Hi Luca,
On 14/11/2023 10:03, Luca Fancellu wrote:
>
>
> Introduce a Kconfig for the dom0less feature, enabled by default,
> to be able to choose if the feature should be compiled or not.
>
> Provide static inline stubs when the option is disabled for the
> functions externally visible.
>
> Us
Hi Luca,
On 14/11/2023 10:03, Luca Fancellu wrote:
>
>
> Move static memory and static shared memory code in separate modules
> so that they are included only when the corresponding feature is
> enabled, doing that we modularise the features and we remove some
> ifdefs from the code to improve r
Hi Luca,
On 14/11/2023 10:03, Luca Fancellu wrote:
>
>
> Currently the dom0less feature code is mostly inside domain_build.c
> and setup.c, it is a feature that may not be useful to everyone so
> put the code in a different compilation module in order to make it
> easier to disable the feature i
Hi Ayan,
On 17/11/2023 16:31, Ayan Kumar Halder wrote:
> The MMU specific code in head.S will not be used on MPU systems.
> Instead of introducing more #ifdefs which will bring complexity
> to the code, move MMU related code to mmu/head.S and keep common
> code in head.S. Two notes while moving:
>
Hi Ayan,
On 17/11/2023 16:31, Ayan Kumar Halder wrote:
> All the MMU related functionality have been clubbed together in
> enable_boot_cpu_mm() for booting primary cpu and enable_secondary_cpu_mm() for
> booting secondary cpus.
> This is done in preparation for moving the code related to MMU in MM
On 20.11.23 03:21, osstest service owner wrote:
flight 183794 linux-linus real [real]
http://logs.test-lab.xenproject.org/osstest/logs/183794/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
test-arm64-arm64-examine 8 reboot
From: Wei Chen
Arm platforms support both ACPI and device tree. We don't
want users to select device tree NUMA or ACPI NUMA manually.
We hope users can just enable NUMA for Arm, and device tree
NUMA and ACPI NUMA can be selected depends on device tree
feature and ACPI feature status automatically
From: Wei Chen
Current numa command in documentation is x86 only. Remove
x86 from numa command's arch limitation in this patch.
Also add related entries in the SUPPORT.md and CHANGELOG.md.
Signed-off-by: Wei Chen
Signed-off-by: Henry Wang
---
v6:
- Add the CHANGELOG.md entry, drop Jan's Acked
From: Wei Chen
A NUMA aware device tree will provide a "distance-map" node to
describe distance between any two nodes. This patch introduce a
new helper to parse this distance map.
Note that, since the NUMA device tree binding does not explicitly
specify the range of valid node distance, hence r
From: Wei Chen
Device tree based NUMA doesn't have the proximity domain like
ACPI. So we can return node id directly as arch nid.
Signed-off-by: Wei Chen
Signed-off-by: Henry Wang
---
v5 -> v6:
- Rebase on top of staging without code changes.
v1 -> v5:
- Use numa_node_to_arch_nid instead of du
From: Wei Chen
The NUMA information provided in the host Device-Tree
are only for Xen. For dom0, we want to hide them as they
may be different (for now, dom0 is still not aware of NUMA)
The CPU and memory nodes are recreated from scratch for the
domain. So we already skip the "numa-node-id" prope
From: Wei Chen
node_online_map in smpboot is still needed for Arm when NUMA is
turned off by Kconfig.
Signed-off-by: Wei Chen
Signed-off-by: Henry Wang
---
v6:
- Rebase on top of staging without code changes.
---
xen/arch/arm/smpboot.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/xen
From: Wei Chen
In this patch, we can start to create NUMA system that is
based on device tree.
Signed-off-by: Wei Chen
Signed-off-by: Henry Wang
---
v5 -> v6:
- Rebase on top of staging without code changes.
v1 -> v5:
- Fix coding style: label indented by 1 space.
