linux/drivers/cxl/pmem.c

466 lines
12 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/* Copyright(c) 2021 Intel Corporation. All rights reserved. */
#include <linux/libnvdimm.h>
#include <linux/unaligned.h>
#include <linux/device.h>
#include <linux/module.h>
#include <linux/ndctl.h>
#include <linux/async.h>
#include <linux/slab.h>
#include <linux/nd.h>
#include "cxlmem.h"
#include "cxl.h"
static __read_mostly DECLARE_BITMAP(exclusive_cmds, CXL_MEM_COMMAND_ID_MAX);
static void clear_exclusive(void *mds)
{
clear_exclusive_cxl_commands(mds, exclusive_cmds);
}
static void unregister_nvdimm(void *nvdimm)
{
nvdimm_delete(nvdimm);
}
static ssize_t provider_show(struct device *dev, struct device_attribute *attr, char *buf)
{
struct nvdimm *nvdimm = to_nvdimm(dev);
struct cxl_nvdimm *cxl_nvd = nvdimm_provider_data(nvdimm);
return sysfs_emit(buf, "%s\n", dev_name(&cxl_nvd->dev));
}
static DEVICE_ATTR_RO(provider);
static ssize_t id_show(struct device *dev, struct device_attribute *attr, char *buf)
{
struct nvdimm *nvdimm = to_nvdimm(dev);
struct cxl_nvdimm *cxl_nvd = nvdimm_provider_data(nvdimm);
struct cxl_dev_state *cxlds = cxl_nvd->cxlmd->cxlds;
return sysfs_emit(buf, "%lld\n", cxlds->serial);
}
static DEVICE_ATTR_RO(id);
static struct attribute *cxl_dimm_attributes[] = {
&dev_attr_id.attr,
&dev_attr_provider.attr,
NULL
};
static const struct attribute_group cxl_dimm_attribute_group = {
.name = "cxl",
.attrs = cxl_dimm_attributes,
};
static const struct attribute_group *cxl_dimm_attribute_groups[] = {
&cxl_dimm_attribute_group,
NULL
};
static int cxl_nvdimm_probe(struct device *dev)
{
struct cxl_nvdimm *cxl_nvd = to_cxl_nvdimm(dev);
struct cxl_memdev *cxlmd = cxl_nvd->cxlmd;
cxl/pmem: Refactor nvdimm device registration, delete the workqueue The three objects 'struct cxl_nvdimm_bridge', 'struct cxl_nvdimm', and 'struct cxl_pmem_region' manage CXL persistent memory resources. The bridge represents base platform resources, the nvdimm represents one or more endpoints, and the region is a collection of nvdimms that contribute to an assembled address range. Their relationship is such that a region is torn down if any component endpoints are removed. All regions and endpoints are torn down if the foundational bridge device goes down. A workqueue was deployed to manage these interdependencies, but it is difficult to reason about, and fragile. A recent attempt to take the CXL root device lock in the cxl_mem driver was reported by lockdep as colliding with the flush_work() in the cxl_pmem flows. Instead of the workqueue, arrange for all pmem/nvdimm devices to be torn down immediately and hierarchically. A similar change is made to both the 'cxl_nvdimm' and 'cxl_pmem_region' objects. For bisect-ability both changes are made in the same patch which unfortunately makes the patch bigger than desired. Arrange for cxl_memdev and cxl_region to register a cxl_nvdimm and cxl_pmem_region as a devres release action of the bridge device. Additionally, include a devres release action of the cxl_memdev or cxl_region device that triggers the bridge's release action if an endpoint exits before the bridge. I.e. this allows either unplugging the bridge, or unplugging and endpoint to result in the same cleanup actions. To keep the patch smaller the cleanup of the now defunct workqueue infrastructure is saved for a follow-on patch. Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993041773.1882361.16444301376147207609.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 21:33:37 +00:00
struct cxl_nvdimm_bridge *cxl_nvb = cxlmd->cxl_nvb;
struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlmd->cxlds);
unsigned long flags = 0, cmd_mask = 0;
struct nvdimm *nvdimm;
int rc;
set_exclusive_cxl_commands(mds, exclusive_cmds);
rc = devm_add_action_or_reset(dev, clear_exclusive, mds);
if (rc)
cxl/pmem: Refactor nvdimm device registration, delete the workqueue The three objects 'struct cxl_nvdimm_bridge', 'struct cxl_nvdimm', and 'struct cxl_pmem_region' manage CXL persistent memory resources. The bridge represents base platform resources, the nvdimm represents one or more endpoints, and the region is a collection of nvdimms that contribute to an assembled address range. Their relationship is such that a region is torn down if any component endpoints are removed. All regions and endpoints are torn down if the foundational bridge device goes down. A workqueue was deployed to manage these interdependencies, but it is difficult to reason about, and fragile. A recent attempt to take the CXL root device lock in the cxl_mem driver was reported by lockdep as colliding with the flush_work() in the cxl_pmem flows. Instead of the workqueue, arrange for all pmem/nvdimm devices to be torn down immediately and hierarchically. A similar change is made to both the 'cxl_nvdimm' and 'cxl_pmem_region' objects. For bisect-ability both changes are made in the same patch which unfortunately makes the patch bigger than desired. Arrange for cxl_memdev and cxl_region to register a cxl_nvdimm and cxl_pmem_region as a devres release action of the bridge device. Additionally, include a devres release action of the cxl_memdev or cxl_region device that triggers the bridge's release action if an endpoint exits before the bridge. I.e. this allows either unplugging the bridge, or unplugging and endpoint to result in the same cleanup actions. To keep the patch smaller the cleanup of the now defunct workqueue infrastructure is saved for a follow-on patch. Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993041773.1882361.16444301376147207609.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 21:33:37 +00:00
return rc;
set_bit(NDD_LABELING, &flags);
set_bit(NDD_REGISTER_SYNC, &flags);
set_bit(ND_CMD_GET_CONFIG_SIZE, &cmd_mask);
set_bit(ND_CMD_GET_CONFIG_DATA, &cmd_mask);
set_bit(ND_CMD_SET_CONFIG_DATA, &cmd_mask);
nvdimm = __nvdimm_create(cxl_nvb->nvdimm_bus, cxl_nvd,
cxl_dimm_attribute_groups, flags,
cmd_mask, 0, NULL, cxl_nvd->dev_id,
cxl_security_ops, NULL);
