December 2015 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] pharo vm + plugins

by Pierce Ng

On Fri, Dec 18, 2015 at 01:55:14PM -0500, Robert Withers wrote: > I tried this but it no longer seems to work. Do you have the lib* > plugins, by chance? What files are you looking for? Pierce

Dec. 19, 2015

Re: [Pharo-dev] pharo vm + plugins

by Pierce Ng

On Fri, Dec 18, 2015 at 07:48:00PM +0100, Sven Van Caekenberghe wrote: > Maybe a silly question, but aren't VM plugins interchangeable between the > Pharo and Squeak VM ? I know for a fact that the SSL plugin used to be like > that, maybe it still is. Pharo's is named libSqueakSSL.so while Squeak's is named SqueakSSL. I don't know whether renaming one plugin file will allow it to be used with the other VM. Pierce

Dec. 19, 2015

Re: [Pharo-dev] Some test where Spur more slow than Cog

by Eliot Miranda

Hi Denis, On Fri, Dec 18, 2015 at 9:36 AM, Denis Kudriashov <dionisiydk(a)gmail.com> wrote: > Hi. > > I found case where Spur more slow than Cog. Not much but. > Well, it might be indicative of a soluble problem. One thing that Spur is slower at is primitive failure; it catches primitive failure in the VM as potentially caused by primitives encountering forwarders. So many primitive failure involves a slow scan of the primitives arguments to some depth to find out if there are any forwarders which should be followed, and if so the primitive retried. So that means some idioms that use primitive failure normally will go slower. One of these is instVarAt: which used to be written instVarAt: index "Primitive. Answer a fixed variable in an object. The numbering of the variables corresponds to the named instance variables. Fail if the index is not an Integer or is not the index of a fixed variable. Essential. See Object documentation whatIsAPrimitive." <primitive: 73> "Access beyond fixed variables." ^self basicAt: index - self class instSize and used in some copying code to access indexable fields. Chris Muller found this in some of his Magma code and so we added a pair of new primitives, primitiveSlotAt and primitiveSlotAtPut that access any slot within the sequence of named followed by indexable inst vars, avoiding the primitive failure. So in Squeak Spur instVarAt: now reads: instVarAt: index "Primitive. Answer a fixed variable in an object. The numbering of the variables corresponds to the named instance variables, followed by the indexed instance variables. Fail if the index is not an Integer or is not the index of a fixed variable. Essential. See Object documentation whatIsAPrimitive." <primitive: 173 error: ec> self primitiveFailed I suspect something like this is happening with the veryDeepCopy code. You might want to use the profiler on both versions and compare. You might find an issue. HTH > prototype := BorderedMorph new. > > result := [prototype veryDeepCopy ] benchFor: 2 seconds. > > In my computer I got: > > Cog: a BenchmarkResult(16,982 iterations in 2 seconds 1 millisecond. 8,487 > per second) > Spur: a BenchmarkResult(12,507 iterations in 2 seconds 33 milliseconds. > 6,152 per second). > > Is it expected behavior for such case? > -- _,,,^..^,,,_ best, Eliot

Dec. 19, 2015

Fwd: picking up a little speed

by Robert Withers

This is without encryption. I think that is 15 MB/sec. Will video fit in that throughput? 100 KB data chunks received: 22 send time: 54 ms, rate: 325.926 Mbs. receive time: 143 ms. rate: 123.077 Mbs. regards, robert -------- Forwarded Message -------- Subject: picking up a little speed Date: Fri, 18 Dec 2015 20:09:30 -0500 From: Robert Withers <robert.w.withers(a)gmail.com> To: Pharo Development List <pharo-dev(a)lists.pharo.org> I was looking at Zinc to see how it works. My SecureSession was built without my knowing about all of Zinc and it looks delicious on the control side. Here is my latest stats: 100 KB data chunks received: 22 send time: 54 ms, rate: 325.926 Mbs. receive time: 143 ms. rate: 123.077 Mbs. -- . .. .. ^,^ robert

Dec. 19, 2015

picking up a little speed

by Robert Withers

I was looking at Zinc to see how it works. My SecureSession was built without my knowing about all of Zinc and it looks delicious on the control side. Here is my latest stats: 100 KB data chunks received: 22 send time: 54 ms, rate: 325.926 Mbs. receive time: 143 ms. rate: 123.077 Mbs. -- . .. .. ^,^ robert

Dec. 19, 2015

Re: [Pharo-dev] TCP server for Pharo?

by Robert Withers

On 12/18/2015 01:37 PM, Sven Van Caekenberghe wrote: >> On 18 Dec 2015, at 18:33, Robert Withers <robert.w.withers(a)gmail.com> wrote: >> >> Hi Sven, I personally had not looked at Zinc. That's looking good. Therefore, I have a question about how to modify for my own client. You see, I implement a secure protocol that is different than HTTPS/SSL. I think you support SSL. How could I replace your rendezvous handshake with mine? I am looking for the secure state-machine entry point. > Have a look at ZdcSecureSocketStream from the Zodiac package, which is a standard part of Pharo too (http://zdc.stfx.eu in particular http://zdc.stfx.eu/zodiac-paper.html) I think you are looking for #connect and #accept which are the setup hooks for the client and server respectively doing the setup. I am taking a look and I have several comments. Firstly, is it using plugin calls to accept:/connect:from:to:into: to test for connected, then looping on reading encrypted data? It seems bound to tcp. I desire to run traffic over different interfaces so either roll my own stack or modify the internals of Zodiak. It seems it rests upon the SSL plugin, though. I will definitely be doing several scans over my implementation to cleanup the control and interaction with the transport. I see I will be looking heavily at the class you've mentioned. Thank you for the information, robert > >> ANother concern was which size frag packet and which size RS encdding, now that that is getting integrated. >> >> Many thanks, and for SmalltalkHub registation, as well. >> Robert >> >> On 12/18/2015 12:21 PM, Sven Van Caekenberghe wrote: >>>> On 18 Dec 2015, at 18:17, Denis Kudriashov <dionisiydk(a)gmail.com> wrote: >>>> >>>> >>>> 2015-12-18 16:43 GMT+01:00 phil(a)highoctane.be <phil(a)highoctane.be>: >>>> Hi Denis, >>>> >>>> This makes me think about the VNC server, which is a TCP server. >>>> >>>> The implementation of it is on top of another generic one. >>>> >>>> http://www.smalltalkhub.com/#!/~PharoExtras/RFB >>>> >>>> I will look at it to extract error handling logic. I not look at it too much for now. But when I do I will put nice tests for this cases. Which is very important to me. Not only loop implementation duplicated in every servers but tests too (and most times no tests exists) >>> ZnServer and subclasses have a quite robust implementation of server socket to connection handler logic with extensive error handling. All there in the standard image. But I assume you already looked at it. >>> >>> >> -- >> . .. .. ^,^ robert >> > -- . .. .. ^,^ robert

Dec. 18, 2015

Version 52... Re: version 52 of SecureSession is released

by Robert Withers

The newest version is cross-environment: squeakj & pharo, version 52. Don't run the perf test without plugins, please, to set the expectation. There is a class var to control whether RS encoding occurs (SecureSessionServer class>>#setUseFEC/useFEC:). By default it is disabled since it seriously affects performance. After profiling and cleanup, without FEC and with plugins, I consistently get over 20.5 Mbs. I think that's decent, though i am not sure where it measures with other crypto connections. 100 KB data chunks received: 22 send time: 417 ms, rate: 42.206 Mbs. receive time: 858 ms. rate: 20.513 Mbs. Hopefully we'll get RS pluginized. Then we could talk over ham radios, with non-commercial use. robert On 12/18/2015 07:15 AM, Robert Withers wrote: > Well, I must correct myself. It encodes/decodes in the absence of > errors, but it does not recover from damage, yet. It must be the > EuclideanAlgorithm it is using. I'll let you know when that's fixed, > but it is usable without damage, in the meantime. > > Sorry about that, > Robert > > On 12/18/2015 06:50 AM, Robert Withers wrote: >> SecureSession is now FEC encoded. >> >> thanks, > -- . .. .. ^,^ robert

Dec. 18, 2015

Re: [Pharo-dev] pharo vm + plugins

by Robert Withers

On 12/18/2015 01:48 PM, Sven Van Caekenberghe wrote: >> On 18 Dec 2015, at 19:42, Robert Withers <robert.w.withers(a)gmail.com> wrote: >> >> I disabled FEC for performance on both squeak and pharo. Still these plugins would really help me. Here is Pharo's performance, without plugins: > Maybe a silly question, but aren't VM plugins interchangeable between the Pharo and Squeak VM ? I know for a fact that the SSL plugin used to be like that, maybe it still is. I tried this but it no longer seems to work. Do you have the lib* plugins, by chance? manythanks, robert >> 10 KB data chunks received: 3 >> send time: 1908 ms, rate: 0.126 Mbs. >> receive time: 3930 ms. rate: 0.061 Mbs. >> >> and squeak's with plugins: >> >> 100 KB data chunks received: 22 >> send time: 412 ms, rate: 42.718 Mbs. >> receive time: 850 ms. rate: 20.706 Mbs. >> >> So you can see there is a significant difference. >> >> Robert >> >> On 12/18/2015 12:42 PM, Robert Withers wrote: >>> What is the link to help me build a 32-bit spur 50 VM? I want to build the crypto plugins. I have been having trouble with ubuntu building but I think CMake worked for me before. I will try, but I could loose the ability to even run the normal vm. sigh. >>> >>> Otherwise, I am really looking for the crypto plugins that are external for Ubuntu 32-bit. If you have generated these plugins, please email them to me, straightwawy, sos I can benefit from them and run the performance test to work on optimizing the FEC code. >>> >>> manythanks, >>> >> -- >> . .. .. ^,^ robert >> > -- . .. .. ^,^ robert

Dec. 18, 2015

Re: [Pharo-dev] pharo vm + plugins

by Sven Van Caekenberghe

> On 18 Dec 2015, at 19:42, Robert Withers <robert.w.withers(a)gmail.com> wrote: > > I disabled FEC for performance on both squeak and pharo. Still these plugins would really help me. Here is Pharo's performance, without plugins: Maybe a silly question, but aren't VM plugins interchangeable between the Pharo and Squeak VM ? I know for a fact that the SSL plugin used to be like that, maybe it still is. > 10 KB data chunks received: 3 > send time: 1908 ms, rate: 0.126 Mbs. > receive time: 3930 ms. rate: 0.061 Mbs. > > and squeak's with plugins: > > 100 KB data chunks received: 22 > send time: 412 ms, rate: 42.718 Mbs. > receive time: 850 ms. rate: 20.706 Mbs. > > So you can see there is a significant difference. > > Robert > > On 12/18/2015 12:42 PM, Robert Withers wrote: >> What is the link to help me build a 32-bit spur 50 VM? I want to build the crypto plugins. I have been having trouble with ubuntu building but I think CMake worked for me before. I will try, but I could loose the ability to even run the normal vm. sigh. >> >> Otherwise, I am really looking for the crypto plugins that are external for Ubuntu 32-bit. If you have generated these plugins, please email them to me, straightwawy, sos I can benefit from them and run the performance test to work on optimizing the FEC code. >> >> manythanks, >> > > -- > . .. .. ^,^ robert >

Dec. 18, 2015

Re: [Pharo-dev] pharo vm + plugins

by Robert Withers

I disabled FEC for performance on both squeak and pharo. Still these plugins would really help me. Here is Pharo's performance, without plugins: 10 KB data chunks received: 3 send time: 1908 ms, rate: 0.126 Mbs. receive time: 3930 ms. rate: 0.061 Mbs. and squeak's with plugins: 100 KB data chunks received: 22 send time: 412 ms, rate: 42.718 Mbs. receive time: 850 ms. rate: 20.706 Mbs. So you can see there is a significant difference. Robert On 12/18/2015 12:42 PM, Robert Withers wrote: > What is the link to help me build a 32-bit spur 50 VM? I want to build > the crypto plugins. I have been having trouble with ubuntu building > but I think CMake worked for me before. I will try, but I could loose > the ability to even run the normal vm. sigh. > > Otherwise, I am really looking for the crypto plugins that are > external for Ubuntu 32-bit. If you have generated these plugins, > please email them to me, straightwawy, sos I can benefit from them and > run the performance test to work on optimizing the FEC code. > > manythanks, > -- . .. .. ^,^ robert

Dec. 18, 2015