Message ID | 20200116141800.9828-1-linux@roeck-us.net (mailing list archive) |
---|---|
Headers | show |
Series | hwmon: k10temp driver improvements | expand |
Tested-By: Darren Salt <devspam@moreofthesa.me.uk>
Linux 5.4.12, Ryzen 5 1600. Patches were applied cleanly. No problems noticed in
$ sensors k10temp-pci-00c3
k10temp-pci-00c3
Adapter: PCI adapter
Vcore: +1.11 V
Vsoc: +0.94 V
Tdie: +42.8°C (high = +70.0°C)
Tctl: +42.8°C
Icore: +15.59 A
Isoc: +12.63 A
$
On Thu, Jan 16, 2020 at 08:55:16PM +0000, Darren Salt wrote: > Tested-By: Darren Salt <devspam@moreofthesa.me.uk> > Thanks! Guenter > Linux 5.4.12, Ryzen 5 1600. Patches were applied cleanly. No problems noticed in > > $ sensors k10temp-pci-00c3 > k10temp-pci-00c3 > Adapter: PCI adapter > Vcore: +1.11 V > Vsoc: +0.94 V > Tdie: +42.8°C (high = +70.0°C) > Tctl: +42.8°C > Icore: +15.59 A > Isoc: +12.63 A > > $ > > -- > | _ | Darren Salt, using Debian GNU/Linux (and Android) > | ( ) | > | X | ASCII Ribbon campaign against HTML e-mail > | / \ |
Tested-by: Bernhard Gebetsberger <bernhard.gebetsberger@gmx.at>
Patches applied cleanly on top of 5.5-rc6, no issues using a Ryzen 3 2200G:
k10temp-pci-00c3
Adapter: PCI adapter
Vcore: 1.29 V
Vsoc: 1.12 V
Tdie: +28.2°C (high = +70.0°C)
Tctl: +28.2°C
Icore: 23.90 A
Isoc: 6.49 A
- Bernhard
On Thu, Jan 16, 2020 at 11:46:47PM +0100, Bernhard Gebetsberger wrote: > Tested-by: Bernhard Gebetsberger <bernhard.gebetsberger@gmx.at> > > Patches applied cleanly on top of 5.5-rc6, no issues using a Ryzen 3 2200G: > k10temp-pci-00c3 > Adapter: PCI adapter > Vcore: 1.29 V > Vsoc: 1.12 V > Tdie: +28.2°C (high = +70.0°C) > Tctl: +28.2°C > Icore: 23.90 A > Isoc: 6.49 A > Thanks! Guenter
On Thu, 16 Jan 2020 at 14:18, Guenter Roeck <linux@roeck-us.net> wrote: > > This patch series implements various improvements for the k10temp driver. > > Patch 1/4 introduces the use of bit operations. > > Patch 2/4 converts the driver to use the devm_hwmon_device_register_with_info > API. This not only simplifies the code and reduces its size, it also > makes the code easier to maintain and enhance. > > Patch 3/4 adds support for reporting Core Complex Die (CCD) temperatures > on Ryzen 3 (Zen2) CPUs. > > Patch 4/4 adds support for reporting core and SoC current and voltage > information on Ryzen CPUs. > > k10temp-pci-00c3 > Adapter: PCI adapter > Vcore: +1.36 V > Vsoc: +1.18 V > Tdie: +86.8°C (high = +70.0°C) > Tctl: +86.8°C > Tccd1: +80.0°C > Tccd2: +81.8°C > Icore: +44.14 A > Isoc: +13.83 A > > The patch series has only been tested with Ryzen 3900 CPUs. Further test > coverage will be necessary before the changes can be applied to the Linux > kernel. I have some Zen1 and Zen1+ here. My Ryzen 3 1300X, applied to 5.5.0-rc5 machine idle, I thought at first the temperature may be a bit low, so I've added other reported temperatures. I now think it is maybe ok. k10temp-pci-00c3 Adapter: PCI adapter Vcore: +1.41 V Vsoc: +0.89 V Tdie: +21.2°C (high = +70.0°C) Tctl: +21.2°C Icore: +30.14 A Isoc: +8.66 A SYSTIN: +29.0°C (high = +0.0°C, hyst = +0.0°C) ALARM sensor = thermistor CPUTIN: +25.5°C (high = +80.0°C, hyst = +75.0°C) sensor = thermistor AUXTIN0: -1.5°C sensor = thermistor AUXTIN1: +87.0°C sensor = thermistor AUXTIN2: +23.0°C sensor = thermistor AUXTIN3: -27.0°C sensor = thermistor SMBUSMASTER 0: +20.5°C After about 2 minutes of make -j8 on kernel, to load it k10temp-pci-00c3 Adapter: PCI adapter Vcore: +1.26 V Vsoc: +0.89 V Tdie: +46.2°C (high = +70.0°C) Tctl: +46.2°C Icore: +45.73 A Isoc: +11.18 A SYSTIN: +29.0°C (high = +0.0°C, hyst = +0.0°C) ALARM sensor = thermistor CPUTIN: +38.5°C (high = +80.0°C, hyst = +75.0°C) sensor = thermistor AUXTIN0: -7.5°C sensor = thermistor AUXTIN1: +85.0°C sensor = thermistor AUXTIN2: +23.0°C sensor = thermistor AUXTIN3: -27.0°C sensor = thermistor SMBUSMASTER 0: +46.0°C So I guess the temperatures *are* in the right area. Interestingly, the Vcore restores to above +1.4V when idle. And my Ryzen 5 3400G (Zen+), applied to 5.4.12, box is idle, also showing the gpu measurements of this APU to confirm the temperature: k10temp-pci-00c3 Adapter: PCI adapter Vcore: +0.94 V Vsoc: +1.09 V Tdie: +34.8°C (high = +70.0°C) Tctl: +34.8°C Icore: +6.24 A Isoc: +8.30 A amdgpu-pci-0900 Adapter: PCI adapter vddgfx: N/A vddnb: N/A edge: +34.0°C (crit = +80.0°C, hyst = +0.0°C) For my Ryzen 5 2500u laptop (Zen1), again showing the gpu: k10temp-pci-00c3 Adapter: PCI adapter Vcore: +0.97 V Vsoc: +0.93 V Tdie: +37.2°C (high = +70.0°C) Tctl: +37.2°C Icore: +19.75 A Isoc: +8.66 A amdgpu-pci-0300 Adapter: PCI adapter vddgfx: N/A vddnb: N/A edge: +37.0°C (crit = +80.0°C, hyst = +0.0°C) Thanks. ĸen
Hi Ken, On 1/16/20 4:38 PM, Ken Moffat wrote: > On Thu, 16 Jan 2020 at 14:18, Guenter Roeck <linux@roeck-us.net> wrote: [ ... ] > I have some Zen1 and Zen1+ here. > > My Ryzen 3 1300X, applied to 5.5.0-rc5 > > machine idle, I thought at first the temperature may be a bit low, so > I've added other reported temperatures. I now think it is maybe ok. > > k10temp-pci-00c3 > Adapter: PCI adapter > Vcore: +1.41 V > Vsoc: +0.89 V > Tdie: +21.2°C (high = +70.0°C) > Tctl: +21.2°C > Icore: +30.14 A > Isoc: +8.66 A > > SYSTIN: +29.0°C (high = +0.0°C, hyst = +0.0°C) > ALARM sensor = thermistor > CPUTIN: +25.5°C (high = +80.0°C, hyst = +75.0°C) > sensor = thermistor > AUXTIN0: -1.5°C sensor = thermistor > AUXTIN1: +87.0°C sensor = thermistor > AUXTIN2: +23.0°C sensor = thermistor > AUXTIN3: -27.0°C sensor = thermistor > SMBUSMASTER 0: +20.5°C > SMBUSMASTER 0 is the CPU, so we have a match with the temperatures. > After about 2 minutes of make -j8 on kernel, to load it > > k10temp-pci-00c3 > Adapter: PCI adapter > Vcore: +1.26 V > Vsoc: +0.89 V > Tdie: +46.2°C (high = +70.0°C) > Tctl: +46.2°C > Icore: +45.73 A > Isoc: +11.18 A > Both Vcore and Icore should be much less when idle, and higher under load. The data from the Super-IO chip suggests that it is a Nuvoton chip. Can you report its first voltage (in0) ? That should roughly match Vcore. > SYSTIN: +29.0°C (high = +0.0°C, hyst = +0.0°C) > ALARM sensor = thermistor > CPUTIN: +38.5°C (high = +80.0°C, hyst = +75.0°C) > sensor = thermistor > AUXTIN0: -7.5°C sensor = thermistor > AUXTIN1: +85.0°C sensor = thermistor > AUXTIN2: +23.0°C sensor = thermistor > AUXTIN3: -27.0°C sensor = thermistor > SMBUSMASTER 0: +46.0°C > > So I guess the temperatures *are* in the right area. > Interestingly, the Vcore restores to above +1.4V when idle. > It should be much lower when idle, actually, not higher. All other data looks ok. Thanks, Guenter
