rickysarraf
commented
3 years ago This looks good to me. The only thing, over time, is that these days we commonly have laptops with non-rotational media. But it still applies to an extent. Just not absolute relevance in this era.
A disk activity can happen for both, reads and writes. On spinning disks, an either activity means spinning up the disk. By this knob, we tell the kernel that if can, it should write the dirty blocks whenever the disk has spun up.
This is the kernel documentation.
Laptop mode is controlled by the knob /proc/sys/vm/laptop_mode. This knob is present for all kernels that have the laptop mode patch, regardless of any configuration options. When the knob is set, any physical disk I/O (that might have caused the hard disk to spin up) causes Linux to flush all dirty blocks. The result of this is that after a disk has spun down, it will not be spun up anymore to write dirty blocks, because those blocks had already been written immediately after the most recent read operation. The value of the laptop_mode knob determines the time between the occurrence of disk I/O and when the flush is triggered. A sensible value for the knob is 5 seconds. Setting the knob to 0 disables laptop mode.
Dear maintainers,
regarding
LM_SECONDS_BEFORE_SYNClaptop-mode.conf(8) says:Unfortunately, it does not to make much sense when read as written. If that would indeed be a timeout started after going idle, then in order to avoid syncs right after spindowns, it should be set to a value more that disk idle timeout (i. e. a timeout that as well starts when disk goes idle and spins it down on expiration).
If I understand Linux documentation [1] for
/proc/sys/vm/laptop_mode(which is what this variable really controls) correctly, itʼs a number of seconds that a machine waits after disk goes active, not idle.[1] file:///usr/share/doc/linux-doc/Documentation/laptops/laptop-mode.txt.gz