FAILED: patch "[PATCH] platform/x86/intel-uncore-freq: Fix current_freq_khz after" failed to apply to 6.6-stable tree

The patch below does not apply to the 6.6-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

To reproduce the conflict and resubmit, you may use the following commands:

git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.6.y
git checkout FETCH_HEAD
git cherry-pick -x 6b63520ed14b17bbe9c2103debbd2152dde1fba3
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<[email protected]>' --in-reply-to '2026072030-guy-reborn-3b8c@gregkh' --subject-prefix 'PATCH 6.6.y' 'HEAD^..'

Possible dependencies:

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From 6b63520ed14b17bbe9c2103debbd2152dde1fba3 Mon Sep 17 00:00:00 2001
From: Guixiong Wei <[email protected]>
Date: Tue, 2 Jun 2026 10:07:52 +0800
Subject: [PATCH] platform/x86/intel-uncore-freq: Fix current_freq_khz after
CPU hotplug
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When the last CPU of a legacy uncore die goes offline, uncore_freq_remove_die_entry() clears control_cpu. During CPU hotplug re-add, uncore_freq_add_entry() still populates sysfs attributes before assigning the new control CPU. As a result, the current frequency read returns -ENXIO and current_freq_khz is omitted from the recreated sysfs group.

Assign control_cpu before the initial read paths and before create_attr_group() so sysfs recreation uses the new online CPU. If sysfs creation fails, restore control_cpu to -1 to keep the error path state consistent.

index 3b554418a7a3..4bcafedf68ef 100644
@@ -285,15 +285,20 @@ int uncore_freq_add_entry(struct uncore_data *data, int cpu)
 			  data->package_id, data->die_id);
 	}
 
+	/*
+	 * Set the control CPU before any read path so entry recreation after CPU
+	 * hotplug can populate read-only attributes from the new online CPU.
+	 */
+	data->control_cpu = cpu;
 	uncore_read(data, &data->initial_min_freq_khz, UNCORE_INDEX_MIN_FREQ);
 	uncore_read(data, &data->initial_max_freq_khz, UNCORE_INDEX_MAX_FREQ);
 
 	ret = create_attr_group(data, data->name);
 	if (ret) {
+		data->control_cpu = -1;
 		if (data->domain_id != UNCORE_DOMAIN_ID_INVALID)
 			ida_free(&intel_uncore_ida, data->seqnum_id);
 	} else {
-		data->control_cpu = cpu;
 		data->valid = true;
 	}
 

commit 6b63520ed14b17bbe9c2103debbd2152dde1fba3 upstream.

When the last CPU of a legacy uncore die goes offline, uncore_freq_remove_die_entry() clears control_cpu. During CPU hotplug re-add, uncore_freq_add_entry() still populates sysfs attributes before assigning the new control CPU. As a result, the current frequency read returns -ENXIO and current_freq_khz is omitted from the recreated sysfs group.

Assign control_cpu before the initial read paths and before create_attr_group() so sysfs recreation uses the new online CPU. If sysfs creation fails, restore control_cpu to -1 to keep the error path state consistent.

---
 .../x86/intel/uncore-frequency/uncore-frequency-common.c   | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

index 33bb58dc3f..b362e0170a 100644
@@ -238,14 +238,19 @@ int uncore_freq_add_entry(struct uncore_data *data, int cpu)
 		sprintf(data->name, "package_%02d_die_%02d", data->package_id, data->die_id);
 	}
 
+	/*
+	 * Set the control CPU before any read path so entry recreation after CPU
+	 * hotplug can populate read-only attributes from the new online CPU.
+	 */
+	data->control_cpu = cpu;
 	uncore_read(data, &data->initial_min_freq_khz, &data->initial_max_freq_khz);
 
 	ret = create_attr_group(data, data->name);
 	if (ret) {
+		data->control_cpu = -1;
 		if (data->domain_id != UNCORE_DOMAIN_ID_INVALID)
 			ida_free(&intel_uncore_ida, data->instance_id);
 	} else {
-		data->control_cpu = cpu;
 		data->valid = true;
 	}
 

2.39.5

Queued for 6.6, thanks.

Thanks,
Sasha