multi-die/package support
Get series with b4
b4 am [email protected]Versions newest first
- v1 Latest11 patches 30 repliesIn review
Patches in v1 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/11 | x86 topology: fix doc typoDocumentation/ | +1 | 0 replies | |
| 2/11 | topolgy: simplify cputopology.txt formatting and wordingDocumentation/ | +17 | 2 replies | |
| 3/11 | x86 topology: Add CPUID.1F multi-die/package supportDocumentation/ arch/ arch/ +1 more | +73 | 12 replies | |
| 4/11 | cpu topology: export die_idDocumentation/ arch/ drivers/ +1 more | +16 | 0 replies | |
| 5/11 | x86 topology: export die_siblingsDocumentation/ arch/ arch/ +4 more | +42 | 7 replies | |
| 6/11 | x86 topology: define topology_unique_die_id()arch/ | +3 | 0 replies | |
| 7/11 | powercap/intel_rapl: simplify rapl_find_package()drivers/ | +9 | 1 reply | |
| 8/11 | powercap/intel_rapl: Support multi-die/packagedrivers/ | +2 | 5 replies | |
| 9/11 | powercap/intel_rapl: update rapl domain name and debug messagesdrivers/ | +32 | 1 reply | |
| 10/11 | thermal/x86_pkg_temp_thermal: Support multi-die/packagedrivers/ | +5 | 0 replies | |
| 11/11 | hwmon/coretemp: Support multi-die/packagedrivers/ | +5 | 2 replies |
Reviews trailers per patch
4 of 11 patches have a Reviewed-by or Acked-by.
| Patch | Rafael J. Wysocki | Guenter Roeck |
|---|---|---|
| 1/11 | ||
| 2/11 | ||
| 3/11 | ||
| 4/11 | ||
| 5/11 | ||
| 6/11 | ||
| 7/11 | A-b | |
| 8/11 | A-b | |
| 9/11 | A-b | |
| 10/11 | ||
| 11/11 | A-b A-b |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.
Cover letter [PATCH 0/11] multi-die/package support
This patch series does 3 things.
1. Parses the new CPUID.1F leaf to discover multi-die/package topology
2. Exports knowledge of multi-die topology inside the kernel, and also to user space via the topology sysfs.
3. Updates parts of the kernel that need to know the difference between a die-scope MSR and a package_scope MSR.
To follow, Kan will send some additional perf-uncore patches for #3.
Also not include here are some updates to user-space utilities, cpuid, lscpu, turbostat, x86_energy_perf_policy.
Rafael suggests that the most practical way to handle this series is via the x86 tree.
Thanks,
-Len