MR: https://gitlab.com/redhat/centos-stream/src/kernel/centos-stream-9/-/merge_requests/8697
s390/cpum_cf: Handle CPU hotplug via prepare/dead callbacks
JIRA: https://redhat.atlassian.net/browse/RHEL-248152
commit 337bd95507a16063687cfc286ea90de5cca48c37
Author: Thomas Richter tmricht@linux.ibm.com
Date: Tue Aug 11 15:39:01 2026 +0200
```
s390/cpum_cf: Handle CPU hotplug via prepare/dead callbacks
The command 'perf stat -e cycles -- <command>' crashes the kernel
when CPUs are hotplug added during that run.
Root cause is the allocation of struct cpu_cf_events at first
event initialization. The allocation is dynamic and the first
event that has task context creates such a structure for
each online CPU. This is not sufficient. CPUs may be offline
during event creation and can be set online during the
perf run time. For example commands
# echo 0 > /sys/devices/system/cpu/cpu1/online
# perf stat -e cycles -i -- stress-ng -t10s --matrix X
# sleep 1
# echo 1 > /sys/devices/system/cpu/cpu1/online
create an event for CPUs 0,2-X. Since the events are created with
task-context, the scheduler will eventually schedule the program
on CPU1. This CPU has not created and initialized any per
CPU event infrastructure as that CPU was not online at the time
of the perf invocation. Thus when the scheduler runs stress-ng
on CPU1, the function cpumf_pmu_add() refers to a NULL pointer:
struct cpu_cf_events *cpuhw = this_cpu_cfhw();
This function call is invoked after the task stress-ng has been
made runnable on CPU1. And this_cpu_cfhw() returns NULL.
The result is a panic:
Unable to handle kernel pointer dereference in virtual kernel address space
Failing address: 0000000000000000 TEID: 0000000000000483
....
Krnl PSW : 0404d00180000000 000003ef8291fd0c (cpumf_pmu_add+0x3c/0x80)
....
Call Trace:
[<000003ef8291fd0c>] cpumf_pmu_add+0x3c/0x80
[<000003ef82bb5e3e>] event_sched_in+0xae/0x190
[<000003ef82bb60d6>] merge_sched_in+0x1b6/0x390
[<000003ef82bb65b8>] visit_groups_merge.constprop.0.isra.0+0x308/0x5b0
[<000003ef82bb689a>] pmu_groups_sched_in+0x3a/0x50
[<000003ef82bb6a30>] ctx_sched_in+0x180/0x260
[<000003ef82bb780c>] perf_event_context_sched_in+0x11c/0x2d0
[<000003ef82bb79ee>] __perf_event_task_sched_in+0x2e/0xc0
[<000003ef82994834>] finish_task_switch.isra.0+0x1a4/0x250
....
Last Breaking-Event-Address:
[<000003ef8291f1d8>] this_cpu_cfhw+0x38/0x40
The issue arises only in per-task context when the CPUMF facility is
used and the scheduler picks a random CPU for such a process to run on.
The scheduler enables the CPUMF infrastructure via PMU callback
functions pmu::add() and pmu::del().
Introduce a CPU hotplug prepare/dead callback pair which creates and
removes the per CPU counter data while the CPU is offline. Count the
users which track every CPU (cpu == -1), that is perf_event_open()
events with task context and /dev/hwctr device sessions, in the new
counter cpu_cf_root::tskcnt, protected by pmc_reserve_mutex.
This ensures the infrastructure is available when
new CPU is selected to run the per-task context process.
In cpum_cf_free_root() and cpum_cf_free_cpu() ensure the reference
pointer to data structures is set to NULL before the data is freed
to prevent interrupt handlers to access stale data.
