b2bf141b6a
perfetto/pps-producer: add optimized cpu/gpu timestamp correlation support
...
The Intel Xe driver added the ability to do cpu/gpu timestamp
correlation giving a much better alignment of timestamps (we use to
have ~20us delta between the 2 samples, just because of the ioctl
barrier potentially sneaking in some work).
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: José Roberto de Souza <jose.souza@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/24591 >
2023-11-09 13:22:48 +00:00
80dd6ed074
intel: Add kmd_type parameter to necessary intel_gem.h functions
...
Here adding kmd_type parameter to
intel_gem_read_render_timestamp(), intel_gem_can_render_on_fd() and
intel_gem_supports_protected_context().
Those 3 functions will have Xe implementations, the other functions
in intel_gem.h will not be called by Xe code paths so not adding
kernel_driver_type to it.
Reviewed-by: Marcin Ślusarz <marcin.slusarz@intel.com >
Signed-off-by: José Roberto de Souza <jose.souza@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20773 >
2023-01-25 09:17:15 -08:00
1e87834980
intel: Share code to read render timestamp
...
Timestamp read is not in any hot path so there is no down-sides in
share the same function between iris, crocus, anv and hasvk.
Also while at it also dropping the functions to read MMIO from kernel,
the only use is read render timestamp so we don't need it.
v2:
- fix compilaton of ds
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Signed-off-by: José Roberto de Souza <jose.souza@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/18920 >
2022-10-03 06:26:50 -07:00
5398c9183e
intel/ds: fix compilation
...
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/6518
Fixes: efc2782f97jordan.l.justen@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16601 >
2022-05-19 16:42:41 +00:00
9d0db8d4c4
intel/perf: deal with OA reports timestamp values on DG2
...
OA reports on XeHP have their timestamp shifted to the left by 1. To
get that back in the same time domain as the REG_READ you need to
shift it back to the right and you're loosing the top bit.
v2: use ull for 64bit constant (Ian)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Ian Romanick <ian.d.romanick@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16144 >
2022-05-17 19:55:10 +00:00
cf9956a8c5
intel/ds: use a per GPU clock ID
...
Multi-GPU setups :)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Rohan Garg <rohan.garg@intel.com >
Acked-by: Antonio Caggiano <antonio.caggiano@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13996 >
2022-01-14 20:17:44 +00:00
61766f9f90
intel/ds: use the right i915_drm.h include location
...
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Rohan Garg <rohan.garg@intel.com >
Acked-by: Antonio Caggiano <antonio.caggiano@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13996 >
2022-01-14 20:17:44 +00:00
e35a65ae45
intel/ds: don't forget to reset upper dword timestamp read
...
This could lead to confusing if the 32bits roll over (every ~6mn or
so).
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Fixes: 4ef6698a26rohan.garg@intel.com >
Acked-by: Antonio Caggiano <antonio.caggiano@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13996 >
2022-01-14 20:17:44 +00:00
6eb554a9c7
intel/ds: allow user to select metric set at start time
...
Rather than using always the same metric set, let the user choose when
starting the producer with :
INTEL_PERFETTO_METRIC_SET=RasterizerAndPixelBackend ./build/src/tool/pps/pps-producer
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Rohan Garg <rohan.garg@intel.com >
Acked-by: Antonio Caggiano <antonio.caggiano@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13996 >
2022-01-14 20:17:44 +00:00
69df00b33b
intel/ds: reuse intel_ioctl()
...
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Rohan Garg <rohan.garg@intel.com >
Acked-by: Antonio Caggiano <antonio.caggiano@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13996 >
2022-01-14 20:17:44 +00:00
d3724de894
intel/dev,perf: Use a single timescale function
...
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Rohan Garg <rohan.garg@intel.com >
Acked-by: Antonio Caggiano <antonio.caggiano@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13996 >
2022-01-14 20:17:44 +00:00
5a2cff9bc8
intel: move timestamp scaling helper to intel/perf
...
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Antonio Caggiano <antonio.caggiano@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13831 >
2021-11-22 11:52:46 +00:00
6126742648
intel/ds: remove verbose messages
...
At high frequency sampling, this generates a lot of messages.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Antonio Caggiano <antonio.caggiano@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13831 >
2021-11-22 11:52:46 +00:00
3d71e35857
intel/ds: isolate intel/perf from the pps-producer
...
Otherwise we need to include intel headers in generic code.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Antonio Caggiano <antonio.caggiano@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13831 >
2021-11-22 11:52:46 +00:00
4ef6698a26
intel/ds: drop timestamp correlation code
...
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Antonio Caggiano <antonio.caggiano@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13831 >
2021-11-22 11:52:46 +00:00
8657fa6b86
pps: allow drivers to report timestamps in their own time domain
...
For this each driver must :
- report its clock_id (if no particular clock just default to cpu
boottime one)
- be able to sample its clock (gpu_timestamp())
The PPSDataSource will then emit timestamp correlation events in the
trace ensuring perfetto is able to display GPU & CPU events
appropriately on its timeline.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Antonio Caggiano <antonio.caggiano@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13831 >
2021-11-22 11:52:46 +00:00
43d5b55bc1
intel/pps: provide accurate min sampling period
...
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Reviewed-by: Antonio Caggiano <antonio.caggiano@collabora.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13571 >
2021-10-28 13:16:56 +03:00
92180a4376
pps: Intel pps driver
...
Add the Intel pps driver using functionalities provided by
libintel_perf.
v2: Fix build with perfetto not enabled.
v3: Open perf stream with no filtering.
v4: Drop usage of inc/dec_n_users.
v5: Isolate intel_perf in its own class.
Signed-off-by: Antonio Caggiano <antonio.caggiano@collabora.com >
Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com >
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/10216 >
2021-05-18 14:28:48 +00:00
Lionel Landwerlin