linux/drivers/cxl/mem.c

250 lines
7 KiB

cxl/mem: Add the cxl_mem driver At this point the subsystem can enumerate all CXL ports (CXL.mem decode resources in upstream switch ports and host bridges) in a system. The last mile is connecting those ports to endpoints. The cxl_mem driver connects an endpoint device to the platform CXL.mem protoctol decode-topology. At ->probe() time it walks its device-topology-ancestry and adds a CXL Port object at every Upstream Port hop until it gets to CXL root. The CXL root object is only present after a platform firmware driver registers platform CXL resources. For ACPI based platform this is managed by the ACPI0017 device and the cxl_acpi driver. The ports are registered such that disabling a given port automatically unregisters all descendant ports, and the chain can only be registered after the root is established. Given ACPI device scanning may run asynchronously compared to PCI device scanning the root driver is tasked with rescanning the bus after the root successfully probes. Conversely if any ports in a chain between the root and an endpoint becomes disconnected it subsequently triggers the endpoint to unregister. Given lock depenedencies the endpoint unregistration happens in a workqueue asynchronously. If userspace cares about synchronizing delayed work after port events the /sys/bus/cxl/flush attribute is available for that purpose. Reported-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Ben Widawsky <ben.widawsky@intel.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> [djbw: clarify changelog, rework hotplug support] Link: https://lore.kernel.org/r/164398782997.903003.9725273241627693186.stgit@dwillia2-desk3.amr.corp.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-02-04 07:18:31 -08:00
cxl/mem: Add the cxl_mem driver At this point the subsystem can enumerate all CXL ports (CXL.mem decode resources in upstream switch ports and host bridges) in a system. The last mile is connecting those ports to endpoints. The cxl_mem driver connects an endpoint device to the platform CXL.mem protoctol decode-topology. At ->probe() time it walks its device-topology-ancestry and adds a CXL Port object at every Upstream Port hop until it gets to CXL root. The CXL root object is only present after a platform firmware driver registers platform CXL resources. For ACPI based platform this is managed by the ACPI0017 device and the cxl_acpi driver. The ports are registered such that disabling a given port automatically unregisters all descendant ports, and the chain can only be registered after the root is established. Given ACPI device scanning may run asynchronously compared to PCI device scanning the root driver is tasked with rescanning the bus after the root successfully probes. Conversely if any ports in a chain between the root and an endpoint becomes disconnected it subsequently triggers the endpoint to unregister. Given lock depenedencies the endpoint unregistration happens in a workqueue asynchronously. If userspace cares about synchronizing delayed work after port events the /sys/bus/cxl/flush attribute is available for that purpose. Reported-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Ben Widawsky <ben.widawsky@intel.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> [djbw: clarify changelog, rework hotplug support] Link: https://lore.kernel.org/r/164398782997.903003.9725273241627693186.stgit@dwillia2-desk3.amr.corp.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-02-04 07:18:31 -08:00
PM: CXL: Disable suspend The CXL specification claims S3 support at a hardware level, but at a system software level there are some missing pieces. Section 9.4 (CXL 2.0) rightly claims that "CXL mem adapters may need aux power to retain memory context across S3", but there is no enumeration mechanism for the OS to determine if a given adapter has that support. Moreover the save state and resume image for the system may inadvertantly end up in a CXL device that needs to be restored before the save state is recoverable. I.e. a circular dependency that is not resolvable without a third party save-area. Arrange for the cxl_mem driver to fail S3 attempts. This still nominaly allows for suspend, but requires unbinding all CXL memory devices before the suspend to ensure the typical DRAM flow is taken. The cxl_mem unbind flow is intended to also tear down all CXL memory regions associated with a given cxl_memdev. It is reasonable to assume that any device participating in a System RAM range published in the EFI memory map is covered by aux power and save-area outside the device itself. So this restriction can be minimized in the future once pre-existing region enumeration support arrives, and perhaps a spec update to clarify if the EFI memory map is sufficent for determining the range of devices managed by platform-firmware for S3 support. Per Rafael, if the CXL configuration prevents suspend then it should fail early before tasks are frozen, and mem_sleep should stop showing 'mem' as an option [1]. Effectively CXL augments the platform suspend ->valid() op since, for example, the ACPI ops are not aware of the CXL / PCI dependencies. Given the split role of platform firmware vs OS provisioned CXL memory it is up to the cxl_mem driver to determine if the CXL configuration has elements that platform firmware may not be prepared to restore. Link: https://lore.kernel.org/r/CAJZ5v0hGVN_=3iU8OLpHY3Ak35T5+JcBM-qs8SbojKrpd0VXsA@mail.gmail.com [1] Cc: "Rafael J. Wysocki" <rafael@kernel.org> Cc: Pavel Machek <pavel@ucw.cz> Cc: Len Brown <len.brown@intel.com> Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Link: https://lore.kernel.org/r/165066828317.3907920.5690432272182042556.stgit@dwillia2-desk3.amr.corp.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-04-22 15:58:11 -07:00
cxl/mem: Add the cxl_mem driver At this point the subsystem can enumerate all CXL ports (CXL.mem decode resources in upstream switch ports and host bridges) in a system. The last mile is connecting those ports to endpoints. The cxl_mem driver connects an endpoint device to the platform CXL.mem protoctol decode-topology. At ->probe() time it walks its device-topology-ancestry and adds a CXL Port object at every Upstream Port hop until it gets to CXL root. The CXL root object is only present after a platform firmware driver registers platform CXL resources. For ACPI based platform this is managed by the ACPI0017 device and the cxl_acpi driver. The ports are registered such that disabling a given port automatically unregisters all descendant ports, and the chain can only be registered after the root is established. Given ACPI device scanning may run asynchronously compared to PCI device scanning the root driver is tasked with rescanning the bus after the root successfully probes. Conversely if any ports in a chain between the root and an endpoint becomes disconnected it subsequently triggers the endpoint to unregister. Given lock depenedencies the endpoint unregistration happens in a workqueue asynchronously. If userspace cares about synchronizing delayed work after port events the /sys/bus/cxl/flush attribute is available for that purpose. Reported-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Ben Widawsky <ben.widawsky@intel.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> [djbw: clarify changelog, rework hotplug support] Link: https://lore.kernel.org/r/164398782997.903003.9725273241627693186.stgit@dwillia2-desk3.amr.corp.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-02-04 07:18:31 -08:00
