Alexandre Oliva has recently removed these files from Linux Libre
with concerns that the sources weren't available.
The sources are available on IGT repository, and only open source
tools are used to generate the {ivb,hsw}_clear_kernel.c files.
However, the remaining concern from Alexandre Oliva was around
GPL license and the source not been present when distributing
the code.
So, it looks like 2 alternatives are possible, the use of
linux-firmware.git repository to store the blob or making sure
that the source is also present in our tree. Since the goal
is to limit the i915 firmware to only the micro-controller blobs
let's make sure that we do include the asm sources here in our tree.
Btw, I tried to have some diligence here and make sure that the
asms that these commits are adding are truly the source for
the mentioned files:
igt$ ./scripts/generate_clear_kernel.sh -g ivb \
-m ~/mesa/build/src/intel/tools/i965_asm
Output file not specified - using default file "ivb-cb_assembled"
Generating gen7 CB Kernel assembled file "ivb_clear_kernel.c"
for i915 driver...
igt$ diff ~/i915/drm-tip/drivers/gpu/drm/i915/gt/ivb_clear_kernel.c \
ivb_clear_kernel.c
< * Generated by: IGT Gpu Tools on Fri 21 Feb 2020 05:29:32 AM UTC
> * Generated by: IGT Gpu Tools on Mon 08 Jun 2020 10:00:54 AM PDT
61c61
< };
> };
\ No newline at end of file
igt$ ./scripts/generate_clear_kernel.sh -g hsw \
-m ~/mesa/build/src/intel/tools/i965_asm
Output file not specified - using default file "hsw-cb_assembled"
Generating gen7.5 CB Kernel assembled file "hsw_clear_kernel.c"
for i915 driver...
igt$ diff ~/i915/drm-tip/drivers/gpu/drm/i915/gt/hsw_clear_kernel.c \
hsw_clear_kernel.c
5c5
< * Generated by: IGT Gpu Tools on Fri 21 Feb 2020 05:30:13 AM UTC
> * Generated by: IGT Gpu Tools on Mon 08 Jun 2020 10:01:42 AM PDT
61c61
< };
> };
\ No newline at end of file
Used IGT and Mesa master repositories from Fri Jun 5 2020)
IGT: 53e8c878a6fb ("tests/kms_chamelium: Force reprobe after replugging
the connector")
Mesa: 5d13c7477eb1 ("radv: set keep_statistic_info with
RADV_DEBUG=shaderstats")
Mesa built with: meson build -D platforms=drm,x11 -D dri-drivers=i965 \
-D gallium-drivers=iris -D prefix=/usr \
-D libdir=/usr/lib64/ -Dtools=intel \
-Dkulkan-drivers=intel && ninja -C build
v2: Header clean-up and include build instructions in a readme (Chris)
Modified commit message to respect check-patch
Reference: http://www.fsfla.org/pipermail/linux-libre/2020-June/003374.html
Reference: http://www.fsfla.org/pipermail/linux-libre/2020-June/003375.html
Fixes: 47f8253d2b ("drm/i915/gen7: Clear all EU/L3 residual contexts")
Cc: <stable@vger.kernel.org> # v5.7+
Cc: Alexandre Oliva <lxoliva@fsfla.org>
Cc: Prathap Kumar Valsan <prathap.kumar.valsan@intel.com>
Cc: Akeem G Abodunrin <akeem.g.abodunrin@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Jon Bloomfield <jon.bloomfield@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200610201807.191440-1-rodrigo.vivi@intel.com
(cherry picked from commit 5a7eeb8ba1)
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
In commit 5ba32c7be8 ("drm/i915/execlists: Always force a context
reload when rewinding RING_TAIL"), we placed the check for rewinding a
context on actually submitting the next request in that context. This
was so that we only had to check once, and could do so with precision
avoiding as many forced restores as possible. For example, to ensure
that we can resubmit the same request a couple of times, we include a
small wa_tail such that on the next submission, the ring->tail will
appear to move forwards when resubmitting the same request. This is very
common as it will happen for every lite-restore to fill the second port
after a context switch.
However, intel_ring_direction() is limited in precision to movements of
upto half the ring size. The consequence being that if we tried to
unwind many requests, we could exceed half the ring and flip the sense
of the direction, so missing a force restore. As no request can be
greater than half the ring (i.e. 2048 bytes in the smallest case), we
can check for rollback incrementally. As we check against the tail that
would be submitted, we do not lose any sensitivity and allow lite
restores for the simple case. We still need to double check upon
submitting the context, to allow for multiple preemptions and
resubmissions.
