[Sightlineapplications] Re: Project Status and Burn Report

Hi Dmitry, I sent a photo of the DB on 12/14 along with a build.zip file. Can you check if you received both (or either)?

Thanks, Kim

-----Original Message----- From: Dmitry Baryshkov dmitry.baryshkov@linaro.org Sent: Monday, December 19, 2022 4:51 PM To: Kim Steiner kim.steiner@sightlineapplications.com; Nicolas Dechesne nicolas.dechesne@linaro.org 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 12/12/2022 20:58, Kim Steiner wrote:

Thanks Nicolas.  This makes sense to use 4096 for pagesize then.

To answer Dmitry’s question, I see a sticker on the back of the DB820C that reads “H100 Rev D.”  I’m not sure if this is the revision you are looking for.

I wanted to check whether you were you able to capture a photo of the labels on your dragonboard?

Regarding the shipment of you hardware.

My shipment address is (note the 'ii' in the name):

Dmitrii Baryshkov Piispanristi 2 A 86 Espoo 02200 Finland

Phone: +358 469526397

Thanks,

Kim

*From:* Nicolas Dechesne nicolas.dechesne@linaro.org *Sent:* Sunday, December 11, 2022 12:35 PM *To:* Dmitry Baryshkov dmitry.baryshkov@linaro.org *Cc:* Kim Steiner kim.steiner@sightlineapplications.com; 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

     >
     > On 03/12/2022 00:45, Kim Steiner wrote:
     > > Hi Dmitry,
     > >
     > > Just letting you know that I tried flashing your boot.img with:
     > >
     > > 'fastboot flash boot boot.img'  and I see the same kernel
    panic message that I was seeing when I built with a pagesize=2048.
     > >
     > > [190]  sys r13 0x00000000 r14 0x00000000 [200] panic (caller
     > > 0x91637134): ASSERT FAILED at
    (platform/msm_shared/mmc_wrapper.c:190):
     > > !(data_len % block_size) [210] HALT: reboot into dload mo
     > > 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
     > >
     > > Are you able to flash the image and reboot?

    Unfortunately I was not able to reproduce the 2048 vs 4096 issue
    here.
    I tested several images including the one I provided you, they
    all use
    pagesize 2048 and all of them work flawlessly.
    It might be that there are differences between the board revisions.
    Could you please check, what revision is printed on the board?

My recollection is that we need to use 4096. It was a bug in the 820
firmware (ABL) where a specific 4K alignment was required. So 2048
might work depending on the data size. but it does not guarantee the
4K alignment/assert in the code which has:

uint32_t mmc_read(uint64_t data_addr, uint32_t *out, uint32_t data_len)
{
         uint32_t ret = 0;
         uint32_t block_size;
         uint32_t read_size = SDHCI_ADMA_MAX_TRANS_SZ;
         void *dev;
         uint8_t *sptr = (uint8_t *)out;

        dev = target_mmc_device();
         block_size = mmc_get_device_blocksize();

         ASSERT(!(data_addr % block_size));
         ASSERT(!(data_len % block_size));

We've always recommended to use 4096, as far as I remember
(block_size if 4K on UFS). 820 was the first platform with UFS, so I
think the only platform with this problem.

-- With best wishes Dmitry