Mac carbon VM 3.8.15b1 pending

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
9 messages Options
Reply | Threaded
Open this post in threaded view
|

Mac carbon VM 3.8.15b1 pending

johnmci
I put a new macintosh carbon VM on my idisk and FTP site
http://www.smalltalkconsulting.com/squeak.html

Squeak 3.8.15beta1U.app.zip


This VM incorporates some changes to fix issues with sound recording/
playback crashes that the Scratch Support team found.
It fixes a problem with untrustedUserDirectory that Tetsuya HAYASHI  
found.
It fixes an issue with the alternate grow VM memory logic that the  
Sophie Team found.


VM changes are:

3.8.15b1 interp.c
                        Alter VMMaker logic to put in missing period in
                                biasToGrow
                                | growSize |
                                growSize :=  growHeadroom*3/2 - (self sizeOfFree: freeBlock).
                                self growObjectMemory: growSize

                        sqMacSecurity.c
                                untrustedUserDirectory  remove fixPath that Tetsuya HAYASHI  
reported stripping last directory
                                off of untrusted directory path that is in info.plis
               
                        sqUnixSoundMacOSXJMM.c
                                        dprintf, change to #define otherwise use of dummy method causes  
compiler to push parms,
                                        resulting in crash if buffer was not allocated on startPlay

3.8.14b8 Basis for MIT Scratch VM
                       
                        sqUnixSoundMacOSXJMM.c
                                        Changes to recording buffer logic to watch for buffer being  
freed by main thread, since
                                        recording buffer logic runs on separate os-x process. Issue was  
crash on stop recording
                                        Where we seem to stop the recording and dispose of the buffer,  
but then the buffer
                                        recording logic on the other thread (osx 10.3) wants to memcopy  
just one more buffer...

3.8.14b7 sqMacMain.c Add signal handler on SIGSEGV, dump all stacks.



--
========================================================================
===
John M. McIntosh <[hidden email]>
Corporate Smalltalk Consulting Ltd.  http://www.smalltalkconsulting.com
========================================================================
===



Reply | Threaded
Open this post in threaded view
|

Re: Mac carbon VM 3.8.15b1 pending

Raymond Asselin
>http://www.smalltalkconsulting.com/squeak.html
Hi John,
    just a note, I try it and it is a lot more fast than the previous
one
Thank for the good work.

Raymond

Reply | Threaded
Open this post in threaded view
|

Re: Mac carbon VM 3.8.15b1 pending

johnmci
How interesting...

Could people run a few tinybenchmarks for me

10 timesRepeat: [Transcript show: 1 tinyBenchmarks printString;cr]

with a 3.8.14 versus 3.8.15b1  VM

let me know if the machine is powerpc, versus macIntel.


On Jan 26, 2007, at 3:05 PM, Asselin Raymond wrote:

>> http://www.smalltalkconsulting.com/squeak.html
> Hi John,
>     just a note, I try it and it is a lot more fast than the previous
> one
> Thank for the good work.
>
> Raymond

--
========================================================================
===
John M. McIntosh <[hidden email]>
Corporate Smalltalk Consulting Ltd.  http://www.smalltalkconsulting.com
========================================================================
===



Reply | Threaded
Open this post in threaded view
|

Re: Mac carbon VM 3.8.15b1 pending

Brian M
John, on a G4 using Squeak3.8.14beta7U:
'197378565 bytecodes/sec; 8295107 sends/sec'
'196923076 bytecodes/sec; 8277048 sends/sec'
'197226502 bytecodes/sec; 8307190 sends/sec'
'197378565 bytecodes/sec; 8307190 sends/sec'
'197074672 bytecodes/sec; 8289079 sends/sec'
'197074672 bytecodes/sec; 8301144 sends/sec'
'190902311 bytecodes/sec; 8277048 sends/sec'
'160905091 bytecodes/sec; 7624179 sends/sec'
'184971098 bytecodes/sec; 7449884 sends/sec'
'179271708 bytecodes/sec; 8301144 sends/sec'