cxl/pmem: Refactor nvdimm device registration, delete the workqueue The three objects 'struct cxl_nvdimm_bridge', 'struct cxl_nvdimm', and 'struct cxl_pmem_region' manage CXL persistent memory resources. The bridge represents base platform resources, the nvdimm represents one or more endpoints, and the region is a collection of nvdimms that contribute to an assembled address range. Their relationship is such that a region is torn down if any component endpoints are removed. All regions and endpoints are torn down if the foundational bridge device goes down. A workqueue was deployed to manage these interdependencies, but it is difficult to reason about, and fragile. A recent attempt to take the CXL root device lock in the cxl_mem driver was reported by lockdep as colliding with the flush_work() in the cxl_pmem flows. Instead of the workqueue, arrange for all pmem/nvdimm devices to be torn down immediately and hierarchically. A similar change is made to both the 'cxl_nvdimm' and 'cxl_pmem_region' objects. For bisect-ability both changes are made in the same patch which unfortunately makes the patch bigger than desired. Arrange for cxl_memdev and cxl_region to register a cxl_nvdimm and cxl_pmem_region as a devres release action of the bridge device. Additionally, include a devres release action of the cxl_memdev or cxl_region device that triggers the bridge's release action if an endpoint exits before the bridge. I.e. this allows either unplugging the bridge, or unplugging and endpoint to result in the same cleanup actions. To keep the patch smaller the cleanup of the now defunct workqueue infrastructure is saved for a follow-on patch. Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993041773.1882361.16444301376147207609.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 21:33:37 +00:00
if (!nvdimm)
return -ENOMEM;
dev_set_drvdata(dev, nvdimm);
cxl/pmem: Refactor nvdimm device registration, delete the workqueue The three objects 'struct cxl_nvdimm_bridge', 'struct cxl_nvdimm', and 'struct cxl_pmem_region' manage CXL persistent memory resources. The bridge represents base platform resources, the nvdimm represents one or more endpoints, and the region is a collection of nvdimms that contribute to an assembled address range. Their relationship is such that a region is torn down if any component endpoints are removed. All regions and endpoints are torn down if the foundational bridge device goes down. A workqueue was deployed to manage these interdependencies, but it is difficult to reason about, and fragile. A recent attempt to take the CXL root device lock in the cxl_mem driver was reported by lockdep as colliding with the flush_work() in the cxl_pmem flows. Instead of the workqueue, arrange for all pmem/nvdimm devices to be torn down immediately and hierarchically. A similar change is made to both the 'cxl_nvdimm' and 'cxl_pmem_region' objects. For bisect-ability both changes are made in the same patch which unfortunately makes the patch bigger than desired. Arrange for cxl_memdev and cxl_region to register a cxl_nvdimm and cxl_pmem_region as a devres release action of the bridge device. Additionally, include a devres release action of the cxl_memdev or cxl_region device that triggers the bridge's release action if an endpoint exits before the bridge. I.e. this allows either unplugging the bridge, or unplugging and endpoint to result in the same cleanup actions. To keep the patch smaller the cleanup of the now defunct workqueue infrastructure is saved for a follow-on patch. Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993041773.1882361.16444301376147207609.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 21:33:37 +00:00
return devm_add_action_or_reset(dev, unregister_nvdimm, nvdimm);
}
static struct cxl_driver cxl_nvdimm_driver = {
.name = "cxl_nvdimm",
.probe = cxl_nvdimm_probe,
.id = CXL_DEVICE_NVDIMM,
.drv = {
.suppress_bind_attrs = true,
},
};
static int cxl_pmem_get_config_size(struct cxl_memdev_state *mds,
struct nd_cmd_get_config_size *cmd,
unsigned int buf_len)
{
struct cxl_mailbox *cxl_mbox = &mds->cxlds.cxl_mbox;
if (sizeof(*cmd) > buf_len)
return -EINVAL;
*cmd = (struct nd_cmd_get_config_size){
.config_size = mds->lsa_size,
.max_xfer =
cxl_mbox->payload_size - sizeof(struct cxl_mbox_set_lsa),
};
return 0;
}
static int cxl_pmem_get_config_data(struct cxl_memdev_state *mds,
struct nd_cmd_get_config_data_hdr *cmd,
unsigned int buf_len)
{
struct cxl_mailbox *cxl_mbox = &mds->cxlds.cxl_mbox;
struct cxl_mbox_get_lsa get_lsa;
struct cxl_mbox_cmd mbox_cmd;
int rc;
if (sizeof(*cmd) > buf_len)
return -EINVAL;
if (struct_size(cmd, out_buf, cmd->in_length) > buf_len)
return -EINVAL;
get_lsa = (struct cxl_mbox_get_lsa) {
.offset = cpu_to_le32(cmd->in_offset),
.length = cpu_to_le32(cmd->in_length),
};
mbox_cmd = (struct cxl_mbox_cmd) {
.opcode = CXL_MBOX_OP_GET_LSA,
.payload_in = &get_lsa,
.size_in = sizeof(get_lsa),
.size_out = cmd->in_length,
.payload_out = cmd->out_buf,
};
rc = cxl_internal_send_cmd(cxl_mbox, &mbox_cmd);
cmd->status = 0;
return rc;
}
static int cxl_pmem_set_config_data(struct cxl_memdev_state *mds,
struct nd_cmd_set_config_hdr *cmd,
unsigned int buf_len)
{
struct cxl_mailbox *cxl_mbox = &mds->cxlds.cxl_mbox;
struct cxl_mbox_set_lsa *set_lsa;
struct cxl_mbox_cmd mbox_cmd;
int rc;
if (sizeof(*cmd) > buf_len)
return -EINVAL;
/* 4-byte status follows the input data in the payload */
if (size_add(struct_size(cmd, in_buf, cmd->in_length), 4) > buf_len)
return -EINVAL;
set_lsa =
kvzalloc(struct_size(set_lsa, data, cmd->in_length), GFP_KERNEL);
if (!set_lsa)
return -ENOMEM;
*set_lsa = (struct cxl_mbox_set_lsa) {
.offset = cpu_to_le32(cmd->in_offset),
};
memcpy(set_lsa->data, cmd->in_buf, cmd->in_length);
mbox_cmd = (struct cxl_mbox_cmd) {
.opcode = CXL_MBOX_OP_SET_LSA,
.payload_in = set_lsa,
.size_in = struct_size(set_lsa, data, cmd->in_length),
};
rc = cxl_internal_send_cmd(cxl_mbox, &mbox_cmd);
/*
* Set "firmware" status (4-packed bytes at the end of the input
* payload.
