Hi Kim
On 20/12/2022 19:01, Kim Steiner wrote:
Hi Dmitry,
With these cpufreq changes to the branch, can I test the original GPU temperature issue?
Yes, please.
I don't think I mentioned it, but boot time seems faster now with the cpufreq patch you last sent 😊
Great :-)
Thanks, Kim
-----Original Message----- From: Dmitry Baryshkov dmitry.baryshkov@linaro.org Sent: Monday, December 19, 2022 4:45 PM To: Kim Steiner kim.steiner@sightlineapplications.com Cc: Jordan Holt jordan.holt@sightlineapplications.com; Paul Neuhardt paul.neuhardt@linaro.org; sightlineapplications@lists.linaro.org Subject: Re: [Sightlineapplications] Re: Project Status and Burn Report
Hi Kim,
On Sat, 3 Dec 2022 at 00:19, Kim Steiner kim.steiner@sightlineapplications.com wrote:
Hi Dmitry,
It was the 'maxcpus=2' cmdline argument. I removed it to get 4 cpu cores. On our system we will need all 4 cores. Also, we will need to be able to set the cores to atleast 1GHz. We used these commands in 4.14 kernel: cpufreq-set --cpu 0 --governor performance --min 1000Mhz --max 1000Mhz cpufreq-set --cpu 2 --governor performance --min 1000Mhz --max 1000Mhz
I have updated the release branch with the patches that fix the cpufreq issue.
Using this kernel and these commands I can set the required frequencies:
# cpufreq-info | grep current current policy: frequency should be within 1.04 GHz and 1.04 GHz. current CPU frequency is 1.04 GHz (asserted by call to hardware). current policy: frequency should be within 1.04 GHz and 1.04 GHz. current CPU frequency is 1.04 GHz (asserted by call to hardware). current policy: frequency should be within 1.04 GHz and 1.04 GHz. current CPU frequency is 1.04 GHz (asserted by call to hardware). current policy: frequency should be within 1.04 GHz and 1.04 GHz. current CPU frequency is 1.04 GHz (asserted by call to hardware).
but they do not work in the 5.15 branch.
Thanks, Kim
-----Original Message----- From: Kim Steiner Sent: Friday, December 2, 2022 12:38 PM To: Dmitry Baryshkov dmitry.baryshkov@linaro.org; Jordan Holt jordan.holt@sightlineapplications.com Cc: Paul Neuhardt paul.neuhardt@linaro.org; sightlineapplications@lists.linaro.org Subject: RE: [Sightlineapplications] Re: Project Status and Burn Report
Hi Dmitry,
When I run 'nproc --all', it reports 4 cpu cores. If I issue 'cpufreq-info', I only see 2. Is there a way to see all 4 cpu cores? It looks like the device tree is ok and I don't see any cpu errors in 'dmesg' (that I can tell are errors).
Thanks, Kim
-----Original Message----- From: Dmitry Baryshkov dmitry.baryshkov@linaro.org Sent: Wednesday, November 30, 2022 12:22 PM To: Kim Steiner kim.steiner@sightlineapplications.com; Jordan Holt jordan.holt@sightlineapplications.com Cc: Paul Neuhardt paul.neuhardt@linaro.org; sightlineapplications@lists.linaro.org Subject: Re: [Sightlineapplications] Re: Project Status and Burn Report
On 28/11/2022 22:19, Kim Steiner wrote:
Hi Dmitry,
With this patch, I built per your instructions. I got this error when I flashed boot.img: [180] sys r13 0x00000000 r14 0x00000000 [190] panic (caller 0x91637134): ASSERT FAILED at (platform/msm_shared/mmc_wrapper.c:190): !(data_len % block_size) [200] HALT: reboot into dload mo
So I change the --pagesize to 4096 and built the image again and it worked. It looks like the db820c should use 4096 per this page: https://github.com/96boards/documentation/wiki/Dragonboard-Boot-Imag e
Maybe there is a difference in the mkbootimg version. I have been using --pagesize 2048 here, see the attached build script that I'm using here.
If 4096 works for you, let's use it.
