Intel Xe OA IGT's
Subsystems (2)
cover Intel Xe OA IGT's
This series adds OA IGT's for Intel Xe KMD. These tests are for the kernel series posted here:
https://patchwork.freedesktop.org/series/121084/
This series is also available at:
https://gitlab.freedesktop.org/adixit/igt-gpu-tools/-/tree/xe-oa.prep
Only LNL support is included in this series. Xe1 support has not been included because Metrics XML's for those platforms are large and cannot be sent over mail.
The complete series, with Xe1 support, is available here:
https://gitlab.freedesktop.org/adixit/igt-gpu-tools/-/tree/xe-oa
Ashutosh Dixit (18):
lib/xe/oa: Import OA metric generation files from i915 lib/xe/oa: Add LNL metric guids
lib/xe/oa: Add OA LNL metrics (oa_lnl.xml)
lib/xe/oa: Generate LNL metrics/registers files drm-uapi/xe: Sync with Perf/OA changes
lib/xe: Complete xe_oa lib changes
tests/intel/xe_query: Add OA units query test tests/intel/xe_oa: Add first tests
tests/intel/xe_oa: Add some negative tests tests/intel/xe_oa: Add "oa-formats" subtest tests/intel/xe_oa: Add oa exponent tests tests/intel/xe_oa: buffer-fill, non-zero-reason, enable-disable tests/intel/xe_oa: blocking and polling tests tests/intel/xe_oa: OAR/OAC tests
tests/intel/xe_oa: Exclusive/concurrent access, rc6 and stress open
close
tests/intel/xe_oa: add remove OA config tests tests/intel/xe_oa: OA buffer mmap tests tests/intel/xe_oa: Register whitelisting and MMIO trigger tests
include/drm-uapi/xe_drm.h | 321 +- lib/intel_device_info.c | 1 + lib/meson.build | 74 + lib/xe/oa-configs/README.md | 115 + lib/xe/oa-configs/codegen.py | 444 ++ lib/xe/oa-configs/guids.xml | 2763 ++++++++++++ lib/xe/oa-configs/mdapi-xml-convert.py | 1221 +++++ lib/xe/oa-configs/oa-equations-codegen.py | 261 ++ lib/xe/oa-configs/oa-lnl.xml | 2799 ++++++++++++ lib/xe/oa-configs/oa-metricset-codegen.py | 304 ++ lib/xe/oa-configs/oa-registers-codegen.py | 118 + lib/xe/oa-configs/oa_guid_registry.py | 131 + lib/xe/oa-configs/update-guids.py | 223 + lib/xe/xe_oa.c | 699 +++ lib/xe/xe_oa.h | 328 ++ lib/xe/xe_query.c | 38 + lib/xe/xe_query.h | 5 + tests/intel/xe_oa.c | 4924 +++++++++++++++++++++ tests/intel/xe_query.c | 53 + tests/meson.build | 2 + 20 files changed, 14823 insertions(+), 1 deletion(-) create mode 100644 lib/xe/oa-configs/README.md create mode 100644 lib/xe/oa-configs/codegen.py create mode 100644 lib/xe/oa-configs/guids.xml create mode 100755 lib/xe/oa-configs/mdapi-xml-convert.py create mode 100644 lib/xe/oa-configs/oa-equations-codegen.py create mode 100644 lib/xe/oa-configs/oa-lnl.xml create mode 100644 lib/xe/oa-configs/oa-metricset-codegen.py create mode 100644 lib/xe/oa-configs/oa-registers-codegen.py create mode 100644 lib/xe/oa-configs/oa_guid_registry.py create mode 100755 lib/xe/oa-configs/update-guids.py create mode 100644 lib/xe/xe_oa.c create mode 100644 lib/xe/xe_oa.h create mode 100644 tests/intel/xe_oa.c
Signature · 1 line
2.41.0
PATCH 1/18 lib/xe/oa: Import OA metric generation files from i915
Import OA metric generation files from i915 to provide a starting point for Xe OA XML processing code.
--- lib/xe/oa-configs/README.md | 115 + lib/xe/oa-configs/codegen.py | 444 ++++ lib/xe/oa-configs/guids.xml | 2749 +++++++++++++++++++++ lib/xe/oa-configs/mdapi-xml-convert.py | 1196 +++++++++ lib/xe/oa-configs/oa-equations-codegen.py | 261 ++ lib/xe/oa-configs/oa-metricset-codegen.py | 257 ++ lib/xe/oa-configs/oa-registers-codegen.py | 118 + lib/xe/oa-configs/oa_guid_registry.py | 117 + lib/xe/oa-configs/update-guids.py | 222 ++ 9 files changed, 5479 insertions(+) create mode 100644 lib/xe/oa-configs/README.md create mode 100644 lib/xe/oa-configs/codegen.py create mode 100644 lib/xe/oa-configs/guids.xml create mode 100755 lib/xe/oa-configs/mdapi-xml-convert.py create mode 100644 lib/xe/oa-configs/oa-equations-codegen.py create mode 100644 lib/xe/oa-configs/oa-metricset-codegen.py create mode 100644 lib/xe/oa-configs/oa-registers-codegen.py create mode 100644 lib/xe/oa-configs/oa_guid_registry.py create mode 100755 lib/xe/oa-configs/update-guids.py
lib/xe/oa-configs/README.md new file +115 -0
@@ -0,0 +1,115 @@ +# About guids.xml + +This is the authoritive registry of unique identifers for different OA unit +hardware configurations. Userspace can reliably use these identifiers to map a +configuration to corresponding normalization equations and counter meta data. + +If a hardware configuration ever changes in a backwards incompatible way +(changing the semantics and/or layout of the raw counters) then it must be +given a new GUID. + +mdapi-xml-convert.py will match metric sets with a GUID from this file based on +an md5 hash of the hardware register configuration and skip a metric set with a +warning if no GUID could be found. + +All new metric sets need to be allocated a GUID here before +mdapi-xml-convert.py will output anything for that +metric set. This ensures we don't automatically import new metric sets without +some explicit review that that's appropriate. + +A failure to find a GUID for an older metric set most likely implies that the +register configuration was changed. It's possible that the change is benign +(e.g. a comment change) and in that case the mdapi_config_hash for the +corresponding metric set below can be updated. + +The update-guids.py script is the recommended way of managing updates to this +file by generate a temporary file with proposed updates that you can compare +with the current guids.xml. + + +# update-guids.xml + +update-guids.py can help with: + +* Recognising new metrics from VPG's MDAPI XML files + + *(NOTE: new guids.xml entries will initially be missing the + config_hash=MD5_HASH attribute until mdapi-xml-convert.py is used to generate + a corresponding oa-*.xml config description)* + +* Adding a config_hash=MD5_HASH attribute to recently added guids.xml entries + after mdapi-xml-convert.py has been run. + +* Allocating a GUID for a custom metric that doesn't have a counterpart in + VPG's MDAPI XML files. + + For this case you can add a stub entry with only a name like `<guid + name="Foo">` to guids.xml and then running update-guids.py will output a + corresponding line with the addition of an id=UUID attribute. + + +# How to sync the oa-\*.xml files with latest internal MDAPI XML files + +1. E.g. copy a new `MetricsXML_BDW.xml` to `mdapi/MetricsXML_BDW.xml` + +*Note: that the `mdapi-xml-convert.py` script will only convert configs that +have a corresponding GUID entry within `guids.xml`. This check helps avoid +unintentionally publishing early, work-in-progress/pre-production configs.* + +The `guids.xml` registry maps each, complex OA unit register configuration to a +unique ID that userspace can recognise and trust the semantics of raw counters +read using that configuration. (Just for reference, this is particularly +valuable for tools that capture raw metrics for later, offline processing since +the IDs effectively provide a compressed description of how to interpret the +data by providing an index into a database of high-level counter descriptions.) + +The registry associates each ID with a hash of the HW register config as found in +MDAPI XML files ('mdapi_config_hash') and also with a hash of the HW config as +found in oa-\*.xml files ('config_hash'). The hashes used for lookups in the +registry also help detect when the register config for a pre-existing metric set +is updated. Note: these hashes are only for the low-level hardware configuration +so updates to counter descriptions used by fronted UIs won't affect indexing +here. + +There is a chicken and egg situation when updating or adding new entries to +guids.xml since we can't hash the configs in oa-\*.xml until successfully running +mdapi-xml-convert.py which depends on a guids.xml registry entry first. The +update-guids.xml script will output registry entries without an oa-\*.xml config +hash if not available and can be re-run after mdapi-xml-convert.py to add the +missing hashes. + +2. Now run: +``` +./update-guids.py --guids=guids.xml mdapi/MetricsXML_BDW.xml > guids.xml2 +``` +*(note the script expects to find oa-\*.xml files in the current directory)* + +Diff `guids.xml` and `guilds.xml2` (easiest with a side-by-side diff editor) and +review the registry changes. *Note: many lines will have a warning like `"Not +found in MDAPI XML file[s]..."` if `update-guids.xml` wasn't given all known +MDAPI XML files but in this case they can be ignored for all non-BDW configs.* + +*Note: for any config that is already supported upstream in the xe_oa driver +we need to be careful if the hash for a metric set changes in case the semantics +for any raw counters were changed. The semantics of raw counters associated with +a given GUID form part of the drm xe_oa uapi contract and must remain +backwards compatible.* + +If the diff shows any `mdapi_config_hash` changes for pre-existing (especially +upstream) configs you should review the MDAPI XML changes for the metric set and +verify the change just relates to a bug fix. If more substantial changes were +made which could mean we need to treat it as a new config. Handling the later +case is left as an exercise to the reader, since it hasn't happened so far :-D. +Assuming all the changes and new entries look good they can be copied into +`guids.xml`, removing any trailing comment left by `update-guids.py`. + +3. Now run mdapi-xml-convert.py: +``` +./mdapi-xml-convert.py --guids=guids.xml mdapi/MetricsXML_BDW.xml > oa-bdw.xml +``` + +4. We can now update new entries in guids.xml with a 'config_hash': +``` +./update-guids.py --guids=guids.xml mdapi/MetricsXML_BDW.xml > guids.xml2 +``` +*(and again diff, check the changes and copy across)*
lib/xe/oa-configs/codegen.py new file +444 -0
444 lines not on this page · raw message
lib/xe/oa-configs/guids.xml new file +2749 -0
2,441 lines not on this page · raw message
lib/xe/oa-configs/mdapi-xml-convert.py new file +1196 -0
1,196 lines not on this page · raw message
lib/xe/oa-configs/oa-equations-codegen.py new file +261 -0
261 lines not on this page · raw message
lib/xe/oa-configs/oa-metricset-codegen.py new file +257 -0
257 lines not on this page · raw message
lib/xe/oa-configs/oa-registers-codegen.py new file +118 -0
118 lines not on this page · raw message
lib/xe/oa-configs/oa_guid_registry.py new file +117 -0
117 lines not on this page · raw message
lib/xe/oa-configs/update-guids.py new file +222 -0
222 lines not on this page · raw message
The rest of this diff is not shown. View the raw message.
Signature · 1 line
2.41.0
Part of the diff is not shown here. View the raw message.
PATCH 2/18 lib/xe/oa: Add LNL metric guids
In reply to Ashutosh Dixit$ update-guids.py --guids=guids.xml MetricsXML_P31.xml > guids.xml2
And copying the new guids to guids.xml (as per the README).
Note: as mentioned in update-guids.py and README.md, in order to generate config_hash in guids.xml, mdapi-convert-xml.py must first be run (as outline in the next commit) and then update-guids.py command above should be repeated.
Note: LNL media guids had issues after running mdapi-convert-xml.py so have not been added, revisit this.
--- lib/xe/oa-configs/guids.xml | 14 ++++++++++++++ lib/xe/oa-configs/oa_guid_registry.py | 16 +++++++++++++++- lib/xe/oa-configs/update-guids.py | 1 + 3 files changed, 30 insertions(+), 1 deletion(-)
lib/xe/oa-configs/guids.xml +14 -0
@@ -2746,4 +2746,18 @@ <guid config_hash="482cebdeeb78bb56e67b42d24644c8db" mdapi_config_hash="11e33e01a371bf15e59b9786f4fe1050" id="4d9b2817-19b9-4461-a54b-dc9fdb7d69cb" chipset="mtlgt3" name="MediaSet1" /> <guid config_hash="5bfab887a79b99cf34a606dcca5f4f8a" mdapi_config_hash="1c238e39e5c73eefd870cef2fb7a18af" id="651a6694-0ac1-4bd5-a4a1-94bcf479d4f4" chipset="mtlgt3" name="MediaSet2" /> <guid config_hash="44bcb438bea8b46b57779739bebfbd69" mdapi_config_hash="fbbd97d1f3b60c17eae57bbd3c42e7fe" id="bac01106-5869-489c-a306-67303a434173" chipset="mtlgt3" name="TestOam" /> + <guid config_hash="208c3c79f1b4db2f86277292566d3f10" mdapi_config_hash="1fbaf5ad061ccf96ff7d4e108cbcf3ae" id="12f20772-0044-44ff-bcc0-d2bc252d140e" chipset="lnl" name="RenderBasic" /> + <guid config_hash="b781b616d31187e7f4982f3a238b0ecd" mdapi_config_hash="35c121ef361006ca639d1e3377b5c6d7" id="72539be3-67d9-4101-b067-766b7811d2bb" chipset="lnl" name="ComputeBasic" /> + <guid config_hash="955739b43332e61729e0a7a032188c02" mdapi_config_hash="d9be0378dd2dd6a5f85f71d156d21b43" id="cae0b2ca-c52a-4111-b8bf-5af070677d57" chipset="lnl" name="DepthProfile" /> + <guid config_hash="55ecb5067bca7d99769a30a14eaceba5" mdapi_config_hash="e444af605670c2230a5796b4ad65b2c0" id="70cc56cf-3a37-4f6c-8911-4455fe441b2a" chipset="lnl" name="DeviceCacheProfile" /> + <guid config_hash="e4a38cb223b20b9c28c5941b2f4979cb" mdapi_config_hash="952b726980ff7e836089c7651070f44f" id="01ffd51d-a6d6-4ff4-93d0-619f04c3a547" chipset="lnl" name="MemoryProfile" /> + <guid config_hash="5d38d137356ca80dca2ea337449675be" mdapi_config_hash="07f0aa1a93f77bc7b6771ba2b68c8325" id="6791470f-f1e6-4f77-9013-bcb0c4be6f57" chipset="lnl" name="RenderPipeProfile" /> + <guid config_hash="72e3ab9e7d727f97c5ea29bef23be40e" mdapi_config_hash="e6b78fe68f406ee2a0cc25293c60bc62" id="d66ce633-4ae5-4de2-8678-9705adfd45bc" chipset="lnl" name="RTProfile" /> + <guid config_hash="6dc1bd4e8d8b49cc9776701db836ffe9" mdapi_config_hash="cee63b3642783febd3bcbf45629c5e1c" id="e7338926-568c-4424-bc44-f90318a6b472" chipset="lnl" name="VectorEngineProfile" /> + <guid config_hash="b87d3a3dfab7193309deb13aeb595589" mdapi_config_hash="1950d62ef4b729af4494d76ca3ec8101" id="4ccd6535-fb9a-440f-b0f5-882879dc4cb0" chipset="lnl" name="VectorEngineStalls" /> + <guid config_hash="9df6fd1c4fd100b5e50cd97e16bd1cb8" mdapi_config_hash="47c19557fd9591c3fc3955e46db4565c" id="24c33f2c-f8d9-40d9-b2db-64e863b4d625" chipset="lnl" name="TestOa" /> + <guid config_hash="9bfa717fd19e8d5b99062ca0101113dd" mdapi_config_hash="f390ce860d404c3f22f9d1a3159ec58e" id="0af24f48-05bc-467b-b561-33dec252e651" chipset="lnl" name="MetricSet1" /> + <guid config_hash="d635631df97322be3d97b9c4e38152e9" mdapi_config_hash="68c19075e48e5128b456a91cd7b95286" id="914ad901-832a-4ccd-a21e-ddc7690ed213" chipset="lnl" name="MetricSet2" /> + <guid config_hash="d63d186f5b83e91f8c0680847bbf4b24" mdapi_config_hash="d31ec4eaa11c8be0d8d72ecece53434f" id="3b3c9f55-e34b-423d-8c05-7b1af03b6685" chipset="lnl" name="MetricSet3" /> + <guid config_hash="76b7587d047c30ecde26e3f6253b4165" mdapi_config_hash="e39d956ed7b8d0cbe7d354cff234af7d" id="e6f21065-a6f5-4ea4-a545-80d6aebaab56" chipset="lnl" name="MetricSet4" /> </guids>
lib/xe/oa-configs/oa_guid_registry.py +15 -1
@@ -100,10 +100,24 @@ class Registry: 'DG1', 'ACM', 'PVC', - 'MTL', ) + 'MTL', + 'LNL', + ) if name in known_chipsets: return name + internal_chipsets = { + 'P22': 'ATS', + 'P24': 'DG2', + 'P25': 'PVC', + 'P26': 'ADL', + 'P27': 'MTL', + 'P31': 'LNL', + } + + if name in internal_chipsets: + return internal_chipsets[name] + # Unknown HW assert 0
lib/xe/oa-configs/update-guids.py +1 -0
@@ -165,6 +165,7 @@ chipsets = [ 'hsw', 'tglgt1', 'tglgt2', 'rkl', 'dg1', 'adl', 'acmgt1', 'acmgt2', 'acmgt3', 'mtlgt2', 'mtlgt3', + 'lnl', ] for chipset in chipsets:
Signature · 1 line
2.41.0
PATCH 3/18 lib/xe/oa: Add OA LNL metrics (oa_lnl.xml)
In reply to Ashutosh Dixit$ mdapi-xml-convert.py --guids=guids.xml MetricsXML_P31.xml > oa-lnl.xml
Note: The metrics XML's consist of (a) the registers (in this case NOA or mux registers) to be programmed (b) the metrics themselves which specify how the counters in the reports should be combined to construct the metrics. The changes in mdapi-xml-convert.py in this patch completely ignore (b) since they are not relevant for IGT. (a) is needed for IGT LNL OAG uses "576B_PEC64LL" format which is not available on previous platforms so we cannot use metrics for a previous platform such as MTL to generate LNL OAG data.
Note: LNL media guids had issues after running mdapi-convert-xml.py so have not been added, revisit this.
--- lib/xe/oa-configs/mdapi-xml-convert.py | 25 + lib/xe/oa-configs/oa-lnl.xml | 2799 ++++++++++++++++++++++++ 2 files changed, 2824 insertions(+) create mode 100644 lib/xe/oa-configs/oa-lnl.xml
lib/xe/oa-configs/mdapi-xml-convert.py +25 -0
@@ -149,6 +149,21 @@ mtl_chipset_oam_samedia_params = { } } +# FIXME: everything except oa_report_size is incorrect here +xe2_chipset_params = { + 'a_offset': 16, + 'b_offset': 192, + 'c_offset': 224, + 'oa_report_size': 576, + 'config_reg_blacklist': { + 0x2364, # OACTXID + }, + 'register_offsets': { + 0x1b0: 'PERFCNT 0', + 0x1b8: 'PERFCNT 1', + } +} + hsw_chipset_oa_formats = { '256B_GENERIC_NOA16': hsw_chipset_params, }
@@ -167,6 +182,10 @@ mtl_chipset_oa_formats = { '128B_MPEC8_NOA16': mtl_chipset_oam_samedia_params, } +xe2_chipset_oa_formats = { + '256B_GENERIC_NOA16': xe2_chipset_params, +} + chipsets = { 'HSW': hsw_chipset_oa_formats, 'BDW': gen8_11_chipset_oa_formats,
@@ -185,10 +204,13 @@ chipsets = { 'ADL': gen8_11_chipset_oa_formats, 'ACM': xehpsdv_chipset_oa_formats, 'MTL': mtl_chipset_oa_formats, + 'LNL': xe2_chipset_oa_formats, } xehp_plus = ( 'ACM', 'MTL' ) +xe2_plus = ( 'LNL' ) + register_types = { 'OA', 'NOA', 'FLEX', 'PM' } default_set_blacklist = { "RenderDX1x", # TODO: rename to something non 'DX'
@@ -352,6 +374,9 @@ def read_token_to_rpn_read_oag(chipset, token, raw_offsets, oa_format): return "C " + str(idx - 48) + " READ" else: return "{0} READ".format(read_value(chipset, offset, oa_format)) + elif chipset in xe2_plus: + # FIXME: skip all metrics to retain just the registers + return "GPU_TIME 0 READ" else: # For Gen8+ the array of accumulated counters is # assumed to start with a GPU_TIME then GPU_CLOCK,
lib/xe/oa-configs/oa-lnl.xml new file +2799 -0
2,799 lines not on this page · raw message
Signature · 1 line
2.41.0
PATCH 4/18 lib/xe/oa: Generate LNL metrics/registers files
In reply to Ashutosh DixitAdd LNL OA formats and generate metrics/registers .c and .h files with the XE_OA_FORMAT_PEC64u64 format (called 576B_PEC64LL) used in MetricsXML_P31.xml.
--- lib/xe/oa-configs/oa-metricset-codegen.py | 53 +++++++++++++++++++++-- 1 file changed, 50 insertions(+), 3 deletions(-)
lib/xe/oa-configs/oa-metricset-codegen.py +50 -3
@@ -81,6 +81,40 @@ def generate_metric_sets(args, gen): #include <stdbool.h> #include <assert.h> +enum xe_oa_format_name { + XE_OA_FORMAT_C4_B8 = 1, + + /* Gen8+ */ + XE_OA_FORMAT_A12, + XE_OA_FORMAT_A12_B8_C8, + XE_OA_FORMAT_A32u40_A4u32_B8_C8, + + /* DG2 */ + XE_OAR_FORMAT_A32u40_A4u32_B8_C8, + XE_OA_FORMAT_A24u40_A14u32_B8_C8, + + /* DG2/MTL OAC */ + XE_OAC_FORMAT_A24u64_B8_C8, + XE_OAC_FORMAT_A22u32_R2u32_B8_C8, + + /* MTL OAM */ + XE_OAM_FORMAT_MPEC8u64_B8_C8, + XE_OAM_FORMAT_MPEC8u32_B8_C8, + + /* Xe2+ */ + XE_OA_FORMAT_PEC64u64, + XE_OA_FORMAT_PEC64u64_B8_C8, + XE_OA_FORMAT_PEC64u32, + XE_OA_FORMAT_PEC32u64_G1, + XE_OA_FORMAT_PEC32u32_G1, + XE_OA_FORMAT_PEC32u64_G2, + XE_OA_FORMAT_PEC32u32_G2, + XE_OA_FORMAT_PEC36u64_G1_32_G2_4, + XE_OA_FORMAT_PEC36u64_G1_4_G2_32, + + XE_OA_FORMAT_MAX, +}; + """)) c("#include \"{0}\"".format(os.path.basename(args.header)))
@@ -116,7 +150,7 @@ def generate_metric_sets(args, gen): if gen.chipset.startswith("acm") or gen.chipset.startswith("mtl"): if set.oa_format == "128B_MPEC8_NOA16": c(textwrap.dedent("""\ - metric_set->perf_oa_format = I915_OAM_FORMAT_MPEC8u32_B8_C8; + metric_set->perf_oa_format = XE_OAM_FORMAT_MPEC8u32_B8_C8; metric_set->perf_raw_size = 128; metric_set->gpu_time_offset = 0;
@@ -128,7 +162,7 @@ def generate_metric_sets(args, gen): """)) else: c(textwrap.dedent("""\ - metric_set->perf_oa_format = I915_OA_FORMAT_A24u40_A14u32_B8_C8; + metric_set->perf_oa_format = XE_OA_FORMAT_A24u40_A14u32_B8_C8; metric_set->perf_raw_size = 256; metric_set->gpu_time_offset = 0;
@@ -138,9 +172,22 @@ def generate_metric_sets(args, gen): metric_set->c_offset = metric_set->b_offset + 8; metric_set->perfcnt_offset = metric_set->c_offset + 8; """)) + elif gen.chipset == "lnl": + c(textwrap.dedent("""\ + metric_set->perf_oa_format = XE_OA_FORMAT_PEC64u64; + + metric_set->perf_raw_size = 576; + metric_set->gpu_time_offset = 0; + metric_set->gpu_clock_offset = 1; + metric_set->a_offset = 2; + // metric_set->b_offset = metric_set->a_offset + 38; + // metric_set->c_offset = metric_set->b_offset + 8; + metric_set->perfcnt_offset = metric_set->a_offset + 64; + + """)) else: c(textwrap.dedent("""\ - metric_set->perf_oa_format = I915_OA_FORMAT_A32u40_A4u32_B8_C8; + metric_set->perf_oa_format = XE_OA_FORMAT_A32u40_A4u32_B8_C8; metric_set->perf_raw_size = 256; metric_set->gpu_time_offset = 0;
Signature · 1 line
2.41.0
PATCH 5/18 drm-uapi/xe: Sync with Perf/OA changes
In reply to Ashutosh DixitInclude Perf/OA changes from the kernel in include/drm-uapi/xe_drm.h.
--- include/drm-uapi/xe_drm.h | 321 +++++++++++++++++++++++++++++++++++++- 1 file changed, 320 insertions(+), 1 deletion(-)
include/drm-uapi/xe_drm.h +320 -1
@@ -100,6 +100,8 @@ extern "C" {#define DRM_XE_EXEC_QUEUE_GET_PROPERTY 0x08#define DRM_XE_EXEC 0x09#define DRM_XE_WAIT_USER_FENCE 0x0a +#define DRM_XE_PERF 0x0e +/* Must be kept compact -- no holes */ #define DRM_IOCTL_XE_DEVICE_QUERY DRM_IOWR(DRM_COMMAND_BASE + DRM_XE_DEVICE_QUERY, struct drm_xe_device_query)
@@ -113,6 +115,7 @@ extern "C" {#define DRM_IOCTL_XE_EXEC_QUEUE_GET_PROPERTY DRM_IOWR(DRM_COMMAND_BASE + DRM_XE_EXEC_QUEUE_GET_PROPERTY, struct drm_xe_exec_queue_get_property)#define DRM_IOCTL_XE_EXEC DRM_IOW(DRM_COMMAND_BASE + DRM_XE_EXEC, struct drm_xe_exec)#define DRM_IOCTL_XE_WAIT_USER_FENCE DRM_IOWR(DRM_COMMAND_BASE + DRM_XE_WAIT_USER_FENCE, struct drm_xe_wait_user_fence) +#define DRM_IOCTL_XE_PERF DRM_IOW(DRM_COMMAND_BASE + DRM_XE_PERF, struct drm_xe_perf_param)/** * DOC: Xe IOCTL Extensions
@@ -574,6 +577,36 @@ struct drm_xe_query_engine_cycles {__u64 cpu_delta; }; +/** +* struct drm_xe_query_uc_fw_version - query a micro-controller firmware version +* +* Given a uc_type this will return the branch, major, minor and patch version +* of the micro-controller firmware. +*/ +struct drm_xe_query_uc_fw_version { +/** @uc_type: The micro-controller type to query firmware version */ +#define XE_QUERY_UC_TYPE_GUC_SUBMISSION 0 +__u16 uc_type; + +/** @pad: MBZ */ +__u16 pad; + +/** @branch_ver: branch uc fw version */ +__u32 branch_ver; +/** @major_ver: major uc fw version */ +__u32 major_ver; +/** @minor_ver: minor uc fw version */ +__u32 minor_ver; +/** @patch_ver: patch uc fw version */ +__u32 patch_ver; + +/** @pad2: MBZ */ +__u32 pad2; + +/** @reserved: Reserved */ +__u64 reserved; +}; +/** * struct drm_xe_device_query - Input of &DRM_IOCTL_XE_DEVICE_QUERY - main * structure to query device information
@@ -643,6 +676,8 @@ struct drm_xe_device_query {#define DRM_XE_DEVICE_QUERY_HWCONFIG 4#define DRM_XE_DEVICE_QUERY_GT_TOPOLOGY 5#define DRM_XE_DEVICE_QUERY_ENGINE_CYCLES 6 +#define DRM_XE_DEVICE_QUERY_UC_FW_VERSION 7 +#define DRM_XE_DEVICE_QUERY_OA_UNITS 8/** @query: The type of data to query */ __u32 query;
@@ -832,7 +867,6 @@ struct drm_xe_vm_destroy {* * and the @flags can be: * - %DRM_XE_VM_BIND_FLAG_READONLY -* - %DRM_XE_VM_BIND_FLAG_ASYNC * - %DRM_XE_VM_BIND_FLAG_IMMEDIATE - Valid on a faulting VM only, do the * MAP operation immediately rather than deferring the MAP to the page * fault handler.
@@ -1340,6 +1374,291 @@ struct drm_xe_wait_user_fence {__u64 reserved[2 ]; }; +/** enum drm_xe_perf_type - Perf stream types */ +enum drm_xe_perf_type { +DRM_XE_PERF_TYPE_OA , +DRM_XE_PERF_TYPE_MAX , +}; + +/** +* enum drm_xe_perf_op - Perf stream ops +*/ +enum drm_xe_perf_op { +/** @DRM_XE_PERF_OP_STREAM_OPEN: Open a perf counter stream */ +DRM_XE_PERF_OP_STREAM_OPEN , + +/** @DRM_XE_PERF_OP_ADD_CONFIG: Add perf stream config */ +DRM_XE_PERF_OP_ADD_CONFIG , + +/** @DRM_XE_PERF_OP_REMOVE_CONFIG: Remove perf stream config */ +DRM_XE_PERF_OP_REMOVE_CONFIG , +}; + +/** +* struct drm_xe_perf_param - Perf layer param +* +* The perf layer enables multiplexing perf counter streams of multiple +* types. The actual params for a particular stream operation are supplied +* via the @param pointer (use __copy_from_user to get these params). +*/ +struct drm_xe_perf_param { +/** @extensions: Pointer to the first extension struct, if any */ +__u64 extensions; +/** @perf_type: Perf stream type, of enum @drm_xe_perf_type */ +__u64 perf_type; +/** @perf_op: Perf op, of enum @drm_xe_perf_op */ +__u64 perf_op; +/** @param: Pointer to actual stream params */ +__u64 param; +}; + +/** +* enum drm_xe_perf_ioctls - Perf fd ioctl's +* +* Information exchanged between userspace and kernel for perf fd ioctl's +* is stream type specific +*/ +enum drm_xe_perf_ioctls { +/** @DRM_XE_PERF_IOCTL_ENABLE: Enable data capture for a stream */ +DRM_XE_PERF_IOCTL_ENABLE =_IO ('i' ,0x0 ), + +/** @DRM_XE_PERF_IOCTL_DISABLE: Disable data capture for a stream */ +DRM_XE_PERF_IOCTL_DISABLE =_IO ('i' ,0x1 ), + +/** @DRM_XE_PERF_IOCTL_CONFIG: Change stream configuration */ +DRM_XE_PERF_IOCTL_CONFIG =_IO ('i' ,0x2 ), + +/** @DRM_XE_PERF_IOCTL_STATUS: Return stream status */ +DRM_XE_PERF_IOCTL_STATUS =_IO ('i' ,0x3 ), + +/** @DRM_XE_PERF_IOCTL_INFO: Return stream info */ +DRM_XE_PERF_IOCTL_INFO =_IO ('i' ,0x4 ), +}; + +/** enum drm_xe_oa_unit_type - OA unit types */ +enum drm_xe_oa_unit_type { +/** +* @DRM_XE_OA_UNIT_TYPE_OAG: OAG OA unit. OAR/OAC are considered +* sub-types of OAG. For OAR/OAC, use OAG. +*/ +DRM_XE_OA_UNIT_TYPE_OAG , + +/** @DRM_XE_OA_UNIT_TYPE_OAM: OAM OA unit */ +DRM_XE_OA_UNIT_TYPE_OAM , +}; + +/** +* struct drm_xe_oa_unit - describe OA unit +*/ +struct drm_xe_oa_unit { +/** @oa_unit_id: OA unit ID */ +__u32 oa_unit_id; + +/** @oa_unit_type: OA unit type of @drm_xe_oa_unit_type */ +__u32 oa_unit_type; + +/** @capabilities: OA capabilities bit-mask */ +__u64 capabilities; +#define DRM_XE_OA_CAPS_BASE (1 << 0) + +/** @oa_timestamp_freq: OA timestamp freq */ +__u64 oa_timestamp_freq; + +/** @reserved: MBZ */ +__u64 reserved[4 ]; + +/** @num_engines: number of engines in @eci array */ +__u64 num_engines; + +/** @eci: engines attached to this OA unit */ +struct drm_xe_engine_class_instance eci[]; +}; + +/** +* struct drm_xe_query_oa_units - describe OA units +* +* If a query is made with a struct drm_xe_device_query where .query +* is equal to DRM_XE_DEVICE_QUERY_OA_UNITS, then the reply uses struct +* drm_xe_query_oa_units in .data. +* +* OA unit properties for all OA units can be accessed using a code block +* such as the one below: +* +* .. code-block:: C +* +* struct drm_xe_query_oa_units *qoa; +* struct drm_xe_oa_unit *oau; +* u8 *poau; +* +* // malloc qoa and issue DRM_XE_DEVICE_QUERY_OA_UNITS. Then: +* poau = (u8 *)&qoa->oa_units[0]; +* for (int i = 0; i < qoa->num_oa_units; i++) { +* oau = (struct drm_xe_oa_unit *)poau; +* // Access 'struct drm_xe_oa_unit' fields here +* poau += sizeof(*oau) + oau->num_engines * sizeof(oau->eci[0]); +* } +*/ +struct drm_xe_query_oa_units { +/** @num_oa_units: number of OA units returned in oau[] */ +__u32 num_oa_units; +/** @pad: MBZ */ +__u32 pad; +/** @oa_units: OA units returned for this device */ +struct drm_xe_oa_unit oa_units[]; +}; + +/** enum drm_xe_oa_format_type - OA format types */ +enum drm_xe_oa_format_type { +DRM_XE_OA_FMT_TYPE_OAG , +DRM_XE_OA_FMT_TYPE_OAR , +DRM_XE_OA_FMT_TYPE_OAM , +DRM_XE_OA_FMT_TYPE_OAC , +DRM_XE_OA_FMT_TYPE_OAM_MPEC , +DRM_XE_OA_FMT_TYPE_PEC , +}; + +/** +* enum drm_xe_oa_property_id - OA stream property id's +* +* Stream params are specified as a chain of @drm_xe_ext_set_property +* struct's, with @property values from enum @drm_xe_oa_property_id and +* @drm_xe_user_extension base.name set to @DRM_XE_OA_EXTENSION_SET_PROPERTY. +* @param field in struct @drm_xe_perf_param points to the first +* @drm_xe_ext_set_property struct. +*/ +enum drm_xe_oa_property_id { +#define DRM_XE_OA_EXTENSION_SET_PROPERTY 0 +/** +* @DRM_XE_OA_PROPERTY_OA_UNIT_ID: ID of the OA unit on which to open +* the OA stream, see @oa_unit_id in 'struct +* drm_xe_query_oa_units'. Defaults to 0 if not provided. + */ +DRM_XE_OA_PROPERTY_OA_UNIT_ID =1 , + +/** +* @DRM_XE_OA_PROPERTY_SAMPLE_OA: A value of 1 requests inclusion of raw +* OA unit reports or stream samples in a global buffer attached to an +* OA unit. +*/ +DRM_XE_OA_PROPERTY_SAMPLE_OA , + +/** +* @DRM_XE_OA_PROPERTY_OA_METRIC_SET: OA metrics defining contents of OA +* reports, previously added via @DRM_XE_PERF_OP_ADD_CONFIG. +*/ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , + +/** @DRM_XE_OA_PROPERTY_OA_FORMAT: Perf counter report format */ +DRM_XE_OA_PROPERTY_OA_FORMAT , +/* +* OA_FORMAT's are specified the same way as in PRM/Bspec 52198/60942, +* in terms of the following quantities: a. enum @drm_xe_oa_format_type +* b. Counter select c. Counter size and d. BC report. Also refer to the +* oa_formats array in drivers/gpu/drm/xe/xe_oa.c. +*/ +#define DRM_XE_OA_FORMAT_MASK_FMT_TYPE (0xff << 0) +#define DRM_XE_OA_FORMAT_MASK_COUNTER_SEL (0xff << 8) +#define DRM_XE_OA_FORMAT_MASK_COUNTER_SIZE (0xff << 16) +#define DRM_XE_OA_FORMAT_MASK_BC_REPORT (0xff << 24) + +/** +* @DRM_XE_OA_PROPERTY_OA_EXPONENT: Requests periodic OA unit sampling +* with sampling frequency proportional to 2^(period_exponent + 1) +*/ +DRM_XE_OA_PROPERTY_OA_EXPONENT , + +/** +* @DRM_XE_OA_PROPERTY_OA_DISABLED: A value of 1 will open the OA +* stream in a DISABLED state (see @DRM_XE_PERF_IOCTL_ENABLE). +*/ +DRM_XE_OA_PROPERTY_OA_DISABLED , + +/** +* @DRM_XE_OA_PROPERTY_EXEC_QUEUE_ID: Open the stream for a specific +* @exec_queue_id. Perf queries can be executed on this exec queue. +*/ +DRM_XE_OA_PROPERTY_EXEC_QUEUE_ID , + +/** +* @DRM_XE_OA_PROPERTY_OA_ENGINE_INSTANCE: Optional engine instance to +* pass along with @DRM_XE_OA_PROPERTY_EXEC_QUEUE_ID or will default to 0. +*/ +DRM_XE_OA_PROPERTY_OA_ENGINE_INSTANCE , + +/** @DRM_XE_OA_PROPERTY_MAX: non-ABI */ +DRM_XE_OA_PROPERTY_MAX +}; + +/** +* struct drm_xe_oa_config - OA metric configuration +* +* Multiple OA configs can be added using @DRM_XE_PERF_OP_ADD_CONFIG. A +* particular config can be specified when opening an OA stream using +* @DRM_XE_OA_PROPERTY_OA_METRIC_SET property. +*/ +struct drm_xe_oa_config { +/** @extensions: Pointer to the first extension struct, if any */ +__u64 extensions; + +/** @uuid: String formatted like "%\08x-%\04x-%\04x-%\04x-%\012x" */ +char uuid[36 ]; + +/** @n_regs: Number of regs in @regs_ptr */ +__u32 n_regs; + +/** +* @regs_ptr: Pointer to (register address, value) pairs for OA config +* registers. Expected length of buffer is: (2 * sizeof(u32) * @n_regs). +*/ +__u64 regs_ptr; +}; + +/** +* struct drm_xe_oa_stream_config - OA stream re-configuration with +* @DRM_XE_PERF_IOCTL_CONFIG perf fd ioctl +*/ +struct drm_xe_oa_stream_config { +/** @extensions: Pointer to the first extension struct, if any */ +__u64 extensions; + +/** +* @metric_set: metric set id, previously added using +* @DRM_XE_PERF_OP_ADD_CONFIG +*/ +__u64 metric_set; + +/** @reserved: reserved for future use */ +__u64 reserved[3 ]; +}; + +/** +* struct drm_xe_oa_stream_status - OA stream status returned from +* @DRM_XE_PERF_IOCTL_STATUS perf fd ioctl +*/ +struct drm_xe_oa_stream_status { +/** @oa_status: OA status register as specified in PRM/Bspec 46717/61226 */ +__u64 oa_status; +#define DRM_XE_OASTATUS_MMIO_TRG_Q_FULL (1 << 6) +#define DRM_XE_OASTATUS_COUNTER_OVERFLOW (1 << 2) +#define DRM_XE_OASTATUS_BUFFER_OVERFLOW (1 << 1) +#define DRM_XE_OASTATUS_REPORT_LOST (1 << 0) + +/** @reserved: reserved for future use */ +__u64 reserved[3 ]; +}; + +/** +* struct drm_xe_oa_stream_info - OA stream info returned from +* @DRM_XE_PERF_IOCTL_INFO perf fd ioctl +*/ +struct drm_xe_oa_stream_info { +/** @oa_buf_size: OA buffer size */ +__u64 oa_buf_size; + +/** @reserved: reserved for future use */ +__u64 reserved[3 ]; +}; +#if defined(__cplusplus) }#endif
Signature · 1 line
2.41.0
PATCH 6/18 lib/xe: Complete xe_oa lib changes
In reply to Ashutosh DixitAdd various functionality in lib/xe for OA. This includes:
* Support for OA metrics generation
* intel_perf_for_devinfo and intel_perf_for_fd support
* intel_perf_load_perf_configs
* xe_perf_ioctl
* drm_xe_query_oa_units
--- lib/intel_device_info.c | 1 + lib/meson.build | 74 +++++ lib/xe/xe_oa.c | 699 ++++++++++++++++++++++++++++++++++++++++ lib/xe/xe_oa.h | 328 +++++++++++++++++++ lib/xe/xe_query.c | 38 +++ lib/xe/xe_query.h | 5 + 6 files changed, 1145 insertions(+) create mode 100644 lib/xe/xe_oa.c create mode 100644 lib/xe/xe_oa.h
lib/intel_device_info.c +1 -0
@@ -510,6 +510,7 @@ static const struct intel_device_info intel_lunarlake_info = { .display_ver = 20, .has_4tile = true, .has_flatccs = true, + .has_oam = true, .is_lunarlake = true, .codename = "lunarlake", .cmds_info = &xe2_cmds_info,
lib/meson.build +74 -0
86 lines not on this page · raw message
lib/xe/xe_oa.c new file +699 -0
699 lines not on this page · raw message
lib/xe/xe_oa.h new file +328 -0
328 lines not on this page · raw message
lib/xe/xe_query.c +38 -0
@@ -114,6 +114,27 @@ static struct drm_xe_query_mem_regions *xe_query_mem_regions_new(int fd)return mem_regions; } +static struct drm_xe_query_oa_units *xe_query_oa_units_new (int fd) +{ +struct drm_xe_query_oa_units *oa_units; +struct drm_xe_device_query query = { + .extensions =0 , + .query =DRM_XE_DEVICE_QUERY_OA_UNITS , + .size =0 , + .data =0 , + }; + +igt_assert_eq (igt_ioctl (fd,DRM_IOCTL_XE_DEVICE_QUERY , &query),0 ); + + oa_units =malloc (query.size); +igt_assert (oa_units); + + query.data =to_user_pointer (oa_units); +igt_assert_eq (igt_ioctl (fd,DRM_IOCTL_XE_DEVICE_QUERY , &query),0 ); + +return oa_units; +} +static uint64_t native_region_for_gt (const struct drm_xe_query_gt_list *gt_list,int gt) {uint64_t region;
@@ -251,6 +272,7 @@ struct xe_device *xe_device_get(int fd) xe_dev->memory_regions =__memory_regions (xe_dev->gt_list); xe_dev->engines =xe_query_engines (fd); xe_dev->mem_regions =xe_query_mem_regions_new (fd); + xe_dev->oa_units =xe_query_oa_units_new (fd); xe_dev->vram_size =calloc (xe_dev->gt_list->num_gt,sizeof (*xe_dev->vram_size)); xe_dev->visible_vram_size =calloc (xe_dev->gt_list->num_gt,sizeof (*xe_dev->visible_vram_size));for (int gt =0 ; gt < xe_dev->gt_list->num_gt; gt++) {
@@ -524,6 +546,22 @@ uint32_t xe_min_page_size(int fd, uint64_t region)*/ xe_dev_FN(xe_config, config, struct drm_xe_query_config *); +/** +* xe_gt_list: +* @fd: xe device fd +* +* Returns query gts of xe device @fd. +*/ +xe_dev_FN(xe_gt_list, gt_list, struct drm_xe_query_gt_list *); + +/** +* xe_oa_units: +* @fd: xe device fd +* +* Returns query gts of xe device @fd. +*/ +xe_dev_FN(xe_oa_units, oa_units, struct drm_xe_query_oa_units *); +/** * xe_number_engine: * @fd: xe device fd
lib/xe/xe_query.h +5 -0
@@ -38,6 +38,9 @@ struct xe_device {/** @mem_regions: regions memory information and usage */ struct drm_xe_query_mem_regions *mem_regions; +/** @oa_units: information about OA units */ +struct drm_xe_query_oa_units *oa_units; +/** @vram_size: array of vram sizes for all gt_list */ uint64_t *vram_size;
@@ -85,6 +88,8 @@ const char *xe_region_name(uint64_t region);uint16_t xe_region_class (int fd,uint64_t region);uint32_t xe_min_page_size (int fd,uint64_t region);struct drm_xe_query_config *xe_config (int fd); +struct drm_xe_query_gt_list *xe_gt_list (int fd); +struct drm_xe_query_oa_units *xe_oa_units (int fd);unsigned int xe_number_engines (int fd);bool xe_has_vram (int fd);uint64_t xe_vram_size (int fd,int gt);
Signature · 1 line
2.41.0
PATCH 7/18 tests/intel/xe_query: Add OA units query test
In reply to Ashutosh DixitAdd test to exercise OA units query.
