linux.git/arch/powerpc/perf, branch master

Merge tag 'bitmap-for-7.3' of https://github.com/norov/linux

2026-08-20T18:41:46+00:00

Pull bitmap updates from Yury Norov:
 "The usual set of fixes, cleanups and performance improvements together
  with a couple of new tests:

   - bitmap_find_next_zero_area_off() optimization (Sunyi)

   - bitmap_find_next_zero_area_off(): return size when no zero area is
     found (Yury)

   - bitmap vs IDA vs Maple Tree performance test (Yury)

   - get rid of cpumap_print_to_pagebuf() (Yury)

   - use nr_node_ids in __nodemask_pr_numnodes() (Li RongQing)

   - bitops: make the *_bit_le functions use unsigned long (Benjamin)

   - bitmap scatter & gather test fix (Christophe)

   - use __ASSEMBLER__ in bitmap header files (Thomas)"

* tag 'bitmap-for-7.3' of https://github.com/norov/linux: (25 commits)
  lib: test bitmap vs IDA vs Maple Tree performance for region allocations
  bitmap: Return size when no zero area is found
  media: s5p-mfc: Treat bitmap size as allocation failure
  crypto: ccp: Treat bitmap size as allocation failure
  powerpc/msi: Treat bitmap size as allocation failure
  ARM: dma-mapping: Treat bitmap size as allocation failure
  bitmap: drop bitmap_next_set_region()
  nodemask: reduce bitmap width to nr_node_ids in __nodemask_pr_numnodes()
  bitmap: Properly initialise destination bitmap for scatter & gather test
  lib/bitmap-str: get rid of cpumap_print_to_pagebuf()
  perf: Use sysfs_emit() for cpumask show callbacks
  PCI/sysfs: Use sysfs_emit() for cpumask show callbacks
  RDMA/hfi1: Use sysfs_emit() for cpumask show helper
  hwtracing: hisi_ptt: Use sysfs_emit() for cpumask show
  fpga: dfl-fme-perf: Use sysfs_emit() for cpumask show
  devfreq: Use sysfs_emit() for cpumask show callbacks
  cpu: Use sysfs_emit() for cpumask show callback
  x86/events: Use sysfs_emit() for cpumask show callbacks
  powerpc: Use sysfs_emit() for cpumask show callbacks
  arm: Use sysfs_emit() for cpumask show callbacks
  ...

Merge tag 'powerpc-7.3-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux

2026-08-18T19:08:13+00:00

Pull powerpc updates from Madhavan Srinivasan:

 - Enable Rust for ppc64le

 - ppc4xx gpio driver updates

 - Add power12 base enablement support

 - Validate arch_compat against host compatibility mode

 - Simplify bootx_scan_dt_build_struct() in powermac platform

 - Implement get_direction() in cpm2

 - Use cpu_relax() in ps3_create_spu()

 - Add NULL guard for cause_ipi in smp_muxed_ipi_message_pass

 - Fixes to handle pseries watchdogs in kdump path

 - Fix missing r2 clobber in PCREL inline assembly

 - Set GPIO chip parent on ppc44x

 - KVM: Introduce KVM_CAP_PPC_COMPAT_CAPS and wire up ioctl

 - KVM: Use generic xfer to guest work function

 - Enable to run posix cpu timers in task context

 - Misc fixes and cleanups

Thanks to Aditya Gupta, Alice Ryhl, Amit Machhiwal, Andrew Morton,
Anushree Mathur, Athira Rajeev, Bartosz Golaszewski, Cédric Le Goater,
Christian König, Christophe Leroy (CS GROUP), Gary Guo, Gaurav Batra,
Gautam Menghani, Gou Hao, Hari Bathini, Harsh Prateek Bora, jiazhenyuan,
Jinjie Ruan, Link Mauve, Linus Walleij, Mahesh Kumar G Mahesh
Salgaonkar, Michael Walle, Michal Suchánek, Mukesh Kumar Chaurasiya
(IBM), Nicholas Piggin, Nikhil Kumar Singh, Praveen K Pandey, Ritesh
Harjani (IBM), Rosen Penev, Saket Kumar Bhaskar, Shrikanth Hegde,
Sourabh Jain, Thorsten Blum, Vaibhav Jain, Venkat Rao Bagalkote, Vishal
Chourasia, Wentao Guan, and Yanfei Xu.

