core/region.c is overloaded with per-region control logic (pmem, dax,
sysram, etc). Move the pmem region driver logic from region.c into
region_pmem.c make it clear that this code only applies to pmem regions.
No functional changes.
Signed-off-by: Gregory Price <gourry@gourry.net>
---
drivers/cxl/core/Makefile | 1 +
drivers/cxl/core/core.h | 1 +
drivers/cxl/core/region.c | 184 --------------------------------
drivers/cxl/core/region_pmem.c | 189 +++++++++++++++++++++++++++++++++
4 files changed, 191 insertions(+), 184 deletions(-)
create mode 100644 drivers/cxl/core/region_pmem.c
diff --git a/drivers/cxl/core/Makefile b/drivers/cxl/core/Makefile
index a639a9499972..d1484a0e5eb4 100644
--- a/drivers/cxl/core/Makefile
+++ b/drivers/cxl/core/Makefile
@@ -16,6 +16,7 @@ cxl_core-y += pmu.o
cxl_core-y += cdat.o
cxl_core-$(CONFIG_TRACING) += trace.o
cxl_core-$(CONFIG_CXL_REGION) += region.o
+cxl_core-$(CONFIG_CXL_REGION) += region_pmem.o
cxl_core-$(CONFIG_CXL_MCE) += mce.o
cxl_core-$(CONFIG_CXL_FEATURES) += features.o
cxl_core-$(CONFIG_CXL_EDAC_MEM_FEATURES) += edac.o
diff --git a/drivers/cxl/core/core.h b/drivers/cxl/core/core.h
index 5539e941782f..ef03966eeabd 100644
--- a/drivers/cxl/core/core.h
+++ b/drivers/cxl/core/core.h
@@ -50,6 +50,7 @@ int cxl_get_poison_by_endpoint(struct cxl_port *port);
struct cxl_region *cxl_dpa_to_region(const struct cxl_memdev *cxlmd, u64 dpa);
u64 cxl_dpa_to_hpa(struct cxl_region *cxlr, const struct cxl_memdev *cxlmd,
u64 dpa);
+int devm_cxl_add_pmem_region(struct cxl_region *cxlr);
#else
static inline u64 cxl_dpa_to_hpa(struct cxl_region *cxlr,
diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c
index f24b7e754727..72aee1408efb 100644
--- a/drivers/cxl/core/region.c
+++ b/drivers/cxl/core/region.c
@@ -2781,46 +2781,6 @@ static ssize_t delete_region_store(struct device *dev,
}
DEVICE_ATTR_WO(delete_region);
-static void cxl_pmem_region_release(struct device *dev)
-{
- struct cxl_pmem_region *cxlr_pmem = to_cxl_pmem_region(dev);
- int i;
-
- for (i = 0; i < cxlr_pmem->nr_mappings; i++) {
- struct cxl_memdev *cxlmd = cxlr_pmem->mapping[i].cxlmd;
-
- put_device(&cxlmd->dev);
- }
-
- kfree(cxlr_pmem);
-}
-
-static const struct attribute_group *cxl_pmem_region_attribute_groups[] = {
- &cxl_base_attribute_group,
- NULL,
-};
-
-const struct device_type cxl_pmem_region_type = {
- .name = "cxl_pmem_region",
- .release = cxl_pmem_region_release,
- .groups = cxl_pmem_region_attribute_groups,
-};
-
-bool is_cxl_pmem_region(struct device *dev)
-{
- return dev->type == &cxl_pmem_region_type;
-}
-EXPORT_SYMBOL_NS_GPL(is_cxl_pmem_region, "CXL");
-
-struct cxl_pmem_region *to_cxl_pmem_region(struct device *dev)
-{
- if (dev_WARN_ONCE(dev, !is_cxl_pmem_region(dev),
- "not a cxl_pmem_region device\n"))
- return NULL;
- return container_of(dev, struct cxl_pmem_region, dev);
-}
-EXPORT_SYMBOL_NS_GPL(to_cxl_pmem_region, "CXL");
-
struct cxl_poison_context {
struct cxl_port *port;
int part;
@@ -3476,64 +3436,6 @@ static int region_offset_to_dpa_result(struct cxl_region *cxlr, u64 offset,
return -ENXIO;
}
-static struct lock_class_key cxl_pmem_region_key;
-
-static int cxl_pmem_region_alloc(struct cxl_region *cxlr)
-{
- struct cxl_region_params *p = &cxlr->params;
- struct cxl_nvdimm_bridge *cxl_nvb;
- struct device *dev;
- int i;
-
- guard(rwsem_read)(&cxl_rwsem.region);
- if (p->state != CXL_CONFIG_COMMIT)
- return -ENXIO;
-
- struct cxl_pmem_region *cxlr_pmem __free(kfree) =
- kzalloc_flex(*cxlr_pmem, mapping, p->nr_targets);
- if (!cxlr_pmem)
- return -ENOMEM;
-
- cxlr_pmem->hpa_range.start = p->res->start;
- cxlr_pmem->hpa_range.end = p->res->end;
-
- /* Snapshot the region configuration underneath the cxl_rwsem.region */
- cxlr_pmem->nr_mappings = p->nr_targets;
- for (i = 0; i < p->nr_targets; i++) {
- struct cxl_endpoint_decoder *cxled = p->targets[i];
- struct cxl_memdev *cxlmd = cxled_to_memdev(cxled);
- struct cxl_pmem_region_mapping *m = &cxlr_pmem->mapping[i];
-
- /*
- * Regions never span CXL root devices, so by definition the
- * bridge for one device is the same for all.
