hws | cpaelzer: Test | 07:59 |
---|---|---|
cpaelzer | hws: exactly | 08:00 |
cpaelzer | and after a long discussion like | 08:00 |
cpaelzer | 1 | 08:00 |
cpaelzer | 2 | 08:00 |
cpaelzer | 3 | 08:00 |
cpaelzer | you could highlight more people to read the above via | 08:00 |
cpaelzer | hws: ^^ | 08:00 |
cpaelzer | hws: btw I also created a bug for a kernel contrib by borntraeger around KVM and zero pages | 08:16 |
cpaelzer | hws: you should have a mail to mirror that back | 08:16 |
cpaelzer | are you the right one to do so? | 08:16 |
cpaelzer | (the mirroring) | 08:16 |
hws | cpaelzer: already done | 08:16 |
cpaelzer | great | 08:16 |
cpaelzer | while we are here - good morning borntraeger | 08:17 |
hws | cpaelzer: already an answer available :) | 08:17 |
cpaelzer | borntraeger: the alignment patch on qemu missed 2.10 as many others did this release | 08:17 |
cpaelzer | borntraeger: did you hear anything on an earlier than usual 2.10.1 and if so any date rumors? | 08:17 |
borntraeger | cpaelzer, only wishes, but no facts | 08:18 |
cpaelzer | hws: the update didn't flow down to me through the tool yet | 08:18 |
cpaelzer | borntraeger: ok, that is my state - thanks | 08:18 |
hws | cpaelzer: just to shortcut..... FWIW, this fix is only necessary if you have guest that use storage keys - e.g. an 16.04 guest should not trigger this problem. | 08:19 |
cpaelzer | hws: do we have any linux guests that use storgae keys in the meantime? | 08:21 |
cpaelzer | I might have lost track | 08:21 |
hws | cpaelzer: question for christian... | 08:21 |
cpaelzer | and as a HV host it is a bit weird, as you can only make "so much" assumption on your guests | 08:21 |
cpaelzer | who says no one uses a Ubuntu Xenial as host in 2020 (still supported then) with a new Guest that suddenly uses stroage keys | 08:22 |
cpaelzer | ... | 08:22 |
cpaelzer | anyway this is just the sort of discussion that I wanted on that bugzilla | 08:22 |
cpaelzer | for a more permanent documentation of the case | 08:22 |
cpaelzer | and since it is mirrored now this will happen | 08:23 |
cpaelzer | I don't mind loosing a few hours for the mirroring | 08:23 |
borntraeger | cpaelzer, hws no current Linux guest uses storage keys | 08:32 |
borntraeger | sles 11 and rhel6 do | 08:32 |
borntraeger | sles12, rhel7 ubuntu 16.04 do not | 08:32 |
cpaelzer | well the old one before we removed it | 08:32 |
cpaelzer | so since there is no intention to add storage keys back in Linux or host non Linux (not intention I'd know of) - the TL;DR is "not needed"? | 08:33 |
borntraeger | cpaelzer, unless you want to support some old debians as guest or so | 08:40 |
cpaelzer | borntraeger: did anybody check if there really is no perf impact? | 08:45 |
cpaelzer | I know that over time all pages will be used | 08:45 |
cpaelzer | but I thought the zero page would still be overloaded | 08:45 |
cpaelzer | or is even the fix | 08:45 |
cpaelzer | only active on storage key using guests | 08:45 |
cpaelzer | borntraeger: and so it is not only "not needed" but even when applied "not actively changing" the zero page handling as long as you have only "new" guests? | 08:46 |
borntraeger | cpaelzer, if a guest has a pattern of only reading a page, it will be a zero page inside the guest | 08:48 |
borntraeger | cpaelzer, the fix is to not use zero pages in the host to back guest pages | 08:49 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!