. A side-effect of the marking is that all forwarders in arrayOfRoots will be followed." self assert: self allObjectsUnmarked. self markObjectsIn: arrayOfRoots. self markObjects: false. "After the mark phase all unreachable weak slots will have been nilled and all active ephemerons fired." self assert: (self isEmptyObjStack: markStack). self assert: (self isEmptyObjStack: weaklingStack). self assert: self noUnscannedEphemerons. "Use the largest free chunk to answer the result." freeChunk := self allocateLargestFreeChunk. ptr := start := freeChunk + self baseHeaderSize. limit := self addressAfter: freeChunk. count := 0. "First put the arrayOfRoots; order is important." count := count + 1. ptr < limit ifTrue: [self longAt: ptr put: arrayOfRoots. ptr := ptr + self bytesPerOop]. 0 to: (self numSlotsOf: arrayOfRoots) - 1 do: [:i| oop := self fetchPointer: i ofObject: arrayOfRoots. (self isNonImmediate: oop) ifTrue: [self noCheckPush: oop onObjStack: markStack]]. "Now collect the unmarked objects reachable from the roots." [self isEmptyObjStack: markStack] whileFalse: [objOop := self popObjStack: markStack. count := count + 1. ptr < limit ifTrue: [self longAt: ptr put: objOop. ptr := ptr + self bytesPerOop]. oop := self fetchClassOfNonImm: objOop. (self isMarked: oop) ifFalse: [self setIsMarkedOf: objOop to: true. self noCheckPush: oop onObjStack: markStack]. ((self isContextNonImm: objOop) and: [coInterpreter isStillMarriedContext: objOop]) "widow now, before the loop" ifTrue: [0 to: (coInterpreter numSlotsOfMarriedContext: objOop) - 1 do: [:i| oop := coInterpreter fetchPointer: i ofMarriedContext: objOop. ((self isImmediate: oop) or: [self isMarked: oop]) ifFalse: [self setIsMarkedOf: objOop to: true. self noCheckPush: oop onObjStack: markStack]]] ifFalse: [0 to: (self numPointerSlotsOf: objOop) - 1 do: [:i| oop := self fetchPointer: i ofObject: objOop. ((self isImmediate: oop) or: [self isMarked: oop]) ifFalse: [self setIsMarkedOf: objOop to: true. self noCheckPush: oop onObjStack: markStack]]]]. self unmarkAllObjects. totalFreeOldSpace := totalFreeOldSpace - (self bytesInObject: freeChunk). "Now try and allocate the result" (count > (ptr - start / self bytesPerOop) "not enough room" or: [limit ~= ptr and: [limit - ptr <= self allocationUnit]]) ifTrue: "can't split a single word" [self freeChunkWithBytes: (self bytesInObject: freeChunk) at: (self startOfObject: freeChunk). self checkFreeSpace: GCModeImageSegment. ^self integerObjectOf: count]. "There's room; set the format, & classIndex and shorten." self setFormatOf: freeChunk to: self arrayFormat. self setClassIndexOf: freeChunk to: ClassArrayCompactIndex. self shorten: freeChunk toIndexableSize: count. (self isForwarded: freeChunk) ifTrue: [freeChunk := self followForwarded: freeChunk]. self possibleRootStoreInto: freeChunk. self checkFreeSpace: GCModeImageSegment. self runLeakCheckerFor: GCModeImageSegment. ^freeChunk! Item was changed: ----- Method: SpurMemoryManager>>storeImageSegmentInto:outPointers:roots: (in category 'image segment in/out') ----- storeImageSegmentInto: segmentWordArrayArg outPointers: outPointerArrayArg roots: arrayOfRootsArg "This primitive is called from Squeak as... <imageSegment> storeSegmentFor: arrayOfRoots into: aWordArray outPointers: anArray. This primitive will store a binary image segment (in the same format as objects in the heap) of the set of objects in arrayOfObjects. All pointers from within the set to objects outside the set will be copied into the array of outPointers. In their place in the image segment will be an oop equal to the offset in the outPointer array (the first would be 8), but with the high bit set. Since Spur has a class table the load primitive must insert classes that have instances into the class table. This primitive marks such classes using the isRemembered bit, which isn't meaningful as a remembered bit in the segment. The primitive expects the segmentWordArray and outPointerArray to be more than adequately long. In this case it returns normally, and truncates the two arrays to exactly the right size. The primitive can fail for the following reasons with the specified failure codes: PrimErrGenericError: the segmentWordArray is too small for the version stamp PrimErrWritePastObject: the segmentWordArray is too small to contain the reachable