User Tools

Site Tools


start

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
Last revision Both sides next revision
start [2018/09/20 14:53]
Barret Rhoden [Quick Setup Guide] added config options for huge pages / FS_DAX_PMD
start [2020/06/12 15:58]
Ira Weiny [Filesystems]
Line 10: Line 10:
   * NDCTL: https://​github.com/​pmem/​ndctl.git   * NDCTL: https://​github.com/​pmem/​ndctl.git
   * NDCTL man pages online: http://​pmem.io/​ndctl/​   * NDCTL man pages online: http://​pmem.io/​ndctl/​
-  * linux-nvdimm Mailing List: https://​lists.01.org/​mailman/listinfo/​linux-nvdimm+  * linux-nvdimm Mailing List: https://​lists.01.org/​postorius/lists/​linux-nvdimm.lists.01.org
   * linux-nvdimm Patchwork: https://​patchwork.kernel.org/​project/​linux-nvdimm/​list/​   * linux-nvdimm Patchwork: https://​patchwork.kernel.org/​project/​linux-nvdimm/​list/​
  
Line 76: Line 76:
  
 <​code>​ <​code>​
-CONFIG_BLK_DEV_RAM_DAX=y +CONFIG_ZONE_DEVICE=y
-CONFIG_FS_DAX=y +
-CONFIG_X86_PMEM_LEGACY=y +
-CONFIG_LIBNVDIMM=y +
-CONFIG_BLK_DEV_PMEM=m +
-CONFIG_ARCH_HAS_PMEM_API=y +
-CONFIG_TRANSPARENT_HUGEPAGE=y+
 CONFIG_MEMORY_HOTPLUG=y CONFIG_MEMORY_HOTPLUG=y
 CONFIG_MEMORY_HOTREMOVE=y CONFIG_MEMORY_HOTREMOVE=y
-CONFIG_ZONE_DEVICE=y +CONFIG_TRANSPARENT_HUGEPAGE=y 
-CONFIG_FS_DAX_PMD=y+CONFIG_ACPI_NFIT=m 
 +CONFIG_X86_PMEM_LEGACY=m 
 +CONFIG_OF_PMEM=m 
 +CONFIG_LIBNVDIMM=m 
 +CONFIG_BLK_DEV_PMEM=m 
 +CONFIG_BTT=y 
 +CONFIG_NVDIMM_PFN=y 
 +CONFIG_NVDIMM_DAX=y 
 +CONFIG_FS_DAX=y 
 +CONFIG_DAX=y 
 +CONFIG_DEV_DAX=m 
 +CONFIG_DEV_DAX_PMEM=m 
 +CONFIG_DEV_DAX_KMEM=m
 </​code>​ </​code>​
  
Line 128: Line 134:
 ndctl create-namespace ties a namespace to a block device or character device: ndctl create-namespace ties a namespace to a block device or character device:
  
-^ mode   ^ description ​   ^ device path ^ label metadata ^ atomicity ^ filesystems ^ DAX ^ PFN metadata ^ former name ^ +^ mode   ^ description ​   ^ device path ^ device type ^ label metadata ^ atomicity ^ filesystems ^ DAX ^ PFN metadata ^ former name ^ 
-| raw    | raw            | /​dev/​pmemN ​ | no             | no        | yes         | no  | no           | | +| raw    | raw            | /​dev/​pmemN  ​| block       | no             | no        | yes         | no  | no           | | 
-| sector | sector atomic ​ | /dev/pmemNs | yes            | yes       | yes         | no  | no           | | +| sector | sector atomic ​ | /​dev/​pmemNs ​| block       | yes            | yes       | yes         | no  | no           | | 
-| fsdax  | filesystem DAX | /​dev/​pmemN ​ | yes            | no        | yes         | yes | yes          | memory | +| fsdax  | filesystem DAX | /​dev/​pmemN  ​| block       | yes            | no        | yes         | yes | yes          | memory | 
-| devdax | device DAX     | /dev/daxN.M | yes            | no        | no          | yes | yes          | dax |+| devdax | device DAX     | /​dev/​daxN.M ​| character ​  | yes            | no        | no          | yes | yes          | dax |
  
-For modes with PFN metadata ("​struct page" metadata), overhead is 64 bytes per 4 KiB of persistent memory. +There are two places to store PFN metadata ("​struct page" metadata):
-  * e.g., 128 MiB for 8 GiB persistent memory +
-  * e.g., 7.45 GiB for 1 TB persistent memory+
   * --map=mem = regular system memory   * --map=mem = regular system memory
     * adequate for small persistent memory capacities     * adequate for small persistent memory capacities
Line 142: Line 146:
     * intended for large persistent memory capacities (there might not be enough regular memory in the system!)     * intended for large persistent memory capacities (there might not be enough regular memory in the system!)
     * persistence of the PFN metadata is not important; this is just convenient because it scales with the persistent memory capacity     * persistence of the PFN metadata is not important; this is just convenient because it scales with the persistent memory capacity
 +
 +The PFN metadata size is 64 bytes per 4 KiB of persistent memory (1.5625%). For some common persistent memory capacities:
 +^  persistent memory capacity ​ ^  PFN metadata size  ^ example ^
 +|  8 GiB    |  128 MiB  | |
 +|  16 GiB   ​| ​ 256 MiB  | |
 +|  96 GiB   ​| ​ 1.5 GiB  | Six 16 GiB NVDIMMs |
 +|  128 GiB  |  2 GiB    | |
 +|  192 GiB  |  3 GiB    | Twelve 16 GiB NVDIMMs |
 +|  256 GiB  |  4 GiB    | |
 +|  512 GiB  |  8 GiB    | |
 +|  768 GiB  |  12 GiB   | Six 128 GiB NVDIMMs |
 +|  1 TiB    |  16 GiB   | |
 +|  1.5 TiB  |  24 GiB   | Six 256 GiB NVDIMMs |
 +|  3 TiB    |  48 GiB   | Six 512 GiB NVDIMMs |
 +|  6 TiB    |  96 GiB   | Six 1 TiB NVDIMMs, or twelve 512 GiB NVDIMMs |
 +
  
