Dear Damien,
Thank you for sending these patches.
Am 23.03.22 um 09:17 schrieb Damien Le Moal:
This series removes the long link debounce delays by default for all
adapters, except for those known to require these delays
(e.g. ata_piix).
Is it know, or just a theory?
The first 2 patches are code cleanup improving the interface of several
functions handling delays.
Patch 3 removes the long delay in sata_link_resume() and reverses the
link flag ATA_LFLAG_NO_DEBOUNCE_DELAY to ATA_LFLAG_DEBOUNCE_DELAY for
adapters to request the delay if needed.
Patch 4 improves sata_link_debounce() by shortening the link stability
test, unless the link has the ATA_LFLAG_DEBOUNCE_DELAY flag set.
This series was tested on a machine with 2 AHCI adapters (Intel and
Marvell) with a port-multiplier box attached to cover that case too.
It’d be great if you gave an example benchmark.
Comments and lots of testing are welcome !
Damien Le Moal (4):
ata: libata-sata: Simplify sata_link_resume() interface
ata: libata-sata: Introduce struct sata_deb_timing
ata: libata-sata: Remove debounce delay by default
ata: libata-sata: Improve sata_link_debounce()
[…]
I am wondering how sure we can be, there won’t be any regressions? Not
having the boot disk detected is quite a serious issue/regression, and
it should be made easy for users to fix that without having to rebuild
the Linux kernel. A Linux kernel CLI parameter to enable the delay would
be helpful for effected users.
Kind regards,
Paul
