From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jaehoon Chung Subject: Re: [PATCH v2] mmc: dw_mmc: read all data in FIFO after Data transfer over interrupt in pio mode Date: Tue, 22 Jan 2013 11:35:12 +0900 Message-ID: <50FDFAE0.4050101@samsung.com> References: <001c01cdf7d2$e1385290$a3a8f7b0$%kim@samsung.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from mailout4.samsung.com ([203.254.224.34]:42197 "EHLO mailout4.samsung.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752447Ab3AVCfX (ORCPT ); Mon, 21 Jan 2013 21:35:23 -0500 Received: from epcpsbgm2.samsung.com (epcpsbgm2 [203.254.230.27]) by mailout4.samsung.com (Oracle Communications Messaging Server 7u4-24.01(7.0.4.24.0) 64bit (built Nov 17 2011)) with ESMTP id <0MH000GE79TJJL10@mailout4.samsung.com> for linux-mmc@vger.kernel.org; Tue, 22 Jan 2013 11:35:21 +0900 (KST) Received: from [10.90.51.55] by mmp2.samsung.com (Oracle Communications Messaging Server 7u4-24.01(7.0.4.24.0) 64bit (built Nov 17 2011)) with ESMTPA id <0MH0002W49UWR6H0@mmp2.samsung.com> for linux-mmc@vger.kernel.org; Tue, 22 Jan 2013 11:35:21 +0900 (KST) In-reply-to: <001c01cdf7d2$e1385290$a3a8f7b0$%kim@samsung.com> Sender: linux-mmc-owner@vger.kernel.org List-Id: linux-mmc@vger.kernel.org To: Kyoungil Kim Cc: linux-mmc@vger.kernel.org, 'Chris Ball' , 'Seungwon Jeon' Hi, i didn't test with this patch. But it makes sense. Just i have minor comment. Acked-by: Jaehoon Chung On 01/21/2013 09:28 PM, Kyoungil Kim wrote: > In dwc manual, the below contents are described. > "During end of packet, interrupt is not generated if threshold > programming is larger than any remaining data. It is responsibility > of host to read remaining bytes on seeing Data Transfer Done > interrupt" > We also have seen the data cannot be read fully > when "sg_miter->length" is less than FIFO size. > > Signed-off-by: Kyoungil Kim > Signed-off-by: Seungwon Jeon > --- > drivers/mmc/host/dw_mmc.c | 11 +++++++---- > 1 files changed, 7 insertions(+), 4 deletions(-) > > diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c > index 323c502..064c010 100644 > --- a/drivers/mmc/host/dw_mmc.c > +++ b/drivers/mmc/host/dw_mmc.c > @@ -1430,7 +1430,7 @@ static void dw_mci_pull_data(struct dw_mci *host, void *buf, int cnt) > host->pull_data(host, buf, cnt); > } > > -static void dw_mci_read_data_pio(struct dw_mci *host) > +static void dw_mci_read_data_pio(struct dw_mci *host, bool int_data_over) > { > struct sg_mapping_iter *sg_miter = &host->sg_miter; > void *buf; > @@ -1465,7 +1465,10 @@ static void dw_mci_read_data_pio(struct dw_mci *host) > sg_miter->consumed = offset; > status = mci_readl(host, MINTSTS); > mci_writel(host, RINTSTS, SDMMC_INT_RXDR); > - } while (status & SDMMC_INT_RXDR); /*if the RXDR is ready read again*/ > + /* if the RXDR is ready read again */ > + } while ((status & SDMMC_INT_RXDR) || > + (int_data_over && > + SDMMC_GET_FCNT(mci_readl(host, STATUS)))); If you can change int_data_over, we can use the one line. Best Regards, Jaehoon Chung > data->bytes_xfered += nbytes; > > if (!remain) { > @@ -1597,7 +1600,7 @@ static irqreturn_t dw_mci_interrupt(int irq, void *dev_id) > smp_wmb(); > if (host->dir_status == DW_MCI_RECV_STATUS) { > if (host->sg != NULL) > - dw_mci_read_data_pio(host); > + dw_mci_read_data_pio(host, 1); > } > set_bit(EVENT_DATA_COMPLETE, &host->pending_events); > tasklet_schedule(&host->tasklet); > @@ -1606,7 +1609,7 @@ static irqreturn_t dw_mci_interrupt(int irq, void *dev_id) > if (pending & SDMMC_INT_RXDR) { > mci_writel(host, RINTSTS, SDMMC_INT_RXDR); > if (host->dir_status == DW_MCI_RECV_STATUS && host->sg) > - dw_mci_read_data_pio(host); > + dw_mci_read_data_pio(host, 0); > } > > if (pending & SDMMC_INT_TXDR) { >