For sharing with Lorenzo.
I'm holding off on posting the CMA patches until the open question around
DOE / auxiliary bus is resolved. Not clear if we'll end up back where we
Note that other than the devm_ fix patch I've not tweaked the DOE series
for review comments.
Base is cxl/next 53989fad1286e652ea3655ae3367ba698da8d2ff
(the tree might have moved again in the meantime)
Ira Weiny (1):
cxl/cdat: Parse out DSMAS data from CDAT table
Jonathan Cameron (9):
PCI: Add vendor ID for the PCI SIG
PCI/DOE: Add Data Object Exchange Aux Driver
cxl/pci: Add DOE Auxiliary Devices
cxl/mem: Add CDAT table reading from DOE
lib/asn1_encoder: Add a function to encode many byte integer values.
spdm: Introduce a library for DMTF SPDM
PCI/CMA: Initial support for Component Measurement and Authentication
cxl/pci: Add really basic CMA authentication support.
cxl: Fixes for wrong device in devm_ calls.
drivers/cxl/Kconfig | 2 +
drivers/cxl/cdat.h | 81 +++
drivers/cxl/core/memdev.c | 157 +++++
drivers/cxl/cxl.h | 10 +
drivers/cxl/cxlmem.h | 50 ++
drivers/cxl/pci.c | 231 ++++++-
drivers/pci/Kconfig | 19 +
drivers/pci/Makefile | 5 +
drivers/pci/cma.c | 105 +++
drivers/pci/doe.c | 701 +++++++++++++++++++
include/linux/asn1_encoder.h | 3 +
include/linux/pci-cma.h | 19 +
include/linux/pci-doe.h | 63 ++
include/linux/pci_ids.h | 1 +
include/linux/spdm.h | 104 +++
include/uapi/linux/pci_regs.h | 29 +-
lib/Kconfig | 3 +
lib/Makefile | 2 +
lib/asn1_encoder.c | 54 ++
lib/spdm.c | 1196 +++++++++++++++++++++++++++++++++
20 files changed, 2833 insertions(+), 2 deletions(-)
create mode 100644 drivers/cxl/cdat.h
create mode 100644 drivers/pci/cma.c
create mode 100644 drivers/pci/doe.c
create mode 100644 include/linux/pci-cma.h
create mode 100644 include/linux/pci-doe.h
create mode 100644 include/linux/spdm.h
create mode 100644 lib/spdm.c
This is the follow-up work to support cluster scheduler. Previously
we have added cluster level in the scheduler to make tasks spread
between clusters to bring more memory bandwidth and decrease
cache contention. But it may hurt some workloads which are sensitive
to the communication latency as they will be placed across clusters.
We modified the select_idle_cpu() on the wake affine path in this
series, expecting the wake affined task to be woken more likely on
the same cluster with the waker. The latency will be decreased as
the waker and wakee in the same cluster may benefit from the hot
L3 cache tag.
Hi Tim and Barry,
This the modified patch of packing path of cluster scheduler and tests
have been done on Kunpeng 920 2-socket 4-NUMA 128core platform, with
8 clusters on each NUMA. Patches based on 5.16-rc1.
Compared to the previous one, we give up scanning the first cpu of the
cluster as the cpu id may not be continuous. So we pickup the way of
scanning cluster first before LLC. The result from tbench and pgbench
are rather positive.
Barry Song (2):
sched: Add per_cpu cluster domain info
sched/fair: Scan cluster before scanning LLC in wake-up path
include/linux/sched/sd_flags.h | 9 ++++++++
include/linux/sched/topology.h | 2 +-
kernel/sched/fair.c | 41 +++++++++++++++++++++++++++++++++-
kernel/sched/sched.h | 1 +
kernel/sched/topology.c | 5 +++++
5 files changed, 56 insertions(+), 2 deletions(-)
OP-TEE Contributions (LOC) monthly meeting is planned for Thursday Nov 25
Following topic is on the agenda:
- Walkthrough of proposal on sharing of hardware resources between multiple
secure entities - Jens Wiklander.
If you have any other topics you'd like to discuss, please let us know and
we can schedule them.
Date/time: Nov 25(a)16.00 (UTC)
Connection details: https://www.trustedfirmware.org/meetings/
Meeting notes: http://bit.ly/loc-notes
Ruchika on behalf of the Linaro OP-TEE team
I was about to test your doe-spdm-v1 branch but it does not compile,
I was wondering if you forgot to push some files.
Have you managed to rebase it on top of the latest DOE patches ?
Please let me know, thank you very much.