On Fri, 17 Jan 2020 at 03:58, Guenter Roeck <linux@roeck-us.net> wrote: > > Hi Ken, > > SMBUSMASTER 0 is the CPU, so we have a match with the temperatures. > OK, thanks for that information. > > Both Vcore and Icore should be much less when idle, and higher under > load. The data from the Super-IO chip suggests that it is a Nuvoton > chip. Can you report its first voltage (in0) ? That should roughly > match Vcore. > > All other data looks ok. > > Thanks, > Guenter Hi Guenter, unfortunately I don't have any report of in0. I'm guessing I need some module(s) which did not seem to do anything useful in the past. All I have in the 'in' area is nct6779-isa-0290 Adapter: ISA adapter Vcore: +0.30 V (min = +0.00 V, max = +1.74 V) in1: +0.00 V (min = +0.00 V, max = +0.00 V) AVCC: +3.39 V (min = +0.00 V, max = +0.00 V) ALARM +3.3V: +3.39 V (min = +0.00 V, max = +0.00 V) ALARM in4: +1.90 V (min = +0.00 V, max = +0.00 V) ALARM in5: +0.90 V (min = +0.00 V, max = +0.00 V) ALARM in6: +1.50 V (min = +0.00 V, max = +0.00 V) ALARM 3VSB: +3.47 V (min = +0.00 V, max = +0.00 V) ALARM Vbat: +3.26 V (min = +0.00 V, max = +0.00 V) ALARM in9: +0.00 V (min = +0.00 V, max = +0.00 V) in10: +0.32 V (min = +0.00 V, max = +0.00 V) ALARM in11: +1.06 V (min = +0.00 V, max = +0.00 V) ALARM in12: +1.70 V (min = +0.00 V, max = +0.00 V) ALARM in13: +0.94 V (min = +0.00 V, max = +0.00 V) ALARM in14: +1.84 V (min = +0.00 V, max = +0.00 V) ALARM and at that point Vcore was reported as 1.41V (system idle) ĸen
Dňa 16. 1. 2020 o 15:17 Guenter Roeck napísal(a): > This patch series implements various improvements for the k10temp driver. > > Patch 1/4 introduces the use of bit operations. > > Patch 2/4 converts the driver to use the devm_hwmon_device_register_with_info > API. This not only simplifies the code and reduces its size, it also > makes the code easier to maintain and enhance. > > Patch 3/4 adds support for reporting Core Complex Die (CCD) temperatures > on Ryzen 3 (Zen2) CPUs. > > Patch 4/4 adds support for reporting core and SoC current and voltage > information on Ryzen CPUs. > > With all patches in place, output on Ryzen 3900 CPUs looks as follows > (with the system under load). > > k10temp-pci-00c3 > Adapter: PCI adapter > Vcore: +1.36 V > Vsoc: +1.18 V > Tdie: +86.8°C (high = +70.0°C) > Tctl: +86.8°C > Tccd1: +80.0°C > Tccd2: +81.8°C > Icore: +44.14 A > Isoc: +13.83 A > > The patch series has only been tested with Ryzen 3900 CPUs. Further test > coverage will be necessary before the changes can be applied to the Linux > kernel. > Hello everyone, I am the author of https://github.com/ocerman/zenpower/ . It is nice to see this merged. I just want to warn you that there have been reported issues with Threadripper CPUs to zenpower issue tracker. Also I think that no-one tested EPYC CPUs. Most of the stuff I was able to figure out by trial-and-error approach and unfortunately because I do not own any Threadripper CPU I was not able to test and fix reported problems. Ondrej.