On same G4 using Squeak3.8.15beta1U:
'190760059 bytecodes/sec; 7844410 sends/sec'
'190760059 bytecodes/sec; 8265052 sends/sec'
'197836166 bytecodes/sec; 8199693 sends/sec'
'197530864 bytecodes/sec; 8289079 sends/sec'
'197836166 bytecodes/sec; 8319309 sends/sec'
'196319018 bytecodes/sec; 8295107 sends/sec'
'197836166 bytecodes/sec; 8295107 sends/sec'
'197683397 bytecodes/sec; 8259068 sends/sec'
'195868400 bytecodes/sec; 8301144 sends/sec'
'197989172 bytecodes/sec; 8307190 sends/sec'

Brian.

On Jan 26, 2007, at 8:04 PM, John M McIntosh wrote:

> 10 timesRepeat: [Transcript show: 1 tinyBenchmarks printString;cr]


Reply | Threaded
Open this post in threaded view
|

Re: Mac carbon VM 3.8.15b1 pending

Michael Haupt-3
In reply to this post by johnmci
Hi John,

On 1/27/07, John M McIntosh <[hidden email]> wrote:
> Could people run a few tinybenchmarks for me
> 10 timesRepeat: [Transcript show: 1 tinyBenchmarks printString;cr]
> with a 3.8.14 versus 3.8.15b1  VM
> let me know if the machine is powerpc, versus macIntel.

(Intel) MacBook Pro (Core Duo, 2 GHz, 2 GB RAM)

3.8.14beta7U
'284602556 bytecodes/sec; 7518637 sends/sec'
'289102202 bytecodes/sec; 7406346 sends/sec'
'285714285 bytecodes/sec; 7563509 sends/sec'
'284760845 bytecodes/sec; 7583625 sends/sec'
'286995515 bytecodes/sec; 7543500 sends/sec'
'283499446 bytecodes/sec; 7558497 sends/sec'
'284760845 bytecodes/sec; 7553492 sends/sec'
'284760845 bytecodes/sec; 7583625 sends/sec'
'284760845 bytecodes/sec; 7548493 sends/sec'
'289756649 bytecodes/sec; 7578586 sends/sec'

3.8.15beta1U
'284602556 bytecodes/sec; 7558497 sends/sec'
'284760845 bytecodes/sec; 7573554 sends/sec'
'284760845 bytecodes/sec; 7568528 sends/sec'
'284602556 bytecodes/sec; 7563509 sends/sec'
'284128745 bytecodes/sec; 7568528 sends/sec'
'284760845 bytecodes/sec; 7391946 sends/sec'
'284760845 bytecodes/sec; 7558497 sends/sec'
'284602556 bytecodes/sec; 7406346 sends/sec'
'284128745 bytecodes/sec; 7563509 sends/sec'
'284760845 bytecodes/sec; 7563509 sends/sec'

(PowerPC) PowerBook G4 (1 GHz, 768 MB RAM)

3.8.14beta7U
'76969332 bytecodes/sec; 3201041 sends/sec'
'76692630 bytecodes/sec; 3156968 sends/sec'
'76692630 bytecodes/sec; 3222349 sends/sec'
'73142857 bytecodes/sec; 3224735 sends/sec'
'75249853 bytecodes/sec; 3222349 sends/sec'
'76830732 bytecodes/sec; 3215215 sends/sec'
'76784643 bytecodes/sec; 3163846 sends/sec'
'76692630 bytecodes/sec; 3159258 sends/sec'
'76738609 bytecodes/sec; 3168448 sends/sec'
'76738609 bytecodes/sec; 3222349 sends/sec'

