Zhenyu Wang writes:
This is first pull request to merge GVT-g device model in i915
which contains core GVT-g device model work to virtualize GPU
resources. This tries to add feature of Intel GVT-g technology
for full GPU virtualization. This version will support KVM based
virtualization solution named as KVMGT.
More background is on official project home: https://01.org/igvt-g
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Having skl_wm_level contain all of the watermarks for each plane is
annoying since it prevents us from having any sort of object to
represent a single watermark level, something we take advantage of in
the next commit to cut down on all of the copy paste code in here.
Changes since v1:
- Style nitpicks
- Fix accidental usage of i vs. PLANE_CURSOR
- Split out skl_pipe_wm_active_state simplification into separate patch
Signed-off-by: Lyude <cpaul@redhat.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
First part of cleaning up all of the skl watermark code. This moves the
structures for storing the ddb allocations of each pipe into
intel_crtc_state, along with moving the structures for storing the
current ddb allocations active on hardware into intel_crtc.
Changes since v1:
- Don't replace alloc->start = alloc->end = 0;
Signed-off-by: Lyude <cpaul@redhat.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Since "Dynamic page table allocations" were introduced, our page tables
can grow (being dynamically allocated) with address space range usage.
Unfortunately, their lifetime is bound to vm. This is not a huge problem
when we're not using softpin - drm_mm is creating an upper bound on used
range by causing addresses for our VMAs to eventually be reused.
With softpin, long lived contexts can drain the system out of memory
even with a single "small" object. For example:
bo = bo_alloc(size);
while(true)
offset += size;
exec(bo, offset);
Will cause us to create new allocations until all memory in the system
is used for tracking GPU pages (even though almost all PTEs in this vm
are pointing to scratch).
Let's free unused page tables in clear_range to prevent this - if no
entries are used, we can safely free it and return this information to
the caller (so that higher-level entry is pointing to scratch).
v2: Document return value and free semantics (Joonas)
v3: No newlines in vars block (Joonas)
v4: Drop redundant local 'reduce' variable
v5: Handle CI fail with enable_ppgtt=2
Cc: Michel Thierry <michel.thierry@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/1476360162-24062-3-git-send-email-michal.winiarski@intel.com
Currently the display INIT power domain disabling/enabling happens in a
mismatched way in the suspend/resume_early hooks respectively. This can
leave display power wells incorrectly disabled in the resume hook if the
suspend sequence is aborted for some reason resulting in the
suspend/resume hooks getting called but the suspend_late/resume_early
hooks being skipped. In particular this change fixes "Unclaimed read
from register 0x1e1204" on BYT/BSW triggered from i915_drm_resume()->
intel_pps_unlock_regs_wa() when suspending with /sys/power/pm_test set
to devices.
Fixes: 85e9067933 ("drm/i915: disable power wells on suspend")
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: David Weinehall <david.weinehall@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1476358446-11621-1-git-send-email-imre.deak@intel.com
As different VM may configure different render MMIOs when executing
workload, to schedule workloads between different VM, the render MMIOs
have to be switched.
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
This patch introduces a vGPU schedule policy framework, with a timer based
schedule policy module for now
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
This patch introduces the vGPU workload scheduler routines.
GVT workload scheduler is responsible for picking and executing GVT workload
from current scheduled vGPU. Before the workload is submitted to host i915,
the guest execlist context will be shadowed in the host GVT shadow context.
the instructions in guest ring buffer will be copied into GVT shadow ring
buffer. Then GVT-g workload scheduler will scan the instructions in guest
ring buffer and submit it to host i915.
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
This patch introduces the vGPU workload submission logics.
Under virtualization environment, guest will submit workload through
virtual execlist submit port. The submitted workload load will be wrapped
into an gvt workload which will be picked by GVT workload scheduler and
executed on host i915 later.
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
This patch introduces the vGPU execlist virtualization.
Under virtulization environment, HW execlist interface are fully emulated
including virtual CSB emulation, virtual execlist emulation. The framework
will emulate the virtual CSB according to the guest workload running status
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
This patch introduces the GVT-g display virtualization.
It consists a collection of display MMIO handlers, like power well register
handler, pipe register handler, plane register handler, which will emulate
all display MMIOs behavior to support virtual mode setting sequence for
guest.
Signed-off-by: Bing Niu <bing.niu@intel.com>
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
This patch introduces the generic vGPU MMIO emulation intercept
framework. The MPT modules will request GVT-g core logic to
emulate MMIO read/write through IO emulation operations
callback when hypervisor trapped a guest GTTMMIO read/write.
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
This patch introduces vGPU PCI configuration space virtualization.
- Adjust the trapped GPFN(Guest Page Frame Number) window of virtual GEN
PCI BAR 0 when guest initializes PCI BAR 0 address.
- Emulate OpRegion when guest touches OpRegion.
- Pass-through a part of aperture to guest when guest initializes
aperture BAR.
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
The vGPU graphics memory emulation framework is responsible for graphics
memory table virtualization. Under virtualization environment, a VM will
populate the page table entry with guest page frame number(GPFN/GFN), while
HW needs a page table filled with MFN(Machine frame number). The
relationship between GFN and MFN(Machine frame number) is managed by
hypervisor, while GEN HW doesn't have such knowledge to translate a GFN.
To solve this gap, shadow GGTT/PPGTT page table is introdcued.
For GGTT, the GFN inside the guest GGTT page table entry will be translated
into MFN and written into physical GTT MMIO registers when guest write
virtual GTT MMIO registers.
