This is the mail archive of the mailing list for the binutils project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: New linker testsuite failures for PR 19162

On Tue, Mar 1, 2016 at 6:00 PM, Alan Modra <> wrote:
> On Tue, Mar 01, 2016 at 08:15:58AM -0800, H.J. Lu wrote:
>> The bug isn't fixed on hppa:
> The reason for the fail on hppa-linux is that Dave Anglin's fix for
> PR12376 makes the data segment always start on a page boundary.
> ie. the test should be xfailed for hppa-linux.

This is what I checked in:

commit 7bcc503f3ef52fcac0d9be31f1b82440ec7ed2ff
Author: H.J. Lu <>
Date:   Wed Mar 2 19:07:01 2016 -0800

    Skip ld-elf/pr19162.d for hppa-*-*

    ld-elf/pr19162.d fails for hppa-*-* since Dave Anglin's fix for PR 12376
    makes the data segment always start on a page boundary.

    * testsuite/ld-elf/pr19162.d: Skip hppa-*-*.

diff --git a/ld/ChangeLog b/ld/ChangeLog
index c839971..328e51d 100644
--- a/ld/ChangeLog
+++ b/ld/ChangeLog
@@ -1,5 +1,9 @@
 2016-03-02  H.J. Lu  <>

+ * testsuite/ld-elf/pr19162.d: Skip hppa-*-*.
+2016-03-02  H.J. Lu  <>
  * testsuite/ld-elf/shared.exp (mix_pic_and_non_pic): Add xfails.
  Xfail mix_pic_and_non_pic on "arm*-*-*" "aarch64*-*-*".

diff --git a/ld/testsuite/ld-elf/pr19162.d b/ld/testsuite/ld-elf/pr19162.d
index eb77c2c..c08bbed 100644
--- a/ld/testsuite/ld-elf/pr19162.d
+++ b/ld/testsuite/ld-elf/pr19162.d
@@ -3,8 +3,9 @@
 #ld: -shared -z max-page-size=0x200000
 #readelf: -l --wide
 #target: *-*-linux* *-*-gnu* *-*-nacl*
-#notarget: arc-*-*
+#notarget: arc-*-* hppa-*-*
 # arc target has an extra 64K stack section.
+# hppa fails due to PR 12376.

  +LOAD +0x0+[0-9a-f]{3} +0x[0-9a-f]+ +0x[0-9a-f]+ +0x0+[0-9a-f]{3}
+0x0+[0-9a-f]{3} +R[WE]+ +0x200000

> It fails on hppa64-linux because .dynamic is put in the text segment,
> making the text segment writable, and then there isn't a reason to
> make a second segment for .data.  A writable text segment is quite
> horrible.  That should really be fixed.
>> i370 backend is broken:
> Yes, I think we should remove i370 from the next binutils release.
> --
> Alan Modra
> Australia Development Lab, IBM


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]