RE: 2.1.123 (Possibly slower)

David B. Rees (dbr@oto.dyn.ml.org)
Mon, 28 Sep 1998 16:26:00 -0700 (PDT)


Are you building the same kernel, or building different kernels. You can't
compare the building times of two different kernels while running two
different kernels. What you want to do is compare the building of the same
kernel (doesn't matter which, just pick one) while running different kernels.
Check this and see if 2.1.123 is still "slower".

-Dave

On 28-Sep-98 Richard B. Johnson wrote:
> It looks as though version 2.1.123 is somewhat slower than usual.
>
> Build using a linux-2.1.108 kernel
>
> make[2]: Leaving directory `/usr/src/linux-2.1.123/arch/i386/boot/compressed'
> gcc -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -o tools/build
> tools/build.c -I/usr/src/linux-2.1.123/include
> objcopy -O binary -R .note -R .comment -S compressed/bvmlinux
> compressed/bvmlinux.out
> tools/build -b bbootsect bsetup compressed/bvmlinux.out CURRENT > bzImage
> Root device is (8, 1)
> Boot sector 512 bytes.
> Setup is 1272 bytes.
> System is 611 kB
> make[1]: Leaving directory `/usr/src/linux-2.1.123/arch/i386/boot'
> Time = 386 second(s).
> ^^^________ Time to build.
> Build using a Linux-2.1.123 kernel.
>
> make[2]: Leaving directory `/usr/src/linux-2.1.123/arch/i386/boot/compressed'
> gcc -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -o tools/build
> tools/build.c -I/usr/src/linux-2.1.123/include
> objcopy -O binary -R .note -R .comment -S compressed/bvmlinux
> compressed/bvmlinux.out
> tools/build -b bbootsect bsetup compressed/bvmlinux.out CURRENT > bzImage
> Root device is (8, 1)
> Boot sector 512 bytes.
> Setup is 1272 bytes.
> System is 611 kB
> make[1]: Leaving directory `/usr/src/linux-2.1.123/arch/i386/boot'
> Time = 400 second(s).
> ^^^_____________ Time to build.
>
> Therefore, I decided to run a bunch of builds in a loop and accumulate
> some times. The kernel was already configured then `make clean` was
> executed. This was not timed.
>
> Then `timer make bzImage &>World.Log &` was executed 8 times.
>
> `timer` is a simple program that gets the time() variable, saves it,
> then executes the rest of the command line (using system()), then
> gets the time again. It subtracts the two times and writes them
> before exit().
>
> More tests appended:
>
> Linux-2.1.108 build times. SMP,
>
> processor : 0
> vendor_id : GenuineIntel
> cpu family : 6
> model : 3
> model name : Pentium II (Klamath)
> stepping : 3
> cache size : 512 KB
> fdiv_bug : no
> hlt_bug : no
> sep_bug : no
> f00f_bug : no
> fpu : yes
> fpu_exception : yes
> cpuid level : 2
> wp : yes
> flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov
mmx
> bogomips : 266.24
>
> processor : 1
> vendor_id : GenuineIntel
> cpu family : 6
> model : 3
> model name : Pentium II (Klamath)
> stepping : 3
> cache size : 512 KB
> fdiv_bug : no
> hlt_bug : no
> sep_bug : no
> f00f_bug : no
> fpu : yes
> fpu_exception : yes
> cpuid level : 2
> wp : yes
> flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov
mmx
> bogomips : 266.24
>
> Build(1) 322 seconds
> Build(2) 355 seconds
> Build(3) 362 seconds
> Build(4) 328 seconds
> Build(5) 311 seconds
> Build(6) 306 seconds
> Build(7) 329 seconds
> Build(8) 341 seconds
> Mean 332 seconds
>
> Linux-2.1.123 build times. SMP,
>
> Build(1) 416 seconds
> Build(2) 408 seconds
> Build(3) 398 seconds
> Build(4) 328 seconds
> Build(5) 422 seconds
> Build(6) 432 seconds
> Build(7) 381 seconds
> Build(8) 396 seconds
> Mean 398 seconds
>
> So it looks as though Linux-2.1.123 is quite a bit slower.
>
>
> Cheers,
> Dick Johnson
> ***** FILE SYSTEM WAS MODIFIED *****
> Penguin : Linux version 2.1.123 on an i586 machine (66.15 BogoMips).
> Warning : It's hard to remain at the trailing edge of technology.
>
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.rutgers.edu
> Please read the FAQ at http://www.tux.org/lkml/

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/