Vm-dev April 2014

vm-dev@lists.squeakfoundation.org

39 participants
112 discussions

Re: [squeak-dev] Help needed to compile a fresh Mac Carbon VM
by David T. Lewis 26 Apr '14

26 Apr '14

On Sat, Apr 26, 2014 at 08:13:41AM -0300, Edgar J. De Cleene wrote: > > On 4/25/14, 9:59 PM, "David T. Lewis" <lewis(a)mail.msen.com> wrote: > > > > If you have the ability to compile a Mac Carbon VM and make it available to > > the community, then please speak up. We need your help. > > > > I think I have how to do it, as have some G4's and Xcode. > But sure I need help for doing this for the first time. > Can you assist or some with experience ? Hi Edgar, Thank you! Yes for sure, I will help in any way I can. I am CCing the vm-dev mailing list also, because many of the experts hang out there. You may want to join the vm-dev list also: http://lists.squeakfoundation.org/mailman/listinfo/vm-dev I do not know much about compatibility with the G4 machines and other Macs. But I think it would be great if you can compile the VM on your G4. That would make it possible for people to use their G4 machines to run new Squeak images saved with a Cog VM. Dave

1 0

Improving buildsqueaktrunkimage.sh in Cog/image SVN tree
by gettimothy 25 Apr '14

25 Apr '14

Hi all. I thought I would get my feet wet on the Squeak CMake stuff by mimicing the script the pharo team has for automatically downloading and configuring an image with the VMaker tools. Well, it turns out that one already exists and I just did not know about it. (: I would like to improve the process such that running the script sets up 'everything' so that when booting the image one can run VMaker out-of-the-box without setting up the oscogvm subdirectories etc. Any objections? thx. tty P.S. my workplan for the CMake project is: 1. get the script going. <-- I am here 2. add support for the existing Standard Interpreter. 3. Build Tim R.'s requirements 4. CMake-ify the remaining VM's/platforms. P.P.S Solaris/OpenIndiana When the Solaris vm is done, If I could get an url to the relevant files, I would like to add that to the script as well. I have an OpenIndiana partition I can test that on.

2 3

Linux Numeric Keyboard issue in Pharo-VM fix
by phil＠highoctane.be 25 Apr '14

25 Apr '14

Hello, As I am now working on a CentOS sytem with Pharo, I got into: https://pharo.fogbugz.com/default.asp?11352 So, I put back the fix that was done in Gitorious #2e6f231 and it now works. May someone integrate that fix into the main Linux repo? Thx Phil

4 3

[commit][2904] Integrate Philippe Back's fix for numberic keypads on X11;
by commits＠squeakvm.org 24 Apr '14