I cycled power 15 or so times and only had one failed boot with the above patch:
[ 2.794094] arm-smmu 1600000.iommu: (IDR0.CTTW overridden by FW configuration) [ 2.799742] arm-smmu 1600000.iommu: stream matching with 15 register groups [ 2.806937] arm-smmu 1600000.iommu: 12 context banks (0 stage-2 only) [ 2.814128] arm-smmu 1600000.iommu: Supported page sizes: 0x63315000 [ 2.820400] arm-smmu 1600000.iommu: Stage-1: 36-bit VA -> 36-bit IPA [ 2.884395] Bluetooth: hci0: Frame reassembly failed (-84)
Interesting. I'll run additional tests on my side.
Format: Log Type - Time(microsec) - Message - Optional Info Log Type: B - Since Boot(Power On Reset), D - Delta, S - Statistic S - QC_IMAGE_VERSION_STRING=BOOT.XF.1.0-00301 S - IMAGE_VARIANT_STRING=M8996LAB S - OEM_IMAGE_VERSION_STRING=crm-ubuntu68 S - Boot Interface: UFS S - Secure Boot: Off S - Boot Config @ 0x00076044 = 0x000001c9 S - JTAG ID @ 0x000760f4 = 0x4003e0e1 S - OEM ID @ 0x000760f8 = 0x00000000 S - Serial Number @ 0x00074138 = 0xbe8b8597 S - OEM Config Row 0 @ 0x00074188 = 0x0000000000000000
I'm not sure if this boot failure is the same as the 4.14 boot failure or not: https://discuss.96boards.org/t/boot-failure-crash-during-the-boot/59 31
No, it looks like a different kind of issue. The reported boot failures happened because of the cpufreq driver and the CPUs being not being properly powered.
Also, would it be possible for me to get your boot.img to compare with the one I built?
I have uploaded my image to: [1]
[1] https://fileserver.linaro.org/s/S27PALCCAN2xAXs
Thanks,
Kim
-----Original Message----- From: Dmitry Baryshkov dmitry.baryshkov@linaro.org Sent: Thursday, November 24, 2022 11:16 AM To: Kim Steiner kim.steiner@sightlineapplications.com; Jordan Holt jordan.holt@sightlineapplications.com Cc: Paul Neuhardt paul.neuhardt@linaro.org; sightlineapplications@lists.linaro.org Subject: Re: [Sightlineapplications] Re: Project Status and Burn Report
Hi Kim,
On 23/11/2022 23:57, Kim Steiner wrote:
Got it, thanks!
-----Original Message----- From: Dmitry Baryshkov dmitry.baryshkov@linaro.org Sent: Wednesday, November 23, 2022 12:04 PM To: Kim Steiner kim.steiner@sightlineapplications.com; Jordan Holt jordan.holt@sightlineapplications.com Cc: Paul Neuhardt paul.neuhardt@linaro.org; sightlineapplications@lists.linaro.org Subject: Re: [Sightlineapplications] Re: Project Status and Burn Report
Hi Kim,
On 23/11/2022 01:52, Kim Steiner wrote:
Hi Dmitry,
I ran our usual temp test with the latest and did not see the gpu failure. We are going to run more thorough of a test next week (full temp range). The glxgears application remains running if I run it from the serial debug window just as you said it would.
This sounds great!
I haven't noticed any improvement to the booting of the DB820c. About 50% of the time it stops at around 5seconds and prints this: [ 5.640755] xhci-hcd xhci-hcd.3.auto: new USB bus registered, assigned bus number 4
Format: Log Type - Time(microsec) - Message - Optional Info Log Type: B - Since Boot(Power On Reset), D - Delta, S - Statistic S - QC_IMAGE_VERSION_STRING=BOOT.XF.1.0-00301
Could you send me your boot.img? It is odd that you do not have the same booting issue as me. I would like to confirm that your image works for me.
Probably there is some misunderstanding. On 17th of November i wrote: "1) I was able to reproduce the failure during the cold boot. I'm investigating it now."
On 18th of November I again stated that "Meanwhile I will continue debugging the 'boot up reboot' issue. "
It took me a while, as I had to 'cool' the board between boots. With the attached patch I can no longer reproduce the reboots during the cold booting (first boot) the board. Could you please give it a try on your side too? It it works for you, I will push corresponding changes to the kernel branch.
