Rather than doing this in the IP code for the SDMA paging
engine, move it up to the core device level init level.
This should fix the scheduler init ordering.
v2: drop extra parens
v3: drop SDMA helpers
v4: Added a Fixes tag because amdgpu dereferences an uninitialized
scheduler without this patch, and this patch fixes this. (Luben)
Tested-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20231025171928.3318505-1-alexander.deucher@amd.com
Acked-by: Christian König <christian.koenig@amd.com>
Fixes: 56e449603f ("drm/sched: Convert the GPU scheduler to variable number of run-queues")
Signed-off-by: Luben Tuikov <ltuikov89@gmail.com>
If DRM_IOCTL_SYNCOBJ_TIMELINE_WAIT is invoked with the
DRM_SYNCOBJ_WAIT_FLAGS_WAIT_AVAILABLE flag set but no fence has yet been
submitted for the given timeline point the call will fail immediately
with EINVAL. This does not match the intended behavior where the call
should wait until the fence has been submitted (or the timeout expires).
The following small example program illustrates the issue. It should
wait for 5 seconds and then print ETIME, but instead it terminates right
away after printing EINVAL.
#include <stdio.h>
#include <fcntl.h>
#include <time.h>
#include <errno.h>
#include <xf86drm.h>
int main(void)
{
int fd = open("/dev/dri/card0", O_RDWR);
uint32_t syncobj;
drmSyncobjCreate(fd, 0, &syncobj);
struct timespec ts;
clock_gettime(CLOCK_MONOTONIC, &ts);
uint64_t point = 1;
if (drmSyncobjTimelineWait(fd, &syncobj, &point, 1,
ts.tv_sec * 1000000000 + ts.tv_nsec + 5000000000, // 5s
DRM_SYNCOBJ_WAIT_FLAGS_WAIT_AVAILABLE, NULL)) {
printf("drmSyncobjTimelineWait failed %d\n", errno);
}
}
Fixes: 01d6c35783 ("drm/syncobj: add support for timeline point wait v8")
Signed-off-by: Erik Kurzinger <ekurzinger@nvidia.com>
Reviewed by: Simon Ser <contact@emersion.fd>
Signed-off-by: Simon Ser <contact@emersion.fr>
Link: https://patchwork.freedesktop.org/patch/msgid/1fac96f1-2f3f-f9f9-4eb0-340f27a8f6c0@nvidia.com
Every know and then we receive the following error when running
for example IGT test kms_flip.
[drm] *ERROR* PHY G Read 0d80 failed after 3 retries.
[drm] *ERROR* PHY G Write 0d81 failed after 3 retries.
Since the error is sporadic in nature, the patch proposes
to reset the message bus after every successful or unsuccessful
read or write operation.
v2: Add FIXME's to indicate the experimental nature of
this workaround (Rodrigo)
v3: Dropping the additional delay as moving reset to *_read_once()
and *_write_once() functions seem unnecessary delay
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20231016125544.719963-1-mika.kahola@intel.com
With the new sharding, the default job timeout is not enough for i915
and their jobs are failing before completing.
See below the current execution time:
🞋 job i915:tgl 8/8 has new status: success (37m3s)
🞋 job i915:tgl 7/8 has new status: success (19m43s)
🞋 job i915:tgl 6/8 has new status: success (21m47s)
🞋 job i915:tgl 5/8 has new status: success (18m16s)
🞋 job i915:tgl 4/8 has new status: success (21m43s)
🞋 job i915:tgl 3/8 has new status: success (17m59s)
🞋 job i915:tgl 2/8 has new status: success (22m15s)
🞋 job i915:tgl 1/8 has new status: success (18m52s)
🞋 job i915:cml 2/2 has new status: success (1h19m58s)
🞋 job i915:cml 1/2 has new status: success (55m45s)
🞋 job i915:whl 2/2 has new status: success (1h8m56s)
🞋 job i915:whl 1/2 has new status: success (54m3s)
🞋 job i915:kbl 3/3 has new status: success (37m43s)
🞋 job i915:kbl 2/3 has new status: success (36m37s)
🞋 job i915:kbl 1/3 has new status: success (34m52s)
🞋 job i915:amly 2/2 has new status: success (1h7m60s)
🞋 job i915:amly 1/2 has new status: success (59m18s)
🞋 job i915:glk 2/2 has new status: success (58m26s)
🞋 job i915:glk 1/2 has new status: success (50m23s)
🞋 job i915:apl 3/3 has new status: success (1h6m39s)
🞋 job i915:apl 2/3 has new status: success (1h4m45s)
🞋 job i915:apl 1/3 has new status: success (1h7m38s)
(generated with ci_run_n_monitor.py script)
The longest job is 1h19m58s, so adjust the timeout.
