.. | .. |
---|
1 | | -For all of the nmem device attributes under nfit/*, see the 'NVDIMM Firmware |
---|
| 1 | +For all of the nmem device attributes under ``nfit/*``, see the 'NVDIMM Firmware |
---|
2 | 2 | Interface Table (NFIT)' section in the ACPI specification |
---|
3 | 3 | (http://www.uefi.org/specifications) for more details. |
---|
4 | 4 | |
---|
.. | .. |
---|
202 | 202 | functions. See the section named 'NVDIMM Root Device _DSMs' in |
---|
203 | 203 | the ACPI specification. |
---|
204 | 204 | |
---|
| 205 | +What: /sys/bus/nd/devices/ndbusX/nfit/firmware_activate_noidle |
---|
| 206 | +Date: Apr, 2020 |
---|
| 207 | +KernelVersion: v5.8 |
---|
| 208 | +Contact: linux-nvdimm@lists.01.org |
---|
| 209 | +Description: |
---|
| 210 | + (RW) The Intel platform implementation of firmware activate |
---|
| 211 | + support exposes an option let the platform force idle devices in |
---|
| 212 | + the system over the activation event, or trust that the OS will |
---|
| 213 | + do it. The safe default is to let the platform force idle |
---|
| 214 | + devices since the kernel is already in a suspend state, and on |
---|
| 215 | + the chance that a driver does not properly quiesce bus-mastering |
---|
| 216 | + after a suspend callback the platform will handle it. However, |
---|
| 217 | + the activation might abort if, for example, platform firmware |
---|
| 218 | + determines that the activation time exceeds the max PCI-E |
---|
| 219 | + completion timeout. Since the platform does not know whether the |
---|
| 220 | + OS is running the activation from a suspend context it aborts, |
---|
| 221 | + but if the system owner trusts driver suspend callback to be |
---|
| 222 | + sufficient then 'firmware_activation_noidle' can be |
---|
| 223 | + enabled to bypass the activation abort. |
---|
205 | 224 | |
---|
206 | 225 | What: /sys/bus/nd/devices/regionX/nfit/range_index |
---|
207 | 226 | Date: Jun, 2015 |
---|