Re: Do we still need server generations ?

> Date: Wed, 26 Jun 2024 18:36:16 +0200 > From: "Enrico Weigelt, metux IT consult" > > Hi folks, > > I wonder whether we really still need server generations. > > I've always fonud that *very* confusing that it does some internal > restart (including VT switch, because graphics also reinitializ

Re: Can we rely on #pragma once ?

> Date: Mon, 19 Feb 2024 12:13:34 +0100 > From: "Enrico Weigelt, metux IT consult" > > Hello folks, Hi Enrico, > we've got a lot of include files, therefore lots of guards. > > Modern C compilers should understand #pragma once, but this isn't > without problems (possibly problematic with symli

Re: tarball types (was: [ANNOUNCE] xf86-input-libinput 1.2.1)

> Date: Mon, 24 Jan 2022 15:53:23 -0800 > From: Alan Coopersmith > > On 1/23/22 21:18, Peter Hutterer wrote: > > xf86-input-libinput 1.2.1 is now available. Primarily a few typos and misc > > minor fixes, the most visible change to distributions is that we now ship an > > xz tarball instead of bz

Re: XChangeProperty accesses 32 bit values as 64 bit

> Date: Wed, 2 Jan 2019 16:31:39 +0100 > From: Hanno Böck > > Hi, > > Trying to debug a crash (in gajim) I discovered that it was due to a > stack buffer overread in gtk+/libX11. > > Digging down I am not entirely sure how to interpret it and whether > it's libX11's or GTK's fault. Here's what'

Re: [PATCH xserver] glamor: Work around GEM usage

> From: Thomas Hellstrom > Date: Wed, 23 May 2018 22:58:05 +0200 > > On 05/23/2018 08:00 PM, Adam Jackson wrote: > > On Wed, 2018-05-23 at 11:14 +0200, Thomas Hellstrom wrote: > >> KMS drivers are not required to support GEM. In particular, vmwgfx > >> doesn't support flink and handles and names

Re: [PATCH xserver 3/4] os: Make CLOCK_REALTIME the CLOCK_MONOTONIC fallback

> From: Jeff Smith > Date: Tue, 26 Dec 2017 22:10:54 -0600 > > When clock_gettime(CLOCK_MONOTONIC) fails, xserver falls back on > gettimeofday(). However, gettimeofday() is deprecated in favor of > clock_gettime(CLOCK_REALTIME). > > Fall back on CLOCK_REALTIME if CLOCK_MONOTONIC fails. As long

Re: Severe performance regression was observed on X Server 1.18 with older graphics cards

> From: "Kevin Brace" > Date: Mon, 31 Jul 2017 08:47:37 +0200 > > Hi, > > I am hoping someone can track down a bad commit that led to severe > performance regression likely caused by X Server 1.18. > The other day, I was trying to figure out why a computer with > Xubuntu 16.04.2 was running so s

Re: [PATCH libpciaccess] linux: support 32 bit PCI domains

> Date: Tue, 11 Jul 2017 10:32:19 -0700 > From: Stephen Hemminger > > On Tue, 11 Jul 2017 10:33:56 +0100 > Emil Velikov wrote: > > > On 11 July 2017 at 05:39, Alan Coopersmith > > wrote: > > > On 07/10/17 11:35 AM, Stephen Hemminger wrote: > > >> > > >> The PCI domain maybe larger than 16 b

Re: [xproto] _X_NONNULL and C++ 11

> Date: Sat, 27 May 2017 13:33:25 +0200 > From: walter harms > > Am 27.05.2017 11:02, schrieb Matthieu Herrb: > > Hi, > > > > Marc Espie recently found out that the X_NONNULL macro in Xfuncproto.h > > is generating spurious warnings when included in C++ code build with > > clang++ -std=c++11. >

Re: [PATCH xserver] Makefile.am: add the meson files to the tarball

> Date: Fri, 28 Apr 2017 16:09:55 +1000 > From: Peter Hutterer > > While we're providing both build systems, we'll likely have 'make dist' > generated tarballs - those tarballs should be buildable with meson to > have more exposure. > > Signed-off-by: Peter Hutterer > --- > triggered by: > http

Re: [PATCH libXdmcp v2] Use getentropy() if arc4random_buf() is not available

l. If the recommended > arc4random_buf() function is not available, emulate it by first trying > to use getentropy() on a supported glibc and kernel. If the call fails, > fall back to the current (vulnerable) code. > > Signed-off-by: Benjamin Tissoires Same comment as the oth

Re: [PATCH libICE v2] Use getentropy() if arc4random_buf() is not available

() name sounds a bit weird. Perhaps call it bad_getrandom_buf() or dangerous_getrandom_buf() or insecure_getrandom_buf() to make it obvious that the fallback code really isn't sufficient? Reviewed-by: Mark Kettenis > --- > > changes in v2: > - use the getentropy() fro

