ubmitted sequence number
> > afterwards.
> >
> > Signed-off-by: Tomeu Vizoso
>
> Does this need a Fixes tag?
Yes - based on the READ_ONCE(..) line, it should be
Fixes: 27b67278e007 ("drm/etnaviv: rework MMU handling")
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
u/drm/v3d/v3d_sched.c| 68 +-
> drivers/gpu/drm/xe/xe_execlist.c | 16 +++--
> drivers/gpu/drm/xe/xe_gpu_scheduler.c | 17 +-
> include/drm/gpu_scheduler.h| 37 ++--
> 14 files changed, 206 insertions(+), 151 deletions(-)
&g
ome bits in there are not static, in this particular case,
> gpu->dev is not static for instance.
>
Ah.. yes - thanks.
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
-- a/drivers/gpu/drm/xe/xe_gpu_scheduler.c
> +++ b/drivers/gpu/drm/xe/xe_gpu_scheduler.c
> @@ -63,13 +63,24 @@ int xe_sched_init(struct xe_gpu_scheduler *sched,
> atomic_t *score, const char *name,
> struct device *dev)
> {
> + const struct drm_sched
>
> Changes made with the following Coccinelle rules:
>
> @@ constant C; @@
>
> - msecs_to_jiffies(C * 1000)
> + secs_to_jiffies(C)
>
> @@ constant C; @@
>
> - msecs_to_jiffies(C * MSEC_PER_SEC)
> + secs_to_jiffies(C)
>
> Signed-off-by: Easwar
From: Christian Gmeiner
Add a new ioctl, DRM_IOCTL_V3D_PERFMON_SET_GLOBAL, to allow
configuration of a global performance monitor (perfmon).
Use the global perfmon for all jobs to ensure consistent
performance tracking across submissions. This feature is
needed to implement a Perfetto
From: Christian Gmeiner
Use the new helper to export stats about memory usage.
Signed-off-by: Christian Gmeiner
---
drivers/gpu/drm/etnaviv/etnaviv_drv.c | 12 +++-
drivers/gpu/drm/etnaviv/etnaviv_gem.c | 12
2 files changed, 23 insertions(+), 1 deletion(-)
diff --git a
Hi Maira,
>
> The patch overall LGTM, I just have one small nit.
>
Hopefully v4 will be the last iteration.
> On 17/11/24 18:41, Christian Gmeiner wrote:
> > From: Christian Gmeiner
> >
> > Add a new ioctl, DRM_IOCTL_V3D_PERFMON_SET_GLOBAL, to allow
> > c
From: Christian Gmeiner
If the active performance monitor (v3d->active_perfmon) is being
destroyed, stop it first. Currently, the active perfmon is not
stopped during destruction, leaving the v3d->active_perfmon pointer
stale. This can lead to undefined behavior and instability.
This
From: Christian Gmeiner
Add a new ioctl, DRM_IOCTL_V3D_PERFMON_SET_GLOBAL, to allow
configuration of a global performance monitor (perfmon).
Use the global perfmon for all jobs to ensure consistent
performance tracking across submissions. This feature is
needed to implement a Perfetto
-by: Maíra Canal
Reviewed-by: Christian Gmeiner
---
drivers/gpu/drm/v3d/v3d_debugfs.c | 4 ++--
drivers/gpu/drm/v3d/v3d_perfmon.c | 15 ---
drivers/gpu/drm/v3d/v3d_regs.h| 29 +
3 files changed, 27 insertions(+), 21 deletions(-)
diff --git a/d
Hi Maíra,
> On 31/10/24 17:51, Christian Gmeiner wrote:
> > From: Christian Gmeiner
> >
> > Add a new ioctl, DRM_IOCTL_V3D_PERFMON_SET_GLOBAL, to allow
> > configuration of a global performance monitor (perfmon).
> > Use the global perfmon for all jobs to
Hi Maíra,
>
> Thanks for your patch! Just some small nits.
>
Thanks for your review.
> On 31/10/24 17:51, Christian Gmeiner wrote:
> > From: Christian Gmeiner
> >
> > Add a new ioctl, DRM_IOCTL_V3D_PERFMON_SET_GLOBAL, to allow
> > configuration of a
Hi André
>
> Em 20/10/2024 17:41, Christian Gmeiner escreveu:
> > From: Christian Gmeiner
> >
> > This patch adds a new ioctl, DRM_IOCTL_V3D_PERFMON_SET_GLOBAL, which
> > allows the configuration of a global performance monitor (perfmon).
