* [Bug 221505] New: REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1
@ 2026-05-11 17:45 bugzilla-daemon
2026-05-11 20:15 ` [Bug 221505] " bugzilla-daemon
` (10 more replies)
0 siblings, 11 replies; 12+ messages in thread
From: bugzilla-daemon @ 2026-05-11 17:45 UTC (permalink / raw)
To: linux-pm
https://bugzilla.kernel.org/show_bug.cgi?id=221505
Bug ID: 221505
Summary: REGRESSION][amd_pstate] "EPP cannot be set under
performance policy" (EBUSY) when switching to
performance profile via PPD on 7.1
Product: Power Management
Version: 2.5
Hardware: AMD
OS: Linux
Status: NEW
Severity: normal
Priority: P3
Component: cpufreq
Assignee: linux-pm@vger.kernel.org
Reporter: sergeyluashkov@gmail.com
Regression: No
Created attachment 310089
--> https://bugzilla.kernel.org/attachment.cgi?id=310089&action=edit
dmesg
Hardware: ASUS TUF Gaming A14 (2026) FA401EA
CPU: AMD Ryzen AI Max+ 932 (Strix Halo platform)
Kernel: 7.1.0-mainline (including patches for Bugzilla #221310)
Power Daemon: power-profiles-daemon 0.30
Description:
On kernel 7.1, power-profiles-daemon fails to activate the "performance"
profile. The D-Bus error returned is Code 26 (Device or resource busy).
After enabling dynamic debug for the amd_pstate module (echo 'module amd_pstate
+p' > /sys/kernel/debug/dynamic_debug/control), I captured the following reason
for the failure in dmesg:
[ 299.585428] amd_pstate: EPP cannot be set under performance policy
Steps to Reproduce:
Use a system with amd_pstate in active mode on kernel 7.1.
Attempt to switch profile: powerprofilesctl set performance.
Failed to communicate with power-profiles-daemon: g-io-error-quark:
GDBus.Error:org.gtk.GDBus.UnmappedGError.Quark._g_2dio_2derror_2dquark.Code26:
Failed to activate CPU driver 'amd_pstate': Error writing
'/sys/devices/system/cpu/cpufreq/policy11/energy_performance_preference':
Device or resource busy (26)
The command fails, and dmesg logs: amd_pstate: EPP cannot be set under
performance policy.
Expected Result:
The driver should allow setting the EPP value even if the policy is set to
performance, or handle the transition gracefully without returning EBUSY to
user-space daemons.
Additional Context:
This issue was observed on the new Strix Halo (Ryzen AI Max+) platform. The
conflict between PPD's profile orchestration and the driver's internal state
checks makes the "Performance" profile unusable via standard desktop tools
(KDE/GNOME power settings).
I can build and test custom patches on this hardware
--
You may reply to this email to add a comment.
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug 221505] REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1
2026-05-11 17:45 [Bug 221505] New: REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1 bugzilla-daemon
@ 2026-05-11 20:15 ` bugzilla-daemon
2026-05-13 6:17 ` bugzilla-daemon
` (9 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: bugzilla-daemon @ 2026-05-11 20:15 UTC (permalink / raw)
To: linux-pm
https://bugzilla.kernel.org/show_bug.cgi?id=221505
--- Comment #1 from Artem S. Tashkinov (aros@gmx.com) ---
*** Bug 221506 has been marked as a duplicate of this bug. ***
--
You may reply to this email to add a comment.
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug 221505] REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1
2026-05-11 17:45 [Bug 221505] New: REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1 bugzilla-daemon
2026-05-11 20:15 ` [Bug 221505] " bugzilla-daemon
@ 2026-05-13 6:17 ` bugzilla-daemon
2026-05-13 11:03 ` bugzilla-daemon
` (8 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: bugzilla-daemon @ 2026-05-13 6:17 UTC (permalink / raw)
To: linux-pm
https://bugzilla.kernel.org/show_bug.cgi?id=221505
The Linux kernel's regression tracker (Thorsten Leemhuis) (regressions@leemhuis.info) changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |mario.limonciello@amd.com,
| |regressions@leemhuis.info
--- Comment #2 from The Linux kernel's regression tracker (Thorsten Leemhuis) (regressions@leemhuis.info) ---
Adding Mario – and also wondering if this is related to
https://lore.kernel.org/linux-pm/14a87c99-785c-4b16-bfce-35ecbf053448@freenet.de/
and https://bugzilla.kernel.org/show_bug.cgi?id=221473 , but it sounds a bit
like this is something different.