Re: [PATCH libXdmcp] Use getrandom() syscall if available

> From: Benjamin Tissoires > Date: Tue, 4 Apr 2017 11:05:47 +0200 > > On Mon, Apr 3, 2017 at 9:17 PM, Mark Kettenis wrote: > > OpenBSD does not have a getrandom(3) function in libc. The OpenBSD > > postion is that the arc4random(3) family of functions is always the

Re: [PATCH libXdmcp] Use getrandom() syscall if available

> From: Benjamin Tissoires > Date: Mon, 3 Apr 2017 17:52:32 +0200 > > On Mon, Apr 3, 2017 at 4:02 PM, Alan Coopersmith > wrote: > > On 04/ 3/17 05:52 AM, Benjamin Tissoires wrote: > >> > >> This allows to fix CVE-2017-2625 on Linux platforms without pulling in > >> libbsd. > >> The syscall getra

Re: [PATCH rendercheck 0/5] Convert to meson.

> From: Eric Anholt > Date: Fri, 24 Mar 2017 13:17:45 -0700 > > Having bitten off a bit more than I can chew in 3 days with the X > Server (hw/xfree86/sdksyms.c is the worst), I decided to take a quick > pass at converting a project that's my own fault. Seems I missed some discussion somewhere..

Re: [PATCH 1/4] xfree86: Don't bother probing -nv on Linux

> From: Timo Aaltonen > Date: Thu, 16 Mar 2017 12:29:41 +0200 > > From: Timo Aaltonen > > For linux this driver is long obsolete now. It may have some relevance > on non-linux systems. Still somewhat relevant on OpenBSD indeed. Diff makes sense to me. Reviewed-by: Mar

Re: [PATCH libdrm v2 2/4] xf86drm: Add USB support

> From: Thierry Reding > Date: Thu, 12 Jan 2017 23:04:27 +0100 > > Allow DRM/KMS devices hosted on USB to be detected by the drmDevice > infrastructure. > > v2: > - make sysfs_uevent_get() more flexible using a format string > > Signed-off-by: Thierry Reding All this sysfs parsing stuff is hi

Re: [PATCH xserver 0/5] Remove (most) 24bpp support

> From: Keith Packard > Date: Thu, 08 Dec 2016 12:25:46 -0800 > > Mark Kettenis writes: > > > It is also unfortunate that this means you won't be able to run X > > directly on the framebuffer set up by UEFI firmware on such servers > > and virtually a

Re: [PATCH xserver 0/5] Remove (most) 24bpp support

> From: Adam Jackson > Date: Thu, 08 Dec 2016 13:34:11 -0500 > > On Thu, 2016-12-08 at 09:19 -0800, Keith Packard wrote: > > > Adam Jackson writes: > > > Assuming those bugs are gone, the only configurations that would break > > > are the two drivers mentioned in the commit message, and any manu

Re: [PATCH libpciaccess] Support for 32-bit domains

> From: Eric Anholt > Date: Thu, 11 Aug 2016 23:32:04 -0700 > > Keith Busch writes: > > > On Thu, Aug 11, 2016 at 12:03:30PM -0700, Eric Anholt wrote: > >> Given that libpciaccess allocates the struct, and the struct isn't > >> embedded in any other public structs, I think you could just tack a

Re: [PATCH libpciaccess] Support for 32-bit domains

> Date: Wed, 10 Aug 2016 22:58:34 + > From: Keith Busch > > On Thu, Aug 11, 2016 at 12:17:48AM +0200, Mark Kettenis wrote: > > > From: Keith Busch > > > Date: Tue, 9 Aug 2016 15:39:35 -0600 > > > > > > A pci "domain" is a purely s

Re: [PATCH libpciaccess] Support for 32-bit domains

> From: Keith Busch > Date: Tue, 9 Aug 2016 15:39:35 -0600 > > A pci "domain" is a purely software construct, and need not be limited > to the 16-bit ACPI defined segment. The Linux kernel currently supports > 32-bit domains, so this patch matches up with those capabilities to make > it usable o

Re: [PATCH] modesetting: fix build with glamor disabled.

> From: Dave Airlie > Date: Fri, 6 May 2016 05:41:58 +1000 > > From: Dave Airlie > > Fix build without --enable-glamor. > > Caught by the arm tinderbox. > > Signed-off-by: Dave Airlie Reviewed-by: Mark Kettenis > --- > hw/xfree86/drivers/modesetting/

Re: [PATCH xserver 3/4] dix: Convert ResourceClientBits to a variable

> From: Adam Jackson > Date: Tue, 3 May 2016 11:24:36 -0400 > > This turns out to be a remarkably hot function in XID lookup. Consider > the expansion of CLIENT_ID. We start with: > > #define RESOURCE_CLIENT_BITS ResourceClientBits() > #define RESOURCE_AND_CLIENT_COUNT 29 > #define CLIENTOF