Thanks, Kim
-----Original Message----- From: Dmitry Baryshkov dmitry.baryshkov@linaro.org Sent: Friday, November 18, 2022 12:42 PM To: Kim Steiner kim.steiner@sightlineapplications.com; Jordan Holt jordan.holt@sightlineapplications.com Cc: Paul Neuhardt paul.neuhardt@linaro.org; sightlineapplications@lists.linaro.org Subject: Re: [Sightlineapplications] Re: Project Status and Burn Report
Hi Kim,
First, I have updated the release branch. The changes included are:
- fix warming up code in the thermal subsystem to behave more aggressively
- add more warming up sources to the CPU and GPU devices
- add separate driver to use MX and CX power rails to warm up the
GPU
Could you please give this release a test? Meanwhile I will continue debugging the 'boot up reboot' issue.
On 17/11/2022 15:35, Kim Steiner wrote:
Hi Dmitry, In response to: 2) If the board is in normal conditions, then glxgears can run like forever. I didn't see any interruption (X11 can blank the screen which would nearly stop glxgears, but that's another story. I think you are running tests with X11 blanking disabled).
What I observe in the serial window is that I see the glxgears process with the 'ps' command and after a few minutes 'ps' shows no glxgears in the process list. Does this sound correct to you?
Yes, this sounds correct. I was just trying to understand if this happens in the cold room test or in the normal working case? Do you see any output in the terminal in such case?
If you need, you can run glxgears from the serial console by using 'XAUTHORITY=/home/linaro/.Xauthority DISPLAY=:0 glxgears'
I'm asking since I've executed several overnight tests which included running the glxgears for hours without any interruption.
Thanks, Kim
-----Original Message----- From: Dmitry Baryshkov dmitry.baryshkov@linaro.org Sent: Wednesday, November 16, 2022 2:15 PM To: Kim Steiner kim.steiner@sightlineapplications.com; Jordan Holt jordan.holt@sightlineapplications.com Cc: Paul Neuhardt paul.neuhardt@linaro.org; sightlineapplications@lists.linaro.org Subject: [Sightlineapplications] Re: Project Status and Burn Report
Hi Kim,
On 15/11/2022 20:15, Kim Steiner wrote: > Hi Dmitry, > > In doing more testing, I cannot get the DB820c to fully boot if I wait a while to turn on power: > > 1.) wait 1 minute with power disconnected. > 2.) turn on power. > 3.) Usually get one of these: > > Format: Log Type - Time(microsec) - Message - Optional Info Log Type: > B - Since Boot(Power On Reset), D - Delta, S - Statistic S - > QC_IMAGE_VERSION_STRING=BOOT.XF.1.0-00301 > S - IMAGE_VARIANT_STRING=M8996LAB S - > OEM_IMAGE_VERSION_STRING=crm-ubuntu68 > S - Boot Interface: UFS > S - Secure Boot: Off > S - Boot Config @ 0x00076044 = 0x000001c9 S - JTAG ID @ > 0x000760f4 = > 0x4003e0e1 S - OEM ID @ 0x000760f8 = 0x00000000
I did a lot of testing. Short summary:
I was able to reproduce the failure during the cold boot. I'm investigating it now.
If the board is in normal conditions, then glxgears can run like forever. I didn't see any interruption (X11 can blank the screen which would nearly stop glxgears, but that's another story. I think you are running tests with X11 blanking disabled).
I'm working on implementing the additional heating for the low temperatures using the PMIC regulators (MX and CX power domains).