objects PrimErrBadIndex: the outPointerArray is too small PrimErrNoMemory: additional allocations failed PrimErrLimitExceeded: there is no room in the hash field to store out pointer indices or class references." <inline: false> | segmentWordArray outPointerArray arrayOfRoots arrayOfObjects savedFirstFields savedOutHashes segStart segAddr endSeg outIndex numClassesInSegment | ((self isObjImmutable: segmentWordArrayArg) or: [self isObjImmutable: outPointerArrayArg]) ifTrue: [^PrimErrNoModification]. "Since segmentWordArrayArg & outPointerArrayArg may get shortened, they can't be pinned." ((self isPinned: segmentWordArrayArg) or: [self isPinned: outPointerArrayArg]) ifTrue: [^PrimErrObjectIsPinned]. (self numSlotsOf: outPointerArrayArg) > self maxIdentityHash ifTrue: [^PrimErrLimitExceeded]. self runLeakCheckerFor: GCModeImageSegment. + "First scavenge to collect any new space garbage that refers to the graph." - "First scavenge to coillect any new space garbage that refers to the graph." self scavengingGC. segmentWordArray := self updatePostScavenge: segmentWordArrayArg. outPointerArray := self updatePostScavenge: outPointerArrayArg. arrayOfRoots := self updatePostScavenge: arrayOfRootsArg. "Now compute the transitive closure, collecting the sequence of objects to be stored in the arrayOfObjects array. Included in arrayOfObjects are the arrayOfRoots and all its contents. All objects have been unmarked." arrayOfObjects := self objectsReachableFromRoots: arrayOfRoots. arrayOfObjects ifNil: [^PrimErrNoMemory]. self assert: self allObjectsUnmarked. "work to be done when the incremental GC is written" "Both to expand the max size of segment and to reduce the length of the load-time pass that adds classes to the class table, move classes to the front of arrayOfObjects, leaving the root array as the first element." numClassesInSegment := self moveClassesForwardsIn: arrayOfObjects. "The scheme is to copy the objects into segmentWordArray, and then map the oops in segmentWordArray. Therefore the primitive needs to both map efficiently originals to copies in segmentWordArray and be able to undo any side-effects if the primitive fails because either segmentWordArray or outPointerArray is too small. The mapping is done by having the objects to be stored in arrayOfObjects refer to their mapped locations through their first field, just like a forwarding pointer, but without becoming a forwarder, saving their first field in savedFirstFields, and the objects in outPointerArray pointing to tehir locations in the outPointerArray through their identityHashes, saved in savedOutHashes. Since arrayOfObjects and its savedFirstFields, and outPointerArray and its saved hashes, can be enumerated side-by-side, the hashes can be restored to the originals. So the hash of an object in arrayOfObjects is set to its offset in segmentWordArray / self allocationUnit, and the hash of an object in outPointerArray is set to its index in outPointerArray plus the top hash bit. Oops in segmentWordArray are therefore mapped by accessing the original oop's identityHash, testing the bottom bit to distinguish between internal and external oops. The saved hash arrays are initialized with an out-of-range hash value so that the first unused entry can be identified." savedFirstFields := self allocateSlots: (self numSlotsOf: arrayOfObjects) format: self wordIndexableFormat classIndex: self wordSizeClassIndexPun. savedOutHashes := self allocateSlots: (self numSlotsForBytes: (self numSlotsOf: outPointerArray) * 4) format: self firstLongFormat classIndex: self thirtyTwoBitLongsClassIndexPun. (savedFirstFields isNil or: [savedOutHashes isNil]) ifTrue: [self freeObject: arrayOfObjects. ^PrimErrNoMemory]. self fillObj: savedFirstFields numSlots: (self numSlotsOf: savedFirstFields) with: 0. self fillObj: savedOutHashes numSlots: (self numSlotsOf: savedOutHashes) with: self savedOutHashFillValue. segAddr := segmentWordArray + self baseHeaderSize. endSeg := self addressAfter: segmentWordArray. "Write a version number for byte order and version check." segAddr >= endSeg ifTrue: [^PrimErrGenericFailure]. self long32At: segAddr put: self imageSegmentVersion. self long32At: segAddr + 4 put: self imageSegmentVersion. segStart := segAddr := segAddr + self allocationUnit. "Copy all reachable objects to the segment, setting the marked bit for all objects (clones) in the segment, and the remembered bit for all classes (clones) in the segment." 