VistaLover Posted August 19, 2020 Share Posted August 19, 2020 (edited) 7 hours ago, RainyShadow said: Latest Serpent 52, page: https://www.kongregate.com/games/baba_s/economical Quote Your browser does not support WebAssembly. I thought Basilisk/Serpent supports that, huh?! Latest St52 has buildID=20200814013442; WASM has been indeed supported since long ago, even MozillaESR 52.6.0, the initial fork-point of UXP, came with native support... Whether that native support is ON/OFF by default is a different matter, though... MozillaESR 52.6.0 had all native WASM support pref'ed OFF by default; in latest St52, the behaviour is determined by following two commits: [Basilisk] Issue MoonchildProductions/UXP#1611 - Use platform default for WASM https://github.com/roytam1/UXP/commit/95c5a64 Issue #1611 - Enable WASM by default but only enable jit when 64bit https://github.com/roytam1/UXP/commit/ed0edf6 Indeed, when I launch latest 32-bit build in a NEW/CLEAN profile, this is what I get: i.e. WASM is ON, but baseline just-in-time (jit) is OFF; if that isn't the case in your set-up, then something altered the default settings, either a (security?) configuration imposed by an extension or you did it yourself (directly in about:config, using a user.js file, etc) in the past, possibly following online advice, and forgot about it... FWIW, visiting the linked URL in the above CLEAN profile, I get no warning of the kind you reported; the site does require Adobe Flash though, and it took a considerable amount of waiting for the game to fully load ; but it did eventually: OTOH, on my very old/DIRTY profile, with many extensions (some of which of the WebExtension format), many userscripts (in GM-for-PM), many userstyles (in Stylem + Stylus), uB0-legacy with many filter lists, the game never loads fully , because of async OOM errors: ... so YMMV even with WASM turned back ON... Edited August 19, 2020 by VistaLover Link to comment Share on other sites More sharing options...
RainyShadow Posted August 19, 2020 Share Posted August 19, 2020 (edited) @VistaLover My settings in about:config are as in your screenshot. I also tested on a clean profile with the same result. I allow Flash on that site, but it didn't make a diference in this case. Later i checked few WASM test pages, but it never worked. What "lack of compiler support" means? Is this a message from the browser, or from the loader script? Maybe it has something to do with my ancient AthlonXP CPU? It won't be a surprise since that stupid Unity is involved... But i don't think the other pages i tried use it too. Edited August 19, 2020 by RainyShadow Link to comment Share on other sites More sharing options...
Sampei.Nihira Posted August 19, 2020 Share Posted August 19, 2020 On 8/17/2020 at 6:58 PM, Sampei.Nihira said: @roytam1 Hi, The new MailNews does not overwrite the Binary Outcast registry values. You can check it in the Start menu by setting the e-mail program. I understand this is a trivial matter, but I changed the register values myself, because the old ones bothered me. Yes I know, I'm paranoid. 1 Link to comment Share on other sites More sharing options...
UCyborg Posted August 19, 2020 Share Posted August 19, 2020 (edited) 16 hours ago, RainyShadow said: What "lack of compiler support" means? Is this a message from the browser, or from the loader script? It comes from the browser. One of the checks that must pass for the browser to even bother with WebAssembly is SSE2 capable CPU. 18 hours ago, VistaLover said: the site does require Adobe Flash though The base site seems to work without it, looks like some hidden Flash file lurking in the background. Flash games obviously won't work without it. Edited August 19, 2020 by UCyborg 2 Link to comment Share on other sites More sharing options...