24 Apr '14

Revision: 2904 Author: eliot Date: 2014-04-24 14:29:08 -0700 (Thu, 24 Apr 2014) Log Message: ----------- Integrate Philippe Back's fix for numberic keypads on X11; see https://pharo.fogbugz.com/f/cases/11352/. Add relevant lines for the Raspberry Pi to the squeak startup script. Modified Paths: -------------- branches/Cog/platforms/unix/config/bin.squeak.sh.in branches/Cog/platforms/unix/config/squeak.sh.in branches/Cog/platforms/unix/vm-display-X11/sqUnixX11.c Property Changed: ---------------- branches/Cog/platforms/Cross/vm/sqSCCSVersion.h Property changes on: branches/Cog/platforms/Cross/vm/sqSCCSVersion.h ___________________________________________________________________ Modified: checkindate - Mon Apr 14 15:29:58 PDT 2014 + Thu Apr 24 14:27:32 PDT 2014 Modified: branches/Cog/platforms/unix/config/bin.squeak.sh.in =================================================================== --- branches/Cog/platforms/unix/config/bin.squeak.sh.in 2014-04-22 21:58:09 UTC (rev 2903) +++ branches/Cog/platforms/unix/config/bin.squeak.sh.in 2014-04-24 21:29:08 UTC (rev 2904) @@ -30,6 +30,8 @@ SVMLLP="/lib/tls/i686/cmov:/lib:/usr/lib/tls/i686/cmov:/usr/lib";; /lib/tls/i686/nosegneg/libc*) \ SVMLLP="/lib/i686/nosegneg/tls:/lib:/usr/lib/i686/nosegneg/tls:/usr/lib";; +/lib/arm-linux-gnueabihf/libc*) \ + SVMLLP="/lib/arm-linux-gnueabihf:/lib:/usr/lib/arm-linux-gnueabihf:/usr/lib";; "") case `/bin/uname -m || /usr/bin/uname -m` in *64*) echo "/usr/bin/ldd didn't produce any output and the system is 64 bit. You may need to (re)install the 32-bit libraries." 1>&2;; *) echo "/usr/bin/ldd didn't produce any output. Can't infer base LD_LIBRARY_PATH. Aborting." 1>&2 Modified: branches/Cog/platforms/unix/config/squeak.sh.in =================================================================== --- branches/Cog/platforms/unix/config/squeak.sh.in 2014-04-22 21:58:09 UTC (rev 2903) +++ branches/Cog/platforms/unix/config/squeak.sh.in 2014-04-24 21:29:08 UTC (rev 2904) @@ -30,6 +30,8 @@ SVMLLP="/lib/tls/i686/cmov:/lib:/usr/lib/tls/i686/cmov:/usr/lib";; /lib/tls/i686/nosegneg/libc*) \ SVMLLP="/lib/i686/nosegneg/tls:/lib:/usr/lib/i686/nosegneg/tls:/usr/lib";; +/lib/arm-linux-gnueabihf/libc*) \ + SVMLLP="/lib/arm-linux-gnueabihf:/lib:/usr/lib/arm-linux-gnueabihf:/usr/lib";; "") case `/bin/uname -m || /usr/bin/uname -m` in *64*) echo "/usr/bin/ldd didn't produce any output and the system is 64 bit. You may need to (re)install the 32-bit libraries." 1>&2;; *) echo "/usr/bin/ldd didn't produce any output. Can't infer base LD_LIBRARY_PATH. Aborting." 1>&2 Modified: branches/Cog/platforms/unix/vm-display-X11/sqUnixX11.c =================================================================== --- branches/Cog/platforms/unix/vm-display-X11/sqUnixX11.c 2014-04-22 21:58:09 UTC (rev 2903) +++ branches/Cog/platforms/unix/vm-display-X11/sqUnixX11.c 2014-04-24 21:29:08 UTC (rev 2904) @@ -1991,6 +1991,7 @@ static int xkeysym2ucs4(KeySym keysym) { + /* Latin 2 Mappings */ static unsigned short const ucs4_01a1_01ff[] = { 0x0104, 0x02d8, 0x0141, 0x0000, 0x013d, 0x015a, 0x0000, /* 0x01a0-0x01a7 */ 0x0000, 0x0160, 0x015e, 0x0164, 0x0179, 0x0000, 0x017d, 0x017b, /* 0x01a8-0x01af */ @@ -2006,6 +2007,7 @@ 0x0159, 0x016f, 0x0000, 0x0171, 0x0000, 0x0000, 0x0163, 0x02d9 /* 0x01f8-0x01ff */ }; + /* Latin 3 Mappings */ static unsigned short const ucs4_02a1_02fe[] = { 0x0126, 0x0000, 0x0000, 0x0000, 0x0000, 0x0124, 0x0000, /* 0x02a0-0x02a7 */ 0x0000, 0x0130, 0x0000, 0x011e, 0x0134, 0x0000, 0x0000, 0x0000, /* 0x02a8-0x02af */ @@ -2021,6 +2023,7 @@ 0x011d, 0x0000, 0x0000, 0x0000, 0x0000, 0x016d, 0x015d /* 