<!DOCTYPE html><html><head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8"> </head> <body> <div class="moz-cite-prefix">On 25-01-2025 12:44, Raag Jadav wrote: </div> <blockquote type="cite" cite="mid:Z5SPWwiB15ptK4hR@black.fi.intel.com"> <pre wrap="" class="moz-quote-pre">On Fri, Jan 24, 2025 at 09:37:55PM +0530, Nilawar, Badal wrote: </pre> <blockquote type="cite"> <pre wrap="" class="moz-quote-pre">On 24-01-2025 20:50, Raag Jadav wrote: </pre> <blockquote type="cite"> <pre wrap="" class="moz-quote-pre">On Fri, Jan 24, 2025 at 08:27:14PM +0530, Nilawar, Badal wrote: </pre> <blockquote type="cite"> <pre wrap="" class="moz-quote-pre">On 24-01-2025 18:03, Raag Jadav wrote: </pre> <blockquote type="cite"> <pre wrap="" class="moz-quote-pre">On Fri, Jan 24, 2025 at 05:29:16PM +0530, Nilawar, Badal wrote: </pre> <blockquote type="cite"> <pre wrap="" class="moz-quote-pre">On 24-01-2025 11:46, Riana Tauro wrote: </pre> <blockquote type="cite"> <pre wrap="" class="moz-quote-pre">Hi Raag On 1/23/2025 8:21 AM, Raag Jadav wrote: </pre> <blockquote type="cite"> <pre wrap="" class="moz-quote-pre">On Tue, Jan 21, 2025 at 01:56:05PM +0530, Riana Tauro wrote: </pre> <blockquote type="cite"> <pre wrap="" class="moz-quote-pre">Hi Raag On 1/8/2025 2:54 PM, Raag Jadav wrote: </pre> <blockquote type="cite"> <pre wrap="" class="moz-quote-pre">Add hwmon support for temp1_input and temp2_input attributes, which will expose package and vram temperature in millidegree Celsius. With this in place we can monitor temperature using lm-sensors tool. Signed-off-by: Raag Jadav <a class="moz-txt-link-rfc2396E" href="mailto:raag.jadav@intel.com"><raag.jadav@intel.com></a> --- .../ABI/testing/sysfs-driver-intel-xe-hwmon | 16 +++++ drivers/gpu/drm/xe/regs/xe_mchbar_regs.h | 3 + drivers/gpu/drm/xe/regs/xe_pcode_regs.h | 2 + drivers/gpu/drm/xe/xe_hwmon.c | 63 +++++++++++++++++++ 4 files changed, 84 insertions(+) diff --git a/Documentation/ABI/testing/sysfs-driver-intel-xe-hwmon b/Documentation/ABI/testing/sysfs-driver-intel-xe-hwmon index d792a56f59ac..998cfb0ee1a6 100644 --- a/Documentation/ABI/testing/sysfs-driver-intel-xe-hwmon +++ b/Documentation/ABI/testing/sysfs-driver-intel-xe-hwmon @@ -108,3 +108,19 @@ Contact: <a class="moz-txt-link-abbreviated" href="mailto:intel-xe@lists.freedesktop.org">intel-xe@lists.freedesktop.org</a> Description: RO. Package current voltage in millivolt. Only supported for particular Intel Xe graphics platforms. + +What: /sys/bus/pci/drivers/xe/.../hwmon/hwmon/temp1_input +Date: April 2025 +KernelVersion: 6.15 +Contact: <a class="moz-txt-link-abbreviated" href="mailto:intel-xe@lists.freedesktop.org">intel-xe@lists.freedesktop.org</a> +Description: RO. Package temperature in millidegree Celsius. + + Only supported for particular Intel Xe graphics platforms. + +What: /sys/bus/pci/drivers/xe/.../hwmon/hwmon/temp2_input +Date: April 2025 +KernelVersion: 6.15 +Contact: <a class="moz-txt-link-abbreviated" href="mailto:intel-xe@lists.freedesktop.org">intel-xe@lists.freedesktop.org</a> +Description: RO. VRAM temperature in millidegree Celsius. + + Only supported for particular Intel Xe graphics platforms. diff --git a/drivers/gpu/drm/xe/regs/xe_mchbar_regs.h b/drivers/gpu/drm/xe/regs/xe_mchbar_regs.h index 519dd1067a19..f5e5234857c1 100644 --- a/drivers/gpu/drm/xe/regs/xe_mchbar_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_mchbar_regs.h @@ -34,6 +34,9 @@ #define PCU_CR_PACKAGE_ENERGY_STATUS XE_REG(MCHBAR_MIRROR_BASE_SNB + 0x593c) +#define PCU_CR_PACKAGE_TEMPERATURE XE_REG(MCHBAR_MIRROR_BASE_SNB + 0x5978) +#define TEMP_MASK REG_GENMASK(7, 0) + #define PCU_CR_PACKAGE_RAPL_LIMIT XE_REG(MCHBAR_MIRROR_BASE_SNB + 0x59a0) #define PKG_PWR_LIM_1 REG_GENMASK(14, 0) #define PKG_PWR_LIM_1_EN REG_BIT(15) diff --git a/drivers/gpu/drm/xe/regs/xe_pcode_regs.h b/drivers/gpu/drm/xe/regs/xe_pcode_regs.h index 0b0b49d850ae..8846eb9ce2a4 100644 --- a/drivers/gpu/drm/xe/regs/xe_pcode_regs.h +++ b/drivers/gpu/drm/xe/regs/xe_pcode_regs.h @@ -21,6 +21,8 @@ #define BMG_PACKAGE_POWER_SKU XE_REG(0x138098) #define BMG_PACKAGE_POWER_SKU_UNIT XE_REG(0x1380dc) #define BMG_PACKAGE_ENERGY_STATUS XE_REG(0x138120) +#define BMG_VRAM_TEMPERATURE XE_REG(0x1382c0) +#define BMG_PACKAGE_TEMPERATURE XE_REG(0x138434) </pre> </blockquote> <pre wrap="" class="moz-quote-pre">indentation. </pre> </blockquote> <pre