Marcin Juszkiewicz
2016-02-10 12:16:04 UTC
Hi
I tried to build Linux kernel 4.5-rc3-git1 today. On APM Mustang. And it
failed:
------------------------------------------------------------------------
/tmp/ccKTs5Cd.s: Assembler messages:
/tmp/ccKTs5Cd.s:1617: Error: selected processor does not support `ldaddal w0,w30,[x1]'
/tmp/ccKTs5Cd.s:1622: Error: attempt to move .org backwards
scripts/Makefile.build:258: recipe for target 'init/do_mounts.o' failed
------------------------------------------------------------------------
Checked what it means and got to commit:
------------------------------------------------------------------------
commit c09d6a04d17d730b0463207a26ece082772b59ee
Author: Will Deacon <***@arm.com>
Date: Tue Feb 3 16:14:13 2015 +0000
arm64: atomics: patch in lse instructions when supported by the CPU
On CPUs which support the LSE atomic instructions introduced in ARMv8.1,
it makes sense to use them in preference to ll/sc sequences.
This patch introduces runtime patching of atomic_t and atomic64_t
routines so that the call-site for the out-of-line ll/sc sequences is
patched with an LSE atomic instruction when we detect that
the CPU supports it.
If binutils is not recent enough to assemble the LSE instructions, then
the ll/sc sequences are inlined as though CONFIG_ARM64_LSE_ATOMICS=n.
Reviewed-by: Catalin Marinas <***@arm.com>
Signed-off-by: Will Deacon <***@arm.com>
------------------------------------------------------------------------
If I revert arch/arm64/Makefile change (so disable use of LSE
instructions) kernel builds.
So I contacted Will Deacon, told him about a bug and that it is with
binutils 2.26 and gcc 6-prerelease.
He replied that failure is gcc6 related and gave me a link:
https://gcc.gnu.org/ml/gcc-patches/2016-02/msg00313.html
Thread mentions also kernel patch for this issue.
I tried to build Linux kernel 4.5-rc3-git1 today. On APM Mustang. And it
failed:
------------------------------------------------------------------------
/tmp/ccKTs5Cd.s: Assembler messages:
/tmp/ccKTs5Cd.s:1617: Error: selected processor does not support `ldaddal w0,w30,[x1]'
/tmp/ccKTs5Cd.s:1622: Error: attempt to move .org backwards
scripts/Makefile.build:258: recipe for target 'init/do_mounts.o' failed
------------------------------------------------------------------------
Checked what it means and got to commit:
------------------------------------------------------------------------
commit c09d6a04d17d730b0463207a26ece082772b59ee
Author: Will Deacon <***@arm.com>
Date: Tue Feb 3 16:14:13 2015 +0000
arm64: atomics: patch in lse instructions when supported by the CPU
On CPUs which support the LSE atomic instructions introduced in ARMv8.1,
it makes sense to use them in preference to ll/sc sequences.
This patch introduces runtime patching of atomic_t and atomic64_t
routines so that the call-site for the out-of-line ll/sc sequences is
patched with an LSE atomic instruction when we detect that
the CPU supports it.
If binutils is not recent enough to assemble the LSE instructions, then
the ll/sc sequences are inlined as though CONFIG_ARM64_LSE_ATOMICS=n.
Reviewed-by: Catalin Marinas <***@arm.com>
Signed-off-by: Will Deacon <***@arm.com>
------------------------------------------------------------------------
If I revert arch/arm64/Makefile change (so disable use of LSE
instructions) kernel builds.
So I contacted Will Deacon, told him about a bug and that it is with
binutils 2.26 and gcc 6-prerelease.
He replied that failure is gcc6 related and gave me a link:
https://gcc.gnu.org/ml/gcc-patches/2016-02/msg00313.html
Thread mentions also kernel patch for this issue.