*/
put_unaligned(0, (u32 *) &cmd->in_buf[cmd->in_length]);
kvfree(set_lsa);
return rc;
}
static int cxl_pmem_nvdimm_ctl(struct nvdimm *nvdimm, unsigned int cmd,
void *buf, unsigned int buf_len)
{
struct cxl_nvdimm *cxl_nvd = nvdimm_provider_data(nvdimm);
unsigned long cmd_mask = nvdimm_cmd_mask(nvdimm);
struct cxl_memdev *cxlmd = cxl_nvd->cxlmd;
struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlmd->cxlds);
if (!test_bit(cmd, &cmd_mask))
return -ENOTTY;
switch (cmd) {
case ND_CMD_GET_CONFIG_SIZE:
return cxl_pmem_get_config_size(mds, buf, buf_len);
case ND_CMD_GET_CONFIG_DATA:
return cxl_pmem_get_config_data(mds, buf, buf_len);
case ND_CMD_SET_CONFIG_DATA:
return cxl_pmem_set_config_data(mds, buf, buf_len);
default:
return -ENOTTY;
}
}
static int cxl_pmem_ctl(struct nvdimm_bus_descriptor *nd_desc,
struct nvdimm *nvdimm, unsigned int cmd, void *buf,
unsigned int buf_len, int *cmd_rc)
{
/*
* No firmware response to translate, let the transport error
* code take precedence.
*/
*cmd_rc = 0;
if (!nvdimm)
return -ENOTTY;
return cxl_pmem_nvdimm_ctl(nvdimm, cmd, buf, buf_len);
}
cxl/pmem: Fix nvdimm unregistration when cxl_pmem driver is absent The cxl_pmem.ko module houses the driver for both cxl_nvdimm_bridge objects and cxl_nvdimm objects. When the core creates a cxl_nvdimm it arranges for it to be autoremoved when the bridge goes down. However, if the bridge never initialized because the cxl_pmem.ko module never loaded, it sets up a the following crash scenario: BUG: kernel NULL pointer dereference, address: 0000000000000478 [..] RIP: 0010:cxl_nvdimm_probe+0x99/0x140 [cxl_pmem] [..] Call Trace: <TASK> cxl_bus_probe+0x17/0x50 [cxl_core] really_probe+0xde/0x380 __driver_probe_device+0x78/0x170 driver_probe_device+0x1f/0x90 __driver_attach+0xd2/0x1c0 bus_for_each_dev+0x79/0xc0 bus_add_driver+0x1b1/0x200 driver_register+0x89/0xe0 cxl_pmem_init+0x50/0xff0 [cxl_pmem] It turns out the recent rework to simplify nvdimm probing obviated the need to unregister cxl_nvdimm objects at cxl_nvdimm_bridge ->remove() time. Leave the cxl_nvdimm device registered until the hosting cxl_memdev departs. The alternative is that the cxl_memdev needs to be reattached whenever the cxl_nvdimm_bridge attach state cycles, which is awkward and unnecessary. The only requirement is to make sure that when the cxl_nvdimm_bridge goes away any dependent cxl_nvdimm objects are shutdown. Handle that in unregister_nvdimm_bus(). With these registration entanglements removed there is no longer a need to pre-load the cxl_pmem module in cxl_acpi. Fixes: cb9cfff82f6a ("cxl/acpi: Simplify cxl_nvdimm_bridge probing") Reported-by: Gregory Price <gregory.price@memverge.com> Debugged-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Tested-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com> Link: https://lore.kernel.org/r/167426077263.3955046.9695309346988027311.stgit@dwillia2-xfh.jf.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2023-01-21 00:26:12 +00:00
static int detach_nvdimm(struct device *dev, void *data)
{
struct cxl_nvdimm *cxl_nvd;
bool release = false;
if (!is_cxl_nvdimm(dev))
return 0;
scoped_guard(device, dev) {
if (dev->driver) {
cxl_nvd = to_cxl_nvdimm(dev);
if (cxl_nvd->cxlmd && cxl_nvd->cxlmd->cxl_nvb == data)
release = true;
}
}
cxl/pmem: Fix nvdimm unregistration when cxl_pmem driver is absent The cxl_pmem.ko module houses the driver for both cxl_nvdimm_bridge objects and cxl_nvdimm objects. When the core creates a cxl_nvdimm it arranges for it to be autoremoved when the bridge goes down. However, if the bridge never initialized because the cxl_pmem.ko module never loaded, it sets up a the following crash scenario: BUG: kernel NULL pointer dereference, address: 0000000000000478 [..] RIP: 0010:cxl_nvdimm_probe+0x99/0x140 [cxl_pmem] [..] Call Trace: <TASK> cxl_bus_probe+0x17/0x50 [cxl_core] really_probe+0xde/0x380 __driver_probe_device+0x78/0x170 driver_probe_device+0x1f/0x90 __driver_attach+0xd2/0x1c0 bus_for_each_dev+0x79/0xc0 bus_add_driver+0x1b1/0x200 driver_register+0x89/0xe0 cxl_pmem_init+0x50/0xff0 [cxl_pmem] It turns out the recent rework to simplify nvdimm probing obviated the need to unregister cxl_nvdimm objects at cxl_nvdimm_bridge ->remove() time. Leave the cxl_nvdimm device registered until the hosting cxl_memdev departs. The alternative is that the cxl_memdev needs to be reattached whenever the cxl_nvdimm_bridge attach state cycles, which is awkward and unnecessary. The only requirement is to make sure that when the cxl_nvdimm_bridge goes away any dependent cxl_nvdimm objects are shutdown. Handle that in unregister_nvdimm_bus(). With these registration entanglements removed there is no longer a need to pre-load the cxl_pmem module in cxl_acpi. Fixes: cb9cfff82f6a ("cxl/acpi: Simplify cxl_nvdimm_bridge probing") Reported-by: Gregory Price <gregory.price@memverge.com> Debugged-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Tested-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com> Link: https://lore.kernel.org/r/167426077263.3955046.9695309346988027311.stgit@dwillia2-xfh.jf.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2023-01-21 00:26:12 +00:00
if (release)
device_release_driver(dev);
return 0;
}