Re: [PATCH xserver] ad hoc fix for mmap's truncated offset parameter on 32bit

> Date: Fri, 29 Apr 2016 09:34:08 +0200 > From: Julien Cristau > > On Thu, Apr 28, 2016 at 16:10:01 +0200, Mark Kettenis wrote: > > > > From: Stefan Dirsch > > > Date: Thu, 28 Apr 2016 14:35:00 +0200 > > > > > > Builtin modesetting driver

Re: [PATCH xserver] ad hoc fix for mmap's truncated offset parameter on 32bit

> From: Stefan Dirsch > Date: Thu, 28 Apr 2016 14:35:00 +0200 > > Builtin modesetting driver didn't work on 32bit on cirrus KMS. See > https://bugzilla.suse.com/show_bug.cgi?id=917385 for more details. I think selectively compiling with -D_FILE_OFFSET_BITS=64 is a seriously bad idea. Either evr

Re: [xserver-xorg][PATCH 1/1] xwayland: ftruncate if posix_fallocate fails

> Date: Tue, 26 Apr 2016 15:58:47 +0300 > From: Ian Ray > > On Mon, Apr 25, 2016 at 04:49:07PM +0300, Pekka Paalanen wrote: > > On Mon, 25 Apr 2016 15:47:09 +0300 > > Ian Ray wrote: > > > > > On a slow system that is configured with SMART_SCHEDULE_POSSIBLE, large > > > posix_fallocate() request

Re: [PATCH] Avoid 10x7 heuristic, handled by server

> From: Stefan Dirsch > Date: Tue, 26 Apr 2016 11:45:15 +0200 > > From: Frederic Crozat > > Remove the 10x7 heuristic, since the server has equivalent code now. > Instead, disable "acceleration" under qemu, since taking the hypercall > trap is really quite expensive and you're better off doing

Re: [PATCH] Use 16bpp when running in virt and on XenSource gfx

> From: Stefan Dirsch > Date: Tue, 26 Apr 2016 11:45:38 +0200 > > From: Frederic Crozat > > Due to graphics corruption default to 16bpp in virt instead of 24 (Fedora). > Do the same on XenSource gfx, which suffers from the same issue. I think using magic numbers like that is unhelpful, especia

Re: [PATCH xserver] os: Treat ssh as a non-local client (v3)

able names starting with > > that prefix. > > * Ignore executable path for the match. > > v3: (Michel Dänzer) > > * Use GetClientCmdName (Mark Kettenis) > > * Perform check on Windows as well, but only ignore path on Cygwin > > (Martin Per

Re: [PATCH xserver 4/8] Create a threaded mechanism for input [v3]

> From: Keith Packard > Date: Sat, 12 Dec 2015 13:19:59 -0800 > > Mark Kettenis writes: > > > I'd say that would be overkill. The use of recursive mutexes is > > somewhat controversal, which is almost certainly why they weren't part > > of POSIX i

Re: [PATCH xserver 4/8] Create a threaded mechanism for input [v3]

> From: Keith Packard > Date: Fri, 11 Dec 2015 15:37:24 -0800 > > Keith Packard writes: > > > Mark Kettenis writes: > > > >> However, is there a reason why you didn't use the > >> PTHREAD_MUTEX_RECURSIVE mtex type that is standardized by POSI

Re: [PATCH xserver] os: Treat ssh as a non-local client (v2)

> From: =?UTF-8?q?Michel=20D=C3=A4nzer?= > Date: Fri, 11 Dec 2015 11:28:51 +0900 > > By the time we get to ComputeLocalClient, we've already done > NextAvailableClient → ReserveClientIds → > DetermineClientCmd (assuming we're built with #define CLIENTIDS), so > we can look up the name of the

Re: [PATCH xserver 4/8] Create a threaded mechanism for input [v3]

> From: Keith Packard > Date: Thu, 10 Dec 2015 07:18:23 -0800 > > Mark Kettenis writes: > > > Ugh. Exporting global variables as part of the ABI is generally not > > such a good idea. Perhaps it is better to use functions to acquire > > and release the inp

Re: [PATCH xserver 4/8] Create a threaded mechanism for input [v3]

> From: Keith Packard > Date: Wed, 09 Dec 2015 16:04:08 -0800 > > >> +extern _X_EXPORT pthread_mutex_t input_mutex; > >> +extern _X_EXPORT __thread int input_mutex_count; > >> + > > Are these really meant to be exported - a wild guess will be that the > > input drivers won't need access to these

Re: [PATCH 0/5] Remove DRI1

> From: Adam Jackson > Date: Tue, 8 Dec 2015 17:41:35 -0500 > > DRI1 hasn't really been a supported option for a while now, Mesa hasn't > supplied any drivers for it in years, and the design really isn't > compatible with KMS-like drivers. Also, Keith has some upcoming work to > thread input han

