Re: [racket-users] Another futures-related bug hunt

2020-05-08 Thread Dominik Pantůček
Hello, The most useful information here is likely to be a stack trace from each OS-level thread at the point where the application is stuck. would this be enough to open an issue for that? (gdb) info threads Id Target IdFrame * 1Thread 0x7

Re: [racket-users] for/set

2020-05-08 Thread Hendrik Boom
On Fri, May 08, 2020 at 04:18:24PM -0700, Sorawee Porncharoenwase wrote: > There’s a search bar in the top left of every doc page. Somehow I hadn't noticed that. It's a huge help to know that. > Typing for/set and > enter will show you this search result page >

Re: [racket-users] for/set

2020-05-08 Thread Stephen De Gabrielle
It’s with the set data type - but I haven’t used it https://docs.racket-lang.org/reference/sets.html#%28form._%28%28lib._racket%2Fset..rkt%29._for%2Fset%29%29 On Sat, 9 May 2020 at 00:07, Hendrik Boom wrote: > Where is for/set documented? What does it do? > > It is mentioned in the typed Rack

Re: [racket-users] for/set

2020-05-08 Thread Sorawee Porncharoenwase
There’s a search bar in the top left of every doc page. Typing for/set and enter will show you this search result page . Here, the first result

[racket-users] for/set

2020-05-08 Thread Hendrik Boom
Where is for/set documented? What does it do? It is mentioned in the typed Racket menual, but oly to say it has the same meaning as the unannotated version. -- hendrik -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this

Re: [racket-users] How to run multiple Racket installations?

2020-05-08 Thread James Platt
Would virtual machines be an option? You do have to have a pretty good host machine with lots of RAM. I do this mainly to have different development and testing environments. It works pretty smoothly on my Mac Pro, with VirtualBox for Linux and Windows guest machines and VMWare for macOS gues

Re: [racket-users] How to avoid all packages when building from source?

2020-05-08 Thread Sam Tobin-Hochstadt
For Racket CS, you'd want `make cs` or `make cs-base`. `PKGS` is what gets built. The `racket/pkgs` directory is packages that are part of the `racket/racket` repository; it includes things like tests, the guide and reference, and a few other packages that make sense to develop in the same reposit

Re: [racket-users] How to avoid all packages when building from source?

2020-05-08 Thread zeRusski
> You can do "make base" instead, which installs no packages. Or you can do > something like 'make PKGS=drracket' which just installs DrRacket and > dependencies, or similar with other packages. > I am installing in-place but racket cs, not racket bc, so IIUC `make base` isn't what I want. P

Re: [racket-users] Another futures-related bug hunt

2020-05-08 Thread Sam Tobin-Hochstadt
You will want to do `handle SIGSEGV nostop noprint` when you start gdb. Racket BC uses the SEGV handler to implement the GC write barrier, so you'll want to skip those. Sam On Fri, May 8, 2020 at 9:36 AM Dominik Pantůček wrote: > > Hello, > > On 08. 05. 20 14:27, Matthew Flatt wrote: > > At Fri

Re: [racket-users] Another futures-related bug hunt

2020-05-08 Thread Dominik Pantůček
Hello, On 08. 05. 20 14:27, Matthew Flatt wrote: At Fri, 8 May 2020 09:34:32 +0200, Dominik Pantůček wrote: Apart from obvious strace (after freeze) and gdb (before/after freeze) debugging to find possible sources of this bug, is there even a remote possibility of getting any clue how can this

Re: [racket-users] Another futures-related bug hunt

2020-05-08 Thread Matthew Flatt
At Fri, 8 May 2020 09:34:32 +0200, Dominik Pantůček wrote: > Apart from obvious strace (after freeze) and gdb (before/after freeze) > debugging to find possible sources of this bug, is there even a remote > possibility of getting any clue how can this happen based on the > information gathered s

Re: [racket-users] How to run multiple Racket installations?

2020-05-08 Thread Matthew Flatt
At Fri, 8 May 2020 01:55:17 -0700 (PDT), zeRusski wrote: > First, does that even work? I noticed that both of them install packages > into ~/Library/Racket/development/ for me. Are both builds so compatible I > don't need to worry about packages stepping on each others toes i.e. > compiled with

Re: [racket-users] How to avoid all packages when building from source?

2020-05-08 Thread Sam Tobin-Hochstadt
The default build installs "main-distribution", which is the same as what you get for a regular (not minimal) Racket install. It does, as you noticed, install a lot of packages. You can do "make base" instead, which installs no packages. Or you can do something like 'make PKGS=drracket' which just

Re: [racket-users] How to run multiple Racket installations?

2020-05-08 Thread Sam Tobin-Hochstadt
This is the tooling I use: https://github.com/takikawa/racket-dev-goodies/ Note that it works primarily with "in-place" installations. Sam On Fri, May 8, 2020, 4:55 AM zeRusski wrote: > I have two builds of Racket on my local machine. Racket CS resides in one > directory and was built with `RA

[racket-users] How to avoid all packages when building from source?

2020-05-08 Thread zeRusski
I just rebuilt Racket from git repo checkout. It takes a while but most of that time is spent in `raco setup` which appears to be building all packages in existence. E.g. games, redex-examples, realm of Racket chapter 6 (really?), plai, algol, etc. Why do I end up with the entire jungle? Is the

[racket-users] How to run multiple Racket installations?

2020-05-08 Thread zeRusski
I have two builds of Racket on my local machine. Racket CS resides in one directory and was built with `RACKETCS_SUFFIX=""` and stardard Racket also built from source in a separate directory. Normally I have my .bashrc setup PATH as needed to use e.g. Racket CS. I ran into a problem with an upst

[racket-users] Another futures-related bug hunt

2020-05-08 Thread Dominik Pantůček
Hello fellow Racketeers, my spare-time out-of-curiosity venture into using HPR (High-Performance Racket) for creating a software 3D rendering pipeline seems to be pushing the futures into rough edges. The scenario is sort of "usual": * 7 futures + 1 in RTT that form a binary tree * GUI threa