cxl/pmem: Refactor nvdimm device registration, delete the workqueue The three objects 'struct cxl_nvdimm_bridge', 'struct cxl_nvdimm', and 'struct cxl_pmem_region' manage CXL persistent memory resources. The bridge represents base platform resources, the nvdimm represents one or more endpoints, and the region is a collection of nvdimms that contribute to an assembled address range. Their relationship is such that a region is torn down if any component endpoints are removed. All regions and endpoints are torn down if the foundational bridge device goes down. A workqueue was deployed to manage these interdependencies, but it is difficult to reason about, and fragile. A recent attempt to take the CXL root device lock in the cxl_mem driver was reported by lockdep as colliding with the flush_work() in the cxl_pmem flows. Instead of the workqueue, arrange for all pmem/nvdimm devices to be torn down immediately and hierarchically. A similar change is made to both the 'cxl_nvdimm' and 'cxl_pmem_region' objects. For bisect-ability both changes are made in the same patch which unfortunately makes the patch bigger than desired. Arrange for cxl_memdev and cxl_region to register a cxl_nvdimm and cxl_pmem_region as a devres release action of the bridge device. Additionally, include a devres release action of the cxl_memdev or cxl_region device that triggers the bridge's release action if an endpoint exits before the bridge. I.e. this allows either unplugging the bridge, or unplugging and endpoint to result in the same cleanup actions. To keep the patch smaller the cleanup of the now defunct workqueue infrastructure is saved for a follow-on patch. Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993041773.1882361.16444301376147207609.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 21:33:37 +00:00
static void unregister_nvdimm_bus(void *_cxl_nvb)
{
struct cxl_nvdimm_bridge *cxl_nvb = _cxl_nvb;
struct nvdimm_bus *nvdimm_bus = cxl_nvb->nvdimm_bus;
cxl/pmem: Fix nvdimm unregistration when cxl_pmem driver is absent The cxl_pmem.ko module houses the driver for both cxl_nvdimm_bridge objects and cxl_nvdimm objects. When the core creates a cxl_nvdimm it arranges for it to be autoremoved when the bridge goes down. However, if the bridge never initialized because the cxl_pmem.ko module never loaded, it sets up a the following crash scenario: BUG: kernel NULL pointer dereference, address: 0000000000000478 [..] RIP: 0010:cxl_nvdimm_probe+0x99/0x140 [cxl_pmem] [..] Call Trace: <TASK> cxl_bus_probe+0x17/0x50 [cxl_core] really_probe+0xde/0x380 __driver_probe_device+0x78/0x170 driver_probe_device+0x1f/0x90 __driver_attach+0xd2/0x1c0 bus_for_each_dev+0x79/0xc0 bus_add_driver+0x1b1/0x200 driver_register+0x89/0xe0 cxl_pmem_init+0x50/0xff0 [cxl_pmem] It turns out the recent rework to simplify nvdimm probing obviated the need to unregister cxl_nvdimm objects at cxl_nvdimm_bridge ->remove() time. Leave the cxl_nvdimm device registered until the hosting cxl_memdev departs. The alternative is that the cxl_memdev needs to be reattached whenever the cxl_nvdimm_bridge attach state cycles, which is awkward and unnecessary. The only requirement is to make sure that when the cxl_nvdimm_bridge goes away any dependent cxl_nvdimm objects are shutdown. Handle that in unregister_nvdimm_bus(). With these registration entanglements removed there is no longer a need to pre-load the cxl_pmem module in cxl_acpi. Fixes: cb9cfff82f6a ("cxl/acpi: Simplify cxl_nvdimm_bridge probing") Reported-by: Gregory Price <gregory.price@memverge.com> Debugged-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Tested-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com> Link: https://lore.kernel.org/r/167426077263.3955046.9695309346988027311.stgit@dwillia2-xfh.jf.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2023-01-21 00:26:12 +00:00
bus_for_each_dev(&cxl_bus_type, NULL, cxl_nvb, detach_nvdimm);
cxl/pmem: Refactor nvdimm device registration, delete the workqueue The three objects 'struct cxl_nvdimm_bridge', 'struct cxl_nvdimm', and 'struct cxl_pmem_region' manage CXL persistent memory resources. The bridge represents base platform resources, the nvdimm represents one or more endpoints, and the region is a collection of nvdimms that contribute to an assembled address range. Their relationship is such that a region is torn down if any component endpoints are removed. All regions and endpoints are torn down if the foundational bridge device goes down. A workqueue was deployed to manage these interdependencies, but it is difficult to reason about, and fragile. A recent attempt to take the CXL root device lock in the cxl_mem driver was reported by lockdep as colliding with the flush_work() in the cxl_pmem flows. Instead of the workqueue, arrange for all pmem/nvdimm devices to be torn down immediately and hierarchically. A similar change is made to both the 'cxl_nvdimm' and 'cxl_pmem_region' objects. For bisect-ability both changes are made in the same patch which unfortunately makes the patch bigger than desired. Arrange for cxl_memdev and cxl_region to register a cxl_nvdimm and cxl_pmem_region as a devres release action of the bridge device. Additionally, include a devres release action of the cxl_memdev or cxl_region device that triggers the bridge's release action if an endpoint exits before the bridge. I.e. this allows either unplugging the bridge, or unplugging and endpoint to result in the same cleanup actions. To keep the patch smaller the cleanup of the now defunct workqueue infrastructure is saved for a follow-on patch. Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993041773.1882361.16444301376147207609.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 21:33:37 +00:00
cxl_nvb->nvdimm_bus = NULL;
nvdimm_bus_unregister(nvdimm_bus);
}
static int cxl_nvdimm_bridge_probe(struct device *dev)
{
struct cxl_nvdimm_bridge *cxl_nvb = to_cxl_nvdimm_bridge(dev);
