Front page | perl.perl5.porters |
Postings from March 2013
(I have been busy with other duties, and is now returning to this.) > Not that that should refrain us from fixing the PerlIO mutex. Fine. So is my 0001 patch attached to the ticket [1] good enough? If yes, can you apply it? If no, what is the next step to get this fixed? /Patrik Hägglund 1. https://rt.perl.org/rt3/Ticket/Attachment/1190997/613241/0001-Add-PL_perlio_mutex-to-atfork_lock-unlock-to-fix-bug.patch -----Original Message----- From: Leon Timmermans via RT [mailto:perlbug-followup@perl.org] Sent: den 14 februari 2013 13:29 To: Patrik Hägglund H Subject: Re: [perl #106212] close and dup hang in threads On Wed, Feb 13, 2013 at 8:46 PM, Dr.Ruud <rvtol+usenet@isolution.nl> wrote: > So we should all do them in (for example) alphabetical order, > or have an (overridable) multilock( LIST ) that does it for us? To make matters more complex, there is also: * PL_hints_mutex /* Mutex for refcounted he refcounting */ * PL_check_mutex /* Mutex for PL_check */ * PL_my_ctx_mutex * PL_dollarzero_mutex /* Modifying $0 */ We don't have three but seven mutexes to take into account. Doing this properly may be a bit of a puzzle. Not that that should refrain us from fixing the PerlIO mutex. Leon