On Thu, 16 Jan 2020, Guenter Roeck wrote: > This patch series implements various improvements for the k10temp driver. > > Patch 1/4 introduces the use of bit operations. > > Patch 2/4 converts the driver to use the devm_hwmon_device_register_with_info > API. This not only simplifies the code and reduces its size, it also > makes the code easier to maintain and enhance. > > Patch 3/4 adds support for reporting Core Complex Die (CCD) temperatures > on Ryzen 3 (Zen2) CPUs. > > Patch 4/4 adds support for reporting core and SoC current and voltage > information on Ryzen CPUs. > > With all patches in place, output on Ryzen 3900 CPUs looks as follows > (with the system under load). > > k10temp-pci-00c3 > Adapter: PCI adapter > Vcore: +1.36 V > Vsoc: +1.18 V > Tdie: +86.8°C (high = +70.0°C) > Tctl: +86.8°C > Tccd1: +80.0°C > Tccd2: +81.8°C > Icore: +44.14 A > Isoc: +13.83 A > > The patch series has only been tested with Ryzen 3900 CPUs. Further test > coverage will be necessary before the changes can be applied to the Linux > kernel. > Here from my little Asrock A300 with a Ryzen 2400G: sensors k10temp-pci-00c3 Adapter: PCI adapter Vcore: +0.78 V Vsoc: +1.11 V Tdie: +44.8°C (high = +70.0°C) Tctl: +44.8°C Icore: +5.20 A Isoc: +2.17 A nvme-pci-0100 Adapter: PCI adapter Composite: +41.9°C (low = -273.1°C, high = +80.8°C) (crit = +80.8°C) Sensor 1: +41.9°C (low = -273.1°C, high = +65261.8°C) Sensor 2: +44.9°C (low = -273.1°C, high = +65261.8°C) nct6793-isa-0290 Adapter: ISA adapter in0: +0.34 V (min = +0.00 V, max = +1.74 V) in1: +1.84 V (min = +0.00 V, max = +0.00 V) ALARM in2: +3.41 V (min = +0.00 V, max = +0.00 V) ALARM in3: +3.39 V (min = +0.00 V, max = +0.00 V) ALARM in4: +0.26 V (min = +0.00 V, max = +0.00 V) ALARM in5: +0.14 V (min = +0.00 V, max = +0.00 V) ALARM in6: +0.67 V (min = +0.00 V, max = +0.00 V) ALARM in7: +3.39 V (min = +0.00 V, max = +0.00 V) ALARM in8: +3.26 V (min = +0.00 V, max = +0.00 V) ALARM in9: +1.84 V (min = +0.00 V, max = +0.00 V) ALARM in10: +0.19 V (min = +0.00 V, max = +0.00 V) ALARM in11: +0.14 V (min = +0.00 V, max = +0.00 V) ALARM in12: +1.85 V (min = +0.00 V, max = +0.00 V) ALARM in13: +1.72 V (min = +0.00 V, max = +0.00 V) ALARM in14: +0.20 V (min = +0.00 V, max = +0.00 V) ALARM fan1: 0 RPM (min = 0 RPM) fan2: 317 RPM (min = 0 RPM) fan3: 0 RPM (min = 0 RPM) fan4: 0 RPM (min = 0 RPM) fan5: 0 RPM (min = 0 RPM) SYSTIN: +113.0°C (high = +0.0°C, hyst = +0.0°C) sensor = thermistor CPUTIN: +59.5°C (high = +80.0°C, hyst = +75.0°C) sensor = thermistor AUXTIN0: +45.0°C (high = +0.0°C, hyst = +0.0°C) ALARM sensor = thermistor AUXTIN1: +107.0°C sensor = thermistor AUXTIN2: +106.0°C sensor = thermistor AUXTIN3: +103.0°C sensor = thermistor SMBUSMASTER 0: +44.5°C PCH_CHIP_CPU_MAX_TEMP: +0.0°C PCH_CHIP_TEMP: +0.0°C PCH_CPU_TEMP: +0.0°C intrusion0: OK intrusion1: ALARM beep_enable: disabled amdgpu-pci-0300 Adapter: PCI adapter vddgfx: N/A vddnb: N/A edge: +44.0°C (crit = +80.0°C, hyst = +0.0°C) Patches applied without any problem against Linus git tree. Many thanks for this work! Regards, Holger