0x02f8-0x02ff */ }; + /* Latin 4 Mappings */ static unsigned short const ucs4_03a2_03fe[] = { 0x0138, 0x0156, 0x0000, 0x0128, 0x013b, 0x0000, /* 0x03a0-0x03a7 */ 0x0000, 0x0000, 0x0112, 0x0122, 0x0166, 0x0000, 0x0000, 0x0000, /* 0x03a8-0x03af */ @@ -2036,6 +2039,7 @@ 0x0000, 0x0173, 0x0000, 0x0000, 0x0000, 0x0169, 0x016b /* 0x03f8-0x03ff */ }; + /* Katakana Mappings */ static unsigned short const ucs4_04a1_04df[] = { 0x3002, 0x3008, 0x3009, 0x3001, 0x30fb, 0x30f2, 0x30a1, /* 0x04a0-0x04a7 */ 0x30a3, 0x30a5, 0x30a7, 0x30a9, 0x30e3, 0x30e5, 0x30e7, 0x30c3, /* 0x04a8-0x04af */ @@ -2047,6 +2051,7 @@ 0x30ea, 0x30eb, 0x30ec, 0x30ed, 0x30ef, 0x30f3, 0x309b, 0x309c /* 0x04d8-0x04df */ }; + /* Arabic mappings */ static unsigned short const ucs4_0590_05fe[] = { 0x06f0, 0x06f1, 0x06f2, 0x06f3, 0x06f4, 0x06f5, 0x06f6, 0x06f7, /* 0x0590-0x0597 */ 0x06f8, 0x06f9, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, /* 0x0598-0x059f */ @@ -2065,6 +2070,7 @@ 0x06a9, 0x06af, 0x06ba, 0x06be, 0x06cc, 0x06d2, 0x06c1 /* 0x05f8-0x05fe */ }; + /* Cyrillic mappings */ static unsigned short ucs4_0680_06ff[] = { 0x0492, 0x0496, 0x049a, 0x049c, 0x04a2, 0x04ae, 0x04b0, 0x04b2, /* 0x0680-0x0687 */ 0x04b6, 0x04b8, 0x04ba, 0x0000, 0x04d8, 0x04e2, 0x04e8, 0x04ee, /* 0x0688-0x068f */ @@ -2084,6 +2090,7 @@ 0x042c, 0x042b, 0x0417, 0x0428, 0x042d, 0x0429, 0x0427, 0x042a /* 0x06f8-0x06ff */ }; + /* Greek mappings */ static unsigned short const ucs4_07a1_07f9[] = { 0x0386, 0x0388, 0x0389, 0x038a, 0x03aa, 0x0000, 0x038c, /* 0x07a0-0x07a7 */ 0x038e, 0x03ab, 0x0000, 0x038f, 0x0000, 0x0000, 0x0385, 0x2015, /* 0x07a8-0x07af */ @@ -2099,6 +2106,7 @@ 0x03c8, 0x03c9 /* 0x07f8-0x07ff */ }; + /* Technical mappings */ static unsigned short const ucs4_08a4_08fe[] = { 0x2320, 0x2321, 0x0000, 0x231c, /* 0x08a0-0x08a7 */ 0x231d, 0x231e, 0x231f, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, /* 0x08a8-0x08af */ @@ -2114,6 +2122,7 @@ 0x0000, 0x0000, 0x0000, 0x2190, 0x2191, 0x2192, 0x2193 /* 0x08f8-0x08ff */ }; + /* Special mappings from the DEC VT100 Special Graphics Character Set */ static unsigned short const ucs4_09df_09f8[] = { 0x2422, /* 0x09d8-0x09df */ 0x2666, 0x25a6, 0x2409, 0x240c, 0x240d, 0x240a, 0x0000, 0x0000, /* 0x09e0-0x09e7 */ @@ -2122,6 +2131,7 @@ 0x2502 /* 0x09f8-0x09ff */ }; + /* Publishing Mappings ? */ static unsigned short const ucs4_0aa1_0afe[] = { 0x2003, 0x2002, 0x2004, 0x2005, 0x2007, 0x2008, 0x2009, /* 0x0aa0-0x0aa7 */ 0x200a, 0x2014, 0x2013, 0x0000, 0x0000, 0x0000, 0x2026, 0x2025, /* 0x0aa8-0x0aaf */ @@ -2137,6 +2147,7 @@ 0x2640, 0x2121, 0x2315, 0x2117, 0x2038, 0x201a, 0x201e /* 0x0af8-0x0aff */ }; + /* Hebrew Mappings */ static unsigned short const ucs4_0cdf_0cfa[] = { 0x2017, /* 0x0cd8-0x0cdf */ 0x05d0, 0x05d1, 0x05d2, 0x05d3, 0x05d4, 0x05d5, 0x05d6, 0x05d7, /* 0x0ce0-0x0ce7 */ @@ -2145,6 +2156,7 @@ 0x05e8, 0x05e9, 0x05ea /* 0x0cf8-0x0cff */ }; + /* Thai Mappings */ static unsigned short const ucs4_0da1_0df9[] = { 0x0e01, 0x0e02, 0x0e03, 0x0e04, 0x0e05, 0x0e06, 0x0e07, /* 0x0da0-0x0da7 */ 0x0e08, 0x0e09, 0x0e0a, 0x0e0b, 0x0e0c, 0x0e0d, 0x0e0e, 0x0e0f, /* 0x0da8-0x0daf */ @@ -2160,6 +2172,7 @@ 0x0e58, 0x0e59 /* 0x0df8-0x0dff */ }; + /* Hangul Mappings */ static unsigned short const ucs4_0ea0_0eff[] = { 0x0000, 0x1101, 0x1101, 0x11aa, 0x1102, 0x11ac, 0x11ad, 0x1103, /* 0x0ea0-0x0ea7 */ 0x1104, 0x1105, 0x11b0, 0x11b1, 0x11b2, 0x11b3, 0x11b4, 0x11b5, /* 0x0ea8-0x0eaf */ @@ -2175,6 +2188,7 @@ 0x11eb, 0x0000, 0x11f9, 0x0000, 0x0000, 0x0000, 0x0000, 0x20a9, /* 0x0ef8-0x0eff */ }; + /* Non existing range in keysymdef.h */ static unsigned short ucs4_12a1_12fe[] = { 0x1e02, 0x1e03, 