> > The global pe
le and disable this state through the IOCTL.
>
> One last thing is: don't forget to stop the perfmons when you don't use
> it anymore :)
>
I've sent v2 of this patch and hope I've addressed all your comments.
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
From: Christian Gmeiner
Add a new ioctl, DRM_IOCTL_V3D_PERFMON_SET_GLOBAL, to allow
configuration of a global performance monitor (perfmon).
Use the global perfmon for all jobs to ensure consistent
performance tracking across submissions.
Signed-off-by: Christian Gmeiner
---
Changes in v2
Hi Lucas
>
> The shader L1 cache is a writeback cache for shader loads/stores
> and thus must be flushed before any BOs backing the shader buffers
> are potentially freed.
>
> Cc: sta...@vger.kernel.org
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
>
han that.
>
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> drivers/gpu/drm/etnaviv/etnaviv_gpu.c | 3 +--
> 1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c
> b/drivers/gpu/drm/etnaviv/etnaviv_gpu
From: Christian Gmeiner
This patch adds a new ioctl, DRM_IOCTL_V3D_PERFMON_SET_GLOBAL, which
allows the configuration of a global performance monitor (perfmon).
The global perfmon is used for all jobs, ensuring consistent performance
tracking across submissions.
Signed-off-by: Christian Gmeiner
Similar to commit f2a4bcb25328 ("drm/v3d: Use v3d_perfmon_find()"),
replace the open-coded `vc4_perfmon_find()` with the real thing.
Cc: Christian Gmeiner
Signed-off-by: Maíra Canal
Reviewed-by: Christian Gmeiner
---
drivers/gpu/drm/vc4/vc4_perfmon.c | 6 +-
1 file
From: Christian Gmeiner
Replace the open coded v3d_perfmon_find(..) with the real
thing.
Signed-off-by: Christian Gmeiner
---
drivers/gpu/drm/v3d/v3d_perfmon.c | 6 +-
1 file changed, 1 insertion(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/v3d/v3d_perfmon.c
b/drivers/gpu/drm/v3d
nd().
>
> Fixes: b72af445cd38 ("drm/etnaviv: request pages from DMA32 zone when needed")
> Suggested-by: Sui Jingfeng
> Signed-off-by: Xiaolei Wang
Reviewed-by: Christian Gmeiner
> ---
>
> change log
>
> v1:
>
> https://patchwork.kernel.org/project/dri-devel
x uninitialized on debug builds, and any subsequent use of the
> mutex is forbidden.
>
> It also could lead to problems if mutex_destroy() gets extended, add
> missing mutex_destroy() to eliminate potential concerns.
>
> Signed-off-by: Sui Jingfeng
Reviewed-by: Christian G
>
> Currently, the etnaviv_gem_submit.c isn't call any runtime power management
> functions. So drop it, we can re-include it when the header really get used
> though.
>
> Signed-off-by: Sui Jingfeng
Reviewed-by: Christian Gmeiner
> ---
> drivers/gpu/drm/etn
t;
> Signed-off-by: Sui Jingfeng
Reviewed-by: Christian Gmeiner
> ---
> drivers/gpu/drm/etnaviv/etnaviv_gem.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.c
> b/drivers/gpu/drm/etnaviv/etnaviv_gem.c
> index
instead of the signed
> type.
>
> Signed-off-by: Sui Jingfeng
Reviewed-by: Christian Gmeiner
> ---
> drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_g
the timeout as long as the
> draw progresses.
>
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> v2:
> - dropped debug register enable/disable
> - locked against concurrent debug register access
> ---
> drivers/gpu/drm/etnaviv/etnaviv_gpu.h | 1 +
>
> Update state_hi.xml.h header from etna_viv commit
> 8f43a34fd9cd ("rndb: document FE current primitve debug reg")
>
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> v2: no changes
> ---
> drivers/gpu/drm/etnaviv/state_hi.xml.h |
they are always accessible performance
> profiling still needs to manage some prerequisites.
>
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> v2: new patch replacing more complex usage tracking from v1
> ---
> drivers/gpu/drm/etnaviv/etnaviv_gpu.
