commit b10a67b252ba8d899b88beb328f040ee9cc1e30f
author Harshit Aghera <haghera@nvidia.com> Tue May 27 12:19:29 2025 +0530
committer Ed Tanous <ed@tanous.net> Tue Jul 22 01:25:38 2025 +0000
tree e7ff7c4104084f4f44a108fe5cc5d3437abc49bb
parent 8951c87e8fca6ba57f78344119ab42866c190c80

nvidia-gpu: add dram temperature sensor

This commit introduces a dram temperature sensor for the GPU, enhancing
the existing temperature monitoring capabilities.

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

```
$ curl -s -k -u 'root:0penBmc' https://10.137.203.137/redfish/v1/Chassis/NVIDIA_GB200_1/Sensors/temperature_NVIDIA_GB200_GPU_0_DRAM_0_TEMP_0
{
  "@odata.id": "/redfish/v1/Chassis/NVIDIA_GB200_1/Sensors/temperature_NVIDIA_GB200_GPU_0_DRAM_0_TEMP_0",
  "@odata.type": "#Sensor.v1_2_0.Sensor",
  "Id": "temperature_NVIDIA_GB200_GPU_0_DRAM_0_TEMP_0",
  "Name": "NVIDIA GB200 GPU 0 DRAM 0 TEMP 0",
  "Reading": 30.0,
  "ReadingRangeMax": 127.0,
  "ReadingRangeMin": -128.0,
  "ReadingType": "Temperature",
  "ReadingUnits": "Cel",
  "Status": {
    "Health": "OK",
    "State": "Enabled"
  },
  "Thresholds": {
    "UpperCritical": {
      "Reading": 95.0
    }
  }
}%
```

Change-Id: I914bb94f85e2d4163397b71a08b4ddd8b171e7d7
Signed-off-by: Harshit Aghera <haghera@nvidia.com>

src/nvidia-gpu/NvidiaGpuDevice.cpp

diff --git a/src/nvidia-gpu/NvidiaGpuDevice.cpp b/src/nvidia-gpu/NvidiaGpuDevice.cpp
index 2c14a1c..d25860d 100644
--- a/src/nvidia-gpu/NvidiaGpuDevice.cpp
+++ b/src/nvidia-gpu/NvidiaGpuDevice.cpp
@@ -59,6 +59,12 @@
         mctpRequester,
         std::bind_front(&GpuDevice::processTLimitThresholds, this));
 
+    dramTempSensor = std::make_shared<NvidiaGpuTempSensor>(
+        conn, mctpRequester, name + "_DRAM_0_TEMP_0", path, eid,
+        gpuDramTempSensorId, objectServer,
+        std::vector<thresholds::Threshold>{thresholds::Threshold{
+            thresholds::Level::CRITICAL, thresholds::Direction::HIGH, 95.0}});
+
     powerSensor = std::make_shared<NvidiaGpuPowerSensor>(
         conn, mctpRequester, name + "_Power_0", path, eid, gpuPowerSensorId,
         objectServer, std::vector<thresholds::Threshold>{});
@@ -107,6 +113,7 @@
     {
         tLimitSensor->update();
     }
+    dramTempSensor->update();
     powerSensor->update();
     energySensor->update();
     voltageSensor->update();

src/nvidia-gpu/NvidiaGpuDevice.hpp

diff --git a/src/nvidia-gpu/NvidiaGpuDevice.hpp b/src/nvidia-gpu/NvidiaGpuDevice.hpp
index 2c81287..5357121 100644
--- a/src/nvidia-gpu/NvidiaGpuDevice.hpp
+++ b/src/nvidia-gpu/NvidiaGpuDevice.hpp
@@ -61,6 +61,7 @@
 
     std::shared_ptr<NvidiaGpuTempSensor> tempSensor;
     std::shared_ptr<NvidiaGpuTempSensor> tLimitSensor;
+    std::shared_ptr<NvidiaGpuTempSensor> dramTempSensor;
     std::shared_ptr<NvidiaGpuPowerSensor> powerSensor;
     std::shared_ptr<NvidiaGpuEnergySensor> energySensor;
     std::shared_ptr<NvidiaGpuVoltageSensor> voltageSensor;

src/nvidia-gpu/NvidiaGpuSensor.hpp

diff --git a/src/nvidia-gpu/NvidiaGpuSensor.hpp b/src/nvidia-gpu/NvidiaGpuSensor.hpp
index e2ae89a..0e9b7af 100644
--- a/src/nvidia-gpu/NvidiaGpuSensor.hpp
+++ b/src/nvidia-gpu/NvidiaGpuSensor.hpp
@@ -22,6 +22,7 @@
 
 constexpr uint8_t gpuTempSensorId{0};
 constexpr uint8_t gpuTLimitSensorId{2};
+constexpr uint8_t gpuDramTempSensorId{1};
 constexpr uint8_t smaTempSensorId{5};
 
 struct NvidiaGpuTempSensor :