Re: [PATCH xserver 0/6] Use a thread for input

> From: Keith Packard > Date: Tue, 8 Dec 2015 15:44:48 -0800 > > Here's a series which revives Tiago's ancient threaded input > patch. This has been cleaned up to make sure that it does locking > correctly, and then patches for xf86 and kdrive/fbdev hook the code to > a couple of X servers to sh

Re: [PATCH:libX11] Remove unused definition of XCONN_CHECK_FREQ

> From: Alan Coopersmith > Date: Sun, 29 Nov 2015 09:48:09 -0800 > > The only use of XCONN_CHECK_FREQ was removed in commit 15e5eaf62897b3179 > when we dropped the old Xlib connection handling in favor of xcb's. > > Signed-off-by: Alan Coopersmith Reviewed-by: Mark K

Re: [PATCH:libxtrans] Remove support for SysV on x86 platforms other than Solaris & SCO

> From: Alan Coopersmith > Date: Sun, 29 Nov 2015 09:47:22 -0800 > > No other x86 SysV platforms have ever been supported in the modular > build systems, so we don't need to keep carrying around a bunch of > ifdef's for them. > > Signed-off-by: Alan Cooper

Re: Direct hardware bashing modules (int10 et al) and recent hardware

> Date: Fri, 20 Nov 2015 17:43:44 +0100 > From: Francois Tigeot > > On Thu, Nov 19, 2015 at 10:06:51AM -0500, Adam Jackson wrote: > > On Thu, 2015-11-19 at 12:07 +, Emil Velikov wrote: > > > > Removing fbdevhw would break mga and r128 and a few others on non-x86. > > FWIW, I have removed su

[PATCH v3] glamor: Make glamor_name_from_pixmap work without DRI3

This function is used by the modesetting driver to implement DRI2 and shouldn't fail on systems that don't support DRI3. Remove the check for DRI3 and rename glamor_egl_dri3_fd_name_from_tex to glamor_egl_fd_name_from_tex. Signed-off-by: Mark Kettenis --- glamor/glamor.c

Re: [PATCH v2] glamor: Make glamor_name_from_pixmap work without DRI3

> From: Eric Anholt > Date: Wed, 18 Nov 2015 12:57:31 -0800 > > Mark Kettenis writes: > > > This function is used by the modesetting driver to implement DRI2 and > > shouldn't fail on systems that don't support DRI3. Remove the check > > for DRI3 and

Re: [PATCH 1/2] modesetting: drop platform_dev pointer.

> From: Dave Airlie > Date: Wed, 18 Nov 2015 09:51:04 +1000 > > This isn't used anywhere, so no point storing it until we need it. Makes sense to me. Reviewed-by: Mark Kettenis > Signed-off-by: Dave Airlie > --- > hw/xfree86/drivers/modesetting/driver.c | 8 +---

Re: [PATCH 2/2] modesetting: create entities for pci and old probe.

> From: Dave Airlie > Date: Wed, 18 Nov 2015 09:51:05 +1000 > > This moves the code from the platform case into > a common function, and calls that from the > other two. Still works for me. FWIW: Reviewed-by: Mark Kettenis > Reported-by: Mark Kettenis > Si

[PATCH] glamor: Make glamor_sync_init work with --disable-xshmfence

Signed-off-by: Mark Kettenis --- glamor/glamor_sync.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/glamor/glamor_sync.c b/glamor/glamor_sync.c index fbc47d4..907e0c6 100644 --- a/glamor/glamor_sync.c +++ b/glamor/glamor_sync.c @@ -97,6 +97,9 @@ glamor_sync_init(ScreenPtr screen

[PATCH v2] glamor: Make glamor_name_from_pixmap work without DRI3

This function is used by the modesetting driver to implement DRI2 and shouldn't fail on systems that don't support DRI3. Remove the check for DRI3 and rename glamor_egl_dri3_fd_name_from_tex to glamor_egl_fd_name_from_tex. Signed-off-by: Mark Kettenis --- glamor/glamor.c | 2

Re: [PATCH] modesetting: and entity setup to generic probe path.

> From: Dave Airlie > Date: Tue, 17 Nov 2015 10:23:37 +1000 > > From: Dave Airlie > > Totally untested, cross fingers hope it works. Seems that does the trick. > Signed-off-by: Dave Airlie > --- > hw/xfree86/drivers/modesetting/driver.c | 25 + > 1 file changed, 21 i

Re: zaphod support broke the modesetting driver

> Date: Mon, 16 Nov 2015 09:05:53 +1000 > From: Dave Airlie > > On 16 November 2015 at 02:57, Mark Kettenis wrote: > > Commit 19e1dc8f6ea6d7ff5ba4a5caa0e2f40a47879408 broke the modesetting > > driver quite badly on systems without XSERVER_PLATFORM_BUS. The > &g

Re: [PATCH] glamor: Make glamor_name_from_pixmap work without DRI3

> From: Keith Packard > Date: Sun, 15 Nov 2015 23:51:04 -0800 > > Mark Kettenis writes: > > > This function is used by the modesetting driver to implement DRI2 and > > shouldn't fail on systems that don't support DRI3. > > If glamor_egl_dri3_fd_na