0 to: (self numSlotsOf: arrayOfObjects) - 1 do: [:i| | newSegAddrOrError objOop | "Check that classes in the segment are addressible. Since the top bit of the hash field is used to tag classes external to the segment, the segment offset must not inadvertently set this bit. This limit still allows for a million or more classes." (i = numClassesInSegment and: [segAddr - segStart / self allocationUnit + self lastClassIndexPun >= TopHashBit]) ifTrue: [^self return: PrimErrLimitExceeded restoringObjectsIn: arrayOfObjects upTo: i savedFirstFields: savedFirstFields and: outPointerArray savedHashes: savedOutHashes]. objOop := self fetchPointer: i ofObject: arrayOfObjects. self deny: ((self isImmediate: objOop) or: [self isForwarded: objOop]). newSegAddrOrError := self copyObj: objOop toAddr: segAddr startAt: segStart stopAt: endSeg savedFirstFields: savedFirstFields index: i. newSegAddrOrError < segStart ifTrue: [^self return: newSegAddrOrError restoringObjectsIn: arrayOfObjects upTo: i savedFirstFields: savedFirstFields and: outPointerArray savedHashes: savedOutHashes]. segAddr := newSegAddrOrError]. "Check that it can be safely shortened." (endSeg ~= segAddr and: [endSeg - segAddr < (self baseHeaderSize + self bytesPerOop)]) ifTrue: [^self return: PrimErrWritePastObject restoringObjectsIn: arrayOfObjects upTo: -1 savedFirstFields: savedFirstFields and: outPointerArray savedHashes: savedOutHashes]. "Now scan, adding out pointers to the outPointersArray; all objects in arrayOfObjects have their first field pointing to the corresponding copy in segmentWordArray." (outIndex := self mapOopsFrom: segStart to: segAddr outPointers: outPointerArray outHashes: savedOutHashes) < 0 ifTrue: "no room in outPointers; fail" [^self return: PrimErrBadIndex restoringObjectsIn: arrayOfObjects upTo: -1 savedFirstFields: savedFirstFields and: outPointerArray savedHashes: savedOutHashes]. "We're done. Shorten the results, restore hashes and return." self shorten: segmentWordArray toIndexableSize: segAddr - (segmentWordArray + self baseHeaderSize) / 4. self shorten: outPointerArray toIndexableSize: outIndex. ^self return: PrimNoErr restoringObjectsIn: arrayOfObjects upTo: -1 savedFirstFields: savedFirstFields and: outPointerArray savedHashes: savedOutHashes! Item was removed: - ----- Method: VMMaker class>>generateSqueakSpurStackSistaVM (in category 'configurations') ----- - generateSqueakSpurStackSistaVM - "No primitives since we can use those for the Cog VM" - ^VMMaker - generate: StackInterpreter - with: #(SistaVM true - ObjectMemory Spur32BitMemoryManager - FailImbalancedPrimitives false - MULTIPLEBYTECODESETS true - bytecodeTableInitializer initializeBytecodeTableForSqueakV3PlusClosuresSistaV1Hybrid) - to: (FileDirectory default pathFromURI: self sourceTree, '/spursistastacksrc') - platformDir: (FileDirectory default pathFromURI: self sourceTree, '/platforms') - including:#()! Item was changed: ----- Method: VMPluginCodeGenerator>>generateRemapOopIn:on:indent: (in category 'C translation') ----- generateRemapOopIn: aNode on: aStream indent: level "Generate the C code for this message onto the given stream." + | arm | + "Avoid nesting #if SPURVM...#else...#endif within arms of an outer #if SPURVM...#else...#endif." + (Notification new tag: #inRemapOopInArm; signal) ifNotNil: + [:inRemapOopInArm| + ^inRemapOopInArm + ifTrue: [self generateSpurRemapOopIn: aNode on: aStream indent: level] + ifFalse: [self generateV3RemapOopIn: aNode on: aStream indent: level]]. + [aStream cr; nextPutAll: '#if SPURVM'; cr. + arm := true. + self generateSpurRemapOopIn: aNode on: aStream indent: level. + aStream cr; nextPutAll: '#else /* SPURVM */'; cr. + arm := false. + self generateV3RemapOopIn: aNode on: aStream indent: level. + aStream cr; nextPutAll: '#endif /* SPURVM */'; cr] + on: Notification + do: [:ex| + ex tag == #inRemapOopInArm + ifTrue: [ex resume: arm] + ifFalse: [ex pass]]! - - aStream cr; nextPutAll: '#if SPURVM'; cr. - self generateSpurRemapOopIn: aNode on: aStream indent: level. - aStream cr; nextPutAll: '#else /* SPURVM */'; cr. - self generateV3RemapOopIn: aNode on: aStream indent: level. - aStream cr; nextPutAll: '#endif /* SPURVM */'; cr!

