Re: [PATCH 14/19] present: Add window flip mode

On 2018-02-02 11:11 PM, Keith Packard wrote: > Michel Dänzer writes: > >> But this seems irrelevant for per-window flips. In this case, the window >> pixmap isn't used for anything else after flipping, so having direct >> access to the pixmap doesn't allow the client to do

Re: [PATCH xrandr v4 2/5] xrandr: stricter --scale argument parsing

Am 05.02.2018 02:47, schrieb Giuseppe Bilotta: > We used to accept something like --scale 2x3junk as a valid input > (scaling x by 2 and y by 3), even though this isn't really a valid > scaling factor. > > Fix by making sure there is nothing after the parsed number(s). > > Signed-off-by:

[PATCH xserver] glx: Only assign 8 bpc fbconfigs for composite visuals.

Commit 91c42093b248 ("glx: Duplicate relevant fbconfigs for compositing visuals") adds many new depth 32 fbconfigs as composite visuals. On a X-Screen running at depth 24, this also adds bgra 10-10-10-2 fbconigs, as they also have config.rgbBits == 32, but these are not displayable on a depth 24

Re: [PATCH xrandr v4 2/5] xrandr: stricter --scale argument parsing

On Mon, Feb 5, 2018 at 10:44 AM, walter harms wrote: > > Am 05.02.2018 02:47, schrieb Giuseppe Bilotta: >> { >> double sx, sy; >> + char junk; >> if (!config_output) argerr ("%s must be used after --output\n", >> argv[i]); >> if (++i >=

Re: [PATCH 14/19] present: Add window flip mode

On 2018-02-02 09:56 PM, Keith Packard wrote: > Michel Dänzer writes: > >> On 2018-02-02 10:42 AM, Roman Gilg wrote: >>> It's because of what you made me aware of in the previous patch set: >>> the window original pixmap needs to have the updated content from the >>> flip

Re: [PATCH xserver] glx: Only assign 8 bpc fbconfigs for composite visuals.

On 02/05/2018 11:20 AM, Mario Kleiner wrote: Commit 91c42093b248 ("glx: Duplicate relevant fbconfigs for compositing visuals") adds many new depth 32 fbconfigs as composite visuals. On a X-Screen running at depth 24, this also adds bgra 10-10-10-2 fbconigs, as they also have config.rgbBits ==

Re: [PATCH xrandr v4 2/5] xrandr: stricter --scale argument parsing

Am 05.02.2018 11:24, schrieb Giuseppe Bilotta: > On Mon, Feb 5, 2018 at 10:44 AM, walter harms wrote: >> >> Am 05.02.2018 02:47, schrieb Giuseppe Bilotta: >>> { >>> double sx, sy; >>> + char junk; >>> if (!config_output) argerr ("%s must be used

xorgproto build system (was Re: [PATCH xserver 00/10] Implement RandR 1.6)

On Sat, 2018-02-03 at 14:28 -0800, Keith Packard wrote: > Adam Jackson writes: > > > Got impatient and hacked on this a bit more: > > > > https://cgit.freedesktop.org/xorg/proto/xorgproto/ > > Thanks for taking this on. It looks great. > > I have a question -- is there any

[PATCH] modesetting: Update fb_id from shadow allocate and destroy if not set

Looks like drmModeDirtyFB() stopped working at some point and we now call it with fb_id of zero for rotated displays. This will stop displays relying on DRM_IOCTL_MODE_DIRTYFB ioctl to display anything. This regression probably with commit 3dcd591fa9b7 ("modesetting: Add support for using RandR

Re: [PATCH] modesetting: Update fb_id from shadow allocate and destroy if not set

Hi, * Keith Packard [180203 20:33]: > Tony Lindgren writes: > > > Looks like drmModeDirtyFB() stopped working at some point and we now > > call it with fb_id of zero for rotated displays. This will stop displays > > relying on DRM_IOCTL_MODE_DIRTYFB ioctl

Re: Making Composite better for interactive apps

On Mon, Feb 5, 2018 at 4:27 AM, Keith Packard wrote: > Giuseppe Bilotta writes: > >> I'm not sure if this could be implemented without making the server >> more susceptible to DOS attacks, though (basically the same issue I >> see in the blocking

