To quote Felix: "For testing KV with current user mode stack, please use
amdgpu. I don't expect this to work with radeon and I'm not planning to
spend any effort on making radeon work with a current user mode stack."
Only compile tested, but should be straight forward.
Signed-off-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
-----BEGIN PGP SIGNATURE-----
iQEcBAABAgAGBQJZUGOmAAoJEHm+PkMAQRiGhX8H/3fIhingPD01MBf98U0xGrJo
yIXmhu6nFs7TM0lDVDcHsKgqLQIT69ll7PrSZrMkc1RGUIPINoCuJVuJqDre0kfB
of5TX2KegqSx8h1vOWjGBCBjdYfPGyMdf9icf6KsGc/SlIdhN6WA99kglAjJA0Ve
qPTNagF0ntUNg1lsXffxyfcHqFpyqw/Z/C4ie/byFsn9iJ1VG9mNlTWSud09vhuM
3tvHzTUVAIWWuRrrgrvgqQpnwL+q5BfSDsXScMjBau0EK3RGGqG8EN6Kbkfa7VQ6
aBoeboQjUijSJnVwvySdQ11MChTIOwZdfrNPra/1HD3WJNsSu4BIRt5JcAKcOhc=
=qmSg
-----END PGP SIGNATURE-----
Backmerge tag 'v4.12-rc7' into drm-next
Linux 4.12-rc7
Needed at least rc6 for drm-misc-next-fixes, may as well go to rc7
New radeon and amdgpu features for 4.13:
- Lots of Vega10 bug fixes
- Preliminary Raven support
- KIQ support for compute rings
- MEC queue management rework from Andres
- Audio support for DCE6
- SR-IOV improvements
- Improved module parameters for controlling radeon vs amdgpu support
for SI and CIK
- Bug fixes
- General code cleanups
[airlied: dropped drmP.h header from one file was needed and build broke]
* 'drm-next-4.13' of git://people.freedesktop.org/~agd5f/linux: (362 commits)
drm/amdgpu: Fix compiler warnings
drm/amdgpu: vm_update_ptes remove code duplication
drm/amd/amdgpu: Port VCN over to new SOC15 macros
drm/amd/amdgpu: Port PSP v10.0 over to new SOC15 macros
drm/amd/amdgpu: Port PSP v3.1 over to new SOC15 macros
drm/amd/amdgpu: Port NBIO v7.0 driver over to new SOC15 macros
drm/amd/amdgpu: Port NBIO v6.1 driver over to new SOC15 macros
drm/amd/amdgpu: Port UVD 7.0 over to new SOC15 macros
drm/amd/amdgpu: Port MMHUB over to new SOC15 macros
drm/amd/amdgpu: Cleanup gfxhub read-modify-write patterns
drm/amd/amdgpu: Port GFXHUB over to new SOC15 macros
drm/amd/amdgpu: Add offset variant to SOC15 macros
drm/amd/powerplay: add avfs control for Vega10
drm/amdgpu: add virtual display support for raven
drm/amdgpu/gfx9: fix compute ring doorbell index
drm/amd/amdgpu: Rename KIQ ring to avoid spaces
drm/amd/amdgpu: gfx9 tidy ups (v2)
drm/amdgpu: add contiguous flag in ucode bo create
drm/amdgpu: fix missed gpu info firmware when cache firmware during S3
drm/amdgpu: export test ib debugfs interface
...
Commit e6b9a6c84b
("drm/radeon: Make display watermark calculations more accurate")
made watermark calculations more accurate, but not for > 4k
resolutions on 32-Bit architectures, as it introduced an integer
overflow for those setups and resolutions.
Fix this by proper u64 casting and division.
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Reported-by: Ben Hutchings <ben.hutchings@codethink.co.uk>
Fixes: e6b9a6c84b ("drm/radeon: Make display watermark calculations more accurate")
Cc: Ben Hutchings <ben.hutchings@codethink.co.uk>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Take ownership of pipe initialization away from KFD.
Note that hpd_eop_gpu_addr was already large enough to accomodate all
pipes.