--- tests/intel/xe_query.c | 53 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 53 insertions(+)
tests/intel/xe_query.c +53 -0
@@ -730,6 +730,58 @@ static void test_engine_cycles_invalid(int fd)query_engine_cycles (fd, &ts); } +/** +* SUBTEST: query-oa-units +* Description: Display fields for OA unit query +* +* SUBTEST: multigpu-query-oa-units +* Description: Display fields for OA unit query for all GPU devices +* Sub-category: MultiGPU +*/ +static void test_query_oa_units (int fd) +{ +struct drm_xe_query_oa_units *qoa; +struct drm_xe_device_query query = { + .extensions =0 , + .query =DRM_XE_DEVICE_QUERY_OA_UNITS , + .size =0 , + .data =0 , + }; +struct drm_xe_oa_unit *oau; +int i, j; +u8 *poau; + +igt_assert_eq (igt_ioctl (fd,DRM_IOCTL_XE_DEVICE_QUERY , &query),0 ); + + qoa =malloc (query.size); +igt_assert (qoa); + + query.data =to_user_pointer (qoa); +igt_assert_eq (igt_ioctl (fd,DRM_IOCTL_XE_DEVICE_QUERY , &query),0 ); + +igt_info ("num_oa_units %d\n" , qoa->num_oa_units); + + poau = (u8 *)&qoa->oa_units[0 ]; +for (i =0 ; i < qoa->num_oa_units; i++) { + oau = (struct drm_xe_oa_unit *)poau; + +igt_info ("-------------------------------\n" ); +igt_info ("oa_unit %d\n" , i); +igt_info ("-------------------------------\n" ); +igt_info ("oa_unit_id %d\n" , oau->oa_unit_id); +igt_info ("oa_unit_type %d\n" , oau->oa_unit_type); +igt_info ("capabilities %#llx\n" , oau->capabilities); +igt_info ("oa_timestamp_freq %lld\n" , oau->oa_timestamp_freq); +igt_info ("num_engines %lld\n" , oau->num_engines); +igt_info ("Engines:" ); +for (j =0 ; j < oau->num_engines; j++) +igt_info (" (%d, %d)" , oau->eci[j].engine_class, + oau->eci[j].engine_instance); +igt_info ("\n" ); + poau +=sizeof (*oau) + j *sizeof (oau->eci[0 ]); + } +} +igt_main {const struct {
@@ -743,6 +795,7 @@ igt_main { "query-hwconfig", test_query_hwconfig }, { "query-topology", test_query_gt_topology }, { "query-cs-cycles", test_query_engine_cycles }, + { "query-oa-units", test_query_oa_units }, { "query-invalid-cs-cycles", test_engine_cycles_invalid }, { "query-invalid-query", test_query_invalid_query }, { "query-invalid-size", test_query_invalid_size },
Signature · 1 line
2.41.0
PATCH 8/18 tests/intel/xe_oa: Add first tests
In reply to Ashutosh DixitAdd "xe-ref-count" and "sysctl-defaults" subtests.
--- tests/intel/xe_oa.c | 842 ++++++++++++++++++++++++++++++++++++++++++++ tests/meson.build | 2 + 2 files changed, 844 insertions(+) create mode 100644 tests/intel/xe_oa.c
tests/intel/xe_oa.c new file +842 -0
@@ -0,0 +1,842 @@ +/* SPDX-License-Identifier: MIT */ +/* +* Copyright © 2024 Intel Corporation +*/ + +#include <stdlib.h> +#include <stdio.h> +#include <string.h> +#include <fcntl.h> +#include <inttypes.h> +#include <errno.h> +#include <signal.h> +#include <sys/stat.h> +#include <sys/time.h> +#include <sys/times.h> +#include <sys/types.h> +#include <dirent.h> +#include <time.h> +#include <poll.h> +#include <math.h> + +#include "drm.h" +#include "igt.h" +#include "igt_device.h" +#include "igt_sysfs.h" +#include "xe/xe_ioctl.h" +#include "xe/xe_query.h" +#include "xe/xe_oa.h" + +/** +* TEST: Test the xe oa metrics streaming interface +* Category: Perf +* Sub-category: OA +* Functionality: OA +* Test category: functionality test +*/ + +#define OASTATUS_MMIO_TRG_Q_FULL REG_BIT(6) +#define OASTATUS_COUNTER_OVERFLOW REG_BIT(2) +#define OASTATUS_BUFFER_OVERFLOW REG_BIT(1) +#define OASTATUS_REPORT_LOST REG_BIT(0) + +#define OA_MI_REPORT_PERF_COUNT ((0x28 << 23) | (4 - 2)) + +#define OAREPORT_REASON_MASK 0x3f +#define OAREPORT_REASON_SHIFT 19 +#define OAREPORT_REASON_TIMER (1<<0) +#define OAREPORT_REASON_INTERNAL (3<<1) +#define OAREPORT_REASON_CTX_SWITCH (1<<3) +#define OAREPORT_REASON_GO (1<<4) +#define OAREPORT_REASON_CLK_RATIO (1<<5) + +#define PIPE_CONTROL_GLOBAL_SNAPSHOT_COUNT_RESET (1 << 19) +#define PIPE_CONTROL_SYNC_GFDT (1 << 17) +#define PIPE_CONTROL_NO_WRITE (0 << 14) +#define PIPE_CONTROL_WRITE_IMMEDIATE (1 << 14) +#define PIPE_CONTROL_WRITE_DEPTH_COUNT (2 << 14) +#define PIPE_CONTROL_RENDER_TARGET_FLUSH (1 << 12) +#define PIPE_CONTROL_INSTRUCTION_INVALIDATE (1 << 11) +#define PIPE_CONTROL_ISP_DIS (1 << 9) +#define PIPE_CONTROL_INTERRUPT_ENABLE (1 << 8) +/* GT */ +#define PIPE_CONTROL_DATA_CACHE_INVALIDATE (1 << 5) +#define PIPE_CONTROL_PPGTT_WRITE (0 << 2) +#define PIPE_CONTROL_GLOBAL_GTT_WRITE (1 << 2) + +#define MAX_OA_BUF_SIZE (16 * 1024 * 1024) +#define OA_BUFFER_SIZE MAX_OA_BUF_SIZE + +#define RING_FORCE_TO_NONPRIV_ADDRESS_MASK 0x03fffffc +/* +* Engine specific registers defined as offsets from engine->mmio_base. For +* these registers, OR bit[0] with 1 so we can add the mmio_base when running +* engine specific test. +*/ +#define MMIO_BASE_OFFSET 0x1 + +#define OAG_OASTATUS (0xdafc) +#define OAG_PERF_COUNTER_B (idx) (0xDA94 + 4 * (idx)) +#define OAG_OATAILPTR (0xdb04) +#define OAG_OATAILPTR_MASK 0xffffffc0 +#define OAG_OABUFFER (0xdb08) + +#define ADD_PROPS (_head, _tail, _key, _value) \ + do { \ + igt_assert((_tail - _head) < (DRM_XE_OA_PROPERTY_MAX * 2)); \ + *_tail++ = DRM_XE_OA_PROPERTY_##_key; \ + *_tail++ = _value; \ + } while (0) + +/* +* If formats are added here remember to change +* lib/xe/oa-configs/oa-metricset-codegen.py +*/ +enum xe_oa_format_name { +XE_OA_FORMAT_C4_B8 =1 , + +/* Gen8+ */ +XE_OA_FORMAT_A12 , +XE_OA_FORMAT_A12_B8_C8 , + XE_OA_FORMAT_A32u40_A4u32_B8_C8, + +/* DG2 */ + XE_OAR_FORMAT_A32u40_A4u32_B8_C8, + XE_OA_FORMAT_A24u40_A14u32_B8_C8, + +/* DG2/MTL OAC */ + XE_OAC_FORMAT_A24u64_B8_C8, + XE_OAC_FORMAT_A22u32_R2u32_B8_C8, + +/* MTL OAM */ + XE_OAM_FORMAT_MPEC8u64_B8_C8, + XE_OAM_FORMAT_MPEC8u32_B8_C8, + +/* Xe2+ */ + XE_OA_FORMAT_PEC64u64, + XE_OA_FORMAT_PEC64u64_B8_C8, + XE_OA_FORMAT_PEC64u32, + XE_OA_FORMAT_PEC32u64_G1, + XE_OA_FORMAT_PEC32u32_G1, + XE_OA_FORMAT_PEC32u64_G2, + XE_OA_FORMAT_PEC32u32_G2, + XE_OA_FORMAT_PEC36u64_G1_32_G2_4, + XE_OA_FORMAT_PEC36u64_G1_4_G2_32, + +XE_OA_FORMAT_MAX , +}; + +struct accumulator { +#define MAX_RAW_OA_COUNTERS 62 +enum xe_oa_format_name format; + +uint64_t deltas[MAX_RAW_OA_COUNTERS ]; +}; + +/* OA unit types */ +enum { +OAG , +OAR , +OAM , + +MAX_OA_TYPE , +}; + +struct oa_format { +const char *name; +size_t size; +int a40_high_off;/* bytes */ +int a40_low_off; +int n_a40; +int a64_off; +int n_a64; +int a_off; +int n_a; +int first_a; +int first_a40; +int b_off; +int n_b; +int c_off; +int n_c; +int oa_type;/* of enum xe_oa_format_name */ +bool report_hdr_64bit; +int counter_select; +int counter_size; +int bc_report; +}; + +static struct oa_format gen12_oa_formats[XE_OA_FORMAT_MAX ] = { + [XE_OA_FORMAT_A32u40_A4u32_B8_C8] = { +"A32u40_A4u32_B8_C8" , .size =256 , + .a40_high_off =160 , .a40_low_off =16 , .n_a40 =32 , + .a_off =144 , .n_a =4 , .first_a =32 , + .b_off =192 , .n_b =8 , + .c_off =224 , .n_c =8 , .oa_type =DRM_XE_OA_FMT_TYPE_OAG , + .counter_select =5 , + }, +}; + +static struct oa_format dg2_oa_formats[XE_OA_FORMAT_MAX ] = { + [XE_OAR_FORMAT_A32u40_A4u32_B8_C8] = { +"A32u40_A4u32_B8_C8" , .size =256 , + .a40_high_off =160 , .a40_low_off =16 , .n_a40 =32 , + .a_off =144 , .n_a =4 , .first_a =32 , + .b_off =192 , .n_b =8 , + .c_off =224 , .n_c =8 , .oa_type =DRM_XE_OA_FMT_TYPE_OAR , + .counter_select =5 , + }, +/* This format has A36 and A37 interleaved with high bytes of some A +* counters, so we will accumulate only subset of counters. +*/ + [XE_OA_FORMAT_A24u40_A14u32_B8_C8] = { +"A24u40_A14u32_B8_C8" , .size =256 , +/* u40: A4 - A23 */ + .a40_high_off =160 , .a40_low_off =16 , .n_a40 =20 , .first_a40 =4 , +/* u32: A0 - A3 */ + .a_off =16 , .n_a =4 , + .b_off =192 , .n_b =8 , + .c_off =224 , .n_c =8 , .oa_type =DRM_XE_OA_FMT_TYPE_OAG , + .counter_select =5 , + }, +/* This format has 24 u64 counters ranging from A0 - A35. Until we come +* up with a better mechanism to define missing counters, we will use a +* subset of counters that are indexed by one-increments - A28 - A35. +*/ + [XE_OAC_FORMAT_A24u64_B8_C8] = { +"OAC_A24u64_B8_C8" , .size =320 , + .a64_off =160 , .n_a64 =8 , + .b_off =224 , .n_b =8 , + .c_off =256 , .n_c =8 , .oa_type =DRM_XE_OA_FMT_TYPE_OAC , + .report_hdr_64bit = true, + .counter_select =1 , }, +}; + +static struct oa_format mtl_oa_formats[XE_OA_FORMAT_MAX ] = { + [XE_OAR_FORMAT_A32u40_A4u32_B8_C8] = { +"A32u40_A4u32_B8_C8" , .size =256 , + .a40_high_off =160 , .a40_low_off =16 , .n_a40 =32 , + .a_off =144 , .n_a =4 , .first_a =32 , + .b_off =192 , .n_b =8 , + .c_off =224 , .n_c =8 , .oa_type =DRM_XE_OA_FMT_TYPE_OAR , + .counter_select =5 , + }, +/* This format has A36 and A37 interleaved with high bytes of some A +* counters, so we will accumulate only subset of counters. +*/ + [XE_OA_FORMAT_A24u40_A14u32_B8_C8] = { +"A24u40_A14u32_B8_C8" , .size =256 , +/* u40: A4 - A23 */ + .a40_high_off =160 , .a40_low_off =16 , .n_a40 =20 , .first_a40 =4 , +/* u32: A0 - A3 */ + .a_off =16 , .n_a =4 , + .b_off =192 , .n_b =8 , + .c_off =224 , .n_c =8 , .oa_type =DRM_XE_OA_FMT_TYPE_OAG , + .counter_select =5 , + }, + +/* Treat MPEC countes as A counters for now */ + [XE_OAM_FORMAT_MPEC8u64_B8_C8] = { +"MPEC8u64_B8_C8" , .size =192 , + .a64_off =32 , .n_a64 =8 , + .b_off =96 , .n_b =8 , + .c_off =128 , .n_c =8 , .oa_type =DRM_XE_OA_FMT_TYPE_OAM_MPEC , + .report_hdr_64bit = true, + .counter_select =1 , + }, + [XE_OAM_FORMAT_MPEC8u32_B8_C8] = { +"MPEC8u32_B8_C8" , .size =128 , + .a_off =32 , .n_a =8 , + .b_off =64 , .n_b =8 , + .c_off =96 , .n_c =8 , .oa_type =DRM_XE_OA_FMT_TYPE_OAM_MPEC , + .report_hdr_64bit = true, + .counter_select =2 , + }, +/* This format has 24 u64 counters ranging from A0 - A35. Until we come +* up with a better mechanism to define missing counters, we will use a +* subset of counters that are indexed by one-increments - A28 - A35. +*/ + [XE_OAC_FORMAT_A24u64_B8_C8] = { +"OAC_A24u64_B8_C8" , .size =320 , + .a64_off =160 , .n_a64 =8 , + .b_off =224 , .n_b =8 , + .c_off =256 , .n_c =8 , .oa_type =DRM_XE_OA_FMT_TYPE_OAC , + .report_hdr_64bit = true, + .counter_select =1 , }, +}; + +static struct oa_format lnl_oa_formats[XE_OA_FORMAT_MAX ] = { + [XE_OA_FORMAT_PEC64u64] = { +"PEC64u64" , .size =576 , + .oa_type =DRM_XE_OA_FMT_TYPE_PEC , + .report_hdr_64bit = true, + .counter_select =1 , + .counter_size =1 , + .bc_report =0 }, + [XE_OA_FORMAT_PEC64u64_B8_C8] = { +"PEC64u64_B8_C8" , .size =640 , + .oa_type =DRM_XE_OA_FMT_TYPE_PEC , + .report_hdr_64bit = true, + .counter_select =1 , + .counter_size =1 , + .bc_report =1 }, + [XE_OA_FORMAT_PEC64u32] = { +"PEC64u32" , .size =320 , + .oa_type =DRM_XE_OA_FMT_TYPE_PEC , + .report_hdr_64bit = true, + .counter_select =1 , + .counter_size =0 , + .bc_report =0 }, + [XE_OA_FORMAT_PEC32u64_G1] = { +"PEC32u64_G1" , .size =320 , + .oa_type =DRM_XE_OA_FMT_TYPE_PEC , + .report_hdr_64bit = true, + .counter_select =5 , + .counter_size =1 , + .bc_report =0 }, + [XE_OA_FORMAT_PEC32u32_G1] = { +"PEC32u32_G1" , .size =192 , + .oa_type =DRM_XE_OA_FMT_TYPE_PEC , + .report_hdr_64bit = true, + .counter_select =5 , + .counter_size =0 , + .bc_report =0 }, + [XE_OA_FORMAT_PEC32u64_G2] = { +"PEC32u64_G2" , .size =320 , + .oa_type =DRM_XE_OA_FMT_TYPE_PEC , + .report_hdr_64bit = true, + .counter_select =6 , + .counter_size =1 , + .bc_report =0 }, + [XE_OA_FORMAT_PEC32u32_G2] = { +"PEC32u64_G2" , .size =192 , + .oa_type =DRM_XE_OA_FMT_TYPE_PEC , + .report_hdr_64bit = true, + .counter_select =6 , + .counter_size =0 , + .bc_report =0 }, + [XE_OA_FORMAT_PEC36u64_G1_32_G2_4] = { +"PEC36u64_G1_32_G2_4" , .size =320 , + .oa_type =DRM_XE_OA_FMT_TYPE_PEC , + .report_hdr_64bit = true, + .counter_select =3 , + .counter_size =1 , + .bc_report =0 }, + [XE_OA_FORMAT_PEC36u64_G1_4_G2_32] = { +"PEC36u64_G1_4_G2_32_G2" , .size =320 , + .oa_type =DRM_XE_OA_FMT_TYPE_PEC , + .report_hdr_64bit = true, + .counter_select =4 , + .counter_size =1 , + .bc_report =0 }, +}; + +static int drm_fd =-1 ; +static int sysfs =-1 ; +static int pm_fd =-1 ; +static int stream_fd =-1 ; +static uint32_t devid; + +struct drm_xe_engine_class_instance default_hwe; + +static struct intel_perf *intel_perf; +static uint64_t oa_exp_1_millisec; +struct intel_mmio_data mmio_data; + +static struct intel_perf_metric_set *metric_set (const struct drm_xe_engine_class_instance *hwe) +{ +const char *test_set_name =NULL ; +struct intel_perf_metric_set *metric_set_iter; +struct intel_perf_metric_set *test_set =NULL ; + +if (hwe->engine_class ==DRM_XE_ENGINE_CLASS_RENDER || + hwe->engine_class ==DRM_XE_ENGINE_CLASS_COMPUTE ) + test_set_name ="TestOa" ; +else if ((hwe->engine_class ==DRM_XE_ENGINE_CLASS_VIDEO_DECODE || + hwe->engine_class ==DRM_XE_ENGINE_CLASS_VIDEO_ENHANCE ) && +HAS_OAM (devid)) + test_set_name ="MediaSet1" ; +else +igt_assert (!"reached" ); + +igt_list_for_each_entry (metric_set_iter, &intel_perf->metric_sets, link) { +if (strcmp (metric_set_iter->symbol_name, test_set_name) ==0 ) { + test_set = metric_set_iter; +break ; + } + } + +igt_assert (test_set); + +/* +* configuration was loaded in init_sys_info() -> +* intel_perf_load_perf_configs(), and test_set->perf_oa_metrics_set +* should point to metric id returned by the config add ioctl. 0 is +* invalid. +*/ +igt_assert_neq_u64 (test_set->perf_oa_metrics_set,0 ); + +igt_debug ("engine %d:%d - %s metric set UUID = %s\n" , + hwe->engine_class, + hwe->engine_instance, + test_set->symbol_name, + test_set->hw_config_guid); + +return test_set; +} +#define default_test_set metric_set(&default_hwe) + +static void set_fd_flags (int fd,int flags) +{ +int old =fcntl (fd,F_GETFL ,0 ); + +igt_assert_lte (0 , old); +igt_assert_eq (0 ,fcntl (fd,F_SETFL , old | flags)); +} + +static u32 get_stream_status (int fd) +{ +struct drm_xe_oa_stream_status status; + +do_ioctl (fd,DRM_XE_PERF_IOCTL_STATUS , &status); + +return status.oa_status; +} + +static void +dump_report (const uint32_t *report,uint32_t size,const char *message) { +uint32_t i; +igt_debug ("%s\n" , message); +for (i =0 ; i < size; i +=4 ) { +igt_debug ("%08x %08x %08x %08x\n" , + report[i], + report[i +1 ], + report[i +2 ], + report[i +3 ]); + } +} + +static struct oa_format +get_oa_format (enum xe_oa_format_name format) +{ +if (IS_DG2 (devid)) +return dg2_oa_formats[format]; +else if (IS_METEORLAKE (devid)) +return mtl_oa_formats[format]; +else if (IS_LUNARLAKE (devid)) +return lnl_oa_formats[format]; +else +return gen12_oa_formats[format]; +} + +static u64 oa_format_fields (u64 name) +{ +#define FIELD_PREP_ULL (_mask, _val) \ + (((_val) << (__builtin_ffsll(_mask) - 1)) & (_mask)) + +struct oa_format f =get_oa_format (name); + +/* 0 format name is invalid */ +if (!name) +memset (&f,0xff ,sizeof (f)); + +return FIELD_PREP_ULL (DRM_XE_OA_FORMAT_MASK_FMT_TYPE , (u64 )f.oa_type) | +FIELD_PREP_ULL (DRM_XE_OA_FORMAT_MASK_COUNTER_SEL , (u64 )f.counter_select) | +FIELD_PREP_ULL (DRM_XE_OA_FORMAT_MASK_COUNTER_SIZE , (u64 )f.counter_size) | +FIELD_PREP_ULL (DRM_XE_OA_FORMAT_MASK_BC_REPORT , (u64 )f.bc_report); +} +#define __ff oa_format_fields + +static void +__perf_close (int fd) +{ +close (fd); + stream_fd =-1 ; + +if (pm_fd >=0 ) { +close (pm_fd); + pm_fd =-1 ; + } +} + +static int +__perf_open (int fd,struct drm_xe_oa_open_prop *param,bool prevent_pm) +{ +int ret; +int32_t pm_value =0 ; + +if (stream_fd >=0 ) +__perf_close (stream_fd); +if (pm_fd >=0 ) { +close (pm_fd); + pm_fd =-1 ; + } + + ret =xe_perf_ioctl (fd,DRM_XE_PERF_OP_STREAM_OPEN , param); + +igt_assert (ret >=0 ); + errno =0 ; + +if (prevent_pm) { + pm_fd =open ("/dev/cpu_dma_latency" ,O_RDWR ); +igt_assert (pm_fd >=0 ); + +igt_assert_eq (write (pm_fd, &pm_value,sizeof (pm_value)),sizeof (pm_value)); + } + +return ret; +} + +static uint64_t +read_u64_file (const char *path) +{ +FILE *f; +uint64_t val; + + f =fopen (path,"r" ); +igt_assert (f); + +igt_assert_eq (fscanf (f,"%" PRIu64, &val),1 ); + +fclose (f); + +return val; +} + +static void +write_u64_file (const char *path,uint64_t val) +{ +FILE *f; + + f =fopen (path,"w" ); +igt_assert (f); + +igt_assert (fprintf (f,"%" PRIu64, val) >0 ); + +fclose (f); +} + +static uint32_t +report_reason (const uint32_t *report) +{ +return ((report[0 ] >>OAREPORT_REASON_SHIFT ) & +OAREPORT_REASON_MASK ); +} + +static uint64_t +oa_timestamp (const uint32_t *report,enum xe_oa_format_name format) +{ +struct oa_format fmt =get_oa_format (format); + +return fmt.report_hdr_64bit ? *(uint64_t *)&report[2 ] : report[1 ]; +} + +static uint64_t +timebase_scale (uint64_t delta) +{ +return (delta *NSEC_PER_SEC ) / intel_perf->devinfo.timestamp_frequency; +} + +/* Returns: the largest OA exponent that will still result in a sampling period +* less than or equal to the given @period. +*/ +static int +max_oa_exponent_for_period_lte (uint64_t period) +{ +/* NB: timebase_scale() takes a uint64_t and an exponent of 30 +* would already represent a period of ~3 minutes so there's +* really no need to consider higher exponents. +*/ +for (int i =0 ; i <30 ; i++) { +uint64_t oa_period =timebase_scale (2 << i); + +if (oa_period > period) +return max (0 , i -1 ); + } + +igt_assert (!"reached" ); +return -1 ; +} + +static bool +oa_report_is_periodic (uint32_t oa_exponent,const uint32_t *report) +{ +if (report_reason (report) &OAREPORT_REASON_TIMER ) +return true; + +return false; +} + +static bool +init_sys_info (void ) +{ +igt_assert_neq (devid,0 ); + + intel_perf =intel_perf_for_fd (drm_fd,0 ); +igt_require (intel_perf); + +igt_debug ("n_eu_slices: %" PRIu64"\n" , intel_perf->devinfo.n_eu_slices); +igt_debug ("n_eu_sub_slices: %" PRIu64"\n" , intel_perf->devinfo.n_eu_sub_slices); +igt_debug ("n_eus: %" PRIu64"\n" , intel_perf->devinfo.n_eus); +igt_debug ("timestamp_frequency = %" PRIu64"\n" , + intel_perf->devinfo.timestamp_frequency); +igt_assert_neq (intel_perf->devinfo.timestamp_frequency,0 ); + +intel_perf_load_perf_configs (intel_perf, drm_fd); + + oa_exp_1_millisec =max_oa_exponent_for_period_lte (1000000 ); + +return true; +} + +static void +read_2_oa_reports (int format_id, +int exponent, +uint32_t *oa_report0, +uint32_t *oa_report1, +bool timer_only) +{ +size_t format_size =get_oa_format (format_id).size; +uint32_t exponent_mask = (1 << (exponent +1 )) -1 ; + +/* Note: we allocate a large buffer so that each read() iteration +* should scrape *all* pending records. +* +* The largest buffer the OA unit supports is 16MB. +* +* Being sure we are fetching all buffered reports allows us to +* potentially throw away / skip all reports whenever we see +* a _REPORT_LOST notification as a way of being sure are +* measurements aren't skewed by a lost report. +* +* Note: that is is useful for some tests but also not something +* applications would be expected to resort to. Lost reports are +* somewhat unpredictable but typically don't pose a problem - except +* to indicate that the OA unit may be over taxed if lots of reports +* are being lost. +*/ +int max_reports =MAX_OA_BUF_SIZE / format_size; +int buf_size = format_size * max_reports *1.5 ; +uint8_t *buf =malloc (buf_size); +int n =0 ; + +for (int i =0 ; i <1000 ; i++) { +ssize_t len; +u32 oa_status; + +while ((len =read (stream_fd, buf, buf_size)) <0 && + errno ==EINTR ) + ; + + oa_status =get_stream_status (stream_fd); +igt_debug ("oa_status %#x\n" , oa_status); + +igt_assert (len >0 ); +igt_debug ("read %d bytes\n" , (int )len); + +/* Need at least 2 reports */ +if (len <2 * format_size) +continue ; + +for (size_t offset =0 ; offset < len; offset += format_size) { +const uint32_t *report = (void *)(buf + offset); + +/* Currently the only test that should ever expect to +* see a _BUFFER_LOST error is the buffer_fill test, +* otherwise something bad has probably happened... +*/ +igt_assert (!(oa_status &OASTATUS_BUFFER_OVERFLOW )); + +/* At high sampling frequencies the OA HW might not be +* able to cope with all write requests and will notify +* us that a report was lost. We restart our read of +* two sequential reports due to the timeline blip this +* implies +*/ +if (oa_status &OASTATUS_REPORT_LOST ) { +igt_debug ("read restart: OA trigger collision / report lost\n" ); + n =0 ; + +/* XXX: break, because we don't know where +* within the series of already read reports +* there could be a blip from the lost report. +*/ +break ; + } + +dump_report (report, format_size /4 ,"oa-formats" ); + +igt_debug ("read report: reason = %x, timestamp = %" PRIx64", exponent mask=%x\n" , + report[0 ],oa_timestamp (report, format_id), exponent_mask); + +/* Don't expect zero for timestamps */ +igt_assert_neq_u64 (oa_timestamp (report, format_id),0 ); + +if (timer_only) { +if (!oa_report_is_periodic (exponent, report)) { +igt_debug ("skipping non timer report\n" ); +continue ; + } + } + +if (n++ ==0 ) +memcpy (oa_report0, report, format_size); +else { +memcpy (oa_report1, report, format_size); +free (buf); +return ; + } + } + } + +free (buf); + +igt_assert (!"reached" ); +} + +static unsigned read_xe_module_ref (void ) +{ +FILE *fp =fopen ("/proc/modules" ,"r" ); +char *line =NULL ; +size_t line_buf_size =0 ; +int len =0 ; +unsigned ref_count; +char mod[8 ]; +int modn =3 ; + +igt_assert (fp); + +strcpy (mod,"xe " ); +while ((len =getline (&line, &line_buf_size, fp)) >0 ) { +if (strncmp (line, mod, modn) ==0 ) { +unsigned long mem; +int ret =sscanf (line +5 ,"%lu %u" , &mem, &ref_count); +igt_assert (ret ==2 ); + goto done; + } + } + +igt_assert (!"reached" ); + +done: +free (line); +fclose (fp); +return ref_count; +} + +/** +* SUBTEST: xe-ref-count +* Description: Check that an open oa stream holds a reference on the xe module +*/ +static void +test_xe_ref_count (void ) +{ +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* Include OA reports in samples */ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_METRIC_SET ,0 /* updated below */ , +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (0 ),/* update below */ +DRM_XE_OA_PROPERTY_OA_EXPONENT ,0 ,/* update below */ + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; +unsigned baseline, ref_count0, ref_count1; +uint32_t oa_report0[64 ]; +uint32_t oa_report1[64 ]; + +/* This should be the first test before the first fixture so no drm_fd +* should have been opened so far... +*/ +igt_assert_eq (drm_fd,-1 ); + + baseline =read_xe_module_ref (); +igt_debug ("baseline ref count (drm fd closed) = %u\n" , baseline); + + drm_fd =__drm_open_driver (DRIVER_XE ); +if (is_xe_device (drm_fd)) +xe_device_get (drm_fd); + devid =intel_get_drm_devid (drm_fd); + sysfs =igt_sysfs_open (drm_fd); + +/* Note: these global variables are only initialized after calling +* init_sys_info()... +*/ +igt_require (init_sys_info ()); + properties[5 ] = default_test_set->perf_oa_metrics_set; + properties[7 ] =__ff (default_test_set->perf_oa_format); + properties[9 ] = oa_exp_1_millisec; + + ref_count0 =read_xe_module_ref (); +igt_debug ("initial ref count with drm_fd open = %u\n" , ref_count0); + + stream_fd =__perf_open (drm_fd, ¶m, false); +set_fd_flags (stream_fd,O_CLOEXEC ); + ref_count1 =read_xe_module_ref (); +igt_debug ("ref count after opening oa stream = %u\n" , ref_count1); + +drm_close_driver (drm_fd); +close (sysfs); + drm_fd =-1 ; + sysfs =-1 ; + ref_count0 =read_xe_module_ref (); +igt_debug ("ref count after closing drm fd = %u\n" , ref_count0); + +read_2_oa_reports (default_test_set->perf_oa_format, + oa_exp_1_millisec, + oa_report0, + oa_report1, + false);/* not just timer reports */ + +__perf_close (stream_fd); + ref_count0 =read_xe_module_ref (); +igt_debug ("ref count after closing oa stream fd = %u\n" , ref_count0); +} + +/** +* SUBTEST: sysctl-defaults +* Description: Test that perf_stream_paranoid sysctl exists +*/ +static void +test_sysctl_defaults (void ) +{ +int paranoid =read_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ); + +igt_assert_eq (paranoid,1 ); +} + +igt_main +{ + igt_fixture { +struct stat sb; + +/* +* Prior tests may have unloaded the module or failed while +* loading/unloading the module. Load xe here before we +* stat the files. +*/ +drm_load_module (DRIVER_XE ); +srandom (time (NULL )); +igt_require (!stat ("/proc/sys/dev/xe/perf_stream_paranoid" , &sb)); + } + + igt_subtest("xe-ref-count") +test_xe_ref_count (); + + igt_subtest("sysctl-defaults") +test_sysctl_defaults (); + + igt_fixture { +/* leave sysctl options in their default state... */ +write_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ,1 ); + +if (intel_perf) +intel_perf_free (intel_perf); + +drm_close_driver (drm_fd); + } +}
tests/meson.build +2 -0
@@ -304,6 +304,7 @@ intel_xe_progs = [ 'xe_mmap', 'xe_module_load', 'xe_noexec_ping_pong', + 'xe_oa', 'xe_pat', 'xe_peer2peer', 'xe_pm',
@@ -377,6 +378,7 @@ extra_dependencies = { 'perf': [ lib_igt_i915_perf ], 'perf_pmu': [ lib_igt_perf ], 'sw_sync': [ libatomic ], + 'xe_oa': [ lib_igt_xe_oa ], } test_executables = []
Signature · 1 line
2.41.0
PATCH 9/18 tests/intel/xe_oa: Add some negative tests
In reply to Ashutosh DixitAdd:
"non-system-wide-paranoid"
"invalid-oa-metric-set-id"
"invalid-oa-format-id"
"missing-sample-flags"
--- tests/intel/xe_oa.c | 192 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 192 insertions(+)
tests/intel/xe_oa.c +192 -0
@@ -341,6 +341,7 @@ struct drm_xe_engine_class_instance default_hwe;static struct intel_perf *intel_perf;static uint64_t oa_exp_1_millisec;struct intel_mmio_data mmio_data; +static igt_render_copyfunc_t render_copy;static struct intel_perf_metric_set *metric_set (const struct drm_xe_engine_class_instance *hwe) {
@@ -588,6 +589,166 @@ init_sys_info(void)return true; } +/** +* SUBTEST: non-system-wide-paranoid +* Description: CAP_SYS_ADMIN is required to open system wide metrics, unless +* sysctl parameter dev.xe.perf_stream_paranoid == 0 +*/ +static void test_system_wide_paranoid (void ) +{ + igt_fork(child, 1) { +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* Include OA reports in samples */ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , default_test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (default_test_set->perf_oa_format), +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exp_1_millisec, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; + +write_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ,1 ); + +igt_drop_root (); + +xe_perf_ioctl_err (drm_fd,DRM_XE_PERF_OP_STREAM_OPEN , ¶m,EACCES ); + } + +igt_waitchildren (); + + igt_fork(child, 1) { +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* Include OA reports in samples */ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , default_test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (default_test_set->perf_oa_format), +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exp_1_millisec, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; +write_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ,0 ); + +igt_drop_root (); + + stream_fd =__perf_open (drm_fd, ¶m, false); +__perf_close (stream_fd); + } + +igt_waitchildren (); + +/* leave in paranoid state */ +write_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ,1 ); +} + +/** +* SUBTEST: invalid-oa-metric-set-id +* Description: Test behavior for invalid metric set id's +*/ +static void test_invalid_oa_metric_set_id (void ) +{ +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* Include OA reports in samples */ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (default_test_set->perf_oa_format), +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exp_1_millisec, +DRM_XE_OA_PROPERTY_OA_METRIC_SET ,UINT64_MAX , + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; + +xe_perf_ioctl_err (drm_fd,DRM_XE_PERF_OP_STREAM_OPEN , ¶m,EINVAL ); + + properties[ARRAY_SIZE (properties) -1 ] =0 ;/* ID 0 is also be reserved as invalid */ +xe_perf_ioctl_err (drm_fd,DRM_XE_PERF_OP_STREAM_OPEN , ¶m,EINVAL ); + +/* Check that we aren't just seeing false positives... */ + properties[ARRAY_SIZE (properties) -1 ] = default_test_set->perf_oa_metrics_set; + stream_fd =__perf_open (drm_fd, ¶m, false); +__perf_close (stream_fd); + +/* There's no valid default OA metric set ID... */ + param.num_properties--; +xe_perf_ioctl_err (drm_fd,DRM_XE_PERF_OP_STREAM_OPEN , ¶m,EINVAL ); +} + +/** +* SUBTEST: invalid-oa-format-id +* Description: Test behavior for invalid OA format fields +*/ +static void test_invalid_oa_format_id (void ) +{ +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* Include OA reports in samples */ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , default_test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exp_1_millisec, +DRM_XE_OA_PROPERTY_OA_FORMAT ,UINT64_MAX ,/* No __ff() here */ + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; + +xe_perf_ioctl_err (drm_fd,DRM_XE_PERF_OP_STREAM_OPEN , ¶m,EINVAL ); + + properties[ARRAY_SIZE (properties) -1 ] =__ff (0 );/* ID 0 is also be reserved as invalid */ +xe_perf_ioctl_err (drm_fd,DRM_XE_PERF_OP_STREAM_OPEN , ¶m,EINVAL ); + +/* Check that we aren't just seeing false positives... */ + properties[ARRAY_SIZE (properties) -1 ] =__ff (default_test_set->perf_oa_format); + stream_fd =__perf_open (drm_fd, ¶m, false); +__perf_close (stream_fd); +/* There's no valid default OA format... */ + param.num_properties--; +xe_perf_ioctl_err (drm_fd,DRM_XE_PERF_OP_STREAM_OPEN , ¶m,EINVAL ); +} + +/** +* SUBTEST: missing-sample-flags +* Description: Test behavior for no SAMPLE_OA and no EXEC_QUEUE_ID +*/ +static void test_missing_sample_flags (void ) +{ +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* No _PROP_SAMPLE_xyz flags */ + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , default_test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exp_1_millisec, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (default_test_set->perf_oa_format), + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; + +xe_perf_ioctl_err (drm_fd,DRM_XE_PERF_OP_STREAM_OPEN , ¶m,EINVAL ); +} + static void read_2_oa_reports(int format_id, int exponent,
@@ -830,6 +991,37 @@ igt_main igt_subtest("sysctl-defaults")test_sysctl_defaults (); + igt_fixture { +/* We expect that the ref count test before these fixtures +* should have closed drm_fd... +*/ +igt_assert_eq (drm_fd,-1 ); + + drm_fd =drm_open_driver (DRIVER_XE ); +xe_device_get (drm_fd); + + devid =intel_get_drm_devid (drm_fd); + sysfs =igt_sysfs_open (drm_fd); + +igt_require (init_sys_info ()); + +write_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ,1 ); + + render_copy =igt_get_render_copyfunc (devid); + } + + igt_subtest("non-system-wide-paranoid") +test_system_wide_paranoid (); + + igt_subtest("invalid-oa-metric-set-id") +test_invalid_oa_metric_set_id (); + + igt_subtest("invalid-oa-format-id") +test_invalid_oa_format_id (); + + igt_subtest("missing-sample-flags") +test_missing_sample_flags (); + igt_fixture {/* leave sysctl options in their default state... */ write_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ,1 );
Signature · 1 line
2.41.0
PATCH 10/18 tests/intel/xe_oa: Add "oa-formats" subtest
In reply to Ashutosh DixitAdd "oa-formats" subtest.
