This is the mail archive of the 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: [PATCH] fallocate: pass off_t in register pair correctly for 64-bit off_t

On Fri, Jun 24, 2016 at 04:30:23AM -0400, Mike Frysinger wrote:
> On 24 Jun 2016 08:43, Yury Norov wrote:
> > sysdeps/unix/sysv/linux/fallocate.c contains wide-spreaded hack for creating
> > register pair from 32-bit signed value:
> > 	LONG_LONG_PAIR (val >> 31, val)
> > 
> > It works well for off_t if it is 32-bit lenght. Modern ABIs requres 64-bit off_t,
> > so this hack doesn't work for them. In this patch, fallocate handler is taken from
> > fallocate64.c, depending on __OFF_T_MATCHES_OFF64_T.
> what ABIs exactly are you referring to ?

AARCH64/ILP32. It would be any modern 32-bit API, as off_t should be
64-bit for them

> why is this change specific to fallocate ?  seems like this argument
> could be made about any of the funcs that operate on off_t's.  what
> about posix_fallocate ?  posix_fadvise ?

Because it's RFC. If this one is OK, I'll prepare correct patch for
all affected functions. I also wrote this small patch to ask what way
is more preferable - with macro, or with redirection.

> > The other option to fix it is to introduce macro that creates the pair correctly
> > for both 32- and 64-bit off_t, like this:
> if off_t & off64_t are the same size, we don't want duplicate symbols
> that do the same thing.
> -mike

So it 1st version is OK, I'll prepare full patch with all needed


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