- */
- if (i == 0) {
- cxl_nvb = cxl_find_nvdimm_bridge(cxlmd->endpoint);
- if (!cxl_nvb)
- return -ENODEV;
- cxlr->cxl_nvb = cxl_nvb;
- }
- m->cxlmd = cxlmd;
- get_device(&cxlmd->dev);
- m->start = cxled->dpa_res->start;
- m->size = resource_size(cxled->dpa_res);
- m->position = i;
- }
-
- dev = &cxlr_pmem->dev;
- device_initialize(dev);
- lockdep_set_class(&dev->mutex, &cxl_pmem_region_key);
- device_set_pm_not_required(dev);
- dev->parent = &cxlr->dev;
- dev->bus = &cxl_bus_type;
- dev->type = &cxl_pmem_region_type;
- cxlr_pmem->cxlr = cxlr;
- cxlr->cxlr_pmem = no_free_ptr(cxlr_pmem);
-
- return 0;
-}
-
static void cxl_dax_region_release(struct device *dev)
{
struct cxl_dax_region *cxlr_dax = to_cxl_dax_region(dev);
@@ -3597,92 +3499,6 @@ static struct cxl_dax_region *cxl_dax_region_alloc(struct cxl_region *cxlr)
return cxlr_dax;
}
-static void cxlr_pmem_unregister(void *_cxlr_pmem)
-{
- struct cxl_pmem_region *cxlr_pmem = _cxlr_pmem;
- struct cxl_region *cxlr = cxlr_pmem->cxlr;
- struct cxl_nvdimm_bridge *cxl_nvb = cxlr->cxl_nvb;
-
- /*
- * Either the bridge is in ->remove() context under the device_lock(),
- * or cxlr_release_nvdimm() is cancelling the bridge's release action
- * for @cxlr_pmem and doing it itself (while manually holding the bridge
- * lock).
- */
- device_lock_assert(&cxl_nvb->dev);
- cxlr->cxlr_pmem = NULL;
- cxlr_pmem->cxlr = NULL;
- device_unregister(&cxlr_pmem->dev);
-}
-
-static void cxlr_release_nvdimm(void *_cxlr)
-{
- struct cxl_region *cxlr = _cxlr;
- struct cxl_nvdimm_bridge *cxl_nvb = cxlr->cxl_nvb;
-
- scoped_guard(device, &cxl_nvb->dev) {
- if (cxlr->cxlr_pmem)
- devm_release_action(&cxl_nvb->dev, cxlr_pmem_unregister,
- cxlr->cxlr_pmem);
- }
- cxlr->cxl_nvb = NULL;
- put_device(&cxl_nvb->dev);
-}
-
-/**
- * devm_cxl_add_pmem_region() - add a cxl_region-to-nd_region bridge
- * @cxlr: parent CXL region for this pmem region bridge device
- *
- * Return: 0 on success negative error code on failure.
- */
-static int devm_cxl_add_pmem_region(struct cxl_region *cxlr)
-{
- struct cxl_pmem_region *cxlr_pmem;
- struct cxl_nvdimm_bridge *cxl_nvb;
- struct device *dev;
- int rc;
-
- rc = cxl_pmem_region_alloc(cxlr);
- if (rc)
- return rc;
- cxlr_pmem = cxlr->cxlr_pmem;
- cxl_nvb = cxlr->cxl_nvb;
-
- dev = &cxlr_pmem->dev;
- rc = dev_set_name(dev, "pmem_region%d", cxlr->id);
- if (rc)
- goto err;
-
- rc = device_add(dev);
- if (rc)
- goto err;
-
- dev_dbg(&cxlr->dev, "%s: register %s\n", dev_name(dev->parent),
- dev_name(dev));
-
- scoped_guard(device, &cxl_nvb->dev) {
- if (cxl_nvb->dev.driver)
- rc = devm_add_action_or_reset(&cxl_nvb->dev,
- cxlr_pmem_unregister,
- cxlr_pmem);
- else
- rc = -ENXIO;
- }
-
- if (rc)
- goto err_bridge;
-
- /* @cxlr carries a reference on @cxl_nvb until cxlr_release_nvdimm */
- return devm_add_action_or_reset(&cxlr->dev, cxlr_release_nvdimm, cxlr);
-
-err:
- put_device(dev);
-err_bridge:
- put_device(&cxl_nvb->dev);
- cxlr->cxl_nvb = NULL;
- return rc;
-}
-
static void cxlr_dax_unregister(void *_cxlr_dax)
{
struct cxl_dax_region *cxlr_dax = _cxlr_dax;
diff --git a/drivers/cxl/core/region_pmem.c b/drivers/cxl/core/region_pmem.c
new file mode 100644
index 000000000000..138c373a5700
--- /dev/null
+++ b/drivers/cxl/core/region_pmem.c
@@ -0,0 +1,189 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright(c) 2022 Intel Corporation. All rights reserved. */
+#include <linux/device.h>
+#include <linux/slab.h>
+#include <cxlmem.h>
+#include <cxl.h>
+#include "core.h"
+
+static void cxl_pmem_region_release(struct device *dev)
+{
+ struct cxl_pmem_region *cxlr_pmem = to_cxl_pmem_region(dev);
+ int i;
+
+ for (i = 0; i < cxlr_pmem->nr_mappings; i++) {
+ struct cxl_memdev *cxlmd = cxlr_pmem->mapping[i].cxlmd;
+
+ put_device(&cxlmd->dev);
+ }
+
+ kfree(cxlr_pmem);
+}
+
+static const struct attribute_group *cxl_pmem_region_attribute_groups[] = {
+ &cxl_base_attribute_group,
+ NULL
+};
+
+const struct device_type cxl_pmem_region_type = {
+ .name = "cxl_pmem_region",
+ .release = cxl_pmem_region_release,
+ .groups = cxl_pmem_region_attribute_groups,
+};
+bool is_cxl_pmem_region(struct device *dev)
+{
+ return dev->type == &cxl_pmem_region_type;
+}
+EXPORT_SYMBOL_NS_GPL(is_cxl_pmem_region, "CXL");
+
+struct cxl_pmem_region *to_cxl_pmem_region(struct device *dev)
+{
+ if (dev_WARN_ONCE(dev, !is_cxl_pmem_region(dev),
+ "not a cxl_pmem_region device\n"))
+ return NULL;
+ return container_of(dev, struct cxl_pmem_region, dev);
+}
+EXPORT_SYMBOL_NS_GPL(to_cxl_pmem_region, "CXL");
+static struct lock_class_key cxl_pmem_region_key;
+
+static int cxl_pmem_region_alloc(struct cxl_region *cxlr)
+{
+ struct cxl_region_params *p = &cxlr->params;
+ struct cxl_nvdimm_bridge *cxl_nvb;
+ struct device *dev;
+ int i;
+
+ guard(rwsem_read)(&cxl_rwsem.region);
+ if (p->state != CXL_CONFIG_COMMIT)
+ return -ENXIO;
+
+ struct cxl_pmem_region *cxlr_pmem __free(kfree) =
+ kzalloc_flex(*cxlr_pmem, mapping, p->nr_targets);
+ if (!cxlr_pmem)
+ return -ENOMEM;
+
+ cxlr_pmem->hpa_range.start = p->res->start;
+ cxlr_pmem->hpa_range.end = p->res->end;
+
+ /* Snapshot the region configuration underneath the cxl_rwsem.region */
+ cxlr_pmem->nr_mappings = p->nr_targets;
+ for (i = 0; i < p->nr_targets; i++) {
+ struct cxl_endpoint_decoder *cxled = p->targets[i];
+ struct cxl_memdev *cxlmd = cxled_to_memdev(cxled);
+ struct cxl_pmem_region_mapping *m = &cxlr_pmem->mapping[i];
+
+ /*
+ * Regions never span CXL root devices, so by definition the
+ * bridge for one device is the same for all.