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> v2: new patch
> ---
> drivers/gpu/drm/etnaviv/etnaviv_perfmon.c | 4
> 1 file changed, 4 insertions(+)
>
> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_perfmon.c
> b/drivers/gpu/drm/etnaviv/etnaviv_perf
lock.
>
> Fixes: 68dc0b295dcb ("drm/etnaviv: use 'sync points' for performance monitor
> requests")
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> v2: new patch
> ---
> drivers/gpu/drm/etnaviv/etnaviv_gpu.c | 20 ++--
> 1 f
ully this series will land anytime.
[6.783267] etnaviv :0c:00.0: enabling device ( -> 0003)
[6.783324] etnaviv :0c:00.0: JingJia Micro JM9100 has 1 GPU cores
[6.783459] etnaviv :0c:00.0: bound etnaviv-gpu,3d.0 (ops
gpu_ops [etnaviv])
[6.783502] etnaviv-gpu
b is already removed from the pending list and
> will not be put back when drm_sched_start() isn't called, we must
> make sure to put the job back on the pending list when extending
> the timeout.
>
> Cc: sta...@vger.kernel.org #6.0
> Signed-off-by: Lucas Stach
Reviewed-by: Christ
>
> The dma sync operation needs to be done with DMA_BIDIRECTIONAL when
> the BO is prepared for both read and write operations.
>
> Fixes: a8c21a5451d8 ("drm/etnaviv: add initial etnaviv DRM driver")
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
>
r has determined that
> there are completed jobs and reuse this for all fences that get
> signalled by the handler.
>
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> drivers/gpu/drm/etnaviv/etnaviv_gpu.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 de
ring node, and if we find out that there is no rendering
> hardware, abort and retry with an accel node.
>
I really love this idea of moving away from a render node. What needs to be done
on the userspace side?
> Signed-off-by: Tomeu Vizoso
> Cc: Oded Gabbay
Reviewed-by: Christian Gme
From: Christian Gmeiner
This reverts commit 1dccdba084897443d116508a8ed71e0ac8a031a4.
In userspace a different approach was choosen - hwdb. As a result, there
is no need for these values.
Signed-off-by: Christian Gmeiner
---
drivers/gpu/drm/etnaviv/etnaviv_gpu.c | 20
due to message size and could potentially slow
down data transfer. However, an alternative
approach could be to allocate a sizable chunk of SRAM memory in
userspace. By utilizing memcpy() to copy data into
this memory, followed by a single rpmsg signal to notify the RTOS that
the data is ready, we can leverage the faster access
speed of SRAM compared to DDR from the remoteproc.
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
>
> Core in platform_driver_register() already sets the .owner, so driver
> does not need to. Whatever is set here will be anyway overwritten by
> main driver calling platform_driver_register().
>
> Signed-off-by: Krzysztof Kozlowski
Reviewed-by: Christian Gmeiner
> -
Hi Lucas,
>
> prefix for etnaviv kernel patches should be "drm/etnaviv: ..."
Ah yeah .. corrected in v2.
>
> Am Donnerstag, dem 29.02.2024 um 20:55 +0100 schrieb Christian Gmeiner:
> > From: Christian Gmeiner
> >
> > The hwdb selection logic as a fe
From: Christian Gmeiner
The hwdb selection logic as a feature that allows it to mark some fields
as 'don't care'. If we match with such a field we memcpy(..)
the current etnaviv_chip_identity into ident.
This step can overwrite some id values read from the GPU with the
'don
From: Christian Gmeiner
The hwdb selection logic as a feature that allows it to mark some fields
as 'don't care'. If we match with such a field we memcpy(..)
the current etnaviv_chip_identity into ident.