1 0

VM Maker: VMMaker.oscog-nice.2214.mcz
by commits＠source.squeak.org 18 May '17

18 May '17

Nicolas Cellier uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-nice.2214.mcz ==================== Summary ==================== Name: VMMaker.oscog-nice.2214 Author: nice Time: 18 May 2017, 5:38:43.506121 pm UUID: 9b032b43-4355-954e-9c08-66df1f3d8138 Ancestors: VMMaker.oscog-eem.2213 Use a special form of sigsetjmp to avoid stack unwinding on Windows 64. Note 1: stack unwinding is for C++ exception handling (for example for calling destructors). We do not need it, and it requires the presence of specific information on stack which is not honoured by our JIT frames. Note 2: this patch is specific to Mingw-w64. It might not work for another compiler (MSVC/Intel/...). Someone making the Vm compile with other tools might have to amend this #preambleCCode again. In mingw, there is a possibility to pass a 2nd parameter to a _setjmp, and if this 2nd argument is NULL, there will be no context unwinding. This knowledge is extracted from these references: http://mingw-w64-public.narkive.com/1mUoWEfG/setjmp-longjmp-crashes-second-… https://patchwork.ozlabs.org/patch/437794/ http://www.agardner.me/golang/windows/cgo/64-bit/setjmp/longjmp/2016/02/29/… https://sourceforge.net/p/mingw-w64/bugs/465/ At current time of writing, _setjmp has 2 arguments #ifndef USE_NO_MINGW_SETJMP_TWO_ARGS and no header defines that by default: grep -r USE_NO_MINGW_SETJMP_TWO_ARGS /usr/x86_64-w64-mingw32/sys-root/mingw/include =============== Diff against VMMaker.oscog-eem.2213 =============== Item was changed: ----- Method: StackInterpreter class>>preambleCCode (in category 'translation') ----- preambleCCode ^ '/* Disable Intel compiler inlining of warning which is used for breakpoints */ #pragma auto_inline(off) sqInt warnpid, erroronwarn; void warning(char *s) { /* Print an error message but don''t necessarily exit. */ if (erroronwarn) error(s); if (warnpid) printf("\n%s pid %ld\n", s, (long)warnpid); else printf("\n%s\n", s); } void warningat(char *s, int l) { /* ditto with line number. */ /* use alloca to call warning so one does not have to remember to set two breakpoints... */ char *sl = alloca(strlen(s) + 16); sprintf(sl, "%s %d", s, l); warning(sl); } #pragma auto_inline(on) void invalidCompactClassError(char *s) { /* Print a (compact) class index error message and exit. */ #if SPURVM printf("\nClass %s does not have the required class index\n", s); #else printf("\nClass %s does not have the required compact class index\n", s); #endif exit(-1); } /* * Define sigsetjmp and siglongjmp to be the most minimal setjmp/longjmp available on the platform. + * Note: on windows 64 via mingw-w64, the 2nd argument NULL to _setjmp prevents stack unwinding */ #undef sigsetjmp #undef siglongjmp + #if _WIN64 + # define sigsetjmp(jb,ssmf) _setjmp(jb,NULL) + # define siglongjmp(jb,v) longjmp(jb,v) + #elif _WIN32 - #if _WIN32 # define sigsetjmp(jb,ssmf) setjmp(jb) # define siglongjmp(jb,v) longjmp(jb,v) #else # define sigsetjmp(jb,ssmf) _setjmp(jb) # define siglongjmp(jb,v) _longjmp(jb,v) #endif #define odd(v) ((int)(v)&1) #define even(v) (!!odd(v)) '!