Signed-off-by: Helen Koike <helen.koike@collabora.com>
Link: https://lore.kernel.org/r/20231024004525.169002-8-helen.koike@collabora.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>
IGT has recently merged a patch that makes code_getversion test to fails
if the driver isn't loaded or if it isn't the expected one defined in
variable IGT_FORCE_DRIVER.
Without this test, jobs were passing when the driver didn't load or
probe for some reason, giving the illusion that everything was ok.
Uprev IGT to include this modification and include core_getversion test
in all the shards.
Signed-off-by: Helen Koike <helen.koike@collabora.com>
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Link: https://lore.kernel.org/r/20231024004525.169002-5-helen.koike@collabora.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>
When building containers, some rust packages were installed without
locking the dependencies version, which got updated and started giving
errors like:
error: failed to compile `bindgen-cli v0.62.0`, intermediate artifacts can be found at `/tmp/cargo-installkNKRwf`
Caused by:
package `rustix v0.38.13` cannot be built because it requires rustc 1.63 or newer, while the currently active rustc version is 1.60.0
A patch to Mesa was added fixing this error, so update it.
Also, commit in linux kernel 6.6 rc3 broke booting in crosvm.
Mesa has upreved crosvm to fix this issue.
Signed-off-by: Helen Koike <helen.koike@collabora.com>
[crosvm mesa update]
Co-Developed-by: Vignesh Raman <vignesh.raman@collabora.com>
Signed-off-by: Vignesh Raman <vignesh.raman@collabora.com>
[v1 container build uprev]
Tested-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Acked-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Reviewed-by: David Heidelberg <david.heidelberg@collabora.com>
Link: https://lore.kernel.org/r/20231024004525.169002-2-helen.koike@collabora.com
Signed-off-by: Maxime Ripard <mripard@kernel.org>
In case of the merge requests it might be useful to push repo-specific
fixes which have not yet propagated to the -external-fixes branch in the
main UPSTREAM_REPO. For example, in case of drm/msm development, we are
staging fixes locally for testing, before pushing them to the drm/drm
repo. Thus, if the CI run was triggered by merge request, also pick up
the -external fixes basing on the the CI_MERGE target repo / and branch.
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Acked-by: Helen Koike <helen.koike@collabora.com>
Link: https://lore.kernel.org/r/20231008132320.762542-1-dmitry.baryshkov@linaro.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
The newly added memset() causes a warning for some reason I could not
figure out:
In file included from arch/x86/include/asm/string.h:3,
from drivers/gpu/drm/i915/gt/intel_rc6.c:6:
In function 'rc6_res_reg_init',
inlined from 'intel_rc6_init' at drivers/gpu/drm/i915/gt/intel_rc6.c:610:2:
arch/x86/include/asm/string_32.h:195:29: error: '__builtin_memset' writing 16 bytes into a region of size 0 overflows the destination [-Werror=stringop-overflow=]
195 | #define memset(s, c, count) __builtin_memset(s, c, count)
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/i915/gt/intel_rc6.c:584:9: note: in expansion of macro 'memset'
584 | memset(rc6->res_reg, INVALID_MMIO_REG.reg, sizeof(rc6->res_reg));
| ^~~~~~
In function 'intel_rc6_init':
Change it to an normal initializer and an added memcpy() that does not have
this problem.
Fixes: 4bb9ca7ee0 ("drm/i915/mtl: C6 residency and C state type for MTL SAMedia")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20231016201012.1022812-1-arnd@kernel.org
Currently all module parameters are handled by i915_param.c/h. This
is a problem for display parameters when Xe driver is used. Add
a mechanism to add parameters specific to the display. This is mainly
copied from i915_[debugfs]_params.[ch]. Parameters are not yet moved. This
is done by subsequent patches.
v2:
- Drop unused predefinition (dentry)
- Clarify need for empty INTEL_DISPLAY_PARAMS_FOR_EACH in comment
Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20231024124109.384973-2-jouni.hogander@intel.com
Today we got a report at [1] for rcu stalls on the i915 testsuite in [2]
due to the conversion of files to SLAB_TYPSSAFE_BY_RCU. Afaict,
get_file_rcu() goes into an infinite loop trying to carefully verify
that i915->gem.mmap_singleton hasn't changed - see the splat below.