0x0000, 0x0000, 0x0000, 0x1e0a, 0x0000, /* 0x12a0-0x12a7 */ 0x1e80, 0x0000, 0x1e82, 0x1e0b, 0x1ef2, 0x0000, 0x0000, 0x0000, /* 0x12a8-0x12af */ @@ -2189,11 +2203,13 @@ 0x0175, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x1e6b, /* 0x12f0-0x12f7 */ 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0177 /* 0x12f0-0x12ff */ }; - + + /* Latin 9 Mappings */ static unsigned short const ucs4_13bc_13be[] = { 0x0152, 0x0153, 0x0178 /* 0x13b8-0x13bf */ }; + /* Non existing range in keysymdef.h */ static unsigned short ucs4_14a1_14ff[] = { 0x2741, 0x00a7, 0x0589, 0x0029, 0x0028, 0x00bb, 0x00ab, /* 0x14a0-0x14a7 */ 0x2014, 0x002e, 0x055d, 0x002c, 0x2013, 0x058a, 0x2026, 0x055c, /* 0x14a8-0x14af */ @@ -2209,6 +2225,7 @@ 0x0554, 0x0584, 0x0555, 0x0585, 0x0556, 0x0586, 0x2019, 0x0027, /* 0x14f8-0x14ff */ }; + /* Non existing range in keysymdef.h */ static unsigned short ucs4_15d0_15f6[] = { 0x10d0, 0x10d1, 0x10d2, 0x10d3, 0x10d4, 0x10d5, 0x10d6, 0x10d7, /* 0x15d0-0x15d7 */ 0x10d8, 0x10d9, 0x10da, 0x10db, 0x10dc, 0x10dd, 0x10de, 0x10df, /* 0x15d8-0x15df */ @@ -2217,6 +2234,7 @@ 0x10f0, 0x10f1, 0x10f2, 0x10f3, 0x10f4, 0x10f5, 0x10f6 /* 0x15f0-0x15f7 */ }; + /* Non existing range in keysymdef.h */ static unsigned short ucs4_16a0_16f6[] = { 0x0000, 0x0000, 0xf0a2, 0x1e8a, 0x0000, 0xf0a5, 0x012c, 0xf0a7, /* 0x16a0-0x16a7 */ 0xf0a8, 0x01b5, 0x01e6, 0x0000, 0x0000, 0x0000, 0x0000, 0x019f, /* 0x16a8-0x16af */ @@ -2231,6 +2249,7 @@ 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0000, 0x0259 /* 0x16f0-0x16f6 */ }; + /* Vietnamesse Mappings */ static unsigned short const ucs4_1e9f_1eff[] = { 0x0303, 0x1ea0, 0x1ea1, 0x1ea2, 0x1ea3, 0x1ea4, 0x1ea5, 0x1ea6, 0x1ea7, /* 0x1ea0-0x1ea7 */ @@ -2247,13 +2266,39 @@ 0x1ef8, 0x1ef9, 0x01a0, 0x01a1, 0x01af, 0x01b0, 0x0309, 0x0323 /* 0x1ef8-0x1eff */ }; + /* Currency */ static unsigned short const ucs4_20a0_20ac[] = { 0x20a0, 0x20a1, 0x20a2, 0x20a3, 0x20a4, 0x20a5, 0x20a6, 0x20a7, /* 0x20a0-0x20a7 */ 0x20a8, 0x20a9, 0x20aa, 0x20ab, 0x20ac /* 0x20a8-0x20af */ }; - static unsigned short const sqSpecialKey[] = {1, 28, 30, 29, 31, 11, 12, 4, 1}; + /* Keypad numbers mapping */ + static unsigned short const ucs4_ffb0_ffb9[] = { 0x30, 0x31, 0x32, 0x33, 0x34, + 0x35, 0x36, 0x37, 0x38, 0x39}; + + /* Keypad operators mapping */ + static unsigned short const ucs4_ffaa_ffaf[] = { + 0x2a, /* Multiply */ + 0x2b, /* Add */ + 0x2c, /* Separator */ + 0x2d, /* Substract */ + 0x2e, /* Decimal */ + 0x2f /* Divide */ + }; + static unsigned short const sqSpecialKey[] = { + 1, /* HOME */ + 28, /* LEFT */ + 30, /* UP */ + 29, /* RIGHT */ + 31, /* DOWN */ + 11, /* PRIOR (page up?) */ + 12, /* NEXT (page down/new page?) */ + 4, /* END */ + 1 /* HOME */ + }; + + /* Latin-1 */ if ( (keysym >= 0x0020 && keysym <= 0x007e) || (keysym >= 0x00a0 && keysym <= 0x00ff)) return keysym; @@ -2275,6 +2320,15 @@ if (keysym == 0xff1b) return keysym & 0x001f; if (keysym == 0xffff) return keysym & 0x007f; + /* Misc mappings */ + if (keysym == XK_Escape) + return keysym & 0x001f; + if (keysym == XK_Delete) + return keysym & 0x007f; + if (keysym == XK_KP_Equal) + return XK_equal; + + /* explicitly mapped */ #define map(lo, hi) if (keysym >= 0x##lo && keysym <= 0x##hi) return ucs4_##lo##_##hi[keysym - 0x##lo]; map(01a1, 01ff); map(02a1, 02fe); map(03a2, 03fe); map(04a1, 04df); @@ -2282,6 +2336,8 @@ map(09df, 09f8); map(0aa1, 0afe); map(0cdf, 0cfa); map(0da1, 0df9); map(0ea0, 0eff); map(12a1, 12fe); map(13bc, 13be); map(14a1, 14ff); map(15d0, 15f6); map(16a0, 16f6); map(1e9f, 1eff); map(20a0, 20ac); + map(ffb0, ffb9); + map(ffaa, ffaf); #undef map #if defined(XF86XK_Start)

