Mailing-List: contact dev-help@corinthia.incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@corinthia.incubator.apache.org
MIME-Version: 1.0
In-Reply-To: <811A01B4-256D-4BF1-BCC7-55E3294DACAB@apache.org>
References: 
 <CALfVaARs0Y_fB=p_B97wNPRy8v1QvbMahMQF6voYJAUdej_n8g@mail.gmail.com>
	<CALfVaASwMWf=2qJDW-7SJLq+qqFL+RqGyJEspDZ7M4cqjBphBg@mail.gmail.com>
	<811A01B4-256D-4BF1-BCC7-55E3294DACAB@apache.org>
Date: Sun, 22 Feb 2015 13:47:07 +0100
Message-ID: 
 <CAK2iWdTAWmF84yQip445313of_5GX2z_KZ6bKLfsX=teo=2RyQ@mail.gmail.com>
Subject: Re: Checking malloc success and adding perror()
From: jan i <jani@apache.org>
To: "dev@corinthia.incubator.apache.org" <dev@corinthia.incubator.apache.org>
Content-Type: multipart/alternative; boundary=089e0160b864f3c968050facaee2

--089e0160b864f3c968050facaee2
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

On Sunday, February 22, 2015, Peter Kelly <pmkelly@apache.org> wrote:

> Sounds great - I would just suggest putting the x* function
> implementations in Wrapper.c rather than having a separate copy in the
> files for each supported platform. Wrapper.c is sort of our generic place
> to put things which work the same way on all platforms but still need to
> live in the platform module (as is the case for the zip functions which a=
re
> currently there).

+1 to use wrapper.c

rgds
jan i

>
> The memory allocation functions are identical for all platforms so I thin=
k
> it would be better to avoid the code duplication - especially since we=E2=
=80=99ll
> likely modify these for error handling/cleanup etc. and only want to do s=
o
> in one place.
>
> If the diffs are really big, you could perhaps post them as zip files to
> cut down on the size. Either that or post them somewhere we can get them;
> whatever is most convenient for you.
>
> =E2=80=94
> Dr Peter M. Kelly
> pmkelly@apache.org <javascript:;>
>
> PGP key: http://www.kellypmk.net/pgp-key <http://www.kellypmk.net/pgp-key=
>
> (fingerprint 5435 6718 59F0 DD1F BFA0 5E46 2523 BAA1 44AE 2966)
>
> > On 22 Feb 2015, at 7:13 pm, Gabriela Gibson <gabriela.gibson@gmail.com
> <javascript:;>> wrote:
> >
> > Hi,
> >
> > I propose to do the following:
> >
> > 1) Replace all the calls to *allocs (malloc, calloc, realloc)
> > {and also free()} with matching x* functions in DocFormats.
> >
> > 2) These new x*alloc functions will live in their respective
> > platform/src files, and will contain a call to the original
> > *alloc function and a simple exit handler.
> >
> > 3) I will create test cases in platform/tests/ for each of those
> > functions.  I can either put them in WrapperTests.c or create a
> > new, MallocTests.c file for them.
> >
> > 4) I will start out with free(), just to be certain I get the shape rig=
ht
> > before I process the rest of the to-do list.
> >
> > 5) I will ship this as a 'git diff' files.
> >
> > What do you think?
> >
> > G
> >
> > On Wed, Feb 18, 2015 at 9:19 PM, Gabriela Gibson <
> gabriela.gibson@gmail.com <javascript:;>>
> > wrote:
> >
> >> Hi,
> >>
> >> Reading through the source, I see that a lot of mallocs() do not have =
a
> >>
> >> [[
> >> if(foo =3D=3D NULL) {
> >>    perror("Foo: Out of memory.\n");
> >>    return _exit(EXIT_FAILURE);
> >> }
> >> ]]
> >>
> >> check.
> >>
> >> I can add those if you like.
> >>
> >> G
> >> --
> >> Visit my Coding Diary: http://gabriela-gibson.blogspot.com/
> >>
> >
> >
> >
> > --
> > Visit my Coding Diary: http://gabriela-gibson.blogspot.com/
>
>

--=20
Sent from My iPad, sorry for any misspellings.

--089e0160b864f3c968050facaee2--