Hi Brendan,

here is my review. Only minor stuff, no real show-stopper.

|| COMPILE_TEST?

+	help
+	  If you say yes to this option, support will be included for the
+	  Aspeed I2C controller.
+
+	  This driver can also be built as a module.  If so, the module
+	  will be called i2c-aspeed.
+
 config I2C_AT91
 	tristate "Atmel AT91 I2C Two-Wire interface (TWI)"
 	depends on ARCH_AT91
+struct aspeed_i2c_bus {
+	struct i2c_adapter		adap;
+	struct device			*dev;
+	void __iomem			*base;
+	/* Synchronizes I/O mem access to base. */
+	spinlock_t			lock;
+	struct completion		cmd_complete;
+	int				irq;

'irq' not really needed, I'd think.

+	unsigned long			parent_clk_frequency;
+	u32				bus_frequency;
+	/* Transaction state. */
+	enum aspeed_i2c_master_state	master_state;
+	struct i2c_msg			*msgs;
+	size_t				buf_index;
+	size_t				msgs_index;
+	size_t				msgs_count;
+	bool				send_stop;
+	int				cmd_err;
+	/* Protected only by i2c_lock_bus */
+	int				master_xfer_result;
+};

...

+static int aspeed_i2c_recover_bus(struct aspeed_i2c_bus *bus)
+{
+	unsigned long time_left, flags;
+	int ret = 0;
+	u32 command;
+
+	spin_lock_irqsave(&bus->lock, flags);
+	command = readl(bus->base + ASPEED_I2C_CMD_REG);
+
+	if (command & ASPEED_I2CD_SDA_LINE_STS) {
+		/* Bus is idle: no recovery needed. */
+		if (command & ASPEED_I2CD_SCL_LINE_STS)
+			goto out;
+		dev_dbg(bus->dev, "bus hung (state %x), attempting recovery\n",
+			command);
+
+		reinit_completion(&bus->cmd_complete);
+		writel(ASPEED_I2CD_M_STOP_CMD, bus->base + ASPEED_I2C_CMD_REG);
+		spin_unlock_irqrestore(&bus->lock, flags);
+
+		time_left = wait_for_completion_timeout(
+				&bus->cmd_complete, bus->adap.timeout);
+
+		spin_lock_irqsave(&bus->lock, flags);
+		if (time_left == 0)
+			goto reset_out;
+		else if (bus->cmd_err)
+			goto reset_out;
+		/* Recovery failed. */
+		else if (!(readl(bus->base + ASPEED_I2C_CMD_REG) &
+			   ASPEED_I2CD_SCL_LINE_STS))
+			goto reset_out;
+	/* Bus error. */
+	} else {
+		dev_dbg(bus->dev, "bus hung (state %x), attempting recovery\n",
+			command);

Same dbg message as in the condition? Move it out of the 'if'?

+
+		reinit_completion(&bus->cmd_complete);
+		writel(ASPEED_I2CD_BUS_RECOVER_CMD,
+		       bus->base + ASPEED_I2C_CMD_REG);

Out of interest: What does the RECOVER_CMD do?

+		spin_unlock_irqrestore(&bus->lock, flags);
+
+		time_left = wait_for_completion_timeout(
+				&bus->cmd_complete, bus->adap.timeout);
+
+		spin_lock_irqsave(&bus->lock, flags);
+		if (time_left == 0)
+			goto reset_out;
+		else if (bus->cmd_err)
+			goto reset_out;
+		/* Recovery failed. */
+		else if (!(readl(bus->base + ASPEED_I2C_CMD_REG) &
+			   ASPEED_I2CD_SDA_LINE_STS))
+			goto reset_out;
+	}
+
+out:
+	spin_unlock_irqrestore(&bus->lock, flags);
+
+	return ret;
+
+reset_out:
+	spin_unlock_irqrestore(&bus->lock, flags);
+
+	return aspeed_i2c_reset(bus);
+}

...

+	case ASPEED_I2C_MASTER_INACTIVE:
+		dev_err(bus->dev,
+			"master received interrupt 0x%08x, but is inactive",
+			irq_status);
+		bus->cmd_err = -EIO;
+		/* Do not STOP as we should be inactive. */
+		goto out_complete;
+	default:
+		WARN(1, "unknown master state\n");
+		bus->master_state = ASPEED_I2C_MASTER_INACTIVE;
+		bus->cmd_err = -EIO;
+		goto out_complete;
+	}
+error_and_stop:
+	bus->cmd_err = -EIO;
+	aspeed_i2c_do_stop(bus);
+	goto out_no_complete;
+out_complete:
+	bus->msgs = NULL;
+	if (bus->cmd_err)
+		bus->master_xfer_result = bus->cmd_err;
+	else
+		bus->master_xfer_result = bus->msgs_index + 1;
+	complete(&bus->cmd_complete);
+out_no_complete:
+	if (irq_status != status_ack)
+		dev_err(bus->dev,
+			"irq handled != irq. expected 0x%08x, but was 0x%08x\n",
+			irq_status, status_ack);
+	spin_unlock(&bus->lock);
+	return !!irq_status;
+}

You return in the interrupt handler -EIO always in case of errors. Can
you check Documentation/i2c/fault-codes and see if you can follow those
guidelines? Especially ENXIO on NACK and EAGAIN...

+
+static irqreturn_t aspeed_i2c_bus_irq(int irq, void *dev_id)
+{
+	struct aspeed_i2c_bus *bus = dev_id;
+
+	if (aspeed_i2c_master_irq(bus))
+		return IRQ_HANDLED;
+	else
+		return IRQ_NONE;

Ternary operator? Your choice, though...

And one more question: Why do you use adapter->algo_data instead of
i2c_{get|set}_adapdata?

Kind regards,

   Wolfram