On 1/16/20 8:47 PM, Ken Moffat wrote: > On Fri, 17 Jan 2020 at 03:58, Guenter Roeck <linux@roeck-us.net> wrote: >> >> Hi Ken, >> >> SMBUSMASTER 0 is the CPU, so we have a match with the temperatures. >> > OK, thanks for that information. > >> >> Both Vcore and Icore should be much less when idle, and higher under >> load. The data from the Super-IO chip suggests that it is a Nuvoton >> chip. Can you report its first voltage (in0) ? That should roughly >> match Vcore. >> >> All other data looks ok. >> >> Thanks, >> Guenter > > Hi Guenter, > > unfortunately I don't have any report of in0. I'm guessing I need some > module(s) which did not seem to do anything useful in the past. > > All I have in the 'in' area is > nct6779-isa-0290 > Adapter: ISA adapter > Vcore: +0.30 V (min = +0.00 V, max = +1.74 V) > in1: +0.00 V (min = +0.00 V, max = +0.00 V) > AVCC: +3.39 V (min = +0.00 V, max = +0.00 V) ALARM > +3.3V: +3.39 V (min = +0.00 V, max = +0.00 V) ALARM > in4: +1.90 V (min = +0.00 V, max = +0.00 V) ALARM > in5: +0.90 V (min = +0.00 V, max = +0.00 V) ALARM > in6: +1.50 V (min = +0.00 V, max = +0.00 V) ALARM > 3VSB: +3.47 V (min = +0.00 V, max = +0.00 V) ALARM > Vbat: +3.26 V (min = +0.00 V, max = +0.00 V) ALARM > in9: +0.00 V (min = +0.00 V, max = +0.00 V) > in10: +0.32 V (min = +0.00 V, max = +0.00 V) ALARM > in11: +1.06 V (min = +0.00 V, max = +0.00 V) ALARM > in12: +1.70 V (min = +0.00 V, max = +0.00 V) ALARM > in13: +0.94 V (min = +0.00 V, max = +0.00 V) ALARM > in14: +1.84 V (min = +0.00 V, max = +0.00 V) ALARM > > and at that point Vcore was reported as 1.41V (system idle) > Looks like someone configured /etc/sensors3.conf on that system which tells it to report in0 as Vcore. So there is a very clear mismatch. Can you report the values seen when the system is under load ? Thanks, Guenter
On Fri, Jan 17, 2020 at 10:46:25AM +0100, Ondrej Čerman wrote: > Dňa 16. 1. 2020 o 15:17 Guenter Roeck napísal(a): > > This patch series implements various improvements for the k10temp driver. > > > > Patch 1/4 introduces the use of bit operations. > > > > Patch 2/4 converts the driver to use the devm_hwmon_device_register_with_info > > API. This not only simplifies the code and reduces its size, it also > > makes the code easier to maintain and enhance. > > > > Patch 3/4 adds support for reporting Core Complex Die (CCD) temperatures > > on Ryzen 3 (Zen2) CPUs. > > > > Patch 4/4 adds support for reporting core and SoC current and voltage > > information on Ryzen CPUs. > > > > With all patches in place, output on Ryzen 3900 CPUs looks as follows > > (with the system under load). > > > > k10temp-pci-00c3 > > Adapter: PCI adapter > > Vcore: +1.36 V > > Vsoc: +1.18 V > > Tdie: +86.8°C (high = +70.0°C) > > Tctl: +86.8°C > > Tccd1: +80.0°C > > Tccd2: +81.8°C > > Icore: +44.14 A > > Isoc: +13.83 A > > > > The patch series has only been tested with Ryzen 3900 CPUs. Further test > > coverage will be necessary before the changes can be applied to the Linux > > kernel. > > > Hello everyone, I am the author of https://github.com/ocerman/zenpower/ . > > It is nice to see this merged. > > I just want to warn you that there have been reported issues with > Threadripper CPUs to zenpower issue tracker. Also I think that no-one tested > EPYC CPUs. > > Most of the stuff I was able to figure out by trial-and-error approach and > unfortunately because I do not own any Threadripper CPU I was not able to > test and fix reported problems. > Thanks a lot for the note. The key problem seems to be that Threadripper doesn't report SoC current and voltage. Is that correct ? If so, that should be easy to solve. On a side note, drivers/gpu/drm/amd/include/asic_reg/thm/thm_10_0_offset.h suggests that two more temperature sensors might be available at 0x0005995C and 0x00059960 (DIE3_TEMP and SW_TEMP). Have you ever tried that ? Thanks, Guenter