* tag 'powerpc-7.3-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux: (62 commits)
  powerpc/pseries/iommu: switch to Default DMA window during kdump
  powerpc/configs: enable CONFIG_RAS to fix EDAC support
  KVM: PPC: Document KVM_PPC_GET_COMPAT_CAPS ioctl
  KVM: PPC: Book3S HV: Add support for compat CPU capabilities for KVM on PowerNV
  KVM: PPC: Book3S HV: Implement compat CPU capability retrieval for KVM on PowerVM
  KVM: PPC: Introduce KVM_CAP_PPC_COMPAT_CAPS and wire up ioctl
  gpio: ppc44x: use dev_name() for chip label
  gpio: ppc44x: fix undefined behavior in GPIO_MASK2 macro
  gpio: ppc44x: drop PPC-specific IO helpers
  gpio: ppc44x: Convert GPIO to generic MMIO
  gpio: ppc44x: Use platform resource helper for GPIO MMIO
  gpio: ppc44x: Use module platform driver helper for GPIO
  gpio: ppc44x: update all 4xx to 44x
  gpio: move ppc4xx gpio driver from arch/powerpc to drivers/gpio
  KVM: PPC: Use min() in kvm_vm_ioctl_check_extension()
  KVM: PPC: booke: Use min() in watchdog_next_timeout()
  powerpc/perf: Add power12 Base Performance Monitoring support
  powerpc: Add Power12 architected mode
  powerpc: Add Power12 raw mode
  powerpc/pseries: Limit PVR list to 16 entries for CAS negotiation
  ...

powerpc/perf: Add power12 Base Performance Monitoring support

2026-08-06T05:30:55+00:00

Base enablement patch to register performance monitoring hardware
support for power12. Patch introduce the raw event encoding format,
defines the supported list of events, config fields for the event
attributes and their corresponding bit values which are exported via
sysfs.

Signed-off-by: Athira Rajeev 
Signed-off-by: Mahesh Salgaonkar 
Tested-by: Praveen K Pandey 
Reviewed-by: Nikhil Kumar Singh 
Signed-off-by: Madhavan Srinivasan 
Link: https://patch.msgid.link/20260806051120.3703698-5-mahesh@linux.ibm.com

powerpc/perf: Use strstarts() to simplify is_thread_imc_pmu()

2026-07-28T04:53:31+00:00

Replace the open-coded implementation with strstarts() to simplify
is_thread_imc_pmu().

Signed-off-by: Thorsten Blum 
Reviewed-by: Athira Rajeev 
Signed-off-by: Madhavan Srinivasan 
Link: https://patch.msgid.link/20260704121353.201583-3-thorsten.blum@linux.dev

powerpc: Use sysfs_emit() for cpumask show callbacks

2026-07-22T19:38:37+00:00

These callbacks are sysfs show paths.

Use sysfs_emit() and cpumask_pr_args() to emit the masks.

This prepares for removing cpumap_print_to_pagebuf().

Signed-off-by: Yury Norov

powerpc/perf: fix preempt count underflow in fsl_emb_pmu_del

2026-06-15T10:55:21+00:00

fsl_emb_pmu_del() unconditionally calls put_cpu_var(cpu_hw_events) at
the 'out:' label, but only calls the matching get_cpu_var() after the
'i < 0' early-return check. When event->hw.idx is negative the
function jumps to 'out:' without having taken get_cpu_var(), and the
trailing put_cpu_var() then issues an unmatched preempt_enable(),
underflowing preempt_count.

On a CONFIG_PREEMPT=y kernel preempt_count would underflow and
eventually present as a 'scheduling while atomic' BUG.

Move put_cpu_var() to pair with get_cpu_var() so the percpu access is
correctly bracketed and the 'out:' label only handles perf_pmu_enable.

Fixes: a11106544f33 ("powerpc/perf: e500 support")
Reviewed-by: Shrikanth Hegde 
Signed-off-by: Aboorva Devarajan 
Signed-off-by: Madhavan Srinivasan 
Link: https://patch.msgid.link/20260605082912.305100-2-aboorvad@linux.ibm.com

powerpc: use sysfs_emit{_at} in sysfs show functions

2026-06-02T05:54:32+00:00

Replace sprintf() with sysfs_emit() and sysfs_emit_at() in sysfs show
functions, which are preferred for formatting sysfs output because they
provide safer bounds checking.

While the current code only emits strings that fit easily within
PAGE_SIZE, use sysfs_emit() and sysfs_emit_at() to follow secure coding
best practices.

This is a mechanical cleanup with a few simple edge cases:

- In domains_show(), drop the redundant n < 0 check since neither
  sprintf() nor sysfs_emit() return negative values.

- In powercap_show() and psr_show(), also drop the dead ret < 0 checks.

