-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 On 1/13/21 3:25 PM, Marek Marczykowski-Górecki wrote: > On Wed, Jan 13, 2021 at 01:21:51PM +0000, Jinoh Kang wrote: >> On 1/11/21 11:03 PM, Marek Marczykowski-Górecki wrote: >>> So, I can confirm the (fixed) 5.10 patch also improves the situation. > >> Sounds good. Thanks for testing! > >>> Have you sent it upstream? > >> No, qubes-users and qubes-devel are the only mailing list where I >> posted this. > >> I guess chances of these patches being merged upstream would not be >> that great. > > If that bug indeed affects only Qubes OS, there is a greater chance to > accept the patch, if the option defaults to false. > >> After all, we're not going to need it with Qubes R4.1. > > Are you sure?
Now that I think about it, I just realised that we are still going to this patch in Qubes R4.1, while migrating old VMs still using MFNDUMP. But then, Fedora has moved away from xf86-video-intel to modesetting, so we aren't going to need this patch either way (unless we are going to switch to Wayland tomorrow)? I believe everything else has been covered in qubes-issues#5909 [1]. > The issue affects dom0 windows, which suggests it still > may be necessary. Does it? I personally haven't experienced this case yet. > On the other hand, your patch description suggests > it's just any VM-mapped window triggers the faulty path in the > xf86-video-intel driver, that later affects all of the output. Yes, but for privcmd-backed pages only. gntdev pages are unaffected. > >>> I do consider including it in our standard >>> kernel package, but I'd like to see i915 driver maintainer opinion >>> first. > >> If you mean you'd prefer to have it upstreamed, I'd appreciate some >> Tested-by: and/or Reviewed-by: lines for the trailer from you. > > Can you send a fixed patch (that builds), rebased on top of recent Linux > (5.11-rc3, or recent 5.10)? I'll re-test and add my Tested-by:. OK, I'll send it upstream. > >> I'm fine as well if you'd rather just submit it yourself. > >> Otherwise I suppose I shall only CC' the maintainers and not the list? > > Generally, Linux patches should be sent to whoever MAINTAINERS file > lists, which do include some mailing lists. I highly recommend using > scripts/get_maintainer.pl script for that purpose (if you use git > send-email, that's as easy as --cc-cmd=scripts/get_maintainer.pl). PS: due to Google SMTP OAuth2 issues, I have this weird workflow where I first stage my changes in my local IMAP daemon's mbox and bounce it via Thunderbird. This is admittedly very annoying, and perhaps I should tinker with (neo)mutt + gmail-oauth2-tools some time. Or maybe switch to a "real" mail account. This will also explain my past embedded PGP signing mistake. Sorry about that...;( > > PS The other (independent) issue I mentioned seems to be > https://bugzilla.suse.com/show_bug.cgi?id=1180543, which is supposed to > be already fixed in >=5.10.6. I've already uploaded 5.10.7, but haven't > tested it on this particular machine yet. Thanks, good to know. - -- [1] https://github.com/QubesOS/qubes-issues/issues/5909 - -- Sincerely, Jinoh Kang -----BEGIN PGP SIGNATURE----- iQJMBAEBCAA2FiEEzGktrvc/U3kFXd9AGlqQRGyEq/UFAmABvzEYHGppbm9oLmth bmcua3JAZ21haWwuY29tAAoJEBpakERshKv1MIcQAI3TZ5Lp1kfIxEW9EasMgLQH lKu+YH/ca5iWkoMGn3/uJKvKF2gEkuJp0u85nAu+ns7DWrBjggJnZ+A0y/uZtAFN FcFQy4OMdURRwPVUTwAOr5WJkkpm04PEhNLZJrgD+8crb/fTYiwu5Ii3XO7egFPd PQB71cxS4rzWoi+Dz8WVrhn1/wIg/jqL15//iXYM44NXLs8355lWI/QT5/L0dfn1 7g4vPJkbpeMUdjCo5xGyDDc59VUyyzauCPskRW7+G0y6JagzFi8wOI4b6kkoABau YUgsC/CPGu+JKYPqbtROmA6M5+Mp8McA3oX2DbyIeYfTK58qd32/+bLI0Q6PY14r 8MGXzLFvN8kBA+e9mGCxX2SdX4sb6bYwlAaUs64B37Eo7ybugapUz6GUpL1U64+o yyQIVoSZ1Hd4zA8KXMP0uocBuH1zrePU0EJ6nlp33X5HPOJXyPdMIFk+CJmPR6yk 3yDmGnQeXxOA+j8uXhH4yOF+zcwyE3+BDOsTKhsVxr99A24TYW92pVUXhN2KTkZQ 3dIYZfaYH2TySA67Z3n9bQ7aoyxe00Q8UcGuO87bFU8vGzUeRGlwyfMYKUzd6Nuz E+w9g8DYfYfT7HXP1TfzMgAoF9hJfMLN3IQb56BBsIU2je5/A1HgF5kRcmulVKU4 CHThsvFTdVjydLZPuPuR =8295 -----END PGP SIGNATURE----- -- You received this message because you are subscribed to the Google Groups "qubes-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to qubes-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/qubes-users/15bfdb86-a55f-f4d3-836b-ca870934ae5a%40gmail.com.
