No PMU support for ACPI-based arm64 systems

Bug #1689661 reported by dann frazier on 2017-05-09
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
High
dann frazier
Zesty
High
dann frazier

Bug Description

[Impact]
ACPI-based arm64 systems, such as the Qualcomm QDF2400 and the Hisilicon D05, do not currently have PMU support. This means that tools like perf are unable to use the built-in hardware event counters.

[Test Case]
On an ACPI-based arm64 system, run the following commands:

sudo perf list pmu
sudo perf list hw

Without ACPI PMU support, these will be empty.

[Regression Risk]
The changes are restricted to ARM-specific code, other than an added enum value in linux/cpuhotplug.h. This reduces the non-negligible regression risk to the ARM architecture.

For ACPI-based ARM platforms, regression risk is limited because they did not have PMU support previously. However, if this code is buggy, previously working commands that just made use of non-PMU events (e.g. perf top) might start causing problems. To mitigate this, I did a perf_fuzzer run on a QDF2400 (ACPI-based) and confirmed it survived (Note: this included a workaround for LP: #1689855, which will need to be fixed separately once upstreamed).

For non-ACPI-based (i.e. devicetree) ARM platforms, there is a chance that the changes have broken PMU support. To mitigate that, I also did a perf_fuzzer run on a Cavium ThunderX system booted in DT mode, which it survived.

Finally, the non-perf related changes in this series are restricted to a change in arch/arm64/kernel/smp.c that saves off pointers to each CPU's MADT GICC tables. The code is populating a static array with existing data - so pretty low risk (no pointer dereferences, etc).

CVE References

dann frazier (dannf) on 2017-05-09
Changed in linux (Ubuntu):
assignee: nobody → dann frazier (dannf)
Changed in linux (Ubuntu Zesty):
assignee: nobody → dann frazier (dannf)
Changed in linux (Ubuntu):
status: New → In Progress
Changed in linux (Ubuntu Zesty):
status: New → In Progress
Changed in linux (Ubuntu):
importance: Undecided → High
Changed in linux (Ubuntu Zesty):
importance: Undecided → High
dann frazier (dannf) on 2017-05-10
description: updated
Seth Forshee (sforshee) on 2017-05-16
Changed in linux (Ubuntu):
status: In Progress → Fix Committed
Changed in linux (Ubuntu Zesty):
status: In Progress → Fix Committed

This bug is awaiting verification that the kernel in -proposed solves the problem. Please test the kernel and update this bug with the results. If the problem is solved, change the tag 'verification-needed-zesty' to 'verification-done-zesty'. If the problem still exists, change the tag 'verification-needed-zesty' to 'verification-failed-zesty'.

If verification is not done by 5 working days from today, this fix will be dropped from the source code, and this bug will be closed.

See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you!

tags: added: verification-needed-zesty
dann frazier (dannf) wrote :

ubuntu@ubuntu:~$ cat /proc/version
Linux version 4.10.0-22-generic (buildd@bos01-arm64-038) (gcc version 6.3.0 20170406 (Ubuntu/Linaro 6.3.0-12ubuntu2) ) #24-Ubuntu SMP Mon May 22 17:43:39 UTC 2017
ubuntu@ubuntu:~$ sudo perf list pmu | wc -l
24
ubuntu@ubuntu:~$ sudo perf list hw | wc -l
6

tags: added: verification-done-zesty
removed: verification-needed-zesty
Launchpad Janitor (janitor) wrote :
Download full text (16.0 KiB)

This bug was fixed in the package linux - 4.10.0-22.24

---------------
linux (4.10.0-22.24) zesty; urgency=low

  * linux: 4.10.0-22.24 -proposed tracker (LP: #1691146)

  * Fix NVLINK2 TCE route (LP: #1690155)
    - powerpc/powernv: Fix TCE kill on NVLink2

  * CVE-2017-0605
    - tracing: Use strlcpy() instead of strcpy() in __trace_find_cmdline()

  * perf: qcom: Add L3 cache PMU driver (LP: #1689856)
    - [Config] CONFIG_QCOM_L3_PMU=y
    - perf: qcom: Add L3 cache PMU driver

  * No PMU support for ACPI-based arm64 systems (LP: #1689661)
    - drivers/perf: arm_pmu: rework per-cpu allocation
    - drivers/perf: arm_pmu: manage interrupts per-cpu
    - drivers/perf: arm_pmu: split irq request from enable
    - drivers/perf: arm_pmu: remove pointless PMU disabling
    - drivers/perf: arm_pmu: define armpmu_init_fn
    - drivers/perf: arm_pmu: fold init into alloc
    - drivers/perf: arm_pmu: factor out pmu registration
    - drivers/perf: arm_pmu: simplify cpu_pmu_request_irqs()
    - drivers/perf: arm_pmu: handle no platform_device
    - drivers/perf: arm_pmu: rename irq request/free functions
    - drivers/perf: arm_pmu: split cpu-local irq request/free
    - drivers/perf: arm_pmu: move irq request/free into probe
    - drivers/perf: arm_pmu: split out platform device probe logic
    - arm64: add function to get a cpu's MADT GICC table
    - [Config] CONFIG_ARM_PMU_ACPI=y
    - drivers/perf: arm_pmu: add ACPI framework
    - arm64: pmuv3: handle !PMUv3 when probing
    - arm64: pmuv3: use arm_pmu ACPI framework

  * [SRU][Zesty]QDF2400 kernel oops on ipmitool fru write 0 fru.bin
    (LP: #1689886)
    - ipmi: Fix kernel panic at ipmi_ssif_thread()

  * tty: pl011: fix earlycon work-around for QDF2400 erratum 44 (LP: #1689818)
    - tty: pl011: fix earlycon work-around for QDF2400 erratum 44
    - tty: pl011: use "qdf2400_e44" as the earlycon name for QDF2400 E44

  * kernel-wedge fails in artful due to leftover squashfs-modules d-i files
    (LP: #1688259)
    - Remove squashfs-modules files from d-i
    - [Config] as squashfs-modules is builtin kernel-image must Provides: it

  * arm64/ACPI support for SBSA watchdog (LP: #1688114)
    - clocksource: arm_arch_timer: clean up printk usage
    - clocksource: arm_arch_timer: rename type macros
    - clocksource: arm_arch_timer: rename the PPI enum
    - clocksource: arm_arch_timer: move enums and defines to header file
    - clocksource: arm_arch_timer: add a new enum for spi type
    - clocksource: arm_arch_timer: rework PPI selection
    - clocksource: arm_arch_timer: split dt-only rate handling
    - clocksource: arm_arch_timer: refactor arch_timer_needs_probing
    - clocksource: arm_arch_timer: move arch_timer_needs_of_probing into DT init
      call
    - clocksource: arm_arch_timer: add structs to describe MMIO timer
    - clocksource: arm_arch_timer: split MMIO timer probing.
    - [Config] CONFIG_ACPI_GTDT=y
    - acpi/arm64: Add GTDT table parse driver
    - clocksource: arm_arch_timer: simplify ACPI support code.
    - acpi/arm64: Add memory-mapped timer support in GTDT driver
    - clocksource: arm_arch_timer: add GTDT support for memory-mapped timer
    - acpi/arm64: Add SBS...

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

Other bug subscribers