cxl/pmem: Refactor nvdimm device registration, delete the workqueue The three objects 'struct cxl_nvdimm_bridge', 'struct cxl_nvdimm', and 'struct cxl_pmem_region' manage CXL persistent memory resources. The bridge represents base platform resources, the nvdimm represents one or more endpoints, and the region is a collection of nvdimms that contribute to an assembled address range. Their relationship is such that a region is torn down if any component endpoints are removed. All regions and endpoints are torn down if the foundational bridge device goes down. A workqueue was deployed to manage these interdependencies, but it is difficult to reason about, and fragile. A recent attempt to take the CXL root device lock in the cxl_mem driver was reported by lockdep as colliding with the flush_work() in the cxl_pmem flows. Instead of the workqueue, arrange for all pmem/nvdimm devices to be torn down immediately and hierarchically. A similar change is made to both the 'cxl_nvdimm' and 'cxl_pmem_region' objects. For bisect-ability both changes are made in the same patch which unfortunately makes the patch bigger than desired. Arrange for cxl_memdev and cxl_region to register a cxl_nvdimm and cxl_pmem_region as a devres release action of the bridge device. Additionally, include a devres release action of the cxl_memdev or cxl_region device that triggers the bridge's release action if an endpoint exits before the bridge. I.e. this allows either unplugging the bridge, or unplugging and endpoint to result in the same cleanup actions. To keep the patch smaller the cleanup of the now defunct workqueue infrastructure is saved for a follow-on patch. Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993041773.1882361.16444301376147207609.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 21:33:37 +00:00
cxl_nvb->nd_desc = (struct nvdimm_bus_descriptor) {
.provider_name = "CXL",
.module = THIS_MODULE,
.ndctl = cxl_pmem_ctl,
};
cxl/pmem: Refactor nvdimm device registration, delete the workqueue The three objects 'struct cxl_nvdimm_bridge', 'struct cxl_nvdimm', and 'struct cxl_pmem_region' manage CXL persistent memory resources. The bridge represents base platform resources, the nvdimm represents one or more endpoints, and the region is a collection of nvdimms that contribute to an assembled address range. Their relationship is such that a region is torn down if any component endpoints are removed. All regions and endpoints are torn down if the foundational bridge device goes down. A workqueue was deployed to manage these interdependencies, but it is difficult to reason about, and fragile. A recent attempt to take the CXL root device lock in the cxl_mem driver was reported by lockdep as colliding with the flush_work() in the cxl_pmem flows. Instead of the workqueue, arrange for all pmem/nvdimm devices to be torn down immediately and hierarchically. A similar change is made to both the 'cxl_nvdimm' and 'cxl_pmem_region' objects. For bisect-ability both changes are made in the same patch which unfortunately makes the patch bigger than desired. Arrange for cxl_memdev and cxl_region to register a cxl_nvdimm and cxl_pmem_region as a devres release action of the bridge device. Additionally, include a devres release action of the cxl_memdev or cxl_region device that triggers the bridge's release action if an endpoint exits before the bridge. I.e. this allows either unplugging the bridge, or unplugging and endpoint to result in the same cleanup actions. To keep the patch smaller the cleanup of the now defunct workqueue infrastructure is saved for a follow-on patch. Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993041773.1882361.16444301376147207609.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 21:33:37 +00:00
cxl_nvb->nvdimm_bus =
nvdimm_bus_register(&cxl_nvb->dev, &cxl_nvb->nd_desc);
cxl/pmem: Refactor nvdimm device registration, delete the workqueue The three objects 'struct cxl_nvdimm_bridge', 'struct cxl_nvdimm', and 'struct cxl_pmem_region' manage CXL persistent memory resources. The bridge represents base platform resources, the nvdimm represents one or more endpoints, and the region is a collection of nvdimms that contribute to an assembled address range. Their relationship is such that a region is torn down if any component endpoints are removed. All regions and endpoints are torn down if the foundational bridge device goes down. A workqueue was deployed to manage these interdependencies, but it is difficult to reason about, and fragile. A recent attempt to take the CXL root device lock in the cxl_mem driver was reported by lockdep as colliding with the flush_work() in the cxl_pmem flows. Instead of the workqueue, arrange for all pmem/nvdimm devices to be torn down immediately and hierarchically. A similar change is made to both the 'cxl_nvdimm' and 'cxl_pmem_region' objects. For bisect-ability both changes are made in the same patch which unfortunately makes the patch bigger than desired. Arrange for cxl_memdev and cxl_region to register a cxl_nvdimm and cxl_pmem_region as a devres release action of the bridge device. Additionally, include a devres release action of the cxl_memdev or cxl_region device that triggers the bridge's release action if an endpoint exits before the bridge. I.e. this allows either unplugging the bridge, or unplugging and endpoint to result in the same cleanup actions. To keep the patch smaller the cleanup of the now defunct workqueue infrastructure is saved for a follow-on patch. Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993041773.1882361.16444301376147207609.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 21:33:37 +00:00
if (!cxl_nvb->nvdimm_bus)
return -ENOMEM;
cxl/pmem: Refactor nvdimm device registration, delete the workqueue The three objects 'struct cxl_nvdimm_bridge', 'struct cxl_nvdimm', and 'struct cxl_pmem_region' manage CXL persistent memory resources. The bridge represents base platform resources, the nvdimm represents one or more endpoints, and the region is a collection of nvdimms that contribute to an assembled address range. Their relationship is such that a region is torn down if any component endpoints are removed. All regions and endpoints are torn down if the foundational bridge device goes down. A workqueue was deployed to manage these interdependencies, but it is difficult to reason about, and fragile. A recent attempt to take the CXL root device lock in the cxl_mem driver was reported by lockdep as colliding with the flush_work() in the cxl_pmem flows. Instead of the workqueue, arrange for all pmem/nvdimm devices to be torn down immediately and hierarchically. A similar change is made to both the 'cxl_nvdimm' and 'cxl_pmem_region' objects. For bisect-ability both changes are made in the same patch which unfortunately makes the patch bigger than desired. Arrange for cxl_memdev and cxl_region to register a cxl_nvdimm and cxl_pmem_region as a devres release action of the bridge device. Additionally, include a devres release action of the cxl_memdev or cxl_region device that triggers the bridge's release action if an endpoint exits before the bridge. I.e. this allows either unplugging the bridge, or unplugging and endpoint to result in the same cleanup actions. To keep the patch smaller the cleanup of the now defunct workqueue infrastructure is saved for a follow-on patch. Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993041773.1882361.16444301376147207609.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 21:33:37 +00:00
