If the user has created a read-only object, they should not be allowed
to circumvent the write protection by using a GGTT mmapping. Deny it.
Also most machines do not support read-only GGTT PTEs, so again we have
to reject attempted writes. Fortunately, this is known a priori, so we
can at least reject in the call to create the mmap (with a sanity check
in the fault handler).
v2: Check the vma->vm_flags during mmap() to allow readonly access.
v3: Remove VM_MAYWRITE to curtail mprotect()
Testcase: igt/gem_userptr_blits/readonly_mmap*
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Jon Bloomfield <jon.bloomfield@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Matthew Auld <matthew.william.auld@gmail.com>
Cc: David Herrmann <dh.herrmann@gmail.com>
Reviewed-by: Matthew Auld <matthew.william.auld@gmail.com> #v1
Reviewed-by: Jon Bloomfield <jon.bloomfield@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180712185315.3288-4-chris@chris-wilson.co.uk
GVT is not propagating the PTE bits, and is always setting the
read-write bit, thus breaking read-only support.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Zhenyu Wang <zhenyuw@linux.intel.com>
Cc: Jon Bloomfield <jon.bloomfield@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Matthew Auld <matthew.william.auld@gmail.com>
Reviewed-by: Jon Bloomfield <jon.bloomfield@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180712185315.3288-3-chris@chris-wilson.co.uk
Hook up the flags to allow read-only ppGTT mappings for gen8+
v2: Include a selftest to check that writes to a readonly PTE are
dropped
v3: Don't duplicate cpu_check() as we can just reuse it, and even worse
don't wholesale copy the theory-of-operation comment from igt_ctx_exec
without changing it to explain the intention behind the new test!
v4: Joonas really likes magic mystery values
Signed-off-by: Jon Bloomfield <jon.bloomfield@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Matthew Auld <matthew.william.auld@gmail.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.william.auld@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180712185315.3288-2-chris@chris-wilson.co.uk
We can set a bit inside the ppGTT PTE to indicate a page is read-only;
writes from the GPU will be discarded. We can use this to protect pages
and in particular support read-only userptr mappings (necessary for
importing PROT_READ vma).
Signed-off-by: Jon Bloomfield <jon.bloomfield@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Matthew Auld <matthew.william.auld@gmail.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.william.auld@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180712185315.3288-1-chris@chris-wilson.co.uk
On GLK NUC platforms the HDMI retiming buffer needs additional disabled
time to correctly sync to a faster incoming signal.
When measured on a scope the highspeed lines of the HDMI clock turn off
for ~400uS during a normal resolution change. The HDMI retimer on the
GLK NUC appears to require at least a full frame of quiet time before a
new faster clock can be correctly sync'd. Wait 100ms due to msleep
inaccuracies while waiting for a completed frame. Add a quirk to the
driver for GLK boards that use ITE66317 HDMI retimers.
V2: Add more devices to the quirk list
V3: Delay increased to 100ms, check to confirm crtc type is HDMI.
V4: crtc type check extended to include _DDI and whitespace fixes
v5: Fix white spaces, remove the macro for delay. Revert the crtc type
check introduced in v4.
Cc: Imre Deak <imre.deak@intel.com>
Cc: <stable@vger.kernel.org> # v4.14+
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105887
Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com>
Tested-by: Daniel Scheller <d.scheller.oss@gmail.com>
Signed-off-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180710200205.1478-1-radhakrishna.sripada@intel.com
This patchs adds the cec_notifier feature to the intel_hdmi part
of the i915 DRM driver. It uses the HDMI DRM connector name to differentiate
between each HDMI ports.
The changes will allow the i915 HDMI code to notify EDID and HPD changes
to an eventual CEC adapter.
Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Reviewed-by: Hans Verkuil <hans.verkuil@cisco.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Reduce the module parameter to enable or disable.
The link stand by vs full link off was used only once.
And it was actually masking another bug fixed by commit
'84bb2916a683 ("drm/i915/psr: Check for SET_POWER_CAPABLE
bit at PSR init time.")'
So, let's remove these options for now. End goal is to
fully remove the mod param, moving it to a debugfs
interface in upcoming patches.
Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Cc: Tarun Vyas <tarun.vyas@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180712052715.8177-1-rodrigo.vivi@intel.com
This allows to read i915_edp_psr_status from tests without triggering
any AUX communication. Take this opportunity to move this under the
eDP-1 connector directory as the status we print is of the sink.
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: José Roberto de Souza <jose.souza@intel.com>
Suggested-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180705003121.2478-1-dhinakaran.pandiyan@intel.com
In commit "drm/i915: Wait for PSR exit before checking for vblank
evasion", the idea was to limit the PSR IDLE checks when PSR is
actually supported. While CAN_PSR does do that check, it doesn't
applies on a per-crtc basis. crtc_state->has_psr is a more granular
check that only applies to pipe(s) that have PSR enabled.
Without the has_psr check, we end up waiting on the eDP transcoder's
PSR_STATUS register irrespective of whether the pipe being updated is
driving it or not.
v2: Remove unnecessary parantheses, make checkpatch happy.
v3: Move the has_psr check to intel_psr_wait_for_idle and commit
message changes (DK).
v4: Derive dev_priv from intel_crtc_state (DK)
v5: Commit message changes to reflect the HW behavior (DK)
Fixes: a608987970 ("drm/i915: Wait for PSR exit before checking for vblank evasion")
Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Signed-off-by: Tarun Vyas <tarun.vyas@intel.com>
Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180712053323.26266-1-tarun.vyas@intel.com
Support for Burst read in HW is added for HDCP2.2 compliance
requirement.
This patch enables the burst read for all the gmbus read of more than
511Bytes, on capable platforms.
v2:
Extra line is removed.
v3:
Macro is added for detecting the BURST_READ Support [Jani]
Runtime detection of the need for burst_read [Jani]
Calculation enhancement.
v4:
GMBUS0 reg val is passed from caller [ville]
Removed a extra var [ville]
Extra brackets are removed [ville]
Implemented the handling of 512Bytes Burst Read.
v5:
Burst read max length is fixed at 767Bytes [Ville]
v6:
Collecting the received reviewed-by.
Signed-off-by: Ramalingam C <ramalingam.c@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/1530192889-5789-3-git-send-email-ramalingam.c@intel.com
GMBUS HW supports 511Bytes as Max Bytes per single RD/WR op. Instead of
enabling the 511Bytes per RD/WR cycle on legacy platforms for no
absolute ROIs, this change allows the max bytes per op upto 511Bytes
from Gen9 onwards.
v2:
No Change.
v3:
Inline function for max_xfer_size and renaming of the macro.[Jani]
v4:
Extra brackets removed [ville]
Commit msg is modified.
v5:
Collecting the Reviewed-By received.
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Ramalingam C <ramalingam.c@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/1530192889-5789-2-git-send-email-ramalingam.c@intel.com
Since:
0d4b78b3d2 ("drm/i915/guc: Assert we have the doorbell before setting it up")
We have asserts in GuC doorbell related functions, which is a good thing.
