RE: [PATCH v4] http.postbuffer: allow full range of ssize_t values

[David Turner <David.Turner@xxxxxxxxxxxx>]

> -----Original Message-----
> From: Ramsay Jones [mailto:ramsay@xxxxxxxxxxxxxxxxxxxx]
> Sent: Monday, April 3, 2017 6:24 PM
> To: David Turner <David.Turner@xxxxxxxxxxxx>; git@xxxxxxxxxxxxxxx
> Subject: Re: [PATCH v4] http.postbuffer: allow full range of ssize_t values
> 
> 
> 
> On 03/04/17 18:30, David Turner wrote:
> > Unfortunately, in order to push some large repos, the http postbuffer
> > must sometimes exceed two gigabytes.  On a 64-bit system, this is OK:
> > we just malloc a larger buffer.
> >
> > This means that we need to use CURLOPT_POSTFIELDSIZE_LARGE to set
> the
> > buffer size.
> >
> > Signed-off-by: David Turner <dturner@xxxxxxxxxxxx>
> > ---
> >  cache.h       |  1 +
> >  config.c      | 17 +++++++++++++++++
> >  http.c        |  4 ++--
> >  http.h        |  2 +-
> >  remote-curl.c |  6 +++---
> >  5 files changed, 24 insertions(+), 6 deletions(-)
> >
> > diff --git a/cache.h b/cache.h
> > index fbdf7a815a..5e6747dbb4 100644
> > --- a/cache.h
> > +++ b/cache.h
> > @@ -1900,6 +1900,7 @@ extern int git_parse_maybe_bool(const char *);
> > extern int git_config_int(const char *, const char *);  extern int64_t
> > git_config_int64(const char *, const char *);  extern unsigned long
> > git_config_ulong(const char *, const char *);
> > +extern ssize_t git_config_ssize_t(const char *, const char *);
> >  extern int git_config_bool_or_int(const char *, const char *, int *);
> > extern int git_config_bool(const char *, const char *);  extern int
> > git_config_maybe_bool(const char *, const char *); diff --git
> > a/config.c b/config.c index 1a4d85537b..de5b155a4e 100644
> > --- a/config.c
> > +++ b/config.c
> > @@ -834,6 +834,15 @@ int git_parse_ulong(const char *value, unsigned
> long *ret)
> >  	return 1;
> >  }
> >
> > +static int git_parse_ssize_t(const char *value, ssize_t *ret) {
> > +	ssize_t tmp;
> > +	if (!git_parse_signed(value, &tmp,
> maximum_signed_value_of_type(ssize_t)))
> > +		return 0;
> > +	*ret = tmp;
> > +	return 1;
> > +}
> > +
> 
> The previous version of this patch causes gcc to complain on a Linux 32bit
> build, like so:
> 
>     CC config.o
> config.c: In function ‘git_parse_ssize_t’:
> config.c:840:31: warning: passing argument 2 of ‘git_parse_signed’ from
> incompatible pointer type [-Wincompatible-pointer-types]
>   if (!git_parse_signed(value, &tmp,
> maximum_signed_value_of_type(ssize_t)))
>                                ^
> config.c:753:12: note: expected ‘intmax_t * {aka long long int *}’ but
> argument is of type ‘ssize_t * {aka int *}’
>  static int git_parse_signed(const char *value, intmax_t *ret, intmax_t max)
>             ^
> 
> ... and I don't think this version would be any different (though I haven't
> tried).

So I guess that means tmp needs to be an intmax_t.