Re: [perl #131889] Segfault / malloc errors testing Uzu with latest Rakudo

Hang on, it's only MVM_JIT_DISABLE sensitive on my branch (even-moar-jit). It appears to happen within the perl6 extension opcodes. Regards, Bart 2017-08-14 10:26 GMT+02:00 Bart Wiegmans <bartwiegm...@gmail.com>: > It's a bus error that goes away with MVM_JIT_DISABLE. > T

Re: [perl #131889] Segfault / malloc errors testing Uzu with latest Rakudo

It's a bus error that goes away with MVM_JIT_DISABLE. This is going to be interesting. 2017-08-14 10:23 GMT+02:00 Bart Wiegmans <bartwiegm...@gmail.com>: > Can reproduce, thank you! > > This also requires the following modules: > > JSON::Fast (broken test on my machine, but

Re: [perl #131889] Segfault / malloc errors testing Uzu with latest Rakudo

Can reproduce, thank you! This also requires the following modules: JSON::Fast (broken test on my machine, but may be for any reason) File::Temp HTTP::Server::Async HTTP::Server::Router Test::Output Template::Mustache YAMLish (I'm going to guess some of these were installed by rakudo star)

Re: [perl #131889] Segfault / malloc errors testing Uzu with latest Rakudo

Hang on, it's only MVM_JIT_DISABLE sensitive on my branch (even-moar-jit). It appears to happen within the perl6 extension opcodes. Regards, Bart 2017-08-14 10:26 GMT+02:00 Bart Wiegmans <bartwiegm...@gmail.com>: > It's a bus error that goes away with MVM_JIT_DISABLE. > T

Re: [perl #131889] Segfault / malloc errors testing Uzu with latest Rakudo

It's a bus error that goes away with MVM_JIT_DISABLE. This is going to be interesting. 2017-08-14 10:23 GMT+02:00 Bart Wiegmans <bartwiegm...@gmail.com>: > Can reproduce, thank you! > > This also requires the following modules: > > JSON::Fast (broken test on my machine, but

Re: [perl #131889] Segfault / malloc errors testing Uzu with latest Rakudo

Can reproduce, thank you! This also requires the following modules: JSON::Fast (broken test on my machine, but may be for any reason) File::Temp HTTP::Server::Async HTTP::Server::Router Test::Output Template::Mustache YAMLish (I'm going to guess some of these were installed by rakudo star)

Re: [perl #131888] No such method 'cache' for invocant of type 'NQPMu'

Hi, Thanks for the report. Can you try running it with the one of the following two variables set: MVM_SPESH_DISABLE=1 MVM_SPESH_INLINE_DISABLE=1 MVM_SPESH_OSR_DISABLE=1 MVM_JIT_DISABLE=1 Last weeks we've seen lots of development in the dynamic optimizer, so there are undoubtedly a few bugs to

Re: [perl #131888] No such method 'cache' for invocant of type 'NQPMu'

Hi, Thanks for the report. Can you try running it with the one of the following two variables set: MVM_SPESH_DISABLE=1 MVM_SPESH_INLINE_DISABLE=1 MVM_SPESH_OSR_DISABLE=1 MVM_JIT_DISABLE=1 Last weeks we've seen lots of development in the dynamic optimizer, so there are undoubtedly a few bugs to

Re: [perl #130540] [BUG] || && and or cannot be "overloaded"

I'm not sure if this adds anything, but: The reason why infix<||> doesn't work as you'd expect, is that operators like '||' and '&&' operate on expressions, that is to say code, rather than on values. The only valid way to extend such operators is to either use macros (which would receive both

Re: [perl #127330] [SLOW] 10_000 lines with ｢say ‘a’;｣ take 16 seconds to run

Yes, precompilation will help. Although we're not ready to make true or fake executables yet, you can compile to the moarvm bytecode file: perl6 --target=mbc --output=print.moarvm print.p6 But, this crashes during deserialization for me, so I'm not sure if that can be made to work as it stands.

Re: [perl #127374] git-protocol not passed to MoarVM/dynasm.git

I humbly suggest that rakudobrew, or perhaps rakudo Configure.pl, edit .gitmodules before running MoarVM Configure.pl if passed a different git protocol than https://. That is probably a better solution than trying to find a lowest-common-denominator protocol which is supported everywhere, since

Re: [perl #127144] Native uint doesn't overflow on ++

Just speculating, but I think this can be fixed by codegen; to be specific, the inc_i op should be followed by a truncate in case of a sized variable. 2016-01-04 11:37 GMT+01:00 Elizabeth Mattijsen : > > On 04 Jan 2016, at 00:25, Juerd Waalboer (via RT) < >

Re: [perl #123796] Bug: copy only copies first 2GB of file

The bug here was that moar expected libuv sendfile to send all bytes that are requested in one go, but it doesn't do that on linux (and perhaps other platforms). Instead, 2G are sent at once. libuv sendfile differs a bit from standard POSIX sendfile, but I think this patch fixes it. Bart

Re: [perl #122896] [BUG] 'is cached' routines break when precompiled

Which backend is that? I'm assuming moar. 2015-01-08 10:14 GMT+01:00 Elizabeth Mattijsen l...@dijkmat.nl: On 04 Oct 2014, at 02:41, (via RT) perl6-bugs-follo...@perl.org wrote: # New Ticket Created by # Please include the string: [perl #122896] # in the subject line of all future

Fwd: Eliminating the spesh codegen for jittable frames

features like deoptimization rely on it, and the JIT relies on those - we can't remove this codegen simply. Any thoughts are quite welcome. -- Forwarded message -- From: Bart Wiegmans bartwiegm...@gmail.com Date: 2015-01-07 0:38 GMT+01:00 Subject: Eliminating the spesh codegen