--- tests/intel/xe_oa.c | 550 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 550 insertions(+)
tests/intel/xe_oa.c +550 -0
@@ -330,6 +330,9 @@ static struct oa_format lnl_oa_formats[XE_OA_FORMAT_MAX] = { .bc_report = 0 }, }; +/* No A counters currently reserved/undefined for gen8+ so far */ +static bool undefined_a_counters[45 ]; +static int drm_fd =-1 ;static int sysfs =-1 ;static int pm_fd =-1 ;
@@ -516,6 +519,57 @@ write_u64_file(const char *path, uint64_t val)fclose (f); } +static uint64_t +read_report_ticks (const uint32_t *report,enum xe_oa_format_name format) +{ + +struct oa_format fmt =get_oa_format (format); + +return fmt.report_hdr_64bit ? *(uint64_t *)&report[6 ] : report[3 ]; +} + +/* +* t0 is a value sampled before t1. width is number of bits used to represent +* t0/t1. Normally t1 is greater than t0. In cases where t1 < t0 use this +* helper. Since the size of t1/t0 is already 64 bits, no special handling is +* needed for width = 64. +*/ +static uint64_t +elapsed_delta (uint64_t t1,uint64_t t0,uint32_t width) +{ +uint32_t max_bits =sizeof (t1) *8 ; + +igt_assert (width <= max_bits); + +if (t1 < t0 && width != max_bits) +return ((1ULL << width) - t0) + t1; + +return t1 - t0; +} + +static uint64_t +oa_tick_delta (const uint32_t *report1, +const uint32_t *report0, +enum xe_oa_format_name format) +{ +return elapsed_delta (read_report_ticks (report1, format), +read_report_ticks (report0, format),32 ); +} + +static void +read_report_clock_ratios (const uint32_t *report, +uint32_t *slice_freq_mhz, +uint32_t *unslice_freq_mhz) +{ +uint32_t unslice_freq = report[0 ] &0x1ff ; +uint32_t slice_freq_low = (report[0 ] >>25 ) &0x7f ; +uint32_t slice_freq_high = (report[0 ] >>9 ) &0x3 ; +uint32_t slice_freq = slice_freq_low | (slice_freq_high <<7 ); + + *slice_freq_mhz = (slice_freq *16666 ) /1000 ; + *unslice_freq_mhz = (unslice_freq *16666 ) /1000 ; +} +static uint32_t report_reason (const uint32_t *report) {
@@ -523,6 +577,27 @@ report_reason(const uint32_t *report) OAREPORT_REASON_MASK); } +static const char * +read_report_reason (const uint32_t *report) +{ +uint32_t reason =report_reason (report); + +if (reason & (1 <<0 )) +return "timer" ; +else if (reason & (1 <<1 )) +return "internal trigger 1" ; +else if (reason & (1 <<2 )) +return "internal trigger 2" ; +else if (reason & (1 <<3 )) +return "context switch" ; +else if (reason & (1 <<4 )) +return "GO 1->0 transition (enter RC6)" ; +else if (reason & (1 <<5 )) +return "[un]slice clock ratio change" ; +else +return "unknown" ; +} +static uint64_t oa_timestamp (const uint32_t *report,enum xe_oa_format_name format) {
@@ -531,6 +606,17 @@ oa_timestamp(const uint32_t *report, enum xe_oa_format_name format)return fmt.report_hdr_64bit ? *(uint64_t *)&report[2 ] : report[1 ]; } +static uint64_t +oa_timestamp_delta (const uint32_t *report1, +const uint32_t *report0, +enum xe_oa_format_name format) +{ +uint32_t width =intel_graphics_ver (devid) >=IP_VER (12 ,55 ) ?56 :32 ; + +return elapsed_delta (oa_timestamp (report1, format), +oa_timestamp (report0, format), width); +} +static uint64_t timebase_scale (uint64_t delta) {
@@ -567,6 +653,191 @@ oa_report_is_periodic(uint32_t oa_exponent, const uint32_t *report)return false; } +static uint64_t +read_40bit_a_counter (const uint32_t *report, +enum xe_oa_format_name fmt,int a_id) +{ +struct oa_format format =get_oa_format (fmt); +uint8_t *a40_high = (((uint8_t *)report) + format.a40_high_off); +uint32_t *a40_low = (uint32_t *)(((uint8_t *)report) + + format.a40_low_off); +uint64_t high = (uint64_t )(a40_high[a_id]) <<32 ; + +return a40_low[a_id] | high; +} + +static uint64_t +xehpsdv_read_64bit_a_counter (const uint32_t *report,enum xe_oa_format_name fmt,int a_id) +{ +struct oa_format format =get_oa_format (fmt); +uint64_t *a64 = (uint64_t *)(((uint8_t *)report) + format.a64_off); + +return a64[a_id]; +} + +static uint64_t +get_40bit_a_delta (uint64_t value0,uint64_t value1) +{ +if (value0 > value1) +return (1ULL <<40 ) + value1 - value0; +else +return value1 - value0; +} + +static void +accumulate_uint64 (int a_index, +const uint32_t *report0, +const uint32_t *report1, +enum xe_oa_format_name format, +uint64_t *delta) +{ +uint64_t value0 =xehpsdv_read_64bit_a_counter (report0, format, a_index), + value1 =xehpsdv_read_64bit_a_counter (report1, format, a_index); + + *delta += (value1 - value0); +} + +/* The TestOa metric set is designed so */ +static void +sanity_check_reports (const uint32_t *oa_report0,const uint32_t *oa_report1, +enum xe_oa_format_name fmt) +{ +struct oa_format format =get_oa_format (fmt); +uint64_t time_delta =timebase_scale (oa_timestamp_delta (oa_report1, + oa_report0, + fmt)); +uint64_t clock_delta =oa_tick_delta (oa_report1, oa_report0, fmt); +uint64_t max_delta; +uint64_t freq; +uint32_t *rpt0_b = (uint32_t *)(((uint8_t *)oa_report0) + + format.b_off); +uint32_t *rpt1_b = (uint32_t *)(((uint8_t *)oa_report1) + + format.b_off); +uint32_t b; +uint32_t ref; + +igt_debug ("report type: %s->%s\n" , +read_report_reason (oa_report0), +read_report_reason (oa_report1)); + + freq = time_delta ? (clock_delta *1000 ) / time_delta :0 ; +igt_debug ("freq = %" PRIu64"\n" , freq); + +igt_debug ("clock delta = %" PRIu64"\n" , clock_delta); + + max_delta = clock_delta * intel_perf->devinfo.n_eus; + +/* Gen8+ has some 40bit A counters... */ +for (int j = format.first_a40; j < format.n_a40 + format.first_a40; j++) { +uint64_t value0 =read_40bit_a_counter (oa_report0, fmt, j); +uint64_t value1 =read_40bit_a_counter (oa_report1, fmt, j); +uint64_t delta =get_40bit_a_delta (value0, value1); + +if (undefined_a_counters[j]) +continue ; + +igt_debug ("A40_%d: delta = %" PRIu64"\n" , j, delta); +igt_assert_f (delta <= max_delta, +"A40_%d: delta = %" PRIu64", max_delta = %" PRIu64"\n" , + j, delta, max_delta); + } + +for (int j =0 ; j < format.n_a64; j++) { +uint64_t delta =0 ; + +accumulate_uint64 (j, oa_report0, oa_report1, fmt, &delta); + +if (undefined_a_counters[j]) +continue ; + +igt_debug ("A64_%d: delta = %" PRIu64"\n" , format.first_a + j, delta); +igt_assert_f (delta <= max_delta, +"A64_%d: delta = %" PRIu64", max_delta = %" PRIu64"\n" , + format.first_a + j, delta, max_delta); + } + +for (int j =0 ; j < format.n_a; j++) { +uint32_t *a0 = (uint32_t *)(((uint8_t *)oa_report0) + + format.a_off); +uint32_t *a1 = (uint32_t *)(((uint8_t *)oa_report1) + + format.a_off); +int a_id = format.first_a + j; +uint32_t delta = a1[j] - a0[j]; + +if (undefined_a_counters[a_id]) +continue ; + +igt_debug ("A%d: delta = %" PRIu32"\n" , a_id, delta); +igt_assert_f (delta <= max_delta, +"A%d: delta = %" PRIu32", max_delta = %" PRIu64"\n" , + a_id, delta, max_delta); + } + +/* The TestOa metric set defines all B counters to be a +* multiple of the gpu clock +*/ +if (format.n_b && (format.oa_type ==DRM_XE_OA_FMT_TYPE_OAG || format.oa_type ==DRM_XE_OA_FMT_TYPE_OAR )) { +if (clock_delta >0 ) { + b = rpt1_b[0 ] - rpt0_b[0 ]; +igt_debug ("B0: delta = %" PRIu32"\n" , b); +igt_assert_eq (b,0 ); + + b = rpt1_b[1 ] - rpt0_b[1 ]; +igt_debug ("B1: delta = %" PRIu32"\n" , b); +igt_assert_eq (b, clock_delta); + + b = rpt1_b[2 ] - rpt0_b[2 ]; +igt_debug ("B2: delta = %" PRIu32"\n" , b); +igt_assert_eq (b, clock_delta); + + b = rpt1_b[3 ] - rpt0_b[3 ]; + ref = clock_delta /2 ; +igt_debug ("B3: delta = %" PRIu32"\n" , b); +igt_assert (b >= ref -1 && b <= ref +1 ); + + b = rpt1_b[4 ] - rpt0_b[4 ]; + ref = clock_delta /3 ; +igt_debug ("B4: delta = %" PRIu32"\n" , b); +igt_assert (b >= ref -1 && b <= ref +1 ); + + b = rpt1_b[5 ] - rpt0_b[5 ]; + ref = clock_delta /3 ; +igt_debug ("B5: delta = %" PRIu32"\n" , b); +igt_assert (b >= ref -1 && b <= ref +1 ); + + b = rpt1_b[6 ] - rpt0_b[6 ]; + ref = clock_delta /6 ; +igt_debug ("B6: delta = %" PRIu32"\n" , b); +igt_assert (b >= ref -1 && b <= ref +1 ); + + b = rpt1_b[7 ] - rpt0_b[7 ]; + ref = clock_delta *2 /3 ; +igt_debug ("B7: delta = %" PRIu32"\n" , b); +igt_assert (b >= ref -1 && b <= ref +1 ); + }else { +for (int j =0 ; j < format.n_b; j++) { + b = rpt1_b[j] - rpt0_b[j]; +igt_debug ("B%i: delta = %" PRIu32"\n" , j, b); +igt_assert_eq (b,0 ); + } + } + } + +for (int j =0 ; j < format.n_c; j++) { +uint32_t *c0 = (uint32_t *)(((uint8_t *)oa_report0) + + format.c_off); +uint32_t *c1 = (uint32_t *)(((uint8_t *)oa_report1) + + format.c_off); +uint32_t delta = c1[j] - c0[j]; + +igt_debug ("C%d: delta = %" PRIu32", max_delta=%" PRIu64"\n" , + j, delta, max_delta); +igt_assert_f (delta <= max_delta, +"C%d: delta = %" PRIu32", max_delta = %" PRIu64"\n" , + j, delta, max_delta); + } +} +static bool init_sys_info (void ) {
@@ -854,6 +1125,264 @@ read_2_oa_reports(int format_id,igt_assert (!"reached" ); } +static void +open_and_read_2_oa_reports (int format_id, +int exponent, +uint32_t *oa_report0, +uint32_t *oa_report1, +bool timer_only, +const struct drm_xe_engine_class_instance *hwe) +{ +struct intel_perf_metric_set *test_set =metric_set (hwe); +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* Include OA reports in samples */ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (format_id), +DRM_XE_OA_PROPERTY_OA_EXPONENT , exponent, +DRM_XE_OA_PROPERTY_OA_ENGINE_INSTANCE , hwe->engine_instance, + + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; + + stream_fd =__perf_open (drm_fd, ¶m, false); +set_fd_flags (stream_fd,O_CLOEXEC ); + +read_2_oa_reports (format_id, exponent, + oa_report0, oa_report1, timer_only); + +__perf_close (stream_fd); +} + +static void +print_reports (uint32_t *oa_report0,uint32_t *oa_report1,int fmt) +{ +struct oa_format format =get_oa_format (fmt); +uint64_t ts0 =oa_timestamp (oa_report0, fmt); +uint64_t ts1 =oa_timestamp (oa_report1, fmt); + +igt_debug ("TIMESTAMP: 1st = %" PRIu64", 2nd = %" PRIu64", delta = %" PRIu64"\n" , + ts0, ts1, ts1 - ts0); + + { +uint64_t clock0 =read_report_ticks (oa_report0, fmt); +uint64_t clock1 =read_report_ticks (oa_report1, fmt); + +igt_debug ("CLOCK: 1st = %" PRIu64", 2nd = %" PRIu64", delta = %" PRIu64"\n" , + clock0, clock1, clock1 - clock0); + } + + { +uint32_t slice_freq0, slice_freq1, unslice_freq0, unslice_freq1; +const char *reason0 =read_report_reason (oa_report0); +const char *reason1 =read_report_reason (oa_report1); + +igt_debug ("CTX ID: 1st = %" PRIu32", 2nd = %" PRIu32"\n" , + oa_report0[2 ], oa_report1[2 ]); + +read_report_clock_ratios (oa_report0, + &slice_freq0, &unslice_freq0); +read_report_clock_ratios (oa_report1, + &slice_freq1, &unslice_freq1); + +igt_debug ("SLICE CLK: 1st = %umhz, 2nd = %umhz, delta = %d\n" , + slice_freq0, slice_freq1, + ((int )slice_freq1 - (int )slice_freq0)); +igt_debug ("UNSLICE CLK: 1st = %umhz, 2nd = %umhz, delta = %d\n" , + unslice_freq0, unslice_freq1, + ((int )unslice_freq1 - (int )unslice_freq0)); + +igt_debug ("REASONS: 1st = \"%s\", 2nd = \"%s\"\n" , reason0, reason1); + } + +/* Gen8+ has some 40bit A counters... */ +for (int j =0 ; j < format.n_a40; j++) { +uint64_t value0 =read_40bit_a_counter (oa_report0, fmt, j); +uint64_t value1 =read_40bit_a_counter (oa_report1, fmt, j); +uint64_t delta =get_40bit_a_delta (value0, value1); + +if (undefined_a_counters[j]) +continue ; + +igt_debug ("A%d: 1st = %" PRIu64", 2nd = %" PRIu64", delta = %" PRIu64"\n" , + j, value0, value1, delta); + } + +for (int j =0 ; j < format.n_a64; j++) { +uint64_t value0 =xehpsdv_read_64bit_a_counter (oa_report0, fmt, j); +uint64_t value1 =xehpsdv_read_64bit_a_counter (oa_report1, fmt, j); +uint64_t delta = value1 - value0; + +if (undefined_a_counters[j]) +continue ; + +igt_debug ("A_64%d: 1st = %" PRIu64", 2nd = %" PRIu64", delta = %" PRIu64"\n" , + format.first_a + j, value0, value1, delta); + } + +for (int j =0 ; j < format.n_a; j++) { +uint32_t *a0 = (uint32_t *)(((uint8_t *)oa_report0) + + format.a_off); +uint32_t *a1 = (uint32_t *)(((uint8_t *)oa_report1) + + format.a_off); +int a_id = format.first_a + j; +uint32_t delta = a1[j] - a0[j]; + +if (undefined_a_counters[a_id]) +continue ; + +igt_debug ("A%d: 1st = %" PRIu32", 2nd = %" PRIu32", delta = %" PRIu32"\n" , + a_id, a0[j], a1[j], delta); + } + +for (int j =0 ; j < format.n_b; j++) { +uint32_t *b0 = (uint32_t *)(((uint8_t *)oa_report0) + + format.b_off); +uint32_t *b1 = (uint32_t *)(((uint8_t *)oa_report1) + + format.b_off); +uint32_t delta = b1[j] - b0[j]; + +igt_debug ("B%d: 1st = %" PRIu32", 2nd = %" PRIu32", delta = %" PRIu32"\n" , + j, b0[j], b1[j], delta); + } + +for (int j =0 ; j < format.n_c; j++) { +uint32_t *c0 = (uint32_t *)(((uint8_t *)oa_report0) + + format.c_off); +uint32_t *c1 = (uint32_t *)(((uint8_t *)oa_report1) + + format.c_off); +uint32_t delta = c1[j] - c0[j]; + +igt_debug ("C%d: 1st = %" PRIu32", 2nd = %" PRIu32", delta = %" PRIu32"\n" , + j, c0[j], c1[j], delta); + } +} + +/* Debug function, only useful when reports don't make sense. */ +#if 0 +static void +print_report (uint32_t *report,int fmt) +{ +struct oa_format format =get_oa_format (fmt); + +igt_debug ("TIMESTAMP: %" PRIu64"\n" ,oa_timestamp (report, fmt)); + + { +uint64_t clock =read_report_ticks (report, fmt); + +igt_debug ("CLOCK: %" PRIu64"\n" , clock); + } + + { +uint32_t slice_freq, unslice_freq; +const char *reason =read_report_reason (report); + +read_report_clock_ratios (report, &slice_freq, &unslice_freq); + +igt_debug ("SLICE CLK: %umhz\n" , slice_freq); +igt_debug ("UNSLICE CLK: %umhz\n" , unslice_freq); +igt_debug ("REASON: \"%s\"\n" , reason); +igt_debug ("CTX ID: %" PRIu32"/%" PRIx32"\n" , report[2 ], report[2 ]); + } + +/* Gen8+ has some 40bit A counters... */ +for (int j =0 ; j < format.n_a40; j++) { +uint64_t value =read_40bit_a_counter (report, fmt, j); + +if (undefined_a_counters[j]) +continue ; + +igt_debug ("A%d: %" PRIu64"\n" , j, value); + } + +for (int j =0 ; j < format.n_a; j++) { +uint32_t *a = (uint32_t *)(((uint8_t *)report) + + format.a_off); +int a_id = format.first_a + j; + +if (undefined_a_counters[a_id]) +continue ; + +igt_debug ("A%d: %" PRIu32"\n" , a_id, a[j]); + } + +for (int j =0 ; j < format.n_b; j++) { +uint32_t *b = (uint32_t *)(((uint8_t *)report) + + format.b_off); + +igt_debug ("B%d: %" PRIu32"\n" , j, b[j]); + } + +for (int j =0 ; j < format.n_c; j++) { +uint32_t *c = (uint32_t *)(((uint8_t *)report) + + format.c_off); + +igt_debug ("C%d: %" PRIu32"\n" , j, c[j]); + } +} +#endif + +static bool +hwe_supports_oa_type (int oa_type,const struct drm_xe_engine_class_instance *hwe) +{ +switch (oa_type) { +case DRM_XE_OA_FMT_TYPE_OAM : +case DRM_XE_OA_FMT_TYPE_OAM_MPEC : +return hwe->engine_class ==DRM_XE_ENGINE_CLASS_VIDEO_DECODE || + hwe->engine_class ==DRM_XE_ENGINE_CLASS_VIDEO_ENHANCE ; +case DRM_XE_OA_FMT_TYPE_OAG : +case DRM_XE_OA_FMT_TYPE_OAR : +return hwe->engine_class ==DRM_XE_ENGINE_CLASS_RENDER ; +case DRM_XE_OA_FMT_TYPE_OAC : +return hwe->engine_class ==DRM_XE_ENGINE_CLASS_COMPUTE ; +case DRM_XE_OA_FMT_TYPE_PEC : +return hwe->engine_class ==DRM_XE_ENGINE_CLASS_RENDER || + hwe->engine_class ==DRM_XE_ENGINE_CLASS_COMPUTE ; +default : +return false; + } + +} + +/** +* SUBTEST: oa-formats +* Description: Test that supported OA formats work as expected +*/ +static void test_oa_formats (const struct drm_xe_engine_class_instance *hwe) +{ +for (int i =0 ; i <XE_OA_FORMAT_MAX ; i++) { +struct oa_format format =get_oa_format (i); +uint32_t oa_report0[format.size /4 ]; +uint32_t oa_report1[format.size /4 ]; + +if (!format.name)/* sparse, indexed by ID */ +continue ; + +if (!hwe_supports_oa_type (format.oa_type, hwe)) +continue ; + +igt_debug ("Checking OA format %s\n" , format.name); + +open_and_read_2_oa_reports (i, + oa_exp_1_millisec, + oa_report0, + oa_report1, + false,/* timer reports only */ + hwe); + +print_reports (oa_report0, oa_report1, i); +sanity_check_reports (oa_report0, oa_report1, i); + } +} + +static unsigned read_xe_module_ref (void ) {FILE *fp =fopen ("/proc/modules" ,"r" );
@@ -970,8 +1499,25 @@ test_sysctl_defaults(void)igt_assert_eq (paranoid,1 ); } +#define __for_one_render_engine_0 (hwe) \ + xe_for_each_engine(drm_fd, hwe) \ + if (hwe->engine_class == DRM_XE_ENGINE_CLASS_RENDER) \ + break; \ + for_each_if(hwe->engine_class == DRM_XE_ENGINE_CLASS_RENDER) \ + igt_dynamic_f("rcs-%d", hwe->engine_instance) + +#define __for_one_render_engine (hwe) \ + for (int m = 0, done = 0; !done; m++) \ + for_each_if(m < xe_number_engines(drm_fd) && \ + (hwe = &xe_engine(drm_fd, m)->instance) && \ + hwe->engine_class == DRM_XE_ENGINE_CLASS_RENDER && \ + (done = 1)) \ + igt_dynamic_f("rcs-%d", hwe->engine_instance) + igt_main { +struct drm_xe_engine_class_instance *hwe =NULL ; + igt_fixture {struct stat sb;
@@ -1022,6 +1568,10 @@ igt_main igt_subtest("missing-sample-flags")test_missing_sample_flags (); + igt_subtest_with_dynamic("oa-formats") +__for_one_render_engine (hwe) +test_oa_formats (hwe); + igt_fixture {/* leave sysctl options in their default state... */ write_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ,1 );
Signature · 1 line
2.41.0
PATCH 11/18 tests/intel/xe_oa: Add oa exponent tests
In reply to Ashutosh DixitAdd "invalid-oa-exponent" and "oa-exponents".
--- tests/intel/xe_oa.c | 414 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 414 insertions(+)
tests/intel/xe_oa.c +414 -0
@@ -450,6 +450,27 @@ static u64 oa_format_fields(u64 name) }#define __ff oa_format_fields +static struct drm_xe_engine_class_instance *oa_unit_engine (int fd,int n) +{ +struct drm_xe_query_oa_units *qoa =xe_oa_units (fd); +struct drm_xe_engine_class_instance *hwe =NULL ; +struct drm_xe_oa_unit *oau; +u8 *poau; + + poau = (u8 *)&qoa->oa_units[0 ]; +for (int i =0 ; i < qoa->num_oa_units; i++) { + oau = (struct drm_xe_oa_unit *)poau; + +if (i == n) { + hwe = &oau->eci[random () % oau->num_engines]; +break ; + } + poau +=sizeof (*oau) + oau->num_engines *sizeof (oau->eci[0 ]); + } + +return hwe; +} +static void __perf_close (int fd) {
@@ -644,6 +665,64 @@ max_oa_exponent_for_period_lte(uint64_t period)return -1 ; } +static uint64_t +oa_exponent_to_ns (int exponent) +{ +return 1000000000ULL * (2ULL << exponent) / intel_perf->devinfo.timestamp_frequency; +} + +static bool +oa_report_ctx_is_valid (uint32_t *report) +{ +return report[0 ] & (1ul <<16 ); +} + +static uint32_t +oa_report_get_ctx_id (uint32_t *report) +{ +if (!oa_report_ctx_is_valid (report)) +return 0xffffffff ; +return report[2 ]; +} + +static void *buf_map (int fd,struct intel_buf *buf,bool write) +{ +void *p; + +if (is_xe_device (fd)) { + buf->ptr =xe_bo_map (fd, buf->handle, buf->surface[0 ].size); + p = buf->ptr; + }else { +if (gem_has_llc (fd)) + p =intel_buf_cpu_map (buf, write); +else + p =intel_buf_device_map (buf, write); + } +return p; +} + +static void +scratch_buf_memset (struct intel_buf *buf,int width,int height,uint32_t color) +{ +buf_map (buf_ops_get_fd (buf->bops), buf, true); + +for (int i =0 ; i < width * height; i++) + buf->ptr[i] = color; + +intel_buf_unmap (buf); +} + +static void +scratch_buf_init (struct buf_ops *bops, +struct intel_buf *buf, +int width,int height, +uint32_t color) +{ +intel_buf_init (bops, buf, width, height,32 ,0 , +I915_TILING_NONE ,I915_COMPRESSION_NONE ); +scratch_buf_memset (buf, width, height, color); +} +static bool oa_report_is_periodic (uint32_t oa_exponent,const uint32_t *report) {
@@ -1383,6 +1462,303 @@ static void test_oa_formats(const struct drm_xe_engine_class_instance *hwe) } +enum load { +LOW , +HIGH +}; + +#define LOAD_HELPER_PAUSE_USEC 500 + +static struct load_helper { +int devid; +struct buf_ops *bops; +uint32_t context_id; +uint32_t vm; +struct intel_bb *ibb; +enum load load; +bool exit; +struct igt_helper_process igt_proc; +struct intel_buf src, dst; +} lh = {0 , }; + +static void load_helper_signal_handler (int sig) +{ +if (sig ==SIGUSR2 ) + lh.load = lh.load ==LOW ?HIGH :LOW ; +else + lh.exit = true; +} + +static void load_helper_set_load (enum load load) +{ +igt_assert (lh.igt_proc.running); + +if (lh.load == load) +return ; + + lh.load = load; +kill (lh.igt_proc.pid,SIGUSR2 ); +} + +static void load_helper_run (enum load load) +{ +if (!render_copy) +return ; + +/* +* FIXME fork helpers won't get cleaned up when started from within a +* subtest, so handle the case where it sticks around a bit too long. +*/ +if (lh.igt_proc.running) { +load_helper_set_load (load); +return ; + } + + lh.load = load; + + igt_fork_helper(&lh.igt_proc) { +signal (SIGUSR1 , load_helper_signal_handler); +signal (SIGUSR2 , load_helper_signal_handler); + +while (!lh.exit) { +render_copy (lh.ibb, + &lh.src,0 ,0 ,1920 ,1080 , + &lh.dst,0 ,0 ); + +intel_bb_sync (lh.ibb); + +/* Lower the load by pausing after every submitted +* write. */ +if (lh.load ==LOW ) +usleep (LOAD_HELPER_PAUSE_USEC ); + } + } +} + +static void load_helper_stop (void ) +{ +if (!render_copy) +return ; + +kill (lh.igt_proc.pid,SIGUSR1 ); +igt_assert (igt_wait_helper (&lh.igt_proc) ==0 ); +} + +static void load_helper_init (void ) +{ +if (!render_copy) { +igt_info ("Running test without render_copy\n" ); +return ; + } + + lh.devid =intel_get_drm_devid (drm_fd); + lh.bops =buf_ops_create (drm_fd); + lh.vm =xe_vm_create (drm_fd,0 ,0 ); + lh.context_id =xe_exec_queue_create (drm_fd, lh.vm, &xe_engine (drm_fd,0 )->instance,0 ); +igt_assert_neq (lh.context_id,0xffffffff ); + + lh.ibb =intel_bb_create_with_context (drm_fd, lh.context_id, lh.vm,NULL ,BATCH_SZ ); + +scratch_buf_init (lh.bops, &lh.dst,1920 ,1080 ,0 ); +scratch_buf_init (lh.bops, &lh.src,1920 ,1080 ,0 ); +} + +static void load_helper_fini (void ) +{ +if (!render_copy) +return ; + +if (lh.igt_proc.running) +load_helper_stop (); + +intel_buf_close (lh.bops, &lh.src); +intel_buf_close (lh.bops, &lh.dst); +intel_bb_destroy (lh.ibb); +xe_exec_queue_destroy (drm_fd, lh.context_id); +xe_vm_destroy (drm_fd, lh.vm); +buf_ops_destroy (lh.bops); +} + +static bool expected_report_timing_delta (uint32_t delta,uint32_t expected_delta) +{ +/* +* On ICL, the OA unit appears to be a bit more relaxed about +* its timing for emitting OA reports (often missing the +* deadline by 1 timestamp). +*/ +if (IS_ICELAKE (devid)) +return delta <= (expected_delta +3 ); +else +return delta <= expected_delta; +} + +/** +* SUBTEST: oa-exponents +* Description: Test that oa exponent values behave as expected +*/ +static void test_oa_exponents (const struct drm_xe_engine_class_instance *hwe) +{ +struct intel_perf_metric_set *test_set =metric_set (hwe); +uint64_t fmt = test_set->perf_oa_format; + +load_helper_init (); +load_helper_run (HIGH ); + +/* It's asking a lot to sample with a 160 nanosecond period and the +* test can fail due to buffer overflows if it wasn't possible to +* keep up, so we don't start from an exponent of zero... +*/ +for (int exponent =5 ; exponent <20 ; exponent++) { +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* Include OA reports in samples */ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (fmt), +DRM_XE_OA_PROPERTY_OA_EXPONENT , exponent, +DRM_XE_OA_PROPERTY_OA_ENGINE_INSTANCE , hwe->engine_instance, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; +uint64_t expected_timestamp_delta =2ULL << exponent; +size_t format_size =get_oa_format (fmt).size; +int max_reports =MAX_OA_BUF_SIZE / format_size; +int buf_size = format_size * max_reports *1.5 ; +uint8_t *buf =calloc (1 , buf_size); +int ret, n_timer_reports =0 ; +uint32_t matches =0 ; +#define NUM_TIMER_REPORTS 30 +uint32_t *reports =malloc (NUM_TIMER_REPORTS * format_size); +uint32_t *timer_reports = reports; + +igt_debug ("testing OA exponent %d," +" expected ts delta = %" PRIu64" (%" PRIu64"ns/%.2fus/%.2fms)\n" , + exponent, expected_timestamp_delta, +oa_exponent_to_ns (exponent), +oa_exponent_to_ns (exponent) /1000.0 , +oa_exponent_to_ns (exponent) / (1000.0 *1000.0 )); + + stream_fd =__perf_open (drm_fd, ¶m, true/* prevent_pm */ ); + +while (n_timer_reports <NUM_TIMER_REPORTS ) { +u32 oa_status; + +while ((ret =read (stream_fd, buf, buf_size)) <0 && + errno ==EINTR ) + ; + +/* igt_debug(" > read %i bytes\n", ret); */ + oa_status =get_stream_status (stream_fd); +igt_debug ("oa_status %#x\n" , oa_status); + +/* We should never have no data. */ +igt_assert (ret >0 ); + +for (int offset =0 ; + offset < ret && n_timer_reports <NUM_TIMER_REPORTS ; + offset += format_size) { +uint32_t *report = (void *)(buf + offset); + +if (oa_status &OASTATUS_BUFFER_OVERFLOW ) { +igt_assert (!"reached" ); +break ; + } + +if (oa_status &OASTATUS_REPORT_LOST ) +igt_debug ("report loss\n" ); + +if (!oa_report_is_periodic (exponent, report)) +continue ; + +memcpy (timer_reports, report, format_size); + n_timer_reports++; + timer_reports += (format_size /4 ); + } + } + +__perf_close (stream_fd); + +igt_debug ("report%04i ts=%" PRIx64" hw_id=0x%08x\n" ,0 , +oa_timestamp (&reports[0 ], fmt), +oa_report_get_ctx_id (&reports[0 ])); +for (int i =1 ; i < n_timer_reports; i++) { +uint64_t delta =oa_timestamp_delta (&reports[i], + &reports[i -1 ], + fmt); + +igt_debug ("report%04i ts=%" PRIx64" hw_id=0x%08x delta=%" PRIu64" %s\n" , i, +oa_timestamp (&reports[i], fmt), +oa_report_get_ctx_id (&reports[i]), + delta,expected_report_timing_delta (delta, + expected_timestamp_delta) ?"" :"******" ); + + matches +=expected_report_timing_delta (delta,expected_timestamp_delta); + } + +igt_debug ("matches=%u/%u\n" , matches, n_timer_reports -1 ); + +/* +* Expect half the reports to match the timing +* expectation. The results are quite erratic because +* the condition under which the HW reaches +* expectations depends on memory controller pressure +* etc... +*/ +igt_assert_lte (n_timer_reports /2 , matches); + +free (reports); + } + +load_helper_stop (); +load_helper_fini (); +} + +/** +* SUBTEST: invalid-oa-exponent +* Description: Test that invalid exponent values are rejected +*/ +/* The OA exponent selects a timestamp counter bit to trigger reports on. +* +* With a 64bit timestamp and least significant bit approx == 80ns then the MSB +* equates to > 40 thousand years and isn't exposed via the xe oa interface. +* +* The max exponent exposed is expected to be 31, which is still a fairly +* ridiculous period (>5min) but is the maximum exponent where it's still +* possible to use periodic sampling as a means for tracking the overflow of +* 32bit OA report timestamps. +*/ +static void test_invalid_oa_exponent (void ) +{ +uint64_t properties[] = { +/* Include OA reports in samples */ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , default_test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (default_test_set->perf_oa_format), +DRM_XE_OA_PROPERTY_OA_EXPONENT ,31 ,/* maximum exponent expected +to be accepted */ + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; + + stream_fd =__perf_open (drm_fd, ¶m, false); + +__perf_close (stream_fd); + +for (int i =32 ; i <65 ; i++) { + properties[7 ] = i; +xe_perf_ioctl_err (drm_fd,DRM_XE_PERF_OP_STREAM_OPEN , ¶m,EINVAL ); + } +} +static unsigned read_xe_module_ref (void ) {FILE *fp =fopen ("/proc/modules" ,"r" );
@@ -1499,6 +1875,37 @@ test_sysctl_defaults(void)igt_assert_eq (paranoid,1 ); } +static const char *xe_engine_class_name (uint32_t engine_class) +{ +switch (engine_class) { +case DRM_XE_ENGINE_CLASS_RENDER : +return "rcs" ; +case DRM_XE_ENGINE_CLASS_COPY : +return "bcs" ; +case DRM_XE_ENGINE_CLASS_VIDEO_DECODE : +return "vcs" ; +case DRM_XE_ENGINE_CLASS_VIDEO_ENHANCE : +return "vecs" ; +case DRM_XE_ENGINE_CLASS_COMPUTE : +return "ccs" ; +default : +igt_warn ("Engine class 0x%x unknown\n" , engine_class); +return "unknown" ; + } +} + +#define __for_one_hwe_in_each_oa_unit (hwe) \ + for (int m = 0; !m || hwe; m++) \ + for_each_if(hwe = oa_unit_engine(drm_fd, m)) \ + igt_dynamic_f("%s-%d", xe_engine_class_name(hwe->engine_class), \ + hwe->engine_instance) + +/* Only OAG (not OAM) is currently supported */ +#define __for_one_hwe_in_oag (hwe) \ + if ((hwe = oa_unit_engine(drm_fd, 0))) \ + igt_dynamic_f("%s-%d", xe_engine_class_name(hwe->engine_class), \ + hwe->engine_instance) +#define __for_one_render_engine_0 (hwe) \ xe_for_each_engine(drm_fd, hwe) \ if (hwe->engine_class == DRM_XE_ENGINE_CLASS_RENDER) \
@@ -1572,6 +1979,13 @@ igt_main__for_one_render_engine (hwe)test_oa_formats (hwe); + igt_subtest("invalid-oa-exponent") +test_invalid_oa_exponent (); + + igt_subtest_with_dynamic("oa-exponents") +__for_one_hwe_in_oag (hwe) +test_oa_exponents (hwe); + igt_fixture {/* leave sysctl options in their default state... */ write_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ,1 );
Signature · 1 line
2.41.0
PATCH 12/18 tests/intel/xe_oa: buffer-fill, non-zero-reason, enable-disable
In reply to Ashutosh DixitAdd the following tests:
"buffer-fill"
"non-zero-reason"
"disabled-read-error"
"enable-disable"
"short-reads"
--- tests/intel/xe_oa.c | 592 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 592 insertions(+)
tests/intel/xe_oa.c +592 -0
