bug#63187: 30.0.50; Tail of longer lines painted after end of nearby lines on macOS

[Aaron Jensen <aaronjensen@gmail.com>]

On Sat, Jun 24, 2023 at 11:41 AM Alan Third <alan@idiocy.org> wrote:
>
> On Sat, Jun 24, 2023 at 10:52:08AM -0400, Aaron Jensen wrote:
> > On Sat, Jun 24, 2023 at 10:14 AM Alan Third <alan@idiocy.org> wrote:
> > >
> > > On Sat, Jun 24, 2023 at 09:34:18AM -0400, Aaron Jensen wrote:
> > > >
> > > > Could you try removing the destination lock as well and see if that
> > > > impacts anything? From what I can tell, locking the destination may be
> > > > a good idea, but I'm curious if Alan has any thoughts as to why it'd
> > > > be a bad idea.
> > >
> > > copyContentsTo is only called by getContext, which has already locked
> > > the destination, so there's no need to lock it again.
> >
> > Ah, I see. Is the kIOSurfaceLockAvoidSync flag valuable for any reason?
>
> I don't think so... Not in our use-case. IIRC it's possible to get the
> GPU to perform actions on the buffer once it's in VRAM. We don't do
> that, so we don't need to ever worry about write-backs to system RAM.
>
> I think.
>
> > If I recall correctly, there's some code in Emacs that optimizes which
> > areas of the screen glyphs are drawn to. Maybe it remembers what was
> > the background color already and doesn't clear it again... I don't
> > remember where I saw it, but I think it was outside of nsterm. Does
> > that ring a bell? I'm thinking about how this manifests. For me, it's
> > always whole characters that are painted, they are always painted in
> > what would otherwise be whitespace, and they tend to get "copied" to
> > the next line in the whitespace as scrolling happens. In other words,
> > it doesn't just seem like a fluke write during the transfer to VRAM.
> > It seems like something in the engine is writing them, that the state
> > is getting "stuck" somehow. Does any of that ring a bell/jog anything?
>
> Its quite possible. I did say before in the thread that it seems quite
> possible to me that something isn't clearing the whitespace correctly.
> But it's obviously rare, and I would expect that it's some piece of NS
> port code rather than somewhere else.
>
> Unless there's something that's #ifdef'd out because at some point in
> the past the NS port has behaved differently...
>
> But more likely if this is what's going on then it's going to be a bug
> in the NS port. Unfortunately I don't really understand how the glyph
> drawing side works, and never did.
>
> But if that's the case, why would removing the asynchronous call to
> getContext fix so many problems?

It's possible we have two very different problems that only appear related:
1. The one i'm seeing, which is sort of ghosting of other lines into
the whitespace of nearby lines. The getContext call removal did not
fix this for me, I saw it happen once.
2. The one Kai is seeing, that is exacerbated by decreasing the
polling interval, but seems to be helped by removing the getContext
call.

> Something perhaps worth trying... Since removing the asynchronous call
> to getContext fixes the problems, perhaps we need to think about the
> "lazy" way we get the next buffer when the current one is displayed.
> At the moment it just forgets about it until we want to draw to the
> screen, at which point we call getContext and it creates the buffer if
> necessary and copies the old one to the new one.
>
> Maybe we should get the new buffer and do the copy when we set the
> current buffer for display...
>
> IIRC I avoided that because there isn't always time for the buffer to
> have been sent to VRAM and unlocked before the *next* call to display,
> so I wanted to leave it as long as possible between display and
> getting the next buffer, but maybe this is just the wrong way to do
> it.
>
> So I suppose putting a call to getContext right after "currentSurface
> == NULL" in display might be a quick and dirty way to test that.

My problem is that at this point it happens so infrequently and I have
no idea if that's because of the patches I'm trying or some other
environmental thing or just luck. I'm going to try running without the
async getContext and without the setNeedsDisplay for a while and see
if it happens. Perhaps the setNeedsDisplay is somehow causing an issue
and that's why changing it from source to dest seemed to help but it
didn't alleviate it.

If I see it again, I'll add the sync getContext call in, though I'll
admit I do not understand your paragraph above starting with IIRC. Are
you suspecting a potential problem with reading from the surface that
is in the process of being copied to vram?

Aaron