For PPGTT, a shadow PPGTT page table will be created and write-protected
translated from guest PPGTT page table. And the shadow page table root
pointers will be written into the shadow context after a guest workload
is shadowed.
vGPU graphics memory emulation framework consists:
- Per-GEN HW platform page table entry bits extract/de-extract routines.
- GTT MMIO register emulation handlers, which will call hypercall to do
GFN->MFN translation when guest write GTT MMIO register
- PPGTT shadow page table routines, e.g. shadow create/destroy/out-of-sync
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
This patch introduces vGPU interrupt emulation framework.
The vGPU intrerrupt emulation framework is an event-based interrupt
emulation framework. It's responsible for emulating GEN hardware interrupts
during emulating other HW behaviour.
It consists several components:
- Descriptions of interrupt register bit
- Upper level <-> lower level interrupt mapping
- GEN HW IER/IMR/IIR register emulation routines
- Event-based interrupt propagation interface
When a GVT-g component wants to inject an interrupt to a VM during a
emulation, first it should specify the event needs to be emulated and the
framework will deal with the rest of emulation:
- Generating related virtual IIR bit according to virtual IER and IMRs,
- Generate related virtual upper level virtual IIR bit accodring to the
per-platform interrupt mapping
- Injecting a MSI to VM
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
A vGPU represents a virtual Intel GEN hardware, which consists following
virtual resources:
- Configuration space (virtualized)
- HW registers (virtualized)
- GGTT memory space (partitioned)
- GPU page table (shadowed)
- Fence registers (partitioned)
* virtualized: fully emulated by GVT-g.
* partitioned: Only a part of the HW resource is allowed to be accessed
by VM.
* shadowed: Resource needs to be translated and shadowed before getting
applied into HW.
This patch introduces vGPU life cycle management framework, which is
responsible for creating/destroying a vGPU and preparing/free resources
related to a vGPU.
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
Each vGPU expects a golden virtual HW state, which is just the state after
system is freshly powered on. GVT-g will try to load the golden virtual HW
state via kernel firmware interface.
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
This patch introduces a framework for tracking HW registers on different
GEN platforms.
Accesses to GEN HW registers from VMs will be trapped by hypervisor. It
will forward these emulation requests to GVT-g device model, which
requires this framework to search for related register descriptions.
Each MMIO entry in this framework describes a GEN HW registers, e.g.
offset, length, whether it contains RO bits, whether it can be accessed by
LRIs...and also emulation handlers for emulating register reading and
writing.
- Use i915 MMIO register definition & statement.(Joonas)
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
This patch introduces the GVT-g vGPU HW resource management. Under
GVT-g virtualizaion environment, each vGPU requires portions HW
resources, including aperture, hidden GM space, and fence registers.
When creating a vGPU, GVT-g will request these HW resources from host,
and return them to host after a vGPU is destroyed.
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
With the possibility of addition of many more number of rings in future,
the drm_i915_private structure could bloat as an array, of type
intel_engine_cs, is embedded inside it.
struct intel_engine_cs engine[I915_NUM_ENGINES];
Though this is still fine as generally there is only a single instance of
drm_i915_private structure used, but not all of the possible rings would be
enabled or active on most of the platforms. Some memory can be saved by
allocating intel_engine_cs structure only for the enabled/active engines.
Currently the engine/ring ID is kept static and dev_priv->engine[] is simply
indexed using the enums defined in intel_engine_id.
To save memory and continue using the static engine/ring IDs, 'engine' is
defined as an array of pointers.
struct intel_engine_cs *engine[I915_NUM_ENGINES];
dev_priv->engine[engine_ID] will be NULL for disabled engine instances.
There is a text size reduction of 928 bytes, from 1028200 to 1027272, for
i915.o file (but for i915.ko file text size remain same as 1193131 bytes).
v2:
- Remove the engine iterator field added in drm_i915_private structure,
instead pass a local iterator variable to the for_each_engine**
macros. (Chris)
- Do away with intel_engine_initialized() and instead directly use the
NULL pointer check on engine pointer. (Chris)
v3:
- Remove for_each_engine_id() macro, as the updated macro for_each_engine()
can be used in place of it. (Chris)
- Protect the access to Render engine Fault register with a NULL check, as
engine specific init is done later in Driver load sequence.
v4:
- Use !!dev_priv->engine[VCS] style for the engine check in getparam. (Chris)
- Kill the superfluous init_engine_lists().
v5:
- Cleanup the intel_engines_init() & intel_engines_setup(), with respect to
allocation of intel_engine_cs structure. (Chris)
v6:
- Rebase.
v7:
- Optimize the for_each_engine_masked() macro. (Chris)
- Change the type of 'iter' local variable to enum intel_engine_id. (Chris)
- Rebase.
v8: Rebase.
v9: Rebase.
v10:
- For index calculation use engine ID instead of pointer based arithmetic in
intel_engine_sync_index() as engine pointers are not contiguous now (Chris)
- For appropriateness, rename local enum variable 'iter' to 'id'. (Joonas)
- Use for_each_engine macro for cleanup in intel_engines_init() and remove
check for NULL engine pointer in cleanup() routines. (Joonas)
v11: Rebase.
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Akash Goel <akash.goel@intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1476378888-7372-1-git-send-email-akash.goel@intel.com
If the user requests a mappable binding to the global GTT, we will first
unbind an existing mapping if it doesn't match. We will unbind even if
there is no possibility that the object can fit in the mappable
aperture. This may lead to a ping-pong migration of the object, for
example igt/gem_exec_big.
v2: Comment upon the reasoning, or lack thereof!, behind the choice of
magic numbers.
Testcase: igt/gem_exec_big
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: http://patchwork.freedesktop.org/patch/msgid/20161013085504.30705-1-chris@chris-wilson.co.uk
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com