> > Thanks, > Kim > -----Original Message----- > From: Kim Steiner kim.steiner@sightlineapplications.com > Sent: Tuesday, November 15, 2022 8:56 AM > To: Kim Steiner kim.steiner@sightlineapplications.com; Dmitry > Baryshkov dmitry.baryshkov@linaro.org; Jordan Holt > jordan.holt@sightlineapplications.com > Cc: Paul Neuhardt paul.neuhardt@linaro.org; > sightlineapplications@lists.linaro.org > Subject: RE: [Sightlineapplications] Re: Project Status and Burn > Report > > Hi Dmitry, > > The build instructions work for me so I will forward them to Penguin Solutions. > > Are you still working on the CPU/GPU messaging? It looks like boot time is quicker (~40s). Did you look into the glxgears application not running for very long? > > Thanks, > Kim > > -----Original Message----- > From: Kim Steiner kim.steiner@sightlineapplications.com > Sent: Tuesday, November 15, 2022 5:43 AM > To: Dmitry Baryshkov dmitry.baryshkov@linaro.org; Jordan Holt > jordan.holt@sightlineapplications.com > Cc: Paul Neuhardt paul.neuhardt@linaro.org; > sightlineapplications@lists.linaro.org > Subject: [Sightlineapplications] Re: Project Status and Burn > Report > > Hi Dmitry, > > I will try these instructions today! Thank you 😊 > > Kim > > -----Original Message----- > From: Dmitry Baryshkov dmitry.baryshkov@linaro.org > Sent: Tuesday, November 15, 2022 5:05 AM > To: Jordan Holt jordan.holt@sightlineapplications.com > Cc: Kim Steiner kim.steiner@sightlineapplications.com; Paul > Neuhardt paul.neuhardt@linaro.org; > sightlineapplications@lists.linaro.org > Subject: Re: [Sightlineapplications] Re: Project Status and Burn > Report > > Hello, Jordan, Kim, > > On Tue, 15 Nov 2022 at 00:39, Jordan Holt jordan.holt@sightlineapplications.com wrote: >> >> Hi Dmitry and Paul, >> Any chance to get these build instructions? We would like to help Penguin to move forward. > > I have updated the kernel branch with board-specific defconfig. The build instructions are attached to this email. > >> Thank you, >> Jordan >> >> Jordan Holt | CTO | SightLine Applications Inc. | Onboard Video >> Processing | he/him >> >> -----Original Message----- >> From: Kim Steiner kim.steiner@sightlineapplications.com >> Sent: Thursday, November 10, 2022 3:34 PM >> To: Dmitry Baryshkov dmitry.baryshkov@linaro.org; Paul >> Neuhardt paul.neuhardt@linaro.org >> Cc: sightlineapplications@lists.linaro.org >> Subject: [Sightlineapplications] Re: Project Status and Burn >> Report >> >> Hi Dmitry, >> >> I asked Penguin Solutions to start looking at creating a patch for us so they are trying to build/install the 5.15 kernel with the work you did so far. They would like some build instructions as well as us. They would also like to know the exact package name of the 11.2 compiler that you are using. I am using gcc-arm-11.2-2022.02-x86_64-aarch64-none-linux-gnu. Is this the same package you are using? >> >> Thanks, >> Kim >> -----Original Message----- >> From: Kim Steiner >> Sent: Friday, November 4, 2022 9:49 AM >> To: Dmitry Baryshkov dmitry.baryshkov@linaro.org; Paul >> Neuhardt paul.neuhardt@linaro.org >> Cc: sightlineapplications@lists.linaro.org >> Subject: RE: [Sightlineapplications] Re: Project Status and >> Burn Report >> >> Hi Dmitry, >> >> Yes, on Wednesday I tested your latest changes to the repo and had the issues mentioned to the group. I will wait for new updates to test. >> >> Thanks, >> Kim >> >> -----Original Message----- >> From: Dmitry Baryshkov dmitry.baryshkov@linaro.org >> Sent: Friday, November 4, 2022 8:03 AM >> To: Paul Neuhardt paul.neuhardt@linaro.org; Kim Steiner >> kim.steiner@sightlineapplications.com >> Cc: sightlineapplications@lists.linaro.org >> Subject: Re: [Sightlineapplications] Re: Project Status and >> Burn Report >> >> Hi Kim, >> >> On 04/11/2022 17:39, Paul Neuhardt wrote: >>> Kim, >>> >>> I had thought that the latest updates had been posted to your >>> repository. My apologies for the misunderstanding. I will ask >>> Dmitry to post his updates as soon as possible. >> >> The most recent code is in the repo. I'll update the repo as soon as I have any updates from my side, including the heating fine-tuning, etc. >> >>> >>> Paul >>> >>> On Fri, 4 Nov 2022 at 14:31, Kim Steiner >>> <kim.steiner@sightlineapplications.com >>> mailto:kim.steiner@sightlineapplications.com> wrote: >>> >>> Hi Paul,____ >>> >>> __ __ >>> >>> 1. In an email to Dmitry on 11/2/2022, I informed him that the gpu >>> temp issue may not be fixed. He stated “Thank you for the test. >>> I'll work on improving the warming of the CPU/GPU >>> then.”