Accelerants: Fix -Wformat-overflow

* Use strlcpy() and strlcat() instead of sprintf()
to suppress -Wformat-overflow pointed by gcc8.
* Add <string.h> to mga_std.h for strlcpy() and strlcat().

Cha

Accelerants: Fix -Wformat-overflow

* Use strlcpy() and strlcat() instead of sprintf()
to suppress -Wformat-overflow pointed by gcc8.
* Add <string.h> to mga_std.h for strlcpy() and strlcat().

Change-Id: Ib038283fd213c9e0ac9f19482402b55d2e3f204a
Reviewed-on: https://review.haiku-os.org/429
Reviewed-by: waddlesplash <waddlesplash@gmail.com>
Reviewed-by: Adrien Destugues <pulkomandy@gmail.com>

show more ...

improved modeline scaling for digitally connected screens. This fixes shifted output on screen mostly happening on 640x480 resolution, due to missing CRTC SYNC pulses (ticket #4442). Bumped version t

improved modeline scaling for digitally connected screens. This fixes shifted output on screen mostly happening on 640x480 resolution, due to missing CRTC SYNC pulses (ticket #4442). Bumped version to 1.04.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@33150 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

driver now sets up internal EDID info related to CRTCs next to earlier introcuded code for EDID info related to connectors which makes it much more useable later on. This commit also hopefully fixes

driver now sets up internal EDID info related to CRTCs next to earlier introcuded code for EDID info related to connectors which makes it much more useable later on. This commit also hopefully fixes the problem introduced in R31183 shutting off some screens in 'invalid' connector setups that the driver tries to correct.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31291 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

doing preparations for using analog widescreen detection. if all is right the driver's behaviour hasn't changed yet.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31183 a95241bf-73f2-0310-859d

doing preparations for using analog widescreen detection. if all is right the driver's behaviour hasn't changed yet.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@31183 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

rewrote kerneldriver and accelerant's retrace INT handling. This fixes the 'driver hanging' bug exhibiting sometimes when CRTC2 is being used as primary CRTC (driver internal feature). INT handling n

rewrote kerneldriver and accelerant's retrace INT handling. This fixes the 'driver hanging' bug exhibiting sometimes when CRTC2 is being used as primary CRTC (driver internal feature). INT handling now exists for both CRTC1 and CRTC2: enabling only the INTs for the head currently being used as primary. (limitation can be removed once we use a driver instance 'per head' instead of 'per card'.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16281 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

same for cursor clear and define functions

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@16112 a95241bf-73f2-0310-859d-f6bbb57e9c96

added timeouts for all retrace-sync waiting loops in accelerant to prevent software from hanging if trying to use a card that isnt' coldstarted.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@1

added timeouts for all retrace-sync waiting loops in accelerant to prevent software from hanging if trying to use a card that isnt' coldstarted.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15924 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

added preliminary DPMS support for laptop panels (disabling/enabling backlight). Not for NV11 though, and only if no external DVI panel is connected to a supported laptop. Please note: this feature c

added preliminary DPMS support for laptop panels (disabling/enabling backlight). Not for NV11 though, and only if no external DVI panel is connected to a supported laptop. Please note: this feature can possibly be extended for NV11 and external panels if enough feedback (logfile) gets to me. If the driver/screen nolonger works for someone please let me know as well. Bumped version to 0.66.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15051 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

updated DPMS programming to take DVI panels into account for TVout modes on the same head. Enabled TVout support for all cards as long as they have a supported encoder (test-image is now shut-off). T

updated DPMS programming to take DVI panels into account for TVout modes on the same head. Enabled TVout support for all cards as long as they have a supported encoder (test-image is now shut-off). TVout support should be complete now, apart from adding support for more encoder-types. Bumped version to 0.63.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14939 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

small comment update

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14886 a95241bf-73f2-0310-859d-f6bbb57e9c96

Added fifo watermark/burst-size programming for NV11 on their secondary heads. This fixes the 'vertical distortion stripes' occuring in 32bit color using secondary TVout modes: NV11 is just perfect h

