[SRU] [Xenial] [PATCH 00/14] Enable NVMe APST for Xenial, take two

Kai-Heng Feng kai.heng.feng at canonical.com
Fri Jan 5 06:06:03 UTC 2018


BugLink: https://bugs.launchpad.net/bugs/1664602

[Impact]
NVME APST feature is not enabled, so the power consumption is higher. 

[Fix]
Enable NVME APST to conserve energy.

[Test]
Use 'nvme get-feature -f 0x0c -H /dev/nvme0' from nvme-cli to query the feature.

Before APST enabled:
get-feature:0xc (Autonomous Power State Transition), Current value:00000000
       Autonomous Power State Transition Enable (APSTE): Disabled

After APST enabled:
get-feature:0xc (Autonomous Power State Transition), Current value:0x000001
       Autonomous Power State Transition Enable (APSTE): Enabled

[Regression Potential]
Low. Now Artful is release for some time now, we only have one APST
issue, which is fixed by a new workaround.

Andy Lutomirski (8):
  nvme/scsi: Remove power management support
  nvme: Fix nvme_get/set_features() with a NULL result pointer
  nvme: Pass pointers, not dma addresses, to nvme_get/set_features()
  nvme: Add a quirk mechanism that uses identify_ctrl
  nvme: Enable autonomous power state transitions
  nvme: Adjust the Samsung APST quirk
  nvme: Quirk APST off on "THNSF5256GPUK TOSHIBA"
  nvme: Quirk APST on Intel 600P/P3100 devices

Christoph Hellwig (3):
  nvme: return the whole CQE through the request passthrough interface
  nvme: factor out a add nvme_is_write helper
  nvme: Modify and export sync command submission for fabrics

Kai-Heng Feng (3):
  nvme: only consider exit latency when choosing useful non-op power
    states
  nvme: relax APST default max latency to 100ms
  nvme-pci: disable APST on Samsung SSD 960 EVO + ASUS PRIME B350M-A

 drivers/nvme/host/core.c | 287 +++++++++++++++++++++++++++++++++++++++++++----
 drivers/nvme/host/nvme.h |  27 ++++-
 drivers/nvme/host/pci.c  |  50 +++++++--
 drivers/nvme/host/scsi.c |  80 +------------
 include/linux/nvme.h     |   6 +
 5 files changed, 341 insertions(+), 109 deletions(-)

-- 
2.14.1





More information about the kernel-team mailing list