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 13:33:37 -08:00
cxl/port: Add RCD endpoint port enumeration Unlike a CXL memory expander in a VH topology that has at least one intervening 'struct cxl_port' instance between itself and the CXL root device, an RCD attaches one-level higher. For example: VH ┌──────────┐ │ ACPI0017 │ │ root0 │ └─────┬────┘ │ ┌─────┴────┐ │ dport0 │ ┌─────┤ ACPI0016 ├─────┐ │ │ port1 │ │ │ └────┬─────┘ │ │ │ │ ┌──┴───┐ ┌──┴───┐ ┌───┴──┐ │dport0│ │dport1│ │dport2│ │ RP0 │ │ RP1 │ │ RP2 │ └──────┘ └──┬───┘ └──────┘ │ ┌───┴─────┐ │endpoint0│ │ port2 │ └─────────┘ ...vs: RCH ┌──────────┐ │ ACPI0017 │ │ root0 │ └────┬─────┘ │ ┌───┴────┐ │ dport0 │ │ACPI0016│ └───┬────┘ │ ┌────┴─────┐ │endpoint0 │ │ port1 │ └──────────┘ So arrange for endpoint port in the RCH/RCD case to appear directly connected to the host-bridge in its singular role as a dport. Compare that to the VH case where the host-bridge serves a dual role as a 'cxl_dport' for the CXL root device *and* a 'cxl_port' upstream port for the Root Ports in the Root Complex that are modeled as 'cxl_dport' instances in the CXL topology. Another deviation from the VH case is that RCDs may need to look up their component registers from the Root Complex Register Block (RCRB). That platform firmware specified RCRB area is cached by the cxl_acpi driver and conveyed via the host-bridge dport to the cxl_mem driver to perform the cxl_rcrb_to_component() lookup for the endpoint port (See 9.11.8 CXL Devices Attached to an RCH for the lookup of the upstream port component registers). Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993045621.1882361.1730100141527044744.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Robert Richter <rrichter@amd.com> Reviewed-by: Jonathan Camerom <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 13:34:16 -08:00
cxl/mem: Add the cxl_mem driver At this point the subsystem can enumerate all CXL ports (CXL.mem decode resources in upstream switch ports and host bridges) in a system. The last mile is connecting those ports to endpoints. The cxl_mem driver connects an endpoint device to the platform CXL.mem protoctol decode-topology. At ->probe() time it walks its device-topology-ancestry and adds a CXL Port object at every Upstream Port hop until it gets to CXL root. The CXL root object is only present after a platform firmware driver registers platform CXL resources. For ACPI based platform this is managed by the ACPI0017 device and the cxl_acpi driver. The ports are registered such that disabling a given port automatically unregisters all descendant ports, and the chain can only be registered after the root is established. Given ACPI device scanning may run asynchronously compared to PCI device scanning the root driver is tasked with rescanning the bus after the root successfully probes. Conversely if any ports in a chain between the root and an endpoint becomes disconnected it subsequently triggers the endpoint to unregister. Given lock depenedencies the endpoint unregistration happens in a workqueue asynchronously. If userspace cares about synchronizing delayed work after port events the /sys/bus/cxl/flush attribute is available for that purpose. Reported-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Ben Widawsky <ben.widawsky@intel.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> [djbw: clarify changelog, rework hotplug support] Link: https://lore.kernel.org/r/164398782997.903003.9725273241627693186.stgit@dwillia2-desk3.amr.corp.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-02-04 07:18:31 -08:00
cxl/mem: Add the cxl_mem driver At this point the subsystem can enumerate all CXL ports (CXL.mem decode resources in upstream switch ports and host bridges) in a system. The last mile is connecting those ports to endpoints. The cxl_mem driver connects an endpoint device to the platform CXL.mem protoctol decode-topology. At ->probe() time it walks its device-topology-ancestry and adds a CXL Port object at every Upstream Port hop until it gets to CXL root. The CXL root object is only present after a platform firmware driver registers platform CXL resources. For ACPI based platform this is managed by the ACPI0017 device and the cxl_acpi driver. The ports are registered such that disabling a given port automatically unregisters all descendant ports, and the chain can only be registered after the root is established. Given ACPI device scanning may run asynchronously compared to PCI device scanning the root driver is tasked with rescanning the bus after the root successfully probes. Conversely if any ports in a chain between the root and an endpoint becomes disconnected it subsequently triggers the endpoint to unregister. Given lock depenedencies the endpoint unregistration happens in a workqueue asynchronously. If userspace cares about synchronizing delayed work after port events the /sys/bus/cxl/flush attribute is available for that purpose. Reported-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Ben Widawsky <ben.widawsky@intel.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> [djbw: clarify changelog, rework hotplug support] Link: https://lore.kernel.org/r/164398782997.903003.9725273241627693186.stgit@dwillia2-desk3.amr.corp.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-02-04 07:18:31 -08:00
cxl/mem: Add the cxl_mem driver At this point the subsystem can enumerate all CXL ports (CXL.mem decode resources in upstream switch ports and host bridges) in a system. The last mile is connecting those ports to endpoints. The cxl_mem driver connects an endpoint device to the platform CXL.mem protoctol decode-topology. At ->probe() time it walks its device-topology-ancestry and adds a CXL Port object at every Upstream Port hop until it gets to CXL root. The CXL root object is only present after a platform firmware driver registers platform CXL resources. For ACPI based platform this is managed by the ACPI0017 device and the cxl_acpi driver. The ports are registered such that disabling a given port automatically unregisters all descendant ports, and the chain can only be registered after the root is established. Given ACPI device scanning may run asynchronously compared to PCI device scanning the root driver is tasked with rescanning the bus after the root successfully probes. Conversely if any ports in a chain between the root and an endpoint becomes disconnected it subsequently triggers the endpoint to unregister. Given lock depenedencies the endpoint unregistration happens in a workqueue asynchronously. If userspace cares about synchronizing delayed work after port events the /sys/bus/cxl/flush attribute is available for that purpose. Reported-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Ben Widawsky <ben.widawsky@intel.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> [djbw: clarify changelog, rework hotplug support] Link: https://lore.kernel.org/r/164398782997.903003.9725273241627693186.stgit@dwillia2-desk3.amr.corp.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-02-04 07:18:31 -08:00