1 0

VM on Solaris (was: Camera sig fault on 64 bits machines.)
by Andreas Wacknitz 24 Apr '14

24 Apr '14

This evening I further dealt with the problems on OpenSolaris (openindiana). I finally got a pthread version running without superuser rights. But I don’t know whether this will really work (ATM it does for me) because I removed the call to pthread_setschedparam in beatStateMachine leaving the heartbeat thread with the same priority than the vm thread. I tried to replace the pthread_setschedparam call with a similar pthread_setschedprio call but with no luck (same problem: failed call with "Not owner"). I don’t know wether this is a general problem with the pthreads implementation on Solaris or just a problem with the gcc version (4.4.4) coming with the openindiana distribution I am using. Maybe this works only with the compilers and libraries that is delivered by Oracle (Solaris 10 ships with gcc 3.4.3; Solaris Studio has its own compilers). Second, I needed to change the implementation of ioUpdateVMTimezone because Solaris does not have time_t->tm_gmtoff. There seem to be copies of this function in all three heartbeat files with the one in sqUnixITimerHeartbeat.c working for those OS’s without tm_gmtoff. NativeBoost doesn’t seem to work yet (at least UnixEnvironment>>environ raises an error: „failed to get a symbol address: environ“). This VM gives me 686787391 bytecodes/sec and 80516849 sends/sec on my 6 years old Sun Ultra 24 (2,4GHz Intel Q9300). I get similar values for the VM with pthread_setschedparam call and superuser rights. My 4 years old iMac (2,8GHz Core i5) gives me 829149797 bytecodes/sec; 117122195 sends/sec. So the results seem comparable. Andreas

