While the end goal is to make this information available to userspace
through a new ioctl, there is no reason we can't display it in a human
readable fashion through debugfs.
slice0: 3 subslice(s) (0x7):
subslice0: 8 EUs (0xff)
subslice1: 8 EUs (0xff)
subslice2: 8 EUs (0xff)
subslice3: 0 EUs (0x0)
slice1: 3 subslice(s) (0x7):
subslice0: 8 EUs (0xff)
subslice1: 8 EUs (0xff)
subslice2: 8 EUs (0xff)
subslice3: 0 EUs (0x0)
slice2: 3 subslice(s) (0x7):
subslice0: 8 EUs (0xff)
subslice1: 8 EUs (0xff)
subslice2: 8 EUs (0xff)
subslice3: 0 EUs (0x0)
v2: Reformat debugfs printing (Tvrtko)
Use the new EU mask helper (Tvrtko)
v3: Move printing code to intel_device_info.c to be shared with error
state (Michal)
v4: Bump u8 to u16 when using sseu_get_eus() (Lionel)
Suggested-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Acked-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180306122857.27317-4-lionel.g.landwerlin@intel.com
Up to now, subslice mask was assumed to be uniform across slices. But
starting with Cannonlake, slices can be asymmetric (for example slice0
has different number of subslices as slice1+). This change stores all
subslices masks for all slices rather than having a single mask that
applies to all slices.
v2: Rework how we store total numbers in sseu_dev_info (Tvrtko)
Fix CHV eu masks, was reading disabled as enabled (Tvrtko)
Readability changes (Tvrtko)
Add EU index helper (Tvrtko)
v3: Turn ALIGN(v, 8) / 8 into DIV_ROUND_UP(v, BITS_PER_BYTE) (Tvrtko)
Reuse sseu_eu_idx() for setting eu_mask on CHV (Tvrtko)
Reformat debug prints for subslices (Tvrtko)
v4: Change eu_mask helper into sseu_set_eus() (Tvrtko)
v5: With Haswell reporting masks & counts, bump sseu_*_eus() functions
to use u16 (Lionel)
v6: Fix sseu_get_eus() for > 8 EUs per subslice (Lionel)
v7: Change debugfs enabels for number of subslices per slice, will
need a small igt/pm_sseu change (Lionel)
Drop subslice_total field from sseu_dev_info, rely on
sseu_subslice_total() to recompute the value instead (Lionel)
v8: Remove unused function compute_subslice_total() (Lionel)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Acked-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180306122857.27317-2-lionel.g.landwerlin@intel.com
During igt, we frequently call into the driver to reset both HW and
driver state (idling the device, waiting for it to become idle and
freeing off old objects) to ensure that we start each test/subtest/pass
from known state. This process incurs an RCU barrier or two to ensure
that any such pending frees are indeed flushed before we return.
However, unconditionally waiting on the RCU barrier adds needless delay
to many callers, which adds up to several seconds when repeated thousands
of times. We can skip the rcu_barrier() if by tracking how many outstanding
frees we have, we know there are none.
The same path is used along suspend, where we may be able to save the
unconditional RCU barrier.
To put it into perspective with a completely meaningless
microbenchmark, igt/gem_sync/idle is improved from 50ms to 30us on bdw.
v2: Remove the extra synchronize_rcu() inside i915_drop_caches_set()
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180219220631.25001-1-chris@chris-wilson.co.uk
i915 is the only driver using those fields in the drm_gem_object
structure, so they only waste memory for all other drivers.
Move the fields into drm_i915_gem_object instead and patch the i915 code
with the following sed commands:
sed -i "s/obj->base.read_domains/obj->read_domains/g" drivers/gpu/drm/i915/*.c drivers/gpu/drm/i915/*/*.c
sed -i "s/obj->base.write_domain/obj->write_domain/g" drivers/gpu/drm/i915/*.c drivers/gpu/drm/i915/*/*.c
Change is only compile tested.
v2: move fields around as suggested by Chris.
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180216124338.9087-1-christian.koenig@amd.com
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
On BXT/GLK GEN6_PCODE_READ_RC6VIDS fails with
MAILBOX_P24C_CC_ILLEGAL_CMD, so don't try to do the query on these
platforms. Do it only on SNB, IVB and HSW, where we use this command
anyway for RC6 enabling.
Based on my tests the command also succeeds on all LLC platforms, but
it's not clear if it's really supported on those (it returns 0 aka 245mv
for all RC6 states everywhere except on SNB). BSpec lists the command as
supported on SKL+ (see P24C_PCODE_MAILBOX_INTERFACE) but that's clearly
incorrect, since on SKL/KBL the same command ID is used for
SKL_PCODE_LOAD_HDCP_KEYS. Since the command fails on BXT/GLK, the BSpec
command list is also incorrect for those platforms (see
P_CR_P24C_PCODE_MAILBOX_INTERFACE_0_2_0_GTTMMADR).
I filed a request to update that info in Bspec, but for now let's
assume a minimal set of platforms where the command is supported.
References: https://bugs.freedesktop.org/show_bug.cgi?id=103337
Signed-off-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180208174102.10240-1-imre.deak@intel.com
By counting the number of times we have woken up, we have a very simple
means of defining an epoch, which will come in handy if we want to
perform deferred tasks at the end of an epoch (i.e. while we are going
to sleep) without imposing on the next activity cycle.
v2: No reason to specify precise number of bits here.
v3: Take Tvrtko's advice and reserve 0 as an invalid epoch.