Reviewed-by: Edward O'Callaghan <funfunctor@folklore1984.net>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Andres Rodriguez <andresx7@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-----BEGIN PGP SIGNATURE-----
iQEcBAABAgAGBQJZK2lrAAoJEHm+PkMAQRiGm3AH/13F1DlIk05aSXHoDr/idIpR
GMHmk3YF+EuFjsL463Sh6s/SSWmz0Lda8euaoB4wCWvQFX2ZjTE+aOd79XlRiZJQ
OTtLkV9I41eXIJUpEOHia7xZiCsbw+usqcHrm1aBoSh5KKV2iQmEOrnJdibqJVOF
eXUMphNK/zFtAd2bKtQSxkaBnOOqsQUgVQSkr2K9rSg25l0KokFC6c5K5IjLn4x9
QgDY4wmMvHrDz0CtpoqlNM4XqbsDJVrFeZGfg6hlMqSRDeXeg4h3Ol0VfIT496RP
QBdrDb6hWO+HKt9B0M+7Q+8a/Fsw+5dtpqv1W/Wlr0i4CS6euU8NChAmrpkrqGo=
=m5ba
-----END PGP SIGNATURE-----
Backmerge tag 'v4.12-rc3' into drm-next
Linux 4.12-rc3
Daniel has requested this for some drm-intel-next work.
We end up reading the interrupt register for HPD5, and then writing it
to HPD6 which on systems without anything using HPD5 results in
permanently disabling hotplug on one of the display outputs after the
first time we acknowledge a hotplug interrupt from the GPU.
This code is really bad. But for now, let's just fix this. I will
hopefully have a large patch series to refactor all of this soon.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Lyude <lyude@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
We end up reading the interrupt register for HPD5, and then writing it
to HPD6 which on systems without anything using HPD5 results in
permanently disabling hotplug on one of the display outputs after the
first time we acknowledge a hotplug interrupt from the GPU.
This code is really bad. But for now, let's just fix this. I will
hopefully have a large patch series to refactor all of this soon.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Lyude <lyude@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Avoid big roundoff errors in scanline/hactive durations for
high pixel clocks, especially for >= 500 Mhz, and thereby
program more accurate display fifo watermarks.
This is a port of the corresponding amdgpu patch.
Implemented for DCE 4,6,8.
Tested on Evergreen/DCE-4 with Radeon HD-5770.
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
At dot clocks > approx. 250 Mhz, some of these calcs will overflow and
cause miscalculation of latency watermarks, and for some overflows also
divide-by-zero driver crash. Make calcs more overflow resistant.
This is a direct port of the corresponding patch from amdgpu-kms,
copy-paste for cik from dce-8 and si from dce-6, with a slightly
simpler variant for evergreen dce-4/5.
Only tested on DCE-4 evergreen with a Radeon HD-5770.
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Use a more common logging style.
Miscellanea:
o Coalesce formats and realign arguments
o Neaten a few macros now using pr_<level>
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Set alignment mode to unaligned on CIK to align with amdgpu. This is
needed for unaligned loads to work properly in mesa. The current setting
requires dword alignment.
Signed-off-by: Marek Olšák <marek.olsak@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Replace packet2's with packet0 writes to UVD_NO_OP. The
value written to UVD_NO_OP does not matter.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
In an if block for (running == 0) running cannot be non-zero.
v2: agd: remove unused variable
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Align to the jump table offset. May fix hangs on some
asics with GFX PG enabled.
Reviewed-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Acked-by: Tom St Denis <tom.stdenis@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The power tables on some variants require different firmware.
This may fix stability issues on some newer CI parts.
bug:
https://bugs.freedesktop.org/show_bug.cgi?id=91880
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
In some cases, like when freezing for hibernation, we need to be
able to force hard reset even if no engine are stuck. This patch
add a bool option to current asic reset callback to allow to force
hard reset on asic that supports it.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This match the exact same control flow as existing code. It just
use goto instead of multiple levels of if/else. It also clarify
early initialization failures by clearing rdev->has_vce doing so
does not change end result from hardware point of view, it only
avoids printing more error messages down the line and thus only
the original error is reported.
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This match the exact same control flow as existing code. It just
use goto instead of multiple levels of if/else. It also clarify
early initialization failures by clearing rdev->has_uvd doing so
does not change end result from hardware point of view, it only
avoids printing more error messages down the line and thus only
the original error is reported.
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
I hate doing this but it hurts my eyes to go over code that does not
comply with indentation rules. Only thing that is not only space change
is in atom.c all other files are space indentation issues.
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Simplify the control flow of cik_tiling_mode_table_init() similar to how
it was done in gfx_v7_0.c and gfx_v8_0.c.