wrap="" class="moz-quote-pre">It's a git quirk, you won't see it in file. </pre> <blockquote type="cite"> <pre wrap="" class="moz-quote-pre">Also you are using the same for DG2. Should have a common name </pre> </blockquote> <pre wrap="" class="moz-quote-pre">Just following the conventions. </pre> </blockquote> <pre wrap="" class="moz-quote-pre">Did not find this convention in the file. BMG_VRAM_TEMPERATURE is used in both dg2 and bmg and has a bmg prefix. Doesn't seem right </pre> <blockquote type="cite"> <blockquote type="cite"> <blockquote type="cite"> <pre wrap="" class="moz-quote-pre"> #define BMG_PACKAGE_RAPL_LIMIT XE_REG(0x138440) #define BMG_PLATFORM_ENERGY_STATUS XE_REG(0x138458) #define BMG_PLATFORM_POWER_LIMIT XE_REG(0x138460) diff --git a/drivers/gpu/drm/xe/xe_hwmon.c b/drivers/gpu/drm/xe/xe_hwmon.c index fde56dad3ab7..5b5c844adf4a 100644 --- a/drivers/gpu/drm/xe/xe_hwmon.c +++ b/drivers/gpu/drm/xe/xe_hwmon.c @@ -6,6 +6,7 @@ #include <linux/hwmon-sysfs.h> #include <linux/hwmon.h> #include <linux/types.h> +#include <linux/units.h> #include <drm/drm_managed.h> #include "regs/xe_gt_regs.h" @@ -20,6 +21,7 @@ #include "xe_pm.h" enum xe_hwmon_reg { + REG_TEMP, </pre> </blockquote> <pre wrap="" class="moz-quote-pre">add to the end </pre> <blockquote type="cite"> <pre wrap="" class="moz-quote-pre"> REG_PKG_RAPL_LIMIT, REG_PKG_POWER_SKU, REG_PKG_POWER_SKU_UNIT, @@ -39,6 +41,11 @@ enum xe_hwmon_channel { CHANNEL_MAX, }; +enum xe_hwmon_temp { + TEMP_PKG, + TEMP_VRAM, +}; </pre> </blockquote> <pre wrap="" class="moz-quote-pre">Can't the existing channel enum be used here? </pre> </blockquote> <pre wrap="" class="moz-quote-pre">Nope, that'd break the indexes. </pre> </blockquote> <pre wrap="" class="moz-quote-pre">@badal/@karthik Are multiple indexes for the same channel okay? In the current code, for dg2 only channel 1 is exposed for power and channel 0 skipped. Something like that needs to be done here too? </pre> </blockquote> <pre wrap="" class="moz-quote-pre">Thanks for looping me in this. Yes, Channel 0 represent card specific attributes and Channel 1 represent package specific attributes. That's how it should be followed. With that BMG_PACKAGE_TEMPERATURE should go under CHANNEL_PKG. For BMG_VRAM_TEMPERATURE new channel (channel 3) should be added in enum xe_hwmon_channel. </pre> </blockquote> <pre wrap="" class="moz-quote-pre">And how does that work with hwmon_channel_info? </pre> </blockquote> <pre wrap="" class="moz-quote-pre">Check curr_crit implementation. HWMON_CHANNEL_INFO(curr, HWMON_C_LABEL, HWMON_C_CRIT | HWMON_C_LABEL) </pre> </blockquote> <pre wrap="" class="moz-quote-pre">Exactly, and hence the separate enums for temp channels. </pre> </blockquote> <pre wrap="" class="moz-quote-pre"> No, we want temp2_input for package (channel 2) temperature and temp3_input (Channel 3) for VRAM temperature. Just for information, in i915 we create separate hwmon node for each layer i.e. package, gt0 and gt1. During review of xe hwmon implementation upstream architect recommended that there should be single node for a device. So we are using channel based approach. So lets stick to that approach. </pre> </blockquote> <pre wrap="" class="moz-quote-pre"> And what about fan channels? Are they to be indexed from 4 and on? I'm not sure if we'd want to abuse hwmon_channel_info in such a way. Raag</pre> </blockquote> Hi Raag, We have adopted a hierarchical approach for HWMON channels and their entries. Channel 1 is designated for card-related information, while Channel 2 is used for package-related information. If new HWMON data does not fit within these channels, we can create a new channel and assign an appropriate label to it. For instance, you can map package_temperature under temp2_xxx. Since there is no temperature data for the card, temp1_xxx can be made invisible. For VRAM temperature, add CHANNEL_VRAM, which will be temp3_xxx with the label temp3_label set to “vram”. Regarding fan channels, based on our offline discussion, it appears that fan data does not fall within the above mappings and pertains to the entire card. For it we can have a separate enum like, enum xe_hwmon_fan_channel { FAN_1, FAN_2, FAN_3 }; for that, with label "card" for all those. <blockquote type="cite" cite="mid:Z5SPWwiB15ptK4hR@black.fi.intel.com"> <pre wrap="" class="moz-quote-pre"> </pre> </blockquote> </body> </html>