Re: [EQL5] Slow startup

2021-01-14 Thread pls.153
On Thursday, January 14, 2021 2:09 PM, Erik Winkels > Anyway, for now the startup time is good enough. FWIW, loading :asdf alone is faster than I previously thought. With the compiler already loaded (require :cmp), I get (on Sailfish with 32bit ECL): > (time (require :asdf)) ;;; Loading #P"/usr

Re: [EQL5] Slow startup

2021-01-14 Thread Erik Winkels
Hi, A bit of a late reply since the discussion has already continued but: On 2021-01-12 18:50, pls.153 wrote: So, my question is: if you don't add ASDF as a dependency, what error messages do you get at app startup? When I do not `(require :sb-bsd-sockets)` I get: Condition of type: SIM

Re: [EQL5] Slow startup

2021-01-13 Thread pls.153
On Wednesday, January 13, 2021 11:39 AM, Daniel Kochmański wrote: > Make sure you have uiop (a library) in your local projects or another > registry - otherwise asdf won't add it to the bundle because it is > 'preloaded' - and ql won't download it - ditto. thanks, I never needed uiop, but tha

Re: [EQL5] Slow startup

2021-01-13 Thread Daniel Kochmański
Make sure you have uiop (a library) in your local projects or another registry - otherwise asdf won't add it to the bundle because it is 'preloaded' - and ql won't download it - ditto. Daniel Wysłano z aplikacji ProtonMail Oryginalna wiadomość ‎ 13 sty 2021, 08:39, pls.153 na

Re: [EQL5] Slow startup

2021-01-12 Thread pls.153
‐‐‐ Original Message ‐‐‐ On Tuesday, January 12, 2021 1:30 PM, Erik Winkels wrote: > I've alleviated my slow startup somewhat by > > diff --git a/make.lisp b/make.lisp > index 41d4e3d..52b05b6 100644 > --- a/make.lisp > +++ b/make.lisp > @@ -13,7 +13,8 @@ > (asdf:make-build "app" > :monoli

Re: [EQL5] Slow startup

2021-01-12 Thread pls.153
On Tuesday, January 12, 2021 1:30 PM, Erik Winkels wrote: > I found this post (of yours?): > https://www.reddit.com/r/Common_Lisp/comments/hicmyt/error_with_uiop_running_ecl_application_built_by/g35fcib/ > > Is this what you meant? Where do you use this code? Yes, that was me. As (hopefully) ever

Re: [EQL5] Slow startup

2021-01-12 Thread Erik Winkels
Hi, On 2021-01-01 12:12, PR wrote: Loading ASDF on mobile _is_ slow, and there is nothing we can do about it, unfortunately. I've alleviated my slow startup somewhat by diff --git a/make.lisp b/make.lisp index 41d4e3d..52b05b6 100644 --- a/make.lisp +++ b/make.lisp @@ -13

Re: [EQL5] Slow startup

2021-01-08 Thread Erik Winkels
On 2021-01-01 13:56, PR wrote: I will try to add some example myself, so in future this (almost empty) repo will hopefully get populated: https://gitlab.com/eql/eql5-sailfish I was a little busy this week so only noticed this mail now. That looks good, glad someone more experienced with Q

Re: [EQL5] Slow startup

2021-01-01 Thread PR
On 1/1/21, Erik Winkels wrote: > Ah yes, welcome to Sailfish! Thanks, yeah it's a great OS, it's hard not to love it, once you've played around with it! I will try to add some example myself, so in future this (almost empty) repo will hopefully get populated: https://gitlab.com/eql/eql5-sailfis

Re: [EQL5] Slow startup

2021-01-01 Thread Erik Winkels
Hi, On Friday, January 1, 2021, PR wrote: > > Loading ASDF on mobile _is_ slow, and there is nothing we can do about it, > unfortunately. > > ASDF should therefore never be a dependency in a final app (there are ways to > avoid it, even if some libraries want it to be present.) Thanks, I'll m

Re: [EQL5] Slow startup

2021-01-01 Thread PR
On 12/28/20, Erik Winkels wrote: > ;;; Loading #P"/usr/lib/ecl-20.4.24/asdf.fas" > --> > Reading config "/home/nemo/.config/pusfofefe/config.lisp"... Config Loading ASDF on mobile _is_ slow, and there is nothing we can do about it, unfortunately. ASDF should therefore never be a depend

[EQL5] Slow startup

2021-01-01 Thread Erik Winkels
Hi, in my app for Sailfish OS I've got a really long startup time compared to other apps. (On the first load it is especially long, I assume because Quicklisp is loading systems.) The longest wait and the one that would make startup time good enough for me if it was gone is marked with "-->":