[Feature] Northpeak Enabling

Bug #1704698 reported by Alice Liu
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
intel
Fix Released
Undecided
Unassigned
linux (Ubuntu)
Fix Released
Medium
Seth Forshee
Artful
Won't Fix
Medium
Seth Forshee

Bug Description

Description

Description: The NPK(North Peak) solution is composed of four major components.Hardware to connect the debug target to the host. DAL (DFX Abstraction Layer) to manage the target to host connection. NPKSDK connects to DAL and processes trace data using a decoder network.
GUI and command line tools provide access to NPKSDK to configure and start / stop tracing.
In CNL we have also Core side tracing that is new enabling.
Stakeholders : To be filled out by Engineering
Status : Base Enabling

HW: Cannon Lake

Target Kernel: v4.14
Target Release: 18.04

Alice Liu (dxliu)
information type: Proprietary → Private
Alice Liu (dxliu)
description: updated
description: updated
Revision history for this message
quanxian (quanxian-wang) wrote :

84331e1390 efb3669e14

Revision history for this message
Alice Liu (dxliu) wrote :

Add an commit for this feature which was merged in kernel v4.14-rc1.
a0e7df335a

Changed in linux (Ubuntu Artful):
status: New → Triaged
importance: Undecided → Medium
assignee: nobody → Seth Forshee (sforshee)
Revision history for this message
Seth Forshee (sforshee) wrote :

84331e1390 efb3669e14 had already come to artful via upstream stable. a0e7df335a has been applied with some minor backporting.

information type: Private → Public
Changed in linux (Ubuntu Artful):
status: Triaged → Fix Committed
Revision history for this message
Seth Forshee (sforshee) wrote :

This patch seems to be causing build errors.

/tmp/kernel-sforshee-e924fd6-hdwi/build/drivers/hwtracing/intel_th/pci.c: In function 'intel_th_pci_activate':
/tmp/kernel-sforshee-e924fd6-hdwi/build/drivers/hwtracing/intel_th/pci.c:39:7: error: implicit declaration of function 'INTEL_TH_CAP'; did you mean '__INTEL_TH_H__'? [-Werror=implicit-function-declaration]
  if (!INTEL_TH_CAP(th, tscu_enable))
       ^~~~~~~~~~~~
       __INTEL_TH_H__
  CC fs/debugfs/inode.o
  LD [M] sound/pci/echoaudio/snd-echo3g.o
  LD [M] net/nfc/nci/nci_uart.o
/tmp/kernel-sforshee-e924fd6-hdwi/build/drivers/hwtracing/intel_th/pci.c:39:24: error: 'tscu_enable' undeclared (first use in this function); did you mean 'pat_enabled'?
  if (!INTEL_TH_CAP(th, tscu_enable))
                        ^~~~~~~~~~~
                        pat_enabled
/tmp/kernel-sforshee-e924fd6-hdwi/build/drivers/hwtracing/intel_th/pci.c:39:24: note: each undeclared identifier is reported only once for each function it appears in
  LD [M] fs/coda/coda.o
  CC [M] fs/afs/vnode.o
  CC [M] fs/ceph/snap.o
  CC arch/x86/kernel/apic/htirq.o
  CC [M] net/irda/irnet/irnet_ppp.o
/tmp/kernel-sforshee-e924fd6-hdwi/build/drivers/hwtracing/intel_th/pci.c: In function 'intel_th_pci_deactivate':
/tmp/kernel-sforshee-e924fd6-hdwi/build/drivers/hwtracing/intel_th/pci.c:60:24: error: 'tscu_enable' undeclared (first use in this function); did you mean 'pat_enabled'?
  if (!INTEL_TH_CAP(th, tscu_enable))
                        ^~~~~~~~~~~
                        pat_enabled
/tmp/kernel-sforshee-e924fd6-hdwi/build/drivers/hwtracing/intel_th/pci.c: In function 'intel_th_pci_probe':
/tmp/kernel-sforshee-e924fd6-hdwi/build/drivers/hwtracing/intel_th/pci.c:92:4: error: 'struct intel_th' has no member named 'activate'
  th->activate = intel_th_pci_activate;
    ^~
/tmp/kernel-sforshee-e924fd6-hdwi/build/drivers/hwtracing/intel_th/pci.c:93:4: error: 'struct intel_th' has no member named 'deactivate'
  th->deactivate = intel_th_pci_deactivate;
    ^~
/tmp/kernel-sforshee-e924fd6-hdwi/build/drivers/hwtracing/intel_th/pci.c: At top level:
/tmp/kernel-sforshee-e924fd6-hdwi/build/drivers/hwtracing/intel_th/pci.c:105:21: error: variable 'intel_th_2x' has initializer but incomplete type
 static const struct intel_th_drvdata intel_th_2x = {
                     ^~~~~~~~~~~~~~~~
/tmp/kernel-sforshee-e924fd6-hdwi/build/drivers/hwtracing/intel_th/pci.c:106:3: error: 'const struct intel_th_drvdata' has no member named 'tscu_enable'
  .tscu_enable = 1,
   ^~~~~~~~~~~
/tmp/kernel-sforshee-e924fd6-hdwi/build/drivers/hwtracing/intel_th/pci.c:106:17: warning: excess elements in struct initializer
  .tscu_enable = 1,
                 ^
/tmp/kernel-sforshee-e924fd6-hdwi/build/drivers/hwtracing/intel_th/pci.c:106:17: note: (near initialization for 'intel_th_2x')
  AR sound/soc/ux500/built-in.o
/tmp/kernel-sforshee-e924fd6-hdwi/build/drivers/hwtracing/intel_th/pci.c:105:38: error: storage size of 'intel_th_2x' isn't known
 static const struct intel_th_drvdata intel_th_2x = {
                                      ^~~~~~~~~~

Revision history for this message
Seth Forshee (sforshee) wrote :

So it looks like we'd need at least 3321371b5d648479058fa6f9441168abbc1467c9 "intel_th: pci: Use drvdata for quirks" which in turn has conflicts in drivers/hwtracing/intel_th/core.c. So I'm dropping this patch for now, we'll either need guidance on exactly which commits should be backported to 4.13 or a backport to resolve the conflicts.

Changed in linux (Ubuntu Artful):
status: Fix Committed → Incomplete
Revision history for this message
quanxian (quanxian-wang) wrote :

got that. Intel is working on that. Thx

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

@canonical, also backport 3321371b5d and 29e15e83a9: they lay the pre-requisite groundwork.

Revision history for this message
Seth Forshee (sforshee) wrote :

Those commits still don't apply, there was substantial reworking of the driver before those commits which make backporting non-trivial.

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

suggest to move it into 18.04. Any comment, let us know.

quanxian (quanxian-wang)
description: updated
Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

I assume this is now resolved for our 4.15 kernel in 18.04? Intel, please confirm.

Changed in intel:
status: New → Incomplete
Revision history for this message
quanxian (quanxian-wang) wrote :

yes. code is now in 4.15.

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

I think this should be done.

Revision history for this message
Andy Whitcroft (apw) wrote : Closing unsupported series nomination.

This bug was nominated against a series that is no longer supported, ie artful. The bug task representing the artful nomination is being closed as Won't Fix.

This change has been made by an automated script, maintained by the Ubuntu Kernel Team.

Changed in linux (Ubuntu Artful):
status: Incomplete → Won't Fix
Changed in intel:
status: Incomplete → Won't Fix
status: Won't Fix → Fix Released
Changed in linux (Ubuntu):
status: Incomplete → 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.