cxl/mem: Add the cxl_mem driver At this point the subsystem can enumerate all CXL ports (CXL.mem decode resources in upstream switch ports and host bridges) in a system. The last mile is connecting those ports to endpoints. The cxl_mem driver connects an endpoint device to the platform CXL.mem protoctol decode-topology. At ->probe() time it walks its device-topology-ancestry and adds a CXL Port object at every Upstream Port hop until it gets to CXL root. The CXL root object is only present after a platform firmware driver registers platform CXL resources. For ACPI based platform this is managed by the ACPI0017 device and the cxl_acpi driver. The ports are registered such that disabling a given port automatically unregisters all descendant ports, and the chain can only be registered after the root is established. Given ACPI device scanning may run asynchronously compared to PCI device scanning the root driver is tasked with rescanning the bus after the root successfully probes. Conversely if any ports in a chain between the root and an endpoint becomes disconnected it subsequently triggers the endpoint to unregister. Given lock depenedencies the endpoint unregistration happens in a workqueue asynchronously. If userspace cares about synchronizing delayed work after port events the /sys/bus/cxl/flush attribute is available for that purpose. Reported-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Ben Widawsky <ben.widawsky@intel.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> [djbw: clarify changelog, rework hotplug support] Link: https://lore.kernel.org/r/164398782997.903003.9725273241627693186.stgit@dwillia2-desk3.amr.corp.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-02-04 07:18:31 -08:00
cxl/mem: Fix no cxl_nvd during pmem region auto-assembling When CXL subsystem is auto-assembling a pmem region during cxl endpoint port probing, always hit below calltrace. BUG: kernel NULL pointer dereference, address: 0000000000000078 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page RIP: 0010:cxl_pmem_region_probe+0x22e/0x360 [cxl_pmem] Call Trace: <TASK> ? __die+0x24/0x70 ? page_fault_oops+0x82/0x160 ? do_user_addr_fault+0x65/0x6b0 ? exc_page_fault+0x7d/0x170 ? asm_exc_page_fault+0x26/0x30 ? cxl_pmem_region_probe+0x22e/0x360 [cxl_pmem] ? cxl_pmem_region_probe+0x1ac/0x360 [cxl_pmem] cxl_bus_probe+0x1b/0x60 [cxl_core] really_probe+0x173/0x410 ? __pfx___device_attach_driver+0x10/0x10 __driver_probe_device+0x80/0x170 driver_probe_device+0x1e/0x90 __device_attach_driver+0x90/0x120 bus_for_each_drv+0x84/0xe0 __device_attach+0xbc/0x1f0 bus_probe_device+0x90/0xa0 device_add+0x51c/0x710 devm_cxl_add_pmem_region+0x1b5/0x380 [cxl_core] cxl_bus_probe+0x1b/0x60 [cxl_core] The cxl_nvd of the memdev needs to be available during the pmem region probe. Currently the cxl_nvd is registered after the endpoint port probe. The endpoint probe, in the case of autoassembly of regions, can cause a pmem region probe requiring the not yet available cxl_nvd. Adjust the sequence so this dependency is met. This requires adding a port parameter to cxl_find_nvdimm_bridge() that can be used to query the ancestor root port. The endpoint port is not yet available, but will share a common ancestor with its parent, so start the query from there instead. Fixes: f17b558d6663 ("cxl/pmem: Refactor nvdimm device registration, delete the workqueue") Co-developed-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Li Ming <ming4.li@intel.com> Tested-by: Alison Schofield <alison.schofield@intel.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Reviewed-by: Alison Schofield <alison.schofield@intel.com> Link: https://patch.msgid.link/20240612064423.2567625-1-ming4.li@intel.com Signed-off-by: Dave Jiang <dave.jiang@intel.com>
2024-06-12 14:44:23 +08:00
cxl/port: Add RCD endpoint port enumeration Unlike a CXL memory expander in a VH topology that has at least one intervening 'struct cxl_port' instance between itself and the CXL root device, an RCD attaches one-level higher. For example: VH ┌──────────┐ │ ACPI0017 │ │ root0 │ └─────┬────┘ │ ┌─────┴────┐ │ dport0 │ ┌─────┤ ACPI0016 ├─────┐ │ │ port1 │ │ │ └────┬─────┘ │ │ │ │ ┌──┴───┐ ┌──┴───┐ ┌───┴──┐ │dport0│ │dport1│ │dport2│ │ RP0 │ │ RP1 │ │ RP2 │ └──────┘ └──┬───┘ └──────┘ │ ┌───┴─────┐ │endpoint0│ │ port2 │ └─────────┘ ...vs: RCH ┌──────────┐ │ ACPI0017 │ │ root0 │ └────┬─────┘ │ ┌───┴────┐ │ dport0 │ │ACPI0016│ └───┬────┘ │ ┌────┴─────┐ │endpoint0 │ │ port1 │ └──────────┘ So arrange for endpoint port in the RCH/RCD case to appear directly connected to the host-bridge in its singular role as a dport. Compare that to the VH case where the host-bridge serves a dual role as a 'cxl_dport' for the CXL root device *and* a 'cxl_port' upstream port for the Root Ports in the Root Complex that are modeled as 'cxl_dport' instances in the CXL topology. Another deviation from the VH case is that RCDs may need to look up their component registers from the Root Complex Register Block (RCRB). That platform firmware specified RCRB area is cached by the cxl_acpi driver and conveyed via the host-bridge dport to the cxl_mem driver to perform the cxl_rcrb_to_component() lookup for the endpoint port (See 9.11.8 CXL Devices Attached to an RCH for the lookup of the upstream port component registers). Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993045621.1882361.1730100141527044744.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Robert Richter <rrichter@amd.com> Reviewed-by: Jonathan Camerom <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 13:34:16 -08:00
cxl/port: Add RCD endpoint port enumeration Unlike a CXL memory expander in a VH topology that has at least one intervening 'struct cxl_port' instance between itself and the CXL root device, an RCD attaches one-level higher. For example: VH ┌──────────┐ │ ACPI0017 │ │ root0 │ └─────┬────┘ │ ┌─────┴────┐ │ dport0 │ ┌─────┤ ACPI0016 ├─────┐ │ │ port1 │ │ │ └────┬─────┘ │ │ │ │ ┌──┴───┐ ┌──┴───┐ ┌───┴──┐ │dport0│ │dport1│ │dport2│ │ RP0 │ │ RP1 │ │ RP2 │ └──────┘ └──┬───┘ └──────┘ │ ┌───┴─────┐ │endpoint0│ │ port2 │ └─────────┘ ...vs: RCH ┌──────────┐ │ ACPI0017 │ │ root0 │ └────┬─────┘ │ ┌───┴────┐ │ dport0 │ │ACPI0016│ └───┬────┘ │ ┌────┴─────┐ │endpoint0 │ │ port1 │ └──────────┘ So arrange for endpoint port in the RCH/RCD case to appear directly connected to the host-bridge in its singular role as a dport. Compare that to the VH case where the host-bridge serves a dual role as a 'cxl_dport' for the CXL root device *and* a 'cxl_port' upstream port for the Root Ports in the Root Complex that are modeled as 'cxl_dport' instances in the CXL topology. Another deviation from the VH case is that RCDs may need to look up their component registers from the Root Complex Register Block (RCRB). That platform firmware specified RCRB area is cached by the cxl_acpi driver and conveyed via the host-bridge dport to the cxl_mem driver to perform the cxl_rcrb_to_component() lookup for the endpoint port (See 9.11.8 CXL Devices Attached to an RCH for the lookup of the upstream port component registers). Tested-by: Robert Richter <rrichter@amd.com> Link: https://lore.kernel.org/r/166993045621.1882361.1730100141527044744.stgit@dwillia2-xfh.jf.intel.com Reviewed-by: Robert Richter <rrichter@amd.com> Reviewed-by: Jonathan Camerom <Jonathan.Cameron@huawei.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-12-01 13:34:16 -08:00