Re: Making Composite better for interactive apps

On Mon, Feb 5, 2018 at 9:24 PM, Keith Packard wrote: > What we want is for the compositing manager to get a clean snapshot of > the system and to then be able to present that on the screen each > frame. That means disallowing *all* changes for the duration of the > construction

[PATCH xorgproto] randr: MONITORINFO has outputs, not crtcs

--- randrproto.txt | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/randrproto.txt b/randrproto.txt index c06bc90..f57301d 100644 --- a/randrproto.txt +++ b/randrproto.txt @@ -2363,14 +2363,14 @@ A.1.1 Common Types added in version 1.5 of the protocol 4 ATOM

Re: [PATCH xorgproto 1/2] randr: Add Leases. [v4]

Adam Jackson writes: > On Mon, 2018-02-05 at 12:39 -0800, Keith Packard wrote: > >> randrproto.pc.in| 2 +- > > Please be sure to bump the pc version in meson.build too. (If there's a > more convenient way to only bump things once, great, let's do that >

Re: [PATCH xrandr v4 2/5] xrandr: stricter --scale argument parsing

Giuseppe Bilotta writes: > I'm not a big fan of strtod because with it it's impossible to know if > a conversion actually happened. xrandr --scale ' ' would actually be > accepted (resulting in a scale value of 0), while the scanf catches > it. strtod takes an

Re: [PATCH xrandr v4 2/5] xrandr: stricter --scale argument parsing

On Tue, Feb 6, 2018 at 1:01 AM, Keith Packard wrote: > Giuseppe Bilotta writes: > >> I'm not a big fan of strtod because with it it's impossible to know if >> a conversion actually happened. xrandr --scale ' ' would actually be >> accepted

Re: xorgproto build system (was Re: [PATCH xserver 00/10] Implement RandR 1.6)

On Mon, 2018-02-05 at 10:38 -0500, Adam Jackson wrote: > Now, can we call the meson build official and do the releases from > that, probably yes with some more work. Update: "some more work" is turning out to be more than I want to spend on this at the moment. release.sh's use of git worktrees

Re: [PATCH 14/19] present: Add window flip mode

Michel Dänzer writes: > Ignoring the Security extension, the client has the same control over > the contents of another application's window *using* the X protocol, > doesn't it? Yeah, good point -- it could easily call DRI3BufferFromPixmap for the window pixmap and have

[PATCH xserver 2/5] glx: Use vnd layer for dispatch (v4)

The big change here is MakeCurrent and context tag tracking. We now delegate context tags entirely to the vnd layer, and simply store a pointer to the context state as the tag data. If a context is deleted while it's current, we allocate a fake ID for the context and move the context state there,

Re: [PATCH 14/19] present: Add window flip mode

On Mon, 2018-02-05 at 12:53 +0100, Michel Dänzer wrote: > > > The inability to queue a presentation to the next MSC is more of a step > > > back compared to the status quo. > > > > I'm about to go write up some ideas I'm working on that will make it > > possible to more regularly display

Anyone have a backup of XFree86 CVS?

I've asked this a couple of times on IRC, repeating it here for wider distribution. I had thought I had a git import laying around somewhere but I can't seem to find it. cvs.xfree86.org hasn't been alive in a long time, and it'd be a shame if that history just got lost, fragmentary as it was. If

Re: [PATCH 14/19] present: Add window flip mode

Michel Dänzer writes: > The motivation for this work is fullscreen applications with Xwayland. I think this could be equally useful for windowed applications as well as it should allow them to avoid the copy to the window pixmap and (perhaps) land in an overlay in whatever

Re: [PATCH xserver 2/4] glx: Use vnd layer for dispatch (v3)

On Mon, 2018-02-05 at 12:05 -0700, Kyle Brenneman wrote: > In __glXDisp_DestroyContext, doesn't it need to record the fake XID that > it generates? If I'm reading it right, if the client deletes a current > context and later unbinds it, then xorgGlxMakeCurrent will call > FreeResourceByType

Re: Making Composite better for interactive apps

Giuseppe Bilotta writes: > While I find this conceptually acceptable, I have an uneasy feeling > about its practicality. Yeah, we'd definitely have to experiment with both X server and Compositing Manager changes together to see what actually works. > I'm coming at

Re: [PATCH xrandr v4 2/5] xrandr: stricter --scale argument parsing

Giuseppe Bilotta writes: > Not in the sense I mean above. If the string is a sequences of > whitespace characters, strtod would return 0 as value (no conversion), > and set endptr to the end of the string, because it would gobble the > whitespace. This would be

