[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-ide
Subject:    Re: [PATCH] libata: fix setting and checking of DMA state.
From:       Damien Le Moal <Damien.LeMoal () wdc ! com>
Date:       2021-08-18 23:30:48
Message-ID: DM6PR04MB7081AB7034F116CDFBA43890E7FF9 () DM6PR04MB7081 ! namprd04 ! prod ! outlook ! com
[Download RAW message or body]

On 2021/08/18 4:03, Reimar Döffinger wrote:
> Checking if DMA is enabled should be done via the
> ata_dma_enabled helper function.
> Update dma_mode initialization so that SATA devices
> are reported as DMA enabled.

Please drop the period at the end of the patch title

> ---
>  drivers/ata/libata-core.c | 4 ++--
>  include/linux/ata.h       | 1 +
>  2 files changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
> index 9934f6c465f4..a5fe20bb22d6 100644
> --- a/drivers/ata/libata-core.c
> +++ b/drivers/ata/libata-core.c
> @@ -2004,7 +2004,7 @@ unsigned int ata_read_log_page(struct ata_device *dev, u8 log,
>  
>  retry:
>  	ata_tf_init(dev, &tf);
> -	if (dev->dma_mode && ata_id_has_read_log_dma_ext(dev->id) &&
> +	if (ata_dma_enabled(dev) && ata_id_has_read_log_dma_ext(dev->id) &&

This change is OK I think. Note that there are many other places where
dev->dma_mode is testing directly. It may be good to check those too.

>  	    !(dev->horkage & ATA_HORKAGE_NO_DMA_LOG)) {
>  		tf.command = ATA_CMD_READ_LOG_DMA_EXT;
>  		tf.protocol = ATA_PROT_DMA;
> @@ -2824,7 +2824,7 @@ int ata_bus_probe(struct ata_port *ap)
>  		 * bus as we may be talking too fast.
>  		 */
>  		dev->pio_mode = XFER_PIO_0;
> -		dev->dma_mode = 0xff;
> +		dev->dma_mode = ap->flags & ATA_FLAG_SATA ? XFER_SATA : 0xff;

This however is not OK I think. You are setting a transfer speed without going
through the link speed negotiation.

>  
>  		/* If the controller has a pio mode setup function
>  		 * then use it to set the chipset to rights. Don't
> diff --git a/include/linux/ata.h b/include/linux/ata.h
> index 1b44f40c7700..7bb2c2acbc42 100644
> --- a/include/linux/ata.h
> +++ b/include/linux/ata.h
> @@ -374,6 +374,7 @@ enum {
>  
>  	/* SETFEATURES stuff */
>  	SETFEATURES_XFER	= 0x03,
> +	XFER_SATA               = 0x48,

Where does this come from ?

>  	XFER_UDMA_7		= 0x47,
>  	XFER_UDMA_6		= 0x46,
>  	XFER_UDMA_5		= 0x45,
> 


-- 
Damien Le Moal
Western Digital Research
=
[prev in list] [next in list] [prev in thread] [next in thread]