VistaLover Posted August 19, 2020 Share Posted August 19, 2020 (edited) 23 hours ago, RainyShadow said: What "lack of compiler support" means? Is this a message from the browser, or from the loader script? The exact URI to that script is https://game314425.konggames.com/gamez/0031/4425/live/Build/UnityLoader.js I downloaded it locally on disk and probed it with an editor ; I couldn't find "Disabled by lack of compiler support", but did find the original warning/error: "Your browser does not support WebAssembly." Looking closer at your Web Console output, it appears the message is generated by the browser's asm.js module, when it tried to compile js code fed to it by the UnityLoader.js script ; BTW, many thanks @UCyborg for your most helpful contribution, as always : 6 hours ago, UCyborg said: It comes from the browser. One of the checks that must pass for the browser to even bother with WebAssembly is SSE2 capable CPU. 23 hours ago, RainyShadow said: Maybe it has something to do with my ancient AthlonXP CPU? You might've mentioned it previously in these forums, but it didn't dawn on me that that was the case... I am running Vista SP2 32-bit and my 2007 era Intel Core 2 Duo is SSE2 capable; so using the "-xpmod" variety of Serpent-52.9.0-win32... By the looks of it, you should be running the "-xpmod-ia32" build, derived from the "ia32" branch of Roytam1's UXP fork: https://github.com/roytam1/UXP/commits/ia32 TBH, I don't follow closely the development of that branch, only that of the custom branch (produces NM28+St52)... 23 hours ago, RainyShadow said: It won't be a surprise since that stupid Unity is involved... But i don't think the other pages i tried use it too. @UCyborg wrote that WASM expects at least SSE2 instructions set , so there's your answer... @roytam1 : Perhaps it could be a good idea to disable WASM and LOCK the related prefs (inside about:config) in your ia32 builds, so as not to create false expectations by the users of those builds on their pretty old processors... Just throwing this out there for consideration... 6 hours ago, UCyborg said: The base site seems to work without it, looks like some hidden Flash file lurking in the background. As posted, I first tried the site with a pristine St52 profile, and it did ask permission to use Flash; Flash is used to deliver ads in the embedded player, just before the game itself loads (when it does... ). Edited August 20, 2020 by VistaLover 2 Link to comment Share on other sites More sharing options...
TRX101 Posted August 19, 2020 Share Posted August 19, 2020 On 8/13/2020 at 6:06 PM, gerwin said: Yeah, I am not really sure yet, but experiencing something similar with NewMoon 28 in the past weeks (using 28.8.0a1 2019-12-06 and 28.10.1a1 2020-07-04). Not really minutes but more like 10 to 20 seconds. Triggering me to try and close the tab that I suspect is causing it. But then it won't let me. Also windows things like show desktop (minimize all) are then unresponsive for a little while. Probably caused by some script acting up and causing problems with GPU. Also Baselisk shows same problems and PM27 recently can be sluggish. My box : Athlon 64 X2 4400+ GeForce 7300 LE Link to comment Share on other sites More sharing options...
roytam1 Posted August 20, 2020 Author Share Posted August 20, 2020 15 hours ago, VistaLover said: @roytam1 : Perhaps it could be a good idea to disable WASM and LOCK the related prefs (inside about:config) in your ia32 builds, so as not to create false expectations by the users of those builds on their pretty old processors... Just throwing this out there for consideration... I did think about adding code to ignore WASM related prefs when SSE2 is not detected. 2 Link to comment Share on other sites More sharing options...
RainyShadow Posted August 20, 2020 Share Posted August 20, 2020 2 hours ago, roytam1 said: I did think about adding code to ignore WASM related prefs when SSE2 is not detected. Let me repeat a question that went by unanswered in the past... Is it possible to use the SSEPlus project as a SSE2/3/4/xxx emulator in your builds? Would it be simple enough change for you to consider it (as in a batch of "Replace All" on the source) or it would need too much work? 1 Link to comment Share on other sites More sharing options...
roytam1 Posted August 21, 2020 Author Share Posted August 21, 2020 9 hours ago, RainyShadow said: Let me repeat a question that went by unanswered in the past... Is it possible to use the SSEPlus project as a SSE2/3/4/xxx emulator in your builds? Would it be simple enough change for you to consider it (as in a batch of "Replace All" on the source) or it would need too much work? it's not that simple as replacing functions, many of them are done in pure assembly. unless sseplus uses system-wide trap or this can't be done. 1 Link to comment Share on other sites More sharing options...
NotHereToPlayGames Posted August 21, 2020 Share Posted August 21, 2020 Does anybody else use v27 and noticed that their Permissions Manager does not work and hasn't worked for over a YEAR ??? Link to comment Share on other sites More sharing options...