1 0

Re: [Vm-dev] [OpenSmalltalk/opensmalltalk-vm] try appveyor (bed86c5)
by Eliot Miranda 18 May '17

18 May '17

Hi Nicolas, can you give a short status report on Cog for Win64? What is not working for the JIT? Given that the stack spur vm works we should be close. So what still needs fixing? _,,,^..^,,,_ (phone) > On May 15, 2017, at 7:13 AM, Nicolas Cellier <notifications(a)github.com> wrote: > > Hi Esteban, currently this fails because with don't build pharo.cog.spur yet in win64. > Could you retry with "${ROOT_DIR}/build.${ARCH}/pharo.stack.spur/build/vm"? > At least, this should turn the appveyor status green again. > > — > You are receiving this because you are subscribed to this thread. > Reply to this email directly, view it on GitHub, or mute the thread. >

2 9

Re: [Vm-dev] OSProcess fork issue with Debian built VM
by Max Leske 18 May '17

18 May '17

> On 18 May 2017, at 00:50, vm-dev-request(a)lists.squeakfoundation.org wrote: > > Hi Max, Hi Alistair, > > On Wed, May 17, 2017 at 1:06 AM, Alistair Grant <akgrant0710(a)gmail.com <mailto:akgrant0710@gmail.com>> > wrote: > >> >> On Tue, May 16, 2017 at 04:59:24PM +0200, Alistair Grant wrote: >>> Hi Max, >>> >>> On 16 May 2017 15:40, "Max Leske" <maxleske(a)gmail.com <mailto:maxleske@gmail.com>> wrote: >>> >>> Hi Alistair, >>> >>> On 16 May 2017, at 15:32, vm-dev-request@lists. >> squeakfoundation.org <http://squeakfoundation.org/> >>> wrote: >>> >>> Hi Max, >>> >>> I can't answer your question directly, but just wondering why >> you are >>> using >>> the itimer VM when the are known issues with external calls, and >> not >>> the >>> heartbeat VM? >>> >>> Because of the root user issue, and also because I don't care about >> that >>> much at the moment. I'm still experimenting and for those >> experiments it >>> doesn't matter which VM I use. Thirdly, the itimer VM is the one I >> get when >>> I use 'curl get.pharo.org/60+vmLatest <http://get.pharo.org/60+vmLatest> | bash', which is convenient >> to get >>> the latest VM, and to minimise differences between the VM's we built >> the >>> same one. I will definitely consider using the threaded VM for >> production. >>> >>> P.S. I would love to see OSProcess working in 32 bit mode. >>> >>> Well, it does work already, just not when we build the VM ourselves >> :/ >>> >>> Interesting, I had the impression that for Pharo 6 OSProcess didn't work >> in >>> 32bits, only 64, but I'm also building my own VM. I'm away from my PC, >> but >>> I'll try and take a look. >> >> I'm seeing the same behaviour as you, i.e. OSProcess works in a VM >> downloaded from get.pharo.org <http://get.pharo.org/>, but locks up when using the VM I >> compiled. >> > > Have you looked at the build logs and eliminated compiler version, command > line flags, etc? One important file is the config.h that is produced in > the build directory. It might be informative to compare the one configure > is producing on your systems and the one that the binary builds creates. Thanks for the pointer. I'll look into it. > > >> >> Both VMs (threaded heartbeat) are based on the same source code, i.e.: >> >> VM: 201705022326 https://github.com/OpenSmalltalk/opensmalltalk-vm.git <https://github.com/OpenSmalltalk/opensmalltalk-vm.git> $ >> Date: Tue May 2 16:26:41 2017 -0700 $ >> >> I'll try and take a look at this eventually, but I'm not sure how long >> that will be (several weeks away, at least). >> >> If you figure it out, please let me know. >> >> Thanks! >> Alistair >> > > > > -- > _,,,^..^,,,_ > best, Eliot