cxl/mem: Add the cxl_mem driver At this point the subsystem can enumerate all CXL ports (CXL.mem decode resources in upstream switch ports and host bridges) in a system. The last mile is connecting those ports to endpoints. The cxl_mem driver connects an endpoint device to the platform CXL.mem protoctol decode-topology. At ->probe() time it walks its device-topology-ancestry and adds a CXL Port object at every Upstream Port hop until it gets to CXL root. The CXL root object is only present after a platform firmware driver registers platform CXL resources. For ACPI based platform this is managed by the ACPI0017 device and the cxl_acpi driver. The ports are registered such that disabling a given port automatically unregisters all descendant ports, and the chain can only be registered after the root is established. Given ACPI device scanning may run asynchronously compared to PCI device scanning the root driver is tasked with rescanning the bus after the root successfully probes. Conversely if any ports in a chain between the root and an endpoint becomes disconnected it subsequently triggers the endpoint to unregister. Given lock depenedencies the endpoint unregistration happens in a workqueue asynchronously. If userspace cares about synchronizing delayed work after port events the /sys/bus/cxl/flush attribute is available for that purpose. Reported-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Ben Widawsky <ben.widawsky@intel.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> [djbw: clarify changelog, rework hotplug support] Link: https://lore.kernel.org/r/164398782997.903003.9725273241627693186.stgit@dwillia2-desk3.amr.corp.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-02-04 07:18:31 -08:00
PM: CXL: Disable suspend The CXL specification claims S3 support at a hardware level, but at a system software level there are some missing pieces. Section 9.4 (CXL 2.0) rightly claims that "CXL mem adapters may need aux power to retain memory context across S3", but there is no enumeration mechanism for the OS to determine if a given adapter has that support. Moreover the save state and resume image for the system may inadvertantly end up in a CXL device that needs to be restored before the save state is recoverable. I.e. a circular dependency that is not resolvable without a third party save-area. Arrange for the cxl_mem driver to fail S3 attempts. This still nominaly allows for suspend, but requires unbinding all CXL memory devices before the suspend to ensure the typical DRAM flow is taken. The cxl_mem unbind flow is intended to also tear down all CXL memory regions associated with a given cxl_memdev. It is reasonable to assume that any device participating in a System RAM range published in the EFI memory map is covered by aux power and save-area outside the device itself. So this restriction can be minimized in the future once pre-existing region enumeration support arrives, and perhaps a spec update to clarify if the EFI memory map is sufficent for determining the range of devices managed by platform-firmware for S3 support. Per Rafael, if the CXL configuration prevents suspend then it should fail early before tasks are frozen, and mem_sleep should stop showing 'mem' as an option [1]. Effectively CXL augments the platform suspend ->valid() op since, for example, the ACPI ops are not aware of the CXL / PCI dependencies. Given the split role of platform firmware vs OS provisioned CXL memory it is up to the cxl_mem driver to determine if the CXL configuration has elements that platform firmware may not be prepared to restore. Link: https://lore.kernel.org/r/CAJZ5v0hGVN_=3iU8OLpHY3Ak35T5+JcBM-qs8SbojKrpd0VXsA@mail.gmail.com [1] Cc: "Rafael J. Wysocki" <rafael@kernel.org> Cc: Pavel Machek <pavel@ucw.cz> Cc: Len Brown <len.brown@intel.com> Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Link: https://lore.kernel.org/r/165066828317.3907920.5690432272182042556.stgit@dwillia2-desk3.amr.corp.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-04-22 15:58:11 -07:00
cxl/mem: Introduce cxl_memdev_attach for CXL-dependent operation Unlike the cxl_pci class driver that opportunistically enables memory expansion with no other dependent functionality, CXL accelerator drivers have distinct PCIe-only and CXL-enhanced operation states. If CXL is available some additional coherent memory/cache operations can be enabled, otherwise traditional DMA+MMIO over PCIe/CXL.io is a fallback. This constitutes a new mode of operation where the caller of devm_cxl_add_memdev() wants to make a "go/no-go" decision about running in CXL accelerated mode or falling back to PCIe-only operation. Part of that decision making process likely also includes additional CXL-acceleration-specific resource setup. Encapsulate both of those requirements into 'struct cxl_memdev_attach' that provides a ->probe() callback. The probe callback runs in cxl_mem_probe() context, after the port topology is successfully attached for the given memdev. It supports a contract where, upon successful return from devm_cxl_add_memdev(), everything needed for CXL accelerated operation has been enabled. Additionally the presence of @cxlmd->attach indicates that the accelerator driver be detached when CXL operation ends. This conceptually makes a CXL link loss event mirror a PCIe link loss event which results in triggering the ->remove() callback of affected devices+drivers. A driver can re-attach to recover back to PCIe-only operation. Live recovery, i.e. without a ->remove()/->probe() cycle, is left as a future consideration. [ dj: Repalce with updated commit log from Dan ] Cc: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com> Reviewed-by: Ben Cheatham <benjamin.cheatham@amd.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com> Tested-by: Alejandro Lucero <alucerop@amd.com> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com> Link: https://patch.msgid.link/20251216005616.3090129-7-dan.j.williams@intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
2025-12-15 16:56:16 -08:00
cxl/edac: Add CXL memory device patrol scrub control feature CXL spec 3.2 section 8.2.10.9.11.1 describes the device patrol scrub control feature. The device patrol scrub proactively locates and makes corrections to errors in regular cycle. Allow specifying the number of hours within which the patrol scrub must be completed, subject to minimum and maximum limits reported by the device. Also allow disabling scrub allowing trade-off error rates against performance. Add support for patrol scrub control on CXL memory devices. Register with the EDAC device driver, which retrieves the scrub attribute descriptors from EDAC scrub and exposes the sysfs scrub control attributes to userspace. For example, scrub control for the CXL memory device "cxl_mem0" is exposed in /sys/bus/edac/devices/cxl_mem0/scrubX/. Additionally, add support for region-based CXL memory patrol scrub control. CXL memory regions may be interleaved across one or more CXL memory devices. For example, region-based scrub control for "cxl_region1" is exposed in /sys/bus/edac/devices/cxl_region1/scrubX/. [dj: A few formatting fixes from Jonathan] Reviewed-by: Dave Jiang <dave.jiang@intel.com> Co-developed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Shiju Jose <shiju.jose@huawei.com> Reviewed-by: Alison Schofield <alison.schofield@intel.com> Acked-by: Dan Williams <dan.j.williams@intel.com> Link: https://patch.msgid.link/20250521124749.817-4-shiju.jose@huawei.com Signed-off-by: Dave Jiang <dave.jiang@intel.com>
