Author neologix
Recipients Aaron.Sherman, giampaolo.rodola, gregory.p.smith, neologix, pitrou, vstinner
Date 2011-03-02.13:01:27
SpamBayes Score 5.57253e-06
Marked as misclassified No
Message-id <AANLkTi=qSziZzpJSjvtoJMx+S0-y1xOUJbCOvWN1H6Lk@mail.gmail.com>
In-reply-to <1299067862.29.0.175389632467.issue11314@psf.upfronthosting.co.za>
Content
> pitrou> I think your analysis is wrong. These mmap() calls are for
> pitrou> anonymous memory, most likely they are emitted by the libc's
> pitrou> malloc() to get some memory from the kernel. In other words
> pitrou> they will be blazingly fast.
>
> Are you sure? :-)

Well, it is fast. It's true that mmap is slower than brk, since the
kernel zero-fills the pages, but the overhead probably doesn't come
from this, but more likely from pymalloc or malloc, and also from the
call to _PyBytes_resize in posix_read when the number of bytes read is
smaller than what has been requested.
History
Date User Action Args
2011-03-02 13:01:28neologixsetrecipients: + neologix, gregory.p.smith, pitrou, vstinner, giampaolo.rodola, Aaron.Sherman
2011-03-02 13:01:27neologixlinkissue11314 messages
2011-03-02 13:01:27neologixcreate