 Sector Atomic mode uses a Block Translation Layer (BTT) to help software that doesn'​t understand sectors might end up with a mix of old and new data if power loss occurs while writes were underway. Sector Atomic mode uses a Block Translation Layer (BTT) to help software that doesn'​t understand sectors might end up with a mix of old and new data if power loss occurs while writes were underway.
Line 323: Line 343:
  
 We then structure the rest of the ''​jq''​ command to operate on normal objects, and it works whether we have one namespace or many. We then structure the rest of the ''​jq''​ command to operate on normal objects, and it works whether we have one namespace or many.
 +
 +==== Persistent Naming ====
 +----
 +The device names chosen by the kernel are subject to creation order and discovery order. Environments can not rely the kernel name being consistent from one boot to the next. For the most part they do not change if the configuration stays static, but if a permanent name is needed use /​dev/​disk/​by-id. Recent versions of udev deploy the following udev rule in 60-persistent-storage.rules:​
 +<​code>​
 +# PMEM devices
 +KERNEL=="​pmem*",​ ENV{DEVTYPE}=="​disk",​ ATTRS{uuid}=="?​*",​ SYMLINK+="​disk/​by-id/​pmem-$attr{uuid}"​
 +</​code>​
 +
 +This rule yields symlinks like the following to be created for namespaces defined by labels:
 +
 +<​code>​
 +ls -l /​dev/​disk/​by-id/​*
 + ​lrwxrwxrwx 1 root root 13 Jul  9 15:24 pmem-206dcdfe-69b7-4e86-a01b-f540621ce62e -> ../​../​pmem1.2
 + ​lrwxrwxrwx 1 root root 13 Jul  9 15:24 pmem-73840bf1-4e74-4ba4-a9c8-8248934c07c8 -> ../​../​pmem1.1
 + ​lrwxrwxrwx 1 root root 13 Jul  9 15:24 pmem-8137bdfd-3c4d-4b26-b326-21da3d4cd4e5 -> ../​../​pmem1.4
 + ​lrwxrwxrwx 1 root root 13 Jul  9 15:24 pmem-f43d1b6e-3300-46cb-8afc-06d66a7c16f6 -> ../​../​pmem1.3
 +</​code>​
 +
 +The persistent name for a pmem namespace is then listed in /etc/fstab like so:
 +<​code>​
 +/​dev/​disk/​by-id/​pmem-206dcdfe-69b7-4e86-a01b-f540621ce62e /mnt/pmem xfs       ​defaults,​dax ​       1 2
 +</​code>​
 +
  
 ==== Partitions ==== ==== Partitions ====
Line 385: Line 429:
 <​code>​ <​code>​
 $ mkfs.ext4 -F /​dev/​pmem0p1 $ mkfs.ext4 -F /​dev/​pmem0p1
-$ mkfs.xfs -f /​dev/​pmem0p2+$ mkfs.xfs -f -m reflink=0 ​/​dev/​pmem0p2
 $ mkfs.btrfs -f /​dev/​pmem0p3 $ mkfs.btrfs -f /​dev/​pmem0p3
-$ mount -o dax /​dev/​pmem0p1 /​mnt/​ext4-pmem0 +$ mount [dax_mount_options] ​/​dev/​pmem0p1 /​mnt/​ext4-pmem0 
-$ mount -o dax /​dev/​pmem0p2 /​mnt/​xfs-pmem0+$ mount [dax_mount_options] ​/​dev/​pmem0p2 /​mnt/​xfs-pmem0
 $ mount /​dev/​pmem0p3 /​mnt/​btrfs-pmem0 $ mount /​dev/​pmem0p3 /​mnt/​btrfs-pmem0
  
Line 410: Line 454:
 /​dev/​pmem0p3 ​                   4.3G   ​17M ​ 4.1G   1% /​mnt/​btrfs-pmem0 /​dev/​pmem0p3 ​                   4.3G   ​17M ​ 4.1G   1% /​mnt/​btrfs-pmem0
 </​code>​ </​code>​
 +
 +Where **[dax_mount_options]** depends on the kernel support you have and the desired behavior. ​ See [[fs_mount_options|fs_mount_options]] for details.
  
 Check the kernel log to ensure the DAX mount option was honored; mount does not print this information. Example failures: Check the kernel log to ensure the DAX mount option was honored; mount does not print this information. Example failures:
start.txt · Last modified: 2020/06/12 15:58 by Ira Weiny