> On Sep 28, 2016, at 7:22 AM, Jaromír Doleček
> wrote:
>
> I think it's far assesment to say that on SATA with NCQ/31 tags (max
> is actually 31, not 32 tags), it's pretty much impossible to have
> acceptable write performance without using write cache. We could
On 2016/09/28 20:40, Nick Hudson wrote:
The problem is here...
https://nxr.netbsd.org/xref/src/sys/uvm/pmap/pmap.c#1275
1275 if (pte_valid_p(opte) && pte_to_paddr(opte) != pa) {
1276 pmap_remove(pmap, va, va + NBPG);
1277 PMAP_COUNT(user_mappings_changed);
1278
On 09/20/16 12:34, Rin Okuyama wrote:
[snip...]
However, unfortunately, something is still wrong. top(1) reports
resources
of some processes are negative:
% top
...
PID USERNAME PRI NICE SIZE RES STATE TIME WCPU CPU COMMAND
...
573 root 850 4304K -3832K
I think it's far assesment to say that on SATA with NCQ/31 tags (max
is actually 31, not 32 tags), it's pretty much impossible to have
acceptable write performance without using write cache. We could never
saturate even drive with 16MB cache with just 31 tags and 64k maxphys.
So it's IMO not