[gor@linux.ibm.com: change commit message]
Fixes: 9b9cf3c77e7e ("s390/cpum_cf: rework PER_CPU_DEFINE of struct cpu_cf_events")
Cc: stable@vger.kernel.org # v6.5+
Suggested-by: Heiko Carstens <hca@linux.ibm.com>
Suggested-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Assisted-by: Claude:claude-sonnet-5
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
```
Signed-off-by: Jan Polensky <jpolensk@redhat.com>
Approved-by: Steve Best <sbest@redhat.com>
Approved-by: Tony Camuso <tcamuso@redhat.com>
Approved-by: CKI KWF Bot <cki-ci-bot+kwf-gitlab-com@redhat.com>
Merged-by: CKI GitLab Kmaint Pipeline Bot <26919896-cki-kmaint-pipeline-bot@users.noreply.gitlab.com>
MR: https://gitlab.com/redhat/centos-stream/src/kernel/centos-stream-9/-/merge_requests/8414
# Merge Request Required Information
## Summary of Changes
Red Hat is seeing multiple reports of Windows memory corruptions
(and consequent BSODs) with hv-tlbflush=on, on AMD processors only.
The crashes, while extremely rare, happen even with a stock configuration,
but with Driver Verifier enabled they can be detected after approximately
200 VM hours.
So, for lack of better ideas, this patch forces a full ASID bump in
svm_flush_tlb_gva(). To avoid paying the price on Intel and also to
avoid unnecessary loops on AMD, the flush_tlb_gva op now returns whether
it did a full flush or not; kvm_hv_vcpu_flush_tlb() takes note and exits
its loops immediately. While there is an obvious performance impact,
about half of the benefit from Hyper-V tlbflush is preserved (10% vs. 20%
on the SQL Server workload).
## Approved Development Ticket(s)
All submissions to CentOS Stream must reference a ticket in [Red Hat Jira](https://issues.redhat.com/).
```
JIRA: https://redhat.atlassian.net/browse/RHEL-214433
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
```
Approved-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Approved-by: awallace9 <awallace@redhat.com>
Approved-by: Maxim Levitsky <mlevitsk@redhat.com>
Merged-by: CKI GitLab Kmaint Pipeline Bot <26919896-cki-kmaint-pipeline-bot@users.noreply.gitlab.com>
JIRA: https://redhat.atlassian.net/browse/RHEL-248152
commit 337bd95507a16063687cfc286ea90de5cca48c37
Author: Thomas Richter <tmricht@linux.ibm.com>
Date: Tue Aug 11 15:39:01 2026 +0200
s390/cpum_cf: Handle CPU hotplug via prepare/dead callbacks
The command 'perf stat -e cycles -- <command>' crashes the kernel
when CPUs are hotplug added during that run.
Root cause is the allocation of struct cpu_cf_events at first
event initialization. The allocation is dynamic and the first
event that has task context creates such a structure for
each online CPU. This is not sufficient. CPUs may be offline
during event creation and can be set online during the
perf run time. For example commands
# echo 0 > /sys/devices/system/cpu/cpu1/online
# perf stat -e cycles -i -- stress-ng -t10s --matrix X
# sleep 1
# echo 1 > /sys/devices/system/cpu/cpu1/online
create an event for CPUs 0,2-X. Since the events are created with
task-context, the scheduler will eventually schedule the program
on CPU1. This CPU has not created and initialized any per
CPU event infrastructure as that CPU was not online at the time
of the perf invocation. Thus when the scheduler runs stress-ng
on CPU1, the function cpumf_pmu_add() refers to a NULL pointer:
struct cpu_cf_events *cpuhw = this_cpu_cfhw();
This function call is invoked after the task stress-ng has been
made runnable on CPU1. And this_cpu_cfhw() returns NULL.
The result is a panic:
Unable to handle kernel pointer dereference in virtual kernel address space
Failing address: 0000000000000000 TEID: 0000000000000483
....
Krnl PSW : 0404d00180000000 000003ef8291fd0c (cpumf_pmu_add+0x3c/0x80)
....