- replace ~0 by INVALID_PADDR
In the common sysctl command XEN_SYSCTL_physinfo, the cores_per_socket
is calculated based on the cpu_core_mask of CPU0. Currently on Arm
this is a fixed value 1 (can be checked via xl info), which is not
correct. This is because during the Arm cpu online process,
set_cpu_sibling_map() only sets th
From: Wei Chen
Memory blocks' NUMA ID information is stored in device tree's
memory nodes as "numa-node-id". We need a new helper to parse
and verify this ID from memory nodes.
Signed-off-by: Wei Chen
Signed-off-by: Henry Wang
---
v5 -> v6:
- Rebase on top of staging without code changes.
v1 -
From: Wei Chen
In this function, we scan the whole device tree to parse CPU node id,
memory node id and distance-map. Though early_scan_node will invoke
a handler to process memory nodes. If we want to parse memory node id
in that handler, we have to embed NUMA parse code in that handler.
But we
From: Wei Chen
In this patch, we make NUMA node online and add cpu to
its NUMA node. This will make NUMA-aware components
have NUMA affinity data to support their work.
To keep the mostly the same behavior of x86, we use
numa_detect_cpu_node to online node. The difference is that,
we have prepar
From: Wei Chen
Implement the same helper "arch_get_ram_range" as x86 for NUMA
code to get memory bank from Arm bootinfo.
Signed-off-by: Wei Chen
Signed-off-by: Henry Wang
---
v2 -> v6:
- Rebase on top of staging without code changes.
v1 -> v2:
- Use arch_get_ram_range instead of arch_get_memor
From: Wei Chen
We will parse NUMA nodes distances from device tree. So we need
a matrix to record the distances between any two nodes we parsed.
Accordingly, we provide this node_set_distance API for device tree
NUMA to set the distance for any two nodes in this patch. When
NUMA initialization fa
From: Wei Chen
As a memory range described in device tree cannot be split across
multiple nodes. And it is very likely than if you have more than
64 nodes, you may need a lot more than 2 regions per node. So the
default NR_NODE_MEMBLKS value (MAX_NUMNODES * 2) makes no sense
on Arm.
So, for Arm,
From: Wei Chen
Processor NUMA ID information is stored in device tree's processor
node as "numa-node-id". We need a new helper to parse this ID from
processor node. If we get this ID from processor node, this ID's
validity still need to be checked. Once we got a invalid NUMA ID
from any processor
From: Wei Chen
NUMA implementation has a cpu_to_node array to store CPU to NODE
map. Xen is using CPU logical ID in runtime components, so we
use CPU logical ID as CPU index in cpu_to_node.
In device tree case, cpu_logical_map is created in dt_smp_init_cpus.
So, when NUMA is enabled, dt_smp_init
From: Wei Chen
NUMA has one global and one implementation specific switches. For
ACPI NUMA implementation, Xen has acpi_numa, so we introduce
device_tree_numa for device tree NUMA implementation. And use
enumerations to indicate init, off and on status.
arch_numa_disabled will get device_tree_nu
The preparation work to support NUMA on Arm has been merged
and can be found at [1] and [2]. The initial discussions of
the Arm NUMA support can be found at [3].
- Background of this series:
Xen memory allocation and scheduler modules are NUMA aware.
But actually, on x86 has implemented the archi
Pipeline #1077843561 has passed!
Project: xen ( https://gitlab.com/xen-project/xen )
Branch: staging ( https://gitlab.com/xen-project/xen/-/commits/staging )
Commit: 98758ae4 (
https://gitlab.com/xen-project/xen/-/commit/98758ae48974d6d24f999e4d9324e463d326f66f
)
Commit Message: xen: introduc
flight 183791 xen-unstable real [real]
http://logs.test-lab.xenproject.org/osstest/logs/183791/
Failures and problems with tests :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
test-amd64-amd64-xl-qemut-stubdom-debianhvm-amd64-xsm broken
Tests whi
26 matches
Mail list logo
