Re: [Pharo-dev] Pharo running with spur (work in process)
Excellent work, Esteban Guillermo (and of course Eliot to begin with), very good to hear. Now, I guess the actual debugging road will be long and dangerous ;-) Apart from the speedup, does this also include 64-bit memory ? On 02 Jul 2014, at 16:13, Esteban Lorenzano esteba...@gmail.com wrote: Hi, I’ve been working on prepare Pharo to run with the new Spur VM… and finally this week Guille and I sit together and make a huge advance :) here a screenshot of pharo 4 running with spur: Screen Shot 2014-07-02 at 15.58.17.png for more information, same image before migration, with a regular cogvm gives this numbers: '1297023432 bytecodes/sec; 161029354 sends/sec’, so that means that the tiny benchmarks run at 166% the speed of the old vm… Of course this is just one benchmark, but I’m very impressed :) Now, it is still not usable but I’m confident I can put a jenkins job to work very soon :) Esteban ps: before you start asking: Spur WILL NOT be available for Pharo3, it will be part of Pharo4.
Re: [Pharo-dev] Pharo running with spur (work in process)
On 02 Jul 2014, at 16:23, Sven Van Caekenberghe s...@stfx.eu wrote: Excellent work, Esteban Guillermo (and of course Eliot to begin with), very good to hear. Now, I guess the actual debugging road will be long and dangerous ;-) Apart from the speedup, does this also include 64-bit memory ? nope :( but this is a necessary step to allow us to start working on the 64bits vm (which I hope to start soon too) On 02 Jul 2014, at 16:13, Esteban Lorenzano esteba...@gmail.com wrote: Hi, I’ve been working on prepare Pharo to run with the new Spur VM… and finally this week Guille and I sit together and make a huge advance :) here a screenshot of pharo 4 running with spur: Screen Shot 2014-07-02 at 15.58.17.png for more information, same image before migration, with a regular cogvm gives this numbers: '1297023432 bytecodes/sec; 161029354 sends/sec’, so that means that the tiny benchmarks run at 166% the speed of the old vm… Of course this is just one benchmark, but I’m very impressed :) Now, it is still not usable but I’m confident I can put a jenkins job to work very soon :) Esteban ps: before you start asking: Spur WILL NOT be available for Pharo3, it will be part of Pharo4.
Re: [Pharo-dev] Pharo running with spur (work in process)
Woohooo! So it has started and it will progress. Wonderful! And it is the path to 64bit as well :) thanks Esteban, Eliot and Guille Norbert Am 02.07.2014 um 16:13 schrieb Esteban Lorenzano esteba...@gmail.com: Hi, I’ve been working on prepare Pharo to run with the new Spur VM… and finally this week Guille and I sit together and make a huge advance :) here a screenshot of pharo 4 running with spur: Screen Shot 2014-07-02 at 15.58.17.png for more information, same image before migration, with a regular cogvm gives this numbers: '1297023432 bytecodes/sec; 161029354 sends/sec’, so that means that the tiny benchmarks run at 166% the speed of the old vm… Of course this is just one benchmark, but I’m very impressed :) Now, it is still not usable but I’m confident I can put a jenkins job to work very soon :) Esteban ps: before you start asking: Spur WILL NOT be available for Pharo3, it will be part of Pharo4.
Re: [Pharo-dev] Pharo running with spur (work in process)
On 02 Jul 2014, at 16:26, Esteban Lorenzano esteba...@gmail.com wrote: On 02 Jul 2014, at 16:23, Sven Van Caekenberghe s...@stfx.eu wrote: Excellent work, Esteban Guillermo (and of course Eliot to begin with), very good to hear. Now, I guess the actual debugging road will be long and dangerous ;-) Apart from the speedup, does this also include 64-bit memory ? nope :( but this is a necessary step to allow us to start working on the 64bits vm (which I hope to start soon too) OK, we'll get there, I am sure. On 02 Jul 2014, at 16:13, Esteban Lorenzano esteba...@gmail.com wrote: Hi, I’ve been working on prepare Pharo to run with the new Spur VM… and finally this week Guille and I sit together and make a huge advance :) here a screenshot of pharo 4 running with spur: Screen Shot 2014-07-02 at 15.58.17.png for more information, same image before migration, with a regular cogvm gives this numbers: '1297023432 bytecodes/sec; 161029354 sends/sec’, so that means that the tiny benchmarks run at 166% the speed of the old vm… Of course this is just one benchmark, but I’m very impressed :) Now, it is still not usable but I’m confident I can put a jenkins job to work very soon :) Esteban ps: before you start asking: Spur WILL NOT be available for Pharo3, it will be part of Pharo4.