RainyShadow Posted August 21, 2020 Share Posted August 21, 2020 @ArcticFoxie It seems to be working here. How it does not work for you? Not shown at all? Link to comment Share on other sites More sharing options...
roytam1 Posted August 22, 2020 Author Share Posted August 22, 2020 New build of Serpent/UXP for XP! Test binary: Win32 https://o.rths.ml/basilisk/basilisk52-g4.6.win32-git-20200822-57e81f0-uxp-dd1ad91cb-xpmod.7z Win64 https://o.rths.ml/basilisk/basilisk52-g4.6.win64-git-20200822-57e81f0-uxp-dd1ad91cb-xpmod.7z source code that is comparable to my current working tree is available here: https://github.com/roytam1/UXP/commits/custom IA32 Win32 https://o.rths.ml/basilisk/basilisk52-g4.6.win32-git-20200822-57e81f0-uxp-dd1ad91cb-xpmod-ia32.7z source code that is comparable to my current working tree is available here: https://github.com/roytam1/UXP/commits/ia32 NM28XP build: Win32 https://o.rths.ml/palemoon/palemoon-28.10.2a1.win32-git-20200822-11ba7a43c-uxp-dd1ad91cb-xpmod.7z Win64 https://o.rths.ml/palemoon/palemoon-28.10.2a1.win64-git-20200822-11ba7a43c-uxp-dd1ad91cb-xpmod.7z Official UXP changes since my last build: - Issue #1629 - Part 1: Implement basic logic in HTMLLinkElement. (a9f337ea7) - Issue #1629 - Part 2: Implement the Explicitly Enabled flag. (686232635) - Issue #1629 - Part 3: Implement behind preference. (931e8d29f) - Fix nits (2372e1902) - Issue #1620 - Intrinsic Aspect Ratio: Debug Follow up. (22c672813) - Merge pull request #1635 from g4jc/1620 (56f8262c8) - Merge pull request #1632 from athenian200/link_element_disabled (7ca7fe802) - [db] Update SQLite lib to 3.33.0 (9074e6928) - Issue #457 - Silence SQLite compiler warning (46595e390) - [js] Reinstate precise floating point model for all js sources. (1b14edc01) - Follow up to eb28b1f32 - Correct inputmethod build file by putting Keyboard.jsm back in EXTRA_JS_MODULES (dd1ad91cb) No official Basilisk changes since my last build. Official Pale-Moon changes since my last build: - Issue #1826 - Restore missing close icon on the undo prompt at the quick dial page (1227e3ce1) - Issue #1717 - Padlock Improvements (f746a96f7) - Issue #1717 - Low Cipher Changes (c3d5f8ae1) - Issue #1717 - 3DES Detection (78aac837d) - Issue #1717 - Localize Padlock Tooltip (9641725f6) - Issue #1717 - Whitespace Cleanup (1ff8522d7) - Issue #1717 - Revert 3DES Check to Original Method for Now (b81e1c204) - Issue #1717 - UXP Says Mixed Content is Broken (b76ba6161) - Issue #1717 - Differentiate DV & EV Mixed Content (ea7f0b6d6) - Issue #1717 - Total Level Rewrite (1f488302c) - Merge pull request #1828 from RealityRipple/master (30987e613) - Update platform and version (Unstable 2020-08-19) (0aa10d730) - Merge pull request #1827 from FranklinDM/quickdial-close-button-work (11ba7a43c) 5 Link to comment Share on other sites More sharing options...