4 11

CMakeMaker--whaddya think?
by gettimothy 24 Apr '14

24 Apr '14

Hi all. Unless I am talked out of it , I am going to generate the CMMake "stuff" from Smalltalk classes similar to the way that VMMaker generates source code. My reasons are: 1. source control is Monticello. (and if you generate the source tree, the pharo folks can use that to pipe into git--so that sorta-kinda-syncs things up a bit) 2. modification and maintenance is in Squeak. 3. it should be fun. 4. Modular (looking at the CMake generators, it just begs for OOP approach) 5. Each plugin is responsible for its own CMakeLists.txt (i.e. cmake configuration file) .... whaddya think? Also, if you are ok with the idea, what do you want it named? Do you want some classes added to VMMaker or do you want something elsewhere? p.s. I am suffering from a bit of Spring Fever and took the weekend and Monday and most of today off. I am still studying CMaker before I even look at the existing project for Interpreter and I will probably not look at the existing Interpreter CMake project until I go through the exercise of writing my own from scratch (unless I fall flat on my face, then I will take a peek (: )

11 27

Clarification needed for sqCompareAndSwapRes
by Nicolas Cellier 24 Apr '14

24 Apr '14

In platforms/Cross/vm/sqAtomicOps.h a comment tells * sqCompareAndSwapRes(var,old,new,res) arranges atomically that if var's value * is equal to old, then var's value is set to new, and that in any case, res * is set to the previous value of var. But for some implementation, if var==old, res is first set to var (old) then to new which is the new value of var: #if defined(__GNUC__) && (defined(i386) || defined(__i386) || defined(__i386__) || defined(_X86_)) #ifdef TARGET_OS_IS_IPHONE # define sqCompareAndSwap(var,old,new) OSAtomicCompareAndSwap32(old, new, &var) /* N.B. This is not atomic in fetching var's old value :( */ # define sqCompareAndSwapRes(var,old,new,res) do { res = var; if (OSAtomicCompareAndSwap32(old, new, &var)) res = new; } while (0) Indeed, comparedAndSwap32 will answer true if var==old as explained here https://www.mikeash.com/pyblog/friday-qa-2011-03-04-a-tour-of-osatomic.html That is going to be a problem in lockSignalQueue in ./platforms/Cross/vm/sqExternalSemaphores.c because variable old will then be 1 in all cases and the loop won't exit once the lock acquired: static inline void lockSignalQueue() { volatile int old; /* spin to obtain a lock */ do { sqLowLevelMFence(); sqCompareAndSwapRes(sigLock, 0, 1, old ); } while (old != 0); } Knowing we are on iphone, I would just write it: static inline void lockSignalQueue() { /* spin to obtain a lock */ do { sqLowLevelMFence(); } while (! OSAtomicCompareInt32(sigLock, 0, 1)); } Esteban "fixed" that by swapping the arguments in the macrosqCompareAndSwapRes But IMO, this is wrong, as I understand it, it completely desactivates the lock... I did not know about the problem, but found the diff with Eliot's branch very suspicious so I opened https://pharo.fogbugz.com/f/cases/13199/Stinky-argument-inversion-of-atomic… and pulled a request https://github.com/pharo-project/pharo-vm/pull/38 for restoring (the supposedly wrong) Eliot's version. Now I need some guru advice, because this kind of code is brainfucking (especially when old in sender is not old in sendee). Do you agree that TARGET_OS_IS_IPHONE branch is wrong? Do you agree that Esteban's fix is wrong? What do you suggest for a fix? Thanks Nicolas