+ */
+ if (i == 0) {
+ cxl_nvb = cxl_find_nvdimm_bridge(cxlmd->endpoint);
+ if (!cxl_nvb)
+ return -ENODEV;
+ cxlr->cxl_nvb = cxl_nvb;
+ }
+ m->cxlmd = cxlmd;
+ get_device(&cxlmd->dev);
+ m->start = cxled->dpa_res->start;
+ m->size = resource_size(cxled->dpa_res);
+ m->position = i;
+ }
+
+ dev = &cxlr_pmem->dev;
+ device_initialize(dev);
+ lockdep_set_class(&dev->mutex, &cxl_pmem_region_key);
+ device_set_pm_not_required(dev);
+ dev->parent = &cxlr->dev;
+ dev->bus = &cxl_bus_type;
+ dev->type = &cxl_pmem_region_type;
+ cxlr_pmem->cxlr = cxlr;
+ cxlr->cxlr_pmem = no_free_ptr(cxlr_pmem);
+
+ return 0;
+}
+
+static void cxlr_pmem_unregister(void *_cxlr_pmem)
+{
+ struct cxl_pmem_region *cxlr_pmem = _cxlr_pmem;
+ struct cxl_region *cxlr = cxlr_pmem->cxlr;
+ struct cxl_nvdimm_bridge *cxl_nvb = cxlr->cxl_nvb;
+
+ /*
+ * Either the bridge is in ->remove() context under the device_lock(),
+ * or cxlr_release_nvdimm() is cancelling the bridge's release action
+ * for @cxlr_pmem and doing it itself (while manually holding the bridge
+ * lock).
+ */
+ device_lock_assert(&cxl_nvb->dev);
+ cxlr->cxlr_pmem = NULL;
+ cxlr_pmem->cxlr = NULL;
+ device_unregister(&cxlr_pmem->dev);
+}
+
+static void cxlr_release_nvdimm(void *_cxlr)
+{
+ struct cxl_region *cxlr = _cxlr;
+ struct cxl_nvdimm_bridge *cxl_nvb = cxlr->cxl_nvb;
+
+ scoped_guard(device, &cxl_nvb->dev) {
+ if (cxlr->cxlr_pmem)
+ devm_release_action(&cxl_nvb->dev, cxlr_pmem_unregister,
+ cxlr->cxlr_pmem);
+ }
+ cxlr->cxl_nvb = NULL;
+ put_device(&cxl_nvb->dev);
+}
+
+/**
+ * devm_cxl_add_pmem_region() - add a cxl_region-to-nd_region bridge
+ * @cxlr: parent CXL region for this pmem region bridge device
+ *
+ * Return: 0 on success negative error code on failure.
+ */
+int devm_cxl_add_pmem_region(struct cxl_region *cxlr)
+{
+ struct cxl_pmem_region *cxlr_pmem;
+ struct cxl_nvdimm_bridge *cxl_nvb;
+ struct device *dev;
+ int rc;
+
+ rc = cxl_pmem_region_alloc(cxlr);
+ if (rc)
+ return rc;
+ cxlr_pmem = cxlr->cxlr_pmem;
+ cxl_nvb = cxlr->cxl_nvb;
+
+ dev = &cxlr_pmem->dev;
+ rc = dev_set_name(dev, "pmem_region%d", cxlr->id);
+ if (rc)
+ goto err;
+
+ rc = device_add(dev);
+ if (rc)
+ goto err;
+
+ dev_dbg(&cxlr->dev, "%s: register %s\n", dev_name(dev->parent),
+ dev_name(dev));
+
+ scoped_guard(device, &cxl_nvb->dev) {
+ if (cxl_nvb->dev.driver)
+ rc = devm_add_action_or_reset(&cxl_nvb->dev,
+ cxlr_pmem_unregister,
+ cxlr_pmem);
+ else
+ rc = -ENXIO;
+ }
+
+ if (rc)
+ goto err_bridge;
+
+ /* @cxlr carries a reference on @cxl_nvb until cxlr_release_nvdimm */
+ return devm_add_action_or_reset(&cxlr->dev, cxlr_release_nvdimm, cxlr);
+
+err:
+ put_device(dev);
+err_bridge:
+ put_device(&cxl_nvb->dev);
+ cxlr->cxl_nvb = NULL;
+ return rc;
+}
--
2.53.0
On Sun, Mar 22, 2026 at 09:16:36AM -0400, Gregory Price wrote: > core/region.c is overloaded with per-region control logic (pmem, dax, > sysram, etc). Move the pmem region driver logic from region.c into > region_pmem.c make it clear that this code only applies to pmem regions. > > No functional changes. Reviewed-by: Alison Schofield <alison.schofield@intel.com>
On Sun, 22 Mar 2026 09:16:36 -0400
Gregory Price <gourry@gourry.net> wrote:
> core/region.c is overloaded with per-region control logic (pmem, dax,
> sysram, etc). Move the pmem region driver logic from region.c into
> region_pmem.c make it clear that this code only applies to pmem regions.