On Fri, 17 Jan 2020 at 14:14, Guenter Roeck <linux@roeck-us.net> wrote: > > On 1/16/20 8:47 PM, Ken Moffat wrote: > > unfortunately I don't have any report of in0. I'm guessing I need some > > module(s) which did not seem to do anything useful in the past. > > > > All I have in the 'in' area is > > nct6779-isa-0290 > > Adapter: ISA adapter > > Vcore: +0.30 V (min = +0.00 V, max = +1.74 V) > > in1: +0.00 V (min = +0.00 V, max = +0.00 V) > > AVCC: +3.39 V (min = +0.00 V, max = +0.00 V) ALARM > > +3.3V: +3.39 V (min = +0.00 V, max = +0.00 V) ALARM > > Looks like someone configured /etc/sensors3.conf on that system which tells it > to report in0 as Vcore. So there is a very clear mismatch. Can you report > the values seen when the system is under load ? > > Thanks, > Guenter I do have sensors3.conf from lm_sensors-3.4.0. Here are the figures under load. Vcore: +0.65 V (min = +0.00 V, max = +1.74 V) k10temp-pci-00c3 Adapter: PCI adapter Vcore: +1.27 V Vsoc: +0.89 V Tdie: +46.2°C (high = +70.0°C) Tctl: +46.2°C Icore: +48.84 A Isoc: +10.10 A ĸen
Hi, On Thu, Jan 16, 2020 at 06:17:56AM -0800, Guenter Roeck wrote: > This patch series implements various improvements for the k10temp driver. > > Patch 1/4 introduces the use of bit operations. > > Patch 2/4 converts the driver to use the devm_hwmon_device_register_with_info > API. This not only simplifies the code and reduces its size, it also > makes the code easier to maintain and enhance. > > Patch 3/4 adds support for reporting Core Complex Die (CCD) temperatures > on Ryzen 3 (Zen2) CPUs. > > Patch 4/4 adds support for reporting core and SoC current and voltage > information on Ryzen CPUs. > > With all patches in place, output on Ryzen 3900 CPUs looks as follows > (with the system under load). > > k10temp-pci-00c3 > Adapter: PCI adapter > Vcore: +1.36 V > Vsoc: +1.18 V > Tdie: +86.8°C (high = +70.0°C) > Tctl: +86.8°C > Tccd1: +80.0°C > Tccd2: +81.8°C > Icore: +44.14 A > Isoc: +13.83 A > > The patch series has only been tested with Ryzen 3900 CPUs. Further test > coverage will be necessary before the changes can be applied to the Linux > kernel. Looks ok on 3800X (idle): $ lscpu | grep "Model name" Model name: AMD Ryzen 7 3800X 8-Core Processor $ sensors "k10temp-*" k10temp-pci-00c3 Adapter: PCI adapter Vcore: 937.00 mV Vsoc: 1.01 V Tdie: +35.2°C (high = +70.0°C) Tctl: +35.2°C Tccd1: +35.8°C Icore: 4.61 A Isoc: 6.18 A And after compiling the kernel with 32 threads for 1 minute: $ sensors "k10temp-*" k10temp-pci-00c3 Adapter: PCI adapter Vcore: 1.29 V Vsoc: 1.01 V Tdie: +77.1°C (high = +70.0°C) Tctl: +77.1°C Tccd1: +78.8°C Icore: 39.53 A Isoc: 6.18 A Board Information during the idle check: $ sudo dmidecode -s system-manufacturer Gigabyte Technology Co., Ltd. $ sudo dmidecode -s system-product-name X570 AORUS ULTRA $ sensors "it8792-*" it8792-isa-0a60 Adapter: ISA adapter in0: 1.79 V (min = +0.00 V, max = +2.78 V) in1: 589.00 mV (min = +0.00 V, max = +2.78 V) in2: 981.00 mV (min = +0.00 V, max = +2.78 V) +3.3V: 1.68 V (min = +0.00 V, max = +2.78 V) in4: 1.79 V (min = +0.00 V, max = +2.78 V) in5: 1.18 V (min = +0.00 V, max = +2.78 V) in6: 2.78 V (min = +0.00 V, max = +2.78 V) ALARM 3VSB: 1.68 V (min = +0.00 V, max = +2.78 V) Vbat: 1.61 V fan1: 0 RPM (min = 0 RPM) fan2: 0 RPM (min = 0 RPM) fan3: 0 RPM (min = 0 RPM) temp1: +37.0°C (low = +127.0°C, high = +127.0°C) sensor = thermistor temp3: +36.0°C (low = +127.0°C, high = +127.0°C) sensor = thermistor intrusion0: ALARM -- Sebastian
Dňa 17. 1. 2020 o 19:46 Guenter Roeck napísal(a): > On Fri, Jan 17, 2020 at 10:46:25AM +0100, Ondrej Čerman wrote: >> Dňa 16. 1. 2020 o 15:17 Guenter Roeck napísal(a): >>> This patch series implements various improvements for the k10temp driver. >>> >>> Patch 1/4 introduces the use of bit operations. >>> >>> Patch 2/4 converts the driver to use the devm_hwmon_device_register_with_info >>> API. This not only simplifies the code and reduces its size, it also >>> makes the code easier to maintain and enhance. >>> >>> Patch 3/4 adds support for reporting Core Complex Die (CCD) temperatures >>> on Ryzen 3 (Zen2) CPUs. >>> >>> Patch 4/4 adds support for reporting core and SoC current and voltage >>> information on Ryzen CPUs. >>> >>> With all patches in place, output on Ryzen 3900 CPUs looks as follows >>> (with the system under load). >>> >>> k10temp-pci-00c3 >>> Adapter: PCI adapter >>> Vcore: +1.36 V >>> Vsoc: +1.18 V >>> Tdie: +86.8°C (high = +70.0°C) >>> Tctl: +86.8°C >>> Tccd1: +80.0°C >>> Tccd2: +81.8°C >>> Icore: +44.14 A >>> Isoc: +13.83 A >>> >>> The patch series has only been tested with Ryzen 3900 CPUs. Further test >>> coverage will be necessary before the changes can be applied to the Linux >>> kernel. >>> >> Hello everyone, I am the author of https://github.com/ocerman/zenpower/ . >> >> It is nice to see this merged. >> >> I just want to warn you that there have been reported issues with >> Threadripper CPUs to zenpower issue tracker. Also I think that no-one tested >> EPYC CPUs. >> >> Most of the stuff I was able to figure out by trial-and-error approach and >> unfortunately because I do not own any Threadripper CPU I was not able to >> test and fix reported problems. >> > Thanks a lot for the note. The key problem seems to be that Threadripper > doesn't report SoC current and voltage. Is that correct ? If so, that > should be easy to solve. Hello, I thought that initially, but I was wrong. It seems like that these multi-node CPUs are reporting SOC and Core voltage/current data at particular node. Look at this HWiNFO64 screenshot of 2990WX for reference: https://i.imgur.com/yM9X5nd.jpg . They also may be using different addresses and/or factors. > On a side note, drivers/gpu/drm/amd/include/asic_reg/thm/thm_10_0_offset.h > suggests that two more temperature sensors might be available at 0x0005995C > and 0x00059960 (DIE3_TEMP and SW_TEMP). Have you ever tried that ? > > Thanks, > Guenter I was aware of 0005995c and I thought that it could be Tdie3 (that's why I have included it in debug output, someone already shared that 3960X is reporting data on that address). I think this one can be safely included. I was not aware of the other address, I will try it. Ondrej.