2025-05-21 13:47:41 +01:00
PM: CXL: Disable suspend The CXL specification claims S3 support at a hardware level, but at a system software level there are some missing pieces. Section 9.4 (CXL 2.0) rightly claims that "CXL mem adapters may need aux power to retain memory context across S3", but there is no enumeration mechanism for the OS to determine if a given adapter has that support. Moreover the save state and resume image for the system may inadvertantly end up in a CXL device that needs to be restored before the save state is recoverable. I.e. a circular dependency that is not resolvable without a third party save-area. Arrange for the cxl_mem driver to fail S3 attempts. This still nominaly allows for suspend, but requires unbinding all CXL memory devices before the suspend to ensure the typical DRAM flow is taken. The cxl_mem unbind flow is intended to also tear down all CXL memory regions associated with a given cxl_memdev. It is reasonable to assume that any device participating in a System RAM range published in the EFI memory map is covered by aux power and save-area outside the device itself. So this restriction can be minimized in the future once pre-existing region enumeration support arrives, and perhaps a spec update to clarify if the EFI memory map is sufficent for determining the range of devices managed by platform-firmware for S3 support. Per Rafael, if the CXL configuration prevents suspend then it should fail early before tasks are frozen, and mem_sleep should stop showing 'mem' as an option [1]. Effectively CXL augments the platform suspend ->valid() op since, for example, the ACPI ops are not aware of the CXL / PCI dependencies. Given the split role of platform firmware vs OS provisioned CXL memory it is up to the cxl_mem driver to determine if the CXL configuration has elements that platform firmware may not be prepared to restore. Link: https://lore.kernel.org/r/CAJZ5v0hGVN_=3iU8OLpHY3Ak35T5+JcBM-qs8SbojKrpd0VXsA@mail.gmail.com [1] Cc: "Rafael J. Wysocki" <rafael@kernel.org> Cc: Pavel Machek <pavel@ucw.cz> Cc: Len Brown <len.brown@intel.com> Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Link: https://lore.kernel.org/r/165066828317.3907920.5690432272182042556.stgit@dwillia2-desk3.amr.corp.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-04-22 15:58:11 -07:00
cxl/mem: Add the cxl_mem driver At this point the subsystem can enumerate all CXL ports (CXL.mem decode resources in upstream switch ports and host bridges) in a system. The last mile is connecting those ports to endpoints. The cxl_mem driver connects an endpoint device to the platform CXL.mem protoctol decode-topology. At ->probe() time it walks its device-topology-ancestry and adds a CXL Port object at every Upstream Port hop until it gets to CXL root. The CXL root object is only present after a platform firmware driver registers platform CXL resources. For ACPI based platform this is managed by the ACPI0017 device and the cxl_acpi driver. The ports are registered such that disabling a given port automatically unregisters all descendant ports, and the chain can only be registered after the root is established. Given ACPI device scanning may run asynchronously compared to PCI device scanning the root driver is tasked with rescanning the bus after the root successfully probes. Conversely if any ports in a chain between the root and an endpoint becomes disconnected it subsequently triggers the endpoint to unregister. Given lock depenedencies the endpoint unregistration happens in a workqueue asynchronously. If userspace cares about synchronizing delayed work after port events the /sys/bus/cxl/flush attribute is available for that purpose. Reported-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Ben Widawsky <ben.widawsky@intel.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> [djbw: clarify changelog, rework hotplug support] Link: https://lore.kernel.org/r/164398782997.903003.9725273241627693186.stgit@dwillia2-desk3.amr.corp.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-02-04 07:18:31 -08:00
cxl/mem: Arrange for always-synchronous memdev attach In preparation for CXL accelerator drivers that have a hard dependency on CXL capability initialization, arrange for cxl_mem_probe() to always run synchronous with the device_add() of cxl_memdev instances. I.e. cxl_mem_driver registration is always complete before the first memdev creation event. At present, cxl_pci does not care about the attach state of the cxl_memdev because all generic memory expansion functionality can be handled by the cxl_core. For accelerators, however, that driver needs to perform driver specific initialization if CXL is available, or execute a fallback to PCIe only operation. This synchronous attach guarantee is also needed for Soft Reserve Recovery, which is an effort that needs to assert that devices have had a chance to attach before making a go / no-go decision on proceeding with CXL subsystem initialization. By moving devm_cxl_add_memdev() to cxl_mem.ko it removes async module loading as one reason that a memdev may not be attached upon return from devm_cxl_add_memdev(). Cc: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com> Cc: Alejandro Lucero <alucerop@amd.com> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com> Tested-by: Alison Schofield <alison.schofield@intel.com> Reviewed-by: Alison Schofield <alison.schofield@intel.com> Reviewed-by: Ben Cheatham <benjamin.cheatham@amd.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com> Tested-by: Alejandro Lucero <alucerop@amd.com> Link: https://patch.msgid.link/20251216005616.3090129-3-dan.j.williams@intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
2025-12-15 16:56:12 -08:00
cxl/mem: Introduce cxl_memdev_attach for CXL-dependent operation Unlike the cxl_pci class driver that opportunistically enables memory expansion with no other dependent functionality, CXL accelerator drivers have distinct PCIe-only and CXL-enhanced operation states. If CXL is available some additional coherent memory/cache operations can be enabled, otherwise traditional DMA+MMIO over PCIe/CXL.io is a fallback. This constitutes a new mode of operation where the caller of devm_cxl_add_memdev() wants to make a "go/no-go" decision about running in CXL accelerated mode or falling back to PCIe-only operation. Part of that decision making process likely also includes additional CXL-acceleration-specific resource setup. Encapsulate both of those requirements into 'struct cxl_memdev_attach' that provides a ->probe() callback. The probe callback runs in cxl_mem_probe() context, after the port topology is successfully attached for the given memdev. It supports a contract where, upon successful return from devm_cxl_add_memdev(), everything needed for CXL accelerated operation has been enabled. Additionally the presence of @cxlmd->attach indicates that the accelerator driver be detached when CXL operation ends. This conceptually makes a CXL link loss event mirror a PCIe link loss event which results in triggering the ->remove() callback of affected devices+drivers. A driver can re-attach to recover back to PCIe-only operation. Live recovery, i.e. without a ->remove()/->probe() cycle, is left as a future consideration. [ dj: Repalce with updated commit log from Dan ] Cc: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com> Reviewed-by: Ben Cheatham <benjamin.cheatham@amd.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com> Tested-by: Alejandro Lucero <alucerop@amd.com> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com> Link: https://patch.msgid.link/20251216005616.3090129-7-dan.j.williams@intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