Fixes: 5ba32c7be8 ("drm/i915/execlists: Always force a context reload when rewinding RING_TAIL")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: <stable@vger.kernel.org> # v5.4+
Reviewed-by: Bruce Chang <yu.bruce.chang@intel.com>
Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200609151723.12971-1-chris@chris-wilson.co.uk
(cherry picked from commit e36ba817fa)
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
We have a I915_REQUEST_NOPREEMPT flag that we set when we must prevent
the HW from preempting during the course of this request. We need to
honour this flag and protect the HW even if we have a heartbeat request,
or other maximum priority barrier, pending. As such, restrict the
timeslicing check to avoid preempting into the topmost priority band,
leaving the unpreemptable requests in blissful peace running
uninterrupted on the HW.
v2: Set the I915_PRIORITY_BARRIER to be less than
I915_PRIORITY_UNPREEMPTABLE so that we never submit a request
(heartbeat or barrier) that can legitimately preempt the current
non-premptable request.
Fixes: 2a98f4e65b ("drm/i915: add infrastructure to hold off preemption on a request")
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/20200527162418.24755-1-chris@chris-wilson.co.uk
(cherry picked from commit b72f02d78e)
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
gcc-9 gets confused by the code flow in check_dirty_whitelist:
drivers/gpu/drm/i915/gt/selftest_workarounds.c: In function 'check_dirty_whitelist':
drivers/gpu/drm/i915/gt/selftest_workarounds.c:492:17: error: 'rsvd' may be used uninitialized in this function [-Werror=maybe-uninitialized]
I could not figure out a good way to do this in a way that gcc
understands better, so initialize the variable to zero, as last
resort.
Fixes: aee20aaed8 ("drm/i915: Implement read-only support in whitelist selftest")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
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/20200527140526.1458215-2-arnd@arndb.de
(cherry picked from commit cc649a9eaf)
Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
The second try at staging the transfer of the breadcrumb. In part one,
we realised we could not simply move to the second engine as we were
only holding the breadcrumb lock on the first. So in commit 6c81e21a47
("drm/i915/gt: Stage the transfer of the virtual breadcrumb"), we
removed it from the first engine and marked up this request to reattach
the signaling on the new engine. However, this failed to take into
account that we only attach the breadcrumb if the new request is added
at the start of the queue, which if we are transferring, it is because
we know there to be a request to be signaled (and hence we would not be
attached).
In this attempt, we try to transfer the completed requests to the
irq_worker on its rq->engine->breadcrumbs. This preserves the coupling
between the rq and its breadcrumbs, so that
i915_request_cancel_breadcrumb() does not attempt to manipulate the list
under the wrong lock.
v2: Code sharing is fun.
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1862
Fixes: 6c81e21a47 ("drm/i915/gt: Stage the transfer of the virtual breadcrumb")
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/20200513074809.18194-1-chris@chris-wilson.co.uk
By providing the default values configured into the kernel via sysfs, it
is much more convenient for userspace to restore those sane defaults, or
at least know what are considered good baseline. This is useful, for
example, to cleanup after any failed userspace prior to commencing new
jobs.
/sys/class/drm/card0/engine/rcs0/
├── capabilities
├── class
├── .defaults
│ ├── heartbeat_interval_ms
│ ├── max_busywait_duration_ns
│ ├── preempt_timeout_ms
│ ├── stop_timeout_ms
│ └── timeslice_duration_ms
├── heartbeat_interval_ms
├── instance
├── known_capabilities
├── max_busywait_duration_ns
├── mmio_base
├── name
├── preempt_timeout_ms
├── stop_timeout_ms
└── timeslice_duration_ms
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Maciej Patelczyk <maciej.patelczyk@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200514062905.28668-1-chris@chris-wilson.co.uk
Now that we have fast timeslicing on semaphores, we no longer need to
prioritise none-semaphore work as we will yield any work blocked on a
semaphore to the next in the queue. Previously with no timeslicing,
blocking on the semaphore caused extremely bad scheduling with multiple
clients utilising multiple rings. Now, there is no impact and we can
remove the complication.
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/20200513173504.28322-1-chris@chris-wilson.co.uk
The initial-breadcrumb is used to mark the end of the awaiting and the
beginning of the user payload. We verify that we do not start the user
payload before all signaler are completed, checking our semaphore setup
by looking for the initial breadcrumb being written too early. We also
want to ensure that we do not add semaphore waits after we have already
closed the semaphore section, an issue for later deferred waits.
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/20200513165937.9508-2-chris@chris-wilson.co.uk
This reverts commit 0b718ba1e8.
There are still some residual issues with asynchronous binding and
execution, but since commit 92581f9fb9 ("drm/i915: Immediately execute
the fenced work") we prefer not to use asynchronous binds, and the
remaining issues do not seem restricted to Cherryview [at least the ones
seen over a few dozen CI runs, less frequent issues are sure to be
discovered!]
These issues seem to be mitigated, if not eliminated entirely, by the
previous commit 84eac0c659 ("drm/i915/gt: Force pte cacheline to main
memory").
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200510102431.21959-3-chris@chris-wilson.co.uk
The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:
struct foo {
int stuff;
struct boo array[];
};
By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.