Call Trace:
[<000003ef8291fd0c>] cpumf_pmu_add+0x3c/0x80
[<000003ef82bb5e3e>] event_sched_in+0xae/0x190
[<000003ef82bb60d6>] merge_sched_in+0x1b6/0x390
[<000003ef82bb65b8>] visit_groups_merge.constprop.0.isra.0+0x308/0x5b0
[<000003ef82bb689a>] pmu_groups_sched_in+0x3a/0x50
[<000003ef82bb6a30>] ctx_sched_in+0x180/0x260
[<000003ef82bb780c>] perf_event_context_sched_in+0x11c/0x2d0
[<000003ef82bb79ee>] __perf_event_task_sched_in+0x2e/0xc0
[<000003ef82994834>] finish_task_switch.isra.0+0x1a4/0x250
....
Last Breaking-Event-Address:
[<000003ef8291f1d8>] this_cpu_cfhw+0x38/0x40
The issue arises only in per-task context when the CPUMF facility is
used and the scheduler picks a random CPU for such a process to run on.
The scheduler enables the CPUMF infrastructure via PMU callback
functions pmu::add() and pmu::del().
Introduce a CPU hotplug prepare/dead callback pair which creates and
removes the per CPU counter data while the CPU is offline. Count the
users which track every CPU (cpu == -1), that is perf_event_open()
events with task context and /dev/hwctr device sessions, in the new
counter cpu_cf_root::tskcnt, protected by pmc_reserve_mutex.
This ensures the infrastructure is available when
new CPU is selected to run the per-task context process.
In cpum_cf_free_root() and cpum_cf_free_cpu() ensure the reference
pointer to data structures is set to NULL before the data is freed
to prevent interrupt handlers to access stale data.
[gor@linux.ibm.com: change commit message]
Fixes: 9b9cf3c77e7e ("s390/cpum_cf: rework PER_CPU_DEFINE of struct cpu_cf_events")
Cc: stable@vger.kernel.org # v6.5+
Suggested-by: Heiko Carstens <hca@linux.ibm.com>
Suggested-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Assisted-by: Claude:claude-sonnet-5
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com>
Acked-by: Heiko Carstens <hca@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
Signed-off-by: Jan Polensky <jpolensk@redhat.com>
JIRA: INTERNAL
Upstream Status: RHEL-9-only
This bumps the value of RHEL_MINOR to 10 so we can formally get RHEL-9.10
merge requests merged and start doing development builds.
Signed-off-by: Julio Faracco <jfaracco@redhat.com>
JIRA: https://redhat.atlassian.net/browse/RHEL-172511
CVE: CVE-2026-31786
commit 27fdbab4221b375de54bf91919798d88520c6e28
Author: Juergen Gross <jgross@suse.com>
Date: Fri Mar 27 14:13:38 2026 +0100
Buffer overflow in drivers/xen/sys-hypervisor.c
The build id returned by HYPERVISOR_xen_version(XENVER_build_id) is
neither NUL terminated nor a string.
The first causes a buffer overflow as sprintf in buildid_show will
read and copy till it finds a NUL.
00000000 f4 91 51 f4 dd 38 9e 9d 65 47 52 eb 10 71 db 50 |..Q..8..eGR..q.P|
00000010 b9 a8 01 42 6f 2e 32 |...Bo.2|
00000017
So use a memcpy instead of sprintf to have the correct value:
00000000 f4 91 51 f4 dd 00 9e 9d 65 47 52 eb 10 71 db 50 |..Q.....eGR..q.P|
00000010 b9 a8 01 42 |...B|
00000014
(the above have a hack to embed a zero inside and check it's
returned correctly).
This is XSA-485 / CVE-2026-31786
Fixes: 84b7625728 ("xen: add sysfs node for hypervisor build id")
Signed-off-by: Frediano Ziglio <frediano.ziglio@citrix.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
JIRA: https://redhat.atlassian.net/browse/RHEL-172489
CVE: CVE-2026-31787
commit 24daca4fc07f3ff8cd0e3f629cd982187f48436a
Author: Juergen Gross <jgross@suse.com>
Date: Fri Apr 10 09:20:04 2026 +0200
xen/privcmd: fix double free via VMA splitting
privcmd_vm_ops defines .close (privcmd_close), but neither .may_split
nor .open. When userspace does a partial munmap() on a privcmd mapping,
the kernel splits the VMA via __split_vma(). Since may_split is NULL,
the split is allowed. vm_area_dup() copies vm_private_data (a pages
array allocated in alloc_empty_pages()) into the new VMA without any
fixup, because there is no .open callback.