--
You may reply to this email to add a comment.
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug 221505] REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1
2026-05-11 17:45 [Bug 221505] New: REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1 bugzilla-daemon
2026-05-11 20:15 ` [Bug 221505] " bugzilla-daemon
2026-05-13 6:17 ` bugzilla-daemon
@ 2026-05-13 11:03 ` bugzilla-daemon
2026-05-13 15:39 ` bugzilla-daemon
` (7 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: bugzilla-daemon @ 2026-05-13 11:03 UTC (permalink / raw)
To: linux-pm
https://bugzilla.kernel.org/show_bug.cgi?id=221505
--- Comment #3 from Mario Limonciello (AMD) (mario.limonciello@amd.com) ---
Did you set CONFIG_X86_AMD_PSTATE_DYNAMIC_EPP
--
You may reply to this email to add a comment.
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug 221505] REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1
2026-05-11 17:45 [Bug 221505] New: REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1 bugzilla-daemon
` (2 preceding siblings ...)
2026-05-13 11:03 ` bugzilla-daemon
@ 2026-05-13 15:39 ` bugzilla-daemon
2026-05-13 15:43 ` bugzilla-daemon
` (6 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: bugzilla-daemon @ 2026-05-13 15:39 UTC (permalink / raw)
To: linux-pm
https://bugzilla.kernel.org/show_bug.cgi?id=221505
--- Comment #4 from Sergey (sergeyluashkov@gmail.com) ---
(In reply to Mario Limonciello (AMD) from comment #3)
> Did you set CONFIG_X86_AMD_PSTATE_DYNAMIC_EPP
After enabling CONFIG_X86_AMD_PSTATE_DYNAMIC_EPP=y, power-profiles-daemon (PPD)
loses the ability to switch to the power-saver profile. Attempting to set it
via powerprofilesctl set power-saver results in the system defaulting back to
balanced.
Technical details:
Governor limitation:
cat /sys/devices/system/cpu/cpufreq/policy0/scaling_available_governors returns
only [performance, powersave].
PPD Failure:
Since only two governors are visible, PPD fails to map the "power-saver"
profile correctly, even though the underlying hardware supports it.
Manual Override:
I discovered that the only way to successfully force the "power-saver" state is
by writing directly to the platform profile:
echo "quiet" | sudo tee
/sys/devices/platform/asus-nb-wmi/platform-profile/platform-profile-25/profile
(Note: The path may vary, but the "quiet" hint works where PPD fails).
Steps to Reproduce:
Enable CONFIG_X86_AMD_PSTATE_DYNAMIC_EPP=y.
Try to set power-saver: powerprofilesctl set power-saver.
Check status: powerprofilesctl get returns balanced.
Expected Result:
PPD should recognize that in active/dynamic_epp mode, "power-saver" should map
to the powersave governor with an EPP hint of power (or quiet at the platform
level)
Actual Result:
PPD is stuck between two profiles (balanced/performance)
--
You may reply to this email to add a comment.
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug 221505] REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1
2026-05-11 17:45 [Bug 221505] New: REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1 bugzilla-daemon
` (3 preceding siblings ...)
2026-05-13 15:39 ` bugzilla-daemon
@ 2026-05-13 15:43 ` bugzilla-daemon
2026-05-13 16:57 ` bugzilla-daemon
` (5 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: bugzilla-daemon @ 2026-05-13 15:43 UTC (permalink / raw)
To: linux-pm
https://bugzilla.kernel.org/show_bug.cgi?id=221505
--- Comment #5 from Mario Limonciello (AMD) (mario.limonciello@amd.com) ---
Yes - this is expected behavior with the series right now. I am proposing to
drop the kconfig and make the dynamic epp mode opt in at runtiem or kernel
command line.
https://lore.kernel.org/linux-pm/20260512221947.1652988-1-mario.limonciello@amd.com/
--
You may reply to this email to add a comment.