Unfortunately, we were using those to check whether GuC FW is refusing
to allocate invalid doorbell - which makes the test fail.
Well, it would make the test WARN, except we fumbled cleanup ordering
and eat the BUG_ON instead.
Let's keep the asserts and use the internal implementation in the test.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107186
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180712112013.3253-1-chris@chris-wilson.co.uk
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Let's reorder things so that we can do onion teardown rather than double
goto.
References: b96f6ebfd0 ("drm/i915: Correctly handle error path in i915_gem_init_hw")
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180712124810.25241-1-michal.winiarski@intel.com
If we fail the module load, we may try and cleanup before we even
allocate the GuC clients. KISS in order to try and re-enable
drv_module_reload for BAT.
Testcase: igt/drv_module_reload/basic-reload-inject
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180712105830.20390-1-chris@chris-wilson.co.uk
The kernel recently gained an augmented rbtree with the purpose of
cacheing the leftmost element of the rbtree, a frequent optimisation to
avoid calls to rb_first() which is also employed by the
execlists->queue. Switch from our open-coded cache to the library.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180629075348.27358-9-chris@chris-wilson.co.uk
Since live_workarounds poke around the w/a registers and checks to see
if they survive across a reset, we are prone to fouling the machine and
leaving it in a non-recoverable state. Wrap the probe inside a timeout
to abort the test if the reset fails.
v2: Include GEM_TRACE on declaring wedged.
v3: Add a few includes to make the header look standalone.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107188
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180711122952.18448-1-chris@chris-wilson.co.uk
Add a mutex into struct i915_address_space to be used while operating on
the vma and their lists for a particular vm. As this may be called from
the shrinker, we taint the mutex with fs_reclaim so that from the start
lockdep warns us if we are caught holding the mutex across an
allocation. (With such small steps we will eventually rid ourselves of
struct_mutex recursion!)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20180711073608.20286-2-chris@chris-wilson.co.uk
Now that our stolen memory is already reserved by the x86 subsystem
(since commit "x86/gpu: reserve ICL's graphics stolen memory"), make
use of it.
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: x86@kernel.org
Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180504203252.28048-2-paulo.r.zanoni@intel.com
On unwinding following a critical failure inside GEM init, we also need
to be sure to flush the workers before unloading the module.
Testcase: igt/drv_module_reload/basic-reload-inject
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180710094421.16223-1-chris@chris-wilson.co.uk
In the next patch, we will make a fairly minor change to flush
outstanding resets before suspend. In order to keep churn to a minimum
in that functional patch, we fix up the comments and coding style now.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180709130208.11730-7-chris@chris-wilson.co.uk
Across a reset, the seqno (and thus hangcheck) should restart and the
hangcheck naturally progress, for when it does not, we want to declare an
emergency. Currently, we only detect if reset and reinit fails, but we
do not detect if the call to reinit succeeds but the HW is fried - as we
are resetting hangcheck on initialisation the engine. Remove that and
rely on the natural progress to reset the hangcheck timer.
References: e21b141376 ("drm/i915: Mark the hangcheck as idle when unparking the engines")
References: 1fd00c0fae ("drm/i915: Declare the driver wedged if hangcheck makes no progress")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180709130208.11730-2-chris@chris-wilson.co.uk
In our swizzling selftests, we cannot predict the physical address of
the target page (at least not simply!) and so skip bit17 swizzles.
However, there are two bit17 swizzle modes and we only skipped one, with
the second being observed on the lab gdg causing the test to fail,
as soon as we hit a page with bit17 set in its address.
Testcase: igt/drv_selftest/live_objects #gdg
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180709194915.5789-1-chris@chris-wilson.co.uk
Be pessimistic and presume that we actually allocate every page we
exercise via the mock_gtt (e.g. for gvt). In which case we have to keep
our working set under the available physical memory to prevent oom.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180710080424.7821-1-chris@chris-wilson.co.uk
Error messages are intended to be addressed to the user; be clear,
succinct, instructive and unambiguous. Adding the function name to
that message does not add any information the user requires and in
the process makes the message less clear.
E.g.
[ 245.539711] i915 0000:00:02.0: [drm:i915_gem_init [i915]] Failed to initialize GPU, declaring it wedged!
becomes
[ 245.539711] i915 0000:00:02.0: Failed to initialize GPU, declaring it wedged!
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180709134858.12446-1-chris@chris-wilson.co.uk
This helps initramfs builder and other tools to know the full dependencies
of i915 and have gvt module loaded with i915.
v2: add condition and change to pre-dependency (Chris)
v3: move declaration to gvt.c. (Chris)
v4: remove xengt (Zhenyu)
Signed-off-by: Hang Yuan <hang.yuan@linux.intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Just like with PIPESTAT, the edge triggered IIR on i965/g4x
also causes problems for hotplug interrupts. To make sure
we don't get the IIR port interrupt bit stuck low with the
ISR bit high we must force an edge in ISR. Unfortunately
we can't borrow the PIPESTAT trick and toggle the enable
bits in PORT_HOTPLUG_EN as that act itself generates hotplug
interrupts. Instead we just have to loop until we've cleared
PORT_HOTPLUG_STAT, or we just give up and WARN.
v2: Don't frob with PORT_HOTPLUG_EN
Cc: stable@vger.kernel.org
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180614175625.1615-1-ville.syrjala@linux.intel.com
Reviewed-by: Imre Deak <imre.deak@intel.com>
(cherry picked from commit 0ba7c51a6f)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
In igt_flush_test() we install a background timer in order to ensure
that the wait completes within a certain time. We can now tell the wait
that it has to complete within a timeout, and so no longer need the
background timer.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180709122044.7028-3-chris@chris-wilson.co.uk
With a broken GPU we expect it to fail during the initial
GPU setup where do a couple of context switches to record the defaults.
This is a task that takes a few milliseconds even on the slowest of
devices, but we may have to wait 60s for hangcheck to give in and
declare the machine inoperable. In this a case where any gpu hang is
unacceptable, both from a timeliness and practical standpoint.
We can therefore set a timeout on our wait-for-idle that is shorter than
the hangcheck (which may be up to 60s for a declaring a wedged driver)
and so detect the broken GPU much more quickly during driver load (and
so prevent stalling userspace for ages).
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180709122044.7028-2-chris@chris-wilson.co.uk
Usually we have no idea about the upper bound we need to wait to catch
up with userspace when idling the device, but in a few situations we
know the system was idle beforehand and can provide a short timeout in
order to very quickly catch a failure, long before hangcheck kicks in.
In the following patches, we will use the timeout to curtain two overly
long waits, where we know we can expect the GPU to complete within a
reasonable time or declare it broken.
In particular, with a broken GPU we expect it to fail during the initial
GPU setup where do a couple of context switches to record the defaults.