This step can overwrite some id values read from the GPU with the
'don
gpu, 0x320, 0x5007) ||
> +etnaviv_is_model_rev(gpu, 0x320, 0x5220)) &&
> gpu_read(gpu, VIVS_HI_CHIP_TIME) != 0x2062400) {
> u32 mc_memory_debug;
>
> @@ -745,7 +747,7 @@ static void etnaviv_gpu_hw_init(struct etnaviv_gpu *gpu)
> VIVS_HI_AXI_CONFIG_ARCACHE(2));
>
> /* GC2000 rev 5108 needs a special bus config */
> - if (etnaviv_is_model_rev(gpu, GC2000, 0x5108)) {
> + if (etnaviv_is_model_rev(gpu, 0x2000, 0x5108)) {
> u32 bus_config = gpu_read(gpu, VIVS_MC_BUS_CONFIG);
> bus_config &= ~(VIVS_MC_BUS_CONFIG_FE_BUS_CONFIG__MASK |
> VIVS_MC_BUS_CONFIG_TX_BUS_CONFIG__MASK);
>
> --
> 2.39.2
>
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
>
> Update the state HI header from the rnndb commit
> 8d7ee714cfe2 ("Merge pull request #24 from pH5/unknown-3950").
>
> Signed-off-by: Philipp Zabel
You missed my R-b from the v1 series for this patch - please include
it the next time!
Reviewed-by: Christian Gmeiner
VIPNano-Si+ NPU on i.MX8MP.
>
> Signed-off-by: Philipp Zabel
Thanks - series is
Reviewed-by: Christian Gmeiner
> ---
> Philipp Zabel (2):
> drm/etnaviv: Update hardware headers from rnndb
> drm/etnaviv: Disable SH_EU clock gating on VIPNano-Si+
>
> dri
Hi Lucas
>
> Am Donnerstag, dem 07.12.2023 um 08:43 +0100 schrieb Christian Gmeiner:
> > From: Christian Gmeiner
> >
> > 0x1540 is the address of 4th render target address pair (two pixel pipes).
> >
> Are there any corresponding TS states for this region?
I wi
From: Christian Gmeiner
0x1540 is the address of 4th render target address pair (two pixel pipes).
Signed-off-by: Christian Gmeiner
---
drivers/gpu/drm/etnaviv/etnaviv_cmd_parser.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_cmd_parser.c
b/drivers/gpu
ll the new values but it looks fine to me.
Acked-by: Christian Gmeiner
> ---
>
> v2: Update a few chipspecs that I had missed before. (Christian)
> ---
> drivers/gpu/drm/etnaviv/etnaviv_gpu.c | 20 +++
> drivers/gpu/drm/etnaviv/etnaviv_gpu.h | 12 +
>
ude/uapi/drm/etnaviv_drm.h
> index af024d90453d..d87410a8443a 100644
> --- a/include/uapi/drm/etnaviv_drm.h
> +++ b/include/uapi/drm/etnaviv_drm.h
> @@ -77,6 +77,11 @@ struct drm_etnaviv_timespec {
> #define ETNAVIV_PARAM_GPU_PRODUCT_ID0x1c
> #define ETNAVIV_PARAM_GPU_CUSTOMER_ID
e who is responsible for this job.
>
> Thanks!
>
> -Kees
>
> >
> > Link:
> > https://www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings
> > [1]
> > Link: https://github.com/KSPP/linux/issues/90
> > Cc: lin
s returning zero in
> the remove callback to the void returning variant.
>
> Reviewed-by: Thomas Zimmermann
> Reviewed-by: Jyri Sarha
> Signed-off-by: Uwe Kleine-König
>
Reviewed-by: Christian Gmeiner
---
> drivers/gpu/drm/etnaviv/etnaviv_drv.c | 6 ++
> dr
Hi Lucas,
Am Mi., 21. Juni 2023 um 17:44 Uhr schrieb Christian Gmeiner
:
>
> Hi Lucas,
>
> Am Mo., 17. Apr. 2023 um 19:42 Uhr schrieb Christian Gmeiner
> :
> >
> > Hi Lucas
> >
> > >
> > > gpu->mmu_context is the MMU context of the last job i
Hi Lucas,
Am Mo., 17. Apr. 2023 um 19:42 Uhr schrieb Christian Gmeiner
:
>
> Hi Lucas
>
> >
> > gpu->mmu_context is the MMU context of the last job in the HW queue, which
> > isn't necessarily the same as the context from the bad job. Dump the MMU
> >
ci output for the GPU? Something like
this: sudo lspci -vvv -s ...
thanks
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> drivers/gpu/drm/etnaviv/etnaviv_gpu.c | 2 ++
> drivers/gpu/drm/etnaviv/etnaviv_gpu.h | 1 +
> drivers/gpu/drm/etnaviv/etnaviv_sched.c | 5 +++--
> 3 files changed, 6 insertions(+), 2 deletions(-)
>
> diff --g
tialized object.
