On 4 February 2016 at 09:26, Ulf Hansson <ulf.hansson@xxxxxxxxxx> wrote: > On 4 February 2016 at 01:52, Nguyen Viet Dung <nv-dung@xxxxxxxxxxx> wrote: >> Hi Ulf >> >> I want to test this problem withc your patch. >> Please sent pacthset for me. >> I have checked below link but detail of patch haven't. > > Sorry for the delay. I haven't been able to rebase my patchset yet. > > Give me another couple of days, then I will post a new version. Hi again, I did a quick retry in re-basing my patches, but realized that it requires quite some work. With a quite small effort I did manage to improve the code in ->probe() and in ->set_ios(), related to runtime PM and clock management. I will post these patches after the weekend, but unfortunate I doubt these will fix the problem you reported here. Although, please help and test them. To fix the problem properly, one should probable re-write the locking and the state machine code in the driver, because it's rather messy and has likely problems. As a quick fix, one could validate pointers before accessing them in the sh_mmcif_irqt() function, because that's not done currently. So to be clear, this is all can help with currently. I wish I could help more, but I have limited bandwidth and no HW to test on. :-) Kind regards Uffe -- To unsubscribe from this list: send the line "unsubscribe linux-mmc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