Also, notice that, dynamic memory allocations won't be affected by
this change:
"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]
sizeof(flexible-array-member) triggers a warning because flexible array
members have incomplete type[1]. There are some instances of code in
which the sizeof operator is being incorrectly/erroneously applied to
zero-length arrays and the result is zero. Such instances may be hiding
some bugs. So, this work (flexible-array member conversions) will also
help to get completely rid of those sorts of issues.
This issue was found with the help of Coccinelle.
[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit 7649773293 ("cxgb3/l2t: Fix undefined behaviour")
Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
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/20200507185408.GA14561@embeddedor
Upon waiting a request (when asked), we gave that request a small
priority boost, not enough for it to cause preemption, but enough for it
to be scheduled next before all equals. We also used that bit to give
new clients a small priority boost, similar to FQ_CODEL, such that we
favoured short interactive tasks ahead of long running streams.
However, this is causing lots of complications with timeslicing where we
both want to honour the boost and yet ignore it. Those complications
cause unexpected user behaviour (tasks not being timesliced and run
concurrently as epxected), and the easiest way to resolve that is to
remove the boost. Hopefully, we can find a compromise again if we need
to, but in theory timeslicing itself and future more advanced schedulers
should give us the interactivity boost we seek.
Testcase: igt/gem_exec_schedule/lateslice
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/20200507152338.7452-3-chris@chris-wilson.co.uk
We recorded the dependencies for WAIT_FOR_SUBMIT in order that we could
correctly perform priority inheritance from the parallel branches to the
common trunk. However, for the purpose of timeslicing and reset
handling, the dependency is weak -- as we the pair of requests are
allowed to run in parallel and not in strict succession.
The real significance though is that this allows us to rearrange
groups of WAIT_FOR_SUBMIT linked requests along the single engine, and
so can resolve user level inter-batch scheduling dependencies from user
semaphores.
Fixes: c81471f5e9 ("drm/i915: Copy across scheduler behaviour flags across submit fences")
Testcase: igt/gem_exec_fence/submit
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: <stable@vger.kernel.org> # v5.6+
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200507155109.8892-1-chris@chris-wilson.co.uk
The older arches did not convert MI_STORE_DATA_IMM to using the GTT, but
left them writing to a physical address. The notes suggest that the
primary reason would be so that the writes were cache coherent, as the
CPU cache uses physical tagging. As such we did not implement the
legacy variant of MI_STORE_DATA_IMM and so left all the relocations
synchronous -- but with a small function to convert from the vma address
into the physical address, we can implement asynchronous relocs on these
older arches, fixing up a few tests that require them.
In order to be able to test the legacy paths, refactor the gpu
relocations so that we can hook them up to a selftest.
v2: Use an array of offsets not enum labels for the selftest
v3: Refactor the common igt_hexdump()
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/757
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/20200504140629.28240-1-chris@chris-wilson.co.uk
Since the introduction of 'soft-rc6', we aim to park the device quickly
and that results in frequent idling of the whole device. Currently upon
idling we free the batch buffer pool, and so this renders the cache
ineffective for many workloads. If we want to have an effective cache of
recently allocated buffers available for reuse, we need to decouple that
cache from the engine powermanagement and make it timer based. As there
is no reason then to keep it within the engine (where it once made
retirement order easier to track), we can move it up the hierarchy to the
owner of the memory allocations.
v2: Hook up to debugfs/drop_caches to clear the cache on demand.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200430111819.10262-2-chris@chris-wilson.co.uk
We treat parking as a manual RPS timeout event, and downclock the GPU
for the next unpark and batch execution. However, having restored the
aggressive downclocking and observed that we have very light workloads
whose only interaction is through the manual parking events, carry over
the aggressive downclocking to the fake RPS events.
References: 21abf0bf16 ("drm/i915/gt: Treat idling as a RPS downclock event")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Andi Shyti <andi.shyti@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200429205446.3259-5-chris@chris-wilson.co.uk
As with the realisation for soft-rc6, we respond to idling the engines
within microseconds, far faster than the response times for HW RC6 and
RPS. Furthermore, our fast parking upon idle, prevents HW RPS from
running for many desktop workloads, as the RPS evaluation intervals are
on the order of tens of milliseconds, but the typical workload is just a
couple of milliseconds, but yet we still need to determine the best
frequency for user latency versus power.
Recognising that the HW evaluation intervals are a poor fit, and that
they were deprecated [in bspec at least] from gen10, start to wean
ourselves off them and replace the EI with a timer and our accurate
busy-stats. The principle benefit of manually evaluating RPS intervals
is that we can be more responsive for better performance and powersaving
for both spiky workloads and steady-state.
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1698
Fixes: 98479ada42 ("drm/i915/gt: Treat idling as a RPS downclock event")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Andi Shyti <andi.shyti@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200429205446.3259-4-chris@chris-wilson.co.uk