1 0

[OpenSmalltalk/opensmalltalk-vm] ca7e8d: Change platform sources to refer to _WIN32 & _WIN6...
by GitHub 18 May '17

18 May '17

Branch: refs/heads/Cog Home: https://github.com/OpenSmalltalk/opensmalltalk-vm Commit: ca7e8db32835f5d2489ebddc7e14a51c9bbd4697 https://github.com/OpenSmalltalk/opensmalltalk-vm/commit/ca7e8db32835f5d248… Author: Eliot Miranda <eliot.miranda(a)gmail.com> Date: 2017-05-16 (Tue, 16 May 2017) Changed paths: M platforms/Cross/plugins/B3DAcceleratorPlugin/B3DAcceleratorPlugin.h M platforms/Cross/plugins/B3DAcceleratorPlugin/sqOpenGLRenderer.c M platforms/Cross/plugins/B3DAcceleratorPlugin/sqOpenGLRenderer.h M platforms/Cross/plugins/BochsIA32Plugin/sqBochsIA32Plugin.cpp M platforms/Cross/plugins/BochsX64Plugin/sqBochsX64Plugin.cpp M platforms/Cross/plugins/IA32ABI/ia32abi.h M platforms/Cross/plugins/IA32ABI/ia32abicc.c M platforms/Cross/plugins/IA32ABI/x64win64stub.c M platforms/Cross/plugins/IA32ABI/xabicc.c M platforms/Cross/plugins/Mpeg3Plugin/libmpeg/changesForSqueak.c M platforms/Cross/plugins/Mpeg3Plugin/libmpeg/mpeg3io.h M platforms/Cross/plugins/SqueakFFIPrims/sqFFIPlugin.c M platforms/Cross/vm/sqVirtualMachine.c M platforms/win32/plugins/CameraPlugin/CameraPlugin.cpp M platforms/win32/vm/sqConfig.h M platforms/win32/vm/sqPlatformSpecific.h M platforms/win32/vm/sqWin32.h M platforms/win32/vm/sqWin32Window.c R spursistastacksrc/examplePlugins.ext R spursistastacksrc/examplePlugins.int R spursistastacksrc/vm/exampleSqNamedPrims.h R spursistastacksrc/vm/gcc3x-interp.c R spursistastacksrc/vm/interp.c R spursistastacksrc/vm/interp.h R spursistastacksrc/vm/vmCallback.h Log Message: ----------- Change platform sources to refer to _WIN32 & _WIN64, not WIN32 & WIN64. Change getSystemAttribute: 1005 (the Windowing system name) to answer 'Windows' insrtead of 'Win32' (we are stuck with answering 'Win32' on 'Win64' unless we want to break /lots/ of image platform-dependent code). If this breaks things then we'll have tio revert. But fingers crossed it is not a problem; there are no uses of 1005 assuming other than 'X11' in a base Squeak image. Nuke spursistastacksrc; it's a misnomer. Sista presupposes a Cogit.

3 3

← Newer
1
...
4
5
6
7
8
9
10
...
16
Older →

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 May 2017