@@ -1759,6 +1759,577 @@ static void test_invalid_oa_exponent(void) } } +/** +* SUBTEST: buffer-fill +* Description: Test filling, wraparound and overflow of OA buffer +*/ +static void +test_buffer_fill (const struct drm_xe_engine_class_instance *hwe) +{ +/* ~5 micro second period */ +int oa_exponent =max_oa_exponent_for_period_lte (5000 ); +uint64_t oa_period =oa_exponent_to_ns (oa_exponent); +struct intel_perf_metric_set *test_set =metric_set (hwe); +uint64_t fmt = test_set->perf_oa_format; +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* Include OA reports in samples */ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (fmt), +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exponent, +DRM_XE_OA_PROPERTY_OA_ENGINE_INSTANCE , hwe->engine_instance, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; +size_t report_size =get_oa_format (fmt).size; +int buf_size =65536 * report_size; +uint8_t *buf =malloc (buf_size); +int len; +size_t oa_buf_size =MAX_OA_BUF_SIZE ; +int n_full_oa_reports = oa_buf_size / report_size; +uint64_t fill_duration = n_full_oa_reports * oa_period; +uint32_t *last_periodic_report =malloc (report_size); +u32 oa_status; + +igt_assert (fill_duration <1000000000 ); + + stream_fd =__perf_open (drm_fd, ¶m, true/* prevent_pm */ ); +set_fd_flags (stream_fd,O_CLOEXEC ); + +for (int i =0 ; i <5 ; i++) { +bool overflow_seen; +uint32_t n_periodic_reports; +uint32_t first_timestamp =0 , last_timestamp =0 ; + +do_ioctl (stream_fd,DRM_XE_PERF_IOCTL_ENABLE ,0 ); + +nanosleep (&(struct timespec ){ .tv_sec =0 , + .tv_nsec = fill_duration *1.25 }, +NULL ); + +while ((len =read (stream_fd, buf, buf_size)) ==-1 && errno ==EINTR ) + ; +igt_assert_neq (len,-1 ); + + oa_status =get_stream_status (stream_fd); +igt_debug ("oa_status %#x\n" , oa_status); + overflow_seen = (oa_status &OASTATUS_BUFFER_OVERFLOW ) ? true : false; + +/* Overrun mode is disabled in the kernel for Xe2+ */ +if (!IS_LUNARLAKE (xe_dev_id (drm_fd))) +igt_assert_eq (overflow_seen, true); + +do_ioctl (stream_fd,DRM_XE_PERF_IOCTL_DISABLE ,0 ); + +igt_debug ("fill_duration = %" PRIu64"ns, oa_exponent = %u\n" , + fill_duration, oa_exponent); + +do_ioctl (stream_fd,DRM_XE_PERF_IOCTL_ENABLE ,0 ); + +nanosleep (&(struct timespec ){ .tv_sec =0 , + .tv_nsec = fill_duration /2 }, +NULL ); + + n_periodic_reports =0 ; + +/* Because of the race condition between notification of new +* reports and reports landing in memory, we need to rely on +* timestamps to figure whether we've read enough of them. +*/ +while (((last_timestamp - first_timestamp) * oa_period) < (fill_duration /2 )) { + +igt_debug ("dts=%u elapsed=%" PRIu64" duration=%" PRIu64"\n" , + last_timestamp - first_timestamp, + (last_timestamp - first_timestamp) * oa_period, + fill_duration /2 ); + +while ((len =read (stream_fd, buf, buf_size)) ==-1 && errno ==EINTR ) + ; +igt_assert_neq (len,-1 ); + + oa_status =get_stream_status (stream_fd); +igt_debug ("oa_status %#x\n" , oa_status); +igt_assert (!(oa_status &OASTATUS_BUFFER_OVERFLOW )); + +for (int offset =0 ; offset < len; offset += report_size) { +uint32_t *report = (void *) (buf + offset); + +igt_debug (" > report ts=%" PRIu64"" +" ts_delta_last_periodic=%" PRIu64" is_timer=%i ctx_id=%8x nb_periodic=%u\n" , +oa_timestamp (report, fmt), + n_periodic_reports >0 ?oa_timestamp_delta (report, last_periodic_report, fmt) :0 , +oa_report_is_periodic (oa_exponent, report), +oa_report_get_ctx_id (report), + n_periodic_reports); + +if (first_timestamp ==0 ) + first_timestamp =oa_timestamp (report, fmt); + last_timestamp =oa_timestamp (report, fmt); + +if (oa_report_is_periodic (oa_exponent, report)) { +memcpy (last_periodic_report, report, report_size); + n_periodic_reports++; + } + } + } + +do_ioctl (stream_fd,DRM_XE_PERF_IOCTL_DISABLE ,0 ); + +igt_debug ("first ts = %u, last ts = %u\n" , first_timestamp, last_timestamp); + +igt_debug ("%f < %zu < %f\n" , + report_size * n_full_oa_reports *0.45 , + n_periodic_reports * report_size, + report_size * n_full_oa_reports *0.55 ); + +igt_assert (n_periodic_reports * report_size > + report_size * n_full_oa_reports *0.45 ); +igt_assert (n_periodic_reports * report_size < + report_size * n_full_oa_reports *0.55 ); + } + +free (last_periodic_report); +free (buf); + +__perf_close (stream_fd); +} + +/** +* SUBTEST: non-zero-reason +* Description: Test reason field is non-zero. Can also check OA buffer wraparound issues +*/ +static void +test_non_zero_reason (const struct drm_xe_engine_class_instance *hwe) +{ +/* ~20 micro second period */ +int oa_exponent =max_oa_exponent_for_period_lte (20000 ); +struct intel_perf_metric_set *test_set =metric_set (hwe); +uint64_t fmt = test_set->perf_oa_format; +size_t report_size =get_oa_format (fmt).size; +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* Include OA reports in samples */ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (fmt), +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exponent, +DRM_XE_OA_PROPERTY_OA_ENGINE_INSTANCE , hwe->engine_instance, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; +uint32_t buf_size =3 *65536 * report_size; +uint8_t *buf =malloc (buf_size); +uint32_t total_len =0 ; +const uint32_t *last_report; +int len; +u32 oa_status; + +igt_assert (buf); + +igt_debug ("Ready to read about %u bytes\n" , buf_size); + +load_helper_init (); +load_helper_run (HIGH ); + + stream_fd =__perf_open (drm_fd, ¶m, true/* prevent_pm */ ); +set_fd_flags (stream_fd,O_CLOEXEC ); + +while (total_len < buf_size && + ((len =read (stream_fd, &buf[total_len], buf_size - total_len)) >0 || + (len ==-1 && errno ==EINTR ))) { +if (len >0 ) + total_len += len; + } + + oa_status =get_stream_status (stream_fd); +igt_debug ("oa_status %#x\n" , oa_status); +igt_assert (!(oa_status &OASTATUS_BUFFER_OVERFLOW )); + +__perf_close (stream_fd); + +load_helper_stop (); +load_helper_fini (); + +igt_debug ("Got %u bytes\n" , total_len); + + last_report =NULL ; +for (uint32_t offset =0 ; offset < total_len; offset += report_size) { +const uint32_t *report = (void *) (buf + offset); +uint32_t reason = (report[0 ] >>OAREPORT_REASON_SHIFT ) &OAREPORT_REASON_MASK ; + +igt_assert_neq (reason,0 ); + +if (last_report) +sanity_check_reports (last_report, report, fmt); + + last_report = report; + } + +free (buf); +} + +/** +* SUBTEST: enable-disable +* Description: Test that OA stream enable/disable works as expected +*/ +static void +test_enable_disable (const struct drm_xe_engine_class_instance *hwe) +{ +/* ~5 micro second period */ +int oa_exponent =max_oa_exponent_for_period_lte (5000 ); +uint64_t oa_period =oa_exponent_to_ns (oa_exponent); +struct intel_perf_metric_set *test_set =metric_set (hwe); +uint64_t fmt = test_set->perf_oa_format; +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* Include OA reports in samples */ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (fmt), +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exponent, +DRM_XE_OA_PROPERTY_OA_DISABLED , true, +DRM_XE_OA_PROPERTY_OA_ENGINE_INSTANCE , hwe->engine_instance, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; +size_t report_size =get_oa_format (fmt).size; +int buf_size =65536 * report_size; +uint8_t *buf =malloc (buf_size); +size_t oa_buf_size =MAX_OA_BUF_SIZE ; +int n_full_oa_reports = oa_buf_size / report_size; +uint64_t fill_duration = n_full_oa_reports * oa_period; +uint32_t *last_periodic_report =malloc (report_size); + +load_helper_init (); +load_helper_run (HIGH ); + + stream_fd =__perf_open (drm_fd, ¶m, true/* prevent_pm */ ); +set_fd_flags (stream_fd,O_CLOEXEC ); + +for (int i =0 ; i <5 ; i++) { +int len; +uint32_t n_periodic_reports; +uint64_t first_timestamp =0 , last_timestamp =0 ; +u32 oa_status; + +/* Giving enough time for an overflow might help catch whether +* the OA unit has been enabled even if the driver might at +* least avoid copying reports while disabled. +*/ +nanosleep (&(struct timespec ){ .tv_sec =0 , + .tv_nsec = fill_duration *1.25 }, +NULL ); + +while ((len =read (stream_fd, buf, buf_size)) ==-1 && errno ==EINTR ) + ; + +igt_assert_eq (len,-1 ); +igt_assert_eq (errno,EIO ); + +do_ioctl (stream_fd,DRM_XE_PERF_IOCTL_ENABLE ,0 ); + +nanosleep (&(struct timespec ){ .tv_sec =0 , + .tv_nsec = fill_duration /2 }, +NULL ); + + n_periodic_reports =0 ; + +/* Because of the race condition between notification of new +* reports and reports landing in memory, we need to rely on +* timestamps to figure whether we've read enough of them. +*/ +while (((last_timestamp - first_timestamp) * oa_period) < (fill_duration /2 )) { + +while ((len =read (stream_fd, buf, buf_size)) ==-1 && errno ==EINTR ) + ; +igt_assert_neq (len,-1 ); + + oa_status =get_stream_status (stream_fd); +igt_debug ("oa_status %#x\n" , oa_status); +igt_assert (!(oa_status &OASTATUS_BUFFER_OVERFLOW )); + +for (int offset =0 ; offset < len; offset += report_size) { +uint32_t *report = (void *) (buf + offset); + +if (first_timestamp ==0 ) + first_timestamp =oa_timestamp (report, fmt); + last_timestamp =oa_timestamp (report, fmt); + +igt_debug (" > report ts=%" PRIx64"" +" ts_delta_last_periodic=%s%" PRIu64"" +" is_timer=%i ctx_id=0x%8x\n" , +oa_timestamp (report, fmt), +oa_report_is_periodic (oa_exponent, report) ?" " :"*" , + n_periodic_reports >0 ?oa_timestamp_delta (report, last_periodic_report, fmt) :0 , +oa_report_is_periodic (oa_exponent, report), +oa_report_get_ctx_id (report)); + +if (oa_report_is_periodic (oa_exponent, report)) { +memcpy (last_periodic_report, report, report_size); + +/* We want to measure only the periodic reports, +* ctx-switch might inflate the content of the +* buffer and skew or measurement. +*/ + n_periodic_reports++; + } + } + } + +do_ioctl (stream_fd,DRM_XE_PERF_IOCTL_DISABLE ,0 ); + +igt_debug ("first ts = %lu, last ts = %lu\n" , first_timestamp, last_timestamp); + +igt_debug ("%f < %zu < %f\n" , + report_size * n_full_oa_reports *0.45 , + n_periodic_reports * report_size, + report_size * n_full_oa_reports *0.55 ); + +igt_assert ((n_periodic_reports * report_size) > + (report_size * n_full_oa_reports *0.45 )); +igt_assert ((n_periodic_reports * report_size) < + report_size * n_full_oa_reports *0.55 ); + + +/* It's considered an error to read a stream while it's disabled +* since it would block indefinitely... +*/ + len =read (stream_fd, buf, buf_size); + +igt_assert_eq (len,-1 ); +igt_assert_eq (errno,EIO ); + } + +free (last_periodic_report); +free (buf); + +__perf_close (stream_fd); + +load_helper_stop (); +load_helper_fini (); +} + +/** +* SUBTEST: short-reads +* Description: Test behavior for short reads +*/ +static void +test_short_reads (void ) +{ +int oa_exponent =max_oa_exponent_for_period_lte (5000 ); +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* Include OA reports in samples */ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , default_test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (default_test_set->perf_oa_format), +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exponent, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; +size_t record_size =get_oa_format (default_test_set->perf_oa_format).size; +size_t page_size =sysconf (_SC_PAGE_SIZE); +int zero_fd =open ("/dev/zero" ,O_RDWR |O_CLOEXEC ); +uint8_t *pages =mmap (NULL , page_size *2 , +PROT_READ |PROT_WRITE ,MAP_PRIVATE , zero_fd,0 ); +u8 *header; +int ret, errnum; +u32 oa_status; + +igt_assert_neq (zero_fd,-1 ); +close (zero_fd); + zero_fd =-1 ; + +igt_assert (pages); + + ret =mprotect (pages + page_size, page_size,PROT_NONE ); +igt_assert_eq (ret,0 ); + + stream_fd =__perf_open (drm_fd, ¶m, false); + +nanosleep (&(struct timespec ){ .tv_sec =0 , .tv_nsec =5000000 },NULL ); + +/* At this point there should be lots of pending reports to read */ + +/* A read that can return at least one record should result in a short +* read not an EFAULT if the buffer is smaller than the requested read +* size... +* +* Expect to see a sample record here, but at least skip over any +* _RECORD_LOST notifications. +*/ +do { + header = (void *)(pages + page_size - record_size); + ret =read (stream_fd, header, page_size); +igt_assert (ret >0 ); + oa_status =get_stream_status (stream_fd); + + }while (oa_status &OASTATUS_REPORT_LOST ); + +igt_assert_eq (ret, record_size); + +/* A read that can't return a single record because it would result +* in a fault on buffer overrun should result in an EFAULT error... +* +* Make sure to weed out all report lost errors before verifying EFAULT. +*/ + header = (void *)(pages + page_size -16 ); +do { + ret =read (stream_fd, header, page_size); + errnum = errno; + oa_status =get_stream_status (stream_fd); + errno = errnum; + }while (ret >0 && oa_status &OASTATUS_REPORT_LOST ); +igt_assert_eq (ret,-1 ); +igt_assert_eq (errno,EFAULT ); + +/* A read that can't return a single record because the buffer is too +* small should result in an ENOSPC error.. +* +* Again, skip over _RECORD_LOST records (smaller than record_size/2) +*/ +do { + header = (void *)(pages + page_size - record_size /2 ); + ret =read (stream_fd, header, record_size /2 ); + errnum = errno; + oa_status =get_stream_status (stream_fd); + errno = errnum; + }while (ret >0 && oa_status &OASTATUS_REPORT_LOST ); + +igt_assert_eq (ret,-1 ); +igt_assert_eq (errno,ENOSPC ); + +__perf_close (stream_fd); + +munmap (pages, page_size *2 ); +} + +/** +* SUBTEST: non-sampling-read-error +* Description: Test that a stream without periodic sampling (no exponent) cannot be read +*/ +static void +test_non_sampling_read_error (void ) +{ +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* XXX: even without periodic sampling we have to +* specify at least one sample layout property... +*/ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , default_test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (default_test_set->perf_oa_format), + +/* XXX: no sampling exponent */ + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; +int ret; +uint8_t buf[1024 ]; + + stream_fd =__perf_open (drm_fd, ¶m, false); +set_fd_flags (stream_fd,O_CLOEXEC ); + + ret =read (stream_fd, buf,sizeof (buf)); +igt_assert_eq (ret,-1 ); +igt_assert_eq (errno,EIO ); + +__perf_close (stream_fd); +} + +/** +* SUBTEST: disabled-read-error +* Description: Test that attempts to read from a stream while it is disable +* will return EIO instead of blocking indefinitely +*/ +static void +test_disabled_read_error (void ) +{ +int oa_exponent =5 ;/* 5 micro seconds */ +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* XXX: even without periodic sampling we have to +* specify at least one sample layout property... +*/ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , default_test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (default_test_set->perf_oa_format), +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exponent, +DRM_XE_OA_PROPERTY_OA_DISABLED , true, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; +uint32_t oa_report0[64 ]; +uint32_t oa_report1[64 ]; +uint32_t buf[128 ] = {0 }; +int ret; + + stream_fd =__perf_open (drm_fd, ¶m, false); + + ret =read (stream_fd, buf,sizeof (buf)); +igt_assert_eq (ret,-1 ); +igt_assert_eq (errno,EIO ); + +__perf_close (stream_fd); + + properties[ARRAY_SIZE (properties) -1 ] = false;/* Set DISABLED to false */ + stream_fd =__perf_open (drm_fd, ¶m, false); +set_fd_flags (stream_fd,O_CLOEXEC ); + +read_2_oa_reports (default_test_set->perf_oa_format, + oa_exponent, + oa_report0, + oa_report1, + false);/* not just timer reports */ + +do_ioctl (stream_fd,DRM_XE_PERF_IOCTL_DISABLE ,0 ); + + ret =read (stream_fd, buf,sizeof (buf)); +igt_assert_eq (ret,-1 ); +igt_assert_eq (errno,EIO ); + +do_ioctl (stream_fd,DRM_XE_PERF_IOCTL_ENABLE ,0 ); + +read_2_oa_reports (default_test_set->perf_oa_format, + oa_exponent, + oa_report0, + oa_report1, + false);/* not just timer reports */ + +__perf_close (stream_fd); +} +static unsigned read_xe_module_ref (void ) {FILE *fp =fopen ("/proc/modules" ,"r" );
@@ -1986,6 +2557,27 @@ igt_main__for_one_hwe_in_oag (hwe)test_oa_exponents (hwe); + igt_subtest_with_dynamic("buffer-fill") +__for_one_hwe_in_oag (hwe) +test_buffer_fill (hwe); + + igt_subtest_with_dynamic("non-zero-reason") { +__for_one_hwe_in_oag (hwe) +test_non_zero_reason (hwe); + } + + igt_subtest("disabled-read-error") +test_disabled_read_error (); + igt_subtest("non-sampling-read-error") +test_non_sampling_read_error (); + + igt_subtest_with_dynamic("enable-disable") +__for_one_hwe_in_oag (hwe) +test_enable_disable (hwe); + + igt_subtest("short-reads") +test_short_reads (); + igt_fixture {/* leave sysctl options in their default state... */ write_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ,1 );
Signature · 1 line
2.41.0
PATCH 13/18 tests/intel/xe_oa: blocking and polling tests
In reply to Ashutosh Dixit"blocking", "polling" and "polling-small-buf".
--- tests/intel/xe_oa.c | 465 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 465 insertions(+)
tests/intel/xe_oa.c +465 -0
@@ -471,6 +471,24 @@ static struct drm_xe_engine_class_instance *oa_unit_engine(int fd, int n)return hwe; } +static char * +pretty_print_oa_period (uint64_t oa_period_ns) +{ +static char result[100 ]; +static const char *units[4 ] = {"ns" ,"us" ,"ms" ,"s" }; +double val = oa_period_ns; +int iter =0 ; + +while (iter < (ARRAY_SIZE (units) -1 ) && + val >=1000.0f ) { + val /=1000.0f ; + iter++; + } + +snprintf (result,sizeof (result),"%.3f%s" , val, units[iter]); +return result; +} +static void __perf_close (int fd) {
@@ -1759,6 +1777,434 @@ static void test_invalid_oa_exponent(void) } } +static int64_t +get_time (void ) +{ +struct timespec ts; + +clock_gettime (CLOCK_MONOTONIC , &ts); + +return ts.tv_sec *1000000000 + ts.tv_nsec; +} + +/** +* SUBTEST: blocking +* Description: Test blocking reads +*/ +/* Note: The interface doesn't currently provide strict guarantees or control +* over the upper bound for how long it might take for a POLLIN event after +* some OA report is written by the OA unit. +* +* The plan is to add a property later that gives some control over the maximum +* latency, but for now we expect it is tuned for a fairly low latency +* suitable for applications wanting to provide live feedback for captured +* metrics. +* +* At the time of writing this test the driver was using a fixed 200Hz hrtimer +* regardless of the OA sampling exponent. +* +* There is no lower bound since a stream configured for periodic sampling may +* still contain other automatically triggered reports. +* +* What we try and check for here is that blocking reads don't return EAGAIN +* and that we aren't spending any significant time burning the cpu in +* kernelspace. +*/ +static void test_blocking (uint64_t requested_oa_period, +bool set_kernel_hrtimer, +uint64_t kernel_hrtimer, +const struct drm_xe_engine_class_instance *hwe) +{ +int oa_exponent =max_oa_exponent_for_period_lte (requested_oa_period); +uint64_t oa_period =oa_exponent_to_ns (oa_exponent); +uint64_t props[DRM_XE_OA_PROPERTY_MAX *2 ]; +uint64_t *idx = props; +struct drm_xe_oa_open_prop param; +uint8_t buf[1024 *1024 ]; +struct tms start_times; +struct tms end_times; +int64_t user_ns, kernel_ns; +int64_t tick_ns =1000000000 /sysconf (_SC_CLK_TCK); +int64_t test_duration_ns = tick_ns *100 ; +int max_iterations = (test_duration_ns / oa_period) +2 ; +int n_extra_iterations =0 ; +int perf_fd; + +/* It's a bit tricky to put a lower limit here, but we expect a +* relatively low latency for seeing reports, while we don't currently +* give any control over this in the api. +* +* We assume a maximum latency of 6 millisecond to deliver a POLLIN and +* read() after a new sample is written (46ms per iteration) considering +* the knowledge that that the driver uses a 200Hz hrtimer (5ms period) +* to check for data and giving some time to read(). +*/ +int min_iterations = (test_duration_ns / (oa_period + kernel_hrtimer + kernel_hrtimer /5 )); +int64_t start, end; +int n =0 ; +struct intel_perf_metric_set *test_set =metric_set (hwe); +size_t format_size =get_oa_format (test_set->perf_oa_format).size; + +ADD_PROPS (props, idx,SAMPLE_OA , true); +ADD_PROPS (props, idx,OA_METRIC_SET , test_set->perf_oa_metrics_set); +ADD_PROPS (props, idx,OA_FORMAT ,__ff (test_set->perf_oa_format)); +ADD_PROPS (props, idx,OA_EXPONENT , oa_exponent); +ADD_PROPS (props, idx,OA_DISABLED , true); +ADD_PROPS (props, idx,OA_UNIT_ID ,0 ); +ADD_PROPS (props, idx,OA_ENGINE_INSTANCE , hwe->engine_instance); + + param.num_properties = (idx - props) /2 ; + param.properties_ptr =to_user_pointer (props); + + perf_fd =__perf_open (drm_fd, ¶m, true/* prevent_pm */ ); +set_fd_flags (perf_fd,O_CLOEXEC ); + +times (&start_times); + +igt_debug ("tick length = %dns, test duration = %" PRIu64"ns, min iter. = %d," +" estimated max iter. = %d, oa_period = %s\n" , + (int )tick_ns, test_duration_ns, + min_iterations, max_iterations, +pretty_print_oa_period (oa_period)); + +/* In the loop we perform blocking polls while the HW is sampling at +* ~25Hz, with the expectation that we spend most of our time blocked +* in the kernel, and shouldn't be burning cpu cycles in the kernel in +* association with this process (verified by looking at stime before +* and after loop). +* +* We're looking to assert that less than 1% of the test duration is +* spent in the kernel dealing with polling and read()ing. +* +* The test runs for a relatively long time considering the very low +* resolution of stime in ticks of typically 10 milliseconds. Since we +* don't know the fractional part of tick values we read from userspace +* so our minimum threshold needs to be >= one tick since any +* measurement might really be +- tick_ns (assuming we effectively get +* floor(real_stime)). +* +* We Loop for 1000 x tick_ns so one tick corresponds to 0.1% +* +* Also enable the stream just before poll/read to minimize +* the error delta. +*/ + start =get_time (); +do_ioctl (perf_fd,DRM_XE_PERF_IOCTL_ENABLE ,0 ); +for (/* nop */ ; ((end =get_time ()) - start) < test_duration_ns;/* nop */ ) { +bool timer_report_read = false; +bool non_timer_report_read = false; +int ret; + +while ((ret =read (perf_fd, buf,sizeof (buf))) <0 && + errno ==EINTR ) + ; + +igt_assert (ret >0 ); + +for (int offset =0 ; offset < ret; offset += format_size) { +uint32_t *report = (void *)(buf + offset); + +if (oa_report_is_periodic (oa_exponent, report)) + timer_report_read = true; +else + non_timer_report_read = true; + } + +if (non_timer_report_read && !timer_report_read) + n_extra_iterations++; + + n++; + } + +times (&end_times); + +/* Using nanosecond units is fairly silly here, given the tick in- +* precision - ah well, it's consistent with the get_time() units. +*/ + user_ns = (end_times.tms_utime - start_times.tms_utime) * tick_ns; + kernel_ns = (end_times.tms_stime - start_times.tms_stime) * tick_ns; + +igt_debug ("%d blocking reads during test with %" PRIu64" Hz OA sampling (expect no more than %d)\n" , + n,NSEC_PER_SEC / oa_period, max_iterations); +igt_debug ("%d extra iterations seen, not related to periodic sampling (e.g. context switches)\n" , + n_extra_iterations); +igt_debug ("time in userspace = %" PRIu64"ns (+-%dns) (start utime = %d, end = %d)\n" , + user_ns, (int )tick_ns, + (int )start_times.tms_utime, (int )end_times.tms_utime); +igt_debug ("time in kernelspace = %" PRIu64"ns (+-%dns) (start stime = %d, end = %d)\n" , + kernel_ns, (int )tick_ns, + (int )start_times.tms_stime, (int )end_times.tms_stime); + +/* With completely broken blocking (but also not returning an error) we +* could end up with an open loop, +*/ +igt_assert (n <= (max_iterations + n_extra_iterations)); + +/* Make sure the driver is reporting new samples with a reasonably +* low latency... +*/ +igt_assert (n > (min_iterations + n_extra_iterations)); + +if (!set_kernel_hrtimer) +igt_assert (kernel_ns <= (test_duration_ns /100ull )); + +__perf_close (perf_fd); +} + +/** +* SUBTEST: polling +* Description: Test polled reads +*/ +static void test_polling (uint64_t requested_oa_period, +bool set_kernel_hrtimer, +uint64_t kernel_hrtimer, +const struct drm_xe_engine_class_instance *hwe) +{ +int oa_exponent =max_oa_exponent_for_period_lte (requested_oa_period); +uint64_t oa_period =oa_exponent_to_ns (oa_exponent); +uint64_t props[DRM_XE_OA_PROPERTY_MAX *2 ]; +uint64_t *idx = props; +struct drm_xe_oa_open_prop param; +uint8_t buf[1024 *1024 ]; +struct tms start_times; +struct tms end_times; +int64_t user_ns, kernel_ns; +int64_t tick_ns =1000000000 /sysconf (_SC_CLK_TCK); +int64_t test_duration_ns = tick_ns *100 ; + +int max_iterations = (test_duration_ns / oa_period) +2 ; +int n_extra_iterations =0 ; + +/* It's a bit tricky to put a lower limit here, but we expect a +* relatively low latency for seeing reports. +* +* We assume a maximum latency of kernel_hrtimer + some margin +* to deliver a POLLIN and read() after a new sample is +* written (40ms + hrtimer + margin per iteration) considering +* the knowledge that that the driver uses a 200Hz hrtimer +* (5ms period) to check for data and giving some time to +* read(). +*/ +int min_iterations = (test_duration_ns / (oa_period + (kernel_hrtimer + kernel_hrtimer /5 ))); +int64_t start, end; +int n =0 ; +struct intel_perf_metric_set *test_set =metric_set (hwe); +size_t format_size =get_oa_format (test_set->perf_oa_format).size; + +ADD_PROPS (props, idx,SAMPLE_OA , true); +ADD_PROPS (props, idx,OA_METRIC_SET , test_set->perf_oa_metrics_set); +ADD_PROPS (props, idx,OA_FORMAT ,__ff (test_set->perf_oa_format)); +ADD_PROPS (props, idx,OA_EXPONENT , oa_exponent); +ADD_PROPS (props, idx,OA_DISABLED , true); +ADD_PROPS (props, idx,OA_UNIT_ID ,0 ); +ADD_PROPS (props, idx,OA_ENGINE_INSTANCE , hwe->engine_instance); + + param.num_properties = (idx - props) /2 ; + param.properties_ptr =to_user_pointer (props); + + stream_fd =__perf_open (drm_fd, ¶m, true/* prevent_pm */ ); +set_fd_flags (stream_fd,O_CLOEXEC |O_NONBLOCK ); + +times (&start_times); + +igt_debug ("tick length = %dns, oa period = %s, " +"test duration = %" PRIu64"ns, min iter. = %d, max iter. = %d\n" , + (int )tick_ns,pretty_print_oa_period (oa_period), test_duration_ns, + min_iterations, max_iterations); + +/* In the loop we perform blocking polls while the HW is sampling at +* ~25Hz, with the expectation that we spend most of our time blocked +* in the kernel, and shouldn't be burning cpu cycles in the kernel in +* association with this process (verified by looking at stime before +* and after loop). +* +* We're looking to assert that less than 1% of the test duration is +* spent in the kernel dealing with polling and read()ing. +* +* The test runs for a relatively long time considering the very low +* resolution of stime in ticks of typically 10 milliseconds. Since we +* don't know the fractional part of tick values we read from userspace +* so our minimum threshold needs to be >= one tick since any +* measurement might really be +- tick_ns (assuming we effectively get +* floor(real_stime)). +* +* We Loop for 1000 x tick_ns so one tick corresponds to 0.1% +* +* Also enable the stream just before poll/read to minimize +* the error delta. +*/ + start =get_time (); +do_ioctl (stream_fd,DRM_XE_PERF_IOCTL_ENABLE ,0 ); +for (/* nop */ ; ((end =get_time ()) - start) < test_duration_ns;/* nop */ ) { +struct pollfd pollfd = { .fd = stream_fd, .events =POLLIN }; +bool timer_report_read = false; +bool non_timer_report_read = false; +int ret; + +while ((ret =poll (&pollfd,1 ,-1 )) <0 && + errno ==EINTR ) + ; +igt_assert_eq (ret,1 ); +igt_assert (pollfd.revents &POLLIN ); + +while ((ret =read (stream_fd, buf,sizeof (buf))) <0 && + errno ==EINTR ) + ; + +/* Don't expect to see EAGAIN if we've had a POLLIN event +* +* XXX: actually this is technically overly strict since we do +* knowingly allow false positive POLLIN events. At least in +* the future when supporting context filtering of metrics for +* Gen8+ handled in the kernel then POLLIN events may be +* delivered when we know there are pending reports to process +* but before we've done any filtering to know for certain that +* any reports are destined to be copied to userspace. +* +* Still, for now it's a reasonable sanity check. +*/ +if (ret <0 ) +igt_debug ("Unexpected error when reading after poll = %d\n" , errno); +igt_assert_neq (ret,-1 ); + +/* For Haswell reports don't contain a well defined reason +* field we so assume all reports to be 'periodic'. For gen8+ +* we want to to consider that the HW automatically writes some +* non periodic reports (e.g. on context switch) which might +* lead to more successful read()s than expected due to +* periodic sampling and we don't want these extra reads to +* cause the test to fail... +*/ +for (int offset =0 ; offset < ret; offset += format_size) { +uint32_t *report = (void *)(buf + offset); + +if (oa_report_is_periodic (oa_exponent, report)) + timer_report_read = true; +else + non_timer_report_read = true; + } + +if (non_timer_report_read && !timer_report_read) + n_extra_iterations++; + +/* At this point, after consuming pending reports (and hoping +* the scheduler hasn't stopped us for too long) we now expect +* EAGAIN on read. While this works most of the times, there are +* some rare failures when the OA period passed to this test is +* very small (say 500 us) and that results in some valid +* reports here. To weed out those rare occurences we assert +* only if the OA period is >= 40 ms because 40 ms has withstood +* the test of time on most platforms (ref: subtest: polling). +*/ +while ((ret =read (stream_fd, buf,sizeof (buf))) <0 && + errno ==EINTR ) + ; + +if (requested_oa_period >=40000000 ) { +igt_assert_eq (ret,-1 ); +igt_assert_eq (errno,EAGAIN ); + } + + n++; + } + +times (&end_times); + +/* Using nanosecond units is fairly silly here, given the tick in- +* precision - ah well, it's consistent with the get_time() units. +*/ + user_ns = (end_times.tms_utime - start_times.tms_utime) * tick_ns; + kernel_ns = (end_times.tms_stime - start_times.tms_stime) * tick_ns; + +igt_debug ("%d non-blocking reads during test with %" PRIu64" Hz OA sampling (expect no more than %d)\n" , + n,NSEC_PER_SEC / oa_period, max_iterations); +igt_debug ("%d extra iterations seen, not related to periodic sampling (e.g. context switches)\n" , + n_extra_iterations); +igt_debug ("time in userspace = %" PRIu64"ns (+-%dns) (start utime = %d, end = %d)\n" , + user_ns, (int )tick_ns, + (int )start_times.tms_utime, (int )end_times.tms_utime); +igt_debug ("time in kernelspace = %" PRIu64"ns (+-%dns) (start stime = %d, end = %d)\n" , + kernel_ns, (int )tick_ns, + (int )start_times.tms_stime, (int )end_times.tms_stime); + +/* With completely broken blocking while polling (but still somehow +* reporting a POLLIN event) we could end up with an open loop. +*/ +igt_assert (n <= (max_iterations + n_extra_iterations)); + +/* Make sure the driver is reporting new samples with a reasonably +* low latency... +*/ +igt_assert (n > (min_iterations + n_extra_iterations)); + +if (!set_kernel_hrtimer) +igt_assert (kernel_ns <= (test_duration_ns /100ull )); + +__perf_close (stream_fd); +} + +/** +* SUBTEST: polling-small-buf +* Description: Test polled read with buffer size smaller than available data +*/ +static void test_polling_small_buf (void ) +{ +int oa_exponent =max_oa_exponent_for_period_lte (40 *1000 );/* 40us */ +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* Include OA reports in samples */ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , default_test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (default_test_set->perf_oa_format), +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exponent, +DRM_XE_OA_PROPERTY_OA_DISABLED , true, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; +uint32_t test_duration =80 *1000 *1000 ; +int sample_size =get_oa_format (default_test_set->perf_oa_format).size; +int n_expected_reports = test_duration /oa_exponent_to_ns (oa_exponent); +int n_expect_read_bytes = n_expected_reports * sample_size; +struct timespec ts = {}; +int n_bytes_read =0 ; +uint32_t n_polls =0 ; + + stream_fd =__perf_open (drm_fd, ¶m, true/* prevent_pm */ ); +set_fd_flags (stream_fd,O_CLOEXEC |O_NONBLOCK ); +do_ioctl (stream_fd,DRM_XE_PERF_IOCTL_ENABLE ,0 ); + +while (igt_nsec_elapsed (&ts) < test_duration) { +struct pollfd pollfd = { .fd = stream_fd, .events =POLLIN }; + +ppoll (&pollfd,1 ,NULL ,NULL ); +if (pollfd.revents &POLLIN ) { +uint8_t buf[1024 ]; +int ret; + + ret =read (stream_fd, buf,sizeof (buf)); +if (ret >=0 ) + n_bytes_read += ret; + } + + n_polls++; + } + +igt_info ("Read %d expected %d (%.2f%% of the expected number), polls=%u\n" , + n_bytes_read, n_expect_read_bytes, + n_bytes_read *100.0f / n_expect_read_bytes, + n_polls); + +__perf_close (stream_fd); + +igt_assert (abs (n_expect_read_bytes - n_bytes_read) < +0.20 * n_expect_read_bytes); +} +/** * SUBTEST: buffer-fill * Description: Test filling, wraparound and overflow of OA buffer
@@ -2575,6 +3021,25 @@ igt_main__for_one_hwe_in_oag (hwe)test_enable_disable (hwe); + igt_subtest_with_dynamic("blocking") { + __for_one_hwe_in_oag(hwe) + test_blocking(40 * 1000 * 1000 /* 40ms oa period */, + false/* set_kernel_hrtimer */ , +5 *1000 *1000 /* default 5ms/200Hz hrtimer */ , + hwe); + } + + igt_subtest_with_dynamic("polling") { + __for_one_hwe_in_oag(hwe) + test_polling(40 * 1000 * 1000 /* 40ms oa period */, + false/* set_kernel_hrtimer */ , +5 *1000 *1000 /* default 5ms/200Hz hrtimer */ , + hwe); + } + + igt_subtest("polling-small-buf") +test_polling_small_buf (); + igt_subtest("short-reads")test_short_reads ();
Signature · 1 line
2.41.0
PATCH 14/18 tests/intel/xe_oa: OAR/OAC tests
In reply to Ashutosh Dixit"mi-rpc", "oa-tlb-invalidate" and "unprivileged-single-ctx-counters" tests.
--- tests/intel/xe_oa.c | 698 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 698 insertions(+)
tests/intel/xe_oa.c +698 -0
@@ -637,6 +637,18 @@ read_report_reason(const uint32_t *report)return "unknown" ; } +static uint32_t +cs_timestamp_frequency (int fd) +{ +return xe_gt_list (drm_fd)->gt_list[0 ].reference_clock; +} + +static uint64_t +cs_timebase_scale (uint32_t u32_delta) +{ +return ((uint64_t )u32_delta *NSEC_PER_SEC ) /cs_timestamp_frequency (drm_fd); +} +static uint64_t oa_timestamp (const uint32_t *report,enum xe_oa_format_name format) {
@@ -703,6 +715,15 @@ oa_report_get_ctx_id(uint32_t *report)return report[2 ]; } +static int +oar_unit_default_format (void ) +{ +if (IS_DG2 (devid) ||IS_METEORLAKE (devid)) +return XE_OAR_FORMAT_A32u40_A4u32_B8_C8; + +return default_test_set->perf_oa_format; +} +static void *buf_map (int fd,struct intel_buf *buf,bool write) {void *p;
@@ -741,6 +762,21 @@ scratch_buf_init(struct buf_ops *bops,scratch_buf_memset (buf, width, height, color); } +static void +emit_report_perf_count (struct intel_bb *ibb, +struct intel_buf *dst, +int dst_offset, +uint32_t report_id) +{ +intel_bb_add_intel_buf (ibb, dst, true); + +intel_bb_out (ibb,OA_MI_REPORT_PERF_COUNT ); +intel_bb_emit_reloc (ibb, dst->handle, +I915_GEM_DOMAIN_INSTRUCTION ,I915_GEM_DOMAIN_INSTRUCTION , + dst_offset, dst->addr.offset); +intel_bb_out (ibb, report_id); +} +static bool oa_report_is_periodic (uint32_t oa_exponent,const uint32_t *report) {
@@ -781,6 +817,31 @@ get_40bit_a_delta(uint64_t value0, uint64_t value1)return value1 - value0; } +static void +accumulate_uint32 (size_t offset, +uint32_t *report0, +uint32_t *report1, +uint64_t *delta) +{ +uint32_t value0 = *(uint32_t *)(((uint8_t *)report0) + offset); +uint32_t value1 = *(uint32_t *)(((uint8_t *)report1) + offset); + + *delta += (uint32_t )(value1 - value0); +} + +static void +accumulate_uint40 (int a_index, +uint32_t *report0, +uint32_t *report1, +enum xe_oa_format_name format, +uint64_t *delta) +{ +uint64_t value0 =read_40bit_a_counter (report0, format, a_index), + value1 =read_40bit_a_counter (report1, format, a_index); + + *delta +=get_40bit_a_delta (value0, value1); +} + static void accumulate_uint64(int a_index, const uint32_t *report0,
@@ -794,6 +855,78 @@ accumulate_uint64(int a_index, *delta += (value1 - value0); } +static void +accumulate_reports (struct accumulator *accumulator, +uint32_t *start, +uint32_t *end) +{ +struct oa_format format =get_oa_format (accumulator->format); +uint64_t *deltas = accumulator->deltas; +int idx =0 ; + +/* timestamp */ + deltas[idx] +=oa_timestamp_delta (end, start, accumulator->format); + idx++; + +/* clock cycles */ + deltas[idx] +=oa_tick_delta (end, start, accumulator->format); + idx++; + +for (int i =0 ; i < format.n_a40; i++) { +accumulate_uint40 (i, start, end, accumulator->format, + deltas + idx++); + } + +for (int i =0 ; i < format.n_a64; i++) { +accumulate_uint64 (i, start, end, accumulator->format, + deltas + idx++); + } + +for (int i =0 ; i < format.n_a; i++) { +accumulate_uint32 (format.a_off +4 * i, + start, end, deltas + idx++); + } + +for (int i =0 ; i < format.n_b; i++) { +accumulate_uint32 (format.b_off +4 * i, + start, end, deltas + idx++); + } + +for (int i =0 ; i < format.n_c; i++) { +accumulate_uint32 (format.c_off +4 * i, + start, end, deltas + idx++); + } +} + +static void +accumulator_print (struct accumulator *accumulator,const char *title) +{ +struct oa_format format =get_oa_format (accumulator->format); +uint64_t *deltas = accumulator->deltas; +int idx =0 ; + +igt_debug ("%s:\n" , title); +igt_debug ("\ttime delta = %" PRIu64"\n" , deltas[idx++]); +igt_debug ("\tclock cycle delta = %" PRIu64"\n" , deltas[idx++]); + +for (int i =0 ; i < format.n_a40; i++) +igt_debug ("\tA%u = %" PRIu64"\n" , i, deltas[idx++]); + +for (int i =0 ; i < format.n_a64; i++) +igt_debug ("\tA64_%u = %" PRIu64"\n" , i, deltas[idx++]); + +for (int i =0 ; i < format.n_a; i++) { +int a_id = format.first_a + i; +igt_debug ("\tA%u = %" PRIu64"\n" , a_id, deltas[idx++]); + } + +for (int i =0 ; i < format.n_a; i++) +igt_debug ("\tB%u = %" PRIu64"\n" , i, deltas[idx++]); + +for (int i =0 ; i < format.n_c; i++) +igt_debug ("\tC%u = %" PRIu64"\n" , i, deltas[idx++]); +} +/* The TestOa metric set is designed so */ static void sanity_check_reports(const uint32_t *oa_report0, const uint32_t *oa_report1,
@@ -2205,6 +2338,92 @@ static void test_polling_small_buf(void) 0.20 * n_expect_read_bytes); } +static int +num_valid_reports_captured (struct drm_xe_oa_open_prop *param, +int64_t *duration_ns,int fmt) +{ +uint8_t buf[1024 *1024 ]; +int64_t start, end; +int num_reports =0 ; +size_t format_size =get_oa_format (fmt).size; + +igt_debug ("Expected duration = %" PRId64"\n" , *duration_ns); + + stream_fd =__perf_open (drm_fd, param, true); + + start =get_time (); +do_ioctl (stream_fd,DRM_XE_PERF_IOCTL_ENABLE ,0 ); +for (/* nop */ ; ((end =get_time ()) - start) < *duration_ns;/* nop */ ) { +int ret; + +while ((ret =read (stream_fd, buf,sizeof (buf))) <0 && + errno ==EINTR ) + ; + +igt_assert (ret >0 ); + +for (int offset =0 ; offset < ret; offset += format_size) { +uint32_t *report = (void *)(buf + offset); + +if (report_reason (report) &OAREPORT_REASON_TIMER ) + num_reports++; + } + } +__perf_close (stream_fd); + + *duration_ns = end - start; + +igt_debug ("Actual duration = %" PRIu64"\n" , *duration_ns); + +return num_reports; +} + +/** +* SUBTEST: oa-tlb-invalidate +* Description: Open OA stream twice to verify OA TLB invalidation +*/ +static void +test_oa_tlb_invalidate (const struct drm_xe_engine_class_instance *hwe) +{ +int oa_exponent =max_oa_exponent_for_period_lte (30000000 ); +struct intel_perf_metric_set *test_set =metric_set (hwe); +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +DRM_XE_OA_PROPERTY_OA_METRIC_SET , test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (test_set->perf_oa_format), +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exponent, +DRM_XE_OA_PROPERTY_OA_DISABLED , true, +DRM_XE_OA_PROPERTY_OA_ENGINE_INSTANCE , hwe->engine_instance, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; +int num_reports1, num_reports2, num_expected_reports; +int64_t duration; + +/* Capture reports for 5 seconds twice and then make sure you get around +* the same number of reports. In the case of failure, the number of +* reports will vary largely since the beginning of the OA buffer +* will have invalid entries. +*/ + duration =5LL *NSEC_PER_SEC ; + num_reports1 =num_valid_reports_captured (¶m, &duration, test_set->perf_oa_format); + num_expected_reports = duration /oa_exponent_to_ns (oa_exponent); +igt_debug ("expected num reports = %d\n" , num_expected_reports); +igt_debug ("actual num reports = %d\n" , num_reports1); +igt_assert (num_reports1 >0.95 * num_expected_reports); + + duration =5LL *NSEC_PER_SEC ; + num_reports2 =num_valid_reports_captured (¶m, &duration, test_set->perf_oa_format); + num_expected_reports = duration /oa_exponent_to_ns (oa_exponent); +igt_debug ("expected num reports = %d\n" , num_expected_reports); +igt_debug ("actual num reports = %d\n" , num_reports2); +igt_assert (num_reports2 >0.95 * num_expected_reports); +} +/** * SUBTEST: buffer-fill * Description: Test filling, wraparound and overflow of OA buffer
@@ -2776,6 +2995,468 @@ test_disabled_read_error(void)__perf_close (stream_fd); } +/** +* SUBTEST: mi-rpc +* Description: Test OAR/OAC using MI_REPORT_PERF_COUNT +*/ +static void +test_mi_rpc (struct drm_xe_engine_class_instance *hwe) + +{ +uint64_t fmt = ((IS_DG2 (devid) ||IS_METEORLAKE (devid)) && + hwe->engine_class ==DRM_XE_ENGINE_CLASS_COMPUTE ) ? + XE_OAC_FORMAT_A24u64_B8_C8 :oar_unit_default_format (); +struct intel_perf_metric_set *test_set =metric_set (hwe); +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* On Gen12, MI RPC uses OAR. OAR is configured only for the +* render context that wants to measure the performance. Hence a +* context must be specified in the gen12 MI RPC when compared +* to previous gens. +* +* Have a random value here for the context id, but initialize +* it once you figure out the context ID for the work to be +* measured +*/ +DRM_XE_OA_PROPERTY_EXEC_QUEUE_ID ,UINT64_MAX , + +/* OA unit configuration: +* DRM_XE_OA_PROPERTY_SAMPLE_OA is no longer required for Gen12 +* because the OAR unit increments counters only for the +* relevant context. No other parameters are needed since we do +* not rely on the OA buffer anymore to normalize the counter +* values. +*/ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (fmt), +DRM_XE_OA_PROPERTY_OA_ENGINE_INSTANCE , hwe->engine_instance, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; +struct buf_ops *bops; +struct intel_bb *ibb; +struct intel_buf *buf; +#define INVALID_CTX_ID 0xffffffff +uint32_t ctx_id =INVALID_CTX_ID ; +uint32_t vm =0 ; +uint32_t *report32; +size_t format_size_32; +struct oa_format format =get_oa_format (fmt); + +/* Ensure perf_stream_paranoid is set to 1 by default */ +write_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ,1 ); + + bops =buf_ops_create (drm_fd); + vm =xe_vm_create (drm_fd,0 ,0 ); + ctx_id =xe_exec_queue_create (drm_fd, vm, hwe,0 ); +igt_assert_neq (ctx_id,INVALID_CTX_ID ); + properties[3 ] = ctx_id; + + ibb =intel_bb_create_with_context (drm_fd, ctx_id, vm,NULL ,BATCH_SZ ); + buf =intel_buf_create (bops,4096 ,1 ,8 ,64 , +I915_TILING_NONE ,I915_COMPRESSION_NONE ); + +buf_map (drm_fd, buf, true); +memset (buf->ptr,0x80 ,4096 ); +intel_buf_unmap (buf); + + stream_fd =__perf_open (drm_fd, ¶m, false); +set_fd_flags (stream_fd,O_CLOEXEC ); + +#define REPORT_ID 0xdeadbeef +#define REPORT_OFFSET 0 +emit_report_perf_count (ibb, + buf, +REPORT_OFFSET , +REPORT_ID ); +intel_bb_flush_render (ibb); +intel_bb_sync (ibb); + +buf_map (drm_fd, buf, false); + report32 = buf->ptr; + format_size_32 = format.size >>2 ; +dump_report (report32, format_size_32,"mi-rpc" ); + +/* Sanity check reports +* reportX_32[0]: report id passed with mi-rpc +* reportX_32[1]: timestamp. NOTE: wraps around in ~6 minutes. +* +* reportX_32[format.b_off]: check if the entire report was filled. +* B0 counter falls in the last 64 bytes of this report format. +* Since reports are filled in 64 byte blocks, we should be able to +* assure that the report was filled by checking the B0 counter. B0 +* counter is defined to be zero, so we can easily validate it. +* +* reportX_32[format_size_32]: outside report, make sure only the report +* size amount of data was written. +*/ +igt_assert_eq (report32[0 ],REPORT_ID ); +igt_assert (oa_timestamp (report32, test_set->perf_oa_format)); +igt_assert_neq (report32[format.b_off >>2 ],0x80808080 ); +igt_assert_eq (report32[format_size_32],0x80808080 ); + +intel_buf_unmap (buf); +intel_buf_destroy (buf); +intel_bb_destroy (ibb); +xe_exec_queue_destroy (drm_fd, ctx_id); +xe_vm_destroy (drm_fd, vm); +buf_ops_destroy (bops); +__perf_close (stream_fd); +} + +static void +emit_stall_timestamp_and_rpc (struct intel_bb *ibb, +struct intel_buf *dst, +int timestamp_offset, +int report_dst_offset, +uint32_t report_id) +{ +uint32_t pipe_ctl_flags = (PIPE_CONTROL_CS_STALL | +PIPE_CONTROL_RENDER_TARGET_FLUSH | +PIPE_CONTROL_WRITE_TIMESTAMP ); + +intel_bb_add_intel_buf (ibb, dst, true); +intel_bb_out (ibb,GFX_OP_PIPE_CONTROL (6 )); +intel_bb_out (ibb, pipe_ctl_flags); +intel_bb_emit_reloc (ibb, dst->handle, +I915_GEM_DOMAIN_INSTRUCTION ,I915_GEM_DOMAIN_INSTRUCTION , + timestamp_offset, dst->addr.offset); +intel_bb_out (ibb,0 );/* imm lower */ +intel_bb_out (ibb,0 );/* imm upper */ + +emit_report_perf_count (ibb, dst, report_dst_offset, report_id); +} + +static void single_ctx_helper (struct drm_xe_engine_class_instance *hwe) +{ +struct intel_perf_metric_set *test_set =metric_set (hwe); +uint64_t fmt =oar_unit_default_format (); +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* Have a random value here for the context id, but initialize +* it once you figure out the context ID for the work to be +* measured +*/ +DRM_XE_OA_PROPERTY_EXEC_QUEUE_ID ,UINT64_MAX , + +/* OA unit configuration: +* DRM_XE_OA_PROPERTY_SAMPLE_OA is no longer required for Gen12 +* because the OAR unit increments counters only for the +* relevant context. No other parameters are needed since we do +* not rely on the OA buffer anymore to normalize the counter +* values. +*/ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (fmt), +DRM_XE_OA_PROPERTY_OA_ENGINE_INSTANCE , hwe->engine_instance, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; +struct buf_ops *bops; +struct intel_bb *ibb0, *ibb1; +struct intel_buf src[3 ], dst[3 ], *dst_buf; +uint32_t context0_id, context1_id, vm =0 ; +uint32_t *report0_32, *report1_32, *report2_32, *report3_32; +uint64_t timestamp0_64, timestamp1_64; +uint64_t delta_ts64, delta_oa32; +uint64_t delta_ts64_ns, delta_oa32_ns; +uint64_t delta_delta; +int width =800 ; +int height =600 ; +#define INVALID_CTX_ID 0xffffffff +uint32_t ctx0_id =INVALID_CTX_ID ; +uint32_t ctx1_id =INVALID_CTX_ID ; +int ret; +struct accumulator accumulator = { + .format = fmt + }; + + bops =buf_ops_create (drm_fd); + +for (int i =0 ; i <ARRAY_SIZE (src); i++) { +scratch_buf_init (bops, &src[i], width, height,0xff0000ff ); +scratch_buf_init (bops, &dst[i], width, height,0x00ff00ff ); + } + + vm =xe_vm_create (drm_fd,0 ,0 ); + context0_id =xe_exec_queue_create (drm_fd, vm, hwe,0 ); + context1_id =xe_exec_queue_create (drm_fd, vm, hwe,0 ); + ibb0 =intel_bb_create_with_context (drm_fd, context0_id, vm,NULL ,BATCH_SZ ); + ibb1 =intel_bb_create_with_context (drm_fd, context1_id, vm,NULL ,BATCH_SZ ); + +igt_debug ("submitting warm up render_copy\n" ); + +/* Submit some early, unmeasured, work to the context we want */ +render_copy (ibb0, + &src[0 ],0 ,0 , width, height, + &dst[0 ],0 ,0 ); + +/* Initialize the context parameter to the perf open ioctl here */ + properties[3 ] = context0_id; + +igt_debug ("opening xe oa stream\n" ); + stream_fd =__perf_open (drm_fd, ¶m, false); +set_fd_flags (stream_fd,O_CLOEXEC ); + + dst_buf =intel_buf_create (bops,4096 ,1 ,8 ,64 , +I915_TILING_NONE , +I915_COMPRESSION_NONE ); + +/* Set write domain to cpu briefly to fill the buffer with 80s */ +buf_map (drm_fd, dst_buf, true/* write enable */ ); +memset (dst_buf->ptr,0x80 ,2048 ); +memset ((uint8_t *) dst_buf->ptr +2048 ,0 ,2048 ); +intel_buf_unmap (dst_buf); + +/* Submit an mi-rpc to context0 before measurable work */ +#define BO_TIMESTAMP_OFFSET0 1024 +#define BO_REPORT_OFFSET0 0 +#define BO_REPORT_ID0 0xdeadbeef +emit_stall_timestamp_and_rpc (ibb0, + dst_buf, +BO_TIMESTAMP_OFFSET0 , +BO_REPORT_OFFSET0 , +BO_REPORT_ID0 ); +intel_bb_flush_render (ibb0); + +/* Remove intel_buf from ibb0 added implicitly in rendercopy */ +intel_bb_remove_intel_buf (ibb0, dst_buf); + +/* This is the work/context that is measured for counter increments */ +render_copy (ibb0, + &src[0 ],0 ,0 , width, height, + &dst[0 ],0 ,0 ); +intel_bb_flush_render (ibb0); + +/* Submit an mi-rpc to context1 before work +* +* On gen12, this measurement should just yield counters that are +* all zeroes, since the counters will only increment for the +* context passed to perf open ioctl +*/ +#define BO_TIMESTAMP_OFFSET2 1040 +#define BO_REPORT_OFFSET2 512 +#define BO_REPORT_ID2 0x00c0ffee +emit_stall_timestamp_and_rpc (ibb1, + dst_buf, +BO_TIMESTAMP_OFFSET2 , +BO_REPORT_OFFSET2 , +BO_REPORT_ID2 ); +intel_bb_flush_render (ibb1); + +/* Submit two copies on the other context to avoid a false +* positive in case the driver somehow ended up filtering for +* context1 +*/ +render_copy (ibb1, + &src[1 ],0 ,0 , width, height, + &dst[1 ],0 ,0 ); + +render_copy (ibb1, + &src[2 ],0 ,0 , width, height, + &dst[2 ],0 ,0 ); +intel_bb_flush_render (ibb1); + +/* Submit an mi-rpc to context1 after all work */ +#define BO_TIMESTAMP_OFFSET3 1048 +#define BO_REPORT_OFFSET3 768 +#define BO_REPORT_ID3 0x01c0ffee +emit_stall_timestamp_and_rpc (ibb1, + dst_buf, +BO_TIMESTAMP_OFFSET3 , +BO_REPORT_OFFSET3 , +BO_REPORT_ID3 ); +intel_bb_flush_render (ibb1); + +/* Remove intel_buf from ibb1 added implicitly in rendercopy */ +intel_bb_remove_intel_buf (ibb1, dst_buf); + +/* Submit an mi-rpc to context0 after all measurable work */ +#define BO_TIMESTAMP_OFFSET1 1032 +#define BO_REPORT_OFFSET1 256 +#define BO_REPORT_ID1 0xbeefbeef +emit_stall_timestamp_and_rpc (ibb0, + dst_buf, +BO_TIMESTAMP_OFFSET1 , +BO_REPORT_OFFSET1 , +BO_REPORT_ID1 ); +intel_bb_flush_render (ibb0); +intel_bb_sync (ibb0); +intel_bb_sync (ibb1); + +buf_map (drm_fd, dst_buf, false); + +/* Sanity check reports +* reportX_32[0]: report id passed with mi-rpc +* reportX_32[1]: timestamp +* reportX_32[2]: context id +* +* report0_32: start of measurable work +* report1_32: end of measurable work +* report2_32: start of other work +* report3_32: end of other work +*/ + report0_32 = dst_buf->ptr; +igt_assert_eq (report0_32[0 ],0xdeadbeef ); +igt_assert (oa_timestamp (report0_32, fmt)); + ctx0_id = report0_32[2 ]; +igt_debug ("MI_RPC(start) CTX ID: %u\n" , ctx0_id); +dump_report (report0_32,64 ,"report0_32" ); + + report1_32 = report0_32 +64 ; +igt_assert_eq (report1_32[0 ],0xbeefbeef ); +igt_assert (oa_timestamp (report1_32, fmt)); + ctx1_id = report1_32[2 ]; +igt_debug ("CTX ID1: %u\n" , ctx1_id); +dump_report (report1_32,64 ,"report1_32" ); + +/* Verify that counters in context1 are all zeroes */ + report2_32 = report0_32 +128 ; +igt_assert_eq (report2_32[0 ],0x00c0ffee ); +igt_assert (oa_timestamp (report2_32, fmt)); +dump_report (report2_32,64 ,"report2_32" ); +igt_assert_eq (0 ,memcmp (&report2_32[4 ], + (uint8_t *) dst_buf->ptr +2048 , +240 )); + + report3_32 = report0_32 +192 ; +igt_assert_eq (report3_32[0 ],0x01c0ffee ); +igt_assert (oa_timestamp (report3_32, fmt)); +dump_report (report3_32,64 ,"report3_32" ); +igt_assert_eq (0 ,memcmp (&report3_32[4 ], + (uint8_t *) dst_buf->ptr +2048 , +240 )); + +/* Accumulate deltas for counters - A0, A21 and A26 */ +memset (accumulator.deltas,0 ,sizeof (accumulator.deltas)); +accumulate_reports (&accumulator, report0_32, report1_32); +igt_debug ("total: A0 = %" PRIu64", A21 = %" PRIu64", A26 = %" PRIu64"\n" , + accumulator.deltas[2 +0 ], + accumulator.deltas[2 +21 ], + accumulator.deltas[2 +26 ]); + +igt_debug ("oa_timestamp32 0 = %" PRIu64"\n" ,oa_timestamp (report0_32, fmt)); +igt_debug ("oa_timestamp32 1 = %" PRIu64"\n" ,oa_timestamp (report1_32, fmt)); +igt_debug ("ctx_id 0 = %u\n" , report0_32[2 ]); +igt_debug ("ctx_id 1 = %u\n" , report1_32[2 ]); + +/* The delta as calculated via the PIPE_CONTROL timestamp or +* the OA report timestamps should be almost identical but +* allow a 500 nanoseconds margin. +*/ + timestamp0_64 = *(uint64_t *)(((uint8_t *)dst_buf->ptr) +BO_TIMESTAMP_OFFSET0 ); + timestamp1_64 = *(uint64_t *)(((uint8_t *)dst_buf->ptr) +BO_TIMESTAMP_OFFSET1 ); + +igt_debug ("ts_timestamp64 0 = %" PRIu64"\n" , timestamp0_64); +igt_debug ("ts_timestamp64 1 = %" PRIu64"\n" , timestamp1_64); + + delta_ts64 = timestamp1_64 - timestamp0_64; + delta_oa32 =oa_timestamp_delta (report1_32, report0_32, fmt); + +/* Sanity check that we can pass the delta to timebase_scale */ + delta_oa32_ns =timebase_scale (delta_oa32); + delta_ts64_ns =cs_timebase_scale (delta_ts64); + +igt_debug ("oa32 delta = %" PRIu64", = %" PRIu64"ns\n" , + delta_oa32, delta_oa32_ns); +igt_debug ("ts64 delta = %" PRIu64", = %" PRIu64"ns\n" , + delta_ts64, delta_ts64_ns); + + delta_delta = delta_ts64_ns > delta_oa32_ns ? + (delta_ts64_ns - delta_oa32_ns) : + (delta_oa32_ns - delta_ts64_ns); +if (delta_delta >500 ) { +igt_debug ("delta_delta = %" PRIu64". exceeds margin, skipping..\n" , + delta_delta); +exit (EAGAIN ); + } + +igt_debug ("n samples written = %" PRIu64"/%" PRIu64" (%ix%i)\n" , + accumulator.deltas[2 +21 ], + accumulator.deltas[2 +26 ], + width, height); +accumulator_print (&accumulator,"filtered" ); + +/* Verify that the work actually happened by comparing the src +* and dst buffers +*/ +buf_map (drm_fd, &src[0 ], false); +buf_map (drm_fd, &dst[0 ], false); + + ret =memcmp (src[0 ].ptr, dst[0 ].ptr,4 * width * height); +intel_buf_unmap (&src[0 ]); +intel_buf_unmap (&dst[0 ]); + +if (ret !=0 ) { +accumulator_print (&accumulator,"total" ); +exit (EAGAIN ); + } + +/* FIXME: can we deduce the presence of A26 from get_oa_format(fmt)? */ +if (IS_LUNARLAKE (devid)) + goto skip_check; + +/* Check that this test passed. The test measures the number of 2x2 +* samples written to the render target using the counter A26. For +* OAR, this counter will only have increments relevant to this specific +* context. The value equals the width * height of the rendered work. +*/ +igt_assert_eq (accumulator.deltas[2 +26 ], width * height); + + skip_check: +/* Clean up */ +for (int i =0 ; i <ARRAY_SIZE (src); i++) { +intel_buf_close (bops, &src[i]); +intel_buf_close (bops, &dst[i]); + } + +intel_buf_unmap (dst_buf); +intel_buf_destroy (dst_buf); +intel_bb_destroy (ibb0); +intel_bb_destroy (ibb1); +xe_exec_queue_destroy (drm_fd, context0_id); +xe_exec_queue_destroy (drm_fd, context1_id); +xe_vm_destroy (drm_fd, vm); +buf_ops_destroy (bops); +__perf_close (stream_fd); +} + +/** +* SUBTEST: unprivileged-single-ctx-counters +* Description: A harder test for OAR/OAC using MI_REPORT_PERF_COUNT +*/ +static void +test_single_ctx_render_target_writes_a_counter (struct drm_xe_engine_class_instance *hwe) +{ +int child_ret; +struct igt_helper_process child = {}; + +/* Ensure perf_stream_paranoid is set to 1 by default */ +write_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ,1 ); + +do { + igt_fork_helper(&child) { +/* A local device for local resources. */ + drm_fd =drm_reopen_driver (drm_fd); + +igt_drop_root (); + +single_ctx_helper (hwe); + +drm_close_driver (drm_fd); + } + child_ret =igt_wait_helper (&child); +igt_assert (WEXITSTATUS (child_ret) ==EAGAIN || +WEXITSTATUS (child_ret) ==0 ); + }while (WEXITSTATUS (child_ret) ==EAGAIN ); +} +static unsigned read_xe_module_ref (void ) {FILE *fp =fopen ("/proc/modules" ,"r" );
@@ -3043,6 +3724,23 @@ igt_main igt_subtest("short-reads")test_short_reads (); + igt_subtest_group { + igt_subtest_with_dynamic("mi-rpc") +__for_one_hwe_in_oag (hwe) +test_mi_rpc (hwe); + + igt_subtest_with_dynamic("oa-tlb-invalidate") +__for_one_hwe_in_oag (hwe) +test_oa_tlb_invalidate (hwe); + + igt_subtest_with_dynamic("unprivileged-single-ctx-counters") { +igt_require_f (render_copy,"no render-copy function\n" ); +igt_require (!IS_LUNARLAKE (devid)); +__for_one_render_engine (hwe) +test_single_ctx_render_target_writes_a_counter (hwe); + } + } + igt_fixture {/* leave sysctl options in their default state... */ write_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ,1 );
Signature · 1 line
2.41.0
PATCH 15/18 tests/intel/xe_oa: Exclusive/concurrent access, rc6 and stress open close
In reply to Ashutosh DixitAdd the following tests:
* "oa-unit-exclusive-stream-sample-oa"
* "oa-unit-exclusive-stream-exec-q"
* "oa-unit-concurrent-oa-buffer-read"
* "rc6-disable"
* "stress-open-close"
--- tests/intel/xe_oa.c | 263 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 263 insertions(+)
tests/intel/xe_oa.c +263 -0
@@ -471,6 +471,23 @@ static struct drm_xe_engine_class_instance *oa_unit_engine(int fd, int n)return hwe; } +static struct drm_xe_oa_unit *nth_oa_unit (int fd,int n) +{ +struct drm_xe_query_oa_units *qoa =xe_oa_units (fd); +struct drm_xe_oa_unit *oau; +u8 *poau; + + poau = (u8 *)&qoa->oa_units[0 ]; +for (int i =0 ; i < qoa->num_oa_units; i++) { + oau = (struct drm_xe_oa_unit *)poau; +if (i == n) +return oau; + poau +=sizeof (*oau) + oau->num_engines *sizeof (oau->eci[0 ]); + } + +return NULL ; +} +static char *pretty_print_oa_period (uint64_t oa_period_ns) {
@@ -558,6 +575,14 @@ write_u64_file(const char *path, uint64_t val)fclose (f); } +static unsigned long rc6_residency_ms (void ) +{ +unsigned long value; + +igt_assert (igt_sysfs_scanf (sysfs,"device/tile0/gt0/gtidle/idle_residency_ms" ,"%lu" , &value) ==1 ); +return value; +} +static uint64_t read_report_ticks (const uint32_t *report,enum xe_oa_format_name format) {
@@ -3487,6 +3512,95 @@ done:return ref_count; } +/** +* SUBTEST: rc6-disable +* Description: Check that opening an OA stream disables RC6 +*/ +static void +test_rc6_disable (void ) +{ +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* Include OA reports in samples */ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , default_test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (default_test_set->perf_oa_format), +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exp_1_millisec, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; +unsigned long rc6_start, rc6_end; + +/* Verify rc6 is functional by measuring residency while idle */ + rc6_start =rc6_residency_ms (); +usleep (50000 ); + rc6_end =rc6_residency_ms (); +igt_require (rc6_end != rc6_start); + +/* While OA is active, we keep rc6 disabled so we don't lose metrics */ + stream_fd =__perf_open (drm_fd, ¶m, false); + + rc6_start =rc6_residency_ms (); +usleep (50000 ); + rc6_end =rc6_residency_ms (); +igt_assert_eq (rc6_end - rc6_start,0 ); + +__perf_close (stream_fd); + +/* But once OA is closed, we expect the device to sleep again */ + rc6_start =rc6_residency_ms (); +usleep (50000 ); + rc6_end =rc6_residency_ms (); +igt_assert_neq (rc6_end - rc6_start,0 ); +} + +/** +* SUBTEST: stress-open-close +* Description: Open/close OA streams in a tight loop +*/ +static void +test_stress_open_close (const struct drm_xe_engine_class_instance *hwe) +{ +struct intel_perf_metric_set *test_set =metric_set (hwe); + +load_helper_init (); +load_helper_run (HIGH ); + + igt_until_timeout(2) { +int oa_exponent =5 ;/* 5 micro seconds */ +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +/* XXX: even without periodic sampling we have to +* specify at least one sample layout property... +*/ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_OA_METRIC_SET , test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (test_set->perf_oa_format), +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exponent, +DRM_XE_OA_PROPERTY_OA_DISABLED , true, +DRM_XE_OA_PROPERTY_OA_ENGINE_INSTANCE , hwe->engine_instance, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; + + stream_fd =__perf_open (drm_fd, ¶m, false); +__perf_close (stream_fd); + } + +load_helper_stop (); +load_helper_fini (); +} +/** * SUBTEST: xe-ref-count * Description: Check that an open oa stream holds a reference on the xe module
@@ -3573,6 +3687,137 @@ test_sysctl_defaults(void)igt_assert_eq (paranoid,1 ); } +/** +* SUBTEST: oa-unit-exclusive-stream-sample-oa +* Description: Check that only a single stream can be opened on an OA unit (with sampling) +* +* SUBTEST: oa-unit-exclusive-stream-exec-q +* Description: Check that only a single stream can be opened on an OA unit (for OAR/OAC) +*/ +/* +* Test if OA buffer streams can be independently opened on OA unit. Once a user +* opens a stream, that oa unit is exclusive to the user, other users get -EBUSY on +* trying to open a stream. +*/ +static void +test_oa_unit_exclusive_stream (bool exponent) +{ +struct drm_xe_query_oa_units *qoa =xe_oa_units (drm_fd); +struct drm_xe_oa_unit *oau; +u8 *poau = (u8 *)&qoa->oa_units[0 ]; +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, +DRM_XE_OA_PROPERTY_OA_METRIC_SET ,0 , +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (0 ), +DRM_XE_OA_PROPERTY_OA_ENGINE_INSTANCE ,0 , +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exp_1_millisec, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; +uint32_t *exec_q =calloc (qoa->num_oa_units,sizeof (u32)); +uint32_t *perf_fd =calloc (qoa->num_oa_units,sizeof (u32)); +u32 vm =xe_vm_create (drm_fd,0 ,0 ); +struct intel_perf_metric_set *test_set; +uint32_t i; + +/* for each oa unit, open one random perf stream with sample OA */ +for (i =0 ; i < qoa->num_oa_units; i++) { +struct drm_xe_engine_class_instance *hwe =oa_unit_engine (drm_fd, i); + + oau = (struct drm_xe_oa_unit *)poau; +if (oau->oa_unit_type !=DRM_XE_OA_UNIT_TYPE_OAG ) +continue ; + test_set =metric_set (hwe); + +igt_debug ("opening OA buffer with c:i %d:%d\n" , + hwe->engine_class, hwe->engine_instance); + exec_q[i] =xe_exec_queue_create (drm_fd, vm, hwe,0 ); +if (!exponent) { + properties[10 ] =DRM_XE_OA_PROPERTY_EXEC_QUEUE_ID ; + properties[11 ] = exec_q[i]; + } + + properties[1 ] = oau->oa_unit_id; + properties[5 ] = test_set->perf_oa_metrics_set; + properties[7 ] =__ff (test_set->perf_oa_format); + properties[9 ] = hwe->engine_instance; + perf_fd[i] =xe_perf_ioctl (drm_fd,DRM_XE_PERF_OP_STREAM_OPEN , ¶m); +igt_assert (perf_fd[i] >=0 ); + poau +=sizeof (*oau) + oau->num_engines *sizeof (oau->eci[0 ]); + } + +/* Xe KMD holds reference to the exec_q's so they shouldn't be really destroyed */ +for (i =0 ; i < qoa->num_oa_units; i++) +if (exec_q[i]) +xe_exec_queue_destroy (drm_fd, exec_q[i]); + +/* for each oa unit make sure no other streams can be opened */ + poau = (u8 *)&qoa->oa_units[0 ]; +for (i =0 ; i < qoa->num_oa_units; i++) { +struct drm_xe_engine_class_instance *hwe =oa_unit_engine (drm_fd, i); +int err; + + oau = (struct drm_xe_oa_unit *)poau; +if (oau->oa_unit_type !=DRM_XE_OA_UNIT_TYPE_OAG ) +continue ; + test_set =metric_set (hwe); + +igt_debug ("try with exp with c:i %d:%d\n" , + hwe->engine_class, hwe->engine_instance); +/* case 1: concurrent access to OAG should fail */ + properties[1 ] = oau->oa_unit_id; + properties[5 ] = test_set->perf_oa_metrics_set; + properties[7 ] =__ff (test_set->perf_oa_format); + properties[9 ] = hwe->engine_instance; + properties[10 ] =DRM_XE_OA_PROPERTY_OA_EXPONENT ; + properties[11 ] = oa_exp_1_millisec; +xe_perf_ioctl_err (drm_fd,DRM_XE_PERF_OP_STREAM_OPEN , ¶m,EBUSY ); + +/* case 2: concurrent access to non-OAG unit should fail */ +igt_debug ("try with exec_q with c:i %d:%d\n" , + hwe->engine_class, hwe->engine_instance); + exec_q[i] =xe_exec_queue_create (drm_fd, vm, hwe,0 ); + properties[10 ] =DRM_XE_OA_PROPERTY_EXEC_QUEUE_ID ; + properties[11 ] = exec_q[i]; + errno =0 ; + err =xe_perf_ioctl (drm_fd,DRM_XE_PERF_OP_STREAM_OPEN , ¶m); +igt_assert (err <0 ); +igt_assert (errno ==EBUSY || errno ==ENODEV ); + poau +=sizeof (*oau) + oau->num_engines *sizeof (oau->eci[0 ]); + } + +for (i =0 ; i < qoa->num_oa_units; i++) { +if (perf_fd[i]) +close (perf_fd[i]); +if (exec_q[i]) +xe_exec_queue_destroy (drm_fd, exec_q[i]); + } +} + +/** +* SUBTEST: oa-unit-concurrent-oa-buffer-read +* Description: Test that we can read streams concurrently on all OA units +*/ +static void +test_oa_unit_concurrent_oa_buffer_read (void ) +{ +struct drm_xe_query_oa_units *qoa =xe_oa_units (drm_fd); + + igt_fork(child, qoa->num_oa_units) { +struct drm_xe_engine_class_instance *hwe =oa_unit_engine (drm_fd, child); + +/* No OAM support yet */ +if (nth_oa_unit (drm_fd, child)->oa_unit_type !=DRM_XE_OA_UNIT_TYPE_OAG ) +exit (0 ); + +test_blocking (40 *1000 *1000 , false,5 *1000 *1000 , hwe); + } +igt_waitchildren (); +} +static const char *xe_engine_class_name (uint32_t engine_class) {switch (engine_class) {
@@ -3740,6 +3985,24 @@ igt_maintest_single_ctx_render_target_writes_a_counter (hwe); } } + igt_subtest_group { + igt_subtest("oa-unit-exclusive-stream-sample-oa") +test_oa_unit_exclusive_stream (true); + + igt_subtest("oa-unit-exclusive-stream-exec-q") +test_oa_unit_exclusive_stream (false); + + igt_subtest("oa-unit-concurrent-oa-buffer-read") +test_oa_unit_concurrent_oa_buffer_read (); + } + + igt_subtest("rc6-disable") +test_rc6_disable (); + + igt_subtest_with_dynamic("stress-open-close") { +__for_one_hwe_in_oag (hwe) +test_stress_open_close (hwe); + } igt_fixture {/* leave sysctl options in their default state... */
Signature · 1 line
2.41.0
PATCH 16/18 tests/intel/xe_oa: add remove OA config tests
In reply to Ashutosh DixitAdd the following tests:
* "invalid-create-userspace-config"
* "invalid-remove-userspace-config"
* "create-destroy-userspace-config"
* "whitelisted-registers-userspace-config"
--- tests/intel/xe_oa.c | 331 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 331 insertions(+)
tests/intel/xe_oa.c +331 -0
@@ -575,6 +575,12 @@ write_u64_file(const char *path, uint64_t val)fclose (f); } +static bool +try_sysfs_read_u64 (const char *path,uint64_t *val) +{ +return igt_sysfs_scanf (sysfs, path,"%" PRIu64, val) ==1 ; +} +static unsigned long rc6_residency_ms (void ) {unsigned long value;
@@ -3601,6 +3607,319 @@ test_stress_open_close(const struct drm_xe_engine_class_instance *hwe)load_helper_fini (); } +static int __xe_oa_add_config (int fd,struct drm_xe_oa_config *config) +{ +int ret =xe_perf_ioctl (fd,DRM_XE_PERF_OP_ADD_CONFIG , config); +if (ret <0 ) + ret = -errno; +return ret; +} + +static int xe_oa_add_config (int fd,struct drm_xe_oa_config *config) +{ +int config_id =__xe_oa_add_config (fd, config); + +igt_debug ("config_id=%i\n" , config_id); +igt_assert (config_id >0 ); + +return config_id; +} + +static void xe_oa_remove_config (int fd,uint64_t config_id) +{ +igt_assert_eq (xe_perf_ioctl (fd,DRM_XE_PERF_OP_REMOVE_CONFIG , &config_id),0 ); +} + +static bool has_xe_oa_userspace_config (int fd) +{ +uint64_t config =0 ; +int ret =xe_perf_ioctl (fd,DRM_XE_PERF_OP_REMOVE_CONFIG , &config); +igt_assert_eq (ret,-1 ); + +igt_debug ("errno=%i\n" , errno); + +return errno !=EINVAL ; +} + +#define SAMPLE_MUX_REG (IS_LUNARLAKE(xe_dev_id(drm_fd)) ? \ + 0x13000 /* PES* */ : 0x9888 /* NOA_WRITE */) + +/** +* SUBTEST: invalid-create-userspace-config +* Description: Test invalid configs are rejected +*/ +static void +test_invalid_create_userspace_config (void ) +{ +struct drm_xe_oa_config config; +const char *uuid ="01234567-0123-0123-0123-0123456789ab" ; +const char *invalid_uuid ="blablabla-wrong" ; +uint32_t mux_regs[] = {SAMPLE_MUX_REG ,0x0 }; +uint32_t invalid_mux_regs[] = {0x12345678 /* invalid register */ ,0x0 }; + +igt_require (has_xe_oa_userspace_config (drm_fd)); + +memset (&config,0 ,sizeof (config)); + +/* invalid uuid */ +strncpy (config.uuid, invalid_uuid,sizeof (config.uuid)); + config.n_regs =1 ; + config.regs_ptr =to_user_pointer (mux_regs); + +igt_assert_eq (__xe_oa_add_config (drm_fd, &config), -EINVAL ); + +/* invalid mux_regs */ +memcpy (config.uuid, uuid,sizeof (config.uuid)); + config.n_regs =1 ; + config.regs_ptr =to_user_pointer (invalid_mux_regs); + +igt_assert_eq (__xe_oa_add_config (drm_fd, &config), -EINVAL ); + +/* empty config */ +memcpy (config.uuid, uuid,sizeof (config.uuid)); + config.n_regs =0 ; + config.regs_ptr =to_user_pointer (mux_regs); + +igt_assert_eq (__xe_oa_add_config (drm_fd, &config), -EINVAL ); + +/* empty config with null pointer */ +memcpy (config.uuid, uuid,sizeof (config.uuid)); + config.n_regs =1 ; + config.regs_ptr =to_user_pointer (NULL ); + +igt_assert_eq (__xe_oa_add_config (drm_fd, &config), -EINVAL ); + +/* invalid pointer */ +memcpy (config.uuid, uuid,sizeof (config.uuid)); + config.n_regs =42 ; + config.regs_ptr =to_user_pointer ((void *)0xDEADBEEF ); + +igt_assert_eq (__xe_oa_add_config (drm_fd, &config), -EFAULT ); +} + +/** +* SUBTEST: invalid-remove-userspace-config +* Description: Test invalid remove configs are rejected +*/ +static void +test_invalid_remove_userspace_config (void ) +{ +struct drm_xe_oa_config config; +const char *uuid ="01234567-0123-0123-0123-0123456789ab" ; +uint32_t mux_regs[] = {SAMPLE_MUX_REG ,0x0 }; +uint64_t config_id, wrong_config_id =999999999 ; +char path[512 ]; + +igt_require (has_xe_oa_userspace_config (drm_fd)); + +snprintf (path,sizeof (path),"metrics/%s/id" , uuid); + +/* Destroy previous configuration if present */ +if (try_sysfs_read_u64 (path, &config_id)) +xe_oa_remove_config (drm_fd, config_id); + +memset (&config,0 ,sizeof (config)); + +memcpy (config.uuid, uuid,sizeof (config.uuid)); + + config.n_regs =1 ; + config.regs_ptr =to_user_pointer (mux_regs); + + config_id =xe_oa_add_config (drm_fd, &config); + +/* Removing configs without permissions should fail. */ + igt_fork(child, 1) { +igt_drop_root (); + +xe_perf_ioctl_err (drm_fd,DRM_XE_PERF_OP_REMOVE_CONFIG , &config_id,EACCES ); + } +igt_waitchildren (); + +/* Removing invalid config ID should fail. */ +xe_perf_ioctl_err (drm_fd,DRM_XE_PERF_OP_REMOVE_CONFIG , &wrong_config_id,ENOENT ); + +xe_oa_remove_config (drm_fd, config_id); +} + +/** +* SUBTEST: create-destroy-userspace-config +* Description: Test add/remove OA configs +*/ +static void +test_create_destroy_userspace_config (void ) +{ +struct drm_xe_oa_config config; +const char *uuid ="01234567-0123-0123-0123-0123456789ab" ; +uint32_t mux_regs[] = {SAMPLE_MUX_REG ,0x0 }; +uint32_t regs[100 ]; +int i; +uint64_t config_id; +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , + +DRM_XE_OA_PROPERTY_OA_METRIC_SET ,0 ,/* Filled later */ + +/* OA unit configuration */ +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (default_test_set->perf_oa_format), +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exp_1_millisec, +DRM_XE_OA_PROPERTY_OA_DISABLED , true, +DRM_XE_OA_PROPERTY_OA_METRIC_SET + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; +char path[512 ]; + +igt_require (has_xe_oa_userspace_config (drm_fd)); + +snprintf (path,sizeof (path),"metrics/%s/id" , uuid); + +/* Destroy previous configuration if present */ +if (try_sysfs_read_u64 (path, &config_id)) +xe_oa_remove_config (drm_fd, config_id); + +memset (&config,0 ,sizeof (config)); +memcpy (config.uuid, uuid,sizeof (config.uuid)); + + regs[0 ] = mux_regs[0 ]; + regs[1 ] = mux_regs[1 ]; +/* Flex EU counters */ +for (i =1 ; i <ARRAY_SIZE (regs) /2 ; i++) { + regs[i *2 ] =0xe458 ;/* EU_PERF_CNTL0 */ + regs[i *2 +1 ] =0x0 ; + } + config.regs_ptr =to_user_pointer (regs); + config.n_regs =ARRAY_SIZE (regs) /2 ; + +/* Creating configs without permissions shouldn't work. */ + igt_fork(child, 1) { +igt_drop_root (); + +igt_assert_eq (__xe_oa_add_config (drm_fd, &config), -EACCES ); + } +igt_waitchildren (); + +/* Create a new config */ + config_id =xe_oa_add_config (drm_fd, &config); + +/* Verify that adding the another config with the same uuid fails. */ +igt_assert_eq (__xe_oa_add_config (drm_fd, &config), -EADDRINUSE ); + +/* Try to use the new config */ + properties[3 ] = config_id; + stream_fd =__perf_open (drm_fd, ¶m, false); + +/* Verify that destroying the config doesn't yield any error. */ +xe_oa_remove_config (drm_fd, config_id); + +/* Read the config to verify shouldn't raise any issue. */ + config_id =xe_oa_add_config (drm_fd, &config); + +__perf_close (stream_fd); + +xe_oa_remove_config (drm_fd, config_id); +} + +/** +* SUBTEST: whitelisted-registers-userspace-config +* Description: Test that an OA config constructed using whitelisted register works +*/ +/* Registers required by userspace. This list should be maintained by +* the OA configs developers and agreed upon with kernel developers as +* some of the registers have bits used by the kernel (for workarounds +* for instance) and other bits that need to be set by the OA configs. +*/ +static void +test_whitelisted_registers_userspace_config (void ) +{ +struct drm_xe_oa_config config; +const char *uuid ="01234567-0123-0123-0123-0123456789ab" ; +uint32_t regs[600 ]; +uint32_t i; +uint32_t oa_start_trig1, oa_start_trig8; +uint32_t oa_report_trig1, oa_report_trig8; +uint64_t config_id; +char path[512 ]; +int ret; +const uint32_t flex[] = { +0xe458 , +0xe558 , +0xe658 , +0xe758 , +0xe45c , +0xe55c , +0xe65c + }; + +igt_require (has_xe_oa_userspace_config (drm_fd)); + +snprintf (path,sizeof (path),"metrics/%s/id" , uuid); + +if (try_sysfs_read_u64 (path, &config_id)) +xe_oa_remove_config (drm_fd, config_id); + +memset (&config,0 ,sizeof (config)); +memcpy (config.uuid, uuid,sizeof (config.uuid)); + + oa_start_trig1 =0xd900 ; + oa_start_trig8 =0xd91c ; + oa_report_trig1 =0xd920 ; + oa_report_trig8 =0xd93c ; + +/* b_counters_regs: OASTARTTRIG[1-8] */ +for (i = oa_start_trig1; i <= oa_start_trig8; i +=4 ) { + regs[config.n_regs *2 ] = i; + regs[config.n_regs *2 +1 ] =0 ; + config.n_regs++; + } +/* b_counters_regs: OAREPORTTRIG[1-8] */ +for (i = oa_report_trig1; i <= oa_report_trig8; i +=4 ) { + regs[config.n_regs *2 ] = i; + regs[config.n_regs *2 +1 ] =0 ; + config.n_regs++; + } + +/* Flex EU registers, only from Gen8+. */ +for (i =0 ; i <ARRAY_SIZE (flex); i++) { + regs[config.n_regs *2 ] = flex[i]; + regs[config.n_regs *2 +1 ] =0 ; + config.n_regs++; + } + +/* Mux registers (too many of them, just checking bounds) */ +/* NOA_WRITE */ + regs[config.n_regs *2 ] =SAMPLE_MUX_REG ; + regs[config.n_regs *2 +1 ] =0 ; + config.n_regs++; + +/* NOA_CONFIG */ +if (!IS_LUNARLAKE (devid)) { + regs[config.n_regs *2 ] =0xD04 ; + regs[config.n_regs *2 +1 ] =0 ; + config.n_regs++; + regs[config.n_regs *2 ] =0xD2C ; + regs[config.n_regs *2 +1 ] =0 ; + config.n_regs++; + } +if (!IS_LUNARLAKE (devid) && !IS_METEORLAKE (devid)) { +/* WAIT_FOR_RC6_EXIT */ + regs[config.n_regs *2 ] =0x20CC ; + regs[config.n_regs *2 +1 ] =0 ; + config.n_regs++; + } + + config.regs_ptr = (uintptr_t ) regs; + +/* Create a new config */ + ret =xe_perf_ioctl (drm_fd,DRM_XE_PERF_OP_ADD_CONFIG , &config); +igt_assert (ret >0 );/* Config 0 should be used by the kernel */ + config_id = ret; + +xe_oa_remove_config (drm_fd, config_id); +} +/** * SUBTEST: xe-ref-count * Description: Check that an open oa stream holds a reference on the xe module