roytam1 Posted August 22, 2020 Author Share Posted August 22, 2020 New build of BOC/UXP for XP! Test binary: MailNews Win32 https://o.rths.ml/boc-uxp/mailnews.win32-20200822-cbe6b975-uxp-dd1ad91cb-xpmod.7z BNavigator Win32 https://o.rths.ml/boc-uxp/bnavigator.win32-20200822-cbe6b975-uxp-dd1ad91cb-xpmod.7z source repo (excluding UXP): https://github.com/roytam1/boc-uxp/tree/custom Official repo changes since my last build: - Move DOMi to its own repo to be used with binoc-build (941e49f3) - Move abprime back to its own repo (40635054) - [Navigator] Add an empty all_panels manifest for the sidebar so it stops trying a long dead URL (ee99d6d2) - [Navigator] Enhance the security status bar panel WITH SECURITY! (6314d848) - Update platform commit pointer (24534bb7) - Update platform commit pointer (7632bc8e) - [Navigator] Add an updated NSIS Windows Installer (7c86790f) - [Navigator] Add the replacement shell service and make it build (42f20e22) - [Navigator] Get Set as Wallpaper working and redesigned (7b06ec1d) - [Navigator] Fix default settings for toolbars (185961f8) - Use brand name for taskNavigator (1fe9384c) - [Navigator] Fix nsModule for Linux and Mac (6bc0adc3) - [Navigator] Fix ShellService.jsm (cbe6b975) * Notice: since official names are not allowed to be used in resulting binaries, the profile prefix (i.e. parent folder names) are also changed, you may rename their names before using new binaries when updating from builds before 2020-08-15. -- New build of IceApe-UXP for XP! Test binary: https://o.rths.ml/gpc/files1.rt/iceape.win32-20200822-id-eed056673-ia-41157bf-uxp-dd1ad91cb-xpmod.7z for UXP changes please see above. 1 Link to comment Share on other sites More sharing options...
roytam1 Posted August 22, 2020 Author Share Posted August 22, 2020 New NewMoon 27 Build! 32bit https://o.rths.ml/palemoon/palemoon-27.9.7.win32-git-20200822-346d6556f-xpmod.7z 32bit SSE https://o.rths.ml/palemoon/palemoon-27.9.7.win32-git-20200822-346d6556f-xpmod-sse.7z 32bit noSSE https://o.rths.ml/palemoon/palemoon-27.9.7.win32-git-20200822-346d6556f-xpmod-ia32.7z 64bit https://o.rths.ml/palemoon/palemoon-27.9.7.win64-git-20200822-346d6556f-xpmod.7z source repo: https://github.com/roytam1/palemoon27 repo changes since my last build: - import changes from `dev' branch of rmottola/Arctic-Fox: - remove SymbolIsWellKnown (24eb6e405) - Bug 1159428 - Use Value and Id type-based dispatching to simplify barriers; r=jonco (ba7eace5b) - pointer style (9ad6a6c0c) - Bug 1146902 - be smarter about inlining atomics. r=jandem (f5755a545) - remove assertion (2c515ccf5) - Bug 1134074 - IonBuilder: Atomize strings when inlining String.split. r=djvj (bb24c18db) - Bug 1148883 - Atomize inlined constants of bound functions. r=h4writer (a956afd3c) - minor cleanup (f73195b88) - cleanup (9503bf8c0) - fix comment (a01abf294) - Bug 1148661 - Optimize accesses to unboxed expandos in Ion, r=jandem. (9bfbd35c0) - pointer style (12871c520) - Bug 1148916 - Remove MaybeSingletonObject NewObjectKind, r=terrence. (daa565d6f) - Bug 1094491 - Generate baseline getElem ICs for accessor properties before calling the getter. r=efaust (43c6515a4) - Bug 1049290 - Optimize JSOP_IN for baseline compiler. r=djvj, r=h4writer (f665c669a) - Bug 1141862 - Part 1: Free up bits in JSFunction for isMethod() and, later, isClassConstructor(). (r=jandem) (c19aa5c98) - Bug 1141862 - Part 2: Allow Lazy script to pass the Method FunctionSyntaxKind during delazification. (r=jorendorff) (b640e7b2a) - pointer style (630ee869d) - Bug 1141862 - Part 3: Remove unused argument from BytecodeEmitter::emitPropLHS. (r=jorendorff) (10690d460) - Bug 1141862 - Part 4: Make everything defined with MethodDefinition syntax use the Method FunctionSyntaxKind while parsing. (r=jorendorff) (81efdc052) - Bug 1141862 - Part 5: Change JSOP_SETCALL to JSOP_THROWMSG, anticipating future use. (r=jorendorff) (73540a35a) - pointer style (e72321035) - Bug 1149769 - Split reflect-parse.js reftest into its own subdirectory. (r=Waldo) (4ca2a1850) - Bug 1149015 - Part 4: Remove some use of expression closure from jstests js1_8_5/reflect-parse/. r=luke (5068e1a53) - pointer style (c038fec03) - Bug 1141862 - Part 6: Implement ES6 SuperProperty and SuperMember. (r=jorendorff) (65535dde8) - Bug 1141862 - Followup: Remove erronously landed debug spew. r=me (bfee06ef5) - Bug 1141862 - Followup: Assure GCC that superBegin cannot, in fact, be used uninitialized. (rs=jorendorff) (d65ef8c05) (346d6556f) Link to comment Share on other sites More sharing options...