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug 221505] REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1
2026-05-11 17:45 [Bug 221505] New: REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1 bugzilla-daemon
` (4 preceding siblings ...)
2026-05-13 15:43 ` bugzilla-daemon
@ 2026-05-13 16:57 ` bugzilla-daemon
2026-05-13 17:12 ` bugzilla-daemon
` (4 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: bugzilla-daemon @ 2026-05-13 16:57 UTC (permalink / raw)
To: linux-pm
https://bugzilla.kernel.org/show_bug.cgi?id=221505
--- Comment #6 from Sergey (sergeyluashkov@gmail.com) ---
(In reply to Mario Limonciello (AMD) from comment #5)
> Yes - this is expected behavior with the series right now. I am proposing
> to drop the kconfig and make the dynamic epp mode opt in at runtiem or
> kernel command line.
>
> https://lore.kernel.org/linux-pm/20260512221947.1652988-1-mario.
> limonciello@amd.com/
I’ve tested the proposed patch on my Ryzen AI Max+ (Strix Halo) system (ASUS
FA401EA), and here are the results:
With amd_dynamic_epp=enable:
The issue described in the Comment 4.
With amd_dynamic_epp disabled (or no parameter):
The issue described in the Description
--
You may reply to this email to add a comment.
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug 221505] REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1
2026-05-11 17:45 [Bug 221505] New: REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1 bugzilla-daemon
` (5 preceding siblings ...)
2026-05-13 16:57 ` bugzilla-daemon
@ 2026-05-13 17:12 ` bugzilla-daemon
2026-05-13 18:41 ` bugzilla-daemon
` (3 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: bugzilla-daemon @ 2026-05-13 17:12 UTC (permalink / raw)
To: linux-pm
https://bugzilla.kernel.org/show_bug.cgi?id=221505
--- Comment #7 from Mario Limonciello (AMD) (mario.limonciello@amd.com) ---
Can you please pair that patch with the patches on this branch?
https://git.kernel.org/pub/scm/linux/kernel/git/superm1/linux.git/log/?h=amd-pstate-fixes
There are multiple bugs at play here.
--
You may reply to this email to add a comment.
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug 221505] REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1
2026-05-11 17:45 [Bug 221505] New: REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1 bugzilla-daemon
` (6 preceding siblings ...)
2026-05-13 17:12 ` bugzilla-daemon
@ 2026-05-13 18:41 ` bugzilla-daemon
2026-05-13 18:44 ` bugzilla-daemon
` (2 subsequent siblings)
10 siblings, 0 replies; 12+ messages in thread
From: bugzilla-daemon @ 2026-05-13 18:41 UTC (permalink / raw)
To: linux-pm
https://bugzilla.kernel.org/show_bug.cgi?id=221505
--- Comment #8 from Sergey (sergeyluashkov@gmail.com) ---
(In reply to Mario Limonciello (AMD) from comment #7)
> Can you please pair that patch with the patches on this branch?
> https://git.kernel.org/pub/scm/linux/kernel/git/superm1/linux.git/log/?h=amd-
> pstate-fixes
>
> There are multiple bugs at play here.
I’ve tested the kernel from your amd-pstate-fixes branch (commit f9f16835d4dc),
but unfortunately, the issues persist, and I’ve discovered a new conflict with
the ASUS platform driver.
Current observations:
With amd_dynamic_epp=disable (or no parameter):
The behavior remains identical to the initial report. powerprofilesctl still
fails with the error described in the original report (Description).
With amd_dynamic_epp=enable:
It breaks the integration with asus-nb-wmi.
Specifically, the /sys/devices/platform/asus-nb-wmi/throttle_thermal_policy
(which controls ASUS thermal profiles) stops switching to value 2
(Silent/Power-save mode).
When I set "Power-saver" via PPD, it only toggles between 0 and 1.
The CPU TDP is not limited to its expected ~55W threshold unless I manually
write 2 to the throttle_thermal_policy file.
