Re: Slow _bfd_strip_section_from_output

[Hans-Peter Nilsson <hp at bitrange dot com>]

On Tue, 3 May 2005, H. J. Lu wrote:
> On Tue, May 03, 2005 at 03:59:05PM -0400, Hans-Peter Nilsson wrote:
> > On Mon, 2 May 2005, H. J. Lu wrote:
> > > This is what I come up with. I added link_order_input so that I
> > > don't have to seatch the whole list to find a bfd_input_link_order.
> > > Otherwise, bfd_new_link_order will be as slow as the old
> > > _bfd_strip_section_from_output.
> > >
> > > Everything seems OK on ia32, ia64 and x86_64. But I did get some
> > > failures on "cris-elf". I don't know if they are harmless.
> > Probably not, but there's no way to tell without details.
> > FWIW, my autotester shows clean results here.
> With my patch applied?

No, I only test what's checked in. If you get failures with your
patch, then your patch is likely flawed (but without details, no
knowing), if it's not meant to actually change the output.
(And if it is, you'd know and wouldn't make the "don't know"
comment.)

brgds, H-P
PS. BTW, thanks a bunch for taking on the sec64k slowness!