linux-kernel - Re: [PATCH] mmc: sdhci: fix an issue of mixing different types

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20200119201243.GA15269@ubuntu-x2-xlarge-x86>
Date:   Sun, 19 Jan 2020 13:12:43 -0700
From:   Nathan Chancellor <natechancellor@...il.com>
To:     Chunyan Zhang <zhang.lyra@...il.com>
Cc:     Ulf Hansson <ulf.hansson@...aro.org>,
        Adrian Hunter <adrian.hunter@...el.com>,
        linux-mmc@...r.kernel.org, Faiz Abbas <faiz_abbas@...com>,
        Chunyan Zhang <zhang.chunyan@...aro.org>,
        linux-kernel@...r.kernel.org, kbuild test robot <lkp@...el.com>,
        clang-built-linux@...glegroups.com
Subject: Re: [PATCH] mmc: sdhci: fix an issue of mixing different types

On Sun, Jan 19, 2020 at 11:49:02AM +0800, Chunyan Zhang wrote:
> Fix an issue reported by sparse, since different type of parameter is
> used on calling dmaengine_prep_slave_sg().
> 
> Fixes: 36e1da441fec (mmc: sdhci: add support for using external DMA devices)
> Reported-by: kbuild test robot <lkp@...el.com>
> Signed-off-by: Chunyan Zhang <zhang.chunyan@...aro.org>
> ---
>  drivers/mmc/host/sdhci.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
> index 30b5a624b50e..4503009f993b 100644
> --- a/drivers/mmc/host/sdhci.c
> +++ b/drivers/mmc/host/sdhci.c
> @@ -1235,8 +1235,8 @@ static int sdhci_external_dma_setup(struct sdhci_host *host,
>  		return -EINVAL;
>  
>  	desc = dmaengine_prep_slave_sg(chan, data->sg, data->sg_len,
> -				       mmc_get_dma_dir(data),
> -				       DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
> +			(enum dma_transfer_direction) mmc_get_dma_dir(data),
> +			DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
>  	if (!desc)
>  		return -EINVAL;
>  
> -- 
> 2.20.1
> 

Clang also noticed this through its -Wenum-conversion. I don't really
like implicit conversions as I think it defeats the point of using the
enum. I had an explicit conversion locally which I attached below.
Regardless, this should resolve this issue (although I am not sure I
like the indentation shift.

Reviewed-by: Nathan Chancellor <natechancellor@...il.com>

diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
7index 30b5a624b50e..98fa25ba3756 100644
--- a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -1201,6 +1201,7 @@ static int sdhci_external_dma_setup(struct sdhci_host *host,
 				    struct mmc_command *cmd)
 {
 	int ret, i;
+	enum dma_transfer_direction dir;
 	struct dma_async_tx_descriptor *desc;
 	struct mmc_data *data = cmd->data;
 	struct dma_chan *chan;
@@ -1234,8 +1235,12 @@ static int sdhci_external_dma_setup(struct sdhci_host *host,
 	if (sg_cnt <= 0)
 		return -EINVAL;
 
-	desc = dmaengine_prep_slave_sg(chan, data->sg, data->sg_len,
-				       mmc_get_dma_dir(data),
+	if (mmc_get_dma_dir(data) == DMA_TO_DEVICE)
+		dir = DMA_MEM_TO_DEV;
+	else
+		dir = DMA_DEV_TO_MEM;
+
+	desc = dmaengine_prep_slave_sg(chan, data->sg, data->sg_len, dir,
 				       DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
 	if (!desc)
 		return -EINVAL;