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When the radeon driver resets a gpu, it attempts to test whether all the
rings can successfully handle an IB. If these rings fail to respond, the
process will wait forever. Another gpu reset can't happen at this point,
as the current reset holds a lock required to do so. Instead, make all
the IB tests run with a timeout, so the system can attempt to recover
in this case.
While this doesn't fix the underlying issue with card resets failing, it
gives the system a higher chance of recovering. These timeouts have been
confirmed to help both a Tathi and Hawaii card recover after a gpu reset.
This also adds a new function, radeon_fence_wait_timeout, that behaves like
fence_wait_timeout. It is used instead of fence_wait_timeout as it continues
to work during a reset. radeon_fence_wait is changed to be implemented
using this function.
V2:
- Changed the timeout to 1s, as the default 10s from radeon_wait_timeout was
too long. A timeout of 100ms was tested and found to be too short.
- Changed radeon_fence_wait_timeout to behave more like fence_wait_timeout.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Matthew Dawson <matthew@mjdsystems.ca>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This merges '5b726e06d6e8309e5c9ef4109a32caf27c71dfc8' into drm-next
Just to resolve some merges to make Daniel's life easier.
Signed-off-by: DAve Airlie <airlied@redhat.com>
[airlied: fixup build problems on arm - added errno.h include]
* 'drm-next-4.5' of git://people.freedesktop.org/~agd5f/linux: (152 commits)
amd/powerplay: fix copy paste typo in hardwaremanager.c
amd/powerplay: disable powerplay by default initially
amd/powerplay: don't enable ucode fan control if vbios has no fan table
drm/amd/powerplay: show gpu load when print gpu performance for Cz. (v2)
drm/amd/powerplay: check whether need to enable thermal control. (v2)
drm/amd/powerplay: add point check to avoid NULL point hang.
drm/amdgpu/powerplay: Program a calculated value as Deep Sleep clock.
drm/amd/powerplay: Don't return an error if fan table is missing
drm/powerplay/hwmgr: log errors in tonga_hwmgr_backend_init
drm/powerplay: add debugging output to processpptables.c
drm/powerplay: add debugging output to tonga_processpptables.c
amd/powerplay: Add structures required to report configuration change
amd/powerplay: Fix get dal power level
amd\powerplay Implement get dal power level
drm/amd/powerplay: display gpu load when print performance for tonga.
drm/amdgpu/powerplay: enable sysfs and debugfs interfaces late
drm/amd/powerplay: move shared function of vi to hwmgr. (v2)
drm/amd/powerplay: check whether enable dpm in powerplay.
drm/amd/powerplay: fix bug that dpm funcs in debugfs/sysfs missing.
drm/amd/powerplay: fix boolreturn.cocci warnings
...
commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many
vblanks were missed" introduced in Linux 4.4-rc1 makes the drm core
more fragile to drivers which don't update hw vblank counters and
vblank timestamps in sync with firing of the vblank irq and
essentially at leading edge of vblank.
This exposed a problem with radeon-kms/amdgpu-kms which do not
satisfy above requirements:
The vblank irq fires a few scanlines before start of vblank, but
programmed pageflips complete at start of vblank and
vblank timestamps update at start of vblank, whereas the
hw vblank counter increments only later, at start of vsync.
This leads to problems like off by one errors for vblank counter
updates, vblank counters apparently going backwards or vblank
timestamps apparently having time going backwards. The net result
is stuttering of graphics in games, or little hangs, as well as
total failure of timing sensitive applications.
See bug #93147 for an example of the regression on Linux 4.4-rc:
https://bugs.freedesktop.org/show_bug.cgi?id=93147
This patch tries to align all above events better from the
viewpoint of the drm core / of external callers to fix the problem:
1. The apparent start of vblank is shifted a few scanlines earlier,
so the vblank irq now always happens after start of this extended
vblank interval and thereby drm_update_vblank_count() always samples
the updated vblank count and timestamp of the new vblank interval.
To achieve this, the reporting of scanout positions by
radeon_get_crtc_scanoutpos() now operates as if the vblank starts
radeon_crtc->lb_vblank_lead_lines before the real start of the hw
vblank interval. This means that the vblank timestamps which are based
on these scanout positions will now update at this earlier start of
vblank.
2. The driver->get_vblank_counter() function will bump the returned
vblank count as read from the hw by +1 if the query happens after
the shifted earlier start of the vblank, but before the real hw increment
at start of vsync, so the counter appears to increment at start of vblank
in sync with the timestamp update.