2025-12-15 16:56:16 -08:00
cxl/mem: Arrange for always-synchronous memdev attach In preparation for CXL accelerator drivers that have a hard dependency on CXL capability initialization, arrange for cxl_mem_probe() to always run synchronous with the device_add() of cxl_memdev instances. I.e. cxl_mem_driver registration is always complete before the first memdev creation event. At present, cxl_pci does not care about the attach state of the cxl_memdev because all generic memory expansion functionality can be handled by the cxl_core. For accelerators, however, that driver needs to perform driver specific initialization if CXL is available, or execute a fallback to PCIe only operation. This synchronous attach guarantee is also needed for Soft Reserve Recovery, which is an effort that needs to assert that devices have had a chance to attach before making a go / no-go decision on proceeding with CXL subsystem initialization. By moving devm_cxl_add_memdev() to cxl_mem.ko it removes async module loading as one reason that a memdev may not be attached upon return from devm_cxl_add_memdev(). Cc: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com> Cc: Alejandro Lucero <alucerop@amd.com> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com> Tested-by: Alison Schofield <alison.schofield@intel.com> Reviewed-by: Alison Schofield <alison.schofield@intel.com> Reviewed-by: Ben Cheatham <benjamin.cheatham@amd.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com> Tested-by: Alejandro Lucero <alucerop@amd.com> Link: https://patch.msgid.link/20251216005616.3090129-3-dan.j.williams@intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
2025-12-15 16:56:12 -08:00
cxl/mem: Introduce cxl_memdev_attach for CXL-dependent operation Unlike the cxl_pci class driver that opportunistically enables memory expansion with no other dependent functionality, CXL accelerator drivers have distinct PCIe-only and CXL-enhanced operation states. If CXL is available some additional coherent memory/cache operations can be enabled, otherwise traditional DMA+MMIO over PCIe/CXL.io is a fallback. This constitutes a new mode of operation where the caller of devm_cxl_add_memdev() wants to make a "go/no-go" decision about running in CXL accelerated mode or falling back to PCIe-only operation. Part of that decision making process likely also includes additional CXL-acceleration-specific resource setup. Encapsulate both of those requirements into 'struct cxl_memdev_attach' that provides a ->probe() callback. The probe callback runs in cxl_mem_probe() context, after the port topology is successfully attached for the given memdev. It supports a contract where, upon successful return from devm_cxl_add_memdev(), everything needed for CXL accelerated operation has been enabled. Additionally the presence of @cxlmd->attach indicates that the accelerator driver be detached when CXL operation ends. This conceptually makes a CXL link loss event mirror a PCIe link loss event which results in triggering the ->remove() callback of affected devices+drivers. A driver can re-attach to recover back to PCIe-only operation. Live recovery, i.e. without a ->remove()/->probe() cycle, is left as a future consideration. [ dj: Repalce with updated commit log from Dan ] Cc: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com> Reviewed-by: Ben Cheatham <benjamin.cheatham@amd.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com> Tested-by: Alejandro Lucero <alucerop@amd.com> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com> Link: https://patch.msgid.link/20251216005616.3090129-7-dan.j.williams@intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
2025-12-15 16:56:16 -08:00
cxl/mem: Arrange for always-synchronous memdev attach In preparation for CXL accelerator drivers that have a hard dependency on CXL capability initialization, arrange for cxl_mem_probe() to always run synchronous with the device_add() of cxl_memdev instances. I.e. cxl_mem_driver registration is always complete before the first memdev creation event. At present, cxl_pci does not care about the attach state of the cxl_memdev because all generic memory expansion functionality can be handled by the cxl_core. For accelerators, however, that driver needs to perform driver specific initialization if CXL is available, or execute a fallback to PCIe only operation. This synchronous attach guarantee is also needed for Soft Reserve Recovery, which is an effort that needs to assert that devices have had a chance to attach before making a go / no-go decision on proceeding with CXL subsystem initialization. By moving devm_cxl_add_memdev() to cxl_mem.ko it removes async module loading as one reason that a memdev may not be attached upon return from devm_cxl_add_memdev(). Cc: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com> Cc: Alejandro Lucero <alucerop@amd.com> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com> Tested-by: Alison Schofield <alison.schofield@intel.com> Reviewed-by: Alison Schofield <alison.schofield@intel.com> Reviewed-by: Ben Cheatham <benjamin.cheatham@amd.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com> Tested-by: Alejandro Lucero <alucerop@amd.com> Link: https://patch.msgid.link/20251216005616.3090129-3-dan.j.williams@intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
2025-12-15 16:56:12 -08:00
cxl/mem: Introduce cxl_memdev_attach for CXL-dependent operation Unlike the cxl_pci class driver that opportunistically enables memory expansion with no other dependent functionality, CXL accelerator drivers have distinct PCIe-only and CXL-enhanced operation states. If CXL is available some additional coherent memory/cache operations can be enabled, otherwise traditional DMA+MMIO over PCIe/CXL.io is a fallback. This constitutes a new mode of operation where the caller of devm_cxl_add_memdev() wants to make a "go/no-go" decision about running in CXL accelerated mode or falling back to PCIe-only operation. Part of that decision making process likely also includes additional CXL-acceleration-specific resource setup. Encapsulate both of those requirements into 'struct cxl_memdev_attach' that provides a ->probe() callback. The probe callback runs in cxl_mem_probe() context, after the port topology is successfully attached for the given memdev. It supports a contract where, upon successful return from devm_cxl_add_memdev(), everything needed for CXL accelerated operation has been enabled. Additionally the presence of @cxlmd->attach indicates that the accelerator driver be detached when CXL operation ends. This conceptually makes a CXL link loss event mirror a PCIe link loss event which results in triggering the ->remove() callback of affected devices+drivers. A driver can re-attach to recover back to PCIe-only operation. Live recovery, i.e. without a ->remove()/->probe() cycle, is left as a future consideration. [ dj: Repalce with updated commit log from Dan ] Cc: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com> Reviewed-by: Ben Cheatham <benjamin.cheatham@amd.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com> Tested-by: Alejandro Lucero <alucerop@amd.com> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com> Link: https://patch.msgid.link/20251216005616.3090129-7-dan.j.williams@intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
