Обсуждение: [HACKERS] Someone might already have done working versions

Поиск
Список
Период
Сортировка

[HACKERS] Someone might already have done working versions

От
Tymm Twillman
Дата:
  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.
  Send mail to mime@docserver.cac.washington.edu for more info.

- ---2133963750-1274176059-865249676=:24416
Content-Type: TEXT/PLAIN; charset=US-ASCII

of these, but here's a diff for working (broken) hton_s, hton_l, etc. for
big endian machines...

    -Tymm



- ---2133963750-1274176059-865249676=:24416
Content-Type: TEXT/PLAIN; charset=US-ASCII; name=diffs
Content-Transfer-Encoding: BASE64
Content-ID: <Pine.SGI.3.96.970602060756.24416H@lucid.coe.missouri.edu>
Content-Description:
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- ---2133963750-1274176059-865249676=:24416--

------------------------------

Re: [HACKERS] Someone might already have done working versions

От
David Friend
Дата:
On Mon, 2 Jun 1997, Tymm Twillman wrote:

> of these, but here's a diff for working (broken) hton_s, hton_l, etc. for
> big endian machines...

This had been a problem on the SPARC/Linux platform but I believe it was
fixed and applied to the source around May 25th.  At any rate, 970531
works on my system without patching.

David Friend                ! cq995@freenet.carleton.ca
Atlantis Scientific Inc.        ! david.friend@atlsci.com
20 Colonnade Rd, Suite 110        ! 613-727-1087 (voice)
Ottawa, Ontario, CANADA  K2E 7M6    ! 800-265-3894 (voice)
ERGOvista Scientific Image Analysis    ! 613-727-5853 (fax)

------------------------------

Re: [HACKERS] Someone might already have done working versions

От
Tymm Twillman
Дата:
On Mon, 2 Jun 1997, David Friend wrote:

> On Mon, 2 Jun 1997, Tymm Twillman wrote:
>
> > of these, but here's a diff for working (broken) hton_s, hton_l, etc. for
> > big endian machines...
>
> This had been a problem on the SPARC/Linux platform but I believe it was
> fixed and applied to the source around May 25th.  At any rate, 970531
> works on my system without patching.
>
> David Friend                ! cq995@freenet.carleton.ca
> Atlantis Scientific Inc.        ! david.friend@atlsci.com
> 20 Colonnade Rd, Suite 110        ! 613-727-1087 (voice)
> Ottawa, Ontario, CANADA  K2E 7M6    ! 800-265-3894 (voice)
> ERGOvista Scientific Image Analysis    ! 613-727-5853 (fax)
>

Admittedly, the current code works... but the macros don't *do* anything,
except eat u_shorts that are passed to hton_s, and give nasty numbers for
u_shorts passed to ntoh_l.  With psql on one system and the database on
another, with different byte ordering, it should break.

- -Tymm

------------------------------

Re: [HACKERS] Someone might already have done working versions

От
David Friend
Дата:
On Mon, 2 Jun 1997, Tymm Twillman wrote:

> On Mon, 2 Jun 1997, David Friend wrote:
>
> > On Mon, 2 Jun 1997, Tymm Twillman wrote:
> >
> > > of these, but here's a diff for working (broken) hton_s, hton_l, etc. for
> > > big endian machines...
> >
> > This had been a problem on the SPARC/Linux platform but I believe it was
> > fixed and applied to the source around May 25th.  At any rate, 970531
> > works on my system without patching.
>
> Admittedly, the current code works... but the macros don't *do* anything,
> except eat u_shorts that are passed to hton_s, and give nasty numbers for
> u_shorts passed to ntoh_l.  With psql on one system and the database on
> another, with different byte ordering, it should break.

I'll take your word for it since I don't have a good understanding for
this piece of code.  Anything to help clean up the mess in this area is
appreciated.

David Friend                ! cq995@freenet.carleton.ca
Atlantis Scientific Inc.        ! david.friend@atlsci.com
20 Colonnade Rd, Suite 110        ! 613-727-1087 (voice)
Ottawa, Ontario, CANADA  K2E 7M6    ! 800-265-3894 (voice)
ERGOvista Scientific Image Analysis    ! 613-727-5853 (fax)

------------------------------

Re: [HACKERS] Someone might already have done working versions

От
The Hermit Hacker
Дата:
On Mon, 2 Jun 1997, Tymm Twillman wrote:

> On Mon, 2 Jun 1997, David Friend wrote:
>
> > On Mon, 2 Jun 1997, Tymm Twillman wrote:
> >
> > > of these, but here's a diff for working (broken) hton_s, hton_l, etc. for
> > > big endian machines...
> >
> > This had been a problem on the SPARC/Linux platform but I believe it was
> > fixed and applied to the source around May 25th.  At any rate, 970531
> > works on my system without patching.
> >
> > David Friend                ! cq995@freenet.carleton.ca
> > Atlantis Scientific Inc.        ! david.friend@atlsci.com
> > 20 Colonnade Rd, Suite 110        ! 613-727-1087 (voice)
> > Ottawa, Ontario, CANADA  K2E 7M6    ! 800-265-3894 (voice)
> > ERGOvista Scientific Image Analysis    ! 613-727-5853 (fax)
> >
>
> Admittedly, the current code works... but the macros don't *do* anything,
> except eat u_shorts that are passed to hton_s, and give nasty numbers for
> u_shorts passed to ntoh_l.  With psql on one system and the database on
> another, with different byte ordering, it should break.

    *should* doesn't equal *does*...can someone out there indicate whether
this patch is required for a v6.1 release?  There have been several hints
that this is all going to be rewritten *after* v6.1 is released, which will
make any patches to the backend communications pretty much useless, and could
possible add some bugs/instabilities that we don't need this close to a
release...

    ...comments?

Marc G. Fournier
Systems Administrator @ hub.org
primary: scrappy@hub.org           secondary: scrappy@{freebsd|postgresql}.org

------------------------------