2020-09-21 09:05:23

by Satheesh Rajendran

[permalink] [raw]
Subject: [PATCH V2] Doc: admin-guide: Add entry for kvm_cma_resv_ratio kernel param

From: Satheesh Rajendran <[email protected]>

Add document entry for kvm_cma_resv_ratio kernel param which
is used to alter the KVM contiguous memory allocation percentage
for hash pagetable allocation used by hash mode PowerPC KVM guests.

Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: Paul Mackerras <[email protected]>
Cc: Michael Ellerman <[email protected]>
Cc: Jonathan Corbet <[email protected]>
Reviewed-by: Randy Dunlap <[email protected]>
Signed-off-by: Satheesh Rajendran <[email protected]>
---

V2:
Addressed review comments from Randy.

V1: https://lkml.org/lkml/2020/9/16/72
---
Documentation/admin-guide/kernel-parameters.txt | 8 ++++++++
1 file changed, 8 insertions(+)

diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index a1068742a6df..932ed45740c9 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -2258,6 +2258,14 @@
[KVM,ARM] Allow use of GICv4 for direct injection of
LPIs.

+ kvm_cma_resv_ratio=n [PPC]
+ Reserves given percentage from system memory area for
+ contiguous memory allocation for KVM hash pagetable
+ allocation.
+ By default it reserves 5% of total system memory.
+ Format: <integer>
+ Default: 5
+
kvm-intel.ept= [KVM,Intel] Disable extended page tables
(virtualized MMU) support on capable Intel chips.
Default is 1 (enabled)
--
2.26.2


2020-09-21 16:02:12

by Randy Dunlap

[permalink] [raw]
Subject: Re: [PATCH V2] Doc: admin-guide: Add entry for kvm_cma_resv_ratio kernel param

On 9/21/20 2:02 AM, [email protected] wrote:
> From: Satheesh Rajendran <[email protected]>
>
> Add document entry for kvm_cma_resv_ratio kernel param which
> is used to alter the KVM contiguous memory allocation percentage
> for hash pagetable allocation used by hash mode PowerPC KVM guests.
>
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
> Cc: Paul Mackerras <[email protected]>
> Cc: Michael Ellerman <[email protected]>
> Cc: Jonathan Corbet <[email protected]>
> Reviewed-by: Randy Dunlap <[email protected]>
> Signed-off-by: Satheesh Rajendran <[email protected]>

Hi,

> ---
>
> V2:
> Addressed review comments from Randy.
>
> V1: https://lkml.org/lkml/2020/9/16/72

In reply to V1, I didn't add a Reviewed-by: tag, but I can now.

Reviewed-by: Randy Dunlap <[email protected]>

> ---
> Documentation/admin-guide/kernel-parameters.txt | 8 ++++++++
> 1 file changed, 8 insertions(+)
>
> diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
> index a1068742a6df..932ed45740c9 100644
> --- a/Documentation/admin-guide/kernel-parameters.txt
> +++ b/Documentation/admin-guide/kernel-parameters.txt
> @@ -2258,6 +2258,14 @@
> [KVM,ARM] Allow use of GICv4 for direct injection of
> LPIs.
>
> + kvm_cma_resv_ratio=n [PPC]
> + Reserves given percentage from system memory area for
> + contiguous memory allocation for KVM hash pagetable
> + allocation.
> + By default it reserves 5% of total system memory.
> + Format: <integer>
> + Default: 5
> +
> kvm-intel.ept= [KVM,Intel] Disable extended page tables
> (virtualized MMU) support on capable Intel chips.
> Default is 1 (enabled)
>

thanks.
--
~Randy

2020-09-22 00:26:56

by Paul Mackerras

[permalink] [raw]
Subject: Re: [PATCH V2] Doc: admin-guide: Add entry for kvm_cma_resv_ratio kernel param

On Mon, Sep 21, 2020 at 02:32:20PM +0530, [email protected] wrote:
> From: Satheesh Rajendran <[email protected]>
>
> Add document entry for kvm_cma_resv_ratio kernel param which
> is used to alter the KVM contiguous memory allocation percentage
> for hash pagetable allocation used by hash mode PowerPC KVM guests.
>
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
> Cc: Paul Mackerras <[email protected]>
> Cc: Michael Ellerman <[email protected]>
> Cc: Jonathan Corbet <[email protected]>
> Reviewed-by: Randy Dunlap <[email protected]>
> Signed-off-by: Satheesh Rajendran <[email protected]>
> ---
>
> V2:
> Addressed review comments from Randy.
>
> V1: https://lkml.org/lkml/2020/9/16/72
> ---
> Documentation/admin-guide/kernel-parameters.txt | 8 ++++++++
> 1 file changed, 8 insertions(+)
>
> diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
> index a1068742a6df..932ed45740c9 100644
> --- a/Documentation/admin-guide/kernel-parameters.txt
> +++ b/Documentation/admin-guide/kernel-parameters.txt
> @@ -2258,6 +2258,14 @@
> [KVM,ARM] Allow use of GICv4 for direct injection of
> LPIs.
>
> + kvm_cma_resv_ratio=n [PPC]
> + Reserves given percentage from system memory area for
> + contiguous memory allocation for KVM hash pagetable
> + allocation.
> + By default it reserves 5% of total system memory.

I am concerned that using the term "reserve" here could give the
impression that this memory is then not available for any other use.
It is in fact available for other uses as long as they are movable
allocations. So this memory is available for uses such as process
anonymous memory and page cache, just not for things like kmalloc.

I'm not sure what would be a better term than "reserve", though.
Perhaps we need to add a sentence something like "The reserved memory
is available for use as process memory and page cache when it is not
being used by KVM."

Paul.

2020-09-24 16:38:53

by Jonathan Corbet

[permalink] [raw]
Subject: Re: [PATCH V2] Doc: admin-guide: Add entry for kvm_cma_resv_ratio kernel param

On Mon, 21 Sep 2020 14:32:20 +0530
[email protected] wrote:

> From: Satheesh Rajendran <[email protected]>
>
> Add document entry for kvm_cma_resv_ratio kernel param which
> is used to alter the KVM contiguous memory allocation percentage
> for hash pagetable allocation used by hash mode PowerPC KVM guests.
>
> Cc: [email protected]
> Cc: [email protected]
> Cc: [email protected]
> Cc: Paul Mackerras <[email protected]>
> Cc: Michael Ellerman <[email protected]>
> Cc: Jonathan Corbet <[email protected]>
> Reviewed-by: Randy Dunlap <[email protected]>
> Signed-off-by: Satheesh Rajendran <[email protected]>
> ---
>
> V2:
> Addressed review comments from Randy.
>
> V1: https://lkml.org/lkml/2020/9/16/72

Applied, thanks.

jon