nonoptimized kernel compilation (an idea)

Phil Rosenthal (lkernel@villaweb.net)
Thu, 25 Jun 1998 18:37:27 -0400 (EDT)

I noticed while compiling my kernel, that it uses the option "-m486" (and
I have a pentium200)
I have pgcc, and -mpentium makes it optimized for pentium, and m486 makes
one optimized for a 486, since I dont have a 486, I wanted to change
that...
after looking into it, I found that arch/i386/Makefile, had this code:

ifdef CONFIG_M586
CFLAGS := $(CFLAGS) -m486 -malign-loops=2 -malign-jumps=2 -malign-functions=2 -DCPU=586
endif

ifdef CONFIG_M686
CFLAGS := $(CFLAGS) -m486 -malign-loops=2 -malign-jumps=2 -malign-functions=2 -DCPU=686
endif

I changed it to this (it works with pgcc, probably egcs and gcc 2.8, but
not with gcc 2.7)

ifdef CONFIG_M586
CFLAGS := $(CFLAGS) -mpentium -malign-loops=2 -malign-jumps=2 -malign-functions=2 -DCPU=586
endif

ifdef CONFIG_M686
CFLAGS := $(CFLAGS) -mpentiumpro -malign-loops=2 -malign-jumps=2 -malign-functions=2 -DCPU=686
endif

now, since it doesnt work with gcc 2.7, it shouldnt be implemented like
this, but maybe someone knows how to make it so that if it supports
-mpentium/-mpentiumpro, it uses it, otherwise it will revert back to -m486

Any ideas?
Anyone know how to code it properly?

I didn't benchmark a speed difference (too lazy)
but it does "feel" more responsive...

--winter

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu