Re: [PATCH 06/10] scsi: add 'am53c974' driver

[Paolo Bonzini <pbonzini@redhat.com>]

Oops, hit send too early.  A small nit:

On 21/11/2014 10:27, Hannes Reinecke wrote:
> +static void pci_esp_dma_drain(struct esp *esp)
> +{
> +	u8 resid;
> +	int lim = 1000;
> +
> +
> +	if ((esp->sreg & ESP_STAT_PMASK) == ESP_DOP ||
> +	    (esp->sreg & ESP_STAT_PMASK) == ESP_DIP)
> +		/* Data-In or Data-Out, nothing to be done */
> +		return;
> +
> +	while (--lim > 0) {
> +		resid = pci_esp_read8(esp, ESP_FFLAGS) & ESP_FF_FBYTES;
> +		if (resid <= 1)
> +			break;

cpu_relax()?

> +	}
> +	if (resid > 1) {
> +		/* FIFO not cleared */
> +		shost_printk(KERN_INFO, esp->host,
> +			     "FIFO not cleared, %d bytes left\n",
> +			     resid);
> +	}
> +
> +	/*
> +	 * When there is a residual BCMPLT will never be set
> +	 * (obviously). But we still have to issue the BLAST
> +	 * command, otherwise the data will not being transferred.
> +	 * But we'll never know when the BLAST operation is
> +	 * finished. So check for some time and give up eventually.
> +	 */
> +	lim = 1000;
> +	pci_esp_write8(esp, ESP_DMA_CMD_DIR | ESP_DMA_CMD_BLAST, ESP_DMA_CMD);
> +	while (pci_esp_read8(esp, ESP_DMA_STATUS) & ESP_DMA_STAT_BCMPLT) {
> +		if (--lim == 0)
> +			break;

cpu_relax()?

Otherwise looks sane.

Paolo