2025-12-15 16:56:16 -08:00
cxl/mem: Arrange for always-synchronous memdev attach In preparation for CXL accelerator drivers that have a hard dependency on CXL capability initialization, arrange for cxl_mem_probe() to always run synchronous with the device_add() of cxl_memdev instances. I.e. cxl_mem_driver registration is always complete before the first memdev creation event. At present, cxl_pci does not care about the attach state of the cxl_memdev because all generic memory expansion functionality can be handled by the cxl_core. For accelerators, however, that driver needs to perform driver specific initialization if CXL is available, or execute a fallback to PCIe only operation. This synchronous attach guarantee is also needed for Soft Reserve Recovery, which is an effort that needs to assert that devices have had a chance to attach before making a go / no-go decision on proceeding with CXL subsystem initialization. By moving devm_cxl_add_memdev() to cxl_mem.ko it removes async module loading as one reason that a memdev may not be attached upon return from devm_cxl_add_memdev(). Cc: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com> Cc: Alejandro Lucero <alucerop@amd.com> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com> Tested-by: Alison Schofield <alison.schofield@intel.com> Reviewed-by: Alison Schofield <alison.schofield@intel.com> Reviewed-by: Ben Cheatham <benjamin.cheatham@amd.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com> Tested-by: Alejandro Lucero <alucerop@amd.com> Link: https://patch.msgid.link/20251216005616.3090129-3-dan.j.williams@intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
2025-12-15 16:56:12 -08:00
cxl/mem: Introduce cxl_memdev_attach for CXL-dependent operation Unlike the cxl_pci class driver that opportunistically enables memory expansion with no other dependent functionality, CXL accelerator drivers have distinct PCIe-only and CXL-enhanced operation states. If CXL is available some additional coherent memory/cache operations can be enabled, otherwise traditional DMA+MMIO over PCIe/CXL.io is a fallback. This constitutes a new mode of operation where the caller of devm_cxl_add_memdev() wants to make a "go/no-go" decision about running in CXL accelerated mode or falling back to PCIe-only operation. Part of that decision making process likely also includes additional CXL-acceleration-specific resource setup. Encapsulate both of those requirements into 'struct cxl_memdev_attach' that provides a ->probe() callback. The probe callback runs in cxl_mem_probe() context, after the port topology is successfully attached for the given memdev. It supports a contract where, upon successful return from devm_cxl_add_memdev(), everything needed for CXL accelerated operation has been enabled. Additionally the presence of @cxlmd->attach indicates that the accelerator driver be detached when CXL operation ends. This conceptually makes a CXL link loss event mirror a PCIe link loss event which results in triggering the ->remove() callback of affected devices+drivers. A driver can re-attach to recover back to PCIe-only operation. Live recovery, i.e. without a ->remove()/->probe() cycle, is left as a future consideration. [ dj: Repalce with updated commit log from Dan ] Cc: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com> Reviewed-by: Ben Cheatham <benjamin.cheatham@amd.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com> Tested-by: Alejandro Lucero <alucerop@amd.com> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com> Link: https://patch.msgid.link/20251216005616.3090129-7-dan.j.williams@intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
2025-12-15 16:56:16 -08:00
cxl/mem: Arrange for always-synchronous memdev attach In preparation for CXL accelerator drivers that have a hard dependency on CXL capability initialization, arrange for cxl_mem_probe() to always run synchronous with the device_add() of cxl_memdev instances. I.e. cxl_mem_driver registration is always complete before the first memdev creation event. At present, cxl_pci does not care about the attach state of the cxl_memdev because all generic memory expansion functionality can be handled by the cxl_core. For accelerators, however, that driver needs to perform driver specific initialization if CXL is available, or execute a fallback to PCIe only operation. This synchronous attach guarantee is also needed for Soft Reserve Recovery, which is an effort that needs to assert that devices have had a chance to attach before making a go / no-go decision on proceeding with CXL subsystem initialization. By moving devm_cxl_add_memdev() to cxl_mem.ko it removes async module loading as one reason that a memdev may not be attached upon return from devm_cxl_add_memdev(). Cc: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com> Cc: Alejandro Lucero <alucerop@amd.com> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com> Tested-by: Alison Schofield <alison.schofield@intel.com> Reviewed-by: Alison Schofield <alison.schofield@intel.com> Reviewed-by: Ben Cheatham <benjamin.cheatham@amd.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com> Tested-by: Alejandro Lucero <alucerop@amd.com> Link: https://patch.msgid.link/20251216005616.3090129-3-dan.j.williams@intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
2025-12-15 16:56:12 -08:00
cxl/mem: Add the cxl_mem driver At this point the subsystem can enumerate all CXL ports (CXL.mem decode resources in upstream switch ports and host bridges) in a system. The last mile is connecting those ports to endpoints. The cxl_mem driver connects an endpoint device to the platform CXL.mem protoctol decode-topology. At ->probe() time it walks its device-topology-ancestry and adds a CXL Port object at every Upstream Port hop until it gets to CXL root. The CXL root object is only present after a platform firmware driver registers platform CXL resources. For ACPI based platform this is managed by the ACPI0017 device and the cxl_acpi driver. The ports are registered such that disabling a given port automatically unregisters all descendant ports, and the chain can only be registered after the root is established. Given ACPI device scanning may run asynchronously compared to PCI device scanning the root driver is tasked with rescanning the bus after the root successfully probes. Conversely if any ports in a chain between the root and an endpoint becomes disconnected it subsequently triggers the endpoint to unregister. Given lock depenedencies the endpoint unregistration happens in a workqueue asynchronously. If userspace cares about synchronizing delayed work after port events the /sys/bus/cxl/flush attribute is available for that purpose. Reported-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Ben Widawsky <ben.widawsky@intel.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> [djbw: clarify changelog, rework hotplug support] Link: https://lore.kernel.org/r/164398782997.903003.9725273241627693186.stgit@dwillia2-desk3.amr.corp.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-02-04 07:18:31 -08:00