roytam1 Posted August 22, 2020 Author Share Posted August 22, 2020 (edited) New regular/weekly KM-Goanna release: https://o.rths.ml/kmeleon/KM76.2.1-Goanna-20200822.7z Changelog: Out-of-tree changes: * update Goanna3 to git bc7c63219..346d6556f: - import changes from `dev' branch of rmottola/Arctic-Fox: - remove SymbolIsWellKnown (24eb6e405) - Bug 1159428 - Use Value and Id type-based dispatching to simplify barriers; r=jonco (ba7eace5b) - pointer style (9ad6a6c0c) - Bug 1146902 - be smarter about inlining atomics. r=jandem (f5755a545) - remove assertion (2c515ccf5) - Bug 1134074 - IonBuilder: Atomize strings when inlining String.split. r=djvj (bb24c18db) - Bug 1148883 - Atomize inlined constants of bound functions. r=h4writer (a956afd3c) - minor cleanup (f73195b88) - cleanup (9503bf8c0) - fix comment (a01abf294) - Bug 1148661 - Optimize accesses to unboxed expandos in Ion, r=jandem. (9bfbd35c0) - pointer style (12871c520) - Bug 1148916 - Remove MaybeSingletonObject NewObjectKind, r=terrence. (daa565d6f) - Bug 1094491 - Generate baseline getElem ICs for accessor properties before calling the getter. r=efaust (43c6515a4) - Bug 1049290 - Optimize JSOP_IN for baseline compiler. r=djvj, r=h4writer (f665c669a) - Bug 1141862 - Part 1: Free up bits in JSFunction for isMethod() and, later, isClassConstructor(). (r=jandem) (c19aa5c98) - Bug 1141862 - Part 2: Allow Lazy script to pass the Method FunctionSyntaxKind during delazification. (r=jorendorff) (b640e7b2a) - pointer style (630ee869d) - Bug 1141862 - Part 3: Remove unused argument from BytecodeEmitter::emitPropLHS. (r=jorendorff) (10690d460) - Bug 1141862 - Part 4: Make everything defined with MethodDefinition syntax use the Method FunctionSyntaxKind while parsing. (r=jorendorff) (81efdc052) - Bug 1141862 - Part 5: Change JSOP_SETCALL to JSOP_THROWMSG, anticipating future use. (r=jorendorff) (73540a35a) - pointer style (e72321035) - Bug 1149769 - Split reflect-parse.js reftest into its own subdirectory. (r=Waldo) (4ca2a1850) - Bug 1149015 - Part 4: Remove some use of expression closure from jstests js1_8_5/reflect-parse/. r=luke (5068e1a53) - pointer style (c038fec03) - Bug 1141862 - Part 6: Implement ES6 SuperProperty and SuperMember. (r=jorendorff) (65535dde8) - Bug 1141862 - Followup: Remove erronously landed debug spew. r=me (bfee06ef5) - Bug 1141862 - Followup: Assure GCC that superBegin cannot, in fact, be used uninitialized. (rs=jorendorff) (d65ef8c05) (346d6556f) * Notice: the changelog above may not always applicable to XULRunner code which K-Meleon uses. A goanna3 source tree that has kmeleon adaption patch applied is available here: https://github.com/roytam1/palemoon27/tree/kmeleon76 Edited August 22, 2020 by roytam1 1 Link to comment Share on other sites More sharing options...
Recommended Posts