On 06/03/2015 09:02, Nicolò Costanza wrote:
> Hi
>
> Kernel 3.18.8 to be tested...
>
> The thing to be tested, if you have a laptop with ATI free driver:
> /"The laptop screen will not turn back after "resume" from the
> suspension!"/

Hi Nicco, I found the problem of the scheduler you told me, about the
fact that

cat /sys/block/sda/queue/scheduler

doesn't allow to set the scheduler. This is due to the fact that more or
less there is a new scheduler
called "MultiQueue Block I/O" which is turned on by default in the new
key CONFIG_SCSI_MQ_DEFAULT=y. This
should improve (i.e. decrease) the latency for SSD (though this is not
yet true);
setting this keyword to "# CONFIG_SCSI_MQ_DEFAULT is not set" would
revert to the
old behaviour. I wouldn't be surprised if this is also fix the USB
transfer problems
(as of
https://forums.openmandriva.org/en/discussion/937/copying-large-amount-of-files-results-in-crash-with-kernel-3-18-5#latest).

I tried to disable the CONFIG_SCSI_MQ_DEFAULT in 3.18.9 for which I
built a quick shot with an "experimental" kernel
(available on ghibo_personal on ABF for Rosa2014.1
[http://abf-downloads.rosalinux.ru/ghibo_personal/repository/rosa2014.1/x86_64/main/release/]
and OMV2014.1
[http://abf-downloads.rosalinux.ru/ghibo_personal/repository/openmandriva2014.0/x86_64/main/release/]).
Note that this had the TOI enabled (but untested).

> Kernel 3.18.8 for ROSA 2012 2012.1 2014.1 OMV2013 2014 2015
> http://mib.pianetalinux.org/forum/viewtopic.php?f=38&t=4533

more details there.

_______________________________________________
OM-Cooker mailing list
[email protected]
http://ml.openmandriva.org/listinfo.cgi/om-cooker-openmandriva.org

Reply via email to