This is a task that takes a few milliseconds even on the slowest of
devices, but we may have to wait 60s for hangcheck to give in and
declare the machine inoperable. In this a case where any gpu hang is
unacceptable, both from a timeliness and practical standpoint.
The other improvement is that in selftests, we do not need to arm an
independent timer to inject a wedge, as we can just limit the timeout on
the wait directly.
v2: Include the timeout parameter in the trace.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180709122044.7028-1-chris@chris-wilson.co.uk
BXT supports EDP. However since GVT-g only simulate DP monitor
to guest and handles EDP_PSR_IMR and EDP_PSR_IIR as default MMIO
r/w. If guest r/w these IMR/IIR, GVT-g won't simulate the real
HW behavior and below warning is printed:
--------
Interrupt register 0x64838 is not zero: 0xffffffff
WARNING: CPU: 1 PID: 1 at drivers/gpu/drm/i915/i915_irq.c:161
gen3_assert_iir_is_zero+0x34/0xa0
Call Trace:
gen8_de_irq_postinstall+0xad/0x330
gen8_irq_postinstall+0x23/0x80
drm_irq_install+0xb5/0x130
i915_driver_load+0xafd/0xf70
--------
Since GVT-g won't simulate EDP to guest, always set EDP_PSR_IMR
and EDP_PSR_IIR IMR/IIR to 0.
Signed-off-by: Colin Xu <colin.xu@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Now GVTg supports shadowing both 2M/64K huge gtt pages. So let's turn on
the cap info bit VGT_CAPS_HUGE_GTT.
v2: Split changes in i915 side into a separated patch.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Don't forget to free allocated spt if shadowing failed.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
If the guest update the 64K gtt entry before changing IPS bit of PDE, we
need to re-shadow the whole page table. Because we have ignored all
updates to unused entries.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
This add 2M huge gtt support for GVTg. Unlike 64K gtt entry, we can
shadow 2M guest entry with real huge gtt. But before that, we have to
check memory physical continuous, alignment and if it is supported on
the host. We can get all supported page sizes from
intel_device_info.page_sizes.
Finally we must split the 2M page into smaller pages if we cannot
satisfy guest Huge Page.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
To support huge gtt, we need to support huge pages in kvmgt first.
This patch adds a 'size' param to the intel_gvt_mpt::dma_map_guest_page
API and implements it in kvmgt.
v2: rebase.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Finally, this add the first huge gtt support for GVTg - 64K pages. Since
64K page and 4K page cannot be mixed on the same page table, so we always
split a 64K entry into small 4K page. And when unshadow guest 64K entry,
we need ensure all the shadowed entries in shadow page table also get
cleared.
For page table which has 64K gtt entry, only PTE#0, PTE#16, PTE#32, ...
PTE#496 are used. Unused PTEs update should be ignored.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
64K PTE is special, only PTE#0, PTE#16, PTE#32, ... PTE#496 are used in
the page table.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
We need a interface to allocate a pure shadow page which doesn't have
a guest page associated with. Such shadow page is used to shadow 2M
huge gtt entry.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Add clear_pse operation in case we need to split huge gtt into small pages.
v2: correct description.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
This add a software PTE flag on the Ignored bit of PTE. It will be used
to identify splited 64K shadow entries.
v2: fix mask definition.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
This change help us detect the real entry type per PSE and IPS setting.
For 64K entry, we also need to check reg GEN8_GAMW_ECO_DEV_RW_IA.
v2: Extend IPS mmio control to Gen10. (Matthew Auld)
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
The register RENDER_HWS_PGA_GEN7 is renamed to GEN8_GAMW_ECO_DEV_RW_IA
from GEN8 which can control IPS enabling.
v3: MMIO control for IPS is not removed from gen9 but gen10 (Matthew Auld)
v2: IPS of all engines must be enabled together for gen9.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Add three IPS operation functions to test/set/clear IPS in PDE.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Add a new entry type GTT_TYPE_PPGTT_PTE_64K_ENTRY. 64K entry is very
different from 2M/1G entry. 64K entry is controlled by IPS bit in upper
PDE. To leverage the current logic, I take IPS bit as 'PSE' for PTE
level. Which means, 64K entries can also processed by get_pse_type().
v2: Make it bisectable.
Signed-off-by: Changbin Du <changbin.du@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
In the next patch, we will want a third distinct class of timeline that
may overlap with the current pair of client and engine timeline classes.
Rather than use the ad hoc markup of SINGLE_DEPTH_NESTING, initialise
the different timeline classes with an explicit subclass.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706210710.16251-1-chris@chris-wilson.co.uk
Inside the mock GEM device, we try to grab the runtime pm for the fake
device to prevent it from ever suspending. However, if CONFIG_PM is not
set, trying to obtain the wakref returns an error which we WARN about.
Suppress the expected warning.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706205947.11209-1-chris@chris-wilson.co.uk
clflush is an unserialised instruction and the IA manual strongly advises
you to serialise it with a mb. To be cautious, apply one before and one
after, so that it is serialised with both writes and reads without
worrying too much about the required direction.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706174926.4712-1-chris@chris-wilson.co.uk
Using a VMA on more than one timeline concurrently is the exception
rather than the rule (using it concurrently on multiple engines). As we
expect to only use one active tracker, store the most recently used
tracker inside the i915_vma itself and only fallback to the rbtree if
we need a second or more concurrent active trackers.
v2: Comments on how we overwrite any existing last_active cache.
v3: __list_del_entry() before list_replace_init() is confusing and, much
more important, entirely redundant.
v4: Note that both last_active and the rbtree may be simultaneously
tracking this timeline, albeit with different requests, and so the vma
may be retired twice for the same timeline.
v5: No, that list_del is required!
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706123157.9645-1-chris@chris-wilson.co.uk
In the next patch, we will want to be able to use more flexible request
timelines that can hop between engines. From the vma pov, we can then
not rely on the binding of this request to an engine and so can not
ensure that different requests are ordered through a per-engine
timeline, and so we must track activity of all timelines. (We track
activity on the vma itself to prevent unbinding from HW before the HW
has finished accessing it.)
v2: Switch to a rbtree for 32b safety (since using u64 as a radixtree
index is fraught with aliasing of unsigned longs).
v3: s/lookup_active/active_instance/ because we can never agree on names
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706103947.15919-5-chris@chris-wilson.co.uk
Handling such a late error in request construction is tricky, but to
accommodate future patches which may allocate here, we potentially could
err. To handle the error after already adjusting global state to track
the new request, we must finish and submit the request. But we don't
want to use the request as not everything is being tracked by it, so we
opt to cancel the commands inside the request.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706103947.15919-3-chris@chris-wilson.co.uk
In the next patch, we will want to start skipping requests on failing to
complete their payloads. So export the utility function current used to
make requests inoperable following a failed gpu reset.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706103947.15919-2-chris@chris-wilson.co.uk
Currently all callers are responsible for adding the vma to the active
timeline and then exporting its fence. Combine the two operations into
i915_vma_move_to_active() to move all the extra handling from the
callers to the single site.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706103947.15919-1-chris@chris-wilson.co.uk
Pull in the malidp writeback implementation for further work on writeback in drm-misc-next.
Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.com>
Replace the magic bit with the proper symbolic name for instructing
MI_STORE_DWORD_IMM to use a virtual address (on gen3) or the global GTT
address (still virtual!) on gen4+.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706142323.25699-1-chris@chris-wilson.co.uk
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Limit the GTT size we try and allocate to ensure that it fits within RAM
and does not trigger the oomkiller indiscriminately.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706125338.24432-1-chris@chris-wilson.co.uk
We already maually control the CPU cache for our page table directories,
so we can tell the dma mapper to skip doing it as well.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706122611.4142-2-chris@chris-wilson.co.uk
As we propagate back the error to the caller for them to handle, we do
not need the lowest level spitting out a redundant warning upon an
allocation failure inside dma_map_page().
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706122611.4142-1-chris@chris-wilson.co.uk
If we have just completed a WC write, we must ensure that the WCB (Write
Combining Buffer) is flushed out to main memory before we can expect to
see the results. This is especially important when mixing WC with GTT as
the physical paths are different and cachelines are not naturally flushed.
Testcase: igt/drv_selftests/live_coherency #gdg
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706115402.18547-1-chris@chris-wilson.co.uk
If the GPU is irrecoverably wedged, we can not execute any requests
making testing execlists (request execution) pointless. Skip!
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706114510.18467-1-chris@chris-wilson.co.uk
If the HW (or driver) doesn't support logical contexts, don't pretend we
gain anything from trying to execute GPU commands with them. At best it
reports -ENODEV, which is an unhelpful failure that we should just skip.
v2: Be more specific and check the driver/engine caps for logical (HW)
context support.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706101923.28548-1-chris@chris-wilson.co.uk
Avoid looking at the magical engines[RCS] to decide if the HW and driver
supports logical contexts, and instead record that knowledge during
initialisation.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706101442.21279-1-chris@chris-wilson.co.uk
We can simplify the encoder's get_power_domains() hook by calling it
only if the encoder is active. That way the hook can return its power
domains unconditionally without checking the active state by calling
encoder::get_hw_state(). This get_hw_state() query is in fact
redundant since it's already done by intel_modeset_readout_hw_state()
setting the encoder's crtc or leaving it NULL accordingly. Let's use
this fact to decide if the encoder is active.
While at it clarify the comment in intel_ddi_get_power_domains() about
primary vs. fake MST encoders and make sure we never do an incorrect
encoder->dig_port cast for fake MST encoders.
Suggested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180705122654.17072-1-imre.deak@intel.com
This interface is deprecated, and has been replaced by the upstream
drm crc interface.
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Tomi Sarvela <tomi.p.sarvela@intel.com>
Cc: Petri Latvala <petri.latvala@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Acked-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628072303.14175-1-maarten.lankhorst@linux.intel.com
If the GPU is terminally wedged we cannot submit any requests into a
context, completely unfulfilling our purpose of doing so. As this
expectedly fails, skip over the test.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706065332.15214-9-chris@chris-wilson.co.uk
We test the GPU handling of huge pages by submitting requests that write
into a huge page, but if the GPU is irrecoverably wedged we cannot
submit any requests. As the test expectedly fails, skip over it.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706065332.15214-8-chris@chris-wilson.co.uk
If the GPU is irrecoverably wedged, we cannot submit any requests and so
cannot make the GTT busy in order to test evicting active objects. As
this expectedly fails, skip over the test.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706065332.15214-7-chris@chris-wilson.co.uk
If the GPU is irrecoverably wedged, we cannot submit any request and
therefore cannot query the register state of the context (which is done
using the GPU command stream). So skip over the test as it expectedly
fails.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706065332.15214-6-chris@chris-wilson.co.uk
As we keep VMA around until the object is destroyed, when testing
partial tiling we instantiate many, many VMA (as the object is huge
allowing for many different partial regions). We test elsewhere our
handling of populating large objects with a full set of VMA and checking
we can retrieve them afterwards, but in this test we incur the cost of
flushing all VMA after every GTT write, dramatically slowing down the
test.
References: https://bugs.freedesktop.org/show_bug.cgi?id=107130
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706065332.15214-2-chris@chris-wilson.co.uk
Having found the error causing the IGT test to fail, downgrade the
verbose logging so that we stop flooding the syslogs as we deliberately
provoke it many thousands of time during selftests.
References: 10195b1e44 ("drm/i915: Show vma allocator stack when in doubt")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180706065332.15214-1-chris@chris-wilson.co.uk
This patch defines AUX lane registers for PORT_PCS_DW1,
PORT_TX_DW2, PORT_TX_DW4, PORT_TX_DW5 used during
dsi enabling.
v2: Review comments from Jani N:
- Define _ICL_PORT_PCS_DW1_AUX_A for consistency
- Three spaces for bitfield definition.
Signed-off-by: Madhav Chauhan <madhav.chauhan@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1530798591-2077-8-git-send-email-madhav.chauhan@intel.com
To save power, unused lanes should be powered
down using the bitfield of PORT_CL_DW10.
v2: Review comments from Jani N
- Put default label next to case 4
- Include the shifts in the macros
Signed-off-by: Madhav Chauhan <madhav.chauhan@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1530798591-2077-7-git-send-email-madhav.chauhan@intel.com
This register used to power down individual lanes for
DDI/DSI ports. Bitfields to power up/down various
combinations of lanes are also added in this patch.
v2: Review comments from Jani N
- Use override instead of "override" for bitfields
- Define mask for override bitfield
- Define PWR_DOWN_LN* macros shifted in place
v3: Correct PWR_DOWN_LN_MASK value (Jani N)
Signed-off-by: Madhav Chauhan <madhav.chauhan@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1530798591-2077-6-git-send-email-madhav.chauhan@intel.com
Escape Clock is used for LP communication across the DSI
Link. To achieve the constant frequency of the escape clock
from the variable DPLL frequency output, a variable divider(M)
is needed. This patch programs the same.
v2: (Jani N) Don't end line with "(".
Signed-off-by: Madhav Chauhan <madhav.chauhan@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1530798591-2077-3-git-send-email-madhav.chauhan@intel.com
Use the more customary order of latest platform first, and don't bother
with an if in the last branch.
Cc: Madhav Chauhan <madhav.chauhan@intel.com>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Madhav Chauhan <madhav.chauhan@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180705132509.12881-3-jani.nikula@intel.com
Avoid confusion with the functions to be added for the new ICL or gen 11
DSI implementation by renaming the current DSI functions. While at it,
permutate the words in the function names to make them all start with
"vlv_dsi" or "vlv_dsi_pll" etc.