Re: [Pharo-dev] Pharo running with spur (work in process)
On 02 Jul 2014, at 16:59, Eliot Miranda eliot.mira...@gmail.com wrote: Hi Esteban, Hi Guille, On Wed, Jul 2, 2014 at 7:56 AM, Esteban Lorenzano esteba...@gmail.com wrote: On 02 Jul 2014, at 16:43, Denis Kudriashov dionisi...@gmail.com wrote: Great! How you built Pharo image for Spur? I read from Eliot mails that for Squeak he has something like 'Sput bootstrap: 'some.image'. Do you implement something similar? If yes then why converting Pharo3 images will not work? we adapted same bootstrap it will not work for pharo3 because we need to add some classes, modify some others and reload the old class builder to bootstrap (and the new class builder needs to be fundamentally modified). so is a lot of work, and does not corresponds to a stable version. also, Spur will be stable some months from now… so it has sense to make it for pharo4 and not for older versions. Great news! Can you send me change sets so I can integrate into VMMaker.oscog? Not yet… it is still very dirty. For now I’m going to the green as fast as I can and I will clean later :P Esteban Best regards, Denis 2014-07-02 18:13 GMT+04:00 Esteban Lorenzano esteba...@gmail.com: Hi, I’ve been working on prepare Pharo to run with the new Spur VM… and finally this week Guille and I sit together and make a huge advance :) here a screenshot of pharo 4 running with spur: Screen Shot 2014-07-02 at 15.58.17.png for more information, same image before migration, with a regular cogvm gives this numbers: '1297023432 bytecodes/sec; 161029354 sends/sec’, so that means that the tiny benchmarks run at 166% the speed of the old vm… Of course this is just one benchmark, but I’m very impressed :) Now, it is still not usable but I’m confident I can put a jenkins job to work very soon :) Esteban ps: before you start asking: Spur WILL NOT be available for Pharo3, it will be part of Pharo4. -- best, Eliot
Re: [Pharo-dev] Pharo running with spur (work in process)
exciting! you guys are magicians sebastian o/ On 02/07/2014, at 11:13, Esteban Lorenzano esteba...@gmail.com wrote: Hi, I’ve been working on prepare Pharo to run with the new Spur VM… and finally this week Guille and I sit together and make a huge advance :) here a screenshot of pharo 4 running with spur: Screen Shot 2014-07-02 at 15.58.17.png for more information, same image before migration, with a regular cogvm gives this numbers: '1297023432 bytecodes/sec; 161029354 sends/sec’, so that means that the tiny benchmarks run at 166% the speed of the old vm… Of course this is just one benchmark, but I’m very impressed :) Now, it is still not usable but I’m confident I can put a jenkins job to work very soon :) Esteban ps: before you start asking: Spur WILL NOT be available for Pharo3, it will be part of Pharo4.
Re: [Pharo-dev] Pharo running with spur (work in process)
Congrats for this giant step forward! 2014-07-02 17:45 GMT+02:00 Sebastian Sastre sebast...@flowingconcept.com: exciting! you guys are magicians sebastian o/ On 02/07/2014, at 11:13, Esteban Lorenzano esteba...@gmail.com wrote: Hi, I’ve been working on prepare Pharo to run with the new Spur VM… and finally this week Guille and I sit together and make a huge advance :) here a screenshot of pharo 4 running with spur: Screen Shot 2014-07-02 at 15.58.17.png for more information, same image before migration, with a regular cogvm gives this numbers: '1297023432 bytecodes/sec; 161029354 sends/sec’, so that means that the tiny benchmarks run at 166% the speed of the old vm… Of course this is just one benchmark, but I’m very impressed :) Now, it is still not usable but I’m confident I can put a jenkins job to work very soon :) Esteban ps: before you start asking: Spur WILL NOT be available for Pharo3, it will be part of Pharo4.