--
You may reply to this email to add a comment.
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug 221505] REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1
2026-05-11 17:45 [Bug 221505] New: REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1 bugzilla-daemon
` (7 preceding siblings ...)
2026-05-13 18:41 ` bugzilla-daemon
@ 2026-05-13 18:44 ` bugzilla-daemon
2026-05-13 19:14 ` bugzilla-daemon
2026-05-14 7:30 ` bugzilla-daemon
10 siblings, 0 replies; 12+ messages in thread
From: bugzilla-daemon @ 2026-05-13 18:44 UTC (permalink / raw)
To: linux-pm
https://bugzilla.kernel.org/show_bug.cgi?id=221505
--- Comment #9 from Mario Limonciello (AMD) (mario.limonciello@amd.com) ---
> With amd_dynamic_epp=enable:
I'm not sure why this is happening. What platform profiles does this other
driver support?
> With amd_dynamic_epp=disable (or no parameter):
So you're still seeing this?
[ 299.585428] amd_pstate: EPP cannot be set under performance policy
Can I see the kernel log with dynamic debug for amd-pstate turned on? And also
the output from the amd-pstate triage tool in amd-debug-tools.
--
You may reply to this email to add a comment.
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug 221505] REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1
2026-05-11 17:45 [Bug 221505] New: REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1 bugzilla-daemon
` (8 preceding siblings ...)
2026-05-13 18:44 ` bugzilla-daemon
@ 2026-05-13 19:14 ` bugzilla-daemon
2026-05-14 7:30 ` bugzilla-daemon
10 siblings, 0 replies; 12+ messages in thread
From: bugzilla-daemon @ 2026-05-13 19:14 UTC (permalink / raw)
To: linux-pm
https://bugzilla.kernel.org/show_bug.cgi?id=221505
--- Comment #10 from Sergey (sergeyluashkov@gmail.com) ---
(In reply to Mario Limonciello (AMD) from comment #9)
> > With amd_dynamic_epp=enable:
>
> I'm not sure why this is happening. What platform profiles does this other
> driver support?
>
> > With amd_dynamic_epp=disable (or no parameter):
>
> So you're still seeing this?
>
> [ 299.585428] amd_pstate: EPP cannot be set under performance policy
>
> Can I see the kernel log with dynamic debug for amd-pstate turned on? And
> also the output from the amd-pstate triage tool in amd-debug-tools.
With amd_dynamic_epp=enable:
cat /sys/devices/platform/asus-nb-wmi/platform-profile/platform-profile-25/name
asus-wmi
cat
/sys/devices/platform/asus-nb-wmi/platform-profile/platform-profile-25/profile
balanced
cat
/sys/devices/platform/asus-nb-wmi/platform-profile/platform-profile-25/choices
quiet balanced performance
With amd_dynamic_epp=disable (or no parameter):
[ 128.393895] amd_pstate: EPP cannot be set under performance policy
🐧 Arch Linux
🐧 Kernel: 7.1.0-rc2-1-mainline-gf9f16835d4dc-dirty
○ 'status': active
○ 'prefcore': enabled
🐧 ITMT: Y
💻 CPU: AMD RYZEN AI MAX+ 392 w/ Radeon 8060S
🔋 Per-CPU sysfs files
+---------+----------------+----------------------+----------------+--------------------+--------------------+---------------------------------+------------+---------+
| CPU # | CPU Min Freq | CPU Nonlinear Freq | CPU Max Freq | Scaling
Min Freq | Scaling Max Freq | Energy Performance Preference | Prefcore |
Boost |
|---------+----------------+----------------------+----------------+--------------------+--------------------+---------------------------------+------------+---------|
| 0 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 206 |
0 |
| 1 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 216 |
0 |
| 2 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 201 |
0 |
