Muli Ben-Yehuda <[EMAIL PROTECTED]> writes:
> I was thinking of kernel compilation, which requires -O2 (force
> inlining)
-O3 turns -finline-functions on. -O2 turns -fforce-mem on, maybe
that's what you meant.
> and thus would break horribly if you compile some of the files with
> O1.
Why? Why
"Nadav Har'El" <[EMAIL PROTECTED]> writes:
> But if you're going to compile some of the code with -O1 and -O3, your
> Makefile or configure script are going to look realy hairy :( It will
> look like black magic. A better (but more time consuming) thing to do is
> to try to find the offending pie
On Thu, Aug 08, 2002 at 09:01:01PM +0300, Shachar Shemesh wrote:
> >>My question is, can I compile in the same command line some files under
> >>-O1 and some under -O3?
> >
> >If they're going to be linked together, that's a very bad
> >idea. Otherwise, you could probably do it with some Makefil
On Thu, 8 Aug 2002, Michael Sternberg wrote:
> Now I want to able to recognize four situations:
> 1. Regular loopback is attached to serial port.
> 2."Special" loopback is attached to serial port.
> 3. Modem is attached to serial port (use AT -> OK ) in previous chat script
> 4. Nothing is attac
On Thu, Aug 08, 2002 at 06:31:11PM +0300, Orr Dunkelman wrote:
> I've stumbled with the following problem regarding gcc. Assume for a
> moment I compile many files, some are header files, some are interface
> files, and some are the "real" program.
>
> When compiling with -O1 optimization things
On Thu, 8 Aug 2002, Orr Dunkelman wrote:
> The code is closed under some legal agreements, so I cannot release it,
> neither send gcc a bug-report.
The possibility that gcc is somehow sensitive to the legal status of
the code in question does not have to be ruled out.
How this might happen
On Thu, Aug 08, 2002, Muli Ben-Yehuda wrote about "Re: gcc question":
> On Thu, Aug 08, 2002 at 06:31:11PM +0300, Orr Dunkelman wrote:
>...
> > My question is, can I compile in the same command line some files under
> > -O1 and some under -O3?
>
> If they're going to be linked together, that's a
Quoth Orr Dunkelman:
> The code is closed under some legal agreements, so I cannot release it,
> neither send gcc a bug-report.
For baseline comparison sake, could you give a compilation round under
icc (Intel's i386 copiler)... It (the icc) is somewhat more pedantic
than gcc and might flush som
On Thu, 8 Aug 2002, Shachar Shemesh wrote:
>
> A. Doing that requires reflushing the modem, making it think it is
> "Speed Touch Pro". Aside from certain claims it is illegal (can anyone
> explain why? It is, after all, a modem I BOUGHT from Bezeq, and it is
> mine to tweak with as I see fit, IANA