>
> No functional changes.
>
> Signed-off-by: Gregory Price <gourry@gourry.net>
> ---
> drivers/cxl/core/Makefile | 1 +
> drivers/cxl/core/core.h | 1 +
> drivers/cxl/core/region.c | 184 --------------------------------
> drivers/cxl/core/region_pmem.c | 189 +++++++++++++++++++++++++++++++++
> 4 files changed, 191 insertions(+), 184 deletions(-)
> create mode 100644 drivers/cxl/core/region_pmem.c
>
> diff --git a/drivers/cxl/core/Makefile b/drivers/cxl/core/Makefile
> index a639a9499972..d1484a0e5eb4 100644
> --- a/drivers/cxl/core/Makefile
> +++ b/drivers/cxl/core/Makefile
> @@ -16,6 +16,7 @@ cxl_core-y += pmu.o
> cxl_core-y += cdat.o
> cxl_core-$(CONFIG_TRACING) += trace.o
> cxl_core-$(CONFIG_CXL_REGION) += region.o
> +cxl_core-$(CONFIG_CXL_REGION) += region_pmem.o
cxl_core-$(CONFIG_CXL_REGION) += region.o region_pmem.o
For me having them on one line makes it more obvious it's
either neither or both. Cost is a bit more churn if you
decide later to split the CONFIG_* up.
> cxl_core-$(CONFIG_CXL_MCE) += mce.o
> cxl_core-$(CONFIG_CXL_FEATURES) += features.o
> cxl_core-$(CONFIG_CXL_EDAC_MEM_FEATURES) += edac.o
Otherwise a couple of white space things. + a passing comment on the lifetime
management being more complex than I'd like.
Fix the white space and you can add
Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com>
> diff --git a/drivers/cxl/core/region_pmem.c b/drivers/cxl/core/region_pmem.c
> new file mode 100644
> index 000000000000..138c373a5700
> --- /dev/null
> +++ b/drivers/cxl/core/region_pmem.c
> +const struct device_type cxl_pmem_region_type = {
> + .name = "cxl_pmem_region",
> + .release = cxl_pmem_region_release,
> + .groups = cxl_pmem_region_attribute_groups,
> +};
Blank line here. There is one in the original and I can't see why
we'd want to get rid of it.
> +bool is_cxl_pmem_region(struct device *dev)
> +{
> + return dev->type == &cxl_pmem_region_type;
> +}
> +EXPORT_SYMBOL_NS_GPL(is_cxl_pmem_region, "CXL");
> +
> +struct cxl_pmem_region *to_cxl_pmem_region(struct device *dev)
> +{
> + if (dev_WARN_ONCE(dev, !is_cxl_pmem_region(dev),
> + "not a cxl_pmem_region device\n"))
> + return NULL;
> + return container_of(dev, struct cxl_pmem_region, dev);
> +}
> +EXPORT_SYMBOL_NS_GPL(to_cxl_pmem_region, "CXL");
Blank line here probably makes sense as the key has nothing to do
with the EXPORT_SYMBOL...
> +static struct lock_class_key cxl_pmem_region_key;
> +
> +
> +/**
> + * devm_cxl_add_pmem_region() - add a cxl_region-to-nd_region bridge
> + * @cxlr: parent CXL region for this pmem region bridge device
> + *
> + * Return: 0 on success negative error code on failure.
FWIW the lifetime management in here feels way to complex to me. Not
a problem for this patch and I'm not immediately sure what we can do about it.
> + */
> +int devm_cxl_add_pmem_region(struct cxl_region *cxlr)
> +{
> + struct cxl_pmem_region *cxlr_pmem;
> + struct cxl_nvdimm_bridge *cxl_nvb;
> + struct device *dev;
> + int rc;
> +
> + rc = cxl_pmem_region_alloc(cxlr);
> + if (rc)
> + return rc;
> + cxlr_pmem = cxlr->cxlr_pmem;
> + cxl_nvb = cxlr->cxl_nvb;
> +
> + dev = &cxlr_pmem->dev;
> + rc = dev_set_name(dev, "pmem_region%d", cxlr->id);
> + if (rc)
> + goto err;
> +
> + rc = device_add(dev);
> + if (rc)
> + goto err;
> +
> + dev_dbg(&cxlr->dev, "%s: register %s\n", dev_name(dev->parent),
> + dev_name(dev));
> +
> + scoped_guard(device, &cxl_nvb->dev) {
> + if (cxl_nvb->dev.driver)
> + rc = devm_add_action_or_reset(&cxl_nvb->dev,
> + cxlr_pmem_unregister,
> + cxlr_pmem);
> + else
> + rc = -ENXIO;
As an example. If we happen to take this path... Where is the device_add() undone?
> + }
> +
> + if (rc)
> + goto err_bridge;
> +
> + /* @cxlr carries a reference on @cxl_nvb until cxlr_release_nvdimm */
> + return devm_add_action_or_reset(&cxlr->dev, cxlr_release_nvdimm, cxlr);
> +
> +err:
> + put_device(dev);
> +err_bridge:
> + put_device(&cxl_nvb->dev);
> + cxlr->cxl_nvb = NULL;
> + return rc;
> +}
On Mon, Mar 23, 2026 at 03:08:47PM +0000, Jonathan Cameron wrote:
> On Sun, 22 Mar 2026 09:16:36 -0400
> Gregory Price <gourry@gourry.net> wrote:
>
> > + scoped_guard(device, &cxl_nvb->dev) {
> > + if (cxl_nvb->dev.driver)
> > + rc = devm_add_action_or_reset(&cxl_nvb->dev,
> > + cxlr_pmem_unregister,
> > + cxlr_pmem);
> > + else
> > + rc = -ENXIO;
> As an example. If we happen to take this path... Where is the device_add() undone?
