On Sun, 13 Oct 2019 21:12:28 +0200 David Bauer <m...@david-bauer.net> wrote:
> Hello Michal, > > I've tested your patch on my Aerohive HiveAP-121 (patch to follow). > While it worked most of the times, in around 1 out of 5 boot > attempts, the driver will fail to probe: > > [ 3.885899] nand: device found, Manufacturer ID: 0xad, Chip ID: > 0xf1 [ 3.961915] nand: Hynix NAND 128MiB 3,3V 8-bit > [ 4.015027] nand: 128 MiB, SLC, erase size: 128 KiB, page size: > 2048, OOB size: 64 [ 4.105743] Scanning device for bad blocks > [ 4.654817] random: fast init done > [ 5.604831] ar934x-nand 1b000200.nand: read operation failed on > page 1664 [ 5.685985] nand_bbt: can't scan flash and build the > RAM-based BBT [ 5.759938] ar934x-nand 1b000200.nand: nand_scan > failed, err:-145 [ 5.832877] ar934x-nand: probe of 1b000200.nand > failed with error -145 > > I've noticed the reset sequence in ar934x_nfc_hw_init is missing > compared to the ar71xx implementation. Adding this back in, the NAND > probes flawlessly (sample size 10). > > See my staging tree for my alteration. If you are good with it, i > would squash it on your commit. [0] [1] > > Best wishes > David > > [0] > https://git.openwrt.org/?p=openwrt/staging/blocktrron.git;a=shortlog;h=refs/heads/ath79-nfc > [1] > https://git.openwrt.org/?p=openwrt/staging/blocktrron.git;a=commitdiff;h=3621acfec49a40fe115598367d2db92855b98d4f Hello David, Yes, please update my commit. Thank you for your feedback, review and tests. Indeed I (wrongly) assumed minimalistic approach and delegated ar71xx reset procedure to a simple DT property. I have only WNDR4300 to test the driver - this router interestingly has no problems whatsoever (I've booted it tens of times). ar934x-nand is my first DT driver port ever, and I thanks to your patch now I know more about implementing reset features :-) Again, thanks and best regards Michal _______________________________________________ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel