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/20180124113608.14909-1-chris@chris-wilson.co.uk
In quite a few places, we have a list iteration over the vma on an
object that only want to inspect GGTT vma. By construction, these are
placed at the start of the list, so we have copied that knowledge into
many callsites. Pull that knowledge back to i915_vma.h and provide a
for_each_ggtt_vma() to tidy up the code.
v2: Add a backreference from vma_create() to remind ourselves why we put
ggtt vma at the head of the obj->vma_list (and ppgtt vma at the tail).
v3: Fixup s/vma/V/
Suggested-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20171207211407.31549-1-chris@chris-wilson.co.uk
Since removing the module parameter to force selection of ringbuffer
emission for gen8, the code is defunct. Remove it.
To put the difference into perspective, a couple of microbenchmarks
(bdw i7-5557u, 20170324):
ring execlists
exec continuous nops on all rings: 1.491us 2.223us
exec sequential nops on each ring: 12.508us 53.682us
single nop + sync: 9.272us 30.291us
vblank_mode=0 glxgears: ~11000fps ~9000fps
Since the earlier submission, gen8 ringbuffer submission has fallen
further and further behind in features. So while ringbuffer may hold the
throughput crown, in terms of interactive latency, execlists is much
better. Alas, we have no convenient metrics for such, other than
demonstrating things we can do with execlists but can not using
legacy ringbuffer submission.
We have made a few improvements to lowlevel execlists throughput,
and ringbuffer currently panics on boot! (bdw i7-5557u, 20171026):
ring execlists
exec continuous nops on all rings: n/a 1.921us
exec sequential nops on each ring: n/a 44.621us
single nop + sync: n/a 21.953us
vblank_mode=0 glxgears: n/a ~18500fps
References: https://bugs.freedesktop.org/show_bug.cgi?id=87725
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Once-upon-a-time-Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171120205504.21892-2-chris@chris-wilson.co.uk
We use to have this fixed per generation, but starting with CNL userspace
cannot tell just off the PCI ID. Let's make this information available. This
is particularly useful for performance monitoring where much of the
normalization work is done using those timestamps (this include pipeline
statistics in both GL & Vulkan as well as OA reports).
v2: Use variables for 24MHz/19.2MHz values (Ewelina)
Renamed function & coding style (Sagar)
v3: Fix frequency read on Broadwell (Sagar)
Fix missing divide by 4 on <= gen4 (Sagar)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Tested-by: Rafael Antognolli <rafael.antognolli@intel.com>
Reviewed-by: Sagar Arun Kamble <sagar.a.kamble@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171110190845.32574-7-lionel.g.landwerlin@intel.com
Take a copy of the HW state after a reset upon module loading by
executing a context switch from a blank context to the kernel context,
thus saving the default hw state over the blank context image.
We can then use the default hw state to initialise any future context,
ensuring that each starts with the default view of hw state.
v2: Unmap our default state from the GTT after stealing it from the
context. This should stop us from accidentally overwriting it via the
GTT (and frees up some precious GTT space).
Testcase: igt/gem_ctx_isolation
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171110142634.10551-7-chris@chris-wilson.co.uk
Currently the DDI encoder->type will change at runtime depending on
what kind of hotplugs we've processed. That's quite bad since we can't
really trust that that current value of encoder->type actually matches
the type of signal we're trying to drive through it.
Let's eliminate that problem by declaring that non-eDP DDI port will
always have the encoder type as INTEL_OUTPUT_DDI. This means the code
can no longer try to distinguish DP vs. HDMI based on encoder->type.
We'll leave eDP as INTEL_OUTPUT_EDP, since it'll never change and
there's a bunch of code that relies on that value to identify eDP
encoders.
We'll introduce a new encoder .compute_output_type() hook. This allows
us to compute the full output_types before any encoder .compute_config()
hooks get called, thus those hooks can rely on output_types being
correct, which is useful for cloning on oldr platforms. For now we'll
just look at the connector type and pick the correct mode based on that.
In the future the new hook could be used to implement dynamic switching
between LS and PCON modes for LSPCON.
v2: Fix BXT/GLK PPS explosion with DSI/MST encoders
v3: Avoid the PPS warn on pure HDMI/DVI DDI encoders by checking dp.output_reg
v4: Rebase
v5: Populate output_types in .get_config() rather than in the caller
v5: Split out populating output_types in .get_config() (Maarten)
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171027193128.14483-3-ville.syrjala@linux.intel.com
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
CNL adds an extra register for slice/subslice information.
Although no SKU is planed with an extra slice let's already
handle this extra piece of information so we don't have the
risk in future of getting a part that might have chosen this
part of the die instead of other slices or anything like that.
Also if subslice is disabled the information of eu ack for that
is garbage, so let's skip checks for eu if subslice is disabled
as we skip the subslice if slice is disabled.
The rest is pretty much like gen9.
v2: Remove IS_CANNONLAKE from gen9 status function.
v3: Consider s_max = 6 and ss_max=4 to run over all possible
slices and subslices possible by spec. Although no real
hardware will have that many slices/subslices.
To match with sseu info init.
v4: Fix offset calculation for slices 4 and 5.
Removed Oscar's rv-b since this change also needs review.
v5: Let's consider only valid bits for SLICE*_PGCTL_ACK.
This looks like wrong in Spec, but seems to be enough
for now. Whenever Spec gets updated and fixed we come
back and properly update the masks. Also add a FIXME,
so we can revisit this later when we find some strange
info on debugfs or when we noitce spec got updated.
Cc: Oscar Mateo <oscar.mateo@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20171026001546.28203-1-rodrigo.vivi@intel.com
