[v3,0/3] Initial BBML2 support for contpte_convert()

Message

Mikołaj Lenczewski March 13, 2025, 10:41 a.m. UTC

Hi All,

This patch series adds adding initial support for eliding
break-before-make requirements on systems that support BBML2 and
additionally guarantee to never raise a conflict abort.

This support elides a TLB invalidation in contpte_convert(). This
leads to a 12% improvement when executing a microbenchmark designed
to force the pathological path where contpte_convert() gets called.
This represents an 80% reduction in the cost of calling
contpte_convert().

This series is based on v6.14-rc4 (d082ecbc71e9).

Patch 1 implements an allow-list of cpus that support BBML2, but with
the additional constraint of never causing TLB conflict aborts. We
settled on this constraint because we will use the feature for kernel
mappings in the future, for which we cannot handle conflict aborts
safely.

Yang Shi has a series at [1] that aims to use BBML2 to enable splitting
the linear map at runtime. This series partially overlaps with it to add
the cpu feature. We believe this series is fully compatible with Yang's
requirements and could go first.

[1]:
  https://lore.kernel.org/linux-arm-kernel/[e-mail]/

Mikołaj Lenczewski (3):
  arm64: Add BBM Level 2 cpu feature
  iommu/arm: Add BBM Level 2 smmu feature
  arm64/mm: Elide tlbi in contpte_convert() under BBML2

 .../admin-guide/kernel-parameters.txt         |  3 +
 arch/arm64/Kconfig                            | 11 +++
 arch/arm64/include/asm/cpucaps.h              |  2 +
 arch/arm64/include/asm/cpufeature.h           |  6 ++
 arch/arm64/kernel/cpufeature.c                | 76 +++++++++++++++++++
 arch/arm64/kernel/pi/idreg-override.c         |  2 +
 arch/arm64/mm/contpte.c                       |  3 +-
 arch/arm64/tools/cpucaps                      |  1 +
 .../iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c   |  3 +
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c   |  3 +
 drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h   |  4 +
 11 files changed, 113 insertions(+), 1 deletion(-)

Comments

Suzuki K Poulose March 13, 2025, 11:22 a.m. UTC | #1

Hi Miko,

On 13/03/2025 10:41, Mikołaj Lenczewski wrote:
> Hi All,
> 
> This patch series adds adding initial support for eliding
> break-before-make requirements on systems that support BBML2 and
> additionally guarantee to never raise a conflict abort.
> 
> This support elides a TLB invalidation in contpte_convert(). This
> leads to a 12% improvement when executing a microbenchmark designed
> to force the pathological path where contpte_convert() gets called.
> This represents an 80% reduction in the cost of calling
> contpte_convert().
> 
> This series is based on v6.14-rc4 (d082ecbc71e9).
> 
> Patch 1 implements an allow-list of cpus that support BBML2, but with
> the additional constraint of never causing TLB conflict aborts. We
> settled on this constraint because we will use the feature for kernel
> mappings in the future, for which we cannot handle conflict aborts
> safely.
> 
> Yang Shi has a series at [1] that aims to use BBML2 to enable splitting
> the linear map at runtime. This series partially overlaps with it to add
> the cpu feature. We believe this series is fully compatible with Yang's
> requirements and could go first.
> 

Nothing about the patch functionality, but :

Minor nit: Generally it is a good idea to add "What changed" from the
previous posting.  That gives the reviewers an idea of what to look for
in the new version. Something like:

Changes since V2:
  {Adding a link to the posting is an added bonus, as we can look up the 
discussions easily}
  - blah blah
  - ..


Cheers
Suzuki


> [1]:
>    https://lore.kernel.org/linux-arm-kernel/[e-mail]/
> 
> Mikołaj Lenczewski (3):
>    arm64: Add BBM Level 2 cpu feature
>    iommu/arm: Add BBM Level 2 smmu feature
>    arm64/mm: Elide tlbi in contpte_convert() under BBML2
> 
>   .../admin-guide/kernel-parameters.txt         |  3 +
>   arch/arm64/Kconfig                            | 11 +++
>   arch/arm64/include/asm/cpucaps.h              |  2 +
>   arch/arm64/include/asm/cpufeature.h           |  6 ++
>   arch/arm64/kernel/cpufeature.c                | 76 +++++++++++++++++++
>   arch/arm64/kernel/pi/idreg-override.c         |  2 +
>   arch/arm64/mm/contpte.c                       |  3 +-
>   arch/arm64/tools/cpucaps                      |  1 +
>   .../iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c   |  3 +
>   drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c   |  3 +
>   drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h   |  4 +
>   11 files changed, 113 insertions(+), 1 deletion(-)
>

Mikołaj Lenczewski March 13, 2025, 11:30 a.m. UTC | #2

On Thu, Mar 13, 2025 at 11:22:07AM +0000, Suzuki K Poulose wrote:
> Hi Miko,
> 
> On 13/03/2025 10:41, Mikołaj Lenczewski wrote:
> > Hi All,
> >  
> Nothing about the patch functionality, but :
> 
> Minor nit: Generally it is a good idea to add "What changed" from the
> previous posting.  That gives the reviewers an idea of what to look for
> in the new version. Something like:
> 
> Changes since V2:
>   {Adding a link to the posting is an added bonus, as we can look up the 
> discussions easily}
>   - blah blah
>   - ..
> 
> 
> Cheers
> Suzuki

Ah, yes. My apologies. Will make sure to include such a changelog with
links to previous versions in future. I do not know why I did not
include it here, other than it not crossing my mind...