Reduce the platform abstractions in the PLL file while at it, moving the
checks to vlv_dsi.c instead, where we typically already have the
necessary if ladders.
Leave the static functions as-is for now; they could be renamed later if
needed.
No functional changes.
v2: use "gen7" prefix.
v3: use "vlv" and "bxt" prefixes, reduce the abstractions.
References: https://patchwork.freedesktop.org/series/44823/
Cc: Madhav Chauhan <madhav.chauhan@intel.com>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Madhav Chauhan <madhav.chauhan@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180705132509.12881-2-jani.nikula@intel.com
Starting from ICL or gen 11 we have a new DSI block which requires
completely different programming from the current implementation. Having
them in the same file would be confusing. Rename the current DSI and DSI
PLL implementation files as vlv_dsi.c and vlv_dsi_pll.c.
No functional changes.
v2: use "gen7" prefix.
v3: use "vlv" prefix.
References: https://patchwork.freedesktop.org/series/44823/
Cc: Madhav Chauhan <madhav.chauhan@intel.com>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Madhav Chauhan <madhav.chauhan@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180705132509.12881-1-jani.nikula@intel.com
If the GPU is irrecoverably wedged on startup, it means that it failed
on initialisation and we have already tried to reset it but failed. We
can ignore all further testing, as it is already dead. Failing early,
prevents us from slowly failing in our endeavours later and timing out.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180705150214.28316-1-chris@chris-wilson.co.uk
i915_gem_detect_bit_6_swizzle() tries to hide unknown swizzling from
userspace (and ourselves) leaving us with the only clue inside
i915->quirks & QUIRK_PIN_SWIZZLED_PAGES. If we see this bit set, it
means that we really have no clue as to what the swizzle pattern is
being used in any one page and so cannot compute what the reference
value should be in our tiling selftests. We have to skip the test.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107133
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180705171523.18462-1-chris@chris-wilson.co.uk
With the fb-helper no longer relying on the non-atomic .best_encoder()
we can eliminate the hook from the MST encoder.
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628131315.14156-3-ville.syrjala@linux.intel.com
In preparation to enabling -Wimplicit-fallthrough, mark switch cases
where we are expecting to fall through.
Addresses-Coverity-ID: 141432
Addresses-Coverity-ID: 141433
Addresses-Coverity-ID: 141434
Addresses-Coverity-ID: 141435
Addresses-Coverity-ID: 141436
Addresses-Coverity-ID: 1357360
Addresses-Coverity-ID: 1357403
Addresses-Coverity-ID: 1357433
Addresses-Coverity-ID: 1392622
Addresses-Coverity-ID: 1415273
Addresses-Coverity-ID: 1435752
Addresses-Coverity-ID: 1441500
Addresses-Coverity-ID: 1454596
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628223541.GA17665@embeddedor.com
This patch adds the new registers and corresponding bit definitions
which will be used for programming/enable DSI PLL.
v2: Review comments from Jani N
- Fix spaces while defining ICL_ESC_CLK_DIV_MASK
- Define shift and mask for bitfields.
Signed-off-by: Madhav Chauhan <madhav.chauhan@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1530795727-28644-2-git-send-email-madhav.chauhan@intel.com
Commit cd7e 61b9"init mmio by lri command in vgpu inhibit context"
initializes registers saved/restored in context with its vreg value
through lri command in ring buffer. It relies on vreg got updated
on every guest access. There is a case found that Linux guest uses
lri command in inhibit-ctx to update the register. This patch adds
vreg update on this case.
v2: move mmio_attribute functions to gvt.h (Zhenyu)
v3: use mask_mmio_write in vreg update
v4: refine codes and add more comments (Zhenyu)
Fixes: cd7e61b9("drm/i915/gvt: init mmio by lri command in vgpu inhibit context")
Signed-off-by: Hang Yuan <hang.yuan@linux.intel.com>
Signed-off-by: Weinan Li <weinan.z.li@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Currently, the wc-stash used for providing flushed WC pages ready for
constructing the page directories is assumed to be protected by the
struct_mutex. However, we want to remove this global lock and so must
install a replacement global lock for accessing the global wc-stash (the
per-vm stash continues to be guarded by the vm).
We need to push ahead on this patch due to an oversight in hastily
removing the struct_mutex guard around the igt_ppgtt_alloc selftest. No
matter, it will prove very useful (i.e. will be required) in the near
future.
v2: Restore the onstack stash so that we can drop the vm->mutex in
future across the allocation.
v3: Restore the lost pagevec_init of the onstack allocation, and repaint
function names.
v4: Reorder init so that we don't try and use i915_address_space before
it is ininitialised.
Fixes: 1f6f00238a ("drm/i915/selftests: Drop struct_mutex around lowlevel pggtt allocation")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180704185518.4193-1-chris@chris-wilson.co.uk
Adjust the EIR clearing to cope with the edge triggered IIR
on i965/g4x. To guarantee an edge in the ISR master error bit
we temporarily mask everything in EMR. As some of the EIR bits
can't even be directly cleared we also borrow a trick from
i915_clear_error_registers() and permanently mask any bit that
remains high. No real thought given to how we might unmask them
again once the cause for the error has been clered. I suppose
on pre-g4x GPU reset will reinitialize EMR from scratch.
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180611200258.27121-3-ville.syrjala@linux.intel.com
Reviewed-by: Imre Deak <imre.deak@intel.com>
Just like with PIPESTAT, the edge triggered IIR on i965/g4x
also causes problems for hotplug interrupts. To make sure
we don't get the IIR port interrupt bit stuck low with the
ISR bit high we must force an edge in ISR. Unfortunately
we can't borrow the PIPESTAT trick and toggle the enable
bits in PORT_HOTPLUG_EN as that act itself generates hotplug
interrupts. Instead we just have to loop until we've cleared
PORT_HOTPLUG_STAT, or we just give up and WARN.
v2: Don't frob with PORT_HOTPLUG_EN
Cc: stable@vger.kernel.org
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180614175625.1615-1-ville.syrjala@linux.intel.com
Reviewed-by: Imre Deak <imre.deak@intel.com>
Two requests have come in for a backmerge,
and I've got some pull reqs on rc2, so this
just makes sense.
Signed-off-by: Dave Airlie <airlied@redhat.com>
For a ppgtt that we are constructing, there is no struct_mutex
dependence so skip it. In the process, also ping the scheduler
frequently to try and avoid the NMI watchdog.
v2: gen6 requires struct_mutex to clean up (currently)
Suggested-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
References: https://bugs.freedesktop.org/show_bug.cgi?id=107094
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180703135331.12265-1-chris@chris-wilson.co.uk
live_gtt is a very slow test to run, simply because it tries to allocate
and use as much as the 48b address space as possibly can and in the
process will try to own all of the system memory. This leads to resource
exhaustion and CPU starvation; the latter impacts us when the NMI
watchdog declares a task hung due to a mutex contention with ourselves.
