This is the mail archive of the libc-help@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: lazy library loading


> So the next question I have is: What problem are you trying to solve?
>
> One of the most expensive operation is symbol binding and that is
> already deferred by lazy symbol resolution.
>
> Why do you need lazy library loading?

I am trying to solve problem of start-up time of applications having a
lot of unused libraries during first window initialization.

> Have you experimented with this and do you have any relative
> performance numbers?

Yes, on several measured applications, lazy library loading can save
from 2% up to 22% of start-up time, depending on application.
Of course not exact numbers, calculated with some assumptions..


--
WBR,
Andrew


2013/2/18 Carlos O'Donell <carlos@systemhalted.org>:
> On Fri, Feb 15, 2013 at 4:24 AM, Andrew Senkevich
> <andrew.n.senkevich@gmail.com> wrote:
>> I am interesting in lazy library loading technology, and how to enable
>> it in libc/eglibc.
>
> Glad to hear that! We are always interested in new developers helping
> with the project.
>
>> May be some implementation plan already exist?
>
> I am not aware of any implementation of lazy library loading for GNU/Linux.
>
> So the next question I have is: What problem are you trying to solve?
>
> One of the most expensive operation is symbol binding and that is
> already deferred by lazy symbol resolution.
>
> Why do you need lazy library loading?
>
> Have you experimented with this and do you have any relative
> performance numbers?
>
> Cheers,
> Carlos.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]