Re: glamor and the sync extension

t a robust xshmfence implementationand should probably be avoided. > On Sun, Nov 15, 2015 at 8:59 AM, Mark Kettenis > wrote: > > Currently glamor hits an assertion on systems that don't have > > xshmfence. This happens when the glamor code calls > > miSyncGetScreenF

glamor and the sync extension

Currently glamor hits an assertion on systems that don't have xshmfence. This happens when the glamor code calls miSyncGetScreenFuncs() because the miSyncScreenPrivateKey has not been set up. For systems with xshmfence, this happens when miSyncShmScreenInit() gets called, but that code is wrapped

zaphod support broke the modesetting driver

Commit 19e1dc8f6ea6d7ff5ba4a5caa0e2f40a47879408 broke the modesetting driver quite badly on systems without XSERVER_PLATFORM_BUS. The problem is that the "entity private" only gets set up when ms_platform_probe() gets called, but is used unconditionally in the driver. So on systems where ms_platf

[PATCH] glamor: Make glamor_name_from_pixmap work without DRI3

This function is used by the modesetting driver to implement DRI2 and shouldn't fail on systems that don't support DRI3. Signed-off-by: Mark Kettenis --- glamor/glamor.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/glamor/glamor.c b/glamor/glamor.c index d4a0236..110bdb8 10

Re: [PATCH] xorg-wrapper: fix build without libdrm

> From: Hans de Goede > Date: Mon, 19 Oct 2015 11:14:21 +0200 > > Hi, > > On 19-10-15 11:03, Samuel Thibault wrote: > > Hans de Goede, le Mon 19 Oct 2015 10:56:26 +0200, a écrit : > >> In which case would you want to build without libdrm anyways ? Can you > >> explain the use-case for this patc

Re: [PATCH xserver 23/24] Remove readmask from screen block/wakeup handler

> From: Keith Packard > Date: Mon, 21 Sep 2015 12:06:22 +0100 > > Mark Kettenis writes: > > > This, of course, has impact on pretty much all the drivers. Many use > > compat-api.h and could probably be fixed by adding an additional > > stanza there. But som

Re: [PATCH xserver 23/24] Remove readmask from screen block/wakeup handler

> From: Keith Packard > Date: Mon, 21 Sep 2015 07:16:34 +0100 > > With no users of the interface needing the readmask anymore, we can > remove it from the argument passed to these functions. > > Signed-off-by: Keith Packard This, of course, has impact on pretty much all the drivers. Many use

Re: [xrandr] Only use the current information when setting modes

> Date: Sun, 13 Sep 2015 12:14:57 +0100 > From: Chris Wilson > > On Sun, Sep 13, 2015 at 01:08:11PM +0200, Mark Kettenis wrote: > > > From: Chris Wilson > > > Date: Sun, 13 Sep 2015 11:40:37 +0100 > > > > > > Before we change the state (e.g.

Re: [xrandr] Only use the current information when setting modes

> From: Chris Wilson > Date: Sun, 13 Sep 2015 11:40:37 +0100 > > Before we change the state (e.g. adding a mode or applying one to an > output), we query the screen resources for the right identifiers. This > should only use the current information rather than force a reprobe on > all hardware -

Re: Fwd: [PATCH] Make configure test for LD_NO_UNDEFINED_FLAG on Solaris work w/autoconf 2.69

> Date: Sat, 12 Sep 2015 09:40:18 -0700 > > Anyone want to review this 4 month old patch? (void *)0 certainly is avalid null pointer constant. So Reviewed-by: Mark Kettenis > Forwarded Message > Subject: [PATCH] Make configure test for LD_NO_UNDEFINED_FLAG

Re: [PATCH xorg-docs] remove bogus \/ escapes

t, having no effect at all in > terminal output, and only changing spacing in a minor way in typeset > output. > > Signed-off-by: Matthieu Herrb Reviewed-by: Mark Kettenis > --- > man/X.man | 6 +++--- > 1 file changed, 3 insertions(+), 3 deletions(-) > > diff -

Re: [PATCH] remove bogus \/ escapes

t, having no effect at all in > terminal output, and only changing spacing in a minor way in typeset > output. > > Signed-off-by: Matthieu Herrb Reviewed-by: Mark Kettenis > --- > hw/xfree86/man/Xorg.man | 2 +- > man/Xserver.man | 2 +- > 2 files changed, 2 inserti

Re: [PATCH xinit] remove bogus \/ escapes

t, having no effect at all in > terminal output, and only changing spacing in a minor way in typeset > output. > > Signed-off-by: Matthieu Herrb Verified that even man on Linux doesn't render \/ as a slash. Reviewed-by: Mark Kettenis > --- > man/xinit.man | 2 +- >