This we can prevent by releasing the struct_mutex and forcing our
i915/rcu workers to run, and in particular flushing the freed object
worker that is the cause for concern.
References: https://bugs.freedesktop.org/show_bug.cgi?id=107094
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180703101829.7360-1-chris@chris-wilson.co.uk
If the whole object is already pinned by HW for use as scanout, we will
fail to move it to the mappable region and so must resort to using a
partial VMA covering the whole object.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104513
Fixes: aa136d9d72 ("drm/i915: Convert partial ggtt vma to full ggtt if it spans the entire object")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Matthew Auld <matthew.william.auld@gmail.com>
Reviewed-by: Matthew Auld <matthew.william.auld@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180630090509.469-1-chris@chris-wilson.co.uk
(cherry picked from commit 7e7367d3bc)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
The PIPEDSL freezes on PSR entry and if PSR hasn't fully exited, then
the pipe_update_start call schedules itself out to check back later.
On ChromeOS-4.4 kernel, which is fairly up-to-date w.r.t drm/i915 but
lags w.r.t core kernel code, hot plugging an external display triggers
tons of "potential atomic update errors" in the dmesg, on *pipe A*. A
closer analysis reveals that we try to read the scanline 3 times and
eventually timeout, b/c PSR hasn't exited fully leading to a PIPEDSL
stuck @ 1599. This issue is not seen on upstream kernels, b/c for *some*
reason we loop inside intel_pipe_update start for ~2+ msec which in this
case is more than enough to exit PSR fully, hence an *unstuck* PIPEDSL
counter, hence no error. On the other hand, the ChromeOS kernel spends
~1.1 msec looping inside intel_pipe_update_start and hence errors out
b/c the source is still in PSR.
Regardless, we should wait for PSR exit (if PSR is disabled, we incur
a ~1-2 usec penalty) before reading the PIPEDSL, b/c if we haven't
fully exited PSR, then checking for vblank evasion isn't actually
applicable.
v4: Comment explaining psr_wait after enabling VBL interrupts (DK)
v5: CAN_PSR() to handle platforms that don't support PSR.
v6: Handle local_irq_disable on early return (Chris)
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Signed-off-by: Tarun Vyas <tarun.vyas@intel.com>
Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180627200250.1515-2-tarun.vyas@intel.com
This is a lockless version of the exisiting psr_wait_for_idle().
We want to wait for PSR to idle out inside intel_pipe_update_start.
At the time of a pipe update, we should never race with any psr
enable or disable code, which is a part of crtc enable/disable.
The follow up patch will use this lockless wait inside pipe_update_
start to wait for PSR to idle out before checking for vblank evasion.
We need to keep the wait in pipe_update_start to as less as it can be.
So,we can live and flourish w/o taking any psr locks at all.
Even if psr is never enabled, psr2_enabled will be false and this
function will wait for PSR1 to idle out, which should just return
immediately, so a very short (~1-2 usec) wait for cases where PSR
is disabled.
v2: Add comment to explain the 25msec timeout (DK)
v3: Rename psr_wait_for_idle to __psr_wait_for_idle_locked to avoid
naming conflicts and propagate err (if any) to the caller (Chris)
v5: Form a series with the next patch
v7: Better explain the need for lockless wait and increase the max
timeout to handle refresh rates < 60 Hz (Daniel Vetter)
v8: Rebase
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Signed-off-by: Tarun Vyas <tarun.vyas@intel.com>
Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180627200250.1515-1-tarun.vyas@intel.com
Prints live state of psr1.Extending the existing
PSR2 live state function to cover psr1.
Tested on KBL with psr2 and psr1 panel.
v2: rebase
v3: DK
Rename psr2_live_status to psr_source_status.
v4: DK
Move EDP_PSR_STATUS_STATE_SHIFT below EDP_PSR_STATUS_STATE_MASK.
Pass seq to psr_source_status, handle source status prints in
psr_source_status.
v5: Fixed CI warning messages
v6:
Remove extra space in the title before the colon.(DK)
Rebase. (Jani)
v7: Use tabs for indenting the values.(Jani)
v8: Addressed dk's review comments.
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Signed-off-by: Vathsala Nagaraju <vathsala.nagaraju@intel.com>
Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1530086910-15914-1-git-send-email-vathsala.nagaraju@intel.com
If the whole object is already pinned by HW for use as scanout, we will
fail to move it to the mappable region and so must resort to using a
partial VMA covering the whole object.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=104513
Fixes: aa136d9d72 ("drm/i915: Convert partial ggtt vma to full ggtt if it spans the entire object")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Matthew Auld <matthew.william.auld@gmail.com>
Reviewed-by: Matthew Auld <matthew.william.auld@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180630090509.469-1-chris@chris-wilson.co.uk
changed gvt display transcode DDI mode from DP_SST to
DVI to address below calltrace issue during guest booting
up which is caused by zero dotclock initial value with DP_SST
mode. transcode DVI mode emulation also align with native with DP
connection.
[drm:drm_calc_timestamping_constants]
ERROR crtc 41: Can't calculate constants, dotclock = 0!
WARNING: at drivers/gpu/drm/drm_vblank.c:620
drm_calc_vbltimestamp_from_scanoutpos
Call Trace:
? drm_calc_timestamping_constants+0x144/0x150 [drm]
drm_get_last_vbltimestamp+0x54/0x90 [drm]
drm_reset_vblank_timestamp+0x59/0xd0 [drm]
drm_crtc_vblank_on+0x7b/0xd0 [drm]
intel_modeset_setup_hw_state+0xb67/0xfd0 [i915]
? gen2_read32+0x110/0x110 [i915]
? drm_modeset_lock+0x30/0xa0 [drm]
intel_modeset_init+0x794/0x19d0 [i915]
? intel_setup_gmbus+0x232/0x2e0 [i915]
i915_driver_load+0xb4a/0xf40 [i915]
Signed-off-by: Xiaolin Zhang <xiaolin.zhang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
when guest writes ggtt entries, it could write 8 bytes a time if
gtt_entry_size is 8. But, qemu could split the 8 bytes into 2 consecutive
4-byte writes.
If each 4-byte partial write could trigger a host ggtt write, it is very
possible that a wrong combination is written to the host ggtt. E.g.
the higher 4 bytes is the old value, but the lower 4 bytes is the new
value, and this 8-byte combination is wrong but written to the ggtt, thus
causing bugs.
To handle this condition, we just record the first 4-byte write, then wait
until the second 4-byte write comes and write the combined 64-bit data to
host ggtt table.
To save memory space and to spot partial write as early as possible, we
don't keep this information for every ggtt index. Instread, we just record
the last ggtt write position, and assume the two 4-byte writes come in
consecutively for each vgpu.