3. Calls from vblank irq-context and regular non-irq calls are now
treated identical, always simulating the shifted vblank start, to
avoid inconsistent results for queries happening from vblank irq vs.
happening from drm_vblank_enable() or vblank_disable_fn().
4. The radeon_flip_work_func will delay mmio programming a pageflip until
the start of the real vblank iff it happens to execute inside the shifted
earlier start of the vblank, so pageflips now also appear to execute at
start of the shifted vblank, in sync with vblank counter and timestamp
updates. This to avoid some races between updates of vblank count and
timestamps that are used for swap scheduling and pageflip execution which
could cause pageflips to execute before the scheduled target vblank.
The lb_vblank_lead_lines "fudge" value is calculated as the size of
the display controllers line buffer in scanlines for the given video
mode: Vblank irq's are triggered by the line buffer logic when the line
buffer refill for a video frame ends, ie. when the line buffer source read
position enters the hw vblank. This means that a vblank irq could fire at
most as many scanlines before the current reported scanout position of the
crtc timing generator as the number of scanlines the line buffer can
maximally hold for a given video mode.
This patch has been successfully tested on a RV730 card with DCE-3 display
engine and on a evergreen card with DCE-4 display engine, in single-display
and dual-display configuration, with different video modes.
A similar patch is needed for amdgpu-kms to fix the same problem.
Limitations:
- Line buffer sizes in pixels are hard-coded on < DCE-4 to a value
i just guessed to be high enough to work ok, lacking info on the true
sizes atm.
Fixes: fdo#93147
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Michel Dänzer <michel.daenzer@amd.com>
Cc: Harry Wentland <Harry.Wentland@amd.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
(v1) Tested-by: Dave Witbrodt <dawitbro@sbcglobal.net>
(v2) Refine radeon_flip_work_func() for better efficiency:
In radeon_flip_work_func, replace the busy waiting udelay(5)
with event lock held by a more performance and energy efficient
usleep_range() until at least predicted true start of hw vblank,
with some slack for scheduler happiness. Release the event lock
during waits to not delay other outputs in doing their stuff, as
the waiting can last up to 200 usecs in some cases.
Retested on DCE-3 and DCE-4 to verify it still works nicely.
(v2) Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This patch makes the IB test on the GFX ring pass for CI-based cards
installed in Big-Endian machines.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
commit 4dfd6486 "drm: Use vblank timestamps to guesstimate how many
vblanks were missed" introduced in Linux 4.4-rc1 makes the drm core
more fragile to drivers which don't update hw vblank counters and
vblank timestamps in sync with firing of the vblank irq and
essentially at leading edge of vblank.
This exposed a problem with radeon-kms/amdgpu-kms which do not
satisfy above requirements:
The vblank irq fires a few scanlines before start of vblank, but
programmed pageflips complete at start of vblank and
vblank timestamps update at start of vblank, whereas the
hw vblank counter increments only later, at start of vsync.
This leads to problems like off by one errors for vblank counter
updates, vblank counters apparently going backwards or vblank
timestamps apparently having time going backwards. The net result
is stuttering of graphics in games, or little hangs, as well as
total failure of timing sensitive applications.
See bug #93147 for an example of the regression on Linux 4.4-rc:
https://bugs.freedesktop.org/show_bug.cgi?id=93147
This patch tries to align all above events better from the
viewpoint of the drm core / of external callers to fix the problem:
1. The apparent start of vblank is shifted a few scanlines earlier,
so the vblank irq now always happens after start of this extended
vblank interval and thereby drm_update_vblank_count() always samples
the updated vblank count and timestamp of the new vblank interval.
To achieve this, the reporting of scanout positions by
radeon_get_crtc_scanoutpos() now operates as if the vblank starts
radeon_crtc->lb_vblank_lead_lines before the real start of the hw
vblank interval. This means that the vblank timestamps which are based
on these scanout positions will now update at this earlier start of
vblank.
2. The driver->get_vblank_counter() function will bump the returned
vblank count as read from the hw by +1 if the query happens after
the shifted earlier start of the vblank, but before the real hw increment
at start of vsync, so the counter appears to increment at start of vblank
in sync with the timestamp update.
