Skip to content

K
Kernel

openeuler / Kernel
大约 2 年 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 09 9月, 2022 3 次提交
    • O
      !108 tools/bpftool: Add BTF_KIND_FLOAT support · d3762912
      由 openeuler-ci-bot 提交于 
      Merge Pull Request from: @zhengzengkai 
 
Some BPF programs compiled on s390 fail to load, because s390
arch-specific linux headers contain float and double types.

Introduce support for such types by representing them using the new
BTF_KIND_FLOAT. This series deals with libbpf, bpftool, in-kernel BTF
parser as well as selftests and documentation. 
 
Link:https://gitee.com/openeuler/kernel/pulls/108 
Reviewed-by: Xu Kuohai <xukuohai@huawei.com> 
Signed-off-by: Xie XiuQi <xiexiuqi@huawei.com>
      d3762912
    • O
      !96 Intel SPR: IPI virtualization support for VM · ed7fbcc4
      由 openeuler-ci-bot 提交于 
      Merge Pull Request from: @x56Jason 
 
This PR is to enable IPI virtualization support for Intel SPR.

## Intel-Kernel Issue
#I5ODSC

## Test

Use IPI benchmark (https://lore.kernel.org/kvm/20171219085010.4081-1-ynorov@caviumnetworks.com) to do unicast IPI testing ("Normal IPI" case in the benchmark):
 - With host disabled IPI virtualization, and guest enable x2apic mode, we can see a lot of MSR_WR_VMEXITs
 - With host enabled IPI virtualization, and guest enable x2apic mode, MSR_WR_VMEXITs from the benchmark disappears.
 - Comparing host disabled IPIv vs. host enabled IPIv, the IPIv enabled case takes shorter time to do the IPI.

With IPIv enabled:
```
[  517.422480] Normal IPI:              203214459,          237340464 ns
[  518.467208] Normal IPI:              203531652,          237656950 ns
[  519.508544] Normal IPI:              210828433,          269128255 ns
[  520.545573] Normal IPI:              219237071,          257442791 ns
[  521.632722] Normal IPI:              222442577,          264457837 ns
[  522.680563] Normal IPI:              214068805,          264828652 ns
[  523.716551] Normal IPI:              206517521,          246817681 ns
[  524.745078] Normal IPI:              216683548,          264798480 ns
[  525.718962] Normal IPI:              207031786,          250300553 ns
[  526.861676] Normal IPI:              296051188,          342197017 ns
```

With IPIv disabled:
```
[  218.838522] Normal IPI:              268385248,          331187170 ns
[  219.865669] Normal IPI:              277581839,          335817862 ns
[  220.906605] Normal IPI:              302618152,          374527951 ns
[  221.914170] Normal IPI:              270566994,          334971550 ns
[  222.965606] Normal IPI:              284867960,          362923781 ns
[  224.077227] Normal IPI:              346076611,          425008936 ns
[  225.087076] Normal IPI:              306762251,          364504064 ns
[  226.283219] Normal IPI:              385979433,          456133361 ns
[  227.359055] Normal IPI:              281473772,          363612958 ns
[  228.423180] Normal IPI:              297768236,          371935725 ns
```

## Known Issue
N/A

## Default Config Change
N/A
 
 
Link:https://gitee.com/openeuler/kernel/pulls/96 
Reviewed-by: Zheng Zengkai <zhengzengkai@huawei.com> 
Reviewed-by: Liu Chao <liuchao173@huawei.com> 
Reviewed-by: Kevin Zhu <zhukeqian1@huawei.com> 
Signed-off-by: Xie XiuQi <xiexiuqi@huawei.com>
      ed7fbcc4
    • O
      !98 Add support for AMD EPYC Genoa platform · 4e726f6a
      由 openeuler-ci-bot 提交于 
      Merge Pull Request from: @haochengxie 
 
This PR includes minimum essential patches to enable AMD EPYC Genoa platform.