Both VMAs now point to the same pages array. When the unmapped portion
is closed, privcmd_close() calls:
- xen_unmap_domain_gfn_range()
- xen_free_unpopulated_pages()
- kvfree(pages)
The surviving VMA still holds the dangling pointer. When it is later
destroyed, the same sequence runs again, which leads to a double free.
Fix this issue by adding a .may_split callback denying the VMA split.
This is XSA-487 / CVE-2026-31787
Fixes: d71f513985 ("xen: privcmd: support autotranslated physmap guests.")
Reported-by: Atharva Vartak <atharva.a.vartak@gmail.com>
Suggested-by: Atharva Vartak <atharva.a.vartak@gmail.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
JIRA: https://redhat.atlassian.net/browse/RHEL-241103
Upstream status: v7.3
commit 62b57396c26a1ce54963709928ea0d01fa522eea
Author: Ma Ke <make_ruc2021@163.com>
Date: Wed Jun 24 09:49:09 2026 +0800
platform/x86: ishtp_eclite: Fix ACPI device reference leak in probe error path
ecl_ishtp_cl_probe() acquires a reference to an ACPI device via
acpi_find_eclite_device() but fails to release it in the error path
when acpi_opregion_init() fails. This results in a reference count
leak, preventing proper cleanup of the ACPI device.
Calling path: acpi_find_eclite_device() ->
acpi_dev_get_first_match_dev() -> acpi_dev_get_next_match_dev() ->
bus_find_device() -> get_device().
Found by code review.
Signed-off-by: Ma Ke <make_ruc2021@163.com>
Acked-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Cc: stable@vger.kernel.org
Fixes: 7b6bf51de974 ("platform/x86: Add Intel ishtp eclite driver")
Link: https://patch.msgid.link/20260624014910.1226446-1-make_ruc2021@163.com
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
(cherry picked from commit 62b57396c26a1ce54963709928ea0d01fa522eea)
Assisted-by: Patchpal
Signed-off-by: Dennis Chen <dechen@redhat.com>
JIRA: https://redhat.atlassian.net/browse/RHEL-241103
Upstream status: v7.2
commit 09b2ae290a241ce1f5f738fb65c35f449dcf663d
Author: Yousef Alhouseen <alhouseenyousef@gmail.com>
Date: Mon Jul 6 09:13:39 2026 -0400
platform/x86/intel/vsec: free ACPI discovery data on early errors
intel_vsec_add_dev() may attach an ACPI discovery table copy to the
intel_vsec_device before passing ownership to intel_vsec_add_aux(). The
normal auxiliary-device release path frees that copy, but the earliest
intel_vsec_add_aux() failures free only the outer structure directly.
Route those direct frees through a common helper so acpi_disc is
released consistently on the parent, xarray, and ID allocation failure
paths.
Fixes: 22fa2ebc11a1 ("platform/x86/intel/vsec: Plumb ACPI PMT discovery tables through vsec")
Signed-off-by: Yousef Alhouseen <alhouseenyousef@gmail.com>
Reviewed-by: David E. Box <david.e.box@linux.intel.com>
Link: https://patch.msgid.link/CAMuQ4bUtJtYNTguKoiXngROJw0QQQcrvW3=3_B0-hpMQOFqvCQ@mail.gmail.com
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Dennis Chen <dechen@redhat.com>
JIRA: https://redhat.atlassian.net/browse/RHEL-241103
Upstream status: v7.2
commit 8a3607ece59fcf84cf4129bdf683cb09122e6b51
Author: David E. Box <david.e.box@linux.intel.com>
Date: Fri Jun 12 14:01:55 2026 -0700
platform/x86/intel/pmc/ssram: Make PMT registration optional
The SSRAM telemetry driver extracts essential PMC device ID and power
management base address information that intel_pmc_core depends on for core
functionality. If PMT registration failure prevents this critical data from
being available, intel_pmc_core operation would break entirely. Therefore,
PMT registration failures must not block access to this data.
