Fix metric type error output + re-align with ROCm SMI metrics

Changes:
* [CLI] Provide fix for "/opt/rocm/bin/amd-smi metric
TypeError: '>' not supported between instances of 'str' and 'i"
--> Python API was updated, CLI needed to reflect these changes
* [API] Updated amdsmi.h's with ROCm SMI
--> Incorrectly added mem_bandwidth_acc & mem_max_bandwidth
--> Realigned wrapper with updates
* [Test] Added metrics not shown in gpu_metrics_read.cc

Change-Id: Ia3a172377fd5a582254dd5a46d81dbec7e763cd9
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
Αυτή η υποβολή περιλαμβάνεται σε:
Charis Poag authored and Maisam Arif committed 2024-01-24 21:23:40 -06:00
1 γονέας 0b67c2ccc4
υποβολή 34bd26c68e
7 files changed
+109 -88

No files matched your search

@@ -1140,10 +1140,6 @@ amdsmi_status_t amdsmi_get_gpu_metrics_info(
(sizeof(pgpu_metrics->jpeg_activity) /
sizeof(pgpu_metrics->jpeg_activity[0])),
std::numeric_limits<uint16_t>::max());
pgpu_metrics->mem_bandwidth_acc =
static_cast<uint64_t>(std::numeric_limits<uint64_t>::max());
pgpu_metrics->mem_max_bandwidth =
static_cast<uint32_t>(std::numeric_limits<uint32_t>::max());
pgpu_metrics->pcie_nak_sent_count_acc =
static_cast<uint32_t>(std::numeric_limits<uint32_t>::max());
pgpu_metrics->pcie_nak_rcvd_count_acc =