return devm_add_action_or_reset(dev, unregister_nvdimm_bus, cxl_nvb);
}
static struct cxl_driver cxl_nvdimm_bridge_driver = {
.name = "cxl_nvdimm_bridge",
.probe = cxl_nvdimm_bridge_probe,
.id = CXL_DEVICE_NVDIMM_BRIDGE,
.drv = {
.suppress_bind_attrs = true,
},
};
static void unregister_nvdimm_region(void *nd_region)
{
nvdimm_region_delete(nd_region);
}
static void cxlr_pmem_remove_resource(void *res)
{
remove_resource(res);
}
struct cxl_pmem_region_info {
u64 offset;
u64 serial;
};
static int cxl_pmem_region_probe(struct device *dev)
{
struct nd_mapping_desc mappings[CXL_DECODER_MAX_INTERLEAVE];
struct cxl_pmem_region *cxlr_pmem = to_cxl_pmem_region(dev);
struct cxl_region *cxlr = cxlr_pmem->cxlr;
cxl/pmem: Refactor nvdimm device registration, delete the workqueue The three objects 'struct cxl_nvdimm_bridge', 'struct cxl_nvdimm', and 'struct cxl_pmem_region' manage CXL persistent memory resources. The bridge represents base platform resources, the nvdimm represents one or more endpoints, and the region is a collection of nvdimms that contribute to an assembled address range. Their relationship is such that a region is torn down if any component endpoints are removed. All regions and endpoints are torn down if the foundational bridge device goes down. A workqueue was deployed to manage these interdependencies, but it is difficult to reason about, and fragile. A recent attempt to take the CXL root device lock in the cxl_mem driver was reported by lockdep as colliding with the flush_work() in the cxl_pmem flows. Instead of the workqueue, arrange for all pmem/nvdimm devices to be torn down immediately and hierarchically. A similar change is made to both the 'cxl_nvdimm' and 'cxl_pmem_region' objects. For bisect-ability both changes are made in the same patch which unfortunately makes the patch bigger than desired. Arrange for cxl_memdev and cxl_region to register a cxl_nvdimm and cxl_pmem_region as a devres release action of the bridge device. Additionally, include a devres release action of the cxl_memdev or cxl_region device that triggers the bridge's release action if an endpoint exits before the bridge. I.e. this allows either unplugging the bridge, or unplugging and endpoint to result in the same cleanup actions. To keep the patch smaller the cleanup of the now defunct workqueue infrastructure is saved for a follow-on patch. Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993041773.1882361.16444301376147207609.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 21:33:37 +00:00
struct cxl_nvdimm_bridge *cxl_nvb = cxlr->cxl_nvb;
struct cxl_pmem_region_info *info = NULL;
struct nd_interleave_set *nd_set;
struct nd_region_desc ndr_desc;
struct cxl_nvdimm *cxl_nvd;
struct nvdimm *nvdimm;
struct resource *res;
int rc, i = 0;
memset(&mappings, 0, sizeof(mappings));
memset(&ndr_desc, 0, sizeof(ndr_desc));
res = devm_kzalloc(dev, sizeof(*res), GFP_KERNEL);
cxl/pmem: Refactor nvdimm device registration, delete the workqueue The three objects 'struct cxl_nvdimm_bridge', 'struct cxl_nvdimm', and 'struct cxl_pmem_region' manage CXL persistent memory resources. The bridge represents base platform resources, the nvdimm represents one or more endpoints, and the region is a collection of nvdimms that contribute to an assembled address range. Their relationship is such that a region is torn down if any component endpoints are removed. All regions and endpoints are torn down if the foundational bridge device goes down. A workqueue was deployed to manage these interdependencies, but it is difficult to reason about, and fragile. A recent attempt to take the CXL root device lock in the cxl_mem driver was reported by lockdep as colliding with the flush_work() in the cxl_pmem flows. Instead of the workqueue, arrange for all pmem/nvdimm devices to be torn down immediately and hierarchically. A similar change is made to both the 'cxl_nvdimm' and 'cxl_pmem_region' objects. For bisect-ability both changes are made in the same patch which unfortunately makes the patch bigger than desired. Arrange for cxl_memdev and cxl_region to register a cxl_nvdimm and cxl_pmem_region as a devres release action of the bridge device. Additionally, include a devres release action of the cxl_memdev or cxl_region device that triggers the bridge's release action if an endpoint exits before the bridge. I.e. this allows either unplugging the bridge, or unplugging and endpoint to result in the same cleanup actions. To keep the patch smaller the cleanup of the now defunct workqueue infrastructure is saved for a follow-on patch. Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993041773.1882361.16444301376147207609.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 21:33:37 +00:00
if (!res)
return -ENOMEM;
res->name = "Persistent Memory";
res->start = cxlr_pmem->hpa_range.start;
res->end = cxlr_pmem->hpa_range.end;
res->flags = IORESOURCE_MEM;
res->desc = IORES_DESC_PERSISTENT_MEMORY;
rc = insert_resource(&iomem_resource, res);
if (rc)