Change the behavior to log a warning when PMT registration fails but
continue with successful driver initialization, ensuring the primary
telemetry data remains accessible to dependent drivers.
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Link: https://patch.msgid.link/4f4c324977951f6082bf2218c8b911e1ae7e0a7b.1781294741.git.david.e.box@linux.intel.com
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Dennis Chen <dechen@redhat.com>
JIRA: https://redhat.atlassian.net/browse/RHEL-241103
Upstream status: v7.2
Conflicts:
- drivers/platform/x86/intel/pmc/ssram_telemetry.c: Remove quotes from
symbol namespace import because RHEL is missing cdd30ebb1b9f
("module: Convert symbol namespace to string literal")
commit 08b94937c78aaa144e3afff659c6112bffea6bf0
Author: David E. Box <david.e.box@linux.intel.com>
Date: Fri Jun 12 14:01:54 2026 -0700
platform/x86/intel/pmc/ssram: Add ACPI discovery scaffolding
Prepare the SSRAM telemetry driver for ACPI-based discovery by adding
support for reading telemetry regions from ACPI _DSD properties.
Add pmc_ssram_telemetry_acpi_init() to parse _DSD for telemetry discovery
tables and register them with the Intel VSEC framework. Extend ssram_type
with a p_index field to specify which PMC index each ACPI device owns
(unlike PCI which discovers all three PMCs from one device).
At this stage, no platform IDs are wired to use ACPI discovery - existing
devices continue using the PCI path. Follow-on patches will add platform
support.
Assisted-by: Claude:claude-sonnet-4-5
Signed-off-by: Xi Pardee <xi.pardee@linux.intel.com>
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Link: https://patch.msgid.link/54850d175993ee38aef99707f954492d24684dcc.1781294741.git.david.e.box@linux.intel.com
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Dennis Chen <dechen@redhat.com>
JIRA: https://redhat.atlassian.net/browse/RHEL-241103
Upstream status: v7.2
commit e3c9200a9c0e05532c495956883ba11fe23a3df0
Author: David E. Box <david.e.box@linux.intel.com>
Date: Fri Jun 12 14:01:53 2026 -0700
platform/x86/intel/pmc/ssram: Switch to static array with per-index probe state
Replace devm-allocated pmc_ssram_telems pointer with a fixed-size static
array and introduce per-index probe state tracking.
This prepares the driver for later per-device probe handling where tying
the PMC tracking storage to one probed PCI device is no longer suitable.
The previous single global device_probed flag cannot describe the state of
individual PMC indices when multiple devices can be probed independently.
Replace it with per-index state (UNPROBED, PROBING, PRESENT, ABSENT) and a
staging cache that publishes discovered values only after probe completes.
This avoids races between probe/unbind and concurrent readers.
Use marked state accesses with release/acquire ordering to prevent compiler
and CPU reordering issues across concurrent probe/unbind cycles.
This patch was substantially rewritten in later revisions. Originally
developed from earlier work by Xi Pardee.
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Assisted-by: Claude:claude-sonnet-4-5
Link: https://patch.msgid.link/a2cccf532bec04fcc370819890d936212bc1b14c.1781294741.git.david.e.box@linux.intel.com
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Dennis Chen <dechen@redhat.com>
JIRA: https://redhat.atlassian.net/browse/RHEL-241103
Upstream status: v7.2
Conflicts:
- drivers/platform/x86/intel/pmc/pwrm_telemetry.c: Remove quotes from
symbol namespace import/exports because RHEL is missing cdd30ebb1b9f
("module: Convert symbol namespace to string literal")
commit adc5d98d9ff8b5d37e89e11a2ac5512595541329
Author: David E. Box <david.e.box@linux.intel.com>
Date: Fri Jun 12 14:01:49 2026 -0700
platform/x86/intel/pmc: Add ACPI PWRM telemetry driver for Nova Lake S
Add an ACPI-based PMC PWRM telemetry driver for Nova Lake S. The driver
locates PMT discovery data in _DSD under the Intel VSEC UUID, parses it,
and registers telemetry regions with the PMT/VSEC framework so PMC
telemetry is exposed via existing PMT interfaces.