____ >>> >>> __ __ >>> >>> 2. We still have a very slow boot up time >>> (+2minutes).____ >>> >>> __ __ >>> >>> 3. I informed Dmitry that the glxgears application only runs for a >>> few minutes before stopping.____ >>> >>> __ __ >>> >>> __ __ >>> >>> As of Wednesday (with latest git repository changes), the above >>> issues existed and I do not plan on doing any more testing until I >>> see new changes or get something new to test.____ >>> >>> __ __ >>> >>> I will inform Penguin Solutions that you are close so they can work >>> on patches for us to apply to our custom board >>> build.____ >>> >>> __ __ >>> >>> I would also like to again request a clear set of build instructions >>> (changes to .config, compiler, initram…) so that I can build/test >>> correctly.____ >>> >>> __ __ >>> >>> Thanks,____ >>> >>> Kim____ >>> >>> __ __ >>> >>> *From:* Paul Neuhardt <paul.neuhardt@linaro.org >>> mailto:paul.neuhardt@linaro.org> >>> *Sent:* Friday, November 4, 2022 7:05 AM >>> *To:* sightlineapplications@lists.linaro.org >>> mailto:sightlineapplications@lists.linaro.org >>> *Subject:* [Sightlineapplications] Project Status and >>> Burn Report____ >>> >>> __ __ >>> >>> All,____ >>> >>> __ __ >>> >>> Dmitry has now delivered what he believes to be the final set of >>> updates to Sightline in support of this project. Sightline is in the >>> process of testing these deliverables and we expect that the results >>> will be at or close to what was expected. If there are still gaps, >>> we will of course work to close them.____ >>> >>> __ __ >>> >>> As of today, there are still 13.5 hours of Dmitry's time left >>> available under the most recent contract extension. This should >>> provide adequate time for any consultation needed and any >>> updates and/or final tuning necessary. I have attached a brief >>> report on that burn to this email. Steve, I will send you a more >>> detailed version that you can share inside Sightline as needed. >>> ____ >>> >>> __ __ >>> >>> Please let me know if you have any questions or concerns. >>> ____ >>> >>> __ __ >>> >>> While I will continue to be in touch as we close down the project, I >>> want to take this opportunity to thank everyone involved for all >>> their work. I am confident I speak for all of us at Linaro when I >>> say I hope we have the opportunity to work with Sightline >>> Applications again in the future.____ >>> >>> __ __ >>> >>> Kind regards,____ >>> >>> Paul____ >>> >>> -- ____ >>> >>> Linaro http://www.linaro.org/____ >>> >>> >>> >>> *Paul Neuhardt*|/Sr. Program Manager/____ >>> >>> T: +44 0771 377 8664 >>> paul.neuhardt@linaro.org mailto:glen.valante@linaro.org >>> IRC: pneuhardt____ >>> >>> __ __ >>> >>> >>> _______________________________________________ >>> Sightlineapplications mailing list -- >>> sightlineapplications@lists.linaro.org >>> To unsubscribe send an email to >>> sightlineapplications-leave@lists.linaro.org >> >> -- >> With best wishes >> Dmitry >> >> _______________________________________________ >> Sightlineapplications mailing list -- >> sightlineapplications@lists.linaro.org >> To unsubscribe send an email to >> sightlineapplications-leave@lists.linaro.org > > > > -- > With best wishes > Dmitry > _______________________________________________ > Sightlineapplications mailing list -- > sightlineapplications@lists.linaro.org > To unsubscribe send an email to > sightlineapplications-leave@lists.linaro.org
-- With best wishes Dmitry
Sightlineapplications mailing list -- sightlineapplications@lists.linaro.org To unsubscribe send an email to sightlineapplications-leave@lists.linaro.org
-- With best wishes Dmitry
-- With best wishes Dmitry
-- With best wishes Dmitry
-- With best wishes Dmitry
-- With best wishes Dmitry