On Wed, 21 Nov 2018 at 14:20, Ezequiel Garcia <ezequiel@xxxxxxxxxxxxx> wrote: > > On Tuesday, November 20, 2018 19:19 -03, Ulf Hansson <ulf.hansson@xxxxxxxxxx> wrote: > > > On 20 November 2018 at 19:21, Ezequiel Garcia <ezequiel@xxxxxxxxxxxxx> wrote: > > > As reported by Aaro, the JZ4740 MMC driver throws a warning > > > when the kernel is built without preemption (CONFIG_PREEMPT_NONE=y). > > > > > > [ 16.461094] jz4740-mmc 13450000.mmc: [jz4740_mmc_prepare_dma_data] invalid cookie: data->host_cookie 567 host->next_data.cookie 568 > > > [ 16.473120] jz4740-mmc 13450000.mmc: [jz4740_mmc_prepare_dma_data] invalid cookie: data->host_cookie 568 host->next_data.cookie 569 > > > [ 16.485144] jz4740-mmc 13450000.mmc: [jz4740_mmc_prepare_dma_data] invalid cookie: data->host_cookie 569 host->next_data.cookie 570 > > > [ 16.497170] jz4740-mmc 13450000.mmc: [jz4740_mmc_prepare_dma_data] invalid cookie: data->host_cookie 570 host->next_data.cookie 571 > > > [ 16.509194] jz4740-mmc 13450000.mmc: [jz4740_mmc_prepare_dma_data] invalid cookie: data->host_cookie 571 host->next_data.cookie 572 > > > [ 16.532421] jz4740-mmc 13450000.mmc: [jz4740_mmc_prepare_dma_data] invalid cookie: data->host_cookie 572 host->next_data.cookie 573 > > > [ 16.544594] jz4740-mmc 13450000.mmc: [jz4740_mmc_prepare_dma_data] invalid cookie: data->host_cookie 573 host->next_data.cookie 574 > > > [ 16.556621] jz4740-mmc 13450000.mmc: [jz4740_mmc_prepare_dma_data] invalid cookie: data->host_cookie 574 host->next_data.cookie 575 > > > [ 16.568638] jz4740-mmc 13450000.mmc: [jz4740_mmc_prepare_dma_data] invalid cookie: data->host_cookie 575 host->next_data.cookie 576 > > > [ 16.601092] jz4740-mmc 13450000.mmc: [jz4740_mmc_prepare_dma_data] invalid cookie: data->host_cookie 582 host->next_data.cookie 583 > > > > > > The problem seems to be related to how pre_req/post_req is implemented. > > > Currently, it seems the driver expects jz4740_mmc_prepare_dma_data() > > > to be called with monotonically increasing host_cookie values, > > > which is wrong. > > > > > > Moreover, the implementation is overly complicated, keeping track > > > of unneeded "next cookie" state. > > > > > > So, instead of attempting to fix the current pre_req/post_req > > > implementation, this commit refactors the driver, dropping > > > the state, following other drivers such as dw_mmc and sdhci. > > > > > > Cc: Paul Cercueil <paul@xxxxxxxxxxxxxxx> > > > Cc: Mathieu Malaterre <malat@xxxxxxxxxx> > > > Reported-by: Aaro Koskinen <aaro.koskinen@xxxxxx> > > > Signed-off-by: Ezequiel Garcia <ezequiel@xxxxxxxxxxxxx> > > > > This looks good to me, however because of the rather big changes I > > doubt this can be tagged for stable - or did you have a try to apply > > it for an old kernel? > > > > In either case, I can queue this up for next, which anyway seems like > > it should be sufficient. Right? > > > > Yes, that is enough for me. > > If anyone is interested in a fix for stable, it might go with > disabling pre_req/post_req altogether. Great! [...] Applied for next then, thanks! Kind regards Uffe