Re: [PATCH libXt] Fix builds on X32

> From: Ross Burton > Date: Mon, 18 May 2015 17:05:46 +0100 > > The x86 X32 ABI is a 32-bit environment on 64-bit processors, so __amd64__ is > defined but pointers and longs are 32-bit. Handle this case by also checking > __LP64__. That's the wrong way to handle it. Not all 64-bit amd64 system

Re: [PATCH] Allow system call restarts upon signal interruption

> From: =?UTF-8?Q?Michel_D=c3=a4nzer?= > Date: Wed, 13 May 2015 11:18:45 +0900 > > On 13.05.2015 07:39, Daniel Drake wrote: > > The X server frequently deals with SIGIO and SIGALRM interruptions. > > If process execution is inside certain blocking system calls > > when these signals arrive, e.g.

Re: [PATCH 1/2] xserver: add xorg.conf support for gpu devices.

> From: Dave Airlie > Date: Tue, 31 Mar 2015 17:12:28 +1000 > > This should allow for easier tweaking of driver options which > currently mess up the GPU device discovery process. Sorry, not a real review, but: > diff --git a/hw/xfree86/common/xf86Config.c b/hw/xfree86/common/xf86Config.c > ind

Re: [PATCH xinit] Remove workaround for xterm -L (#89653)

> Date: Tue, 24 Mar 2015 11:33:28 +0100 > From: Hans de Goede > > Hi, > > On 24-03-15 05:49, Peter Hutterer wrote: > > The -L flag was removed in 1989. > > > > This enables the legacy keyboard driver again when the server is started > > with -keeptty (bd6cacdd3661) > > > > X.Org Bug 89653

Re: [PATCH:xf86-video-mga] Fix pointer cast warning.

> Date: Wed, 25 Feb 2015 11:31:50 +0100 > From: Tormod Volden > > On Tue, Feb 24, 2015 at 7:36 AM, Mark Kettenis > wrote: > >> cast to pointer from integer of different size > >> [-Werror=int-to-pointer-cast] > >> > >> Signed-off-by:

Re: [PATCH:xf86-video-mga] Fix pointer cast warning.

> Date: Mon, 23 Feb 2015 22:55:51 -0800 > From: Alan Coopersmith > > On 02/23/15 10:36 PM, Mark Kettenis wrote: > >> cast to pointer from integer of different size > >> [-Werror=int-to-pointer-cast] > >> > >> Signed-off-by: Thomas Klausner &

[PATCH xinit] startx: Don't use GNU expr extensions

Use the ':' operator instead of "match" and avoid the use of "\+". Both constructions aren't specified by POSIX and not supported in BSD expr. Also drop the '^' from the regular expressions as it is implicit and POSIX leaves its behaviour undefined.

Re: [PATCH:xauth v2] include POSIX-standard limits.h for PATH_MAX instead of sys/syslimits.h

> From: Alan Coopersmith > Date: Sat, 3 Jan 2015 10:52:28 -0800 > > Signed-off-by: Alan Coopersmith Reviewed-by: Mark Kettenis > --- > gethost.c |2 +- > parsedpy.c |2 +- > 2 files changed, 2 insertions(+), 2 deletions(-) __

Re: [PATCH:xauth] include limits.h for PATH_MAX on OS'es without sys/syslimits.h

> From: Alan Coopersmith > Date: Fri, 2 Jan 2015 09:50:36 -0800 Are there any systems that have but don't have a that provides PATH_MAX? POSIX requires that provides PATH_MAX. And as far as I can tell is a BSD-ism, and has always been included by . So I think you can just replace with a

Re: [PATCH 1/5] dbe: Call to DDX SwapBuffers requires address of int, not unsigned int

> From: Keith Packard > Date: Tue, 9 Dec 2014 09:30:57 -0800 > > When the local types used to walk the DBE request were changed, this > changed the type of the parameter passed to the DDX SwapBuffers API, > but there wasn't a matching change in the API definition. > > At this point, with the AP

Re: [PATCH 4/5] dix: GetHosts bounds check using wrong pointer value

clared as void **, > so we need a cast to turn it into a byte pointer so we can actually do > pointer comparisons. Not sure it is needed for pointer *comparisons*, but it certainly is necessary to be able to do pointer arithmetic without relying on GCC extensions. > Signed-off-by: Keith Pac

Re: [PATCH] Solaris: delete undocumented, unuseful -protect0 flag

c:89:9: warning: shadowed declaration is here [-Wshadow] > int fd; > ^ > > Signed-off-by: Alan Coopersmith Reviewed-by: Mark Kettenis in case you need one. > --- > hw/xfree86/os-support/solaris/sun_init.c | 31 > -- > 1 file chan

Re: [PATCH 2/4] xfree86: Remove implicit xf86EnableIO from BSD for PowerPC

on a PowerMac G4 with an r128 on OpenBSD. Although I think that on OpenBSD we actually avoid accessing VGA registers. Reviewed-by: Mark Kettenis > --- > hw/xfree86/os-support/bsd/ppc_video.c | 1 - > 1 file changed, 1 deletion(-) > > diff --git a/hw/xfree86/os-support/bsd/ppc_vi