- In ps3_fw_version_show(), normalize the output by adding a terminating
  newline as suggested by checkpatch.

- In vio's modalias_show(), replace the deprecated strcpy() [1] followed
  by strlen() with sysfs_emit().

Leave validate_show() and the variable-length hv-gpci helpers unchanged
since they already have explicit bounds handling, and converting those
would be more than a mechanical conversion.

[1] https://www.kernel.org/doc/html/latest/process/deprecated.html#strcpy

Signed-off-by: Thorsten Blum 
Reviewed-by: Christophe Leroy (CS GROUP) 
Signed-off-by: Madhavan Srinivasan 
Link: https://patch.msgid.link/20260524130002.793476-2-thorsten.blum@linux.dev

powerpc/perf: Use cpumask_intersects api for checking disable path

2026-06-02T05:54:32+00:00

First online CPU in the node disables the nest counters by
making an OPAL call. Any other CPU in that node, will bail out.

Instead of using a temporary mask to find out if any cpu in the
node is visited or not, it is better to use the cpumask_intersects
api to achieve the same.

Similarly a temporary cpumask is used to check if a core is already part
of core_imc_cpumask. Use the same cpumask_intersects api there.

Signed-off-by: Shrikanth Hegde 
Reviewed-by: Yury Norov 
Signed-off-by: Madhavan Srinivasan 
Link: https://patch.msgid.link/20260427044715.559137-4-sshegde@linux.ibm.com

powerpc: rename arch_irq_disabled_regs

2026-05-20T01:27:13+00:00

Rename arch_irq_disabled_regs() to regs_irqs_disabled() to align with the
naming used in the generic irqentry framework. This makes the function
available for use both in the PowerPC architecture code and in the
common entry/exit paths shared with other architectures.

This is a preparatory change for enabling the generic irqentry framework
on PowerPC.

Signed-off-by: Mukesh Kumar Chaurasiya 
Tested-by: Samir M 
Tested-by: David Gow 
Tested-by: Venkat Rao Bagalkote 
Reviewed-by: Shrikanth Hegde 
Reviewed-by: Jinjie Ruan 
Signed-off-by: Madhavan Srinivasan 
Link: https://patch.msgid.link/20260427122742.210074-2-mkchauras@gmail.com

powerpc/hv-gpci: fix preempt count leak in sysfs show paths

2026-05-12T06:21:17+00:00

Four sysfs show() callbacks in hv-gpci take get_cpu_var(hv_gpci_reqb)
(which calls preempt_disable()) but only call the matching put_cpu_var()
on the error path under the 'out:' label. Every successful read leaks
one preempt_disable():

  processor_bus_topology_show()
  processor_config_show()
  affinity_domain_via_virtual_processor_show()
  affinity_domain_via_domain_show()

(affinity_domain_via_partition_show() was already correct.)

On a CONFIG_PREEMPT=y kernel, repeated reads raise preempt_count and
eventually return to userspace with preemption still disabled. The
next user-mode page fault then hits faulthandler_disabled() == 1,
gets forced to SIGSEGV, and the resulting coredump trips
'BUG: scheduling while atomic' in call_usermodehelper_exec ->
wait_for_completion_state -> schedule:

  BUG: scheduling while atomic: //0x00000004
  ...
  __schedule_bug+0x6c/0x90
  __schedule+0x58c/0x13a0
  schedule+0x48/0x1a0
  schedule_timeout+0x104/0x170
  wait_for_completion_state+0x16c/0x330
  call_usermodehelper_exec+0x254/0x2d0
  vfs_coredump+0x1050/0x2590
  get_signal+0xb9c/0xc80
  do_notify_resume+0xf8/0x470

Add an out_success label that calls put_cpu_var() before returning
the byte count, mirroring affinity_domain_via_partition_show().

Fixes: 71f1c39647d8 ("powerpc/hv_gpci: Add sysfs file inside hv_gpci device to show processor bus topology information")
Fixes: 1a160c2a13c6 ("powerpc/hv_gpci: Add sysfs file inside hv_gpci device to show processor config information")
Fixes: 71a7ccb478fc ("powerpc/hv_gpci: Add sysfs file inside hv_gpci device to show affinity domain via virtual processor information")
Fixes: a69a57cac1ec ("powerpc/hv_gpci: Add sysfs file inside hv_gpci device to show affinity domain via domain information")
Signed-off-by: Aboorva Devarajan 
Reviewed-by: Ritesh Harjani (IBM) 
Signed-off-by: Madhavan Srinivasan 
Link: https://patch.msgid.link/20260508041256.3447113-1-aboorvad@linux.ibm.com