Home Original page

FIXED: Re: New test ld-elf/header fails for cris-elf (likely elsewhere)

Hans-Peter Nilsson hans-peter.nilsson@axis.com
Fri Jan 12 01:58:00 GMT 2007 
> Date: Thu, 11 Jan 2007 18:51:09 +0100
> From: Hans-Peter Nilsson <hp@axis.com>

> > Hans-Peter Nilsson wrote:
> > > regexp "^    LOAD off    0x0*0000000 vaddr 0x0*0010000 paddr 0x0*0010000 align 2..16$"
> 
> So "align .*" ?

Forget that; I didn't notice that this should've been taken care
of by max-page-size, which was ignored by ld-new; with error
message, but with output still generated.

> Ok, I was mostly interested in whether the "2**16" check was
> important.  I suggest adding "#target: *-*-linux*" plus the changes
> above.

For cris-axis-linux-gnu, there's also a need to match an
additional line where private header flags are output.  With the
purpose of the test now sufficiently clarified and obvious, I
committed the following fix after checking on native
i686-pc-linux-gnu and cross to e.g. (targets that failed below
and) cris-axis-linux-gnu, cris-axis-elf.

FWIW, some other "failed targets" were arm-unknown-elf,
d10v-unknown-elf, frv-unknown-elf, h8300-unknown-elf.
iq2000-unknown-elf, m32c-unknown-elf, m32r-unknown-elf,
mips-unknown-elf, v850-unknown-elf.

ld/testsuite:

	* ld-elf/header.d: Allow arbitrary lines between "Program Header"
	and "Sections".  Only run on *-*-linux*.

Index: ld-elf/header.d
===================================================================
RCS file: /cvs/src/src/ld/testsuite/ld-elf/header.d,v
retrieving revision 1.1
diff -p -u -r1.1 header.d
--- ld-elf/header.d	11 Jan 2007 12:23:53 -0000	1.1
+++ ld-elf/header.d	12 Jan 2007 01:52:16 -0000
@@ -1,3 +1,4 @@
+# target: *-*-linux*
 # ld: -T header.t -z max-page-size=0x10000
 # objdump: -hpw
 
@@ -5,7 +6,7 @@
 Program Header:
     LOAD off    0x0*0000000 vaddr 0x0*0010000 paddr 0x0*0010000 align 2..16
          filesz 0x0*001002[48] memsz 0x0*001002[48] flags rwx
-
+#...
 Sections:
 Idx Name          Size      VMA       *LMA       *File off  Algn  Flags
   0 .text         0*000ffac  0*001007[48]  0*001007[48]  0*000007[48]  2...  CONTENTS, ALLOC, LOAD, READONLY, CODE


brgds, H-P
More information about the Binutils mailing list