This assumption is right based on the characteristic of ggtt entry which
stores memory address. When gtt_entry_size is 8, the guest memory physical
address should be 64 bits, so any sane guest driver should write 8-byte
long data at a time, so 2 consecutive 4-byte writes at the same ggtt index
should be trapped in gvt.
v2:
when incomplete ggtt entry write is located, e.g.
1. guest only writes 4 bytes at a ggtt offset and no long writes the
rest 4 bytes.
2. guest writes 4 bytes of a ggtt offset, then write at other ggtt
offsets, then return back to write the left 4 bytes of the first
ggtt offset.
add error handling logic to remap host entry to scratch page, and mark
guest virtual ggtt entry as not present. (zhenyu wang)
Signed-off-by: Zhao Yan <yan.y.zhao@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
-----BEGIN PGP SIGNATURE-----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=0uqV
-----END PGP SIGNATURE-----
Merge tag 'for-linus-20180629' of git://git.kernel.dk/linux-block
Pull block fixes from Jens Axboe:
"Small set of fixes for this series. Mostly just minor fixes, the only
oddball in here is the sg change.
The sg change came out of the stall fix for NVMe, where we added a
mempool and limited us to a single page allocation. CONFIG_SG_DEBUG
sort-of ruins that, since we'd need to account for that. That's
actually a generic problem, since lots of drivers need to allocate SG
lists. So this just removes support for CONFIG_SG_DEBUG, which I added
back in 2007 and to my knowledge it was never useful.
Anyway, outside of that, this pull contains:
- clone of request with special payload fix (Bart)
- drbd discard handling fix (Bart)
- SATA blk-mq stall fix (me)
- chunk size fix (Keith)
- double free nvme rdma fix (Sagi)"
* tag 'for-linus-20180629' of git://git.kernel.dk/linux-block:
sg: remove ->sg_magic member
drbd: Fix drbd_request_prepare() discard handling
blk-mq: don't queue more if we get a busy return
block: Fix cloning of requests with a special payload
nvme-rdma: fix possible double free of controller async event buffer
block: Fix transfer when chunk sectors exceeds max
While debugging we may want to examine params passed to GuC.
v2: drop #ifdef DEBUG_GUC - Michal
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Michel Thierry <michel.thierry@intel.com> #1
Cc: Michal Winiarski <michal.winiarski@intel.com>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180618111821.47088-1-michal.wajdeczko@intel.com
make_obj_busy() makes a dummy busy object, but didn't attach the fence
to the reservation object, so it would not have registered as busy. For
completeness, attach the dummy request as the exclusive fence and mark
the object as written (in i915_vma_move_to_active)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180629133717.11761-2-chris@chris-wilson.co.uk
We correctly attach the exclusive fetch for the scratch object when
emitting a request that writes into it, but for completeness we should
also declared the write to i915_vma_move_to_active()
Reported-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180629133717.11761-1-chris@chris-wilson.co.uk
This was introduced more than a decade ago when sg chaining was
added, but we never really caught anything with it. The scatterlist
entry size can be critical, since drivers allocate it, so remove
the magic member. Recently it's been triggering allocation stalls
and failures in NVMe.
Tested-by: Jordan Glover <Golden_Miller83@protonmail.ch>
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
The only time we should start FBC is when we have waited a vblank
after the atomic update. We've already forced a vblank wait by doing
wait_for_flip_done before intel_post_plane_update(), so we don't need
to wait a second time before enabling.
Removing the worker simplifies the code and removes possible race
conditions, like happening in 103167.
Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=103167
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180625163758.10871-2-maarten.lankhorst@linux.intel.com
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Back in commit 27af5eea54 ("drm/i915: Move execlists irq handler to a
bottom half"), we came to the conclusion that running our CSB processing
and ELSP submission from inside the irq handler was a bad idea. A really
bad idea as we could impose nearly 1s latency on other users of the
system, on average! Deferring our work to a tasklet allowed us to do the
processing with irqs enabled, reducing the impact to an average of about
50us.
We have since eradicated the use of forcewaked mmio from inside the CSB
processing and ELSP submission, bringing the impact down to around 5us
(on Kabylake); an order of magnitude better than our measurements 2
years ago on Broadwell and only about 2x worse on average than the
gem_syslatency on an unladen system.
In this iteration of the tasklet-vs-direct submission debate, we seek a
compromise where by we submit new requests immediately to the HW but
defer processing the CS interrupt onto a tasklet. We gain the advantage
of low-latency and ksoftirqd avoidance when waking up the HW, while
avoiding the system-wide starvation of our CS irq-storms.
Comparing the impact on the maximum latency observed (that is the time
stolen from an RT process) over a 120s interval, repeated several times
(using gem_syslatency, similar to RT's cyclictest) while the system is
fully laden with i915 nops, we see that direct submission an actually
improve the worse case.
Maximum latency in microseconds of a third party RT thread
(gem_syslatency -t 120 -f 2)
x Always using tasklets (a couple of >1000us outliers removed)
+ Only using tasklets from CS irq, direct submission of requests
+------------------------------------------------------------------------+
| + |
| + |
| + |
| + + |
| + + + |
| + + + + x x x |
| +++ + + + x x x x x x |
| +++ + ++ + + *x x x x x x |
| +++ + ++ + * *x x * x x x |
| + +++ + ++ * * +*xxx * x x xx |
| * +++ + ++++* *x+**xx+ * x x xxxx x |
| **x++++*++**+*x*x****x+ * +x xx xxxx x x |
|x* ******+***************++*+***xxxxxx* xx*x xxx + x+|
| |__________MA___________| |
| |______M__A________| |
+------------------------------------------------------------------------+
N Min Max Median Avg Stddev
x 118 91 186 124 125.28814 16.279137
+ 120 92 187 109 112.00833 13.458617
Difference at 95.0% confidence
-13.2798 +/- 3.79219
-10.5994% +/- 3.02677%
(Student's t, pooled s = 14.9237)
However the mean latency is adversely affected:
Mean latency in microseconds of a third party RT thread
(gem_syslatency -t 120 -f 1)
x Always using tasklets
+ Only using tasklets from CS irq, direct submission of requests
+------------------------------------------------------------------------+
| xxxxxx + ++ |
| xxxxxx + ++ |
| xxxxxx + +++ ++ |
| xxxxxxx +++++ ++ |
| xxxxxxx +++++ ++ |
| xxxxxxx +++++ +++ |
| xxxxxxx + ++++++++++ |
| xxxxxxxx ++ ++++++++++ |
| xxxxxxxx ++ ++++++++++ |
| xxxxxxxxxx +++++++++++++++ |
| xxxxxxxxxxx x +++++++++++++++ |
|x xxxxxxxxxxxxx x + + ++++++++++++++++++ +|
| |__A__| |
| |____A___| |
+------------------------------------------------------------------------+
N Min Max Median Avg Stddev
x 120 3.506 3.727 3.631 3.6321417 0.02773109
+ 120 3.834 4.149 4.039 4.0375167 0.041221676
Difference at 95.0% confidence
0.405375 +/- 0.00888913
11.1608% +/- 0.244735%
(Student's t, pooled s = 0.03513)
However, since the mean latency corresponds to the amount of irqsoff
processing we have to do for a CS interrupt, we only need to speed that
up to benefit not just system latency but our own throughput.
v2: Remember to defer submissions when under reset.
v4: Only use direct submission for new requests
v5: Be aware that with mixing direct tasklet evaluation and deferred
tasklets, we may end up idling before running the deferred tasklet.
v6: Remove the redudant likely() from tasklet_is_enabled(), restrict the
annotation to reset_in_progress().
v7: Take the full timeline.lock when enabling perf_pmu stats as the
tasklet is no longer a valid guard. A consequence is that the stats are
now only valid for engines also using the timeline.lock to process
state.