| 3 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 216 |
0 |
| 4 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 196 |
0 |
| 5 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 211 |
0 |
| 6 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 176 |
0 |
| 7 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 166 |
0 |
| 8 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 186 |
0 |
| 9 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 181 |
0 |
| 10 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 171 |
0 |
| 11 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 191 |
0 |
| 12 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 206 |
0 |
| 13 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 216 |
0 |
| 14 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 201 |
0 |
| 15 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 216 |
0 |
| 16 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 196 |
0 |
| 17 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 211 |
0 |
| 18 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 176 |
0 |
| 19 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 166 |
0 |
| 20 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 186 |
0 |
| 21 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 181 |
0 |
| 22 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 171 |
0 |
| 23 | 609334 | 1980334 | 3199000 |
609334 | 3199000 | power | 191 |
0 |
+---------+----------------+----------------------+----------------+--------------------+--------------------+---------------------------------+------------+---------+
🔋 CPPC MSRs
+---------+----------+----------+--------------------+---------+------------+
| CPU # | Enable | Status | Cap 1 | Cap 2 | Request |
|---------+----------+----------+--------------------+---------+------------|
| 0 | 1 | 0 | 0xce694114ce694114 | 0x0 | 0xff001569 |
| 1 | 1 | 0 | 0xd8694114d8694114 | 0x0 | 0xff001569 |
| 2 | 1 | 0 | 0xc9694114c9694114 | 0x0 | 0xff001569 |
| 3 | 1 | 0 | 0xd8694114d8694114 | 0x0 | 0xff001569 |
| 4 | 1 | 0 | 0xc4694114c4694114 | 0x0 | 0xff001569 |
| 5 | 1 | 0 | 0xd3694114d3694114 | 0x0 | 0xff001569 |
| 6 | 1 | 0 | 0xb0694114b0694114 | 0x0 | 0xff001569 |
| 7 | 1 | 0 | 0xa6694114a6694114 | 0x0 | 0xff001569 |
| 8 | 1 | 0 | 0xba694114ba694114 | 0x0 | 0xff001569 |
| 9 | 1 | 0 | 0xb5694114b5694114 | 0x0 | 0xff001569 |
| 10 | 1 | 0 | 0xab694114ab694114 | 0x0 | 0xff001569 |
| 11 | 1 | 0 | 0xbf694114bf694114 | 0x0 | 0xff001569 |
| 12 | 1 | 0 | 0xce694114 | 0x0 | 0xff001569 |
| 13 | 1 | 0 | 0xd8694114 | 0x0 | 0xff001569 |
| 14 | 1 | 0 | 0xc9694114 | 0x0 | 0xff001569 |
| 15 | 1 | 0 | 0xd8694114 | 0x0 | 0xff001569 |
| 16 | 1 | 0 | 0xc4694114 | 0x0 | 0xff001569 |
| 17 | 1 | 0 | 0xd3694114 | 0x0 | 0xff001569 |
| 18 | 1 | 0 | 0xb0694114 | 0x0 | 0xff001569 |
| 19 | 1 | 0 | 0xa6694114 | 0x0 | 0xff001569 |
| 20 | 1 | 0 | 0xba694114 | 0x0 | 0xff001569 |
| 21 | 1 | 0 | 0xb5694114 | 0x0 | 0xff001569 |
| 22 | 1 | 0 | 0xab694114 | 0x0 | 0xff001569 |
| 23 | 1 | 0 | 0xbf694114 | 0x0 | 0xff001569 |
+---------+----------+----------+--------------------+---------+------------+
🔋 MSR_AMD_CPPC_CAP1 (decoded)
+---------+---------------+------------------+----------------+----------------+
| CPU # | Lowest Perf | Nonlinear Perf | Nominal Perf | Highest Perf
|
|---------+---------------+------------------+----------------+----------------|
| 0 | 20 | 65 | 105 | 206
|
| 1 | 20 | 65 | 105 | 216
|
| 2 | 20 | 65 | 105 | 201
|
| 3 | 20 | 65 | 105 | 216
|
| 4 | 20 | 65 | 105 | 196
|
| 5 | 20 | 65 | 105 | 211
|
| 6 | 20 | 65 | 105 | 176