3.8.15beta1U
'76923076 bytecodes/sec; 3198690 sends/sec'
'76509264 bytecodes/sec; 3205752 sends/sec'
'76555023 bytecodes/sec; 3212844 sends/sec'
'76646706 bytecodes/sec; 3217590 sends/sec'
'76190476 bytecodes/sec; 3163846 sends/sec'
'76600837 bytecodes/sec; 3224735 sends/sec'
'76738609 bytecodes/sec; 3224735 sends/sec'
'76692630 bytecodes/sec; 2799882 sends/sec'
'76600837 bytecodes/sec; 3161550 sends/sec'
'76600837 bytecodes/sec; 2375472 sends/sec'

There you are. :-)

Best,

Michael

Reply | Threaded
Open this post in threaded view
|

Re: Mac carbon VM 3.8.15b1 pending

Michael Haupt-3
In reply to this post by Brian M
Hi Brian,

what clock speed does your G4 have?

Best,

Michael

On 1/27/07, Brian Murphy-Dye <[hidden email]> wrote:

> John, on a G4 using Squeak3.8.14beta7U:
> '197378565 bytecodes/sec; 8295107 sends/sec'
> '196923076 bytecodes/sec; 8277048 sends/sec'
> '197226502 bytecodes/sec; 8307190 sends/sec'
> '197378565 bytecodes/sec; 8307190 sends/sec'
> '197074672 bytecodes/sec; 8289079 sends/sec'
> '197074672 bytecodes/sec; 8301144 sends/sec'
> '190902311 bytecodes/sec; 8277048 sends/sec'
> '160905091 bytecodes/sec; 7624179 sends/sec'
> '184971098 bytecodes/sec; 7449884 sends/sec'
> '179271708 bytecodes/sec; 8301144 sends/sec'
>
>
> On same G4 using Squeak3.8.15beta1U:
> '190760059 bytecodes/sec; 7844410 sends/sec'
> '190760059 bytecodes/sec; 8265052 sends/sec'
> '197836166 bytecodes/sec; 8199693 sends/sec'
> '197530864 bytecodes/sec; 8289079 sends/sec'
> '197836166 bytecodes/sec; 8319309 sends/sec'
> '196319018 bytecodes/sec; 8295107 sends/sec'
> '197836166 bytecodes/sec; 8295107 sends/sec'
> '197683397 bytecodes/sec; 8259068 sends/sec'
> '195868400 bytecodes/sec; 8301144 sends/sec'
> '197989172 bytecodes/sec; 8307190 sends/sec'
>
> Brian.
>
> On Jan 26, 2007, at 8:04 PM, John M McIntosh wrote:
>
> > 10 timesRepeat: [Transcript show: 1 tinyBenchmarks printString;cr]
>
>
>

Reply | Threaded
Open this post in threaded view
|

Re: Mac carbon VM 3.8.15b1 pending

Sachin Desai
In reply to this post by johnmci

Hello John,

This is what I get on my laptop.

(Intel) MacBook Pro (Core 2 Duo, 2.33 GHz, 2 GB RAM)

3.8.14beta7U
'525937339 bytecodes/sec; 11643488 sends/sec'
'523785166 bytecodes/sec; 11784756 sends/sec'
'528652555 bytecodes/sec; 11476945 sends/sec'
'524053224 bytecodes/sec; 11462688 sends/sec'
'522448979 bytecodes/sec; 11636146 sends/sec'
'448336252 bytecodes/sec; 11519930 sends/sec'
'345479082 bytecodes/sec; 11512744 sends/sec'
'526207605 bytecodes/sec; 11732313 sends/sec'
'528107271 bytecodes/sec; 11717415 sends/sec'
'455111111 bytecodes/sec; 11807376 sends/sec'


3.8.15beta1U
'524590163 bytecodes/sec; 11512744 sends/sec'
'528379772 bytecodes/sec; 11875758 sends/sec'
'528379772 bytecodes/sec; 11680334 sends/sec'
'443290043 bytecodes/sec; 11301242 sends/sec'
'525937339 bytecodes/sec; 11695138 sends/sec'
'525937339 bytecodes/sec; 11563238 sends/sec'
'527020072 bytecodes/sec; 11643488 sends/sec'
'527563111 bytecodes/sec; 11799826 sends/sec'
'525667351 bytecodes/sec; 11695138 sends/sec'
'525937339 bytecodes/sec; 11398967 sends/sec'

