Re: [PATCH] x86/build: Move _etext to actual end of .text

2019-06-20 Thread Kees Cook
On Wed, Jun 19, 2019 at 12:37:11PM -0600, Ross Zwisler wrote: > On Sun, Jun 9, 2019 at 1:00 PM Johannes Hirte > wrote: > > On 2019 Jun 09, Klaus Kusche wrote: > > > Hello, > > > > > > Same problem for linux 5.1.7: > > > Kernel building fails with the same relocation error. > > > > > > 5.1.5 does

Re: [PATCH] x86/build: Move _etext to actual end of .text

2019-06-19 Thread Ross Zwisler
On Sun, Jun 9, 2019 at 1:00 PM Johannes Hirte wrote: > On 2019 Jun 09, Klaus Kusche wrote: > > Hello, > > > > Same problem for linux 5.1.7: > > Kernel building fails with the same relocation error. > > > > 5.1.5 does not have the problem, builds fine for me. > > > > Is there anything I can do to

Re: [PATCH] x86/build: Move _etext to actual end of .text

2019-06-09 Thread Johannes Hirte
On 2019 Jun 09, Klaus Kusche wrote: > > Hello, > > Same problem for linux 5.1.7: > Kernel building fails with the same relocation error. > > 5.1.5 does not have the problem, builds fine for me. > > Is there anything I can do to investigate the problem? > Please try linux 5.1.8. The

Re: [PATCH] x86/build: Move _etext to actual end of .text

2019-06-09 Thread Klaus Kusche
Hello, Same problem for linux 5.1.7: Kernel building fails with the same relocation error. 5.1.5 does not have the problem, builds fine for me. Is there anything I can do to investigate the problem? -- Prof. Dr. Klaus Kusche Private address: Rosenberg 41, 07546 Gera, Germany +49 365

Re: [PATCH] x86/build: Move _etext to actual end of .text

2019-06-05 Thread Greg KH
On Wed, Jun 05, 2019 at 11:16:06AM -0700, Kees Cook wrote: > On Wed, Jun 05, 2019 at 11:08:13AM -0500, Alec Ari wrote: > > I'm having this problem too, build is failing: > > > > Invalid absolute R_X86_64_32S relocation: _etext > > > > I stayed on the 4.14 branch to help prevent these kind of

Re: [PATCH] x86/build: Move _etext to actual end of .text

2019-06-05 Thread Kees Cook
On Wed, Jun 05, 2019 at 11:08:13AM -0500, Alec Ari wrote: > I'm having this problem too, build is failing: > > Invalid absolute R_X86_64_32S relocation: _etext > > I stayed on the 4.14 branch to help prevent these kind of breakages, > so much for that idea. Gentoo GCC 8.3.0. It seems to be a

Re: [PATCH] x86/build: Move _etext to actual end of .text

2019-06-05 Thread Alec Ari
Hi, I'm having this problem too, build is failing: Invalid absolute R_X86_64_32S relocation: _etext I stayed on the 4.14 branch to help prevent these kind of breakages, so much for that idea. Gentoo GCC 8.3.0. Alec

Re: [PATCH] x86/build: Move _etext to actual end of .text

2019-06-01 Thread Klaus Kusche
Hello, same problem here. gcc version 9.1.0 (Gentoo 9.1.0 p1.0) linux-5.1.6 RELOCS arch/x86/boot/compressed/vmlinux.relocs Invalid absolute R_X86_64_32S relocation: _etext make[2]: *** [arch/x86/boot/compressed/Makefile:130: arch/x86/boot/compressed/vmlinux.relocs] Error 1 make[2]: ***

Re: [PATCH] x86/build: Move _etext to actual end of .text

2019-05-16 Thread Kees Cook
On Thu, May 16, 2019 at 03:56:07PM +0200, Johannes Hirte wrote: > On 2019 Mai 15, Kees Cook wrote: > > Various stupid questions: did you wipe the whole bulid tree and start > > clean? > > No I didn't. And this fixed it now. After a distclean I'm unable to > reproduce it. So sorry for the noise.

Re: [PATCH] x86/build: Move _etext to actual end of .text

2019-05-16 Thread Johannes Hirte
On 2019 Mai 15, Kees Cook wrote: > On Tue, May 14, 2019 at 06:10:55PM +0200, Johannes Hirte wrote: > > On 2019 Mai 14, Kees Cook wrote: > > > On Tue, May 14, 2019 at 02:04:21PM +0200, Johannes Hirte wrote: > > > > This breaks the build on my system: > > > > > > > > RELOCS

Re: [PATCH] x86/build: Move _etext to actual end of .text

2019-05-15 Thread Kees Cook
On Tue, May 14, 2019 at 06:10:55PM +0200, Johannes Hirte wrote: > On 2019 Mai 14, Kees Cook wrote: > > On Tue, May 14, 2019 at 02:04:21PM +0200, Johannes Hirte wrote: > > > This breaks the build on my system: > > > > > > RELOCS arch/x86/boot/compressed/vmlinux.relocs > > > CC

Re: [PATCH] x86/build: Move _etext to actual end of .text

2019-05-14 Thread Johannes Hirte
On 2019 Mai 14, Kees Cook wrote: > On Tue, May 14, 2019 at 02:04:21PM +0200, Johannes Hirte wrote: > > On 2019 Apr 23, Kees Cook wrote: > > > When building x86 with Clang LTO and CFI, CFI jump regions are > > > automatically added to the end of the .text section late in linking. As a > > > result,

Re: [PATCH] x86/build: Move _etext to actual end of .text

2019-05-14 Thread Kees Cook
On Tue, May 14, 2019 at 02:04:21PM +0200, Johannes Hirte wrote: > On 2019 Apr 23, Kees Cook wrote: > > When building x86 with Clang LTO and CFI, CFI jump regions are > > automatically added to the end of the .text section late in linking. As a > > result, the _etext position was being labelled

Re: [PATCH] x86/build: Move _etext to actual end of .text

2019-05-14 Thread Johannes Hirte
On 2019 Apr 23, Kees Cook wrote: > When building x86 with Clang LTO and CFI, CFI jump regions are > automatically added to the end of the .text section late in linking. As a > result, the _etext position was being labelled before the appended jump > regions, causing confusion about where the

[PATCH] x86/build: Move _etext to actual end of .text

2019-04-23 Thread Kees Cook
When building x86 with Clang LTO and CFI, CFI jump regions are automatically added to the end of the .text section late in linking. As a result, the _etext position was being labelled before the appended jump regions, causing confusion about where the boundaries of the executable region actually