Here is another weird observation; Adding a printf statement to end of board_init_f() in arch/arm/mach-sunxi/board.c causes the hang to almost entirely disappear. The board now boots about 99% of the time.
Any ideas on why printf affects anything ?? Brandon. On Wednesday, October 26, 2022 at 4:57:03 PM UTC+1 Cheo Fusi wrote: > Hi, > > Applied the patch. No change in behaviour. > > Even increasing the delay to half a second has no effect. > > Brandon. > > On Tuesday, October 25, 2022 at 2:32:07 PM UTC+1 huseyi...@gmail.com > wrote: > >> I also have the same problem, did not test out yet but, >> >> >> https://patchwork.ozlabs.org/project/uboot/patch/CACqvRUbw1wg7BL+3XWV4MX=7t9njn6oufc81...@mail.gmail.com/ >> >> <https://patchwork.ozlabs.org/project/uboot/patch/CACqvRUbw1wg7BL+3XWV4MX=7t9njn6oufc81wz7wfom-1hm...@mail.gmail.com/> >> >> seems related, i can also confirm the ESD relevance. When I touch some >> metal groud, issue gets worsened. >> >> Uart shows below on hang: >> U-Boot SPL 2022.10 (Oct 22 2022 - 18:54:33 +0200) >> DRAM: 1024 MiB >> >> 29 Eylül 2022 Perşembe tarihinde saat 16:02:34 UTC+2 itibarıyla >> fusibr...@gmail.com şunları yazdı: >> >>> *DRAM_CLK* seems to affect this. Reducing to say 432 lets the board >>> boot about 70% of the time. >>> >>> The hang seems to be coming from the *spl_load_image *function defined >>> in common/spl/spl.c, with the SPL sometimes failing to load the U-boot >>> image. >>> >>> Could this be an issue with MMC controller timings ?? Looking at the A33 >>> and R16 user manuals show the Physical Layer Spec for the former is Ver3.00 >>> that for the latter is Ver2.00. Maybe this explains why the boot on the R16 >>> never hangs. My current knowledge of MMC controllers is still ~nil to be >>> able to read the specs and comprehend the difference. >>> >>> Brandon. >>> >>> On Thursday, September 22, 2022 at 11:08:32 AM UTC+1 Cheo Fusi wrote: >>> >>>> Hi Simon, >>>> >>>> Did you finally solve this issue?? I have the same problem with the M2M >>>> variant that has no eMMC chip. The same image works well with the variant >>>> that includes the eMMC. >>>> >>>> Brandon >>>> On Tuesday, November 30, 2021 at 3:09:21 PM UTC+1 simon...@gmail.com >>>> wrote: >>>> >>>>> Hi, >>>>> >>>>> I'm trying to get a banana pi M2M working on a slightly more recent >>>>> version of the kernel than the 3.4 that's provided. I got some things >>>>> working but u-boot isn't reliably booting from the sd card. It sometimes >>>>> works but it mostly hangs after displaying the DRAM message (as shown >>>>> below) >>>>> >>>>> ``` >>>>> U-Boot SPL 2022.01-rc3 (Nov 30 2021 - 10:46:41 +0000) >>>>> DRAM: 512 MiB >>>>> >>>>> <pressed reset> >>>>> >>>>> U-Boot SPL 2022.01-rc3 (Nov 30 2021 - 10:46:41 +0000) >>>>> DRAM: 512 MiB >>>>> >>>>> <pressed reset> >>>>> >>>>> U-Boot SPL 2022.01-rc3 (Nov 30 2021 - 10:46:41 +0000) >>>>> DRAM: 512 MiB >>>>> >>>>> <pressed reset> >>>>> >>>>> U-Boot SPL 2022.01-rc3 (Nov 30 2021 - 10:46:41 +0000) >>>>> DRAM: 512 MiB >>>>> >>>>> <pressed reset> >>>>> >>>>> U-Boot SPL 2022.01-rc3 (Nov 30 2021 - 10:46:41 +0000) >>>>> DRAM: 512 MiB >>>>> >>>>> <pressed reset> >>>>> >>>>> U-Boot SPL 2022.01-rc3 (Nov 30 2021 - 10:46:41 +0000) >>>>> DRAM: 512 MiB >>>>> >>>>> <pressed reset> >>>>> >>>>> U-Boot SPL 2022.01-rc3 (Nov 30 2021 - 10:46:41 +0000) >>>>> DRAM: 512 MiB >>>>> Trying to boot from MMC1 >>>>> >>>>> >>>>> U-Boot 2022.01-rc3 (Nov 30 2021 - 10:46:41 +0000) Allwinner Technology >>>>> >>>>> CPU: Allwinner A33 (SUN8I 1667) >>>>> Model: BananaPi M2 Magic >>>>> DRAM: 512 MiB >>>>> WDT: Not starting watchdog@1c20ca0 >>>>> MMC: mmc@1c0f000: 0, mmc@1c10000: 1, mmc@1c11000: 2 >>>>> ``` >>>>> >>>>> Looking through the wiki https://linux-sunxi.org/U-Boot it looks like >>>>> it could be because of the device tree, but considering it's booting some >>>>> times I don't think this is the issue? It looks like SPL is always >>>>> working >>>>> but it's often failing to pass to u-boot proper. >>>>> >>>>> Any pointers on what could be wrong here? >>>>> >>>>> Has anyone got a bananapi m2m or similar working well with a recent >>>>> kernel? I managed to get SPI working fine (i might submit a patch) but so >>>>> far I haven't managed to get wifi working (a bit stuck on that) and I >>>>> haven't looked much beyond that. >>>>> >>>>> Best regards, >>>>> Simon >>>>> >>>> -- You received this message because you are subscribed to the Google Groups "linux-sunxi" group. To unsubscribe from this group and stop receiving emails from it, send an email to linux-sunxi+unsubscr...@googlegroups.com. To view this discussion on the web, visit https://groups.google.com/d/msgid/linux-sunxi/662f5da1-c879-4f78-902e-9b4616d22fa9n%40googlegroups.com.