Added fifo watermark/burst-size programming for NV11 on their secondary heads. This fixes the 'vertical distortion stripes' occuring in 32bit color using secondary TVout modes: NV11 is just perfect here now :). Modified TNT2-M64 fifo watermark/burst-size programming to be executed independant of being coldstarted, which is more 'all-round'. It's upto users to confirm/deny my strong suspision this is fully OK to do (already tested on 3 cards over here). Bumped version to 0.62.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14884 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

updated DPMS programming to be compatible with secondary TVout modes.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14662 a95241bf-73f2-0310-859d-f6bbb57e9c96

modified custom flags behaviour, mostly TV_PRIMARY flag stuff: more in proposemode, less in other places. For singlehead TV modes on dualhead cards TVout chip is always assigned to the primary head n

modified custom flags behaviour, mostly TV_PRIMARY flag stuff: more in proposemode, less in other places. For singlehead TV modes on dualhead cards TVout chip is always assigned to the primary head now.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14660 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

added/rewrote TVout enabling/disabling routines: moved CRTC specific stuff to the respective files. Driver now has support for TVout head switching, although it's not completed yet (switch mode will

added/rewrote TVout enabling/disabling routines: moved CRTC specific stuff to the respective files. Driver now has support for TVout head switching, although it's not completed yet (switch mode will likely mess up as that's not taken into account yet :). On singlehead cards TVout is on the one and only head, on dualhead cards TVout defaults to the secondary head (use dualhead setupthere ). Tested OK on NV11 in dualhead clone and stretch mode.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14545 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

changed fetching DPMS method: now just returning the earlier set flags instead of querying a CRTC. This fixes DPMS state getting lost after using TVout, and simplifies code as well.

git-svn-id: file

changed fetching DPMS method: now just returning the earlier set flags instead of querying a CRTC. This fixes DPMS state getting lost after using TVout, and simplifies code as well.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14512 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

completed 4 Desktop TVout modelines. Also established that the comments in the crtc modesetting functions were not applicable (about 'offsets' on timings before programming in respective registers.

completed 4 Desktop TVout modelines. Also established that the comments in the crtc modesetting functions were not applicable (about 'offsets' on timings before programming in respective registers.

git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@14299 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

shutdown panel DPMS for laptops: I don't like the looks of that!. Confirmed DPMS working OK for all DVI panels though.


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@11406 a95241bf-73f2-0310

shutdown panel DPMS for laptops: I don't like the looks of that!. Confirmed DPMS working OK for all DVI panels though.


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@11406 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

added preliminary DPMS support for all panels (still in test)


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@11401 a95241bf-73f2-0310-859d-f6bbb57e9c96

fixed cursor visibility update on show/hide_cursor commands for secondary head on NV40 architecture (hardware bug workaround). Confirmed Dualhead up and working correctly with two analog connected sc

fixed cursor visibility update on show/hide_cursor commands for secondary head on NV40 architecture (hardware bug workaround). Confirmed Dualhead up and working correctly with two analog connected screens on PCIe 6600 GT.


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@11368 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

fixed a truncate error with casting between floats and ints: this makes the centering calculation for WS panels work more precise, removing some 'noise' pixels on the right side of the screen when di

fixed a truncate error with casting between floats and ints: this makes the centering calculation for WS panels work more precise, removing some 'noise' pixels on the right side of the screen when displaying non native aspect modes.


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@11235 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

finalized HW cursor sync: now tested and found OK on all my pre-NV40 cards.


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@10650 a95241bf-73f2-0310-859d-f6bbb57e9c96

completed cursor hardware retrace syncing, confirmed working fully OK on FX5200. Other cards will be tested shortly. Note: this also seems to fix losing screen sync for short bursts with DVI connecto

completed cursor hardware retrace syncing, confirmed working fully OK on FX5200. Other cards will be tested shortly. Note: this also seems to fix losing screen sync for short bursts with DVI connectors on some DVI encoders.


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@10646 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

updated modeline tuning for panels: fixes overlay distortions on native mode on NV34Go (at least). Please report if this introduces trouble on your panel(s)!


git-svn-id: file:///srv/svn/repos/haiku

updated modeline tuning for panels: fixes overlay distortions on native mode on NV34Go (at least). Please report if this introduces trouble on your panel(s)!


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@8914 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

fixed get_timing_constraints hook, also added extra modeline validity checks befor modes are actually set in case user does not adhere to info provided in that hook.


git-svn-id: file:///srv/svn/rep

fixed get_timing_constraints hook, also added extra modeline validity checks befor modes are actually set in case user does not adhere to info provided in that hook.


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@8250 a95241bf-73f2-0310-859d-f6bbb57e9c96

show more ...

panel update for 'right-shifted' picture on some cards


git-svn-id: file:///srv/svn/repos/haiku/trunk/current@8175 a95241bf-73f2-0310-859d-f6bbb57e9c96