### Issue

Enable AMD EPYC Genoa platform support for openEuler
https://gitee.com/openeuler/kernel/issues/I5NGRU

### Backporting Patches included
crypto:
3438de03 5.15	crypto: ccp - Add support for new CCP/PSP device ID

hwmon：
2ade8fc6 5.14 x86/amd_nb: Add AMD family 19h model 50h PCI ids (front patch)
c8d0d3fa 5.12 hwmon: (k10temp) Zen3 Ryzen Desktop CPUs support(front patch)
02c9dce4 5.15 hwmon: (k10temp) support Zen3 APUs
128066c8 5.15 hwmon: (k10temp) Add additional missing Zen2 and Zen3 APUs
02a2484c 5.15 hwmon: (k10temp) Don't show Tdie for all Zen/Zen2/Zen3 CPU/APU
0e3f52bb 5.15 hwmon: (k10temp) Rework the temperature offset calculation
25572c81 5.15 hwmon: (k10temp) Add support for yellow carp
23c69b90 5.15 hwmon: (k10temp) Remove residues of current and voltage
f707bcb5 5.17 hwmon: (k10temp) Remove unused definitions
4fb0abfe 5.17 x86/amd_nb: Add AMD Family 19h Models (10h-1Fh) and (A0h-AFh) PCI IDs
3cf90efa 5.17 hwmon: (k10temp) Add support for AMD Family 19h Models 10h-1Fh and A0h-AFh
8bb050cd 5.17 hwmon: (k10temp) Support up to 12 CCDs on AMD Family of processors

EDAC：
f9571124	5.17	EDAC: Add RDDR5 and LRDDR5 memory types
e2be5955	5.17	EDAC/amd64: Add support for AMD Family 19h Models 10h-1Fh and A0h-AFh
75aeaaf2	5.18	EDAC/amd64: Set memory type per DIMM
2151c84e	5.18	EDAC/amd64: Add new register offset support and related changes

MCE:
94a311ce    5.14    x86/MCE/AMD, EDAC/mce_amd: Add new SMCA bank types
5176a93a	5.17	x86/MCE/AMD, EDAC/mce_amd: Add new SMCA bank types
f38ce910    5.16    x86/MCE/AMD: Export smca_get_bank_type symbol
0b746e8c    5.17    x86/MCE/AMD, EDAC/amd64: Move address translation to AMD64 EDAC
91f75eb4	5.17	x86/MCE/AMD, EDAC/mce_amd: Support non-uniform MCA bank type enumeration 

kvm Essential:
03ca4589 5.13 KVM: x86: Prevent KVM SVM from loading on kernels with 5-level paging
43e540cc 5.15 KVM: SVM: Add 5-level page table support for SVM
cb0f722a 5.15 KVM: x86/mmu: Support shadowing NPT when 5-level paging is enabled
39150352 5.17 KVM: x86: SVM: move avic definitions from AMD's spec to svm.h
4a204f78 5.18-rc1 KVM: SVM: Allow AVIC support on system w/ physical APIC ID > 255

 
 
Link:https://gitee.com/openeuler/kernel/pulls/98 
Reviewed-by: Kevin Zhu <zhukeqian1@huawei.com> 
Signed-off-by: Xie XiuQi <xiexiuqi@huawei.com>
      4e726f6a
  3. 08 9月, 2022 6 次提交
  5. 06 9月, 2022 14 次提交
  7. 05 9月, 2022 12 次提交
  9. 02 9月, 2022 1 次提交
    • O
      !69 samples: bpf: Add sample BMC for Redis v5 · d8768ca2
      由 openeuler-ci-bot 提交于 
      Merge Pull Request from: @zhengzengkai 
 
BMC is an in-kernel key-value cache implemented in BPF and proposed by
paper [1]. The paper discussed BMC for memcached, obtaining at least
6x performance speedup.

This patch implements a sample BMC for Redis. Paper [1] implements BMC
in XDP, bypassing the kernel network stack totally. Since Redis is based
on TCP protocol, and it's almost impossible to fully process TCP traffic
in XDP, so this patch implements BMC in sockmap, which locates at the
top of kernel network stack. Since kernel network stack is not bypassed,
the speedup is not significant. Any way, this is only a sample
implementation, and performance improvements can be continuously
optimized.

See [2] for details on how to build samples/bpf.

Output files:
 samples/bpf/bmctool
 samples/bpf/bmc/bpf.o

Sample usage:
 bmctool prog load -p 6379 ./bmc/bpf.o  # load bmc bpf prog and attach it
                                        # to sockets with listen port 6379

 bmctool stat                           # dump bmc status

 bmctool prog unload                    # detach and unload bmc prog

[1] https://www.usenix.org/conference/nsdi21/presentation/ghigoff
[2] https://www.kernel.org/doc/readme/samples-bpf-README.rst 
 
Link:https://gitee.com/openeuler/kernel/pulls/69 
Reviewed-by: Xie XiuQi <xiexiuqi@huawei.com> 
Signed-off-by: Xie XiuQi <xiexiuqi@huawei.com>
      d8768ca2
  11. 01 9月, 2022 1 次提交
  13. 31 8月, 2022 3 次提交