In many other parts of the code, device_add is not rolled back either
"for reasons", and - depending on who you talk to - this is
intentional... for debugging?
I dunno, but the "hanging objects on failure" thing is common throughout
cxl/ - :shrug:
~Gregory
On Mon, 23 Mar 2026 10:47:51 -0500
Gregory Price <gourry@gourry.net> wrote:
> On Mon, Mar 23, 2026 at 03:08:47PM +0000, Jonathan Cameron wrote:
> > On Sun, 22 Mar 2026 09:16:36 -0400
> > Gregory Price <gourry@gourry.net> wrote:
> >
> > > + scoped_guard(device, &cxl_nvb->dev) {
> > > + if (cxl_nvb->dev.driver)
> > > + rc = devm_add_action_or_reset(&cxl_nvb->dev,
> > > + cxlr_pmem_unregister,
> > > + cxlr_pmem);
> > > + else
> > > + rc = -ENXIO;
> > As an example. If we happen to take this path... Where is the device_add() undone?
>
> In many other parts of the code, device_add is not rolled back either
> "for reasons", and - depending on who you talk to - this is
> intentional... for debugging?
>
> I dunno, but the "hanging objects on failure" thing is common throughout
> cxl/ - :shrug:
If it were consistent maybe, but in other paths in this function it is cleaned up...
So I call bug even if it's one that is relatively unlikely to actually happen.
J
>
> ~Gregory
On Sun, Mar 22, 2026 at 09:16:36AM -0400, Gregory Price wrote:
> core/region.c is overloaded with per-region control logic (pmem, dax,
> sysram, etc). Move the pmem region driver logic from region.c into
> region_pmem.c make it clear that this code only applies to pmem regions.
>
> No functional changes.
This always adds region_pmem.o regardless of CONFIG_CXL_PMEM.
Are you heading towards conditionally compiling, where region_pmem.o is
omitted if !CONFIG_CXL_PMEM?
Similar question for region_dax.o
--Alison
>
> Signed-off-by: Gregory Price <gourry@gourry.net>
> ---
> drivers/cxl/core/Makefile | 1 +
> drivers/cxl/core/core.h | 1 +
> drivers/cxl/core/region.c | 184 --------------------------------
> drivers/cxl/core/region_pmem.c | 189 +++++++++++++++++++++++++++++++++
> 4 files changed, 191 insertions(+), 184 deletions(-)
> create mode 100644 drivers/cxl/core/region_pmem.c
>
> diff --git a/drivers/cxl/core/Makefile b/drivers/cxl/core/Makefile
> index a639a9499972..d1484a0e5eb4 100644
> --- a/drivers/cxl/core/Makefile
> +++ b/drivers/cxl/core/Makefile
> @@ -16,6 +16,7 @@ cxl_core-y += pmu.o
> cxl_core-y += cdat.o
> cxl_core-$(CONFIG_TRACING) += trace.o
> cxl_core-$(CONFIG_CXL_REGION) += region.o
> +cxl_core-$(CONFIG_CXL_REGION) += region_pmem.o
> cxl_core-$(CONFIG_CXL_MCE) += mce.o
> cxl_core-$(CONFIG_CXL_FEATURES) += features.o
> cxl_core-$(CONFIG_CXL_EDAC_MEM_FEATURES) += edac.o
> diff --git a/drivers/cxl/core/core.h b/drivers/cxl/core/core.h
> index 5539e941782f..ef03966eeabd 100644
> --- a/drivers/cxl/core/core.h
> +++ b/drivers/cxl/core/core.h
> @@ -50,6 +50,7 @@ int cxl_get_poison_by_endpoint(struct cxl_port *port);
> struct cxl_region *cxl_dpa_to_region(const struct cxl_memdev *cxlmd, u64 dpa);
> u64 cxl_dpa_to_hpa(struct cxl_region *cxlr, const struct cxl_memdev *cxlmd,
> u64 dpa);
> +int devm_cxl_add_pmem_region(struct cxl_region *cxlr);
>
> #else
> static inline u64 cxl_dpa_to_hpa(struct cxl_region *cxlr,
> diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c
> index f24b7e754727..72aee1408efb 100644
> --- a/drivers/cxl/core/region.c
> +++ b/drivers/cxl/core/region.c
> @@ -2781,46 +2781,6 @@ static ssize_t delete_region_store(struct device *dev,
> }
> DEVICE_ATTR_WO(delete_region);
>
> -static void cxl_pmem_region_release(struct device *dev)
> -{
> - struct cxl_pmem_region *cxlr_pmem = to_cxl_pmem_region(dev);
> - int i;
> -
> - for (i = 0; i < cxlr_pmem->nr_mappings; i++) {
> - struct cxl_memdev *cxlmd = cxlr_pmem->mapping[i].cxlmd;
> -
> - put_device(&cxlmd->dev);
> - }
> -
> - kfree(cxlr_pmem);
> -}
> -
> -static const struct attribute_group *cxl_pmem_region_attribute_groups[] = {
> - &cxl_base_attribute_group,
> - NULL,
> -};
> -
> -const struct device_type cxl_pmem_region_type = {
