Hide Forgot
Description of problem: RHEL-7.5 binutils has been rebased: RHBA-2017:30118-01 binutils bug fix and enhancement update https://errata.devel.redhat.com/advisory/30118 binutils-2.25.1-31.base.el7.ppc64 -> binutils-2.27-21.base.el7.ppc64 The rebase includes upstream commit: 647e4d46495f2bfb0950fd1066c8a660173cca40 is the first bad commit commit 647e4d46495f2bfb0950fd1066c8a660173cca40 Author: H.J. Lu <hjl.tools> Date: Wed Jun 22 05:37:24 2016 -0700 ld: Add a linker configure option --enable-relro Which made relro the default on various targets incl. ppc64. binutils.spec now even contains explicit override for all arches: +# Default to read-only-relocations (relro) in shared binaries. +%define default_relro 1 I haven't seen any request for relro in the Bugs listed for the binutils errata, was there? Other binary tools were not ready for relro when it became default upstream in 2016-06 as RHEL-7.0 is from 2014. Still even trunk GDB cannot handle ppc64 relro properly as I have filed now: ppc64: gdbarch_skip_trampoline_code fails for relro https://sourceware.org/bugzilla/show_bug.cgi?id=22576 I believe the default relro setting should remain the same as in RHEL-7.4. Version-Release number of selected component (if applicable): PASS: binutils-2.25.1-31.base.el7.ppc64 FAIL: binutils-2.27-21.base.el7.ppc64 How reproducible: Always. Steps to Reproduce: 1. GDB: runtest gdb.reverse/solib-precsave.exp Actual results: FAIL: gdb.reverse/solib-precsave.exp: reverse-step back to main one FAIL: gdb.reverse/solib-precsave.exp: reverse-step into solib function two FAIL: gdb.reverse/solib-precsave.exp: reverse-step within solib function two FAIL: gdb.reverse/solib-precsave.exp: reverse-step back to main two FAIL: gdb.reverse/solib-precsave.exp: run until end part two FAIL: gdb.reverse/solib-precsave.exp: reverse-next over solib function one Expected results: Testfile reports PASS. Additional info:
Hi Jan, This was a snafu. When I reviewed the patches for the RHEL binutils I noticed that we had a patch to enable relro by default. Since there is now a configure option that does the same thing, there was no need for the patch. So I removed it and added the code to the binutils.spec file. Unfortunately I totally forgot about the PowerPC64's problems. Especially since relro is enabled by default for PowerPC64 in the FSF binutils sources. Anyway I have checked in a patch to not enable relro by default for the PowerPC64, so please try out: binutils-2.27-22.base.el7 Cheers Nick
Thanks, confirming binutils-2.27-22.base.el7.ppc64 does fix the GDB Red HatBug 1522798 regression. I did not do more regression testing, up to QA.
After some internal discussion it has been decided that this BZ is wrong. RELRO support for PowerPC is enabled in RHEL 7.4 and it should continue to be enabled in RHEL 7.5. (The GDB problem was a separate bug that has now been fixed). So I am reverting the patch previously applied for this BZ. The binutils-2.27-27.base.el7 rpm now contains this change.
binutils-2.29-13.fc27 has been submitted as an update to Fedora 27. https://bodhi.fedoraproject.org/updates/FEDORA-2018-1e994c659e
binutils-2.29-13.fc27 has been pushed to the Fedora 27 testing repository. If problems still persist, please make note of it in this bug report. See https://fedoraproject.org/wiki/QA:Updates_Testing for instructions on how to install test updates. You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2018-1e994c659e
reclosed