So I stared at this code to figure out what it actually does. It seems
that the i915->gem.mmap_singleton pointer itself never had rcu semantics.
The i915->gem.mmap_singleton is replaced in
file->f_op->release::singleton_release():
static int singleton_release(struct inode *inode, struct file *file)
{
struct drm_i915_private *i915 = file->private_data;
cmpxchg(&i915->gem.mmap_singleton, file, NULL);
drm_dev_put(&i915->drm);
return 0;
}
The cmpxchg() is ordered against a concurrent update of
i915->gem.mmap_singleton from mmap_singleton(). IOW, when
mmap_singleton() fails to get a reference on i915->gem.mmap_singleton:
While mmap_singleton() does
rcu_read_lock();
file = get_file_rcu(&i915->gem.mmap_singleton);
rcu_read_unlock();
it allocates a new file via anon_inode_getfile() and does
smp_store_mb(i915->gem.mmap_singleton, file);
So, then what happens in the case of this bug is that at some point
fput() is called and drops the file->f_count to zero leaving the pointer
in i915->gem.mmap_singleton in tact.
Now, there might be delays until
file->f_op->release::singleton_release() is called and
i915->gem.mmap_singleton is set to NULL.
Say concurrently another task hits mmap_singleton() and does:
rcu_read_lock();
file = get_file_rcu(&i915->gem.mmap_singleton);
rcu_read_unlock();
When get_file_rcu() fails to get a reference via atomic_inc_not_zero()
it will try the reload from i915->gem.mmap_singleton expecting it to be
NULL, assuming it has comparable semantics as we expect in
__fget_files_rcu().
But it hasn't so it reloads the same pointer again, trying the same
atomic_inc_not_zero() again and doing so until
file->f_op->release::singleton_release() of the old file has been
called.
So, in contrast to __fget_files_rcu() here we want to not retry when
atomic_inc_not_zero() has failed. We only want to retry in case we
managed to get a reference but the pointer did change on reload.
<3> [511.395679] rcu: INFO: rcu_preempt detected stalls on CPUs/tasks:
<3> [511.395716] rcu: Tasks blocked on level-1 rcu_node (CPUs 0-9): P6238
<3> [511.395934] rcu: (detected by 16, t=65002 jiffies, g=123977, q=439 ncpus=20)
<6> [511.395944] task:i915_selftest state:R running task stack:10568 pid:6238 tgid:6238 ppid:1001 flags:0x00004002
<6> [511.395962] Call Trace:
<6> [511.395966] <TASK>
<6> [511.395974] ? __schedule+0x3a8/0xd70
<6> [511.395995] ? asm_sysvec_apic_timer_interrupt+0x1a/0x20
<6> [511.396003] ? lockdep_hardirqs_on+0xc3/0x140
<6> [511.396013] ? asm_sysvec_apic_timer_interrupt+0x1a/0x20
<6> [511.396029] ? get_file_rcu+0x10/0x30
<6> [511.396039] ? get_file_rcu+0x10/0x30
<6> [511.396046] ? i915_gem_object_mmap+0xbc/0x450 [i915]
<6> [511.396509] ? i915_gem_mmap+0x272/0x480 [i915]
<6> [511.396903] ? mmap_region+0x253/0xb60
<6> [511.396925] ? do_mmap+0x334/0x5c0
<6> [511.396939] ? vm_mmap_pgoff+0x9f/0x1c0
<6> [511.396949] ? rcu_is_watching+0x11/0x50
<6> [511.396962] ? igt_mmap_offset+0xfc/0x110 [i915]
<6> [511.397376] ? __igt_mmap+0xb3/0x570 [i915]
<6> [511.397762] ? igt_mmap+0x11e/0x150 [i915]
<6> [511.398139] ? __trace_bprintk+0x76/0x90
<6> [511.398156] ? __i915_subtests+0xbf/0x240 [i915]
<6> [511.398586] ? __pfx___i915_live_setup+0x10/0x10 [i915]
<6> [511.399001] ? __pfx___i915_live_teardown+0x10/0x10 [i915]
<6> [511.399433] ? __run_selftests+0xbc/0x1a0 [i915]
<6> [511.399875] ? i915_live_selftests+0x4b/0x90 [i915]
<6> [511.400308] ? i915_pci_probe+0x106/0x200 [i915]
<6> [511.400692] ? pci_device_probe+0x95/0x120
<6> [511.400704] ? really_probe+0x164/0x3c0
<6> [511.400715] ? __pfx___driver_attach+0x10/0x10
<6> [511.400722] ? __driver_probe_device+0x73/0x160
<6> [511.400731] ? driver_probe_device+0x19/0xa0
<6> [511.400741] ? __driver_attach+0xb6/0x180
<6> [511.400749] ? __pfx___driver_attach+0x10/0x10
<6> [511.400756] ? bus_for_each_dev+0x77/0xd0
<6> [511.400770] ? bus_add_driver+0x114/0x210
<6> [511.400781] ? driver_register+0x5b/0x110
<6> [511.400791] ? i915_init+0x23/0xc0 [i915]
<6> [511.401153] ? __pfx_i915_init+0x10/0x10 [i915]
<6> [511.401503] ? do_one_initcall+0x57/0x270
<6> [511.401515] ? rcu_is_watching+0x11/0x50
<6> [511.401521] ? kmalloc_trace+0xa3/0xb0
<6> [511.401532] ? do_init_module+0x5f/0x210
<6> [511.401544] ? load_module+0x1d00/0x1f60
<6> [511.401581] ? init_module_from_file+0x86/0xd0
<6> [511.401590] ? init_module_from_file+0x86/0xd0
<6> [511.401613] ? idempotent_init_module+0x17c/0x230
<6> [511.401639] ? __x64_sys_finit_module+0x56/0xb0
<6> [511.401650] ? do_syscall_64+0x3c/0x90
<6> [511.401659] ? entry_SYSCALL_64_after_hwframe+0x6e/0xd8
<6> [511.401684] </TASK>
Link: [1]: https://lore.kernel.org/intel-gfx/SJ1PR11MB6129CB39EED831784C331BAFB9DEA@SJ1PR11MB6129.namprd11.prod.outlook.com
Link: [2]: https://intel-gfx-ci.01.org/tree/linux-next/next-20231013/bat-dg2-11/igt@i915_selftest@live@mman.html#dmesg-warnings10963
Cc: Jann Horn <jannh@google.com>,
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/20231025-formfrage-watscheln-84526cd3bd7d@brauner
Signed-off-by: Christian Brauner <brauner@kernel.org>
The steering control and semaphore registers are inside an "always on"
power domain with respect to RC6. However there are some issues if
higher-level platform sleep states are entering/exiting at the same time
these registers are accessed. Grabbing GT forcewake and holding it over
the entire lock/steer/unlock cycle ensures that those sleep states have
been fully exited before we access these registers.
This is expected to become a formally documented/numbered workaround
soon.
Note that this patch alone isn't expected to have an immediately
noticeable impact on MCR (mis)behavior; an upcoming pcode firmware
update will also be necessary to provide the other half of this
workaround.
v2:
- Move the forcewake inside the Xe_LPG-specific IP version check. This
should only be necessary on platforms that have a steering semaphore.
Fixes: 3100240bf8 ("drm/i915/mtl: Add hardware-level lock for steering")
Cc: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Cc: Jonathan Cavitt <jonathan.cavitt@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Reviewed-by: Radhakrishna Sripada <radhakrishna.sripada@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20231019170241.2102037-2-matthew.d.roper@intel.com
(cherry picked from commit 8fa1c7cd1f)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
drm/logicvc driver is depend on REGMAP and REGMAP_MMIO, should select this
two kconfig option, otherwise the driver failed to compile on platform
without REGMAP_MMIO selected:
ERROR: modpost: "__devm_regmap_init_mmio_clk" [drivers/gpu/drm/logicvc/logicvc-drm.ko] undefined!
make[1]: *** [scripts/Makefile.modpost:136: Module.symvers] Error 1
make: *** [Makefile:1978: modpost] Error 2
Signed-off-by: Sui Jingfeng <suijingfeng@loongson.cn>
Acked-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
Fixes: efeeaefe9b ("drm: Add support for the LogiCVC display controller")
Link: https://patchwork.freedesktop.org/patch/msgid/20230608024207.581401-1-suijingfeng@loongson.cn
Signed-off-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