Re: [PATCH 1/4] xfree86: Remove no-op /dev/mem checks from xf86OSInitVidMem

. We can add the necessary bits back if we ever add it back (which isn't all that likely). Reviewed-by: Mark Kettenis > Signed-off-by: Adam Jackson > --- > hw/xfree86/os-support/bsd/alpha_video.c | 122 > > hw/xfree86/os-support/bsd/arm_

Re: [PATCH 0/4] Finish killing VidMemInfo

> From: Adam Jackson > Date: Wed, 8 Oct 2014 17:38:55 +0200 > > This finishes the job from: > > http://lists.freedesktop.org/archives/xorg-devel/2014-July/043343.html > > OpenBSD retains its special case for privsep setup, but otherwise this is > the same as before. Please give me a chance to

[PATCH pciaccess] Use PCIOCREADMASK on OpenBSD.

devices was initially probed. Reviewed-by: Matthieu Herrb Signed-off-by: Mark Kettenis --- Given that Matthieu already reviewed this diff, I'll push it in a couple of days unless somebody speaks up. src/openbsd_pci.c | 31 +-- 1 file changed, 25 insertions(

Re: [PATCH 19/19] x86emu: Undefine _NO_INLINE

> From: Adam Jackson > Date: Thu, 25 Sep 2014 13:37:35 -0400 > > Never defined by the server. Fairly certain it was never intended to be defined by the server, but used as a compile-time option to make debugging easier. > Signed-off-by: Adam Jackson > --- > hw/xfree86/x86emu/x86emu/x86emui.h

Re: [PATCH 18/19] xfree86: Remove driver entity hooks and private

> From: Adam Jackson > Date: Thu, 25 Sep 2014 13:37:34 -0400 > > No driver is using these, as far as I know. The xf86-video-mga driver uses these, but only if DISABLE_VGA_IO is defined, and by default it isn't, and there is no configure option to turn it on. I'm a bit worried about silently ign

Re: [PATCH 04/19] loader: Deobfuscate RTLD_* macro stuff

> From: Adam Jackson > Date: Thu, 25 Sep 2014 13:37:20 -0400 > > POSIX requires that these be named correctly, no need to be clever. None of the BSDs have needed those compatibility defines for quite some time, and I doubt anything else still supported cares. Reviewed-by: M

Re: [PATCH] os: Don't listen to 'tcp' or 'unix' by default. Add '-listen' option.

> From: Keith Packard > Date: Fri, 12 Sep 2014 11:35:41 -0700 > > This disables tcp and unix listen ports by default (the unix port is > the non-abstract /tmp/.X11-unix port that xcb doesn't use). Then, it > uses a new xtrans interface, TRANS(Listen), to provide a command line > option to re-enab

Re: [PATCH] os: Don't listen to 'tcp' or 'unix' by default. Add '-listen' option.

> From: Keith Packard > Date: Fri, 12 Sep 2014 15:19:53 -0700 > > Mark Kettenis writes: > > > Unconditionally disabling the "unix" listen ports by default might be > > a step too far. Abstract sockets are only available on Linux. > > Yes, of cour

Re: [PATCH] os: Don't listen to 'tcp' or 'unix' by default. Add '-listen' option.

> From: Keith Packard > Date: Fri, 12 Sep 2014 11:35:41 -0700 > > This disables tcp and unix listen ports by default (the unix port is > the non-abstract /tmp/.X11-unix port that xcb doesn't use). Then, it > uses a new xtrans interface, TRANS(Listen), to provide a command line > option to re-enab

Re: [PATCH xinit 0/1] startx: Pass "-nolisten tcp" by default

> Date: Fri, 12 Sep 2014 12:17:24 +0200 > From: Hans de Goede > > Hi, > > On 09/12/2014 11:12 AM, Mark Kettenis wrote: > >> From: Hans de Goede > >> Date: Fri, 12 Sep 2014 09:25:17 +0200 > >> > >> Hi All, > >> > >> A

Re: [PATCH xinit 0/1] startx: Pass "-nolisten tcp" by default

> From: Hans de Goede > Date: Fri, 12 Sep 2014 09:25:17 +0200 > > Hi All, > > After doing the 1.3.4 release yesterday, I've started working on updating the > Fedora packages to 1.3.4. While looking at our open bug list against xinit, > I found one bug which is not yet resolved in 1.3.4 . > > Th

Re: [PATCH xinit 0/1] startx: Pass "-nolisten tcp" by default

> From: "Colin Harrison" > Date: Fri, 12 Sep 2014 09:37:05 +0100 > > Hi, > > 'this day and age' > 'this time and age' > > Time is good: it allows evolution (or erosion) to slowly happen. > > But I for one vote that X11 network transparency remains default. In this day^H^H^Htime and age sane p