3. Calls from vblank irq-context and regular non-irq calls are now
treated identical, always simulating the shifted vblank start, to
avoid inconsistent results for queries happening from vblank irq vs.
happening from drm_vblank_enable() or vblank_disable_fn().
4. The radeon_flip_work_func will delay mmio programming a pageflip until
the start of the real vblank iff it happens to execute inside the shifted
earlier start of the vblank, so pageflips now also appear to execute at
start of the shifted vblank, in sync with vblank counter and timestamp
updates. This to avoid some races between updates of vblank count and
timestamps that are used for swap scheduling and pageflip execution which
could cause pageflips to execute before the scheduled target vblank.
The lb_vblank_lead_lines "fudge" value is calculated as the size of
the display controllers line buffer in scanlines for the given video
mode: Vblank irq's are triggered by the line buffer logic when the line
buffer refill for a video frame ends, ie. when the line buffer source read
position enters the hw vblank. This means that a vblank irq could fire at
most as many scanlines before the current reported scanout position of the
crtc timing generator as the number of scanlines the line buffer can
maximally hold for a given video mode.
This patch has been successfully tested on a RV730 card with DCE-3 display
engine and on a evergreen card with DCE-4 display engine, in single-display
and dual-display configuration, with different video modes.
A similar patch is needed for amdgpu-kms to fix the same problem.
Limitations:
- Line buffer sizes in pixels are hard-coded on < DCE-4 to a value
i just guessed to be high enough to work ok, lacking info on the true
sizes atm.
Fixes: fdo#93147
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Michel Dänzer <michel.daenzer@amd.com>
Cc: Harry Wentland <Harry.Wentland@amd.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
(v1) Tested-by: Dave Witbrodt <dawitbro@sbcglobal.net>
(v2) Refine radeon_flip_work_func() for better efficiency:
In radeon_flip_work_func, replace the busy waiting udelay(5)
with event lock held by a more performance and energy efficient
usleep_range() until at least predicted true start of hw vblank,
with some slack for scheduler happiness. Release the event lock
during waits to not delay other outputs in doing their stuff, as
the waiting can last up to 200 usecs in some cases.
Retested on DCE-3 and DCE-4 to verify it still works nicely.
(v2) Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
HPD signals on DVI ports can be fired off before the pins required for
DDC probing actually make contact, due to the pins for HPD making
contact first. This results in a HPD signal being asserted but DDC
probing failing, resulting in hotplugging occasionally failing.
This is somewhat rare on most cards (depending on what angle you plug
the DVI connector in), but on some cards it happens constantly. The
Radeon R5 on the machine used for testing this patch for instance, runs
into this issue just about every time I try to hotplug a DVI monitor and
as a result hotplugging almost never works.
Rescheduling the hotplug work for a second when we run into an HPD
signal with a failing DDC probe usually gives enough time for the rest
of the connector's pins to make contact, and fixes this issue.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Lyude <cpaul@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Trying to resolve issues with missed vblanks and impossible
values inside delivered kms pageflip completion events showed
that radeon's irq handling sometimes doesn't handle valid irqs,
but silently skips them. This was observed for vblank interrupts.
Although those irqs have corresponding events queued in the gpu's
irq ring at time of interrupt, and therefore the corresponding
handling code gets triggered by these events, the handling code
sometimes silently skipped processing the irq. The reason for those
skips is that the handling code double-checks for each irq event if
the corresponding irq status bits in the irq status registers
are set. Sometimes those bits are not set at time of check
for valid irqs, maybe due to some hardware race on some setups?
The problem only seems to happen on some machine + card combos
sometimes, e.g., never happened during my testing of different PC
cards of the DCE-2/3/4 generation a year ago, but happens consistently
now on two different Apple Mac cards (RV730, DCE-3, Apple iMac and
Evergreen JUNIPER, DCE-4 in a Apple MacPro). It also doesn't happen
at each interrupt but only occassionally every couple of
hundred or thousand vblank interrupts.
This results in XOrg warning messages like
"[ 7084.472] (WW) RADEON(0): radeon_dri2_flip_event_handler:
Pageflip completion event has impossible msc 420120 < target_msc 420121"
as well as skipped frames and problems for applications that
use kms pageflip events or vblank events, e.g., users of DRI2 and
DRI3/Present, Waylands Weston compositor, etc. See also
https://bugs.freedesktop.org/show_bug.cgi?id=85203
After some talking to Alex and Michel, we decided to fix this
by turning the double-check for asserted irq status bits into a
warning. Whenever a irq event is queued in the IH ring, always
execute the corresponding interrupt handler. Still check the irq
status bits, but only to log a DRM_DEBUG message on a mismatch.
