[Feature] SKX: Support crystall ridge / far / near memory indication in PEBS

Bug #1591813 reported by XiongZhang
4
This bug affects 1 person
Affects Status Importance Assigned to Milestone
intel
Fix Released
Undecided
Unassigned
linux (Ubuntu)
Fix Released
Medium
Seth Forshee

Bug Description

On SKX PEBS bits should report whether a memory address DLA sample is crystal ridge:
hey have two PEBS encodings for L4_HIT Local and L4_HIT Remote. Since we have no L4, we setup the PRE for Server so that the PRE encodings that map onto L4_HIT are actually crystal ridge. So PEBS 0x8 (L4 HIT Local) should mean serviced by local crystal ridge memory on a SKX part, and PEBS 0x9 (L4 HIT Remote) should mean serviced by remote crystal ridge memory on a SKX part.
That said, there are multiple steps in the chain that gets that information from the memory controller to the core, and I don't think any of those steps has been formally validated.
We need to extend perf to report these distinctions. First needs some testing if it actually works. Would require kernel changes, ABI changes, perf report changes.

HW: Purley

Upstream schedule:
   kernel: TBD

Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

Any updates on the kernel target here? Did this make v4.8? Or should we defer this to 17.10?

Revision history for this message
XiongZhang (xiong-y-zhang) wrote :

Please defer it to 17.10.

Revision history for this message
quanxian (quanxian-wang) wrote :

95298355143f 6ae5fa61d27d

Seth Forshee (sforshee)
information type: Proprietary → Public
Changed in linux (Ubuntu):
assignee: nobody → Seth Forshee (sforshee)
importance: Undecided → Medium
status: New → Triaged
Seth Forshee (sforshee)
Changed in linux (Ubuntu):
status: Triaged → Fix Committed
Revision history for this message
Launchpad Janitor (janitor) wrote :
Download full text (7.8 KiB)

This bug was fixed in the package linux - 4.13.0-12.13

---------------
linux (4.13.0-12.13) artful; urgency=low

  * linux: 4.13.0-12.13 -proposed tracker (LP: #1718980)

  * [Feature] SKX: Support crystall ridge / far / near memory indication in PEBS
    (LP: #1591813)
    - perf/x86: Move Nehalem PEBS code to flag
    - perf/x86: Fix data source decoding for Skylake

  * Upgrade to 4.13.0-11.12 in artful amd64 VM breaks display on wayland
    (LP: #1718679)
    - [Config] CONFIG_DRM_VBOXVIDEO=n

  * ipmmu-vmsa driver breaks arm64 boots (LP: #1718734)
    - [Config] Disable CONFIG_IPMMU_VMSA on arm64

  * Vlun resize request could fail with cxlflash driver (LP: #1713575)
    - scsi: cxlflash: Fix vlun resize failure in the shrink path

  * multipath -ll is not showing the disks which are actually multipath
    (LP: #1718397)
    - fs: aio: fix the increment of aio-nr and counting against aio-max-nr

  * [Feature] Crystal Ridge - BTT - Rework error clearing (LP: #1704350)
    - libnvdimm, btt: fix a missed NVDIMM_IO_ATOMIC case in the write path
    - libnvdimm, btt: refactor map entry operations with macros
    - libnvdimm, btt: ensure that flags were also unchanged during a map_read
    - libnvdimm, btt: cache sector_size in arena_info
    - libnvdimm: fix potential deadlock while clearing errors
    - libnvdimm, btt: rework error clearing

  * [Feature] Crystal Ridge - have 4k DAX faults use a common zero page
    (LP: #1704439)
    - mm: add vm_insert_mixed_mkwrite()
    - dax: relocate some dax functions
    - dax: use common 4k zero page for dax mmap reads
    - dax: remove DAX code from page_cache_tree_insert()
    - dax: move all DAX radix tree defs to fs/dax.c

  * [bug] 17.10: CDP test fail on platform of Purley-2S/4S/Neoncity,BDW-
    de/ep/ex, (LP: #1716843)
    - SAUCE: (no-up) x86/intel_rdt: Fix cdp info directory files issue

  * [featue] GPIO support for Denverton (LP: #1591829)
    - pinctrl: intel: Add Intel Denverton pin controller support

  * ETPS/2 Elantech Touchpad inconsistently detected (Gigabyte P57W laptop)
    (LP: #1594214)
    - Input: i8042 - add Gigabyte P57 to the keyboard reset table

  * autopkgtest profile fails to build on armhf (LP: #1717920)
    - [Packaging] autopkgtest -- disable d-i when dropping flavours

  * Artful update to v4.13.3 stable release (LP: #1718412)
    - Revert "net: use lib/percpu_counter API for fragmentation mem accounting"
    - Revert "net: fix percpu memory leaks"
    - gianfar: Fix Tx flow control deactivation
    - vhost_net: correctly check tx avail during rx busy polling
    - ip6_gre: update mtu properly in ip6gre_err
    - udp: drop head states only when all skb references are gone
    - ipv6: fix memory leak with multiple tables during netns destruction
    - ipv6: fix typo in fib6_net_exit()
    - sctp: fix missing wake ups in some situations
    - tcp: fix a request socket leak
    - ip_tunnel: fix setting ttl and tos value in collect_md mode
    - f2fs: let fill_super handle roll-forward errors
    - f2fs: check hot_data for roll-forward recovery
    - x86/fsgsbase/64: Fully initialize FS and GS state in start_thread_common
    - x86/fsgsbase/64: Report FSBASE and GSBA...

Read more...

Changed in linux (Ubuntu):
status: Fix Committed → Fix Released
Changed in intel:
status: New → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.