> - .name = "cxl_pmem_region",
> - .release = cxl_pmem_region_release,
> - .groups = cxl_pmem_region_attribute_groups,
> -};
> -
> -bool is_cxl_pmem_region(struct device *dev)
> -{
> - return dev->type == &cxl_pmem_region_type;
> -}
> -EXPORT_SYMBOL_NS_GPL(is_cxl_pmem_region, "CXL");
> -
> -struct cxl_pmem_region *to_cxl_pmem_region(struct device *dev)
> -{
> - if (dev_WARN_ONCE(dev, !is_cxl_pmem_region(dev),
> - "not a cxl_pmem_region device\n"))
> - return NULL;
> - return container_of(dev, struct cxl_pmem_region, dev);
> -}
> -EXPORT_SYMBOL_NS_GPL(to_cxl_pmem_region, "CXL");
> -
> struct cxl_poison_context {
> struct cxl_port *port;
> int part;
> @@ -3476,64 +3436,6 @@ static int region_offset_to_dpa_result(struct cxl_region *cxlr, u64 offset,
> return -ENXIO;
> }
>
> -static struct lock_class_key cxl_pmem_region_key;
> -
> -static int cxl_pmem_region_alloc(struct cxl_region *cxlr)
> -{
> - struct cxl_region_params *p = &cxlr->params;
> - struct cxl_nvdimm_bridge *cxl_nvb;
> - struct device *dev;
> - int i;
> -
> - guard(rwsem_read)(&cxl_rwsem.region);
> - if (p->state != CXL_CONFIG_COMMIT)
> - return -ENXIO;
> -
> - struct cxl_pmem_region *cxlr_pmem __free(kfree) =
> - kzalloc_flex(*cxlr_pmem, mapping, p->nr_targets);
> - if (!cxlr_pmem)
> - return -ENOMEM;
> -
> - cxlr_pmem->hpa_range.start = p->res->start;
> - cxlr_pmem->hpa_range.end = p->res->end;
> -
> - /* Snapshot the region configuration underneath the cxl_rwsem.region */
> - cxlr_pmem->nr_mappings = p->nr_targets;
> - for (i = 0; i < p->nr_targets; i++) {
> - struct cxl_endpoint_decoder *cxled = p->targets[i];
> - struct cxl_memdev *cxlmd = cxled_to_memdev(cxled);
> - struct cxl_pmem_region_mapping *m = &cxlr_pmem->mapping[i];
> -
> - /*
> - * Regions never span CXL root devices, so by definition the
> - * bridge for one device is the same for all.
> - */
> - if (i == 0) {
> - cxl_nvb = cxl_find_nvdimm_bridge(cxlmd->endpoint);
> - if (!cxl_nvb)
> - return -ENODEV;
> - cxlr->cxl_nvb = cxl_nvb;
> - }
> - m->cxlmd = cxlmd;
> - get_device(&cxlmd->dev);
> - m->start = cxled->dpa_res->start;
> - m->size = resource_size(cxled->dpa_res);
> - m->position = i;
> - }
> -
> - dev = &cxlr_pmem->dev;
> - device_initialize(dev);
> - lockdep_set_class(&dev->mutex, &cxl_pmem_region_key);
> - device_set_pm_not_required(dev);
> - dev->parent = &cxlr->dev;
> - dev->bus = &cxl_bus_type;
> - dev->type = &cxl_pmem_region_type;
> - cxlr_pmem->cxlr = cxlr;
> - cxlr->cxlr_pmem = no_free_ptr(cxlr_pmem);
> -
> - return 0;
> -}
> -
> static void cxl_dax_region_release(struct device *dev)
> {
> struct cxl_dax_region *cxlr_dax = to_cxl_dax_region(dev);
> @@ -3597,92 +3499,6 @@ static struct cxl_dax_region *cxl_dax_region_alloc(struct cxl_region *cxlr)
> return cxlr_dax;
> }
>
> -static void cxlr_pmem_unregister(void *_cxlr_pmem)
> -{
> - struct cxl_pmem_region *cxlr_pmem = _cxlr_pmem;
> - struct cxl_region *cxlr = cxlr_pmem->cxlr;
> - struct cxl_nvdimm_bridge *cxl_nvb = cxlr->cxl_nvb;
> -
> - /*
> - * Either the bridge is in ->remove() context under the device_lock(),
> - * or cxlr_release_nvdimm() is cancelling the bridge's release action
> - * for @cxlr_pmem and doing it itself (while manually holding the bridge
> - * lock).
> - */
> - device_lock_assert(&cxl_nvb->dev);
> - cxlr->cxlr_pmem = NULL;
> - cxlr_pmem->cxlr = NULL;
> - device_unregister(&cxlr_pmem->dev);
> -}
> -
> -static void cxlr_release_nvdimm(void *_cxlr)
> -{
> - struct cxl_region *cxlr = _cxlr;
> - struct cxl_nvdimm_bridge *cxl_nvb = cxlr->cxl_nvb;
> -
> - scoped_guard(device, &cxl_nvb->dev) {
> - if (cxlr->cxlr_pmem)
> - devm_release_action(&cxl_nvb->dev, cxlr_pmem_unregister,
> - cxlr->cxlr_pmem);
> - }
> - cxlr->cxl_nvb = NULL;
> - put_device(&cxl_nvb->dev);
> -}
> -
> -/**
> - * devm_cxl_add_pmem_region() - add a cxl_region-to-nd_region bridge
> - * @cxlr: parent CXL region for this pmem region bridge device
> - *
> - * Return: 0 on success negative error code on failure.
