nvidia-gpu: add Power Sensor PeakReading Property
Add support for Sensor Properties PeakReading and PeakRedingTime.
Current Limitation -
The ResetMetrics action is currently not supported for Redfish URIs in
bmcweb. As a result, the ability to clear PeakReading values for GPU
Power Sensors has not been implemented.
Future Consideration -
If ResetMetrics action support is added to bmcweb in the future, the
corresponding functionality will also need to be implemented in the
dbus-sensor application to ensure full compatibility.
Tested: Build an image for gb200nvl-obmc machine with the following
patches cherry picked. This patches are needed to enable the mctp stack.
https://gerrit.openbmc.org/c/openbmc/openbmc/+/79422
```
root@gb200nvl-obmc:~# busctl introspect xyz.openbmc_project.GpuSensor /xyz/openbmc_project/sensors/power/NVIDIA_GB200_GPU_0_Power_0
NAME TYPE SIGNATURE RESULT/VALUE FLAGS
org.freedesktop.DBus.Introspectable interface - - -
.Introspect method - s -
org.freedesktop.DBus.Peer interface - - -
.GetMachineId method - s -
.Ping method - - -
org.freedesktop.DBus.Properties interface - - -
.Get method ss v -
.GetAll method s a{sv} -
.Set method ssv - -
.PropertiesChanged signal sa{sv}as - -
xyz.openbmc_project.Association.Definitions interface - - -
.Associations property a(sss) 1 "chassis" "all_sensors" "/xyz/openb... emits-change
xyz.openbmc_project.Sensor.Value interface - - -
.MaxValue property d 5000 emits-change
.MinValue property d 0 emits-change
.Unit property s "xyz.openbmc_project.Sensor.Value.Uni... emits-change
.Value property d 29.194 emits-change writable
xyz.openbmc_project.Sensor.ValueMutability interface - - -
.Mutable property b true emits-change
xyz.openbmc_project.State.Decorator.Availability interface - - -
.Available property b true emits-change writable
xyz.openbmc_project.State.Decorator.OperationalStatus interface - - -
.Functional property b true emits-change
xyz.openbmc_project.Telemetry.Report interface - - -
.Readings property (ta(ssdt)) 0 1 "PeakReading" "" 80.933 0 emits-change
```
Change-Id: I0a4f7eb0a5db688f32bf80954839140da9bb7e2a
Signed-off-by: Harshit Aghera <haghera@nvidia.com>
diff --git a/src/nvidia-gpu/NvidiaGpuMctpVdm.cpp b/src/nvidia-gpu/NvidiaGpuMctpVdm.cpp
index ebfa76b..51a6755 100644
--- a/src/nvidia-gpu/NvidiaGpuMctpVdm.cpp
+++ b/src/nvidia-gpu/NvidiaGpuMctpVdm.cpp
@@ -218,16 +218,16 @@
return 0;
}
-int encodeGetCurrentPowerDrawRequest(uint8_t instanceId, uint8_t sensorId,
- uint8_t averagingInterval,
- std::span<uint8_t> buf)
+int encodeGetPowerDrawRequest(PlatformEnvironmentalCommands commandCode,
+ uint8_t instanceId, uint8_t sensorId,
+ uint8_t averagingInterval, std::span<uint8_t> buf)
{
- if (buf.size() < sizeof(GetCurrentPowerDrawRequest))
+ if (buf.size() < sizeof(GetPowerDrawRequest))
{
return EINVAL;
}
- auto* msg = reinterpret_cast<GetCurrentPowerDrawRequest*>(buf.data());
+ auto* msg = reinterpret_cast<GetPowerDrawRequest*>(buf.data());
ocp::accelerator_management::BindingPciVidInfo header{};
header.ocp_accelerator_management_msg_type =
@@ -243,8 +243,7 @@
return rc;
}
- msg->hdr.command = static_cast<uint8_t>(
- PlatformEnvironmentalCommands::GET_CURRENT_POWER_DRAW);
+ msg->hdr.command = static_cast<uint8_t>(commandCode);
msg->hdr.data_size = sizeof(sensorId) + sizeof(averagingInterval);
msg->sensorId = sensorId;
msg->averagingInterval = averagingInterval;
@@ -252,10 +251,9 @@
return 0;
}
-int decodeGetCurrentPowerDrawResponse(
- std::span<const uint8_t> buf,
- ocp::accelerator_management::CompletionCode& cc, uint16_t& reasonCode,
- uint32_t& power)
+int decodeGetPowerDrawResponse(std::span<const uint8_t> buf,
+ ocp::accelerator_management::CompletionCode& cc,
+ uint16_t& reasonCode, uint32_t& power)
{
auto rc =
ocp::accelerator_management::decodeReasonCodeAndCC(buf, cc, reasonCode);
@@ -265,13 +263,13 @@
return rc;
}
- if (buf.size() < sizeof(GetCurrentPowerDrawResponse))
+ if (buf.size() < sizeof(GetPowerDrawResponse))
{
return EINVAL;
}
const auto* response =
- reinterpret_cast<const GetCurrentPowerDrawResponse*>(buf.data());
+ reinterpret_cast<const GetPowerDrawResponse*>(buf.data());
const uint16_t dataSize = le16toh(response->hdr.data_size);
@@ -330,7 +328,7 @@
return rc;
}
- if (buf.size() < sizeof(GetCurrentPowerDrawResponse))
+ if (buf.size() < sizeof(GetPowerDrawResponse))
{
return EINVAL;
}