cxl/mem: Arrange for always-synchronous memdev attach In preparation for CXL accelerator drivers that have a hard dependency on CXL capability initialization, arrange for cxl_mem_probe() to always run synchronous with the device_add() of cxl_memdev instances. I.e. cxl_mem_driver registration is always complete before the first memdev creation event. At present, cxl_pci does not care about the attach state of the cxl_memdev because all generic memory expansion functionality can be handled by the cxl_core. For accelerators, however, that driver needs to perform driver specific initialization if CXL is available, or execute a fallback to PCIe only operation. This synchronous attach guarantee is also needed for Soft Reserve Recovery, which is an effort that needs to assert that devices have had a chance to attach before making a go / no-go decision on proceeding with CXL subsystem initialization. By moving devm_cxl_add_memdev() to cxl_mem.ko it removes async module loading as one reason that a memdev may not be attached upon return from devm_cxl_add_memdev(). Cc: Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com> Cc: Alejandro Lucero <alucerop@amd.com> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com> Tested-by: Alison Schofield <alison.schofield@intel.com> Reviewed-by: Alison Schofield <alison.schofield@intel.com> Reviewed-by: Ben Cheatham <benjamin.cheatham@amd.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com> Tested-by: Alejandro Lucero <alucerop@amd.com> Link: https://patch.msgid.link/20251216005616.3090129-3-dan.j.williams@intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
2025-12-15 16:56:12 -08:00
cxl/mem: Add the cxl_mem driver At this point the subsystem can enumerate all CXL ports (CXL.mem decode resources in upstream switch ports and host bridges) in a system. The last mile is connecting those ports to endpoints. The cxl_mem driver connects an endpoint device to the platform CXL.mem protoctol decode-topology. At ->probe() time it walks its device-topology-ancestry and adds a CXL Port object at every Upstream Port hop until it gets to CXL root. The CXL root object is only present after a platform firmware driver registers platform CXL resources. For ACPI based platform this is managed by the ACPI0017 device and the cxl_acpi driver. The ports are registered such that disabling a given port automatically unregisters all descendant ports, and the chain can only be registered after the root is established. Given ACPI device scanning may run asynchronously compared to PCI device scanning the root driver is tasked with rescanning the bus after the root successfully probes. Conversely if any ports in a chain between the root and an endpoint becomes disconnected it subsequently triggers the endpoint to unregister. Given lock depenedencies the endpoint unregistration happens in a workqueue asynchronously. If userspace cares about synchronizing delayed work after port events the /sys/bus/cxl/flush attribute is available for that purpose. Reported-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Ben Widawsky <ben.widawsky@intel.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> [djbw: clarify changelog, rework hotplug support] Link: https://lore.kernel.org/r/164398782997.903003.9725273241627693186.stgit@dwillia2-desk3.amr.corp.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-02-04 07:18:31 -08:00
cxl/mem: Add the cxl_mem driver At this point the subsystem can enumerate all CXL ports (CXL.mem decode resources in upstream switch ports and host bridges) in a system. The last mile is connecting those ports to endpoints. The cxl_mem driver connects an endpoint device to the platform CXL.mem protoctol decode-topology. At ->probe() time it walks its device-topology-ancestry and adds a CXL Port object at every Upstream Port hop until it gets to CXL root. The CXL root object is only present after a platform firmware driver registers platform CXL resources. For ACPI based platform this is managed by the ACPI0017 device and the cxl_acpi driver. The ports are registered such that disabling a given port automatically unregisters all descendant ports, and the chain can only be registered after the root is established. Given ACPI device scanning may run asynchronously compared to PCI device scanning the root driver is tasked with rescanning the bus after the root successfully probes. Conversely if any ports in a chain between the root and an endpoint becomes disconnected it subsequently triggers the endpoint to unregister. Given lock depenedencies the endpoint unregistration happens in a workqueue asynchronously. If userspace cares about synchronizing delayed work after port events the /sys/bus/cxl/flush attribute is available for that purpose. Reported-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Ben Widawsky <ben.widawsky@intel.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> [djbw: clarify changelog, rework hotplug support] Link: https://lore.kernel.org/r/164398782997.903003.9725273241627693186.stgit@dwillia2-desk3.amr.corp.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-02-04 07:18:31 -08:00
cxl/mem: Add the cxl_mem driver At this point the subsystem can enumerate all CXL ports (CXL.mem decode resources in upstream switch ports and host bridges) in a system. The last mile is connecting those ports to endpoints. The cxl_mem driver connects an endpoint device to the platform CXL.mem protoctol decode-topology. At ->probe() time it walks its device-topology-ancestry and adds a CXL Port object at every Upstream Port hop until it gets to CXL root. The CXL root object is only present after a platform firmware driver registers platform CXL resources. For ACPI based platform this is managed by the ACPI0017 device and the cxl_acpi driver. The ports are registered such that disabling a given port automatically unregisters all descendant ports, and the chain can only be registered after the root is established. Given ACPI device scanning may run asynchronously compared to PCI device scanning the root driver is tasked with rescanning the bus after the root successfully probes. Conversely if any ports in a chain between the root and an endpoint becomes disconnected it subsequently triggers the endpoint to unregister. Given lock depenedencies the endpoint unregistration happens in a workqueue asynchronously. If userspace cares about synchronizing delayed work after port events the /sys/bus/cxl/flush attribute is available for that purpose. Reported-by: Randy Dunlap <rdunlap@infradead.org> Signed-off-by: Ben Widawsky <ben.widawsky@intel.com> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> [djbw: clarify changelog, rework hotplug support] Link: https://lore.kernel.org/r/164398782997.903003.9725273241627693186.stgit@dwillia2-desk3.amr.corp.intel.com Signed-off-by: Dan Williams <dan.j.williams@intel.com>
2022-02-04 07:18:31 -08:00