-- Sachin



On Jan 26, 2007, at 7:04 PM, John M McIntosh wrote:

> How interesting...
>
> Could people run a few tinybenchmarks for me
>
> 10 timesRepeat: [Transcript show: 1 tinyBenchmarks printString;cr]
>
> with a 3.8.14 versus 3.8.15b1  VM
>
> let me know if the machine is powerpc, versus macIntel.
>
>
> On Jan 26, 2007, at 3:05 PM, Asselin Raymond wrote:
>
>>> http://www.smalltalkconsulting.com/squeak.html
>> Hi John,
>>     just a note, I try it and it is a lot more fast than the previous
>> one
>> Thank for the good work.
>>
>> Raymond
>
> --
> ======================================================================
> =====
> John M. McIntosh <[hidden email]>
> Corporate Smalltalk Consulting Ltd.  http://
> www.smalltalkconsulting.com
> ======================================================================
> =====
>
>
>


Reply | Threaded
Open this post in threaded view
|

Re: Mac carbon VM 3.8.15b1 pending

Brian M
In reply to this post by Michael Haupt-3
Oops, these numbers are for a 2.5 GHz G5 (not G4).


On Jan 27, 2007, at 2:27 AM, Michael Haupt wrote:

> Hi Brian,
>
> what clock speed does your G4 have?
>
> Best,
>
> Michael
>
> On 1/27/07, Brian Murphy-Dye <[hidden email]> wrote:
>> John, on a G4 using Squeak3.8.14beta7U:
>> '197378565 bytecodes/sec; 8295107 sends/sec'
>> '196923076 bytecodes/sec; 8277048 sends/sec'
>> '197226502 bytecodes/sec; 8307190 sends/sec'
>> '197378565 bytecodes/sec; 8307190 sends/sec'
>> '197074672 bytecodes/sec; 8289079 sends/sec'
>> '197074672 bytecodes/sec; 8301144 sends/sec'
>> '190902311 bytecodes/sec; 8277048 sends/sec'
>> '160905091 bytecodes/sec; 7624179 sends/sec'
>> '184971098 bytecodes/sec; 7449884 sends/sec'
>> '179271708 bytecodes/sec; 8301144 sends/sec'
>>
>>
>> On same G4 using Squeak3.8.15beta1U:
>> '190760059 bytecodes/sec; 7844410 sends/sec'
>> '190760059 bytecodes/sec; 8265052 sends/sec'
>> '197836166 bytecodes/sec; 8199693 sends/sec'
>> '197530864 bytecodes/sec; 8289079 sends/sec'
>> '197836166 bytecodes/sec; 8319309 sends/sec'
>> '196319018 bytecodes/sec; 8295107 sends/sec'
>> '197836166 bytecodes/sec; 8295107 sends/sec'
>> '197683397 bytecodes/sec; 8259068 sends/sec'
>> '195868400 bytecodes/sec; 8301144 sends/sec'
>> '197989172 bytecodes/sec; 8307190 sends/sec'
>>
>> Brian.
>>
>> On Jan 26, 2007, at 8:04 PM, John M McIntosh wrote:
>>
>> > 10 timesRepeat: [Transcript show: 1 tinyBenchmarks printString;cr]
>>
>>
>>
>


Reply | Threaded
Open this post in threaded view
|

Re: Mac carbon VM 3.8.15b1 pending

Michael Haupt-3
Hi Brian,

On 1/29/07, Brian Murphy-Dye <[hidden email]> wrote:
> Oops, these numbers are for a 2.5 GHz G5 (not G4).

phew. I was wondering what I have been doing to my G4. ;-)

Thanks,

Michael