You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by Brian Behlendorf <br...@hyperreal.org> on 1998/05/04 10:17:31 UTC
Re: LGPL code in Apache?
At 05:42 AM 4/21/98 -0400, Jim Jagielski wrote:
>Chuck Murcko wrote:
>>
>> Panos had gotten permissions for all the stuff that was in xinetd before I
>> started the other ports (FreeBSD, Linux, Solaris [by Jan Wedekind],and
IRIX. I
>> and later Jan are both working under the structure of the copyright that
Panos
>> originally used. Snprintf() and cvt() were there already when I started
hacking
>> the first BSD port. Nothing's been added from GNU since then by Jan or I.
>
>The snprintf() stuff (actually the strx/Sio stuff) is safe to use...
>the *cvt() stuff is unknown, at least to me
Okay, Panos wrote me back saying:
>The comment that you are referring to is *NOT* part of the source of
>the SIO library, because the associated code is *NOT* in the SIO
>library and never was. Specifically, the SIO code needs the routines
>ecvt(), fcvt(), and gcvt(), which it presumes are in the system's C
>library.
So it's not part of SIO. Chuck, Jim, who would have been the one to
integrate it into xinetd? Looks like the lineage is now glibc -> xinetd ->
apache rather than glibc -> SIO -> xinetd -> apache.
Brian
--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
pure chewing satisfaction brian@apache.org
brian@hyperreal.org
Re: LGPL code in Apache?
Posted by Chuck Murcko <ch...@topsail.org>.
It's in /usr/lib/libc_p.a, on 2.2.5, at least. Reentrant alone isn't thread
safe, unfortunately.
Brian Behlendorf wrote:
> At 12:14 PM 5/4/98 -0700, Dean Gaudet wrote:
> >On Mon, 4 May 1998, Chuck Murcko wrote:
> >
> >> Actually, I can't see why cvt() should be in SIO at all. every libc I
> >> can find contains it. Perhaps there was some mutant BSDI or something.
>
> Where is it in FreeBSD? I can't find it...
>
> >The libc versions aren't threadsafe.
>
> Even in libc_r's? I thought the "r" meant reentrant.
>
> Brian
>
> --=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
> pure chewing satisfaction brian@apache.org
> brian@hyperreal.org
Re: LGPL code in Apache?
Posted by Dean Gaudet <dg...@arctic.org>.
On Mon, 4 May 1998, Brian Behlendorf wrote:
> At 12:14 PM 5/4/98 -0700, Dean Gaudet wrote:
> >On Mon, 4 May 1998, Chuck Murcko wrote:
> >
> >> Actually, I can't see why cvt() should be in SIO at all. every libc I
> >> can find contains it. Perhaps there was some mutant BSDI or something.
>
> Where is it in FreeBSD? I can't find it...
>
> >The libc versions aren't threadsafe.
>
> Even in libc_r's? I thought the "r" meant reentrant.
Perhaps. They can be made threadsafe at the expense of a bit of speed,
and by wasting memory in all threads for static buffers. It depends on
what your libc_r does. It'll certainly be different elsewhere, and
certainly different on win32.
Dean
Re: LGPL code in Apache?
Posted by Brian Behlendorf <br...@hyperreal.org>.
At 12:14 PM 5/4/98 -0700, Dean Gaudet wrote:
>On Mon, 4 May 1998, Chuck Murcko wrote:
>
>> Actually, I can't see why cvt() should be in SIO at all. every libc I
>> can find contains it. Perhaps there was some mutant BSDI or something.
Where is it in FreeBSD? I can't find it...
>The libc versions aren't threadsafe.
Even in libc_r's? I thought the "r" meant reentrant.
Brian
--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
pure chewing satisfaction brian@apache.org
brian@hyperreal.org
Re: LGPL code in Apache?
Posted by Dean Gaudet <dg...@arctic.org>.
On Mon, 4 May 1998, Chuck Murcko wrote:
> Actually, I can't see why cvt() should be in SIO at all. every libc I
> can find contains it. Perhaps there was some mutant BSDI or something.
The libc versions aren't threadsafe.
Dean
Re: LGPL code in Apache?
Posted by Chuck Murcko <ch...@topsail.org>.
Actually, I can't see why cvt() should be in SIO at all. every libc I
can find contains it. Perhaps there was some mutant BSDI or something.
Brian Behlendorf wrote:
>
> At 05:42 AM 4/21/98 -0400, Jim Jagielski wrote:
> >Chuck Murcko wrote:
> >>
> >> Panos had gotten permissions for all the stuff that was in xinetd before I
> >> started the other ports (FreeBSD, Linux, Solaris [by Jan Wedekind],and
> IRIX. I
> >> and later Jan are both working under the structure of the copyright that
> Panos
> >> originally used. Snprintf() and cvt() were there already when I started
> hacking
> >> the first BSD port. Nothing's been added from GNU since then by Jan or I.
> >
> >The snprintf() stuff (actually the strx/Sio stuff) is safe to use...
> >the *cvt() stuff is unknown, at least to me
>
> Okay, Panos wrote me back saying:
>
> >The comment that you are referring to is *NOT* part of the source of
> >the SIO library, because the associated code is *NOT* in the SIO
> >library and never was. Specifically, the SIO code needs the routines
> >ecvt(), fcvt(), and gcvt(), which it presumes are in the system's C
> >library.
>
> So it's not part of SIO. Chuck, Jim, who would have been the one to
> integrate it into xinetd? Looks like the lineage is now glibc -> xinetd ->
> apache rather than glibc -> SIO -> xinetd -> apache.
>
> Brian
>
> --=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
> pure chewing satisfaction brian@apache.org
> brian@hyperreal.org
--
chuck
Chuck Murcko
The Topsail Group, West Chester PA USA
chuck@topsail.org
Re: LGPL code in Apache?
Posted by Chuck Murcko <ch...@topsail.org>.
I checked the first copy of xinetd that I started working on, and cvt() was
already there. I'm assuming it got added between Panos' last touch of the code
and my first. It looks like I'll also have to write a replacement, then.
Brian Behlendorf wrote:
> At 05:42 AM 4/21/98 -0400, Jim Jagielski wrote:
> >Chuck Murcko wrote:
> >>
> >> Panos had gotten permissions for all the stuff that was in xinetd before I
> >> started the other ports (FreeBSD, Linux, Solaris [by Jan Wedekind],and
> IRIX. I
> >> and later Jan are both working under the structure of the copyright that
> Panos
> >> originally used. Snprintf() and cvt() were there already when I started
> hacking
> >> the first BSD port. Nothing's been added from GNU since then by Jan or I.
> >
> >The snprintf() stuff (actually the strx/Sio stuff) is safe to use...
> >the *cvt() stuff is unknown, at least to me
>
> Okay, Panos wrote me back saying:
>
> >The comment that you are referring to is *NOT* part of the source of
> >the SIO library, because the associated code is *NOT* in the SIO
> >library and never was. Specifically, the SIO code needs the routines
> >ecvt(), fcvt(), and gcvt(), which it presumes are in the system's C
> >library.
>
> So it's not part of SIO. Chuck, Jim, who would have been the one to
> integrate it into xinetd? Looks like the lineage is now glibc -> xinetd ->
> apache rather than glibc -> SIO -> xinetd -> apache.
>
> Brian
>
> --=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
> pure chewing satisfaction brian@apache.org
> brian@hyperreal.org