Re: specs file substitue chapter 5.7 lfs_udev branch

2006-03-09 Thread Dan Nicholson
On 3/9/06, Dragos Ionita <[EMAIL PROTECTED]> wrote: > > With regards to Chris' comment: > > > Except LFS isn't expected to be used on other platforms. That's what > > CLFS is for. > > I know. I decided to not use CLFS as I am compiling from a PPC Linux > system, so I didn't see the need to use the

RE: specs file substitue chapter 5.7 lfs_udev branch

2006-03-09 Thread Dragos Ionita
Dan Nicholson wrote: > Dragos, when you get to the Ch. 6 Readjustment, you will have to change > a couple of the commands because they specifically referenct ld-linux.so.2, > which you obviously don't have. Probably have to drop a ^ there, too. The chapter 6 readjustment already leaves the "^" o

Re: specs file substitue chapter 5.7 lfs_udev branch

2006-03-09 Thread Chris Staub
Dan Nicholson wrote: Very interesting. So, on gcc-4 with PPC, you don't have a separate spec called *dynamic_linker ? Because on x86, it looks like this: *dynamic_linker: /tools/lib/ld-linux.so.2 (Obviously, I already did the adjustment.) I suppose LFS could leave the ^ out. Just seemed mos

Re: specs file substitue chapter 5.7 lfs_udev branch

2006-03-09 Thread Dan Nicholson
On 3/9/06, Dragos Ionita <[EMAIL PROTECTED]> wrote: > Archaic wrote: > > Sorry about that, I didn't have a look at 6.1. > The specsfile code snippet is the following: > > "-m elf32ppclinux %{!shared: %{!static: %{rdynamic:-export-dynamic} > %{!dynamic-linker:-dynamic-linker /tools/lib/ld.so.1}}

RE: RE: specs file substitue chapter 5.7 lfs_udev branch

2006-03-09 Thread Dragos Ionita
Archaic wrote: > Notice the difference between 6.1 and current trunk. Gcc-4 does > have this at the beginning and the sed works. I would recommend > pasting the relevant line from your ppc specs file so we can have a > look. Sorry about that, I didn't have a look at 6.1. The specsfile code snippe

Re: specs file substitue chapter 5.7 lfs_udev branch

2006-03-09 Thread Archaic
On Thu, Mar 09, 2006 at 09:51:15AM +0100, Dragos Ionita wrote: > > I've found that the reason for this is, that the sed substitute looks > for "^/lib", however, in the specsfile created by dumpspecs, "/lib" > isn't at the beginning of a line. Notice the difference between 6.1 and current trunk. G