5 21

VM Maker: CMakeVMMaker-tty.215.mcz
by commits＠source.squeak.org 23 Apr '14

23 Apr '14

Timothy M uploaded a new version of CMakeVMMaker to project VM Maker: http://source.squeak.org/VMMaker/CMakeVMMaker-tty.215.mcz ==================== Summary ==================== Name: CMakeVMMaker-tty.215 Author: tty Time: 23 April 2014, 4:32:27.764414 pm UUID: 807800ba-ed49-4d2c-858b-e3c0d81fe7cf Ancestors: CMakeVMMaker-tty.214 Modified class comment in CPlatformConfig to refer to the correct repository on github instead of gittorious =============== Diff against CMakeVMMaker-EstebanLorenzano.213 =============== Item was changed: Object subclass: #CPlatformConfig instanceVariableNames: 'topDir buildDir srcDir platformsDir outputDir internalPlugins externalPlugins pluginsDir externalLibs thirdpartyLibs' classVariableNames: '' poolDictionaries: '' category: 'CMakeVMMaker'! + !CPlatformConfig commentStamp: 'tty 4/23/2014 16:30' prior: 0! - !CPlatformConfig commentStamp: 'MarianoMartinezPeck 4/24/2011 15:29' prior: 0! Base class, defining some common properties and behavior for all kinds of VM build configurations. These classes are used to generate the VM sources and the CMMake files. + Notice that this configuration classes expect that you have download the "platofroms code" which is hosted here: + https://github.com/pharo-project/pharo-vm. - Notice that this configuration classes expect that you have download the "platofroms code" which is in Gitorious: https://gitorious.org/cogvm - You can load it executing: git clone git://gitorious.org/cogvm/blessed.git + You can clone it using this command: git clone https://github.com/pharo-project/pharo-vm + Once that is done it should have created a directory /blessed. In order to get all the directories by default, place this image under /blessed/image To generate a sources/build configuration use one of my subclasses with: <config> generate. - to generate a build configuration <config> generateWithSources - to generate VMMaker sources and then build configuration. Once you do that, go to /blessed/build and execute: cmake . or cmake -G"MSYS Makefiles" if you are in Windows Since some plugins require extra settings (like additional source files etc), there is two ways to add that: - add custom rules on plugin class side: PluginClass>>generateFor: aCMakeVMGenerator internal: aBoolean ^ aCMakeVMGenerator generatePlugin: self internal: aBoolean extraRules: [:maker | ... your rules come here ... ] - define rules by adding #configure<PluginClass>: method in config class (see my 'plugin extra rules' category) The last one is more compact and also avoids inter-package dependency, so you can load and use CMakeVMMaker even if some 3rd-party plugins are not loaded into image. Links of interest: Official Cog Website: http://www.mirandabanda.org/cog Official Cog Blog: http://www.mirandabanda.org/cogblog Cog issue tracker: http://code.google.com/p/cog Cog Wiki: http://code.google.com/p/cog/w/list Cog binaries: http://www.mirandabanda.org/files/Cog/VM Cog binaries in Pharo Hudson server: https://pharo-ic.lille.inria.fr/hudson/view/Cog Cog SVN branch: http://squeakvm.org/svn/squeak/branches/Cog VM mailing list: http://lists.squeakfoundation.org/mailman/listinfo/vm-dev VM-beginners mailing list: http://lists.squeakfoundation.org/mailman/listinfo/vm-beginners Guide step by step about how to build the Cog VM using CMakeVMMaker and Git: http://code.google.com/p/cog/wiki/Guide !

1 0

Spur performance report...
by Eliot Miranda 23 Apr '14

23 Apr '14

Hi All, I should write a blog post on this, but I can't wait... In recent days I've written a script to build a Cog VMMaker image from Squeak 4.5. See http://www.squeakvm.org/svn/squeak/branches/Cog/image. This has allowed me to run the current Cog VM against Spur side-to-side. To build a Cog VMMaker image on my 2.2GHz Intel Core i7 MacBook Pro using the current Cog VM takes about 2 and a half minutes: McStalker.image$ time oscfvm CogVMMaker.image BuildSqueak45Image.st real 2m30.671s user 2m15.683s sys 0m5.283s To build the equivalent image using Spur takes about 1 and a half minutes: McStalker.image$ time spurcfvm CogVMMaker-spur.image BuildSqueak45Image.st real 1m34.943s user 1m23.666s sys 0m6.810s Comparing: 94.943 - 150.671 / 150.671 * 100 -36.99 83.666 - 135.681 / 135.681 * 100 -38.34 150.671 / 94.943 1.59 135.681 / 83.666 1.62 that's about a -37% speedup, or 1.6x faster. -- best, Eliot

9 12

[commit][2903] Remember to load ImageFormat and the refactoring engine.
by commits＠squeakvm.org 22 Apr '14

22 Apr '14

Revision: 2903 Author: eliot Date: 2014-04-22 14:58:09 -0700 (Tue, 22 Apr 2014) Log Message: ----------- Remember to load ImageFormat and the refactoring engine. Simplify squashing the update completed informer (thanks Tobias!). Modified Paths: -------------- branches/Cog/image/BuildSqueak45Image.st branches/Cog/image/BuildSqueakTrunkImage.st Modified: branches/Cog/image/BuildSqueak45Image.st =================================================================== --- branches/Cog/image/BuildSqueak45Image.st 2014-04-22 20:32:33 UTC (rev 2902) +++ branches/Cog/image/BuildSqueak45Image.st 2014-04-22 21:58:09 UTC (rev 2903) @@ -1,13 +1,14 @@ | manifest load | manifest := #( ('http://source.squeak.org/FFI' 1 ('FFI-Pools' 'FFI-Kernel')) - ('http://source.squeak.org/VMMaker' 5 ('Balloon-Engine-Pools' 'VMMaker.oscog' 'Cog' 'CogTools')) + ('http://source.squeak.org/VMMaker' 6 ('Balloon-Engine-Pools' 'VMMaker.oscog' 'Cog' 'CogTools' 'ImageFormat')) ('http://ss3.gemstone.com/ss/MethodMassage' 3 ('MethodMassage' 'MethodMassageCompatibility')) - ('http://www.squeaksource.com/AioPlugin' 6 ('VMConstruction-Plugins-AioPlugin')) + ('http://www.squeaksource.com/AioPlugin' 7 ('VMConstruction-Plugins-AioPlugin')) ('http://www.squeaksource.com/Alien' 0 ('Alien')) - "('http://www.squeaksource.com/FreeTypePlus' 7 ('FreeType')) can't load this. it is toxic to Squeak 4.5" + "('http://www.squeaksource.com/FreeTypePlus' 5 ('FreeType')) can't load this. it is toxic to Squeak 4.5" ('http://www.squeaksource.com/FreetypePlugin' 8 ('Freetype-Plugin')) ('http://www.squeaksource.com/OSProcess' 4 ('OSProcess')) ('http://www.squeaksource.com/OSProcessPlugin' 9 ('VMConstruction-Plugins-OSProcessPlugin.oscog')) + ('http://www.squeaksource.com/rb' 5 ('AST-Core' 'AST-Semantic' 'AST-Tests-Core' 'AST-Tests-Semantic' 'Refactoring-Changes' 'Refactoring-Core' 'Refactoring-Environment' 'Refactoring-Tests-Changes' 'Refactoring-Tests-Core' 'Refactoring-Tests-Environment')) ('http://www.squeaksource.com/Speech' 2 ('SharedPool-Speech' ))). load := (manifest collect: Modified: branches/Cog/image/BuildSqueakTrunkImage.st =================================================================== --- branches/Cog/image/BuildSqueakTrunkImage.st 2014-04-22 20:32:33 UTC (rev 2902) +++ branches/Cog/image/BuildSqueakTrunkImage.st 2014-04-22 21:58:09 UTC (rev 2903) @@ -1,12 +1,7 @@ -"Change the defaut update URL to that for trunk, then install a DummyUIManager - to avoid the post-update OK box, then update, the restore the UIManager." -| uim | -uim := Project current uiManager. -[Project current instVarNamed: 'uiManager' put: DummyUIManager new. - MCMcmUpdater +"Change the update stream to trunk, update, then build as per 4.5" +[MCMcmUpdater defaultUpdateURL: 'http://source.squeak.org/trunk'; - updateFromServer] - ensure: [Project current instVarNamed: 'uiManager' put: uim]. + updateFromServer] valueSupplyingAnswer: true. "don't use fileIn; it causes an error when the saved image is started" Compiler evaluate: (FileStream oldFileNamed: 'BuildSqueak45Image.st') contentsOfEntireFile

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Vm-dev April 2014