Messages in this thread | | | From | Andi Kleen <> | Subject | Re: [Patch 2/4] Basic reorder infrastructure - makes linking very slow | Date | Fri, 10 Mar 2006 10:45:57 +0100 |
| |
On Monday 27 February 2006 18:19, Arjan van de Ven wrote: > On Mon, 2006-02-27 at 17:31 +0100, sam@ravnborg.org wrote: > > > This patch puts the infrastructure in place to allow for a reordering of > > > functions based inside the vmlinux. > > > > Can we make this general instead of x86_64 only? > > Then we can use Kconfig to enable it for the architectures where we want it. > > Actually Linus had pretty good arguments to make this per-architecture: > the list will be different on each architecture. > > (eg my first patch had it more generic; but Linus asked it to be per > arch, and I agree with the reasons he gave) > > Also I doubt it can be enabled "blindly" for all architectures; I expect > more to need hacks similar to the x86_64 entry.S fix before it can > work...
Hi,
I just discovered that this patch is the reason why my compiles slowed down so dramatically (thanks to Michael M. for the hint) The SUSE 10.0 ld goes from running in seconds to more than a minute.
I think I will drop this patch for now because I doubt the runtime improvement is worth the compile slowdown.
If there is some binutils release that handles this without dramatic slowdown we can test for it in the Makefile, but i don't want it to be enabled by default.
-Andi
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |