amdgpu_dpm_get_temp_metrics(9)

💅 amdgpu_dpm_get_temp_metrics — Kernel API Manual ✨

okay so bestie, welcome to the amdgpu_dpm_get_temp_metrics(9) page 🎀 this is the kernel API manual for it, and we're gonna get you fully caught up, no cap.

---

🌸 NAME 🌸

amdgpu_dpm_get_temp_metrics — Retrieve metrics for a specific compute partition 🫶

---

⭐ SYNOPSIS ⭐

ssize_t amdgpu_dpm_get_temp_metrics (struct amdgpu_device *adev , enum
smu_temp_metric_type type , void *table );

like, that's the whole signature, verbatim, don't come for me 💅

---

🦋 ARGUMENTS 🦋

• adev — Pointer to the device. literally just the device, she's the main character here 👑
• type — Identifier for the temperature type metrics to be fetched. tells the function which temp-type metrics you're after 🌡️
• table — Pointer to a buffer where the metrics will be stored. and here's the iconic part: if it's NULL, the function returns the size of the metrics structure instead of filling anything in 🎀

---

💖 DESCRIPTION 💖

So what she does is retrieve metrics for a specific temperature type ✨. And if the table parameter is NULL? She doesn't populate anything — she just returns the size of the metrics structure. it's giving "vibe check first, fill the buffer later" honestly 🙈

---

🔥 RETURN 🔥

Size of the metrics structure on success 🫶 — or a negative error code on failure 😭 (yes, negative, that's the "something went wrong" energy, don't ignore it).

---

🌈 SEE ALSO 🌈

• Kernel file ./drivers/gpu/drm/amd/pm/amdgpu_dpm.c — the source, go read her, iykyk 📖
• amdgpu_dpm_reset_sdma_is_supported(9)
• amdgpu_dpm_is_temp_metrics_supported(9)
• amdgpu_dpm_get_xcp_metrics(9)

honestly the whole friend group ate 💀

---

ngl bestie, this page came from the ./drivers/gpu/drm/amd/pm source, and it's dated October 2026 🌸 and yes that's also the manual name up in the corner — amdgpu_dpm_get_temp_metrics(9), periodt 💜 mwah, go read the kernel code, ily 🧚