|
| 7 | 20 | 65 | 105 | 166
|
| 8 | 20 | 65 | 105 | 186
|
| 9 | 20 | 65 | 105 | 181
|
| 10 | 20 | 65 | 105 | 171
|
| 11 | 20 | 65 | 105 | 191
|
| 12 | 20 | 65 | 105 | 206
|
| 13 | 20 | 65 | 105 | 216
|
| 14 | 20 | 65 | 105 | 201
|
| 15 | 20 | 65 | 105 | 216
|
| 16 | 20 | 65 | 105 | 196
|
| 17 | 20 | 65 | 105 | 211
|
| 18 | 20 | 65 | 105 | 176
|
| 19 | 20 | 65 | 105 | 166
|
| 20 | 20 | 65 | 105 | 186
|
| 21 | 20 | 65 | 105 | 181
|
| 22 | 20 | 65 | 105 | 171
|
| 23 | 20 | 65 | 105 | 191
|
+---------+---------------+------------------+----------------+----------------+
🔋 MSR_AMD_CPPC_REQ (decoded)
+---------+------------+------------+----------------+---------------------------+
| CPU # | Min Perf | Max Perf | Desired Perf | Energy Performance
Perf |
|---------+------------+------------+----------------+---------------------------|
| 0 | 21 | 105 | 0 |
255 |
| 1 | 21 | 105 | 0 |
255 |
| 2 | 21 | 105 | 0 |
255 |
| 3 | 21 | 105 | 0 |
255 |
| 4 | 21 | 105 | 0 |
255 |
| 5 | 21 | 105 | 0 |
255 |
| 6 | 21 | 105 | 0 |
255 |
| 7 | 21 | 105 | 0 |
255 |
| 8 | 21 | 105 | 0 |
255 |
| 9 | 21 | 105 | 0 |
255 |
| 10 | 21 | 105 | 0 |
255 |
| 11 | 21 | 105 | 0 |
255 |
| 12 | 21 | 105 | 0 |
255 |
| 13 | 21 | 105 | 0 |
255 |
| 14 | 21 | 105 | 0 |
255 |
| 15 | 21 | 105 | 0 |
255 |
| 16 | 21 | 105 | 0 |
255 |
| 17 | 21 | 105 | 0 |
255 |
| 18 | 21 | 105 | 0 |
255 |
| 19 | 21 | 105 | 0 |
255 |
| 20 | 21 | 105 | 0 |
255 |
| 21 | 21 | 105 | 0 |
255 |
| 22 | 21 | 105 | 0 |
255 |
| 23 | 21 | 105 | 0 |
255 |
+---------+------------+------------+----------------+---------------------------+
--
You may reply to this email to add a comment.
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 12+ messages in thread
* [Bug 221505] REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1
2026-05-11 17:45 [Bug 221505] New: REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1 bugzilla-daemon
` (9 preceding siblings ...)
2026-05-13 19:14 ` bugzilla-daemon
@ 2026-05-14 7:30 ` bugzilla-daemon
10 siblings, 0 replies; 12+ messages in thread
From: bugzilla-daemon @ 2026-05-14 7:30 UTC (permalink / raw)
To: linux-pm
https://bugzilla.kernel.org/show_bug.cgi?id=221505
Rhys Edwards (rhys.r.edwards@gmail.com) changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |rhys.r.edwards@gmail.com
--- Comment #11 from Rhys Edwards (rhys.r.edwards@gmail.com) ---
Not sure if i'm adding much additional info here not already known but i also
noticed this issue recently and believe it is also linked to the frequency
limiting problem in https://bugzilla.kernel.org/show_bug.cgi?id=221473 as i
have/had both issues
I've had cachyos installed using rc kernel on a desktop computer and last week
or so noticed cpu temps/wattage/frequency was much lower than normal, frequency
was specifically capping out at exactly 3Ghz regardless of workload.
I don't recall changing any pstate profiles at any point since i installed a
few months ago so i assume it was on performance(?)
CPU: AMD Ryzen 7 5800x
Mobo: MSI x570 Tomahawk
Ram: 32GB DDR4 3600Mhz
GPU: AMD RX570 4GB
It seems mostly fixed when i switched back to balanced using powerprofilesctl,
but now i also can't set back to performance getting the error mentioned here:
Failed to communicate with power-profiles-daemon: g-io-error-quark:
GDBus.Error:org.gtk.GDBus.UnmappedGError.Quark._g_2dio_2derror_2dquark.Code26:
Failed to activate CPU driver 'amd_pstate': Error writing
'/sys/devices/system/cpu/cpufreq/policy11/energy_performance_preference':
Device or resource busy (26)
Here is my p-state triage if it's any use...