Re: [Pharo-dev] Pharo running with spur (work in process)
A spur image :) awesome !! Thanks Esteban Guille! I'll start adapting the new class builder to Spur next wednesday when I come back from holidays. On 2 juil. 2014, at 16:13, Esteban Lorenzano esteba...@gmail.com wrote: Hi, I’ve been working on prepare Pharo to run with the new Spur VM… and finally this week Guille and I sit together and make a huge advance :) here a screenshot of pharo 4 running with spur: Screen Shot 2014-07-02 at 15.58.17.png for more information, same image before migration, with a regular cogvm gives this numbers: '1297023432 bytecodes/sec; 161029354 sends/sec’, so that means that the tiny benchmarks run at 166% the speed of the old vm… Of course this is just one benchmark, but I’m very impressed :) Now, it is still not usable but I’m confident I can put a jenkins job to work very soon :) Esteban ps: before you start asking: Spur WILL NOT be available for Pharo3, it will be part of Pharo4.
Re: [Pharo-dev] Pharo running with spur (work in process)
0.549181 seconds per message is just insanely low number . This is not even in the realm of nanoseconds. Is this number real ? Because I am very skeptical that one can get this kind of performance from a dynamic language even with a JIT VM. On Wed, Jul 2, 2014 at 9:08 PM, Camille Teruel camille.ter...@gmail.com wrote: A spur image :) awesome !! Thanks Esteban Guille! I'll start adapting the new class builder to Spur next wednesday when I come back from holidays. On 2 juil. 2014, at 16:13, Esteban Lorenzano esteba...@gmail.com wrote: Hi, I’ve been working on prepare Pharo to run with the new Spur VM… and finally this week Guille and I sit together and make a huge advance :) here a screenshot of pharo 4 running with spur: Screen Shot 2014-07-02 at 15.58.17.png for more information, same image before migration, with a regular cogvm gives this numbers: '1297023432 bytecodes/sec; 161029354 sends/sec’, so that means that the tiny benchmarks run at 166% the speed of the old vm… Of course this is just one benchmark, but I’m very impressed :) Now, it is still not usable but I’m confident I can put a jenkins job to work very soon :) Esteban ps: before you start asking: Spur WILL NOT be available for Pharo3, it will be part of Pharo4.
Re: [Pharo-dev] Pharo running with spur (work in process)
On Jul 2, 2014, at 3:43 PM, kilon alios kilon.al...@gmail.com wrote: 0.549181 seconds per message is just insanely low number . This is not even in the realm of nanoseconds. Is this number real ? Because I am very skeptical that one can get this kind of performance from a dynamic language even with a JIT VM. Nice! careful guys, as we’re start dealing with things in the Planck scale with all this power we can accidentally open a portal to another multiverse :D
Re: [Pharo-dev] Pharo running with spur (work in process)
On Wed, Jul 2, 2014 at 11:43 AM, kilon alios kilon.al...@gmail.com wrote: 0.549181 seconds per message is just insanely low number . This is not even in the realm of nanoseconds. Is this number real ? Because I am very skeptical that one can get this kind of performance from a dynamic language even with a JIT VM. Here's a measurement on my 2.2GHz Mac Book Pro Core i7: | r | {[r := 34 benchFib] timeToRun. r} #(140 18454929) benchFib is nfib that adds one for each call, so the answer is the number of calls necessary to compute the answer. benchFib ^self 2 ifTrue: [1] ifFalse: [(self-1) benchFib + (self-2) benchFib + 1] So that's 18 million calls in 140 milliseconds, or about 132 calls per microsecond. That's indeed in the realm of nanoseconds, 13.2 nsecs per call. On Wed, Jul 2, 2014 at 9:08 PM, Camille Teruel camille.ter...