> - */
> -static int devm_cxl_add_pmem_region(struct cxl_region *cxlr)
> -{
> - struct cxl_pmem_region *cxlr_pmem;
> - struct cxl_nvdimm_bridge *cxl_nvb;
> - struct device *dev;
> - int rc;
> -
> - rc = cxl_pmem_region_alloc(cxlr);
> - if (rc)
> - return rc;
> - cxlr_pmem = cxlr->cxlr_pmem;
> - cxl_nvb = cxlr->cxl_nvb;
> -
> - dev = &cxlr_pmem->dev;
> - rc = dev_set_name(dev, "pmem_region%d", cxlr->id);
> - if (rc)
> - goto err;
> -
> - rc = device_add(dev);
> - if (rc)
> - goto err;
> -
> - dev_dbg(&cxlr->dev, "%s: register %s\n", dev_name(dev->parent),
> - dev_name(dev));
> -
> - scoped_guard(device, &cxl_nvb->dev) {
> - if (cxl_nvb->dev.driver)
> - rc = devm_add_action_or_reset(&cxl_nvb->dev,
> - cxlr_pmem_unregister,
> - cxlr_pmem);
> - else
> - rc = -ENXIO;
> - }
> -
> - if (rc)
> - goto err_bridge;
> -
> - /* @cxlr carries a reference on @cxl_nvb until cxlr_release_nvdimm */
> - return devm_add_action_or_reset(&cxlr->dev, cxlr_release_nvdimm, cxlr);
> -
> -err:
> - put_device(dev);
> -err_bridge:
> - put_device(&cxl_nvb->dev);
> - cxlr->cxl_nvb = NULL;
> - return rc;
> -}
> -
> static void cxlr_dax_unregister(void *_cxlr_dax)
> {
> struct cxl_dax_region *cxlr_dax = _cxlr_dax;
> diff --git a/drivers/cxl/core/region_pmem.c b/drivers/cxl/core/region_pmem.c
> new file mode 100644
> index 000000000000..138c373a5700
> --- /dev/null
> +++ b/drivers/cxl/core/region_pmem.c
> @@ -0,0 +1,189 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +/* Copyright(c) 2022 Intel Corporation. All rights reserved. */
> +#include <linux/device.h>
> +#include <linux/slab.h>
> +#include <cxlmem.h>
> +#include <cxl.h>
> +#include "core.h"
> +
> +static void cxl_pmem_region_release(struct device *dev)
> +{
> + struct cxl_pmem_region *cxlr_pmem = to_cxl_pmem_region(dev);
> + int i;
> +
> + for (i = 0; i < cxlr_pmem->nr_mappings; i++) {
> + struct cxl_memdev *cxlmd = cxlr_pmem->mapping[i].cxlmd;
> +
> + put_device(&cxlmd->dev);
> + }
> +
> + kfree(cxlr_pmem);
> +}
> +
> +static const struct attribute_group *cxl_pmem_region_attribute_groups[] = {
> + &cxl_base_attribute_group,
> + NULL
> +};
> +
> +const struct device_type cxl_pmem_region_type = {
> + .name = "cxl_pmem_region",
> + .release = cxl_pmem_region_release,
> + .groups = cxl_pmem_region_attribute_groups,
> +};
> +bool is_cxl_pmem_region(struct device *dev)
> +{
> + return dev->type == &cxl_pmem_region_type;
> +}
> +EXPORT_SYMBOL_NS_GPL(is_cxl_pmem_region, "CXL");
> +
> +struct cxl_pmem_region *to_cxl_pmem_region(struct device *dev)
> +{
> + if (dev_WARN_ONCE(dev, !is_cxl_pmem_region(dev),
> + "not a cxl_pmem_region device\n"))
> + return NULL;
> + return container_of(dev, struct cxl_pmem_region, dev);
> +}
> +EXPORT_SYMBOL_NS_GPL(to_cxl_pmem_region, "CXL");
> +static struct lock_class_key cxl_pmem_region_key;
> +
> +static int cxl_pmem_region_alloc(struct cxl_region *cxlr)
> +{
> + struct cxl_region_params *p = &cxlr->params;
> + struct cxl_nvdimm_bridge *cxl_nvb;
> + struct device *dev;
> + int i;
> +
> + guard(rwsem_read)(&cxl_rwsem.region);
> + if (p->state != CXL_CONFIG_COMMIT)
> + return -ENXIO;
> +
> + struct cxl_pmem_region *cxlr_pmem __free(kfree) =
> + kzalloc_flex(*cxlr_pmem, mapping, p->nr_targets);
> + if (!cxlr_pmem)
> + return -ENOMEM;
> +
> + cxlr_pmem->hpa_range.start = p->res->start;
> + cxlr_pmem->hpa_range.end = p->res->end;
> +
> + /* Snapshot the region configuration underneath the cxl_rwsem.region */
> + cxlr_pmem->nr_mappings = p->nr_targets;
> + for (i = 0; i < p->nr_targets; i++) {
> + struct cxl_endpoint_decoder *cxled = p->targets[i];
> + struct cxl_memdev *cxlmd = cxled_to_memdev(cxled);
> + struct cxl_pmem_region_mapping *m = &cxlr_pmem->mapping[i];
> +
> + /*
> + * Regions never span CXL root devices, so by definition the
> + * bridge for one device is the same for all.
> + */
> + if (i == 0) {
> + cxl_nvb = cxl_find_nvdimm_bridge(cxlmd->endpoint);
> + if (!cxl_nvb)
> + return -ENODEV;
> + cxlr->cxl_nvb = cxl_nvb;
> + }
> + m->cxlmd = cxlmd;
> + get_device(&cxlmd->dev);
> + m->start = cxled->dpa_res->start;
> + m->size = resource_size(cxled->dpa_res);
> + m->position = i;
> + }
> +
> + dev = &cxlr_pmem->dev;
> + device_initialize(dev);
> + lockdep_set_class(&dev->mutex, &cxl_pmem_region_key);
> + device_set_pm_not_required(dev);
> + dev->parent = &cxlr->dev;
> + dev->bus = &cxl_bus_type;
> + dev->type = &cxl_pmem_region_type;
> + cxlr_pmem->cxlr = cxlr;
> + cxlr->cxlr_pmem = no_free_ptr(cxlr_pmem);
> +
> + return 0;
> +}
> +
> +static void cxlr_pmem_unregister(void *_cxlr_pmem)
> +{
> + struct cxl_pmem_region *cxlr_pmem = _cxlr_pmem;
> + struct cxl_region *cxlr = cxlr_pmem->cxlr;
> + struct cxl_nvdimm_bridge *cxl_nvb = cxlr->cxl_nvb;
> +
> + /*
> + * Either the bridge is in ->remove() context under the device_lock(),
> + * or cxlr_release_nvdimm() is cancelling the bridge's release action
> + * for @cxlr_pmem and doing it itself (while manually holding the bridge
> + * lock).
