[Bug-gnulib] Re: CVS update [cvs1-11-x-branch]: /ccvs/src/

[Derek Robert Price]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Paul Eggert wrote:

>Derek Robert Price <address@hidden> writes:
>
>>The alternative would seem to be a fix at the OS level to prevent
>>stdout & stderr from being set nonblock,
>
>
>That sounds unlikely.  Among other things, I don't think POSIX
>allows it.
>
>I recall proposing a less-intrusive solution; see
><http://lists.gnu.org/archive/html/bug-cvs/2002-08/msg00032.html>.
>You demurred because it involves a separate process to copy stderr and
>you were worried about the inefficiencies involved.  Can you quantify
>these inefficiencies?  Perhaps they're not worth worrying about, in
>practice.  If so, the extra-process solution may be the way to go,
>despite its minor inefficiencies.


I think we recently settled in this thread that a wrapper around the
fwrite() function would prove as effective and is almost certainly
more efficient.  Why would the extra-process be preferrable to an
fwrite() wrapper that watched for incomplete writes and looped until
all the data was written?

>>Perhaps having a substitute for the stdio routines is still the best
>>preventative measure for this sort of problem.
>
>
>But this would require rewriting/auditing every part of CVS that can
>write to the affected descriptors, no?  A thankless and error/prone
>job.  No wonder nobody wants to do it.  It'd likely introduce bugs.


A function like:

size_t fwrite_block (const void *ptr, size_t size, size_t nmemb, FILE
*stream)
{
    size_t nwritten = 0;
    size_t nbytes = 0;

    while (nmemb > nwritten)
    {
        /* Convert to bytes since the system fwrite() is not
guaranteed to return
         * an object count we can equate to our file position for any
other SIZE.
         * Also, write a maximum of SIZE_MAX bytes each pass.  This is
also
         * a limitation of the POSIX fwrite().
         */
        size_t towrite = xtimes (nmemb - nwritten, size) - nbytes;
        size_t written;

        select (FILENO(stream)+1, ...);    /* Pseudocode select here */

        written = fwrite (ptr, 1, towrite, stream);
        ptr += written;
        nwritten += (written + nbytes) / size;
        nbytes = (written + nbytes) % size;
        if (ferror (stream)) break;
    }

    return nwritten;
}

and a:

#define fwrite fwrite_block

seems as effective and even less bug-prone than the extra-process.


I've already volunteered to write it.  I just wanted to get some
feedback on it, like if anyone else could spot problems I'd missed and
whether it sounded like a viable GNULIB module, before I did.

Cheers,

Derek

- --
                *8^)

Email: address@hidden

Get CVS support at <http://ximbiot.com>!
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFBNz7dLD1OTBfyMaQRAhTOAJ9WZ65nFNvIBlL2R7rq3j7/26he/QCcDdSW
3axDyAsi2BDcq2ANSDFhoko=
=ndyj
-----END PGP SIGNATURE-----