On 16/1/20 10:17 pm, Guenter Roeck wrote: > This patch series implements various improvements for the k10temp driver. > Looks good here. Identical motherboards (ASUS x370 Prime-Pro), different CPUs. 3950x k10temp-pci-00c3 Adapter: PCI adapter Vcore: +1.38 V Vsoc: +1.08 V Tdie: +69.1°C (high = +70.0°C) Tctl: +69.1°C Tccd1: +54.2°C Tccd2: +57.0°C Icore: +27.67 A Isoc: +14.13 A it8665-isa-0290 Adapter: ISA adapter Vcore: +1.41 V (min = +0.83 V, max = +1.65 V) in1: +2.51 V (min = +1.98 V, max = +2.73 V) +12V: +11.98 V (min = +11.20 V, max = +12.40 V) +5V: +5.01 V (min = +4.74 V, max = +5.61 V) 3VSB: +6.67 V (min = +2.83 V, max = +3.40 V) Vbat: +6.58 V +3.3V: +3.33 V CPU Fan: 3409 RPM (min = 1500 RPM) Back Fan: 0 RPM (min = 0 RPM) MB CPU Temp: +56.0°C (low = +13.0°C, high = +88.0°C) Ambient: +35.0°C (low = +13.0°C, high = +43.0°C) sensor = thermistor PCH: +46.0°C (low = +18.0°C, high = +61.0°C) sensor = thermistor 1800x k10temp-pci-00c3 Adapter: PCI adapter Vcore: +1.26 V Vsoc: +0.91 V Tdie: +36.0°C (high = +70.0°C) Tctl: +56.0°C Icore: +15.59 A Isoc: +7.94 A it8665-isa-0290 Adapter: ISA adapter Vcore: +1.25 V (min = +0.83 V, max = +1.65 V) in1: +2.48 V (min = +1.98 V, max = +2.73 V) +12V: +11.98 V (min = +11.20 V, max = +12.40 V) +5V: +4.96 V (min = +4.74 V, max = +5.61 V) 3VSB: +6.54 V (min = +2.83 V, max = +3.40 V) Vbat: +6.37 V +3.3V: +3.31 V CPU Fan: 1171 RPM (min = 1500 RPM) ALARM Back Fan: 0 RPM (min = 0 RPM) MB CPU Temp: +36.0°C (low = +13.0°C, high = +88.0°C) Ambient: +44.0°C (low = +13.0°C, high = +43.0°C) sensor = thermistor PCH: +38.0°C (low = +18.0°C, high = +61.0°C) sensor = thermistor Regards, Brad
On 1/18/20 12:52 AM, Brad Campbell wrote: > On 16/1/20 10:17 pm, Guenter Roeck wrote: >> This patch series implements various improvements for the k10temp driver. >> > > Looks good here. Identical motherboards (ASUS x370 Prime-Pro), different CPUs. > > 3950x > Interesting. I thought the 3950X needs a newer motherboard. Is that CPU as amazing as everyone says it is ? And does it really need liquid cooling ? Anyway, thanks a lot for testing! Guenter
On 19/1/20 1:14 am, Guenter Roeck wrote: > On 1/18/20 12:52 AM, Brad Campbell wrote: >> On 16/1/20 10:17 pm, Guenter Roeck wrote: >>> This patch series implements various improvements for the k10temp >>> driver. >>> >> >> Looks good here. Identical motherboards (ASUS x370 Prime-Pro), >> different CPUs. >> >> 3950x >> > Interesting. I thought the 3950X needs a newer motherboard. This board has the 3950x listed on the compatibility matrix, so I took the punt. I am running a beta BIOS with the 1.0.0.4 AGESA but it's running in a stock production environment and has been stable. I don't overclock or game, it's predominantly a VM host. > Is that CPU as amazing as everyone says it is ? It is fairly impressive and a significant update over the 1800x it replaced. Kernel compiles are pretty quick :) > And does it really need liquid cooling ? No. I'm using a stock AMD Wraith Prism cooler in a 4U rack case. It might reach higher boost clocks with better cooling, but under an all-core load I still see > 4.1GHz across all cores. Regards, Brad