> + */
> + device_lock_assert(&cxl_nvb->dev);
> + cxlr->cxlr_pmem = NULL;
> + cxlr_pmem->cxlr = NULL;
> + device_unregister(&cxlr_pmem->dev);
> +}
> +
> +static void cxlr_release_nvdimm(void *_cxlr)
> +{
> + struct cxl_region *cxlr = _cxlr;
> + struct cxl_nvdimm_bridge *cxl_nvb = cxlr->cxl_nvb;
> +
> + scoped_guard(device, &cxl_nvb->dev) {
> + if (cxlr->cxlr_pmem)
> + devm_release_action(&cxl_nvb->dev, cxlr_pmem_unregister,
> + cxlr->cxlr_pmem);
> + }
> + cxlr->cxl_nvb = NULL;
> + put_device(&cxl_nvb->dev);
> +}
> +
> +/**
> + * devm_cxl_add_pmem_region() - add a cxl_region-to-nd_region bridge
> + * @cxlr: parent CXL region for this pmem region bridge device
> + *
> + * Return: 0 on success negative error code on failure.
> + */
> +int devm_cxl_add_pmem_region(struct cxl_region *cxlr)
> +{
> + struct cxl_pmem_region *cxlr_pmem;
> + struct cxl_nvdimm_bridge *cxl_nvb;
> + struct device *dev;
> + int rc;
> +
> + rc = cxl_pmem_region_alloc(cxlr);
> + if (rc)
> + return rc;
> + cxlr_pmem = cxlr->cxlr_pmem;
> + cxl_nvb = cxlr->cxl_nvb;
> +
> + dev = &cxlr_pmem->dev;
> + rc = dev_set_name(dev, "pmem_region%d", cxlr->id);
> + if (rc)
> + goto err;
> +
> + rc = device_add(dev);
> + if (rc)
> + goto err;
> +
> + dev_dbg(&cxlr->dev, "%s: register %s\n", dev_name(dev->parent),
> + dev_name(dev));
> +
> + scoped_guard(device, &cxl_nvb->dev) {
> + if (cxl_nvb->dev.driver)
> + rc = devm_add_action_or_reset(&cxl_nvb->dev,
> + cxlr_pmem_unregister,
> + cxlr_pmem);
> + else
> + rc = -ENXIO;
> + }
> +
> + if (rc)
> + goto err_bridge;
> +
> + /* @cxlr carries a reference on @cxl_nvb until cxlr_release_nvdimm */
> + return devm_add_action_or_reset(&cxlr->dev, cxlr_release_nvdimm, cxlr);
> +
> +err:
> + put_device(dev);
> +err_bridge:
> + put_device(&cxl_nvb->dev);
> + cxlr->cxl_nvb = NULL;
> + return rc;
> +}
> --
> 2.53.0
>
On Sun, Mar 22, 2026 at 07:37:28PM -0700, Alison Schofield wrote: > On Sun, Mar 22, 2026 at 09:16:36AM -0400, Gregory Price wrote: > > core/region.c is overloaded with per-region control logic (pmem, dax, > > sysram, etc). Move the pmem region driver logic from region.c into > > region_pmem.c make it clear that this code only applies to pmem regions. > > > > No functional changes. > > This always adds region_pmem.o regardless of CONFIG_CXL_PMEM. > Are you heading towards conditionally compiling, where region_pmem.o is > omitted if !CONFIG_CXL_PMEM? > > Similar question for region_dax.o > I hadn't considered going that far, but it doesn't seem unreasonable? It just seemed natural to break out the code this way, but I don't know if adding more Kconfig cruft is worth it just to save some .o's and a smaller amount of code. The recent pmem/nvdimm Kconfig change does kind of show that we might have some bloat for some deployments which might not care about pmem - for example. ~Gregory
On Mon, Mar 23, 2026 at 01:37:49AM -0400, Gregory Price wrote: > On Sun, Mar 22, 2026 at 07:37:28PM -0700, Alison Schofield wrote: > > On Sun, Mar 22, 2026 at 09:16:36AM -0400, Gregory Price wrote: > > > core/region.c is overloaded with per-region control logic (pmem, dax, > > > sysram, etc). Move the pmem region driver logic from region.c into > > > region_pmem.c make it clear that this code only applies to pmem regions. > > > > > > No functional changes. > > > > This always adds region_pmem.o regardless of CONFIG_CXL_PMEM. > > Are you heading towards conditionally compiling, where region_pmem.o is > > omitted if !CONFIG_CXL_PMEM? > > > > Similar question for region_dax.o > > > > I hadn't considered going that far, but it doesn't seem unreasonable? > > It just seemed natural to break out the code this way, but I don't know > if adding more Kconfig cruft is worth it just to save some .o's and a > smaller amount of code. > > The recent pmem/nvdimm Kconfig change does kind of show that we might > have some bloat for some deployments which might not care about pmem - > for example. I guess my original thought was, what is your long term intent with the move? And I was thinking you are probably heading to conditionally compiling. HOWEVER, now the scaredy-cat-me is thinking, shoot, that would add test config complexity and test misses. ie. Folks make a change in region_dax.c, don't test with CONFIG_CXL_PMEM enabled, and boom. So, I lean towards keeping them all (working) together, like you have. -- Alison > > ~Gregory
On Mon, Mar 23, 2026 at 08:49:02PM -0700, Alison Schofield wrote: > On Mon, Mar 23, 2026 at 01:37:49AM -0400, Gregory Price wrote: > > > > The recent pmem/nvdimm Kconfig change does kind of show that we might > > have some bloat for some deployments which might not care about pmem - > > for example. > > I guess my original thought was, what is your long term intent with > the move? And I was thinking you are probably heading to conditionally > compiling. HOWEVER, now the scaredy-cat-me is thinking, shoot, that > would add test config complexity and test misses. ie. Folks make a > change in region_dax.c, don't test with CONFIG_CXL_PMEM enabled, and > boom. > > So, I lean towards keeping them all (working) together, like you have. > fwiw i think this is the right take. It doesn't really cost us anything to keep in, and probably prevents adding more Kconfig footguns. ~Gregory
© 2016 - 2026 Red Hat, Inc.