This fixed the problems reliably on both previously failing
cards, RV-730 dual-head tested on both crtcs (pipes D1 and D2)
and a triple-output Juniper HD-5770 card tested on all three
available crtcs (D1/D2/D3). The r600 and evergreen irq handling
is therefore tested, but the cik an si handling is only compile
tested due to lack of hw.
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Mario Kleiner <mario.kleiner.de@gmail.com>
CC: Michel Dänzer <michel.daenzer@amd.com>
CC: Alex Deucher <alexander.deucher@amd.com>
CC: <stable@vger.kernel.org> # v3.16+
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
In order for hibernation to reliably work we need to cleanup more
thoroughly the compute ring. Hibernation is different from suspend
resume as when we resume from hibernation the hardware is first
fully initialize by regular kernel then freeze callback happens
(which correspond to a suspend inside the radeon kernel driver)
and turn off each of the block. It turns out we were not cleanly
shutting down the compute ring. This patch fix that.
Hibernation and suspend to ram were tested (several times) on :
Bonaire
Hawaii
Mullins
Kaveri
Kabini
Changed since v1:
- Factor the ring stop logic into a function taking ring as arg.
Cc: stable@vger.kernel.org
Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEcBAABAgAGBQJVa7zvAAoJEHm+PkMAQRiGtfMIAILs3sxFtrC1hApgcfRLF/7z
K34bwTRqErzqUO/orTwakEr9kSIpIL0zIPSryTCOTPZLfMGkQjhHXO3KR/DSbbTV
MZ8y/BM/yelFA/Np+1LjbiYjTNRnTRvCoaQihkIH8Rn02g7ob9HyL4gIGKpuGFcZ
04GacL2cgChqsRSACdNef948jCoJXKgcuDpe39DXphDWZnBKNZ3HFuJ6bryGJf9A
1/eCI4is85BNwKPemQUYR0xx83UIzDfrghatZP2mOCDDSA2MNg8HNxLTd12LGoQD
tfgX4B7aftzW9Y7GSEDfZ0IKm2NRzgPmCVj6PjVR/iI0lIK4Aq0Z/lDJxxEq3XQ=
=AJM5
-----END PGP SIGNATURE-----
Merge tag 'v4.1-rc6' into drm-next
Linux 4.1-rc6
backmerge 4.1-rc6 as some of the later pull reqs are based on newer bases
and I'd prefer to do the fixup myself.
We have that bug for years and some users report side effects when fixing it on older hardware.
So revert it for VM_CONTEXT0_PAGE_TABLE_END_ADDR, but keep it for VM 1-15.
Signed-off-by: Christian König <christian.koenig@amd.com>
CC: stable@vger.kernel.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The mapping range is inclusive between starting and ending addresses.
Signed-off-by: Christian König <christian.koenig@amd.com>
CC: stable@vger.kernel.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Registers that can be fetched from the info ioctl.
Tested-by: Marek Olšák <marek.olsak@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Enable at init and disable on fini. Workaround for hardware problems.
v2 (chk): extend commit message
v3: add new function
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com> (v2)
Cc: stable@vger.kernel.org
Emit the EOP twice to avoid cache flushing problems.
Signed-off-by: Christian König <christian.koenig@amd.com>
Cc: stable@vger.kernel.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Slava Grigorev <slava.grigorev@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Slava Grigorev <slava.grigorev@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Backmerge Linus tree after rc5 + drm-fixes went in.
There were a few amdkfd conflicts I wanted to avoid,
and Ben requested this for nouveau also.
Conflicts:
drivers/gpu/drm/amd/amdkfd/Makefile
drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c
drivers/gpu/drm/amd/amdkfd/kfd_priv.h
drivers/gpu/drm/amd/include/kgd_kfd_interface.h
drivers/gpu/drm/i915/intel_runtime_pm.c
drivers/gpu/drm/radeon/radeon_kfd.c
We need to wait for the GPUVM flush to complete. There
was some confusion as to how this mechanism was supposed
to work. The operation is not atomic. For GPU initiated
invalidations you need to read back a VM register to
introduce enough latency for the update to complete.
v2: drop gart changes
v3: just read back rather than polling
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Christian König