Export pmc_parse_telem_dsd() and pmc_find_telem_guid() to support ACPI
discovery in other PMC drivers (e.g., ssram_telemetry) without duplicating
ACPI parsing logic. Also export acpi_disc_t typedef from core.h for callers
to properly declare discovery table arrays.
Selected by INTEL_PMC_CORE. Existing PCI functionality is preserved.
Assisted-by: GitHub-Copilot:claude-opus-4.7
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Link: https://patch.msgid.link/09b8211d8a5a79fa019ee2397137a6a43cf19430.1781294741.git.david.e.box@linux.intel.com
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Dennis Chen <dechen@redhat.com>
JIRA: https://redhat.atlassian.net/browse/RHEL-241103
Upstream status: v7.2
commit 13793c7f9e9ff30042a54bb680662b5cfa0f58fa
Author: David E. Box <david.e.box@linux.intel.com>
Date: Fri Jun 12 14:01:47 2026 -0700
platform/x86/intel/pmt: Unify header fetch and add ACPI source
Allow the PMT class to read discovery headers from either PCI MMIO or
ACPI-provided entries, depending on the discovery source. The new
source-aware fetch helper caches the canonical discovery header for both
paths, capping PCI MMIO reads to the mapped resource size, while keeping
the mapped PCI discovery table available for users such as crashlog.
Split intel_pmt_populate_entry() into source-specific resolvers:
- pmt_resolve_access_pci(): handles both ACCESS_LOCAL and ACCESS_BARID
for PCI-backed devices and sets entry->pcidev. Same existing
functionality.
- pmt_resolve_access_acpi(): handles only ACCESS_BARID for ACPI-backed
devices, rejecting ACCESS_LOCAL which has no valid semantics without
a physical discovery resource.
This maintains existing PCI behavior and makes no functional changes
for PCI devices.
Assisted-by: GitHub-Copilot:claude-opus-4.7
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Link: https://patch.msgid.link/4b33b04ffaf0943b67d330f48b5d1dfcb6d1be5d.1781294741.git.david.e.box@linux.intel.com
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Dennis Chen <dechen@redhat.com>
JIRA: https://redhat.atlassian.net/browse/RHEL-241103
Upstream status: v7.2
commit 4dfc7dca6e934ca414d8d3c70a84e79d13d9e750
Author: David E. Box <david.e.box@linux.intel.com>
Date: Fri Jun 12 14:01:46 2026 -0700
platform/x86/intel/pmt: Cache the telemetry discovery header
pmt_telem_header_decode() only needs the discovery header dwords, but it
currently decodes them by reading directly from entry->disc_table.
Cache the discovery header in intel_pmt_entry when the device is created
and have telemetry decode use the cached values instead of performing MMIO
reads at decode time.
The DVSEC discovery resource for a namespace is sized by its per-entry
entry_size (in dwords), which can be less than the 4-dword cache (e.g.
telemetry uses entry_size = 3, i.e. 12 bytes). Cap the memcpy_fromio()
to resource_size(disc_res) so the new cache does not read past the
mapped region. Any unread dwords stay zero from the zero-initialized
allocation of the containing struct.
This keeps the telemetry header decode path independent of how the
discovery data is backed and avoids baking a direct MMIO assumption into
the feature-specific decode logic.
Assisted-by: GitHub-Copilot:claude-opus-4.7
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Link: https://patch.msgid.link/f805e2ada52dc0661761cda7f692e76e6ea2d257.1781294741.git.david.e.box@linux.intel.com
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Dennis Chen <dechen@redhat.com>
CKI KWF Bot
Jarod Wilson
Jan Polensky
Julio Faracco
Vitaly Kuznetsov
Dennis Chen