🐧 CachyOS
🐧 Kernel: 7.1.0-rc3-1-cachyos-rc
○ 'status': active
○ 'prefcore': enabled
🐧 ITMT: Y
💻 CPU: AMD Ryzen 7 5800X 8-Core Processor
🔋 Per-CPU sysfs files
+---------+----------------+----------------------+----------------+--------------------+--------------------+---------------------------------+------------+---------+
| CPU # | CPU Min Freq | CPU Nonlinear Freq | CPU Max Freq | Scaling
Min Freq | Scaling Max Freq | Energy Performance Preference | Prefcore |
Boost |
|---------+----------------+----------------------+----------------+--------------------+--------------------+---------------------------------+------------+---------|
| 0 | 555531 | 1754308 | 4853585 |
1754308 | 4853585 | balance_performance | 176 |
1 |
| 1 | 555531 | 1754308 | 4853585 |
1754308 | 4853585 | balance_performance | 196 |
1 |
| 2 | 555531 | 1754308 | 4853585 |
1754308 | 4853585 | balance_performance | 186 |
1 |
| 3 | 555531 | 1754308 | 4853585 |
1754308 | 4853585 | balance_performance | 181 |
1 |
| 4 | 555531 | 1754308 | 4853585 |
1754308 | 4853585 | balance_performance | 171 |
1 |
| 5 | 555531 | 1754308 | 4853585 |
1754308 | 4853585 | balance_performance | 191 |
1 |
| 6 | 555531 | 1754308 | 4853585 |
1754308 | 4853585 | balance_performance | 166 |
1 |
| 7 | 555531 | 1754308 | 4853585 |
1754308 | 4853585 | balance_performance | 196 |
1 |
| 8 | 555531 | 1754308 | 4853585 |
1754308 | 4853585 | balance_performance | 176 |
1 |
| 9 | 555531 | 1754308 | 4853585 |
1754308 | 4853585 | balance_performance | 196 |
1 |
| 10 | 555531 | 1754308 | 4853585 |
1754308 | 4853585 | balance_performance | 186 |
1 |
| 11 | 555531 | 1754308 | 4853585 |
1754308 | 4853585 | balance_performance | 181 |
1 |
| 12 | 555531 | 1754308 | 4853585 |
1754308 | 4853585 | balance_performance | 171 |
1 |
| 13 | 555531 | 1754308 | 4853585 |
1754308 | 4853585 | balance_performance | 191 |
1 |
| 14 | 555531 | 1754308 | 4853585 |
1754308 | 4853585 | balance_performance | 166 |
1 |
| 15 | 555531 | 1754308 | 4853585 |
1754308 | 4853585 | balance_performance | 196 |
1 |
+---------+----------------+----------------------+----------------+--------------------+--------------------+---------------------------------+------------+---------+
🔋 CPPC MSRs
+---------+----------+----------+---------+---------+-----------+
| CPU # | Enable | Status | Cap 1 | Cap 2 | Request |
|---------+----------+----------+---------+---------+-----------|
| 0 | 0 | 0 | 0x0 | 0x0 | 0x0 |
| 1 | 0 | 0 | 0x0 | 0x0 | 0x0 |
| 2 | 0 | 0 | 0x0 | 0x0 | 0x0 |
| 3 | 0 | 0 | 0x0 | 0x0 | 0x0 |
| 4 | 0 | 0 | 0x0 | 0x0 | 0x0 |
| 5 | 0 | 0 | 0x0 | 0x0 | 0x0 |
| 6 | 0 | 0 | 0x0 | 0x0 | 0x0 |
| 7 | 0 | 0 | 0x0 | 0x0 | 0x0 |
