diff mbox series

[v1,4/4] config/x86: enable RTE_USE_C11_MEM_MODEL by default

Message ID 20250521111432.207936-5-konstantin.ananyev@huawei.com (mailing list archive)
State New
Delegated to: Thomas Monjalon
Headers
Series ring: some fixes and improvements |

Checks

Context Check Description
ci/checkpatch success coding style OK
ci/loongarch-compilation success Compilation OK
ci/loongarch-unit-testing success Unit Testing PASS
ci/Intel-compilation success Compilation OK
ci/intel-Testing success Testing PASS
ci/intel-Functional success Functional PASS
ci/github-robot: build success github build: passed
ci/iol-marvell-Functional success Functional Testing PASS
ci/iol-mellanox-Performance success Performance Testing PASS
ci/iol-intel-Performance success Performance Testing PASS
ci/iol-intel-Functional success Functional Testing PASS
ci/iol-broadcom-Performance success Performance Testing PASS
ci/iol-abi-testing success Testing PASS
ci/iol-unit-amd64-testing success Testing PASS
ci/iol-compile-amd64-testing warning Testing issues
ci/iol-unit-arm64-testing fail Testing issues
ci/iol-compile-arm64-testing success Testing PASS
ci/iol-sample-apps-testing success Testing PASS
ci/aws-unit-testing fail Unit Testing FAILURE

Commit Message

Konstantin Ananyev May 21, 2025, 11:14 a.m. UTC 
As an attempt to reduce legacy code usage within rte_ring -
enable RTE_USE_C11_MEM_MODEL by default on all x86 platforms.

Signed-off-by: Konstantin Ananyev <konstantin.ananyev@huawei.com>
---
 config/x86/meson.build | 1 +
 1 file changed, 1 insertion(+)

Comments

Morten Brørup May 21, 2025, 7:47 p.m. UTC | #1 
> From: Konstantin Ananyev [mailto:konstantin.ananyev@huawei.com]
> Sent: Wednesday, 21 May 2025 13.15
> 
> As an attempt to reduce legacy code usage within rte_ring -
> enable RTE_USE_C11_MEM_MODEL by default on all x86 platforms.

I theory, after requiring C11, RTE_USE_C11_MEM_MODEL should be the default on all platforms, and the legacy memory model should be optional for performance.

But IIRC, significant performance differences made the legacy memory model better for some platforms, so keeping legacy the default for those platforms makes sense.

If this patch series solved the C11 performance issues on those other platforms, we can get rid of the ring operations using the legacy memory model. Such a change would be suitable for an LTS release.

That said, we could postpone this patch 4 to the 25.11 release.
In theory, patch 3 replaced a rte_atomic_thread_fence() - affecting all memory operations - with some atomic functions affecting only the ring head/tail memory operations.
However, if that theoretical change is 100 % irrelevant on any x86 CPU in reality, then it's acceptable to not wait for 25.11.

> 
> Signed-off-by: Konstantin Ananyev <konstantin.ananyev@huawei.com>
> ---
>  config/x86/meson.build | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/config/x86/meson.build b/config/x86/meson.build
> index c3564b0011..5528eb4960 100644
> --- a/config/x86/meson.build
> +++ b/config/x86/meson.build
> @@ -21,6 +21,7 @@ endif
> 
>  dpdk_conf.set('RTE_CACHE_LINE_SIZE', 64)
>  dpdk_conf.set('RTE_MAX_LCORE', 128)
> +dpdk_conf.set('RTE_USE_C11_MEM_MODEL', true)
> 
>  epyc_zen_cores = {
>      '__znver5__':768,
> --
> 2.43.0
diff mbox series

Patch

diff --git a/config/x86/meson.build b/config/x86/meson.build
index c3564b0011..5528eb4960 100644
--- a/config/x86/meson.build
+++ b/config/x86/meson.build
@@ -21,6 +21,7 @@  endif
 
 dpdk_conf.set('RTE_CACHE_LINE_SIZE', 64)
 dpdk_conf.set('RTE_MAX_LCORE', 128)
+dpdk_conf.set('RTE_USE_C11_MEM_MODEL', true)
 
 epyc_zen_cores = {
     '__znver5__':768,