cxl/pmem: Refactor nvdimm device registration, delete the workqueue The three objects 'struct cxl_nvdimm_bridge', 'struct cxl_nvdimm', and 'struct cxl_pmem_region' manage CXL persistent memory resources. The bridge represents base platform resources, the nvdimm represents one or more endpoints, and the region is a collection of nvdimms that contribute to an assembled address range. Their relationship is such that a region is torn down if any component endpoints are removed. All regions and endpoints are torn down if the foundational bridge device goes down. A workqueue was deployed to manage these interdependencies, but it is difficult to reason about, and fragile. A recent attempt to take the CXL root device lock in the cxl_mem driver was reported by lockdep as colliding with the flush_work() in the cxl_pmem flows. Instead of the workqueue, arrange for all pmem/nvdimm devices to be torn down immediately and hierarchically. A similar change is made to both the 'cxl_nvdimm' and 'cxl_pmem_region' objects. For bisect-ability both changes are made in the same patch which unfortunately makes the patch bigger than desired. Arrange for cxl_memdev and cxl_region to register a cxl_nvdimm and cxl_pmem_region as a devres release action of the bridge device. Additionally, include a devres release action of the cxl_memdev or cxl_region device that triggers the bridge's release action if an endpoint exits before the bridge. I.e. this allows either unplugging the bridge, or unplugging and endpoint to result in the same cleanup actions. To keep the patch smaller the cleanup of the now defunct workqueue infrastructure is saved for a follow-on patch. Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993041773.1882361.16444301376147207609.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 21:33:37 +00:00
return rc;
rc = devm_add_action_or_reset(dev, cxlr_pmem_remove_resource, res);
if (rc)
cxl/pmem: Refactor nvdimm device registration, delete the workqueue The three objects 'struct cxl_nvdimm_bridge', 'struct cxl_nvdimm', and 'struct cxl_pmem_region' manage CXL persistent memory resources. The bridge represents base platform resources, the nvdimm represents one or more endpoints, and the region is a collection of nvdimms that contribute to an assembled address range. Their relationship is such that a region is torn down if any component endpoints are removed. All regions and endpoints are torn down if the foundational bridge device goes down. A workqueue was deployed to manage these interdependencies, but it is difficult to reason about, and fragile. A recent attempt to take the CXL root device lock in the cxl_mem driver was reported by lockdep as colliding with the flush_work() in the cxl_pmem flows. Instead of the workqueue, arrange for all pmem/nvdimm devices to be torn down immediately and hierarchically. A similar change is made to both the 'cxl_nvdimm' and 'cxl_pmem_region' objects. For bisect-ability both changes are made in the same patch which unfortunately makes the patch bigger than desired. Arrange for cxl_memdev and cxl_region to register a cxl_nvdimm and cxl_pmem_region as a devres release action of the bridge device. Additionally, include a devres release action of the cxl_memdev or cxl_region device that triggers the bridge's release action if an endpoint exits before the bridge. I.e. this allows either unplugging the bridge, or unplugging and endpoint to result in the same cleanup actions. To keep the patch smaller the cleanup of the now defunct workqueue infrastructure is saved for a follow-on patch. Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993041773.1882361.16444301376147207609.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 21:33:37 +00:00
return rc;
ndr_desc.res = res;
ndr_desc.provider_data = cxlr_pmem;
ndr_desc.numa_node = memory_add_physaddr_to_nid(res->start);
ndr_desc.target_node = phys_to_target_node(res->start);
if (ndr_desc.target_node == NUMA_NO_NODE) {
ndr_desc.target_node = ndr_desc.numa_node;
dev_dbg(&cxlr->dev, "changing target node from %d to %d",
NUMA_NO_NODE, ndr_desc.target_node);
}
nd_set = devm_kzalloc(dev, sizeof(*nd_set), GFP_KERNEL);
cxl/pmem: Refactor nvdimm device registration, delete the workqueue The three objects 'struct cxl_nvdimm_bridge', 'struct cxl_nvdimm', and 'struct cxl_pmem_region' manage CXL persistent memory resources. The bridge represents base platform resources, the nvdimm represents one or more endpoints, and the region is a collection of nvdimms that contribute to an assembled address range. Their relationship is such that a region is torn down if any component endpoints are removed. All regions and endpoints are torn down if the foundational bridge device goes down. A workqueue was deployed to manage these interdependencies, but it is difficult to reason about, and fragile. A recent attempt to take the CXL root device lock in the cxl_mem driver was reported by lockdep as colliding with the flush_work() in the cxl_pmem flows. Instead of the workqueue, arrange for all pmem/nvdimm devices to be torn down immediately and hierarchically. A similar change is made to both the 'cxl_nvdimm' and 'cxl_pmem_region' objects. For bisect-ability both changes are made in the same patch which unfortunately makes the patch bigger than desired. Arrange for cxl_memdev and cxl_region to register a cxl_nvdimm and cxl_pmem_region as a devres release action of the bridge device. Additionally, include a devres release action of the cxl_memdev or cxl_region device that triggers the bridge's release action if an endpoint exits before the bridge. I.e. this allows either unplugging the bridge, or unplugging and endpoint to result in the same cleanup actions. To keep the patch smaller the cleanup of the now defunct workqueue infrastructure is saved for a follow-on patch. Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993041773.1882361.16444301376147207609.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 21:33:37 +00:00
if (!nd_set)
return -ENOMEM;
ndr_desc.memregion = cxlr->id;
set_bit(ND_REGION_CXL, &ndr_desc.flags);
set_bit(ND_REGION_PERSIST_MEMCTRL, &ndr_desc.flags);
info = kmalloc_array(cxlr_pmem->nr_mappings, sizeof(*info), GFP_KERNEL);
