Claude has a code review mode. One of the phases does performace improvements. Also, CLAUDE has done comparisons with the timing of the full test suite on the Mac standard vm vs our vm. Some digital math factorial functions were missing. But after that, it all seems to come down to the lack of JIT. It hurts a lot.
Claude suggested pre-compiling (at build time) all hotspots into C code. This would be part of a fake JIT. You could change the Smalltalk code, and it would invalidate the JIT and get slow. But only for that function. I suppose if you made a change to the Object, all JIT would be lost. But I haven't looked into it yet. There was also a suggestion to have a serious Smalltalk group petition Apple for a workaround to the JIT ban. Originaly Apple banned all code that can write code. But I recently published a Z80 CP/M emulator, and Apple had no problem with that. Although they did hold up the Mac version half a month due to a zork1 copyright inquiry (but the iPad with zork1 was fine). So Apple can loosen up and still be squirrely at the same time. ----- Original message ----- From: Tim Mackinnon <[email protected]> To: Any question about pharo is welcome <[email protected]>, Richard O'Keefe <[email protected]> Cc: Aaron Wohl <[email protected]> Subject: Re: [Pharo-users] Re: iPad Pharo test release Date: Monday, March 02, 2026 5:21 AM Its worth adding to this, to set expectations - I have been testing on an iPhone 13 pro (so a few years old) and an iPad Pro 2nd Gen (from about the same time period). While the iPad has more CPU cores, the A15 bionic chip of the iPhone does run this better, and its a bit sluggish on the ipad and a bit nippier on the iPhone. If I change the Color Theme from the startup screen, both devices will pause for 15seconds or more while they process that - however things like code lookup or evaluating factorial, or dragging windows all are acceptable. An iPad 6 is 30-70% slower than my iPad pro, so I suspect it will work but be quite laggy without the jitter. Still its a worth a shot. I think that modern M series ipads will probably work quite well - and the lack of a jitter might not be a big impediment. It worth a shot, and maybe Claude might spot some interesting optimisations - its kind of amazing how its churning through my feedback to Aaron so far. Today its added useful toolbar icons and its starting to look like a viable part time coding environment - i'm quite stunned. Tim On Mon, 2 Mar 2026, at 12:53 AM, Aaron Wohl via Pharo-users wrote: > The motion code is not released yet. Probably tomorrow > > can you reasearch the earliest iPad and or iOS this will run on. a user > writes ... > Here's the research summary: > > > *iPad 6 (A1893) — Yes, it will work.* > > > > - iPad 6 has an A10 Fusion (arm64) and supports up to iPadOS 17 > > - Our deployment target is iOS/iPadOS 15.0, so it's well within range > > - It won't get iPadOS 18+, but iPadOS 17 is fine for us > > > > *Minimum* *supported* *devices* *(iOS* *15.0* *deployment* *target):* > > > > iPads: iPad 5th gen (2017) and newer, iPad Air 2 (2014), iPad mini 4 > (2015) > iPhones: iPhone 6s / 6s Plus (2015), iPhone SE 1st gen (2016) > > *Storage* *requirements:* > > * * *App* *itself* *(debug* *build):* *~15* *MB* > * * *Pharo* *13* *image:* *~56* *MB* > * * *Pharo* *sources* *file:* *~43* *MB* > * * *Changes* *file:* *~20* *KB* > * * *TestFlight* *app:* *~30* *MB* *(from* *App* *Store)* > > Total: ~150 MB for everything > > So even a 32 GB iPad 6 has plenty of room. > > *Re:* *TestFlight* *instructions* — that's a good idea for the README. > *Re:* *Core* *Motion* *binding* — that would be a fun Pharo FFI > project (accelerometer, gyroscope, compass on iPad). > > Want me to update the README with TestFlight install instructions? > ... > Added the following: > | data | > │ > │ "Start motion updates" > │ > │ <primitive: 'primitiveMotionStart' module: 'CoreMotionPlugin'>. > │ > │ > │ > │ "Read latest sample — Float64Array of 13 elements" > │ > │ data := <primitive: 'primitiveMotionData' module: 'CoreMotionPlugin'>. > │ > │ "Elements: accelX,Y,Z (1-3), gyroX,Y,Z (4-6), magX,Y,Z (7-9), > │ > │ roll,pitch,yaw (10-12), timestamp (13)" > │ > │ > │ > │ "Stop updates" > │ > │ <primitive: 'primitiveMotionStop' module: 'CoreMotionPlugin'>. > │ > > ----- Original message ----- > From: Richard O'Keefe <[email protected]> > To: Any question about pharo is welcome <[email protected]> > Cc: Aaron Wohl <[email protected]> > Subject: Re: [Pharo-users] iPad Pharo test release > Date: Sunday, March 01, 2026 7:10 PM > > I have an iPad 6 (model A1893 if anyone cares). > Will Pharo run on that device? > How much storage do I need to install Pharo on this iPad? > (And TestFlight itself.) > I've never heard of the test store before; could the README in GitHub > be extended with instructions about how to install the beta? > > If you had infinite time and I had money to pay for it ... > it would be nice to have a binding for Core Motion. > But having Pharo on the iPad? Huge thanks. > > On Thu, 26 Feb 2026 at 02:19, Aaron Wohl via Pharo-users > <[email protected]> wrote: >> Pharo 13 for iPad in the test Apple Store (open in the Apple TestFlight app); >> Phone or Mac https://testflight.apple.com/join/kGmPQFr9 >> Due to Apple’s no JIT, it's not snappy. But maybe it's still useful? >> Source https://github.com/avwohl/iospharo about 100K lines of new source >> code. >> >> This is my first venture into using Pharo. I could use advice on where this >> project fits in Pharo/Smalltalk land. In the early 80s, I used the Xerox >> Parc Alto at Carnegie Mellon to run Smalltalk. I got laid off, went on a >> cruise, sat down in a beach chair with LEARN PHARO and an iPad. But, no >> Pharo for iPad. So I did the natural thing and told Claude Code to port >> Pharo to iPad. 54 days and 1347 commits later, out popped a usable iPad >> app. I do have a background in compilers, working on the Production Quality >> Compiler project at CMU in the early 1980s, mostly on lexers. But I didn’t >> write any code for this. I mostly told CLAUDE a few times a day stop doing >> workarounds, do what the real VM does, and that’s a workaround. CLAUDE >> would never have succeeded I think if I hadn’t started insisting a focus on >> getting the test suite to pass rather than the app to work. >> >> The main snag using the real VM was encoding the encoding of type info in >> the high bits of pointers for immediate types. It seemed simple to me to >> shift the type info to the low 3 bits. But Claude could not get the real VM >> to stop looking in the high bits; perhaps I should have pushed harder on >> that. But I had always heard how easy it was to port Smalltalk, so I just >> went with a new VM with the type info in the low bits. Loading a standard >> image translates the addresses. >> The other issue was that Apple forbids JIT on iPad. So this new VM has no >> JIT. >> Issues: >> - Should this be a separate VM? Should I push harder on getting the standard >> VM to work with type info in the low bits and no JIT? Even if that would be >> better, is it worth the bother now that this works? >> >> - I don’t want to suck up the PharoSmalltalk name in the Apple stores if it >> causes offense. I can call it VibeTalk or something to leave the name free >> for a hand-coded version. Or move off the name if a real contender ever >> shows. >> >> - Without JIT, is it too slow to bother releasing? Is there some other >> optimization that could bridge the gap if it is too slow? >> >> - Are there any users who can give the iPad version a really good workout in >> real work? The VM passes the test suite. But it just started working on >> the iPad today. There will be UI issues. >> >> - On the iphone the menu bar is cut off, and the welcome window is mostly >> off the screen. The menu bar layout is in the image, not the VM. I am not >> sure the VM should work around it. Should these be fixed in the image to >> have a multi-line menu bar as needed, and reshape initial window layouts on >> phones? If so is anyone willing to do the Smalltalk side or help me do it >> (I have written mostly other languages) >