| 8 | 0 | 0 | 0x0 | 0x0 | 0x0 |
| 9 | 0 | 0 | 0x0 | 0x0 | 0x0 |
| 10 | 0 | 0 | 0x0 | 0x0 | 0x0 |
| 11 | 0 | 0 | 0x0 | 0x0 | 0x0 |
| 12 | 0 | 0 | 0x0 | 0x0 | 0x0 |
| 13 | 0 | 0 | 0x0 | 0x0 | 0x0 |
| 14 | 0 | 0 | 0x0 | 0x0 | 0x0 |
| 15 | 0 | 0 | 0x0 | 0x0 | 0x0 |
+---------+----------+----------+---------+---------+-----------+
🔋 MSR_AMD_CPPC_CAP1 (decoded)
+---------+---------------+------------------+----------------+----------------+
| CPU # | Lowest Perf | Nonlinear Perf | Nominal Perf | Highest Perf
|
|---------+---------------+------------------+----------------+----------------|
| 0 | 0 | 0 | 0 | 0
|
| 1 | 0 | 0 | 0 | 0
|
| 2 | 0 | 0 | 0 | 0
|
| 3 | 0 | 0 | 0 | 0
|
| 4 | 0 | 0 | 0 | 0
|
| 5 | 0 | 0 | 0 | 0
|
| 6 | 0 | 0 | 0 | 0
|
| 7 | 0 | 0 | 0 | 0
|
| 8 | 0 | 0 | 0 | 0
|
| 9 | 0 | 0 | 0 | 0
|
| 10 | 0 | 0 | 0 | 0
|
| 11 | 0 | 0 | 0 | 0
|
| 12 | 0 | 0 | 0 | 0
|
| 13 | 0 | 0 | 0 | 0
|
| 14 | 0 | 0 | 0 | 0
|
| 15 | 0 | 0 | 0 | 0
|
+---------+---------------+------------------+----------------+----------------+
🔋 MSR_AMD_CPPC_REQ (decoded)
+---------+------------+------------+----------------+---------------------------+
| CPU # | Min Perf | Max Perf | Desired Perf | Energy Performance
Perf |
|---------+------------+------------+----------------+---------------------------|
| 0 | 0 | 0 | 0 |
0 |
| 1 | 0 | 0 | 0 |
0 |
| 2 | 0 | 0 | 0 |
0 |
| 3 | 0 | 0 | 0 |
0 |
| 4 | 0 | 0 | 0 |
0 |
| 5 | 0 | 0 | 0 |
0 |
| 6 | 0 | 0 | 0 |
0 |
| 7 | 0 | 0 | 0 |
0 |
| 8 | 0 | 0 | 0 |
0 |
| 9 | 0 | 0 | 0 |
0 |
| 10 | 0 | 0 | 0 |
0 |
| 11 | 0 | 0 | 0 |
0 |
| 12 | 0 | 0 | 0 |
0 |
| 13 | 0 | 0 | 0 |
0 |
| 14 | 0 | 0 | 0 |
0 |
| 15 | 0 | 0 | 0 |
0 |
+---------+------------+------------+----------------+---------------------------+
--
You may reply to this email to add a comment.
You are receiving this mail because:
You are the assignee for the bug.
^ permalink raw reply [flat|nested] 12+ messages in thread
end of thread, other threads:[~2026-05-14 7:30 UTC | newest]
Thread overview: 12+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-05-11 17:45 [Bug 221505] New: REGRESSION][amd_pstate] "EPP cannot be set under performance policy" (EBUSY) when switching to performance profile via PPD on 7.1 bugzilla-daemon
2026-05-11 20:15 ` [Bug 221505] " bugzilla-daemon
2026-05-13 6:17 ` bugzilla-daemon
2026-05-13 11:03 ` bugzilla-daemon
2026-05-13 15:39 ` bugzilla-daemon
2026-05-13 15:43 ` bugzilla-daemon
2026-05-13 16:57 ` bugzilla-daemon
2026-05-13 17:12 ` bugzilla-daemon
2026-05-13 18:41 ` bugzilla-daemon
2026-05-13 18:44 ` bugzilla-daemon
2026-05-13 19:14 ` bugzilla-daemon
2026-05-14 7:30 ` bugzilla-daemon
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox