Date: Tue, 12 Apr 2005 17:10:22 +0200 From: Divacky Roman <xdivac02@stud.fit.vutbr.cz> To: Vladislav Shabanov <vs@rambler-co.ru> Cc: hackers@freebsd.org Subject: Re: do-nothing code in sysv_shm.c Message-ID: <20050412151022.GB17895@stud.fit.vutbr.cz> In-Reply-To: <200504112003.25096.vs@rambler-co.ru> References: <200504112003.25096.vs@rambler-co.ru>
index | next in thread | previous in thread | raw e-mail
On Mon, Apr 11, 2005 at 08:03:24PM +0400, Vladislav Shabanov wrote:
> I`ve found the following code in sysv_shm.c :
>
> ............
> static void
> shminit()
> {
> int i;
>
> TUNABLE_INT_FETCH("kern.ipc.shmmaxpgs", &shminfo.shmall);
> for (i = PAGE_SIZE; i > 0; i--) {
> shminfo.shmmax = shminfo.shmall * PAGE_SIZE;
> if (shminfo.shmmax >= shminfo.shmall)
> break;
> }
> TUNABLE_INT_FETCH("kern.ipc.shmmin", &shminfo.shmmin);
> .........
>
> IMHO, loop does nothing. Possible variant:
alfred perlstein commited this saying:
Bring shm functions closer the the opengroup standards.
he is the person you should ask why is it coded this way. imho compiler just
optimizes this out leaving plain shminfo.shmmax = shminfo.shmall * PAGE_SIZE;
roman
help
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050412151022.GB17895>