cxl/pmem: Refactor nvdimm device registration, delete the workqueue The three objects 'struct cxl_nvdimm_bridge', 'struct cxl_nvdimm', and 'struct cxl_pmem_region' manage CXL persistent memory resources. The bridge represents base platform resources, the nvdimm represents one or more endpoints, and the region is a collection of nvdimms that contribute to an assembled address range. Their relationship is such that a region is torn down if any component endpoints are removed. All regions and endpoints are torn down if the foundational bridge device goes down. A workqueue was deployed to manage these interdependencies, but it is difficult to reason about, and fragile. A recent attempt to take the CXL root device lock in the cxl_mem driver was reported by lockdep as colliding with the flush_work() in the cxl_pmem flows. Instead of the workqueue, arrange for all pmem/nvdimm devices to be torn down immediately and hierarchically. A similar change is made to both the 'cxl_nvdimm' and 'cxl_pmem_region' objects. For bisect-ability both changes are made in the same patch which unfortunately makes the patch bigger than desired. Arrange for cxl_memdev and cxl_region to register a cxl_nvdimm and cxl_pmem_region as a devres release action of the bridge device. Additionally, include a devres release action of the cxl_memdev or cxl_region device that triggers the bridge's release action if an endpoint exits before the bridge. I.e. this allows either unplugging the bridge, or unplugging and endpoint to result in the same cleanup actions. To keep the patch smaller the cleanup of the now defunct workqueue infrastructure is saved for a follow-on patch. Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993041773.1882361.16444301376147207609.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 21:33:37 +00:00
if (!info)
return -ENOMEM;
for (i = 0; i < cxlr_pmem->nr_mappings; i++) {
struct cxl_pmem_region_mapping *m = &cxlr_pmem->mapping[i];
struct cxl_memdev *cxlmd = m->cxlmd;
struct cxl_dev_state *cxlds = cxlmd->cxlds;
cxl/pmem: Refactor nvdimm device registration, delete the workqueue The three objects 'struct cxl_nvdimm_bridge', 'struct cxl_nvdimm', and 'struct cxl_pmem_region' manage CXL persistent memory resources. The bridge represents base platform resources, the nvdimm represents one or more endpoints, and the region is a collection of nvdimms that contribute to an assembled address range. Their relationship is such that a region is torn down if any component endpoints are removed. All regions and endpoints are torn down if the foundational bridge device goes down. A workqueue was deployed to manage these interdependencies, but it is difficult to reason about, and fragile. A recent attempt to take the CXL root device lock in the cxl_mem driver was reported by lockdep as colliding with the flush_work() in the cxl_pmem flows. Instead of the workqueue, arrange for all pmem/nvdimm devices to be torn down immediately and hierarchically. A similar change is made to both the 'cxl_nvdimm' and 'cxl_pmem_region' objects. For bisect-ability both changes are made in the same patch which unfortunately makes the patch bigger than desired. Arrange for cxl_memdev and cxl_region to register a cxl_nvdimm and cxl_pmem_region as a devres release action of the bridge device. Additionally, include a devres release action of the cxl_memdev or cxl_region device that triggers the bridge's release action if an endpoint exits before the bridge. I.e. this allows either unplugging the bridge, or unplugging and endpoint to result in the same cleanup actions. To keep the patch smaller the cleanup of the now defunct workqueue infrastructure is saved for a follow-on patch. Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993041773.1882361.16444301376147207609.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 21:33:37 +00:00
cxl_nvd = cxlmd->cxl_nvd;
nvdimm = dev_get_drvdata(&cxl_nvd->dev);
if (!nvdimm) {
dev_dbg(dev, "[%d]: %s: no nvdimm found\n", i,
dev_name(&cxlmd->dev));
rc = -ENODEV;
goto out_nvd;
}
m->cxl_nvd = cxl_nvd;
mappings[i] = (struct nd_mapping_desc) {
.nvdimm = nvdimm,
.start = m->start,
.size = m->size,
.position = i,
};
info[i].offset = m->start;
info[i].serial = cxlds->serial;
}
ndr_desc.num_mappings = cxlr_pmem->nr_mappings;
ndr_desc.mapping = mappings;
/*
* TODO enable CXL labels which skip the need for 'interleave-set cookie'
*/
nd_set->cookie1 =
nd_fletcher64(info, sizeof(*info) * cxlr_pmem->nr_mappings, 0);
nd_set->cookie2 = nd_set->cookie1;
ndr_desc.nd_set = nd_set;
cxlr_pmem->nd_region =
nvdimm_pmem_region_create(cxl_nvb->nvdimm_bus, &ndr_desc);
if (!cxlr_pmem->nd_region) {
rc = -ENOMEM;
goto out_nvd;
}
rc = devm_add_action_or_reset(dev, unregister_nvdimm_region,
cxlr_pmem->nd_region);
out_nvd:
kfree(info);
return rc;
}
static struct cxl_driver cxl_pmem_region_driver = {
.name = "cxl_pmem_region",
.probe = cxl_pmem_region_probe,
.id = CXL_DEVICE_PMEM_REGION,
.drv = {
.suppress_bind_attrs = true,
},
};
static __init int cxl_pmem_init(void)
{
int rc;
set_bit(CXL_MEM_COMMAND_ID_SET_SHUTDOWN_STATE, exclusive_cmds);
set_bit(CXL_MEM_COMMAND_ID_SET_LSA, exclusive_cmds);
rc = cxl_driver_register(&cxl_nvdimm_bridge_driver);
if (rc)
return rc;
rc = cxl_driver_register(&cxl_nvdimm_driver);
if (rc)
goto err_nvdimm;
rc = cxl_driver_register(&cxl_pmem_region_driver);
if (rc)
goto err_region;
return 0;
err_region:
cxl_driver_unregister(&cxl_nvdimm_driver);
err_nvdimm:
cxl_driver_unregister(&cxl_nvdimm_bridge_driver);
return rc;
}
static __exit void cxl_pmem_exit(void)
{
cxl_driver_unregister(&cxl_pmem_region_driver);
cxl_driver_unregister(&cxl_nvdimm_driver);
cxl_driver_unregister(&cxl_nvdimm_bridge_driver);
}
MODULE_DESCRIPTION("CXL PMEM: Persistent Memory Support");
MODULE_LICENSE("GPL v2");
module_init(cxl_pmem_init);
module_exit(cxl_pmem_exit);
MODULE_IMPORT_NS(CXL);
MODULE_ALIAS_CXL(CXL_DEVICE_NVDIMM_BRIDGE);
MODULE_ALIAS_CXL(CXL_DEVICE_NVDIMM);
MODULE_ALIAS_CXL(CXL_DEVICE_PMEM_REGION);