@gmail.com wrote: A spur image :) awesome !! Thanks Esteban Guille! I'll start adapting the new class builder to Spur next wednesday when I come back from holidays. On 2 juil. 2014, at 16:13, Esteban Lorenzano esteba...@gmail.com wrote: Hi, I’ve been working on prepare Pharo to run with the new Spur VM… and finally this week Guille and I sit together and make a huge advance :) here a screenshot of pharo 4 running with spur: Screen Shot 2014-07-02 at 15.58.17.png for more information, same image before migration, with a regular cogvm gives this numbers: '1297023432 bytecodes/sec; 161029354 sends/sec’, so that means that the tiny benchmarks run at 166% the speed of the old vm… Of course this is just one benchmark, but I’m very impressed :) Now, it is still not usable but I’m confident I can put a jenkins job to work very soon :) Esteban ps: before you start asking: Spur WILL NOT be available for Pharo3, it will be part of Pharo4. -- best, Eliot
Re: [Pharo-dev] Pharo running with spur (work in process)
On Wed, Jul 2, 2014 at 12:22 PM, Eliot Miranda eliot.mira...@gmail.com wrote: On Wed, Jul 2, 2014 at 11:43 AM, kilon alios kilon.al...@gmail.com wrote: 0.549181 seconds per message is just insanely low number . This is not even in the realm of nanoseconds. Is this number real ? Because I am very skeptical that one can get this kind of performance from a dynamic language even with a JIT VM. Here's a measurement on my 2.2GHz Mac Book Pro Core i7: | r | {[r := 34 benchFib] timeToRun. r} #(140 18454929) benchFib is nfib that adds one for each call, so the answer is the number of calls necessary to compute the answer. benchFib ^self 2 ifTrue: [1] ifFalse: [(self-1) benchFib + (self-2) benchFib + 1] So that's 18 million calls in 140 milliseconds, or about 132 calls per microsecond. That's indeed in the realm of nanoseconds, 13.2 nsecs per call. Ugh, let me try again... that's 184549290 / 0.14 = 1.3182092142857141e9 calls per second, which is 0.76 nsecs per call. On Wed, Jul 2, 2014 at 9:08 PM, Camille Teruel camille.ter...@gmail.com wrote: A spur image :) awesome !! Thanks Esteban Guille! I'll start adapting the new class builder to Spur next wednesday when I come back from holidays. On 2 juil. 2014, at 16:13, Esteban Lorenzano esteba...@gmail.com wrote: Hi, I’ve been working on prepare Pharo to run with the new Spur VM… and finally this week Guille and I sit together and make a huge advance :) here a screenshot of pharo 4 running with spur: Screen Shot 2014-07-02 at 15.58.17.png for more information, same image before migration, with a regular cogvm gives this numbers: '1297023432 bytecodes/sec; 161029354 sends/sec’, so that means that the tiny benchmarks run at 166% the speed of the old vm… Of course this is just one benchmark, but I’m very impressed :) Now, it is still not usable but I’m confident I can put a jenkins job to work very soon :) Esteban ps: before you start asking: Spur WILL NOT be available for Pharo3, it will be part of Pharo4. -- best, Eliot -- best, Eliot
Re: [Pharo-dev] Pharo running with spur (work in process)