@@ -4004,6 +4323,18 @@ igt_maintest_stress_open_close (hwe); } + igt_subtest("invalid-create-userspace-config") +test_invalid_create_userspace_config (); + + igt_subtest("invalid-remove-userspace-config") +test_invalid_remove_userspace_config (); + + igt_subtest("create-destroy-userspace-config") +test_create_destroy_userspace_config (); + + igt_subtest("whitelisted-registers-userspace-config") +test_whitelisted_registers_userspace_config (); + igt_fixture {/* leave sysctl options in their default state... */ write_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ,1 );
Signature · 1 line
2.41.0
PATCH 17/18 tests/intel/xe_oa: OA buffer mmap tests
In reply to Ashutosh DixitAdd the following tests:
* "map-oa-buffer"
* "invalid-map-oa-buffer"
* "non-privileged-map-oa-buffer"
* "non-privileged-access-vaddr"
* "privileged-forked-access-vaddr"
* "closed-fd-and-unmapped-access"
--- tests/intel/xe_oa.c | 251 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 251 insertions(+)
tests/intel/xe_oa.c +251 -0
@@ -4137,6 +4137,231 @@ test_oa_unit_concurrent_oa_buffer_read(void)igt_waitchildren (); } +static void *map_oa_buffer (u32 *size) +{ +void *vaddr =mmap (0 ,OA_BUFFER_SIZE ,PROT_READ ,MAP_PRIVATE , stream_fd,0 ); + +igt_assert (vaddr !=NULL ); + *size =OA_BUFFER_SIZE ; +return vaddr; +} + +static void invalid_param_map_oa_buffer (const struct drm_xe_engine_class_instance *hwe) +{ +void *oa_vaddr =NULL ; + +/* try a couple invalid mmaps */ +/* bad prots */ + oa_vaddr =mmap (0 ,OA_BUFFER_SIZE ,PROT_WRITE ,MAP_PRIVATE , stream_fd,0 ); +igt_assert (oa_vaddr ==MAP_FAILED ); + + oa_vaddr =mmap (0 ,OA_BUFFER_SIZE ,PROT_EXEC ,MAP_PRIVATE , stream_fd,0 ); +igt_assert (oa_vaddr ==MAP_FAILED ); + +/* bad MAPs */ + oa_vaddr =mmap (0 ,OA_BUFFER_SIZE ,PROT_READ ,MAP_SHARED , stream_fd,0 ); +igt_assert (oa_vaddr ==MAP_FAILED ); + +/* bad size */ + oa_vaddr =mmap (0 ,OA_BUFFER_SIZE +1 ,PROT_READ ,MAP_PRIVATE , stream_fd,0 ); +igt_assert (oa_vaddr ==MAP_FAILED ); + +/* do the right thing */ + oa_vaddr =mmap (0 ,OA_BUFFER_SIZE ,PROT_READ ,MAP_PRIVATE , stream_fd,0 ); +igt_assert (oa_vaddr !=MAP_FAILED && oa_vaddr !=NULL ); + +munmap (oa_vaddr,OA_BUFFER_SIZE ); +} + +static void unprivileged_try_to_map_oa_buffer (void ) +{ +void *oa_vaddr; + + oa_vaddr =mmap (0 ,OA_BUFFER_SIZE ,PROT_READ ,MAP_PRIVATE , stream_fd,0 ); +igt_assert (oa_vaddr ==MAP_FAILED ); +igt_assert_eq (errno,EACCES ); +} + +static void unprivileged_map_oa_buffer (const struct drm_xe_engine_class_instance *hwe) +{ + igt_fork(child, 1) { +igt_drop_root (); +unprivileged_try_to_map_oa_buffer (); + } +igt_waitchildren (); +} + +static jmp_buf jmp; +static void __attribute__((noreturn))sigtrap (int sig) +{ +siglongjmp (jmp, sig); +} + +static void try_invalid_access (void *vaddr) +{ +sighandler_t old_sigsegv; +uint32_t dummy; + + old_sigsegv =signal (SIGSEGV , sigtrap); +switch (sigsetjmp (jmp,SIGSEGV )) { +case SIGSEGV : +break ; +case 0 : + dummy =READ_ONCE (*((uint32_t *)vaddr +1 )); + (void ) dummy; +default : +igt_assert (!"reached" ); +break ; + } +signal (SIGSEGV , old_sigsegv); +} + +static void map_oa_buffer_unprivilege_access (const struct drm_xe_engine_class_instance *hwe) +{ +void *vaddr; +uint32_t size; + + vaddr =map_oa_buffer (&size); + + igt_fork(child, 1) { +igt_drop_root (); +try_invalid_access (vaddr); + } +igt_waitchildren (); + +munmap (vaddr, size); +} + +static void map_oa_buffer_forked_access (const struct drm_xe_engine_class_instance *hwe) +{ +void *vaddr; +uint32_t size; + + vaddr =map_oa_buffer (&size); + + igt_fork(child, 1) { +try_invalid_access (vaddr); + } +igt_waitchildren (); + +munmap (vaddr, size); +} + +static void check_reports (void *oa_vaddr,uint32_t oa_size, +const struct drm_xe_engine_class_instance *hwe) +{ +struct intel_perf_metric_set *test_set =metric_set (hwe); +uint64_t fmt = test_set->perf_oa_format; +struct oa_format format =get_oa_format (fmt); +size_t report_words = format.size >>2 ; +uint32_t *reports; +uint32_t timer_reports =0 ; + +for (reports = (uint32_t *)oa_vaddr; + timer_reports <20 && reports[0 ] &&oa_timestamp (reports, fmt); + reports += report_words) { +if (!oa_report_is_periodic (oa_exp_1_millisec, reports)) +continue ; + + timer_reports++; +if (timer_reports >=3 ) +sanity_check_reports (reports -2 * report_words, + reports - report_words, fmt); + } + +igt_assert (timer_reports >=3 ); +} + +static void check_reports_from_mapped_buffer (const struct drm_xe_engine_class_instance *hwe) +{ +void *vaddr; +uint32_t size; +uint32_t period_us =oa_exponent_to_ns (oa_exp_1_millisec) /1000 ; + + vaddr =map_oa_buffer (&size); + +/* wait for approx 100 reports */ +usleep (100 * period_us); +check_reports (vaddr, size, hwe); + +munmap (vaddr, size); +} + +/** +* SUBTEST: closed-fd-and-unmapped-access +* Description: Unmap buffer, close fd and try to access +*/ +static void closed_fd_and_unmapped_access (const struct drm_xe_engine_class_instance *hwe) +{ +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, +DRM_XE_OA_PROPERTY_OA_METRIC_SET , default_test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (default_test_set->perf_oa_format), +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exp_1_millisec, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; +void *vaddr; +uint32_t size; +uint32_t period_us =oa_exponent_to_ns (oa_exp_1_millisec) /1000 ; + + stream_fd =__perf_open (drm_fd, ¶m, false); + vaddr =map_oa_buffer (&size); + +usleep (100 * period_us); +check_reports (vaddr, size, hwe); + +munmap (vaddr, size); +__perf_close (stream_fd); + +try_invalid_access (vaddr); +} + +/** +* SUBTEST: map-oa-buffer +* Description: Verify mapping of oa buffer +* +* SUBTEST: invalid-map-oa-buffer +* Description: Verify invalid mappings of oa buffer +* +* SUBTEST: non-privileged-map-oa-buffer +* Description: Verify if non-privileged user can map oa buffer +* +* SUBTEST: non-privileged-access-vaddr +* Description: Verify if non-privileged user can map oa buffer +* +* SUBTEST: privileged-forked-access-vaddr +* Description: Verify that forked access to mapped buffer fails +*/ +typedef void (*map_oa_buffer_test_t )(const struct drm_xe_engine_class_instance *hwe); +static void test_mapped_oa_buffer (map_oa_buffer_test_t test_with_fd_open, +const struct drm_xe_engine_class_instance *hwe) +{ +struct intel_perf_metric_set *test_set =metric_set (hwe); +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, +DRM_XE_OA_PROPERTY_OA_METRIC_SET , test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (test_set->perf_oa_format), +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exp_1_millisec, +DRM_XE_OA_PROPERTY_OA_ENGINE_INSTANCE , hwe->engine_instance, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =ARRAY_SIZE (properties) /2 , + .properties_ptr =to_user_pointer (properties), + }; + + stream_fd =__perf_open (drm_fd, ¶m, false); + +igt_assert (test_with_fd_open); +test_with_fd_open (hwe); + +__perf_close (stream_fd); +} +static const char *xe_engine_class_name (uint32_t engine_class) {switch (engine_class) {
@@ -4335,6 +4560,32 @@ igt_main igt_subtest("whitelisted-registers-userspace-config")test_whitelisted_registers_userspace_config (); + igt_subtest_group { + igt_subtest_with_dynamic("map-oa-buffer") +__for_one_hwe_in_oag (hwe) +test_mapped_oa_buffer (check_reports_from_mapped_buffer, hwe); + + igt_subtest_with_dynamic("invalid-map-oa-buffer") +__for_one_hwe_in_oag (hwe) +test_mapped_oa_buffer (invalid_param_map_oa_buffer, hwe); + + igt_subtest_with_dynamic("non-privileged-map-oa-buffer") +__for_one_hwe_in_oag (hwe) +test_mapped_oa_buffer (unprivileged_map_oa_buffer, hwe); + + igt_subtest_with_dynamic("non-privileged-access-vaddr") +__for_one_hwe_in_oag (hwe) +test_mapped_oa_buffer (map_oa_buffer_unprivilege_access, hwe); + + igt_subtest_with_dynamic("privileged-forked-access-vaddr") +__for_one_hwe_in_oag (hwe) +test_mapped_oa_buffer (map_oa_buffer_forked_access, hwe); + + igt_subtest_with_dynamic("closed-fd-and-unmapped-access") +__for_one_hwe_in_oag (hwe) +closed_fd_and_unmapped_access (hwe); + } + igt_fixture {/* leave sysctl options in their default state... */ write_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ,1 );
Signature · 1 line
2.41.0
PATCH 18/18 tests/intel/xe_oa: Register whitelisting and MMIO trigger tests
In reply to Ashutosh DixitAdd the following tests:
* "oa-regs-whitelisted"
* "mmio-triggered-reports"
--- tests/intel/xe_oa.c | 326 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 326 insertions(+)
tests/intel/xe_oa.c +326 -0
@@ -3920,6 +3920,316 @@ test_whitelisted_registers_userspace_config(void)xe_oa_remove_config (drm_fd, config_id); } +#define OAG_OASTATUS (0xdafc) +#define OAG_MMIOTRIGGER (0xdb1c) + +static const uint32_t oa_wl[] = { +OAG_MMIOTRIGGER , +OAG_OASTATUS , +}; + +static const uint32_t nonpriv_slot_offsets[] = { +0x4d0 ,0x4d4 ,0x4d8 ,0x4dc ,0x4e0 ,0x4e4 ,0x4e8 ,0x4ec , +0x4f0 ,0x4f4 ,0x4f8 ,0x4fc ,0x010 ,0x014 ,0x018 ,0x01c , +0x1e0 ,0x1e4 ,0x1e8 ,0x1ec , +}; + +struct test_perf { +const uint32_t *slots; +uint32_t num_slots; +const uint32_t *wl; +uint32_t num_wl; +} perf; + +#define HAS_OA_MMIO_TRIGGER (__d) \ + (IS_DG2(__d) || IS_PONTEVECCHIO(__d) || IS_METEORLAKE(__d) || IS_LUNARLAKE(__d)) + +static void perf_init_whitelist (void ) +{ + perf.slots = nonpriv_slot_offsets; + perf.num_slots =20 ; + perf.wl = oa_wl; + perf.num_wl =ARRAY_SIZE (oa_wl); +} + +static void +emit_oa_reg_read (struct intel_bb *ibb,struct intel_buf *dst,uint32_t offset, +uint32_t reg) +{ +intel_bb_add_intel_buf (ibb, dst, true); + +intel_bb_out (ibb,MI_STORE_REGISTER_MEM |2 ); +intel_bb_out (ibb, reg); +intel_bb_emit_reloc (ibb, dst->handle, +I915_GEM_DOMAIN_INSTRUCTION , +I915_GEM_DOMAIN_INSTRUCTION , + offset, dst->addr.offset); +intel_bb_out (ibb,lower_32_bits (offset)); +intel_bb_out (ibb,upper_32_bits (offset)); +} + +static void +emit_mmio_triggered_report (struct intel_bb *ibb,uint32_t value) +{ +intel_bb_out (ibb,MI_LOAD_REGISTER_IMM (1 )); +intel_bb_out (ibb,OAG_MMIOTRIGGER ); +intel_bb_out (ibb, value); +} + +static void dump_whitelist (uint32_t mmio_base,const char *msg) +{ +int i; + +igt_debug ("%s\n" , msg); + +for (i =0 ; i < perf.num_slots; i++) +igt_debug ("FORCE_TO_NON_PRIV_%02d = %08x\n" , + i,intel_register_read (&mmio_data, + mmio_base + perf.slots[i])); +} + +static bool in_whitelist (uint32_t mmio_base,uint32_t reg) +{ +int i; + +if (reg &MMIO_BASE_OFFSET ) + reg = (reg & ~MMIO_BASE_OFFSET ) + mmio_base; + +for (i =0 ; i < perf.num_slots; i++) { +uint32_t fpriv =intel_register_read (&mmio_data, + mmio_base + perf.slots[i]); + +if ((fpriv &RING_FORCE_TO_NONPRIV_ADDRESS_MASK ) == reg) +return true; + } + +return false; +} + +static void oa_regs_in_whitelist (uint32_t mmio_base,bool are_present) +{ +int i; + +if (are_present) { +for (i =0 ; i < perf.num_wl; i++) +igt_assert (in_whitelist (mmio_base, perf.wl[i])); + }else { +for (i =0 ; i < perf.num_wl; i++) +igt_assert (!in_whitelist (mmio_base, perf.wl[i])); + } +} + +static u32 oa_get_mmio_base (const struct drm_xe_engine_class_instance *hwe) +{ +u32 mmio_base =0x2000 ; + +switch (hwe->engine_class) { +case DRM_XE_ENGINE_CLASS_RENDER : + mmio_base =0x2000 ; +break ; +case DRM_XE_ENGINE_CLASS_COMPUTE : +switch (hwe->engine_instance) { +case 0 : + mmio_base =0x1a000 ; +break ; +case 1 : + mmio_base =0x1c000 ; +break ; +case 2 : + mmio_base =0x1e000 ; +break ; +case 3 : + mmio_base =0x26000 ; +break ; + } + } + +return mmio_base; +} + +/** +* SUBTEST: oa-regs-whitelisted +* Description: Verify that OA registers are whitelisted +*/ +static void test_oa_regs_whitelist (const struct drm_xe_engine_class_instance *hwe) +{ +struct intel_perf_metric_set *test_set =metric_set (hwe); +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_OA_UNIT_ID ,0 , +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, +DRM_XE_OA_PROPERTY_OA_METRIC_SET , test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (test_set->perf_oa_format), +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exp_1_millisec, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =sizeof (properties) /16 , + .properties_ptr =to_user_pointer (properties), + }; +// uint32_t mmio_base = gem_engine_mmio_base(drm_fd, e->name); +u32 mmio_base; + +/* FIXME: Add support for OAM whitelist testing */ +if (hwe->engine_class !=DRM_XE_ENGINE_CLASS_RENDER && + hwe->engine_class !=DRM_XE_ENGINE_CLASS_COMPUTE ) +return ; + + mmio_base =oa_get_mmio_base (hwe); + +intel_register_access_init (&mmio_data, +igt_device_get_pci_device (drm_fd), +0 , drm_fd); + stream_fd =__perf_open (drm_fd, ¶m, false); + +dump_whitelist (mmio_base,"oa whitelisted" ); + +oa_regs_in_whitelist (mmio_base, true); + +__perf_close (stream_fd); + +dump_whitelist (mmio_base,"oa remove whitelist" ); + +/* +* after perf close, check that registers are removed from the nonpriv +* slots +* FIXME if needed: currently regs remain added forever +*/ +// oa_regs_in_whitelist(mmio_base, false); + +intel_register_access_fini (&mmio_data); +} + +static void +__test_mmio_triggered_reports (struct drm_xe_engine_class_instance *hwe) +{ +struct intel_perf_metric_set *test_set = default_test_set; +int oa_exponent =max_oa_exponent_for_period_lte (2 *NSEC_PER_SEC ); +uint64_t properties[] = { +DRM_XE_OA_PROPERTY_SAMPLE_OA , true, +DRM_XE_OA_PROPERTY_OA_METRIC_SET , test_set->perf_oa_metrics_set, +DRM_XE_OA_PROPERTY_OA_FORMAT ,__ff (test_set->perf_oa_format), +DRM_XE_OA_PROPERTY_OA_EXPONENT , oa_exponent, +DRM_XE_OA_PROPERTY_OA_ENGINE_INSTANCE , hwe->engine_instance, + }; +struct drm_xe_oa_open_prop param = { + .num_properties =sizeof (properties) /16 , + .properties_ptr =to_user_pointer (properties), + }; +size_t format_size =get_oa_format (test_set->perf_oa_format).size; +uint32_t oa_buffer, offset_tail1, offset_tail2; +struct intel_buf src, dst, *dst_buf; +uint32_t mmio_triggered_reports =0 ; +uint32_t *start, *end; +struct buf_ops *bops; +struct intel_bb *ibb; +uint32_t context, vm; +int height =600 ; +int width =800 ; +uint8_t *buf; + + bops =buf_ops_create (drm_fd); + + dst_buf =intel_buf_create (bops,4096 ,1 ,8 ,64 , +I915_TILING_NONE , +I915_COMPRESSION_NONE ); +buf_map (drm_fd, dst_buf, true); +memset (dst_buf->ptr,0 ,4096 ); +intel_buf_unmap (dst_buf); + +scratch_buf_init (bops, &src, width, height,0xff0000ff ); +scratch_buf_init (bops, &dst, width, height,0x00ff00ff ); + + vm =xe_vm_create (drm_fd,0 ,0 ); + context =xe_exec_queue_create (drm_fd, vm, hwe,0 ); +igt_assert (context); + ibb =intel_bb_create_with_context (drm_fd, context, vm,NULL ,BATCH_SZ ); + + stream_fd =__perf_open (drm_fd, ¶m, false); +set_fd_flags (stream_fd,O_CLOEXEC ); + + buf =mmap (0 ,OA_BUFFER_SIZE ,PROT_READ ,MAP_PRIVATE , stream_fd,0 ); +igt_assert (buf !=NULL ); + +emit_oa_reg_read (ibb, dst_buf,0 ,OAG_OABUFFER ); +emit_oa_reg_read (ibb, dst_buf,4 ,OAG_OATAILPTR ); +emit_mmio_triggered_report (ibb,0xc0ffee11 ); + +if (render_copy) +render_copy (ibb, + &src,0 ,0 , width, height, + &dst,0 ,0 ); + +emit_mmio_triggered_report (ibb,0xc0ffee22 ); + +emit_oa_reg_read (ibb, dst_buf,8 ,OAG_OATAILPTR ); + +intel_bb_flush_render (ibb); +intel_bb_sync (ibb); + +buf_map (drm_fd, dst_buf, false); + + oa_buffer = dst_buf->ptr[0 ] &OAG_OATAILPTR_MASK ; + offset_tail1 = (dst_buf->ptr[1 ] &OAG_OATAILPTR_MASK ) - oa_buffer; + offset_tail2 = (dst_buf->ptr[2 ] &OAG_OATAILPTR_MASK ) - oa_buffer; + +igt_debug ("oa_buffer = %08x, tail1 = %08x, tail2 = %08x\n" , + oa_buffer, offset_tail1, offset_tail2); + + start = (uint32_t *)(buf + offset_tail1); + end = (uint32_t *)(buf + offset_tail2); +while (start < end) { +if (!report_reason (start)) + mmio_triggered_reports++; + +if (get_oa_format (test_set->perf_oa_format).report_hdr_64bit) { +u64 *start64 = (u64 *)start; + +igt_debug ("hdr: %016lx %016lx %016lx %016lx\n" , + start64[0 ], start64[1 ], start64[2 ], start64[3 ]); + }else { +igt_debug ("hdr: %08x %08x %08x %08x\n" , + start[0 ], start[1 ], start[2 ], start[3 ]); + } + + start += format_size /4 ; + } + +igt_assert_eq (mmio_triggered_reports,2 ); + +munmap (buf,OA_BUFFER_SIZE ); +intel_buf_close (bops, &src); +intel_buf_close (bops, &dst); +intel_buf_unmap (dst_buf); +intel_buf_destroy (dst_buf); +intel_bb_destroy (ibb); +xe_exec_queue_destroy (drm_fd, context); +xe_vm_destroy (drm_fd, vm); +buf_ops_destroy (bops); +__perf_close (stream_fd); +} + +/** +* SUBTEST: mmio-triggered-reports +* Description: Test MMIO trigger functionality +*/ +static void +test_mmio_triggered_reports (struct drm_xe_engine_class_instance *hwe) +{ +struct igt_helper_process child = {}; +int ret; + +write_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ,0 ); + igt_fork_helper(&child) { +igt_drop_root (); + +__test_mmio_triggered_reports (hwe); + } + ret =igt_wait_helper (&child); +write_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ,1 ); + +igt_assert (WEXITSTATUS (ret) ==EAGAIN || +WEXITSTATUS (ret) ==0 ); +} +/** * SUBTEST: xe-ref-count * Description: Check that an open oa stream holds a reference on the xe module
@@ -4586,6 +4896,22 @@ igt_mainclosed_fd_and_unmapped_access (hwe); } + igt_subtest_group { + igt_fixture { +perf_init_whitelist (); + } + + igt_subtest_with_dynamic("oa-regs-whitelisted") +__for_one_hwe_in_oag (hwe) +test_oa_regs_whitelist (hwe); + + igt_subtest_with_dynamic("mmio-triggered-reports") { +igt_require (HAS_OA_MMIO_TRIGGER (devid)); +__for_one_hwe_in_oag (hwe) +test_mmio_triggered_reports (hwe); + } + } + igt_fixture {/* leave sysctl options in their default state... */ write_u64_file ("/proc/sys/dev/xe/perf_stream_paranoid" ,1 );
Signature · 1 line
2.41.0
Patchwork
CI✗ GitLab.Pipeline: warning for Intel Xe OA IGT's
In reply to Ashutosh Dixit== Series Details ==
== Series Details ==
Series: Intel Xe OA IGT's
URL : https://patchwork.freedesktop.org/series/130033/
State : warning
== Summary ==
Pipeline status: FAILED.
see https://gitlab.freedesktop.org/gfx-ci/igt-ci-tags/-/pipelines/1105477 for the overview.
build:tests-debian-meson-armhf has failed (https://gitlab.freedesktop.org/gfx-ci/igt-ci-tags/-/jobs/55152791):
^
../lib/xe/xe_oa.c: In function ‘xe_oa_prop_to_ext’:
../lib/xe/xe_oa.c:653:16: error: cast to pointer from integer of different size [-Werror=int-to-pointer-cast]
__u64 *prop = (__u64 *)properties->properties_ptr;
^
../lib/xe/xe_oa.c:667:32: error: cast from pointer to integer of different size [-Werror=pointer-to-int-cast]
ext[j].base.next_extension = (__u64)&ext[j + 1];
^
../lib/xe/xe_oa.c: In function ‘xe_perf_ioctl’:
../lib/xe/xe_oa.c:681:12: error: cast from pointer to integer of different size [-Werror=pointer-to-int-cast]
.param = (__u64)((op == DRM_XE_PERF_OP_STREAM_OPEN) ? ext : arg),
^ cc1: some warnings being treated as errors
ninja: build stopped: subcommand failed. section_end:1708126620:step_script section_start:1708126620:cleanup_file_variables Cleaning up project directory and file based variables section_end:1708126621:cleanup_file_variables
ERROR: Job failed: exit code 1
build:tests-debian-meson-mips has failed (https://gitlab.freedesktop.org/gfx-ci/igt-ci-tags/-/jobs/55152793):
^
../lib/xe/xe_oa.c: In function ‘xe_oa_prop_to_ext’:
../lib/xe/xe_oa.c:653:16: error: cast to pointer from integer of different size [-Werror=int-to-pointer-cast]
__u64 *prop = (__u64 *)properties->properties_ptr;
^
../lib/xe/xe_oa.c:667:32: error: cast from pointer to integer of different size [-Werror=pointer-to-int-cast]
ext[j].base.next_extension = (__u64)&ext[j + 1];
^
../lib/xe/xe_oa.c: In function ‘xe_perf_ioctl’:
../lib/xe/xe_oa.c:681:12: error: cast from pointer to integer of different size [-Werror=pointer-to-int-cast]
.param = (__u64)((op == DRM_XE_PERF_OP_STREAM_OPEN) ? ext : arg),
^ cc1: some warnings being treated as errors
ninja: build stopped: subcommand failed. section_end:1708126612:step_script section_start:1708126612:cleanup_file_variables Cleaning up project directory and file based variables section_end:1708126612:cleanup_file_variables
ERROR: Job failed: exit code 1
== Logs ==
For more details see: https://gitlab.freedesktop.org/gfx-ci/igt-ci-tags/-/pipelines/1105477
Patchwork
CI✓ Fi.CI.BAT: success for Intel Xe OA IGT's
In reply to Ashutosh Dixit== Series Details ==
== Series Details ==
Series: Intel Xe OA IGT's
URL : https://patchwork.freedesktop.org/series/130033/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_14287 -> IGTPW_10690
====================================================
Summary
-------
**SUCCESS**
No regressions found.
External URL: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/index.html
Participating hosts (36 -> 35)
------------------------------
Missing (1): fi-snb-2520m
Known issues
------------
Here are the changes found in IGTPW_10690 that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@debugfs_test@basic-hwmon:
- bat-jsl-1: NOTRUN -> [SKIP][1] ([i915#9318])
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/bat-jsl-1/igt@[email protected]
* igt@gem_huc_copy@huc-copy:
- bat-jsl-1: NOTRUN -> [SKIP][2] ([i915#2190])
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/bat-jsl-1/igt@[email protected]
* igt@gem_lmem_swapping@verify-random:
- bat-jsl-1: NOTRUN -> [SKIP][3] ([i915#4613]) +3 other tests skip
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/bat-jsl-1/igt@[email protected]
* igt@kms_cursor_legacy@basic-busy-flip-before-cursor-legacy:
- bat-jsl-1: NOTRUN -> [SKIP][4] ([i915#4103]) +1 other test skip
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/bat-jsl-1/igt@[email protected]
* igt@kms_dsc@dsc-basic:
- bat-jsl-1: NOTRUN -> [SKIP][5] ([i915#3555] / [i915#9886])
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/bat-jsl-1/igt@[email protected]
* igt@kms_force_connector_basic@force-load-detect:
- bat-jsl-1: NOTRUN -> [SKIP][6] ([fdo#109285])
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/bat-jsl-1/igt@[email protected]
* igt@kms_setmode@basic-clone-single-crtc:
- bat-jsl-1: NOTRUN -> [SKIP][7] ([i915#3555])
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/bat-jsl-1/igt@[email protected]
#### Possible fixes ####
* igt@gem_exec_create@basic@smem:
- {bat-arls-1}: [DMESG-WARN][8] ([i915#10194]) -> [PASS][9]
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/bat-arls-1/igt@gem_exec_create@[email protected]
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/bat-arls-1/igt@gem_exec_create@[email protected]
{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).
[fdo#109285]: https://bugs.freedesktop.org/show_bug.cgi?id=109285
[i915#10194]: https://gitlab.freedesktop.org/drm/intel/issues/10194
[i915#10213]: https://gitlab.freedesktop.org/drm/intel/issues/10213
[i915#10215]: https://gitlab.freedesktop.org/drm/intel/issues/10215
[i915#2190]: https://gitlab.freedesktop.org/drm/intel/issues/2190
[i915#3555]: https://gitlab.freedesktop.org/drm/intel/issues/3555
[i915#4103]: https://gitlab.freedesktop.org/drm/intel/issues/4103
[i915#4613]: https://gitlab.freedesktop.org/drm/intel/issues/4613
[i915#9318]: https://gitlab.freedesktop.org/drm/intel/issues/9318
[i915#9886]: https://gitlab.freedesktop.org/drm/intel/issues/9886
Build changes
-------------
* CI: CI-20190529 -> None
* IGT: IGT_7716 -> IGTPW_10690
CI-20190529: 20190529
CI_DRM_14287: b09fd1bd9f8b4a8e5cd8c1de8dbba4d4eb15b4e9 @ git://anongit.freedesktop.org/gfx-ci/linux
IGTPW_10690: 10690
IGT_7716: 7716
Testlist changes
----------------
+igt@xe_oa@blocking
+igt@xe_oa@buffer-fill
+igt@xe_oa@closed-fd-and-unmapped-access
+igt@xe_oa@create-destroy-userspace-config
+igt@xe_oa@disabled-read-error
+igt@xe_oa@enable-disable
+igt@xe_oa@invalid-create-userspace-config
+igt@xe_oa@invalid-map-oa-buffer
+igt@xe_oa@invalid-oa-exponent
+igt@xe_oa@invalid-oa-format-id
+igt@xe_oa@invalid-oa-metric-set-id
+igt@xe_oa@invalid-remove-userspace-config
+igt@xe_oa@map-oa-buffer
+igt@xe_oa@missing-sample-flags
+igt@xe_oa@mi-rpc
+igt@xe_oa@mmio-triggered-reports
+igt@xe_oa@non-privileged-access-vaddr
+igt@xe_oa@non-privileged-map-oa-buffer
+igt@xe_oa@non-sampling-read-error
+igt@xe_oa@non-system-wide-paranoid
+igt@xe_oa@non-zero-reason
+igt@xe_oa@oa-exponents
+igt@xe_oa@oa-formats
+igt@xe_oa@oa-regs-whitelisted
+igt@xe_oa@oa-tlb-invalidate
+igt@xe_oa@oa-unit-concurrent-oa-buffer-read
+igt@xe_oa@oa-unit-exclusive-stream-exec-q
+igt@xe_oa@oa-unit-exclusive-stream-sample-oa
+igt@xe_oa@polling
+igt@xe_oa@polling-small-buf
+igt@xe_oa@privileged-forked-access-vaddr
+igt@xe_oa@rc6-disable
+igt@xe_oa@short-reads
+igt@xe_oa@stress-open-close
+igt@xe_oa@sysctl-defaults
+igt@xe_oa@unprivileged-single-ctx-counters
+igt@xe_oa@whitelisted-registers-userspace-config
+igt@xe_oa@xe-ref-count
+igt@xe_query@multigpu-query-oa-units
+igt@xe_query@query-oa-units
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/index.html
Patchwork
CI✗ CI.xeBAT: failure for Intel Xe OA IGT's
In reply to Ashutosh Dixit== Series Details ==
== Series Details ==
Series: Intel Xe OA IGT's
URL : https://patchwork.freedesktop.org/series/130033/
State : failure
== Summary ==
CI Bug Log - changes from XEIGT_7716_BAT -> XEIGTPW_10690_BAT
====================================================
Summary
-------
**FAILURE**
Serious unknown changes coming with XEIGTPW_10690_BAT absolutely need to be verified manually.
If you think the reported changes have nothing to do with the changes
introduced in XEIGTPW_10690_BAT, please notify your bug team ([email protected]) to allow them to document this new failure mode, which will reduce false positives in CI.
Participating hosts (4 -> 4)
------------------------------
No changes in participating hosts
Possible new issues
-------------------
Here are the unknown changes that may have been introduced in XEIGTPW_10690_BAT:
### IGT changes ###
#### Possible regressions ####
* igt@xe_compute@compute-square:
- bat-atsm-2: [PASS][1] -> [FAIL][2] +150 other tests fail
[1]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-atsm-2/igt@[email protected]
[2]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-atsm-2/igt@[email protected]
* igt@xe_exec_fault_mode@twice-userptr-invalidate-prefetch:
- bat-pvc-2: [PASS][3] -> [FAIL][4] +171 other tests fail
[3]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[4]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
* igt@xe_intel_bb@create-in-region:
- bat-dg2-oem2: [PASS][5] -> [FAIL][6] +201 other tests fail
[5]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-dg2-oem2/igt@[email protected]
[6]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-dg2-oem2/igt@[email protected]
* igt@xe_intel_bb@offset-control:
- bat-adlp-7: [PASS][7] -> [FAIL][8] +184 other tests fail
[7]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-adlp-7/igt@[email protected]
[8]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-adlp-7/igt@[email protected]
#### Warnings ####
* igt@kms_addfb_basic@addfb25-x-tiled-legacy:
- bat-pvc-2: [SKIP][9] ([i915#6077]) -> [FAIL][10] +30 other tests fail
[9]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[10]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
* igt@kms_addfb_basic@addfb25-y-tiled-small-legacy:
- bat-dg2-oem2: [SKIP][11] ([Intel XE#623]) -> [FAIL][12]
[11]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-dg2-oem2/igt@[email protected]
[12]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-dg2-oem2/igt@[email protected]
* igt@kms_addfb_basic@invalid-set-prop-any:
- bat-atsm-2: [SKIP][13] ([i915#6077]) -> [FAIL][14] +30 other tests fail
[13]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-atsm-2/igt@[email protected]
[14]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-atsm-2/igt@[email protected]
* igt@kms_cursor_legacy@basic-flip-after-cursor-atomic:
- bat-pvc-2: [SKIP][15] ([Intel XE#1024] / [Intel XE#782]) -> [FAIL][16] +5 other tests fail
[15]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[16]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
* igt@kms_cursor_legacy@basic-flip-before-cursor-legacy:
- bat-atsm-2: [SKIP][17] ([Intel XE#1024] / [Intel XE#782]) -> [FAIL][18] +5 other tests fail
[17]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-atsm-2/igt@[email protected]
[18]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-atsm-2/igt@[email protected]
* igt@kms_dsc@dsc-basic:
- bat-atsm-2: [SKIP][19] ([Intel XE#1024] / [Intel XE#784]) -> [FAIL][20]
[19]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-atsm-2/igt@[email protected]
[20]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-atsm-2/igt@[email protected]
- bat-pvc-2: [SKIP][21] ([Intel XE#1024] / [Intel XE#784]) -> [FAIL][22]
[21]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[22]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
- bat-dg2-oem2: [SKIP][23] ([Intel XE#455]) -> [FAIL][24]
[23]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-dg2-oem2/igt@[email protected]
[24]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-dg2-oem2/igt@[email protected]
- bat-adlp-7: [SKIP][25] ([Intel XE#455]) -> [FAIL][26]
[25]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-adlp-7/igt@[email protected]
[26]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-adlp-7/igt@[email protected]
* igt@kms_flip@basic-flip-vs-dpms:
- bat-pvc-2: [SKIP][27] ([Intel XE#1024] / [Intel XE#947]) -> [FAIL][28] +3 other tests fail
[27]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[28]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
* igt@kms_flip@basic-flip-vs-modeset:
- bat-atsm-2: [SKIP][29] ([Intel XE#1024] / [Intel XE#947]) -> [FAIL][30] +3 other tests fail
[29]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-atsm-2/igt@[email protected]
[30]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-atsm-2/igt@[email protected]
* igt@kms_force_connector_basic@force-connector-state:
- bat-pvc-2: [SKIP][31] ([Intel XE#540]) -> [FAIL][32] +3 other tests fail
[31]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[32]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
- bat-atsm-2: [SKIP][33] ([Intel XE#540]) -> [FAIL][34] +3 other tests fail
[33]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-atsm-2/igt@[email protected]
[34]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-atsm-2/igt@[email protected]
* igt@kms_force_connector_basic@prune-stale-modes:
- bat-dg2-oem2: [SKIP][35] ([i915#5274]) -> [FAIL][36]
[35]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-dg2-oem2/igt@[email protected]
[36]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-dg2-oem2/igt@[email protected]
* igt@kms_frontbuffer_tracking@basic:
- bat-pvc-2: [SKIP][37] ([Intel XE#1024] / [Intel XE#783]) -> [FAIL][38]
[37]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[38]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
- bat-adlp-7: [FAIL][39] ([Intel XE#616]) -> [FAIL][40]
[39]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-adlp-7/igt@[email protected]
[40]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-adlp-7/igt@[email protected]
- bat-atsm-2: [SKIP][41] ([Intel XE#1024] / [Intel XE#783]) -> [FAIL][42]
[41]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-atsm-2/igt@[email protected]
[42]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-atsm-2/igt@[email protected]
* igt@kms_hdmi_inject@inject-audio:
- bat-dg2-oem2: [SKIP][43] ([Intel XE#417]) -> [FAIL][44]
[43]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-dg2-oem2/igt@[email protected]
[44]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-dg2-oem2/igt@[email protected]
- bat-adlp-7: [SKIP][45] ([Intel XE#417]) -> [FAIL][46]
[45]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-adlp-7/igt@[email protected]
[46]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-adlp-7/igt@[email protected]
* igt@kms_pipe_crc_basic@compare-crc-sanitycheck-xr24:
- bat-atsm-2: [SKIP][47] ([i915#1836]) -> [FAIL][48] +6 other tests fail
[47]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-atsm-2/igt@[email protected]
[48]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-atsm-2/igt@[email protected]
* igt@kms_pipe_crc_basic@nonblocking-crc-frame-sequence:
- bat-pvc-2: [SKIP][49] ([Intel XE#829]) -> [FAIL][50] +6 other tests fail
[49]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[50]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
* igt@kms_prop_blob@basic:
- bat-pvc-2: [SKIP][51] ([Intel XE#780]) -> [FAIL][52]
[51]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[52]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
- bat-atsm-2: [SKIP][53] ([Intel XE#780]) -> [FAIL][54]
[53]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-atsm-2/igt@[email protected]
[54]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-atsm-2/igt@[email protected]
* igt@xe_evict@evict-beng-small-external:
- bat-pvc-2: [FAIL][55] ([Intel XE#1000]) -> [FAIL][56] +3 other tests fail
[55]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[56]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
- bat-adlp-7: [SKIP][57] ([Intel XE#261] / [Intel XE#688]) -> [FAIL][58] +15 other tests fail
[57]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-adlp-7/igt@[email protected]
[58]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-adlp-7/igt@[email protected]
* igt@xe_evict@evict-small-cm:
- bat-pvc-2: [DMESG-FAIL][59] ([Intel XE#482]) -> [FAIL][60] +3 other tests fail
[59]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[60]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
* igt@xe_evict_ccs@evict-overcommit-parallel-nofree-samefd:
- bat-pvc-2: [INCOMPLETE][61] ([Intel XE#392]) -> [FAIL][62]
[61]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[62]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
* igt@xe_evict_ccs@evict-overcommit-simple:
- bat-adlp-7: [SKIP][63] ([Intel XE#688]) -> [FAIL][64] +1 other test fail
[63]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-adlp-7/igt@[email protected]
[64]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-adlp-7/igt@[email protected]
* igt@xe_exec_fault_mode@many-basic:
- bat-dg2-oem2: [SKIP][65] ([Intel XE#288]) -> [FAIL][66] +32 other tests fail
[65]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-dg2-oem2/igt@[email protected]
[66]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-dg2-oem2/igt@[email protected]
* igt@xe_exec_fault_mode@twice-userptr:
- bat-adlp-7: [SKIP][67] ([Intel XE#288]) -> [FAIL][68] +32 other tests fail
[67]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-adlp-7/igt@[email protected]
[68]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-adlp-7/igt@[email protected]
* igt@xe_exec_fault_mode@twice-userptr-invalidate-imm:
- bat-atsm-2: [SKIP][69] ([Intel XE#288]) -> [FAIL][70] +32 other tests fail
[69]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-atsm-2/igt@[email protected]
[70]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-atsm-2/igt@[email protected]
* igt@xe_huc_copy@huc_copy:
- bat-pvc-2: [SKIP][71] ([Intel XE#255]) -> [FAIL][72]
[71]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@xe_huc_copy@huc_copy.html
[72]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@xe_huc_copy@huc_copy.html
- bat-dg2-oem2: [SKIP][73] ([Intel XE#255]) -> [FAIL][74]
[73]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-dg2-oem2/igt@xe_huc_copy@huc_copy.html
[74]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-dg2-oem2/igt@xe_huc_copy@huc_copy.html
- bat-atsm-2: [SKIP][75] ([Intel XE#255]) -> [FAIL][76]
[75]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-atsm-2/igt@xe_huc_copy@huc_copy.html
[76]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-atsm-2/igt@xe_huc_copy@huc_copy.html
* igt@xe_intel_bb@render:
- bat-pvc-2: [SKIP][77] ([Intel XE#532]) -> [FAIL][78]
[77]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[78]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
* igt@xe_mmap@vram:
- bat-adlp-7: [SKIP][79] ([Intel XE#1008]) -> [FAIL][80]
[79]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-adlp-7/igt@[email protected]
[80]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-adlp-7/igt@[email protected]
* igt@xe_pat@pat-index-xe2:
- bat-pvc-2: [SKIP][81] ([Intel XE#977]) -> [FAIL][82] +1 other test fail
[81]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[82]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
- bat-adlp-7: [SKIP][83] ([Intel XE#977]) -> [FAIL][84]
[83]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-adlp-7/igt@[email protected]
[84]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-adlp-7/igt@[email protected]
- bat-atsm-2: [SKIP][85] ([Intel XE#977]) -> [FAIL][86]
[85]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-atsm-2/igt@[email protected]
[86]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-atsm-2/igt@[email protected]
- bat-dg2-oem2: [SKIP][87] ([Intel XE#977]) -> [FAIL][88]
[87]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-dg2-oem2/igt@[email protected]
[88]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-dg2-oem2/igt@[email protected]
* igt@xe_pat@pat-index-xehpc:
- bat-dg2-oem2: [SKIP][89] ([Intel XE#979]) -> [FAIL][90] +1 other test fail
[89]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-dg2-oem2/igt@[email protected]
[90]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-dg2-oem2/igt@[email protected]
- bat-adlp-7: [SKIP][91] ([Intel XE#979]) -> [FAIL][92] +1 other test fail
[91]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-adlp-7/igt@[email protected]
[92]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-adlp-7/igt@[email protected]
* igt@xe_pat@pat-index-xelpg:
- bat-atsm-2: [SKIP][93] ([Intel XE#979]) -> [FAIL][94] +1 other test fail
[93]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-atsm-2/igt@[email protected]
[94]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-atsm-2/igt@[email protected]
- bat-pvc-2: [SKIP][95] ([Intel XE#979]) -> [FAIL][96]
[95]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[96]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
* igt@xe_pm_residency@gt-c6-on-idle:
- bat-pvc-2: [SKIP][97] ([Intel XE#531]) -> [FAIL][98]
[97]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[98]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
#### Suppressed ####
The following results come from untrusted machines, tests, or statuses. They do not affect the overall result.