Testcase: igt/gem_exec_latency/*rthog*
References: 27af5eea54 ("drm/i915: Move execlists irq handler to a bottom half")
Suggested-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-9-chris@chris-wilson.co.uk
Now that we use the CSB stored in the CPU friendly HWSP, we do not need
to track interrupts for when the mmio CSB registers are valid and can
just check where we read up to last from the cached HWSP. This means we
can forgo the atomic bit tracking from interrupt, and in the next patch
it means we can check the CSB at any time.
v2: Change the splitting inside reset_prepare, we only want to lose
testing the interrupt in this patch, the next patch requires the change
in locking
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-8-chris@chris-wilson.co.uk
As we now never read back our current head position from the CSB
pointers register, and the HW itself doesn't use it to prevent
overwriting unread CSB entries, we do not need to keep updating the
register. As it turns out this register is not listed as being shadowed,
and so requires forcewake -- but we haven't been taking forcewake around
it so the writes has probably been regularly dropped. Fortuitously, we
only read the value after a reset where it did not matter, and zero was
the right answer (well, close enough).
Mika pointed out that this was how we used to do it (accidentally!)
before he fixed it in commit cc53699b25 ("drm/i915: Use masked write
for Context Status Buffer Pointer").
References: cc53699b25 ("drm/i915: Use masked write for Context Status Buffer Pointer")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-7-chris@chris-wilson.co.uk
On HW reset, the HW clears the write pointer (to 0). But since it also
writes its first CSB entry to slot 0, we need to reset the write pointer
back to the element before (so the first entry we read is 0).
This is required for the next patch, where we trust the CSB completely!
v2: Use _MASKED_FIELD
v3: Store the reset value, so that we differentiate between mmio/hwsp
transparently and without pretense.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-6-chris@chris-wilson.co.uk
Following the removal of the last workarounds, the only CSB mmio access
is for the old vGPU interface. The mmio registers presented by vGPU do
not require forcewake and can be treated as ordinary volatile memory,
i.e. they behave just like the HWSP access just at a different location.
We can reduce the CSB access to a set of read/write/buffer pointers and
treat the various paths identically and not worry about forcewake.
(Forcewake is nightmare for worstcase latency, and we want to process
this all with irqsoff -- no latency allowed!)
v2: Comments, comments, comments. Well, 2 bonus comments.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-5-chris@chris-wilson.co.uk
In the next patch, we will process the CSB events directly from the
submission path, rather than only after a CS interrupt. Hence, we will
no longer have the need for a loop until the has-interrupt bit is clear,
and in the meantime can remove that small optimisation.
v2: Tvrtko pointed out it was safer to unconditionally kick the tasklet
after each irq, when assuming that the tasklet is called for each irq.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-4-chris@chris-wilson.co.uk
In the following patch, we will process the CSB events under the
timeline.lock and not serialised by the tasklet. This also means that we
will need to protect access to common variables such as
execlists->csb_head with the timeline.lock during reset.
v2: Move sync_irq to avoid deadlocks between taking timeline.lock from
our interrupt handler.
v3: Kill off the synchronize_hardirq as it raises more questions than
answered; now we use the timeline.lock entirely for CSB serialisation
between the irq and elsewhere, we don't need to be so heavy handed with
flushing
v4: Treat request cancellation (wedging after failed reset) similarly
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-3-chris@chris-wilson.co.uk
In the next patch, we will begin processing the CSB from inside the
submission path (underneath an irqsoff section, and even from inside
interrupt handlers). This means that updating the execlists->port[] will
no longer be serialised by the tasklet but needs to be locked by the
engine->timeline.lock instead. Pull dequeue and submit under the same
lock for protection. (An alternate future plan is to keep the in/out
arrays separate for concurrent processing and reduced lock coverage.)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-2-chris@chris-wilson.co.uk
We do not need to do a posting read of our uncached mmio write to
re-enable the master interrupt lines after handling an interrupt, so
don't. This saves us a slow UC read before we can process the interrupt,
most noticeable in execlists where any stalls imposes extra latency on
GPU command execution.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Ville Syrjala <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-1-chris@chris-wilson.co.uk
We're fetching GuC/HuC firmwares directly from uc level during
init_early stage but this breaks guc/huc struct isolation and
also strict SW-only initialization rule for init_early. Move fw
fetching to init phase and do it separately per guc/huc struct.
v2: don't forget to move wopcm_init - Michele
v3: fetch in init_misc phase - Michal
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Michel Thierry <michel.thierry@intel.com> #2
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628141522.62788-2-michal.wajdeczko@intel.com
We will add more init steps to misc phase and there is no need
to expose them separately for use in uc_init_misc function.
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Michel Thierry <michel.thierry@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628141522.62788-1-michal.wajdeczko@intel.com
If we have more interrupts pending (because we know there are more
breadcrumb signals before the completion), then we do not need to
trigger an irq_seqno_barrier or even wakeup the task on this interrupt
as there will be another. To allow some margin of error (we are trying
to work around incoherent seqno after all), we wakeup the breadcrumb
before the target as well as on the target.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180627201304.15817-2-chris@chris-wilson.co.uk
By taking advantage of the RCU protection of the task struct, we can find
the appropriate signaler under the spinlock and then release the spinlock
before waking the task and signaling the fence.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180627201304.15817-1-chris@chris-wilson.co.uk
At the moment, gem_exec_gttfill fails with a sporadic EBUSY due to us
wanting to unbind a pinned batch. Let's dump who first bound that vma to
see if that helps us identify who still unexpectedly has it pinned.
v2: We cannot allocate inside the printer (as it may be on an fs-reclaim
path), so hope for the best and build the string on the stack
v3: stack depth of 16 routinely overflows a 512 character string, limit
it to 12 to avoid unsightly truncation.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628132206.8329-1-chris@chris-wilson.co.uk
This patch unifies the naming of DRM functions for reference counting
of struct drm_device. The resulting code is more aligned with the rest
of the Linux kernel interfaces.
Signed-off-by: Thomas Zimmermann <tdz@users.sourceforge.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20180618110154.30462-6-tdz@users.sourceforge.net