>
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> drivers/gpu/drm/etnaviv/etnaviv_cmdbuf.c | 3 +++
> drivers/gpu/drm/etnaviv/etnaviv_gpu.c| 9 ++---
> drivers/gpu/drm/etnaviv/etnaviv_gpu.h| 1 -
> drivers/gpu/drm/etnaviv
in that
> state.
>
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> drivers/gpu/drm/etnaviv/etnaviv_gpu.c | 25 ++---
> drivers/gpu/drm/etnaviv/etnaviv_gpu.h | 10 +-
> 2 files changed, 27 insertions(+), 8 deletions(-)
>
s also removes the need to guard against the state where the driver
> isn't fully initialized yet in the runtime PM resume handler.
>
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> drivers/gpu/drm/etnaviv/etnaviv_gpu.c | 15 +--
> 1 file c
Hi Lucas,
Am Do., 15. Juni 2023 um 11:37 Uhr schrieb Lucas Stach :
>
> Hi Christian,
>
> Am Mittwoch, dem 14.06.2023 um 20:41 +0200 schrieb Christian Gmeiner:
> > Hi Lucas
> >
> > >
> > > Conceptually events are the right abstraction to handle the GPU
&g
);
> }
>
> static int etnaviv_hw_reset(struct etnaviv_gpu *gpu)
> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.h
> b/drivers/gpu/drm/etnaviv/etnaviv_gpu.h
> index 98c6f9c320fc..e1e1de59c38d 100644
> --- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.h
> +++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.h
> @@ -150,6 +150,7 @@ struct etnaviv_gpu {
> struct clk *clk_shader;
>
> unsigned int freq_scale;
> + unsigned int fe_waitcycles;
> unsigned long base_rate_core;
> unsigned long base_rate_shader;
> };
> --
> 2.39.2
>
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
have some unknown discrete steps in the wait cycles
> >> add a comma here.
> >>>so
> >>> we may over/undershoot the target a bit, but that should be harmless.
> >> overshoot or undershoot
> >>> Signed-off-by: Lucas Stach
> >>&g
d-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> drivers/gpu/drm/etnaviv/etnaviv_gpu.c | 9 +
> 1 file changed, 5 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c
> b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c
> index
tnaviv_gem_submit *submit)
> unsigned int i, nr_events = 1, event[3];
> int ret;
>
> - if (!submit->runtime_resumed) {
> - ret = pm_runtime_get_sync(gpu->dev);
> - if (ret < 0) {
> - pm_runtime_put_noidle(gpu->dev);
> - return NULL;
> - }
> - submit->runtime_resumed = true;
> - }
> -
> /*
> * if there are performance monitor requests we need to have
> * - a sync point to re-configure gpu and process ETNA_PM_PROCESS_PRE
> --
> 2.39.2
>
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
gpu->event_spinlock);
> for_each_set_bit(i, gpu->event_bitmap, ETNA_NR_EVENTS)
> - complete(&gpu->event_free);
> - bitmap_zero(gpu->event_bitmap, ETNA_NR_EVENTS);
> + event_free(gpu, i);
> spin_unlock(&gpu->event_spi
Hi Lucas
>
> So it can use the event_free function without adding another
> forward declaration. No functional change.
>
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> drivers/gpu/drm/etnaviv/etnaviv_gpu.c | 88 +--
> 1 fil
should not hurt most use-cases. The FE WAIT
> command seems to have some unknown discrete steps in the wait cycles so
> we may over/undershoot the target a bit, but that should be harmless.
>
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> drivers/gpu/drm/etna
>
> Module level clock gating and the pulse eater might interfere with
> the GPU reset, as they both have the potential to stop the clock
> and thus reset propagation to parts of the GPU.
>
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> I'm
17e4660ae3d7 ("drm/etnaviv: implement per-process address spaces on
> MMUv2")
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> drivers/gpu/drm/etnaviv/etnaviv_dump.c | 14 +++---
> 1 file changed, 7 insertions(+), 7 deletions(-)
>
>
Hi Andrew
Am Di., 4. Apr. 2023 um 17:02 Uhr schrieb Christian Gmeiner
:
>
> > > Hi Andrew
> > >
> > >>
> > >>
> > >> Okay, will split for v2.
> > >>
> > >>
> > >
> > > Was there a follow-up v2 of
rk on it? If not I would like to help out
> > as we have a use case where we want to
> > use a dma-buf sram exporter.
> >
> >
>
> Sure, I've been keeping it alive in our evil vendor tree, but if
> there is interest upstream now I'll post a v2 and CC you.
That would be great!
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
f sram exporter.
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
d-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c | 10 +-
> 1 file changed, 9 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c
> b/drivers/gpu/drm/etnaviv/etnaviv_gem_pri
Hi Lucas
Am Mi., 30. Nov. 2022 um 19:53 Uhr schrieb Lucas Stach :
>
> From: Christian Gmeiner
>
> The MMU tells us the fault status. While the raw register value is
> already printed, it's a bit more user friendly to translate the
> fault reasons into human readable fo
rs describe the interpretation of the tag bits in this
> buffer.
>
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
r possible solutions:
1) Add two seperate helpers ala gpu_read_power() and gpu_write_power()
where we do the if beast.
2) Add a power register offset variable to etnaviv_gpu and explicitly
use it on for reads and writes - like the Vivante driver does.
But that's just my personal opinion. Can't wait to hear what Lucas thinks.
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
>
> The GC300's features register doesn't specify that a 2D pipe is
> available, and like the GC600, its idle register reports zero bits where
> modules aren't present.
>
> Signed-off-by: Doug Brown
Reviewed-by: Christian Gmeiner
> ---
> dri
o the same check
> to bundle them.
>
> Signed-off-by: Marco Felsch
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> drivers/gpu/drm/etnaviv/etnaviv_gpu.c | 9 +
> 1 file changed, 5 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/gpu
han
forcing users to update their kernels to get this new database entry.
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
s a starting point. I have no problem introducing per
process load avg
values .. lets see how the next version of this series will look like,
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
Am Fr., 24. Juni 2022 um 11:38 Uhr schrieb Lucas Stach :
>
> Am Dienstag, dem 21.06.2022 um 09:20 +0200 schrieb Christian Gmeiner:
> > The GPU has an idle state register where each bit represents the idle
> > state of a sub-GPU component like FE or TX. Sample this register
Hi Lucas
>
> Am Dienstag, dem 21.06.2022 um 09:20 +0200 schrieb Christian Gmeiner:
> > This adds a SMA algorithm inspired by Exponentially weighted moving
> > average (EWMA) algorithm found in the kernel.
> >
> Still not sure about this one. I _feel_ that a simple mov
Make it possible to access the sub-GPU component load value from
user space with the perfmon infrastructure.
Signed-off-by: Christian Gmeiner
---
drivers/gpu/drm/etnaviv/etnaviv_perfmon.c | 79 +++
1 file changed, 79 insertions(+)
diff --git a/drivers/gpu/drm/etnaviv
percentage based load of each sub-GPU
component.
Signed-off-by: Christian Gmeiner
---
drivers/gpu/drm/etnaviv/etnaviv_drv.c | 14 ++
drivers/gpu/drm/etnaviv/etnaviv_gpu.c | 64 ++-
drivers/gpu/drm/etnaviv/etnaviv_gpu.h | 37
3 files changed, 114 insertions
Might be helpful to see the loadavg in debugfs.
Signed-off-by: Christian Gmeiner
---
drivers/gpu/drm/etnaviv/etnaviv_gpu.c | 12 +++-
1 file changed, 11 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c
b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c
index
This adds a SMA algorithm inspired by Exponentially weighted moving
average (EWMA) algorithm found in the kernel.
Signed-off-by: Christian Gmeiner
---
drivers/gpu/drm/etnaviv/etnaviv_sma.h | 53 +++
1 file changed, 53 insertions(+)
create mode 100644 drivers/gpu/drm
This patch series add support for loadavg values for GPU
sub-components. I am adding a SMA algorithm as I was not
really sure if EWMA would be a good fit for this use case.
Changes v2:
- Addressed feedback from Lucas
Christian Gmeiner (4):
drm/etnaviv: add simple moving average (SMA)
drm
Track the pid per submit, so we can print the name and cmdline of
the task which submitted the batch that caused the gpu to hang.
Signed-off-by: Christian Gmeiner
---
drivers/gpu/drm/etnaviv/etnaviv_gem.h| 1 +
drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c | 6 ++
drivers/gpu/drm
8ddd ("drm/etnaviv: limit submit sizes")
> Cc: sta...@vger.kernel.org
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
U resets, also remember and consider the last completed fence seqno
> in the hang check.
>
> Reported-by: Joerg Albert
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
ngle submit seems like a reasonably future-proof number for now. This
> number can be bumped if needed without breaking the interface.
>
> Cc: sta...@vger.kernel.org
> Reported-by: Dan Carpenter
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
put it in read-only memory.
>
> Signed-off-by: Rikard Falkeborn
Reviewed-by: Christian Gmeiner
> ---
> drivers/gpu/drm/etnaviv/etnaviv_gpu.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c
> b/driver
ed with coccinelle:
>
> @@
> expression E;
> @@
> - flush_workqueue(E);
> destroy_workqueue(E);
>
> Signed-off-by: Christophe JAILLET
For drm/etnaviv:
Reviewed-by: Christian Gmeiner
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
Am Do., 26. Aug. 2021 um 14:10 Uhr schrieb Michael Walle :
>
> There is already a macro for the magic value. Use it.
>
> Signed-off-by: Michael Walle
Reviewed-by: Christian Gmeiner
I will wait for v2 for the rest of the changes to review.
> ---
> drivers/gpu/drm/etnaviv
Add the
> mssing context put to fix the leak.
>
> Cc: sta...@vger.kernel.org # 5.4
> Signed-off-by: Lucas Stach
> Tested-by: Michael Walle
Series is:
Reviewed-by: Christian Gmeiner
> ---
> drivers/gpu/drm/etnaviv/etnaviv_mmu.c | 1 +
> 1 file changed, 1 insertion(+)
Hi Lucas
Am Di., 24. Aug. 2021 um 09:54 Uhr schrieb Lucas Stach :
>
> Am Dienstag, dem 24.08.2021 um 09:24 +0200 schrieb Christian Gmeiner:
> > Am Fr., 20. Aug. 2021 um 22:18 Uhr schrieb Lucas Stach
> > :
> > >
> > > Move the refcount manipulation of th
&
> VIVS_MMUv2_SEC_CONTROL_ENABLE)
> return;
>
> + if (gpu->mmu_context)
> + etnaviv_iommu_context_put(gpu->mmu_context);
> + gpu->mmu_context = etnaviv_iommu_context_get(context);
> +
I have seen this pattern now more than two times - maybe put the
assignment of a new mmu context into its own function?
> gpu_write(gpu, VIVS_MMUv2_PTA_ADDRESS_LOW,
> lower_32_bits(context->global->v2.pta_dma));
> gpu_write(gpu, VIVS_MMUv2_PTA_ADDRESS_HIGH,
> --
> 2.30.2
>
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
kvfree(etnaviv_obj->pages);
> + kvfree(etnaviv_obj->pages);
>
> drm_prime_gem_destroy(&etnaviv_obj->base, etnaviv_obj->sgt);
> }
> --
> 2.31.0
>
Thanks for the patch, but there is an other one queued up in
etnaviv/next that fixes the same issue:
https://git.pengutronix.de/c
2.31.0
>
Thanks for the patch, but there is an other one queued up in
etnaviv/next that fixes the same issue:
https://git.pengutronix.de/cgit/lst/linux/commit/?h=etnaviv/next&id=bdf622e0fade2cec72c948c708763378b656c01d
--
greets
--
Christian Gmeiner, MSc
https://christian-gmeiner.info/privacypolicy
Am Di., 1. Juni 2021 um 16:06 Uhr schrieb Zheng Yongjun
:
>
> Eliminate the following coccicheck warning:
>
> drivers/gpu/drm/etnaviv/etnaviv_gem_prime.c:84:2-8: WARNING:
> NULL check before some freeing functions is not needed.
>
> Signed-off-by: Zheng Yongjun
Reviewed-
t; is obviously bad.
>
> Rework the linear window offset calculation to be based on the command buffer
> physical address, making sure that the command buffer is always mappable.
>
> Signed-off-by: Lucas Stach
Reviewed-by: Christian Gmeiner
> ---
> driver
1 - 100 of 478 matches
Mail list logo