* {igt@kms_psr@psr-cursor-plane-move}:
- bat-pvc-2: [SKIP][99] ([Intel XE#1024]) -> [FAIL][100] +2 other tests fail
[99]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[100]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
- bat-dg2-oem2: [SKIP][101] ([Intel XE#929]) -> [FAIL][102] +2 other tests fail
[101]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-dg2-oem2/igt@[email protected]
[102]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-dg2-oem2/igt@[email protected]
- bat-adlp-7: [PASS][103] -> [FAIL][104] +8 other tests fail
[103]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-adlp-7/igt@[email protected]
[104]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-adlp-7/igt@[email protected]
* {igt@kms_psr@psr-primary-page-flip}:
- bat-atsm-2: [SKIP][105] ([Intel XE#1024]) -> [FAIL][106] +2 other tests fail
[105]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-atsm-2/igt@[email protected]
[106]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-atsm-2/igt@[email protected]
* {igt@xe_create@create-invalid-mbz}:
- bat-atsm-2: [PASS][107] -> [FAIL][108] +5 other tests fail
[107]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-atsm-2/igt@[email protected]
[108]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-atsm-2/igt@[email protected]
* {igt@xe_exec_atomic@basic-dec-all}:
- bat-pvc-2: [PASS][109] -> [FAIL][110] +3 other tests fail
[109]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@[email protected]
[110]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@[email protected]
- bat-dg2-oem2: [PASS][111] -> [FAIL][112] +5 other tests fail
[111]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-dg2-oem2/igt@[email protected]
[112]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-dg2-oem2/igt@[email protected]
* {igt@xe_gt_freq@freq_range_idle}:
- bat-pvc-2: [SKIP][113] ([Intel XE#1021]) -> [FAIL][114] +1 other test fail
[113]: https://intel-gfx-ci.01.org/tree/intel-xe/IGT_7716/bat-pvc-2/igt@xe_gt_freq@freq_range_idle.html
[114]: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/bat-pvc-2/igt@xe_gt_freq@freq_range_idle.html
{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).
[Intel XE#1000]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1000
[Intel XE#1008]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1008
[Intel XE#1021]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1021
[Intel XE#1024]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1024
[Intel XE#255]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/255
[Intel XE#261]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/261
[Intel XE#288]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/288
[Intel XE#392]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/392
[Intel XE#417]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/417
[Intel XE#455]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/455
[Intel XE#482]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/482
[Intel XE#531]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/531
[Intel XE#532]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/532
[Intel XE#540]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/540
[Intel XE#616]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/616
[Intel XE#623]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/623
[Intel XE#688]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/688
[Intel XE#780]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/780
[Intel XE#782]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/782
[Intel XE#783]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/783
[Intel XE#784]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/784
[Intel XE#829]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/829
[Intel XE#929]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/929
[Intel XE#947]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/947
[Intel XE#977]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/977
[Intel XE#979]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/979
[i915#1836]: https://gitlab.freedesktop.org/drm/intel/issues/1836
[i915#5274]: https://gitlab.freedesktop.org/drm/intel/issues/5274
[i915#6077]: https://gitlab.freedesktop.org/drm/intel/issues/6077
Build changes
-------------
* IGT: IGT_7716 -> IGTPW_10690
* Linux: xe-789-55724f2a1075fcace79fd49ea0c2b7b5600d7ba8 -> xe-794-b09fd1bd9f8b4a8e5cd8c1de8dbba4d4eb15b4e9
IGTPW_10690: 10690
IGT_7716: 7716
xe-789-55724f2a1075fcace79fd49ea0c2b7b5600d7ba8: 55724f2a1075fcace79fd49ea0c2b7b5600d7ba8 xe-794-b09fd1bd9f8b4a8e5cd8c1de8dbba4d4eb15b4e9: b09fd1bd9f8b4a8e5cd8c1de8dbba4d4eb15b4e9
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/intel-xe/IGTPW_10690/index.html
Patchwork
CI✗ Fi.CI.IGT: failure for Intel Xe OA IGT's
In reply to Ashutosh Dixit== Series Details ==
== Series Details ==
Series: Intel Xe OA IGT's
URL : https://patchwork.freedesktop.org/series/130033/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_14287_full -> IGTPW_10690_full
====================================================
Summary
-------
**FAILURE**
Serious unknown changes coming with IGTPW_10690_full absolutely need to be verified manually.
If you think the reported changes have nothing to do with the changes
introduced in IGTPW_10690_full, please notify your bug team ([email protected]) to allow them to document this new failure mode, which will reduce false positives in CI.
External URL: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/index.html
Participating hosts (7 -> 8)
------------------------------
Additional (1): shard-glk-0
Possible new issues
-------------------
Here are the unknown changes that may have been introduced in IGTPW_10690_full:
### IGT changes ###
#### Possible regressions ####
* igt@api_intel_bb@render-ccs:
- shard-dg2: NOTRUN -> [FAIL][1]
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
#### Suppressed ####
The following results come from untrusted machines, tests, or statuses. They do not affect the overall result.
* {igt@kms_psr@psr2-no-drrs@edp-1}:
- shard-mtlp: [PASS][2] -> [FAIL][3]
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-mtlp-7/igt@kms_psr@[email protected]
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-8/igt@kms_psr@[email protected]
Known issues
------------
Here are the changes found in IGTPW_10690_full that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@api_intel_bb@crc32:
- shard-tglu: NOTRUN -> [SKIP][4] ([i915#6230])
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-2/igt@[email protected]
* igt@drm_fdinfo@most-busy-check-all@bcs0:
- shard-dg1: NOTRUN -> [SKIP][5] ([i915#8414]) +4 other tests skip
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-16/igt@drm_fdinfo@[email protected]
* igt@drm_fdinfo@most-busy-idle-check-all@rcs0:
- shard-rkl: [PASS][6] -> [FAIL][7] ([i915#7742])
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-rkl-5/igt@drm_fdinfo@[email protected]
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@drm_fdinfo@[email protected]
* igt@drm_fdinfo@virtual-busy-all:
- shard-dg2: NOTRUN -> [SKIP][8] ([i915#8414]) +4 other tests skip
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-5/igt@[email protected]
* igt@gem_basic@multigpu-create-close:
- shard-dg1: NOTRUN -> [SKIP][9] ([i915#7697])
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-17/igt@[email protected]
* igt@gem_caching@read-writes:
- shard-mtlp: NOTRUN -> [SKIP][10] ([i915#4873])
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-6/igt@[email protected]
* igt@gem_ccs@ctrl-surf-copy:
- shard-rkl: NOTRUN -> [SKIP][11] ([i915#3555] / [i915#9323])
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-3/igt@[email protected]
* igt@gem_create@create-ext-cpu-access-big:
- shard-dg2: NOTRUN -> [ABORT][12] ([i915#10183])
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-3/igt@[email protected]
- shard-rkl: NOTRUN -> [SKIP][13] ([i915#6335])
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-7/igt@[email protected]
* igt@gem_create@create-ext-set-pat:
- shard-dg1: NOTRUN -> [SKIP][14] ([i915#8562])
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-19/igt@[email protected]
* igt@gem_ctx_param@set-priority-not-supported:
- shard-dg2: NOTRUN -> [SKIP][15] ([fdo#109314])
[15]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-5/igt@[email protected]
* igt@gem_ctx_persistence@engines-mixed-process:
- shard-snb: NOTRUN -> [SKIP][16] ([fdo#109271] / [i915#1099]) +1 other test skip
[16]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-snb6/igt@[email protected]
* igt@gem_ctx_sseu@engines:
- shard-rkl: NOTRUN -> [SKIP][17] ([i915#280])
[17]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-1/igt@[email protected]
- shard-tglu: NOTRUN -> [SKIP][18] ([i915#280])
[18]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-2/igt@[email protected]
- shard-dg2: NOTRUN -> [SKIP][19] ([i915#280])
[19]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-10/igt@[email protected]
* igt@gem_eio@hibernate:
- shard-dg2: NOTRUN -> [ABORT][20] ([i915#10030] / [i915#7975] / [i915#8213])
[20]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-1/igt@[email protected]
* igt@gem_exec_balancer@noheartbeat:
- shard-dg2: NOTRUN -> [SKIP][21] ([i915#8555])
[21]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-5/igt@[email protected]
* igt@gem_exec_balancer@parallel-keep-in-fence:
- shard-rkl: NOTRUN -> [SKIP][22] ([i915#4525]) +1 other test skip
[22]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-4/igt@[email protected]
* igt@gem_exec_capture@capture-invisible@smem0:
- shard-rkl: NOTRUN -> [SKIP][23] ([i915#6334])
[23]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-4/igt@gem_exec_capture@[email protected]
* igt@gem_exec_capture@capture-recoverable:
- shard-tglu: NOTRUN -> [SKIP][24] ([i915#6344])
[24]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-8/igt@[email protected]
* igt@gem_exec_capture@many-4k-incremental:
- shard-dg1: NOTRUN -> [FAIL][25] ([i915#9606])
[25]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-19/igt@[email protected]
* igt@gem_exec_endless@dispatch@bcs0:
- shard-dg2: [PASS][26] -> [TIMEOUT][27] ([i915#3778] / [i915#7016])
[26]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-dg2-3/igt@gem_exec_endless@[email protected]
[27]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-6/igt@gem_exec_endless@[email protected]
* igt@gem_exec_fair@basic-none:
- shard-dg1: NOTRUN -> [SKIP][28] ([i915#3539] / [i915#4852])
[28]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-19/igt@[email protected]
* igt@gem_exec_fair@basic-none-rrul:
- shard-dg2: NOTRUN -> [SKIP][29] ([i915#3539] / [i915#4852]) +4 other tests skip
[29]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-10/igt@[email protected]
* igt@gem_exec_fair@basic-none-solo@rcs0:
- shard-rkl: NOTRUN -> [FAIL][30] ([i915#2842])
[30]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@gem_exec_fair@[email protected]
* igt@gem_exec_fair@basic-none@bcs0:
- shard-rkl: [PASS][31] -> [FAIL][32] ([i915#2842]) +1 other test fail
[31]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-rkl-4/igt@gem_exec_fair@[email protected]
[32]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-7/igt@gem_exec_fair@[email protected]
* igt@gem_exec_fair@basic-pace-share:
- shard-mtlp: NOTRUN -> [SKIP][33] ([i915#4473] / [i915#4771])
[33]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-6/igt@[email protected]
* igt@gem_exec_fair@basic-pace-share@rcs0:
- shard-glk: [PASS][34] -> [FAIL][35] ([i915#2842])
[34]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-glk8/igt@gem_exec_fair@[email protected]
[35]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-glk3/igt@gem_exec_fair@[email protected]
* igt@gem_exec_flush@basic-uc-prw-default:
- shard-dg2: NOTRUN -> [SKIP][36] ([i915#3539])
[36]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-6/igt@[email protected]
* igt@gem_exec_flush@basic-uc-set-default:
- shard-dg1: NOTRUN -> [SKIP][37] ([i915#3539])
[37]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-19/igt@[email protected]
* igt@gem_exec_reloc@basic-active:
- shard-dg2: NOTRUN -> [SKIP][38] ([i915#3281]) +11 other tests skip
[38]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-5/igt@[email protected]
* igt@gem_exec_reloc@basic-gtt-read-noreloc:
- shard-rkl: NOTRUN -> [SKIP][39] ([i915#3281]) +8 other tests skip
[39]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-7/igt@[email protected]
* igt@gem_exec_reloc@basic-wc-read-active:
- shard-mtlp: NOTRUN -> [SKIP][40] ([i915#3281])
[40]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-1/igt@[email protected]
* igt@gem_exec_reloc@basic-write-read-active:
- shard-dg1: NOTRUN -> [SKIP][41] ([i915#3281]) +3 other tests skip
[41]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-13/igt@[email protected]
* igt@gem_exec_suspend@basic-s4-devices@lmem0:
- shard-dg2: NOTRUN -> [ABORT][42] ([i915#7975] / [i915#8213])
[42]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-1/igt@gem_exec_suspend@[email protected]
* igt@gem_fenced_exec_thrash@no-spare-fences-busy-interruptible:
- shard-mtlp: NOTRUN -> [SKIP][43] ([i915#4860])
[43]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-1/igt@[email protected]
* igt@gem_fenced_exec_thrash@too-many-fences:
- shard-dg2: NOTRUN -> [SKIP][44] ([i915#4860]) +2 other tests skip
[44]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-3/igt@[email protected]
* igt@gem_lmem_swapping@parallel-random-verify-ccs:
- shard-rkl: NOTRUN -> [SKIP][45] ([i915#4613]) +4 other tests skip
[45]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-1/igt@[email protected]
* igt@gem_lmem_swapping@smem-oom@lmem0:
- shard-dg1: [PASS][46] -> [TIMEOUT][47] ([i915#5493])
[46]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-dg1-15/igt@gem_lmem_swapping@[email protected]
[47]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-12/igt@gem_lmem_swapping@[email protected]
* igt@gem_lmem_swapping@verify-ccs:
- shard-tglu: NOTRUN -> [SKIP][48] ([i915#4613]) +2 other tests skip
[48]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-9/igt@[email protected]
* igt@gem_mmap_gtt@basic:
- shard-mtlp: NOTRUN -> [SKIP][49] ([i915#4077]) +3 other tests skip
[49]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-2/igt@[email protected]
* igt@gem_mmap_gtt@cpuset-big-copy-odd:
- shard-dg1: NOTRUN -> [SKIP][50] ([i915#4077])
[50]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-17/igt@[email protected]
* igt@gem_mmap_gtt@hang:
- shard-dg2: NOTRUN -> [SKIP][51] ([i915#4077]) +9 other tests skip
[51]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-3/igt@[email protected]
* igt@gem_mmap_wc@bad-size:
- shard-dg2: NOTRUN -> [SKIP][52] ([i915#4083]) +2 other tests skip
[52]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
* igt@gem_mmap_wc@write-gtt-read-wc:
- shard-mtlp: NOTRUN -> [SKIP][53] ([i915#4083]) +2 other tests skip
[53]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-1/igt@[email protected]
* igt@gem_partial_pwrite_pread@write:
- shard-dg2: NOTRUN -> [SKIP][54] ([i915#3282]) +1 other test skip
[54]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-10/igt@[email protected]
* igt@gem_partial_pwrite_pread@write-display:
- shard-mtlp: NOTRUN -> [SKIP][55] ([i915#3282]) +2 other tests skip
[55]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-8/igt@[email protected]
* igt@gem_partial_pwrite_pread@writes-after-reads:
- shard-rkl: NOTRUN -> [SKIP][56] ([i915#3282]) +3 other tests skip
[56]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-1/igt@[email protected]
* igt@gem_pxp@create-valid-protected-context:
- shard-mtlp: NOTRUN -> [SKIP][57] ([i915#4270]) +1 other test skip
[57]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-1/igt@[email protected]
* igt@gem_pxp@display-protected-crc:
- shard-dg2: NOTRUN -> [SKIP][58] ([i915#4270]) +3 other tests skip
[58]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-1/igt@[email protected]
* igt@gem_pxp@reject-modify-context-protection-off-3:
- shard-dg1: NOTRUN -> [SKIP][59] ([i915#4270])
[59]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-16/igt@[email protected]
* igt@gem_pxp@verify-pxp-execution-after-suspend-resume:
- shard-rkl: NOTRUN -> [SKIP][60] ([i915#4270]) +1 other test skip
[60]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-2/igt@[email protected]
* igt@gem_render_copy@y-tiled-ccs-to-y-tiled:
- shard-dg2: NOTRUN -> [SKIP][61] ([i915#5190]) +9 other tests skip
[61]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
* igt@gem_set_tiling_vs_blt@tiled-to-untiled:
- shard-dg1: NOTRUN -> [SKIP][62] ([i915#4079])
[62]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-17/igt@[email protected]
* igt@gem_set_tiling_vs_pwrite:
- shard-dg2: NOTRUN -> [SKIP][63] ([i915#4079])
[63]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-6/igt@gem_set_tiling_vs_pwrite.html
* igt@gem_userptr_blits@access-control:
- shard-dg1: NOTRUN -> [SKIP][64] ([i915#3297])
[64]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-16/igt@[email protected]
* igt@gem_userptr_blits@dmabuf-unsync:
- shard-dg2: NOTRUN -> [SKIP][65] ([i915#3297]) +2 other tests skip
[65]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
* igt@gem_userptr_blits@forbidden-operations:
- shard-dg1: NOTRUN -> [SKIP][66] ([i915#3282] / [i915#3297])
[66]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-17/igt@[email protected]
* igt@gem_userptr_blits@invalid-mmap-offset-unsync:
- shard-tglu: NOTRUN -> [SKIP][67] ([i915#3297])
[67]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-3/igt@[email protected]
* igt@gem_userptr_blits@map-fixed-invalidate-busy:
- shard-dg2: NOTRUN -> [SKIP][68] ([i915#3297] / [i915#4880])
[68]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
* igt@gem_userptr_blits@readonly-pwrite-unsync:
- shard-mtlp: NOTRUN -> [SKIP][69] ([i915#3297])
[69]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-3/igt@[email protected]
* igt@gem_userptr_blits@unsync-overlap:
- shard-rkl: NOTRUN -> [SKIP][70] ([i915#3297])
[70]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-1/igt@[email protected]
* igt@gen3_render_mixed_blits:
- shard-rkl: NOTRUN -> [SKIP][71] ([fdo#109289]) +1 other test skip
[71]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@gen3_render_mixed_blits.html
* igt@gen7_exec_parse@basic-offset:
- shard-dg2: NOTRUN -> [SKIP][72] ([fdo#109289]) +3 other tests skip
[72]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-5/igt@[email protected]
* igt@gen9_exec_parse@batch-zero-length:
- shard-mtlp: NOTRUN -> [SKIP][73] ([i915#2856])
[73]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-1/igt@[email protected]
* igt@gen9_exec_parse@bb-secure:
- shard-dg1: NOTRUN -> [SKIP][74] ([i915#2527]) +1 other test skip
[74]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-13/igt@[email protected]
* igt@gen9_exec_parse@secure-batches:
- shard-dg2: NOTRUN -> [SKIP][75] ([i915#2856]) +3 other tests skip
[75]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-6/igt@[email protected]
* igt@gen9_exec_parse@shadow-peek:
- shard-rkl: NOTRUN -> [SKIP][76] ([i915#2527]) +2 other tests skip
[76]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-1/igt@[email protected]
* igt@i915_module_load@reload-with-fault-injection:
- shard-rkl: NOTRUN -> [ABORT][77] ([i915#9820])
[77]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@[email protected]
- shard-tglu: NOTRUN -> [INCOMPLETE][78] ([i915#10137] / [i915#9200])
[78]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-4/igt@[email protected]
- shard-dg2: NOTRUN -> [INCOMPLETE][79] ([i915#10137] / [i915#9820] / [i915#9849])
[79]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
* igt@i915_pm_freq_api@freq-reset-multiple:
- shard-rkl: NOTRUN -> [SKIP][80] ([i915#8399]) +1 other test skip
[80]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-7/igt@[email protected]
* igt@i915_pm_rc6_residency@media-rc6-accuracy:
- shard-dg1: NOTRUN -> [SKIP][81] ([fdo#109289])
[81]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-12/igt@[email protected]
* igt@i915_pm_rps@thresholds-idle@gt0:
- shard-mtlp: NOTRUN -> [SKIP][82] ([i915#8925])
[82]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-4/igt@i915_pm_rps@[email protected]
* igt@i915_pm_rps@thresholds-idle@gt1:
- shard-mtlp: NOTRUN -> [SKIP][83] ([i915#3555] / [i915#8925])
[83]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-4/igt@i915_pm_rps@[email protected]
* igt@i915_pm_sseu@full-enable:
- shard-dg2: NOTRUN -> [SKIP][84] ([i915#4387])
[84]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
* igt@i915_power@sanity:
- shard-rkl: NOTRUN -> [SKIP][85] ([i915#7984])
[85]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-7/igt@[email protected]
* igt@i915_query@query-topology-unsupported:
- shard-dg2: NOTRUN -> [SKIP][86] ([fdo#109302])
[86]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-6/igt@[email protected]
* igt@i915_suspend@basic-s3-without-i915:
- shard-rkl: [PASS][87] -> [FAIL][88] ([i915#10031])
[87]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-rkl-1/igt@[email protected]
[88]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-6/igt@[email protected]
* igt@intel_hwmon@hwmon-read:
- shard-rkl: NOTRUN -> [SKIP][89] ([i915#7707])
[89]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-1/igt@[email protected]
* igt@kms_addfb_basic@basic-x-tiled-legacy:
- shard-mtlp: NOTRUN -> [SKIP][90] ([i915#4212])
[90]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-1/igt@[email protected]
* igt@kms_addfb_basic@clobberred-modifier:
- shard-dg2: NOTRUN -> [SKIP][91] ([i915#4212]) +1 other test skip
[91]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
* igt@kms_async_flips@async-flip-with-page-flip-events@pipe-a-hdmi-a-1-y-rc-ccs-cc:
- shard-rkl: NOTRUN -> [SKIP][92] ([i915#8709]) +3 other tests skip
[92]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@kms_async_flips@async-flip-with-page-flip-events@pipe-a-hdmi-a-1-y-rc-ccs-cc.html
* igt@kms_async_flips@async-flip-with-page-flip-events@pipe-d-hdmi-a-2-4-mc-ccs:
- shard-dg2: NOTRUN -> [SKIP][93] ([i915#8709]) +11 other tests skip
[93]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@kms_async_flips@[email protected]
* igt@kms_async_flips@invalid-async-flip:
- shard-dg2: NOTRUN -> [SKIP][94] ([i915#6228])
[94]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-6/igt@[email protected]
* igt@kms_atomic_transition@plane-all-modeset-transition:
- shard-mtlp: NOTRUN -> [SKIP][95] ([i915#1769] / [i915#3555])
[95]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-8/igt@[email protected]
* igt@kms_big_fb@4-tiled-32bpp-rotate-0:
- shard-rkl: NOTRUN -> [SKIP][96] ([i915#5286]) +3 other tests skip
[96]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@[email protected]
* igt@kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-0-hflip-async-flip:
- shard-dg1: NOTRUN -> [SKIP][97] ([i915#4538] / [i915#5286])
[97]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-18/igt@kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-0-hflip-async-flip.html
* igt@kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-180:
- shard-tglu: NOTRUN -> [SKIP][98] ([fdo#111615] / [i915#5286]) +1 other test skip
[98]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-2/igt@[email protected]
* igt@kms_big_fb@linear-16bpp-rotate-0:
- shard-mtlp: [PASS][99] -> [FAIL][100] ([i915#5138])
[99]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-mtlp-6/igt@[email protected]
[100]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-3/igt@[email protected]
* igt@kms_big_fb@linear-16bpp-rotate-90:
- shard-rkl: NOTRUN -> [SKIP][101] ([fdo#111614] / [i915#3638])
[101]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-6/igt@[email protected]
* igt@kms_big_fb@x-tiled-16bpp-rotate-270:
- shard-dg1: NOTRUN -> [SKIP][102] ([i915#3638])
[102]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-17/igt@[email protected]
* igt@kms_big_fb@x-tiled-16bpp-rotate-90:
- shard-dg2: NOTRUN -> [SKIP][103] ([fdo#111614]) +4 other tests skip
[103]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-6/igt@[email protected]
* igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-async-flip:
- shard-tglu: [PASS][104] -> [FAIL][105] ([i915#3743]) +1 other test fail
[104]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-tglu-2/igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-async-flip.html
[105]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-3/igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-async-flip.html
* igt@kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-0-hflip:
- shard-mtlp: NOTRUN -> [SKIP][106] ([fdo#111615]) +2 other tests skip
[106]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-1/igt@[email protected]
* igt@kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-0-hflip-async-flip:
- shard-dg2: NOTRUN -> [SKIP][107] ([i915#4538] / [i915#5190]) +13 other tests skip
[107]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-10/igt@kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-0-hflip-async-flip.html
* igt@kms_big_fb@yf-tiled-32bpp-rotate-90:
- shard-rkl: NOTRUN -> [SKIP][108] ([fdo#110723]) +1 other test skip
[108]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@[email protected]
* igt@kms_big_fb@yf-tiled-addfb:
- shard-rkl: NOTRUN -> [SKIP][109] ([fdo#111615]) +1 other test skip
[109]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-4/igt@[email protected]
* igt@kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-180-hflip-async-flip:
- shard-tglu: NOTRUN -> [SKIP][110] ([fdo#111615]) +3 other tests skip
[110]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-6/igt@kms_big_fb@yf-tiled-max-hw-stride-32bpp-rotate-180-hflip-async-flip.html
* igt@kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-0:
- shard-dg1: NOTRUN -> [SKIP][111] ([i915#4538]) +1 other test skip
[111]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-17/igt@[email protected]
* igt@kms_big_joiner@invalid-modeset:
- shard-rkl: NOTRUN -> [SKIP][112] ([i915#2705])
[112]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@[email protected]
* igt@kms_ccs@pipe-a-bad-pixel-format-4-tiled-dg2-rc-ccs-cc:
- shard-mtlp: NOTRUN -> [SKIP][113] ([i915#5354] / [i915#6095]) +11 other tests skip
[113]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-8/igt@[email protected]
* igt@kms_ccs@pipe-b-bad-rotation-90-4-tiled-mtl-rc-ccs-cc:
- shard-rkl: NOTRUN -> [SKIP][114] ([i915#5354] / [i915#6095]) +17 other tests skip
[114]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-7/igt@[email protected]
* igt@kms_ccs@pipe-b-crc-primary-basic-4-tiled-mtl-rc-ccs-cc:
- shard-dg2: NOTRUN -> [SKIP][115] ([i915#5354]) +74 other tests skip
[115]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-6/igt@[email protected]
* igt@kms_ccs@pipe-b-random-ccs-data-4-tiled-mtl-rc-ccs-cc:
- shard-dg1: NOTRUN -> [SKIP][116] ([i915#5354] / [i915#6095]) +13 other tests skip
[116]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-12/igt@[email protected]
* igt@kms_ccs@pipe-c-missing-ccs-buffer-y-tiled-gen12-mc-ccs:
- shard-tglu: NOTRUN -> [SKIP][117] ([i915#5354] / [i915#6095]) +16 other tests skip
[117]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-6/igt@[email protected]
* igt@kms_ccs@pipe-d-bad-pixel-format-4-tiled-mtl-rc-ccs:
- shard-snb: NOTRUN -> [SKIP][118] ([fdo#109271]) +121 other tests skip
[118]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-snb2/igt@[email protected]
* igt@kms_ccs@pipe-d-random-ccs-data-4-tiled-mtl-rc-ccs-cc:
- shard-rkl: NOTRUN -> [SKIP][119] ([i915#5354]) +27 other tests skip
[119]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-7/igt@[email protected]
* igt@kms_cdclk@mode-transition:
- shard-dg1: NOTRUN -> [SKIP][120] ([i915#3742])
[120]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-18/igt@[email protected]
* igt@kms_cdclk@mode-transition@pipe-d-hdmi-a-1:
- shard-dg2: NOTRUN -> [SKIP][121] ([i915#7213]) +3 other tests skip
[121]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-10/igt@kms_cdclk@[email protected]
* igt@kms_cdclk@plane-scaling:
- shard-rkl: NOTRUN -> [SKIP][122] ([i915#3742])
[122]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-2/igt@[email protected]
* igt@kms_cdclk@plane-scaling@pipe-d-hdmi-a-1:
- shard-dg2: NOTRUN -> [SKIP][123] ([i915#4087]) +3 other tests skip
[123]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-10/igt@kms_cdclk@[email protected]
* igt@kms_chamelium_audio@dp-audio:
- shard-tglu: NOTRUN -> [SKIP][124] ([i915#7828]) +4 other tests skip
[124]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-5/igt@[email protected]
* igt@kms_chamelium_color@ctm-0-75:
- shard-dg2: NOTRUN -> [SKIP][125] ([fdo#111827]) +1 other test skip
[125]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-5/igt@[email protected]
* igt@kms_chamelium_color@ctm-limited-range:
- shard-rkl: NOTRUN -> [SKIP][126] ([fdo#111827])
[126]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-1/igt@[email protected]
* igt@kms_chamelium_color@ctm-red-to-blue:
- shard-mtlp: NOTRUN -> [SKIP][127] ([fdo#111827])
[127]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-6/igt@[email protected]
* igt@kms_chamelium_edid@hdmi-edid-stress-resolution-non-4k:
- shard-dg1: NOTRUN -> [SKIP][128] ([i915#7828])
[128]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-17/igt@[email protected]
* igt@kms_chamelium_frames@dp-crc-fast:
- shard-dg2: NOTRUN -> [SKIP][129] ([i915#7828]) +9 other tests skip
[129]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-5/igt@[email protected]
* igt@kms_chamelium_hpd@vga-hpd-for-each-pipe:
- shard-rkl: NOTRUN -> [SKIP][130] ([i915#7828]) +6 other tests skip
[130]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@[email protected]
* igt@kms_chamelium_hpd@vga-hpd-without-ddc:
- shard-mtlp: NOTRUN -> [SKIP][131] ([i915#7828]) +1 other test skip
[131]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-6/igt@[email protected]
* igt@kms_color@deep-color:
- shard-dg2: NOTRUN -> [SKIP][132] ([i915#3555]) +3 other tests skip
[132]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
* igt@kms_content_protection@atomic-dpms:
- shard-rkl: NOTRUN -> [SKIP][133] ([i915#7118] / [i915#9424])
[133]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-1/igt@[email protected]
* igt@kms_content_protection@content-type-change:
- shard-tglu: NOTRUN -> [SKIP][134] ([i915#6944] / [i915#9424])
[134]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-3/igt@[email protected]
* igt@kms_content_protection@mei-interface:
- shard-dg1: NOTRUN -> [SKIP][135] ([i915#9433])
[135]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-17/igt@[email protected]
* igt@kms_cursor_crc@cursor-offscreen-512x512:
- shard-dg1: NOTRUN -> [SKIP][136] ([i915#3359])
[136]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-15/igt@[email protected]
* igt@kms_cursor_crc@cursor-random-32x10:
- shard-mtlp: NOTRUN -> [SKIP][137] ([i915#3555] / [i915#8814])
[137]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-2/igt@[email protected]
* igt@kms_cursor_crc@cursor-random-512x512:
- shard-dg2: NOTRUN -> [SKIP][138] ([i915#3359]) +2 other tests skip
[138]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
* igt@kms_cursor_crc@cursor-rapid-movement-512x170:
- shard-rkl: NOTRUN -> [SKIP][139] ([i915#3359])
[139]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-1/igt@[email protected]
* igt@kms_cursor_crc@cursor-sliding-32x10:
- shard-rkl: NOTRUN -> [SKIP][140] ([i915#3555]) +3 other tests skip
[140]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@[email protected]
* igt@kms_cursor_legacy@2x-long-flip-vs-cursor-atomic:
- shard-dg2: NOTRUN -> [SKIP][141] ([fdo#109274] / [i915#5354]) +4 other tests skip
[141]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-3/igt@[email protected]
* igt@kms_cursor_legacy@2x-long-nonblocking-modeset-vs-cursor-atomic:
- shard-rkl: NOTRUN -> [SKIP][142] ([fdo#111825]) +10 other tests skip
[142]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@kms_cursor_legacy@2x-long-nonblocking-modeset-vs-cursor-atomic.html
* igt@kms_cursor_legacy@basic-busy-flip-before-cursor-varying-size:
- shard-rkl: NOTRUN -> [SKIP][143] ([i915#4103]) +1 other test skip
[143]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-4/igt@kms_cursor_legacy@basic-busy-flip-before-cursor-varying-size.html
* igt@kms_cursor_legacy@cursorb-vs-flipa-toggle:
- shard-tglu: NOTRUN -> [SKIP][144] ([fdo#109274])
[144]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-3/igt@[email protected]
* igt@kms_cursor_legacy@cursorb-vs-flipb-toggle:
- shard-rkl: NOTRUN -> [SKIP][145] ([fdo#111767] / [fdo#111825]) +1 other test skip
[145]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-7/igt@[email protected]
* igt@kms_cursor_legacy@short-busy-flip-before-cursor-toggle:
- shard-tglu: NOTRUN -> [SKIP][146] ([i915#4103])
[146]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-4/igt@[email protected]
- shard-glk: NOTRUN -> [SKIP][147] ([fdo#109271]) +17 other tests skip
[147]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-glk4/igt@[email protected]
- shard-dg2: NOTRUN -> [SKIP][148] ([i915#4103] / [i915#4213]) +1 other test skip
[148]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
* igt@kms_dirtyfb@fbc-dirtyfb-ioctl@a-hdmi-a-1:
- shard-rkl: NOTRUN -> [SKIP][149] ([fdo#110189] / [i915#9723])
[149]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@kms_dirtyfb@[email protected]
* igt@kms_dirtyfb@fbc-dirtyfb-ioctl@a-vga-1:
- shard-snb: NOTRUN -> [SKIP][150] ([fdo#109271] / [fdo#110189])
[150]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-snb7/igt@kms_dirtyfb@[email protected]
* igt@kms_dirtyfb@psr-dirtyfb-ioctl:
- shard-dg2: NOTRUN -> [SKIP][151] ([i915#9833])
[151]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
* igt@kms_dsc@dsc-with-bpc-formats:
- shard-dg2: NOTRUN -> [SKIP][152] ([i915#3555] / [i915#3840])
[152]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-6/igt@[email protected]
* igt@kms_dsc@dsc-with-formats:
- shard-rkl: NOTRUN -> [SKIP][153] ([i915#3555] / [i915#3840])
[153]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-4/igt@[email protected]
* igt@kms_dsc@dsc-with-output-formats-with-bpc:
- shard-dg2: NOTRUN -> [SKIP][154] ([i915#3840] / [i915#9053])
[154]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-10/igt@[email protected]
* igt@kms_fbcon_fbt@psr:
- shard-dg1: NOTRUN -> [SKIP][155] ([i915#3469])
[155]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-15/igt@[email protected]
* igt@kms_feature_discovery@display-2x:
- shard-dg2: NOTRUN -> [SKIP][156] ([i915#1839]) +1 other test skip
[156]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-5/igt@[email protected]
* igt@kms_feature_discovery@psr2:
- shard-rkl: NOTRUN -> [SKIP][157] ([i915#658])
[157]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-2/igt@[email protected]
* igt@kms_flip@2x-blocking-absolute-wf_vblank-interruptible:
- shard-dg1: NOTRUN -> [SKIP][158] ([fdo#111825] / [i915#9934])
[158]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-19/igt@kms_flip@2x-blocking-absolute-wf_vblank-interruptible.html
* igt@kms_flip@2x-busy-flip:
- shard-tglu: NOTRUN -> [SKIP][159] ([fdo#109274] / [i915#3637]) +1 other test skip
[159]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-5/igt@[email protected]
* igt@kms_flip@2x-flip-vs-blocking-wf-vblank:
- shard-tglu: NOTRUN -> [SKIP][160] ([fdo#109274] / [fdo#111767] / [i915#3637]) +2 other tests skip
[160]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-7/igt@[email protected]
* igt@kms_flip@2x-flip-vs-fences:
- shard-dg2: NOTRUN -> [SKIP][161] ([i915#8381])
[161]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-5/igt@[email protected]
* igt@kms_flip@2x-modeset-vs-vblank-race:
- shard-dg2: NOTRUN -> [SKIP][162] ([fdo#109274]) +5 other tests skip
[162]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-10/igt@[email protected]
* igt@kms_flip_scaled_crc@flip-32bpp-yftile-to-64bpp-yftile-downscaling@pipe-a-valid-mode:
- shard-tglu: NOTRUN -> [SKIP][163] ([i915#2587] / [i915#2672]) +1 other test skip
[163]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-3/igt@kms_flip_scaled_crc@flip-32bpp-yftile-to-64bpp-yftile-downscaling@pipe-a-valid-mode.html
* igt@kms_flip_scaled_crc@flip-32bpp-yftileccs-to-64bpp-yftile-downscaling@pipe-a-valid-mode:
- shard-dg1: NOTRUN -> [SKIP][164] ([i915#2587] / [i915#2672])
[164]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-13/igt@kms_flip_scaled_crc@flip-32bpp-yftileccs-to-64bpp-yftile-downscaling@pipe-a-valid-mode.html
* igt@kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling@pipe-a-valid-mode:
- shard-rkl: NOTRUN -> [SKIP][165] ([i915#2672]) +1 other test skip
[165]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-7/igt@kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling@pipe-a-valid-mode.html
* igt@kms_flip_scaled_crc@flip-64bpp-yftile-to-16bpp-yftile-downscaling@pipe-a-default-mode:
- shard-mtlp: NOTRUN -> [SKIP][166] ([i915#2672])
[166]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-1/igt@kms_flip_scaled_crc@flip-64bpp-yftile-to-16bpp-yftile-downscaling@pipe-a-default-mode.html
* igt@kms_flip_scaled_crc@flip-64bpp-ytile-to-16bpp-ytile-upscaling@pipe-a-valid-mode:
- shard-dg2: NOTRUN -> [SKIP][167] ([i915#2672]) +2 other tests skip
[167]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-3/igt@kms_flip_scaled_crc@[email protected]
* igt@kms_frontbuffer_tracking@fbc-2p-primscrn-cur-indfb-draw-mmap-wc:
- shard-dg2: NOTRUN -> [SKIP][168] ([i915#8708]) +21 other tests skip
[168]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@kms_frontbuffer_tracking@fbc-2p-primscrn-cur-indfb-draw-mmap-wc.html
* igt@kms_frontbuffer_tracking@fbc-2p-primscrn-pri-indfb-draw-mmap-wc:
- shard-snb: [PASS][169] -> [SKIP][170] ([fdo#109271]) +7 other tests skip
[169]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-snb7/igt@kms_frontbuffer_tracking@fbc-2p-primscrn-pri-indfb-draw-mmap-wc.html
[170]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-snb6/igt@kms_frontbuffer_tracking@fbc-2p-primscrn-pri-indfb-draw-mmap-wc.html
* igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-indfb-draw-mmap-gtt:
- shard-mtlp: NOTRUN -> [SKIP][171] ([i915#8708]) +1 other test skip
[171]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-8/igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-indfb-draw-mmap-gtt.html
* igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-shrfb-draw-pwrite:
- shard-tglu: NOTRUN -> [SKIP][172] ([fdo#109280]) +11 other tests skip
[172]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-5/igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-pri-shrfb-draw-pwrite.html
* igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-indfb-draw-mmap-cpu:
- shard-tglu: NOTRUN -> [SKIP][173] ([fdo#110189]) +10 other tests skip
[173]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-7/igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-pri-indfb-draw-mmap-cpu.html
* igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-cur-indfb-draw-render:
- shard-dg2: NOTRUN -> [SKIP][174] ([fdo#111767] / [i915#5354])
[174]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-10/igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-cur-indfb-draw-render.html
* igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-spr-indfb-draw-pwrite:
- shard-dg1: NOTRUN -> [SKIP][175] ([fdo#111825]) +5 other tests skip
[175]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-13/igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-spr-indfb-draw-pwrite.html
* igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-cur-indfb-onoff:
- shard-tglu: NOTRUN -> [SKIP][176] ([fdo#109280] / [fdo#111767])
[176]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-5/igt@[email protected]
* igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-shrfb-draw-mmap-wc:
- shard-rkl: NOTRUN -> [SKIP][177] ([fdo#111767] / [fdo#111825] / [i915#1825])
[177]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-2/igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-pri-shrfb-draw-mmap-wc.html
* igt@kms_frontbuffer_tracking@fbcpsr-tiling-4:
- shard-rkl: NOTRUN -> [SKIP][178] ([i915#5439])
[178]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-7/igt@[email protected]
* igt@kms_frontbuffer_tracking@plane-fbc-rte:
- shard-dg2: NOTRUN -> [SKIP][179] ([i915#10070])
[179]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
- shard-rkl: NOTRUN -> [SKIP][180] ([i915#10070])
[180]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-3/igt@[email protected]
* igt@kms_frontbuffer_tracking@psr-1p-primscrn-cur-indfb-move:
- shard-dg2: NOTRUN -> [SKIP][181] ([i915#3458]) +16 other tests skip
[181]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
* igt@kms_frontbuffer_tracking@psr-1p-primscrn-indfb-plflip-blt:
- shard-rkl: NOTRUN -> [SKIP][182] ([i915#3023]) +16 other tests skip
[182]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-6/igt@[email protected]
* igt@kms_frontbuffer_tracking@psr-1p-primscrn-pri-shrfb-draw-mmap-gtt:
- shard-dg1: NOTRUN -> [SKIP][183] ([i915#8708]) +6 other tests skip
[183]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-12/igt@kms_frontbuffer_tracking@psr-1p-primscrn-pri-shrfb-draw-mmap-gtt.html
* igt@kms_frontbuffer_tracking@psr-2p-primscrn-indfb-pgflip-blt:
- shard-mtlp: NOTRUN -> [SKIP][184] ([i915#1825]) +6 other tests skip
[184]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-1/igt@[email protected]
* igt@kms_frontbuffer_tracking@psr-2p-primscrn-spr-indfb-draw-mmap-wc:
- shard-rkl: NOTRUN -> [SKIP][185] ([fdo#111825] / [i915#1825]) +28 other tests skip
[185]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-7/igt@kms_frontbuffer_tracking@psr-2p-primscrn-spr-indfb-draw-mmap-wc.html
* igt@kms_frontbuffer_tracking@psr-2p-scndscrn-cur-indfb-onoff:
- shard-dg1: NOTRUN -> [SKIP][186] ([fdo#111767] / [fdo#111825]) +1 other test skip
[186]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-13/igt@[email protected]
* igt@kms_frontbuffer_tracking@psr-suspend:
- shard-dg1: NOTRUN -> [SKIP][187] ([i915#3458]) +2 other tests skip
[187]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-13/igt@[email protected]
* igt@kms_hdr@bpc-switch-suspend:
- shard-dg2: NOTRUN -> [SKIP][188] ([i915#3555] / [i915#8228])
[188]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-6/igt@[email protected]
* igt@kms_hdr@invalid-hdr:
- shard-tglu: NOTRUN -> [SKIP][189] ([i915#3555] / [i915#8228])
[189]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-8/igt@[email protected]
* igt@kms_invalid_mode@clock-too-high@pipe-a-edp-1:
- shard-mtlp: NOTRUN -> [SKIP][190] ([i915#9457]) +3 other tests skip
[190]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-8/igt@kms_invalid_mode@[email protected]
* igt@kms_multipipe_modeset@basic-max-pipe-crc-check:
- shard-dg2: NOTRUN -> [SKIP][191] ([i915#4816])
[191]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-6/igt@[email protected]
* igt@kms_plane@pixel-format-source-clamping@pipe-b:
- shard-rkl: NOTRUN -> [INCOMPLETE][192] ([i915#10137])
[192]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-7/igt@kms_plane@[email protected]
* igt@kms_plane_lowres@tiling-y:
- shard-dg2: NOTRUN -> [SKIP][193] ([i915#8821])
[193]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
* igt@kms_plane_lowres@tiling-yf:
- shard-dg2: NOTRUN -> [SKIP][194] ([i915#3555] / [i915#8821])
[194]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-6/igt@[email protected]
* igt@kms_plane_scaling@intel-max-src-size@pipe-a-hdmi-a-2:
- shard-rkl: NOTRUN -> [FAIL][195] ([i915#8292])
[195]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-3/igt@kms_plane_scaling@[email protected]
* igt@kms_plane_scaling@plane-downscale-factor-0-25-with-modifiers@pipe-b-edp-1:
- shard-mtlp: NOTRUN -> [SKIP][196] ([i915#5176]) +3 other tests skip
[196]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-4/igt@kms_plane_scaling@[email protected]
* igt@kms_plane_scaling@plane-downscale-factor-0-25-with-pixel-format@pipe-b-hdmi-a-3:
- shard-dg2: NOTRUN -> [SKIP][197] ([i915#9423]) +7 other tests skip
[197]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-1/igt@kms_plane_scaling@plane-downscale-factor-0-25-with-pixel-format@pipe-b-hdmi-a-3.html
* igt@kms_plane_scaling@plane-downscale-factor-0-25-with-rotation@pipe-c-hdmi-a-3:
- shard-dg1: NOTRUN -> [SKIP][198] ([i915#9423]) +19 other tests skip
[198]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-13/igt@kms_plane_scaling@[email protected]
* igt@kms_plane_scaling@plane-downscale-factor-0-5-with-rotation@pipe-a-hdmi-a-1:
- shard-rkl: NOTRUN -> [SKIP][199] ([i915#9423]) +5 other tests skip
[199]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-2/igt@kms_plane_scaling@[email protected]
* igt@kms_plane_scaling@plane-downscale-factor-0-75-with-rotation@pipe-d-hdmi-a-1:
- shard-tglu: NOTRUN -> [SKIP][200] ([i915#9423]) +3 other tests skip
[200]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-7/igt@kms_plane_scaling@[email protected]
* igt@kms_plane_scaling@planes-downscale-factor-0-25-unity-scaling@pipe-c-hdmi-a-3:
- shard-dg1: NOTRUN -> [SKIP][201] ([i915#5235]) +15 other tests skip
[201]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-12/igt@kms_plane_scaling@[email protected]
* igt@kms_plane_scaling@planes-downscale-factor-0-25-upscale-20x20@pipe-d-hdmi-a-3:
- shard-dg2: NOTRUN -> [SKIP][202] ([i915#5235] / [i915#9423]) +3 other tests skip
[202]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-6/igt@kms_plane_scaling@[email protected]
* igt@kms_plane_scaling@planes-unity-scaling-downscale-factor-0-25@pipe-a-hdmi-a-1:
- shard-rkl: NOTRUN -> [SKIP][203] ([i915#5235]) +5 other tests skip
[203]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@kms_plane_scaling@[email protected]
* igt@kms_pm_dc@dc3co-vpb-simulation:
- shard-dg2: NOTRUN -> [SKIP][204] ([i915#9685]) +3 other tests skip
[204]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-10/igt@[email protected]
* igt@kms_pm_dc@dc5-psr:
- shard-rkl: NOTRUN -> [SKIP][205] ([i915#9685])
[205]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-3/igt@[email protected]
- shard-tglu: NOTRUN -> [SKIP][206] ([i915#9685])
[206]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-4/igt@[email protected]
* igt@kms_pm_dc@dc6-dpms:
- shard-dg2: NOTRUN -> [SKIP][207] ([i915#5978])
[207]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-3/igt@[email protected]
- shard-rkl: NOTRUN -> [SKIP][208] ([i915#3361])
[208]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-7/igt@[email protected]
* igt@kms_pm_dc@dc9-dpms:
- shard-rkl: NOTRUN -> [SKIP][209] ([i915#4281])
[209]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@[email protected]
* igt@kms_pm_lpsp@kms-lpsp:
- shard-dg2: NOTRUN -> [SKIP][210] ([i915#9340])
[210]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
* igt@kms_pm_lpsp@kms-lpsp@pipe-a-hdmi-a-1:
- shard-tglu: NOTRUN -> [FAIL][211] ([i915#9301])
[211]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-9/igt@kms_pm_lpsp@[email protected]
* igt@kms_pm_rpm@i2c:
- shard-dg2: [PASS][212] -> [FAIL][213] ([i915#8717])
[212]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-dg2-6/igt@[email protected]
[213]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
* igt@kms_pm_rpm@modeset-lpsp:
- shard-dg2: [PASS][214] -> [SKIP][215] ([i915#9519]) +1 other test skip
[214]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-dg2-10/igt@[email protected]
[215]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-6/igt@[email protected]
* igt@kms_pm_rpm@modeset-lpsp-stress-no-wait:
- shard-dg1: NOTRUN -> [SKIP][216] ([i915#9519])
[216]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-15/igt@[email protected]
* igt@kms_pm_rpm@pc8-residency:
- shard-dg1: NOTRUN -> [SKIP][217] ([fdo#109293] / [fdo#109506])
[217]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-17/igt@[email protected]
* igt@kms_psr2_sf@cursor-plane-move-continuous-exceed-fully-sf:
- shard-rkl: NOTRUN -> [SKIP][218] ([i915#9683]) +1 other test skip
[218]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-4/igt@[email protected]
- shard-tglu: NOTRUN -> [SKIP][219] ([i915#9683])
[219]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-6/igt@[email protected]
* igt@kms_psr2_sf@primary-plane-update-sf-dmg-area:
- shard-tglu: NOTRUN -> [SKIP][220] ([fdo#111068] / [i915#9683])
[220]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-5/igt@[email protected]
- shard-rkl: NOTRUN -> [SKIP][221] ([fdo#111068] / [i915#9683])
[221]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@[email protected]
* igt@kms_psr2_su@page_flip-p010:
- shard-dg2: NOTRUN -> [SKIP][222] ([i915#9683]) +4 other tests skip
[222]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-6/igt@kms_psr2_su@page_flip-p010.html
* igt@kms_rotation_crc@bad-tiling:
- shard-dg2: NOTRUN -> [SKIP][223] ([i915#4235]) +1 other test skip
[223]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@[email protected]
* igt@kms_rotation_crc@primary-4-tiled-reflect-x-180:
- shard-rkl: NOTRUN -> [SKIP][224] ([i915#5289])
[224]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@[email protected]
- shard-tglu: NOTRUN -> [SKIP][225] ([i915#5289])
[225]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-2/igt@[email protected]
* igt@kms_rotation_crc@primary-y-tiled-reflect-x-0:
- shard-rkl: [PASS][226] -> [ABORT][227] ([i915#8875] / [i915#9926])
[226]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-rkl-7/igt@[email protected]
[227]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@[email protected]
* igt@kms_rotation_crc@primary-yf-tiled-reflect-x-0:
- shard-rkl: NOTRUN -> [SKIP][228] ([fdo#111615] / [i915#5289])
[228]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@[email protected]
- shard-tglu: NOTRUN -> [SKIP][229] ([fdo#111615] / [i915#5289])
[229]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-4/igt@[email protected]
* igt@kms_rotation_crc@primary-yf-tiled-reflect-x-90:
- shard-dg2: NOTRUN -> [SKIP][230] ([i915#4235] / [i915#5190])
[230]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-10/igt@[email protected]
- shard-dg1: NOTRUN -> [SKIP][231] ([fdo#111615] / [i915#5289])
[231]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-18/igt@[email protected]
* igt@kms_scaling_modes@scaling-mode-none:
- shard-dg1: NOTRUN -> [SKIP][232] ([i915#3555])
[232]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-18/igt@[email protected]
* igt@kms_setmode@clone-exclusive-crtc:
- shard-tglu: NOTRUN -> [SKIP][233] ([i915#3555])
[233]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-6/igt@[email protected]
* igt@kms_tiled_display@basic-test-pattern-with-chamelium:
- shard-mtlp: NOTRUN -> [SKIP][234] ([i915#8623])
[234]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-1/igt@[email protected]
* igt@kms_tv_load_detect@load-detect:
- shard-rkl: NOTRUN -> [SKIP][235] ([fdo#109309])
[235]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@[email protected]
* igt@kms_universal_plane@cursor-fb-leak@pipe-a-hdmi-a-1:
- shard-tglu: [PASS][236] -> [FAIL][237] ([i915#9196])
[236]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-tglu-7/igt@kms_universal_plane@[email protected]
[237]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-3/igt@kms_universal_plane@[email protected]
* igt@kms_vrr@flip-basic-fastset:
- shard-mtlp: NOTRUN -> [SKIP][238] ([i915#8808] / [i915#9906])
[238]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-8/igt@[email protected]
* igt@kms_writeback@writeback-fb-id-xrgb2101010:
- shard-dg2: NOTRUN -> [SKIP][239] ([i915#2437] / [i915#9412])
[239]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-1/igt@[email protected]
* igt@kms_writeback@writeback-pixel-formats:
- shard-tglu: NOTRUN -> [SKIP][240] ([i915#2437] / [i915#9412])
[240]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-2/igt@[email protected]
* igt@perf@mi-rpc:
- shard-dg2: NOTRUN -> [SKIP][241] ([i915#2434])
[241]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-3/igt@[email protected]
- shard-tglu: NOTRUN -> [SKIP][242] ([fdo#109289]) +2 other tests skip
[242]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-6/igt@[email protected]
* igt@perf@per-context-mode-unprivileged:
- shard-rkl: NOTRUN -> [SKIP][243] ([i915#2435])
[243]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-7/igt@[email protected]
* igt@perf_pmu@busy-double-start@vecs1:
- shard-dg2: NOTRUN -> [FAIL][244] ([i915#4349]) +3 other tests fail
[244]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-5/igt@perf_pmu@[email protected]
* igt@perf_pmu@cpu-hotplug:
- shard-rkl: NOTRUN -> [SKIP][245] ([i915#8850])
[245]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-7/igt@[email protected]
* igt@perf_pmu@event-wait@rcs0:
- shard-rkl: NOTRUN -> [SKIP][246] ([fdo#112283])
[246]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@perf_pmu@[email protected]
* igt@perf_pmu@rc6-suspend:
- shard-tglu: [PASS][247] -> [INCOMPLETE][248] ([i915#9853])
[247]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-tglu-5/igt@[email protected]
[248]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-2/igt@[email protected]
* igt@perf_pmu@rc6@other-idle-gt0:
- shard-dg2: NOTRUN -> [SKIP][249] ([i915#8516])
[249]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-6/igt@perf_pmu@[email protected]
* igt@prime_udl:
- shard-tglu: NOTRUN -> [SKIP][250] ([fdo#109291])
[250]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-8/igt@prime_udl.html
* igt@prime_vgem@basic-fence-mmap:
- shard-dg2: NOTRUN -> [SKIP][251] ([i915#3708] / [i915#4077])
[251]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-3/igt@[email protected]
* igt@prime_vgem@basic-gtt:
- shard-dg1: NOTRUN -> [SKIP][252] ([i915#3708] / [i915#4077])
[252]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-15/igt@[email protected]
* igt@prime_vgem@basic-write:
- shard-dg2: NOTRUN -> [SKIP][253] ([i915#3291] / [i915#3708])
[253]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-1/igt@[email protected]
- shard-rkl: NOTRUN -> [SKIP][254] ([fdo#109295] / [i915#3291] / [i915#3708])
[254]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-1/igt@[email protected]
* igt@prime_vgem@fence-flip-hang:
- shard-rkl: NOTRUN -> [SKIP][255] ([fdo#109295] / [i915#3708])
[255]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-3/igt@[email protected]
* igt@prime_vgem@fence-read-hang:
- shard-dg2: NOTRUN -> [SKIP][256] ([i915#3708])
[256]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-3/igt@[email protected]
* igt@sriov_basic@enable-vfs-bind-unbind-each:
- shard-rkl: NOTRUN -> [SKIP][257] ([i915#9917])
[257]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-2/igt@[email protected]
* igt@syncobj_timeline@invalid-wait-zero-handles:
- shard-dg1: NOTRUN -> [FAIL][258] ([i915#9781])
[258]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-15/igt@[email protected]
* igt@v3d/v3d_perfmon@create-two-perfmon:
- shard-dg2: NOTRUN -> [SKIP][259] ([i915#2575]) +13 other tests skip
[259]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-10/igt@v3d/[email protected]
* igt@v3d/v3d_submit_cl@single-in-sync:
- shard-rkl: NOTRUN -> [SKIP][260] ([fdo#109315]) +12 other tests skip
[260]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-2/igt@v3d/[email protected]
- shard-tglu: NOTRUN -> [SKIP][261] ([fdo#109315] / [i915#2575]) +3 other tests skip
[261]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-3/igt@v3d/[email protected]
* igt@v3d/v3d_wait_bo@bad-bo:
- shard-dg1: NOTRUN -> [SKIP][262] ([i915#2575]) +2 other tests skip
[262]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-16/igt@v3d/[email protected]
* igt@v3d/v3d_wait_bo@bad-pad:
- shard-mtlp: NOTRUN -> [SKIP][263] ([i915#2575]) +2 other tests skip
[263]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-6/igt@v3d/[email protected]
* igt@vc4/vc4_label_bo@set-bad-handle:
- shard-tglu: NOTRUN -> [SKIP][264] ([i915#2575]) +2 other tests skip
[264]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-2/igt@vc4/[email protected]
* igt@vc4/vc4_label_bo@set-bad-name:
- shard-dg1: NOTRUN -> [SKIP][265] ([i915#7711]) +1 other test skip
[265]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-18/igt@vc4/[email protected]
* igt@vc4/vc4_label_bo@set-label:
- shard-mtlp: NOTRUN -> [SKIP][266] ([i915#7711]) +1 other test skip
[266]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-3/igt@vc4/[email protected]
* igt@vc4/vc4_perfmon@create-perfmon-invalid-events:
- shard-dg2: NOTRUN -> [SKIP][267] ([i915#7711]) +8 other tests skip
[267]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@vc4/[email protected]
* igt@vc4/vc4_tiling@get-bad-flags:
- shard-rkl: NOTRUN -> [SKIP][268] ([i915#7711]) +7 other tests skip
[268]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-2/igt@vc4/[email protected]
#### Possible fixes ####
* igt@gem_ctx_exec@basic-nohangcheck:
- shard-rkl: [FAIL][269] ([i915#6268]) -> [PASS][270]
[269]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-rkl-6/igt@[email protected]
[270]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@[email protected]
* igt@gem_eio@unwedge-stress:
- shard-dg1: [FAIL][271] ([i915#5784]) -> [PASS][272]
[271]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-dg1-16/igt@[email protected]
[272]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-15/igt@[email protected]
* igt@gem_exec_fair@basic-pace@rcs0:
- shard-rkl: [FAIL][273] ([i915#2842]) -> [PASS][274] +1 other test pass
[273]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-rkl-7/igt@gem_exec_fair@[email protected]
[274]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-7/igt@gem_exec_fair@[email protected]
* igt@gem_exec_suspend@basic-s4-devices@smem:
- shard-tglu: [ABORT][275] ([i915#7975] / [i915#8213]) -> [PASS][276]
[275]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-tglu-10/igt@gem_exec_suspend@[email protected]
[276]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-3/igt@gem_exec_suspend@[email protected]
* igt@gem_mmap_gtt@hang-user:
- shard-tglu: [INCOMPLETE][277] ([i915#10137]) -> [PASS][278]
[277]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-tglu-6/igt@[email protected]
[278]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-9/igt@[email protected]
* igt@i915_module_load@reload-with-fault-injection:
- shard-snb: [INCOMPLETE][279] ([i915#10137] / [i915#9200] / [i915#9849]) -> [PASS][280]
[279]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-snb7/igt@[email protected]
[280]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-snb7/igt@[email protected]
- shard-glk: [INCOMPLETE][281] ([i915#10137] / [i915#9200] / [i915#9849]) -> [PASS][282]
[281]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-glk4/igt@[email protected]
[282]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-glk4/igt@[email protected]
* igt@i915_pm_rps@reset:
- shard-snb: [INCOMPLETE][283] ([i915#10137] / [i915#7790]) -> [PASS][284]
[283]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-snb2/igt@[email protected]
[284]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-snb1/igt@[email protected]
* igt@i915_power@sanity:
- shard-mtlp: [SKIP][285] ([i915#7984]) -> [PASS][286]
[285]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-mtlp-7/igt@[email protected]
[286]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-8/igt@[email protected]
* igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-hflip-async-flip:
- shard-tglu: [FAIL][287] ([i915#3743]) -> [PASS][288]
[287]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-tglu-8/igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-hflip-async-flip.html
[288]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-5/igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-hflip-async-flip.html
* igt@kms_cursor_legacy@2x-flip-vs-cursor-atomic:
- shard-snb: [SKIP][289] ([fdo#109271] / [fdo#111767]) -> [PASS][290]
[289]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-snb5/igt@[email protected]
[290]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-snb7/igt@[email protected]
* igt@kms_cursor_legacy@2x-flip-vs-cursor-legacy:
- shard-snb: [SKIP][291] ([fdo#109271]) -> [PASS][292] +11 other tests pass
[291]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-snb6/igt@[email protected]
[292]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-snb7/igt@[email protected]
* igt@kms_flip@flip-vs-expired-vblank-interruptible@b-hdmi-a4:
- shard-dg1: [FAIL][293] ([i915#79]) -> [PASS][294]
[293]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-dg1-16/igt@kms_flip@[email protected]
[294]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-17/igt@kms_flip@[email protected]
* igt@kms_frontbuffer_tracking@fbc-1p-offscren-pri-shrfb-draw-blt:
- shard-dg2: [FAIL][295] ([i915#6880]) -> [PASS][296]
[295]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-dg2-3/igt@[email protected]
[296]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-5/igt@[email protected]
* igt@kms_plane@pixel-format-source-clamping@pipe-a:
- shard-rkl: [INCOMPLETE][297] ([i915#10056]) -> [PASS][298]
[297]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-rkl-4/igt@kms_plane@[email protected]
[298]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-7/igt@kms_plane@[email protected]
* igt@kms_pm_rpm@modeset-lpsp:
- shard-rkl: [SKIP][299] ([i915#9519]) -> [PASS][300] +2 other tests pass
[299]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-rkl-1/igt@[email protected]
[300]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-5/igt@[email protected]
* igt@kms_pm_rpm@modeset-lpsp-stress:
- shard-dg2: [SKIP][301] ([i915#9519]) -> [PASS][302]
[301]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-dg2-3/igt@[email protected]
[302]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-10/igt@[email protected]
* igt@kms_universal_plane@cursor-fb-leak@pipe-a-edp-1:
- shard-mtlp: [FAIL][303] ([i915#9196]) -> [PASS][304]
[303]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-mtlp-6/igt@kms_universal_plane@[email protected]
[304]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-mtlp-8/igt@kms_universal_plane@[email protected]
* igt@kms_universal_plane@cursor-fb-leak@pipe-c-hdmi-a-4:
- shard-dg1: [FAIL][305] ([i915#9196]) -> [PASS][306]
[305]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-dg1-16/igt@kms_universal_plane@[email protected]
[306]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-19/igt@kms_universal_plane@[email protected]
* igt@perf@non-zero-reason@0-rcs0:
- shard-dg2: [FAIL][307] ([i915#7484]) -> [PASS][308]
[307]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-dg2-1/igt@perf@[email protected]
[308]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-5/igt@perf@[email protected]
#### Warnings ####
* igt@device_reset@unbind-reset-rebind:
- shard-dg1: [INCOMPLETE][309] ([i915#10137] / [i915#9618]) -> [INCOMPLETE][310] ([i915#10137] / [i915#9408] / [i915#9618])
[309]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-dg1-18/igt@[email protected]
[310]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-15/igt@[email protected]
* igt@kms_ccs@pipe-a-missing-ccs-buffer-4-tiled-mtl-rc-ccs:
- shard-dg1: [SKIP][311] ([i915#5354] / [i915#6095]) -> [SKIP][312] ([i915#4423] / [i915#5354] / [i915#6095]) +1 other test skip
[311]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-dg1-17/igt@[email protected]
[312]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-18/igt@[email protected]
* igt@kms_content_protection@mei-interface:
- shard-rkl: [SKIP][313] ([i915#8063]) -> [SKIP][314] ([i915#9424])
[313]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-rkl-6/igt@[email protected]
[314]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-2/igt@[email protected]
- shard-tglu: [SKIP][315] ([i915#8063]) -> [SKIP][316] ([i915#6944] / [i915#9424])
[315]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-tglu-9/igt@[email protected]
[316]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-tglu-6/igt@[email protected]
* igt@kms_content_protection@uevent:
- shard-snb: [INCOMPLETE][317] ([i915#8816]) -> [SKIP][318] ([fdo#109271])
[317]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-snb7/igt@[email protected]
[318]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-snb5/igt@[email protected]
* igt@kms_fbcon_fbt@psr-suspend:
- shard-rkl: [SKIP][319] ([i915#3955]) -> [SKIP][320] ([fdo#110189] / [i915#3955]) +1 other test skip
[319]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-rkl-6/igt@[email protected]
[320]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-rkl-1/igt@[email protected]
* igt@kms_frontbuffer_tracking@psr-2p-scndscrn-spr-indfb-draw-mmap-cpu:
- shard-dg1: [SKIP][321] ([fdo#111825]) -> [SKIP][322] ([fdo#111825] / [i915#4423]) +1 other test skip
[321]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-dg1-19/igt@kms_frontbuffer_tracking@psr-2p-scndscrn-spr-indfb-draw-mmap-cpu.html
[322]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg1-18/igt@kms_frontbuffer_tracking@psr-2p-scndscrn-spr-indfb-draw-mmap-cpu.html
* igt@prime_mmap@test_aperture_limit@test_aperture_limit-smem:
- shard-dg2: [CRASH][323] ([i915#9351]) -> [INCOMPLETE][324] ([i915#5493])
[323]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_14287/shard-dg2-3/igt@prime_mmap@test_aperture_limit@test_aperture_limit-smem.html
[324]: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/shard-dg2-2/igt@prime_mmap@test_aperture_limit@test_aperture_limit-smem.html
{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).
[fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
[fdo#109274]: https://bugs.freedesktop.org/show_bug.cgi?id=109274
[fdo#109280]: https://bugs.freedesktop.org/show_bug.cgi?id=109280
[fdo#109289]: https://bugs.freedesktop.org/show_bug.cgi?id=109289
[fdo#109291]: https://bugs.freedesktop.org/show_bug.cgi?id=109291
[fdo#109293]: https://bugs.freedesktop.org/show_bug.cgi?id=109293
[fdo#109295]: https://bugs.freedesktop.org/show_bug.cgi?id=109295
[fdo#109302]: https://bugs.freedesktop.org/show_bug.cgi?id=109302
[fdo#109309]: https://bugs.freedesktop.org/show_bug.cgi?id=109309
[fdo#109314]: https://bugs.freedesktop.org/show_bug.cgi?id=109314
[fdo#109315]: https://bugs.freedesktop.org/show_bug.cgi?id=109315
[fdo#109506]: https://bugs.freedesktop.org/show_bug.cgi?id=109506
[fdo#110189]: https://bugs.freedesktop.org/show_bug.cgi?id=110189
[fdo#110723]: https://bugs.freedesktop.org/show_bug.cgi?id=110723
[fdo#111068]: https://bugs.freedesktop.org/show_bug.cgi?id=111068
[fdo#111614]: https://bugs.freedesktop.org/show_bug.cgi?id=111614
[fdo#111615]: https://bugs.freedesktop.org/show_bug.cgi?id=111615
[fdo#111767]: https://bugs.freedesktop.org/show_bug.cgi?id=111767
[fdo#111825]: https://bugs.freedesktop.org/show_bug.cgi?id=111825
[fdo#111827]: https://bugs.freedesktop.org/show_bug.cgi?id=111827
[fdo#112283]: https://bugs.freedesktop.org/show_bug.cgi?id=112283
[i915#10030]: https://gitlab.freedesktop.org/drm/intel/issues/10030
[i915#10031]: https://gitlab.freedesktop.org/drm/intel/issues/10031
[i915#10056]: https://gitlab.freedesktop.org/drm/intel/issues/10056
[i915#10070]: https://gitlab.freedesktop.org/drm/intel/issues/10070
[i915#10137]: https://gitlab.freedesktop.org/drm/intel/issues/10137
[i915#10183]: https://gitlab.freedesktop.org/drm/intel/issues/10183
[i915#10278]: https://gitlab.freedesktop.org/drm/intel/issues/10278
[i915#1099]: https://gitlab.freedesktop.org/drm/intel/issues/1099
[i915#1769]: https://gitlab.freedesktop.org/drm/intel/issues/1769
[i915#1825]: https://gitlab.freedesktop.org/drm/intel/issues/1825
[i915#1839]: https://gitlab.freedesktop.org/drm/intel/issues/1839
[i915#2434]: https://gitlab.freedesktop.org/drm/intel/issues/2434
[i915#2435]: https://gitlab.freedesktop.org/drm/intel/issues/2435
[i915#2437]: https://gitlab.freedesktop.org/drm/intel/issues/2437
[i915#2527]: https://gitlab.freedesktop.org/drm/intel/issues/2527
[i915#2575]: https://gitlab.freedesktop.org/drm/intel/issues/2575
[i915#2587]: https://gitlab.freedesktop.org/drm/intel/issues/2587
[i915#2672]: https://gitlab.freedesktop.org/drm/intel/issues/2672
[i915#2705]: https://gitlab.freedesktop.org/drm/intel/issues/2705
[i915#280]: https://gitlab.freedesktop.org/drm/intel/issues/280
[i915#2842]: https://gitlab.freedesktop.org/drm/intel/issues/2842
[i915#2856]: https://gitlab.freedesktop.org/drm/intel/issues/2856
[i915#3023]: https://gitlab.freedesktop.org/drm/intel/issues/3023
[i915#3281]: https://gitlab.freedesktop.org/drm/intel/issues/3281
[i915#3282]: https://gitlab.freedesktop.org/drm/intel/issues/3282
[i915#3291]: https://gitlab.freedesktop.org/drm/intel/issues/3291
[i915#3297]: https://gitlab.freedesktop.org/drm/intel/issues/3297
[i915#3359]: https://gitlab.freedesktop.org/drm/intel/issues/3359
[i915#3361]: https://gitlab.freedesktop.org/drm/intel/issues/3361
[i915#3458]: https://gitlab.freedesktop.org/drm/intel/issues/3458
[i915#3469]: https://gitlab.freedesktop.org/drm/intel/issues/3469
[i915#3539]: https://gitlab.freedesktop.org/drm/intel/issues/3539
[i915#3555]: https://gitlab.freedesktop.org/drm/intel/issues/3555
[i915#3637]: https://gitlab.freedesktop.org/drm/intel/issues/3637
[i915#3638]: https://gitlab.freedesktop.org/drm/intel/issues/3638
[i915#3708]: https://gitlab.freedesktop.org/drm/intel/issues/3708
[i915#3742]: https://gitlab.freedesktop.org/drm/intel/issues/3742
[i915#3743]: https://gitlab.freedesktop.org/drm/intel/issues/3743
[i915#3778]: https://gitlab.freedesktop.org/drm/intel/issues/3778
[i915#3840]: https://gitlab.freedesktop.org/drm/intel/issues/3840
[i915#3955]: https://gitlab.freedesktop.org/drm/intel/issues/3955
[i915#4077]: https://gitlab.freedesktop.org/drm/intel/issues/4077
[i915#4079]: https://gitlab.freedesktop.org/drm/intel/issues/4079
[i915#4083]: https://gitlab.freedesktop.org/drm/intel/issues/4083
[i915#4087]: https://gitlab.freedesktop.org/drm/intel/issues/4087
[i915#4103]: https://gitlab.freedesktop.org/drm/intel/issues/4103
[i915#4212]: https://gitlab.freedesktop.org/drm/intel/issues/4212
[i915#4213]: https://gitlab.freedesktop.org/drm/intel/issues/4213
[i915#4235]: https://gitlab.freedesktop.org/drm/intel/issues/4235
[i915#4270]: https://gitlab.freedesktop.org/drm/intel/issues/4270
[i915#4281]: https://gitlab.freedesktop.org/drm/intel/issues/4281
[i915#4349]: https://gitlab.freedesktop.org/drm/intel/issues/4349
[i915#4387]: https://gitlab.freedesktop.org/drm/intel/issues/4387
[i915#4423]: https://gitlab.freedesktop.org/drm/intel/issues/4423
[i915#4473]: https://gitlab.freedesktop.org/drm/intel/issues/4473
[i915#4525]: https://gitlab.freedesktop.org/drm/intel/issues/4525
[i915#4538]: https://gitlab.freedesktop.org/drm/intel/issues/4538
[i915#4613]: https://gitlab.freedesktop.org/drm/intel/issues/4613
[i915#4771]: https://gitlab.freedesktop.org/drm/intel/issues/4771
[i915#4816]: https://gitlab.freedesktop.org/drm/intel/issues/4816
[i915#4852]: https://gitlab.freedesktop.org/drm/intel/issues/4852
[i915#4860]: https://gitlab.freedesktop.org/drm/intel/issues/4860
[i915#4873]: https://gitlab.freedesktop.org/drm/intel/issues/4873
[i915#4880]: https://gitlab.freedesktop.org/drm/intel/issues/4880
[i915#5138]: https://gitlab.freedesktop.org/drm/intel/issues/5138
[i915#5176]: https://gitlab.freedesktop.org/drm/intel/issues/5176
[i915#5190]: https://gitlab.freedesktop.org/drm/intel/issues/5190
[i915#5235]: https://gitlab.freedesktop.org/drm/intel/issues/5235
[i915#5286]: https://gitlab.freedesktop.org/drm/intel/issues/5286
[i915#5289]: https://gitlab.freedesktop.org/drm/intel/issues/5289
[i915#5354]: https://gitlab.freedesktop.org/drm/intel/issues/5354
[i915#5439]: https://gitlab.freedesktop.org/drm/intel/issues/5439
[i915#5493]: https://gitlab.freedesktop.org/drm/intel/issues/5493
[i915#5784]: https://gitlab.freedesktop.org/drm/intel/issues/5784
[i915#5978]: https://gitlab.freedesktop.org/drm/intel/issues/5978
[i915#6095]: https://gitlab.freedesktop.org/drm/intel/issues/6095
[i915#6228]: https://gitlab.freedesktop.org/drm/intel/issues/6228
[i915#6230]: https://gitlab.freedesktop.org/drm/intel/issues/6230
[i915#6268]: https://gitlab.freedesktop.org/drm/intel/issues/6268
[i915#6334]: https://gitlab.freedesktop.org/drm/intel/issues/6334
[i915#6335]: https://gitlab.freedesktop.org/drm/intel/issues/6335
[i915#6344]: https://gitlab.freedesktop.org/drm/intel/issues/6344
[i915#658]: https://gitlab.freedesktop.org/drm/intel/issues/658
[i915#6880]: https://gitlab.freedesktop.org/drm/intel/issues/6880
[i915#6944]: https://gitlab.freedesktop.org/drm/intel/issues/6944
[i915#7016]: https://gitlab.freedesktop.org/drm/intel/issues/7016
[i915#7118]: https://gitlab.freedesktop.org/drm/intel/issues/7118
[i915#7213]: https://gitlab.freedesktop.org/drm/intel/issues/7213
[i915#7484]: https://gitlab.freedesktop.org/drm/intel/issues/7484
[i915#7697]: https://gitlab.freedesktop.org/drm/intel/issues/7697
[i915#7707]: https://gitlab.freedesktop.org/drm/intel/issues/7707
[i915#7711]: https://gitlab.freedesktop.org/drm/intel/issues/7711
[i915#7742]: https://gitlab.freedesktop.org/drm/intel/issues/7742
[i915#7790]: https://gitlab.freedesktop.org/drm/intel/issues/7790
[i915#7828]: https://gitlab.freedesktop.org/drm/intel/issues/7828
[i915#79]: https://gitlab.freedesktop.org/drm/intel/issues/79
[i915#7975]: https://gitlab.freedesktop.org/drm/intel/issues/7975
[i915#7984]: https://gitlab.freedesktop.org/drm/intel/issues/7984
[i915#8063]: https://gitlab.freedesktop.org/drm/intel/issues/8063
[i915#8213]: https://gitlab.freedesktop.org/drm/intel/issues/8213
[i915#8228]: https://gitlab.freedesktop.org/drm/intel/issues/8228
[i915#8292]: https://gitlab.freedesktop.org/drm/intel/issues/8292
[i915#8381]: https://gitlab.freedesktop.org/drm/intel/issues/8381
[i915#8399]: https://gitlab.freedesktop.org/drm/intel/issues/8399
[i915#8414]: https://gitlab.freedesktop.org/drm/intel/issues/8414
[i915#8516]: https://gitlab.freedesktop.org/drm/intel/issues/8516
[i915#8555]: https://gitlab.freedesktop.org/drm/intel/issues/8555
[i915#8562]: https://gitlab.freedesktop.org/drm/intel/issues/8562
[i915#8623]: https://gitlab.freedesktop.org/drm/intel/issues/8623
[i915#8708]: https://gitlab.freedesktop.org/drm/intel/issues/8708
[i915#8709]: https://gitlab.freedesktop.org/drm/intel/issues/8709
[i915#8717]: https://gitlab.freedesktop.org/drm/intel/issues/8717
[i915#8808]: https://gitlab.freedesktop.org/drm/intel/issues/8808
[i915#8814]: https://gitlab.freedesktop.org/drm/intel/issues/8814
[i915#8816]: https://gitlab.freedesktop.org/drm/intel/issues/8816
[i915#8821]: https://gitlab.freedesktop.org/drm/intel/issues/8821
[i915#8850]: https://gitlab.freedesktop.org/drm/intel/issues/8850
[i915#8875]: https://gitlab.freedesktop.org/drm/intel/issues/8875
[i915#8925]: https://gitlab.freedesktop.org/drm/intel/issues/8925
[i915#9053]: https://gitlab.freedesktop.org/drm/intel/issues/9053
[i915#9196]: https://gitlab.freedesktop.org/drm/intel/issues/9196
[i915#9200]: https://gitlab.freedesktop.org/drm/intel/issues/9200
[i915#9301]: https://gitlab.freedesktop.org/drm/intel/issues/9301
[i915#9323]: https://gitlab.freedesktop.org/drm/intel/issues/9323
[i915#9340]: https://gitlab.freedesktop.org/drm/intel/issues/9340
[i915#9351]: https://gitlab.freedesktop.org/drm/intel/issues/9351
[i915#9408]: https://gitlab.freedesktop.org/drm/intel/issues/9408
[i915#9412]: https://gitlab.freedesktop.org/drm/intel/issues/9412
[i915#9423]: https://gitlab.freedesktop.org/drm/intel/issues/9423
[i915#9424]: https://gitlab.freedesktop.org/drm/intel/issues/9424
[i915#9433]: https://gitlab.freedesktop.org/drm/intel/issues/9433
[i915#9457]: https://gitlab.freedesktop.org/drm/intel/issues/9457
[i915#9519]: https://gitlab.freedesktop.org/drm/intel/issues/9519
[i915#9606]: https://gitlab.freedesktop.org/drm/intel/issues/9606
[i915#9618]: https://gitlab.freedesktop.org/drm/intel/issues/9618
[i915#9683]: https://gitlab.freedesktop.org/drm/intel/issues/9683
[i915#9685]: https://gitlab.freedesktop.org/drm/intel/issues/9685
[i915#9688]: https://gitlab.freedesktop.org/drm/intel/issues/9688
[i915#9723]: https://gitlab.freedesktop.org/drm/intel/issues/9723
[i915#9732]: https://gitlab.freedesktop.org/drm/intel/issues/9732
[i915#9781]: https://gitlab.freedesktop.org/drm/intel/issues/9781
[i915#9808]: https://gitlab.freedesktop.org/drm/intel/issues/9808
[i915#9820]: https://gitlab.freedesktop.org/drm/intel/issues/9820
[i915#9833]: https://gitlab.freedesktop.org/drm/intel/issues/9833
[i915#9849]: https://gitlab.freedesktop.org/drm/intel/issues/9849
[i915#9853]: https://gitlab.freedesktop.org/drm/intel/issues/9853
[i915#9906]: https://gitlab.freedesktop.org/drm/intel/issues/9906
[i915#9917]: https://gitlab.freedesktop.org/drm/intel/issues/9917
[i915#9926]: https://gitlab.freedesktop.org/drm/intel/issues/9926
[i915#9934]: https://gitlab.freedesktop.org/drm/intel/issues/9934
Build changes
-------------
* CI: CI-20190529 -> None
* IGT: IGT_7716 -> IGTPW_10690
* Piglit: piglit_4509 -> None
CI-20190529: 20190529
CI_DRM_14287: b09fd1bd9f8b4a8e5cd8c1de8dbba4d4eb15b4e9 @ git://anongit.freedesktop.org/gfx-ci/linux
IGTPW_10690: 10690
IGT_7716: 7716
piglit_4509: fdc5a4ca11124ab8413c7988896eec4c97336694 @ git://anongit.freedesktop.org/piglit
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/IGTPW_10690/index.html