Ah ah, insanely sweet! Le 2 juil. 2014 21:29, Eliot Miranda eliot.mira...@gmail.com a écrit : On Wed, Jul 2, 2014 at 12:22 PM, Eliot Miranda eliot.mira...@gmail.com wrote: On Wed, Jul 2, 2014 at 11:43 AM, kilon alios kilon.al...@gmail.com wrote: 0.549181 seconds per message is just insanely low number . This is not even in the realm of nanoseconds. Is this number real ? Because I am very skeptical that one can get this kind of performance from a dynamic language even with a JIT VM. Here's a measurement on my 2.2GHz Mac Book Pro Core i7: | r | {[r := 34 benchFib] timeToRun. r} #(140 18454929) benchFib is nfib that adds one for each call, so the answer is the number of calls necessary to compute the answer. benchFib ^self 2 ifTrue: [1] ifFalse: [(self-1) benchFib + (self-2) benchFib + 1] So that's 18 million calls in 140 milliseconds, or about 132 calls per microsecond. That's indeed in the realm of nanoseconds, 13.2 nsecs per call. Ugh, let me try again... that's 184549290 / 0.14 = 1.3182092142857141e9 calls per second, which is 0.76 nsecs per call. On Wed, Jul 2, 2014 at 9:08 PM, Camille Teruel camille.ter...@gmail.com wrote: A spur image :) awesome !! Thanks Esteban Guille! I'll start adapting the new class builder to Spur next wednesday when I come back from holidays. On 2 juil. 2014, at 16:13, Esteban Lorenzano esteba...@gmail.com wrote: Hi, I’ve been working on prepare Pharo to run with the new Spur VM… and finally this week Guille and I sit together and make a huge advance :) here a screenshot of pharo 4 running with spur: Screen Shot 2014-07-02 at 15.58.17.png for more information, same image before migration, with a regular cogvm gives this numbers: '1297023432 bytecodes/sec; 161029354 sends/sec’, so that means that the tiny benchmarks run at 166% the speed of the old vm… Of course this is just one benchmark, but I’m very impressed :) Now, it is still not usable but I’m confident I can put a jenkins job to work very soon :) Esteban ps: before you start asking: Spur WILL NOT be available for Pharo3, it will be part of Pharo4. -- best, Eliot -- best, Eliot
Re: [Pharo-dev] Pharo running with spur (work in process)
For me its 99 milliseconds with my iMac (Pharo 3) which is a 3.2 Intel Core i5 and 8GB 1600Mhz DDR3 Ram. Very impressive stuff. On Wed, Jul 2, 2014 at 10:22 PM, Eliot Miranda eliot.mira...@gmail.com wrote: On Wed, Jul 2, 2014 at 11:43 AM, kilon alios kilon.al...@gmail.com wrote: 0.549181 seconds per message is just insanely low number . This is not even in the realm of nanoseconds. Is this number real ? Because I am very skeptical that one can get this kind of performance from a dynamic language even with a JIT VM. Here's a measurement on my 2.2GHz Mac Book Pro Core i7: | r | {[r := 34 benchFib] timeToRun. r} #(140 18454929) benchFib is nfib that adds one for each call, so the answer is the number of calls necessary to compute the answer. benchFib ^self 2 ifTrue: [1] ifFalse: [(self-1) benchFib + (self-2) benchFib + 1] So that's 18 million calls in 140 milliseconds, or about 132 calls per microsecond. That's indeed in the realm of nanoseconds, 13.2 nsecs per call. On Wed, Jul 2, 2014 at 9:08 PM, Camille Teruel camille.ter...@gmail.com wrote: A spur image :) awesome !! Thanks Esteban Guille! I'll start adapting the new class builder to Spur next wednesday when I come back from holidays. On 2 juil. 2014, at 16:13, Esteban Lorenzano esteba...@gmail.com wrote: Hi, I’ve been working on prepare Pharo to run with the new Spur VM… and finally this week Guille and I sit together and make a huge advance :) here a screenshot of pharo 4 running with spur: Screen Shot 2014-07-02 at 15.58.17.png for more information, same image before migration, with a regular cogvm gives this numbers: '1297023432 bytecodes/sec; 161029354 sends/sec’, so that means that the tiny benchmarks run at 166% the speed of the old vm… Of course this is just one benchmark, but I’m very impressed :) Now, it is still not usable but I’m confident I can put a jenkins job to work very soon :) Esteban ps: before you start asking: Spur WILL NOT be available for Pharo3, it will be part of Pharo4. -- best, Eliot