Re: JITs and 52-bit VA

On 28 April 2016 at 14:24, Peter Maydell wrote: > On 28 April 2016 at 14:17, Arnd Bergmann wrote: >> One simple (from the kernel's perspective, not from the JIT) approach >> might be to always use MAP_FIXED whenever an allocation is made for >> memory that needs these special pointers, and then m

Cortex A57 Optimisation Guide

For those interested in optimising code for the Cortex-A57, the following guide has just been released: http://infocenter.arm.com/help/topic/com.arm.doc.uan0015a/cortex_a57_software_optimisation_guide_external.pdf Cheers, -- Steve ___ linaro-dev mailing

Re: [PATCH v3 Ceph] crc32c: add aarch64 optimized crc32c implementation

est shows ~4x performance increase versus sctp. Hi Yazen, Thanks, this is looking good. > > Signed-off-by: Yazen Ghannam Please add: Reviewed-by: Steve Capper This can be sent to ceph-de...@vger.kernel.org, let's see what upstream think ;-). Cheers, -- Steve >

Re: [PATCH v2 Ceph] crc32c: add aarch64 optimized crc32c implementation

Hi Yazen, This is looking good, just a few minor comments below. Cheers, -- Steve On Fri, Jan 23, 2015 at 09:13:42AM -0600, Yazen Ghannam wrote: > ARMv8 defines a set of optional CRC32/CRC32C instructions. > This patch defines an optimized function that uses these > instructions when available

Re: [PATCH Ceph] crc32c: add arm64/aarch64 optimized crc32c implementation

Hi Yazen, I have a few comments below. Cheers, -- Steve On Wed, Jan 21, 2015 at 02:17:47PM -0600, Yazen Ghannam wrote: > Tested on AMD Seattle. > Passes all crc32c unit tests. > ~4x performance increase versus sctp. Was that perf result from the unit test? Also describe the reason for the pa

Re: [Libhugetlbfs-devel] [PATCH 0/5] libhugetlbfs - Aarch64 support and some fixes

No probs, thank you Eric. Cheers, -- Steve ___ linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev

Re: [PATCH 0/5] libhugetlbfs - Aarch64 support and some fixes

On Tue, Sep 10, 2013 at 02:11:27PM +0100, Steve Capper wrote: > Hello, > This series adds Aarch64 support and makes some minor tweaks. > > The first two patches of this series add Aarch64 support to > libhugetlbfs. (Starting from 3.11-rc1, the Linux Kernel supports > HugeTLB

[PATCH 5/5] Cleanup ARM ld.hugetlbfs HTLB_LINK logic

this. Signed-off-by: Steve Capper --- ld.hugetlbfs | 5 - 1 file changed, 5 deletions(-) diff --git a/ld.hugetlbfs b/ld.hugetlbfs index 5128aa2..ba9e00a 100755 --- a/ld.hugetlbfs +++ b/ld.hugetlbfs @@ -79,11 +79,6 @@ if [ -n "$HTLB_LINK" ]; then HTLB_ALIGN=""

[PATCH 3/5] Add PROT_NONE to the mprotect test.

hould succeed. Signed-off-by: Steve Capper --- tests/mprotect.c | 6 ++ 1 file changed, 6 insertions(+) diff --git a/tests/mprotect.c b/tests/mprotect.c index aa4673e..db6a662 100644 --- a/tests/mprotect.c +++ b/tests/mprotect.c @@ -213,5 +213,11 @@ int main(int argc, char

[PATCH 2/5] Aarch64 unit test fixes.

256MB slices. Signed-off-by: Steve Capper --- tests/icache-hygiene.c| 7 --- tests/mremap-expand-slice-collision.c | 2 +- 2 files changed, 5 insertions(+), 4 deletions(-) diff --git a/tests/icache-hygiene.c b/tests/icache-hygiene.c index 51792b3..876ce10 100644 --- a/tests

[PATCH 0/5] libhugetlbfs - Aarch64 support and some fixes

unit test, and the linkhuge_rw test is enabled for 64 bit where there aren't any custom ldscripts. The final patch clears up the superfluous ARM ld.hugetlbfs HTLB_LINK logic. Any comments would be appreciated. Cheers, -- Steve Steve Capper (5): Aarch64 support. Aarch64 unit test fixes.

[PATCH 1/5] Aarch64 support.

This patch adds support for Aarch64. As with ARMv7, We do not add the xBT/xBDT style linker scripts as these have been deprecated in favour of adjusting the page sizes via command line parameter to ld. Signed-off-by: Steve Capper --- Makefile | 7 +++ ld.hugetlbfs | 2 +- sys

[PATCH 4/5] Add linkhuge_rw test to 64 bit && !CUSTOM_LDSCIPTS

: Steve Capper --- tests/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/Makefile b/tests/Makefile index 231e3b0..9140e72 100644 --- a/tests/Makefile +++ b/tests/Makefile @@ -54,7 +54,7 @@ ifeq ($(CUSTOM_LDSCRIPTS),yes) TESTS += $(LDSCRIPT_TESTS) $(HUGELINK_TESTS

[PATCH 2/2] ARM: Add IT instructions to inline assembler

s may fail to build for Thumb under older build systems. To remedy this, we add the IT instructions to gcc_armv7.h that would normally be added implicitly by the assembler. Signed-off-by: Steve Capper --- include/tbb/machine/gcc_armv7.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/i

[PATCH 0/2] Fixes for TBB ARM support

Wheezy (with `make test' and `make examples'). Any comments/critique welcome, as I aim to send these patches upstream. Leif Lindholm (1): ARM: Add IT instructions to inline assembler Steve Capper (1): ARM: Correct build architecture detection logic build/linux.inc

[PATCH 1/2] ARM: Correct build architecture detection logic

Our initial patches appear to have been mangled a bit. We check for armv7l (not seventy-one), and correct the whitespace. Signed-off-by: Steve Capper --- build/linux.inc | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/build/linux.inc b/build/linux.inc index 1c96d86