Re: [PATCH 09/12] xfree86: Remove leaky /dev/mem checks from xf86OSInitVidMem

> From: Adam Jackson > Date: Tue, 29 Jul 2014 15:00:16 -0400 > > This is mostly a no-op, the checks didn't have much effect since > pciaccess didn't end up using the fd we opened. Except for OpenBSD, > where you have to pass that in from above, which is sort of a weird API > decision. So this w

Re: [PATCH] modesetting: Support native primary plane rotation

> From: Chris Wilson > Date: Wed, 9 Jul 2014 12:14:41 +0100 > > With the advent of universal drm planes and the introduction of generic > plane properties for rotations, we can query and program the hardware > for native rotation support. > > NOTE: this depends upon the next release of libdrm t

Re: [PATCH] modesetting: Support native primary plane rotation

> Date: Wed, 09 Jul 2014 09:28:31 +0200 > From: walter harms > > Am 09.07.2014 09:00, schrieb Chris Wilson: > > With the advent of universal drm planes and the introduction of generic > > plane properties for rotations, we can query and program the hardware > > for native rotation support. > > >

Re: [PATCH] libx11:lcDefConv.c:fix use before check

> Date: Wed, 04 Jun 2014 13:28:35 +0200 > From: walter harms > > Am 04.06.2014 12:23, schrieb Mark Kettenis: > >> Date: Wed, 04 Jun 2014 12:07:25 +0200 > >> From: walter harms > >> > >> Do not use variables before checke

Re: [PATCH] libx11:lcDefConv.c:fix use before check

> Date: Wed, 04 Jun 2014 12:07:25 +0200 > From: walter harms > > Do not use variables before checked for NULL. > > > Signed-off-by: wharms > > --- > modules/lc/def/lcDefConv.c | 35 +-- > 1 file changed, 25 insertions(+), 10 deletions(-) > > diff --git a/modu

Re: [PATCH 0/9] Rewrite include/servermd.h

> From: Adam Jackson > Date: Wed, 21 May 2014 15:31:16 -0400 > > We really don't want to be in the business of knowing about CPU > architectures. servermd.h is one of the worst offenders for that in > the tree; this series attempts to clean out the gunk. Probably the most > contentious bit is a

Re: [PATCH 1/3] xfree86: Report DRI3 as a built-in module

> From: Chris Wilson > Date: Mon, 12 May 2014 08:12:37 +0100 > > This is so that drivers can do a runtime check that DRI3 is available, > similar to existing runtime checks performed by the drivers for DRI and > DRI2. Does that run-time check actually work? It seems any check based on compiled_

Re: xserver ABI freeze policy (was: [PATCH] hw/xfree86: Restore API compatibility for cursor loading functions)

> From: Keith Packard > Date: Mon, 28 Apr 2014 15:59:32 -0700 > > Aaron Plattner writes: > > > Changing the ABI at the last minute, even if you bump the ABI version > > number, defeats the purpose of that. In this case, I just added support > > for ABI 17 so that driver that in theory will s

Re: [PATCH 0/4] Xorg.wrap: Misc cleanups and porting

> From: Guillem Jover > Date: Mon, 14 Apr 2014 18:13:21 +0200 > > Hi! > > I've adapted the porting patches I submitted to the Debian Xorg wrapper > long ago to the native one, and in the process fixed some other issues. > > I've only tested that it builds, and that it fails on a non-console as

Re: libdrm: ERESTART undefined

> Date: Sat, 12 Apr 2014 09:21:06 +0200 > From: Thomas Klausner > > After updating libdrm in pkgsrc, I got a bug report from David Shao > > --- begin quote --- > > For current cvs pkgsrc on DragonFly 3.7-DEVELOPMENT and perhaps other > BSDs, ERESTART is not defined for userland programs instead

Re: [PATCH libxtrans] Increase UNIX socket buffer size

> Date: Sat, 29 Mar 2014 17:26:00 -0700 > From: Alan Coopersmith > > Do we need to do the same for the SO_RCVBUF as well? Not really. These connections are local by defenition so in principle it doesn't really matter whether we queue on the sending side or on the receiving side. On Linux SO_RC

Re: [PATCH] dix/dispatch: DoGetImage: Use a single buffer

> From: Keith Packard > Date: Sat, 29 Mar 2014 12:40:07 -0700 > > > In my case the problem shows up with Firefox, which apparently is > > rendering images to the screen when downloading them, then grabs the > > pixels with XGetImage() and re-renders using those grabbed pixels. > > Seems utterly re

[PATCH libxtrans] Increase UNIX socket buffer size

From: Mark Kettenis Some systems provide a really small default buffer size for UNIX sockets. Bump it up to 64k if necessary such that large transfers (such as XGetImage() on a 8-megapixel image) don't take tens of seconds. --- Xtranssock.c | 21 + 1 file change

  1   2   3   4   5   6   >