Re: [PATCH xorgproto] randr: MONITORINFO has outputs, not crtcs

Giuseppe Bilotta writes: > - 2 CARD16 ncrtcs > + 2 CARD16 noutputs Reviewed-by: Keith Packard -- -keith signature.asc Description: PGP signature ___

Re: Making Composite better for interactive apps

Giuseppe Bilotta writes: > Maybe we don't even need to change the AutoList. What we want is to > ensure that the Compositing Manager has _processed_ the notification > of the geometry change for the windows in the AutoList. This could be > achieved for exampe by

Re: [PATCH xserver 00/10] Implement RandR 1.6 (leases and non-desktop)

On Sat, Feb 03, 2018 at 02:28:58PM -0800, Keith Packard wrote: > Adam Jackson writes: > > > Got impatient and hacked on this a bit more: > > > > https://cgit.freedesktop.org/xorg/proto/xorgproto/ > > Thanks for taking this on. It looks great. > > I have a question -- is there

Re: Anyone have a backup of XFree86 CVS?

On Mon, 2018-02-05 at 15:24 -0500, Adam Jackson wrote: > I've asked this a couple of times on IRC, repeating it here for wider > distribution. I had thought I had a git import laying around somewhere > but I can't seem to find it. cvs.xfree86.org hasn't been alive in a > long time, and it'd be a

[PATCH xserver] xwayland: Don't process cursor warping without an xwl_seat

Unfortunately, on my machine Xwayland immediately crashes when I try to start it. gdb backtrace: #0 0x774f0e79 in wl_proxy_marshal () from target:/lib64/libwayland-client.so.0 #1 0x00413172 in zwp_confined_pointer_v1_destroy (zwp_confined_pointer_v1=0x7) at

Re: [PATCH xserver 3/3] xfree86: Only call PreInit handler if it exists for device

On Sun, 2018-02-04 at 23:17 -0600, Jeff Smith wrote: > DoConfigure() attempts to call the PreInit handler on a device without > checking that the handler exists. > > Check that the PreInit handler exists for a device before attempting to > call it. This is a bit strange to me as the PreInit hook

Re: [PATCH xorgproto 1/2] randr: Add Leases. [v4]

On Mon, 2018-02-05 at 12:39 -0800, Keith Packard wrote: > randrproto.pc.in| 2 +- Please be sure to bump the pc version in meson.build too. (If there's a more convenient way to only bump things once, great, let's do that instead.) > +7.6. Extension Requests added in version

Re: Making Composite better for interactive apps

Giuseppe Bilotta writes: > So the only difference between the two approaches is that in the > current async way the server can cover the latency that would come > from waiting for the WM to complete the mapping (or in general other > such requests) by processing other

[PATCH xorgproto 2/2] randr: Add non-desktop output property and behaviors [v3]

non-desktop devices are those to which the normal desktop environment should not be extended. Examples are Head-mounted displays and the Apple Touch Bar. How an output device is set to non-desktop is not part of this proposal; it is expected that the underlying operating system will provide this

[PATCH xorgproto 1/2] randr: Add Leases. [v4]

A "lease" is a set of crtc and output resources granted to another application for use outside of X. These will not be usable through the X protocol until the lease terminates. Leased outputs will be seen as disconnected, leased CRTCs will be seen as not usable with any output. v2: Delete

[PATCH xorgproto 0/2] randr: Add leases and non-desktop property

I've just taken my randrproto patches and cherry-picked them into xorgproto, then cleaned up the bits for the randrproto version number. ___ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info:

Re: [PATCH xserver 2/4] glx: Use vnd layer for dispatch (v3)

On 02/02/2018 12:15 PM, Adam Jackson wrote: The big change here is MakeCurrent and context tag tracking. We now delegate context tags entirely to the vnd layer, and simply store a pointer to the context state as the tag data. If a context is deleted while it's current, we allocate a fake ID for

Re: xorgproto build system (was Re: [PATCH xserver 00/10] Implement RandR 1.6)

On 02/ 5/18 11:06 AM, Adam Jackson wrote: > On Mon, 2018-02-05 at 10:38 -0500, Adam Jackson wrote: >> Would anyone have a problem with only making xz tarballs? > > This would still be nice to know the answer to. I wouldn't, since we deal with those for GNOME already. -- -Alan