RE: [PATCH v3 5/7] i2c: aspeed: Add aspeed_set_slave_busy()

From: Ryan Chen <ryan_chen@aspeedtech.com>
Date: 2021-05-28 02:58:15
Also in: linux-arm-kernel, linux-aspeed, linux-i2c, lkml, openbmc

-----Original Message-----
From: Quan Nguyen <quan@os.amperecomputing.com>
Sent: Friday, May 28, 2021 8:53 AM
To: Ryan Chen <ryan_chen@aspeedtech.com>; Corey Minyard
[off-list ref]; Rob Herring [off-list ref]; Joel Stanley
[off-list ref]; Andrew Jeffery [off-list ref]; Brendan Higgins
[off-list ref]; Benjamin Herrenschmidt
[off-list ref]; Wolfram Sang [off-list ref]; Philipp Zabel
[off-list ref]; openipmi-developer@lists.sourceforge.net;
devicetree@vger.kernel.org; linux-arm-kernel@lists.infradead.org;
linux-aspeed@lists.ozlabs.org; linux-kernel@vger.kernel.org;
linux-i2c@vger.kernel.org
Cc: Open Source Submission <redacted>; Thang Q .
Nguyen [off-list ref]; Phong Vo
[off-list ref]; openbmc@lists.ozlabs.org
Subject: Re: [PATCH v3 5/7] i2c: aspeed: Add aspeed_set_slave_busy()

On 25/05/2021 17:30, Ryan Chen wrote:

quoted

-----Original Message-----
From: Quan Nguyen <quan@os.amperecomputing.com>
Sent: Monday, May 24, 2021 6:49 PM
To: Ryan Chen <ryan_chen@aspeedtech.com>; Corey Minyard
[off-list ref]; Rob Herring [off-list ref]; Joel Stanley
[off-list ref]; Andrew Jeffery [off-list ref]; Brendan Higgins
[off-list ref]; Benjamin Herrenschmidt
[off-list ref]; Wolfram Sang [off-list ref]; Philipp
Zabel [off-list ref];
openipmi-developer@lists.sourceforge.net;
devicetree@vger.kernel.org; linux-arm-kernel@lists.infradead.org;
linux-aspeed@lists.ozlabs.org; linux-kernel@vger.kernel.org;
linux-i2c@vger.kernel.org
Cc: Open Source Submission <redacted>; Thang Q .
Nguyen [off-list ref]; Phong Vo
[off-list ref]; openbmc@lists.ozlabs.org
Subject: Re: [PATCH v3 5/7] i2c: aspeed: Add aspeed_set_slave_busy()

On 24/05/2021 17:36, Ryan Chen wrote:

quoted

-----Original Message-----
From: Quan Nguyen <quan@os.amperecomputing.com>
Sent: Monday, May 24, 2021 6:20 PM
To: Ryan Chen <ryan_chen@aspeedtech.com>; Corey Minyard
[off-list ref]; Rob Herring [off-list ref]; Joel Stanley
[off-list ref]; Andrew Jeffery [off-list ref]; Brendan Higgins
[off-list ref]; Benjamin Herrenschmidt
[off-list ref]; Wolfram Sang [off-list ref]; Philipp
Zabel [off-list ref];
openipmi-developer@lists.sourceforge.net;
devicetree@vger.kernel.org; linux-arm-kernel@lists.infradead.org;
linux-aspeed@lists.ozlabs.org; linux-kernel@vger.kernel.org;
linux-i2c@vger.kernel.org
Cc: Open Source Submission <redacted>; Thang

Q .

quoted

Nguyen [off-list ref]; Phong Vo
[off-list ref]; openbmc@lists.ozlabs.org
Subject: Re: [PATCH v3 5/7] i2c: aspeed: Add
aspeed_set_slave_busy()

On 24/05/2021 17:06, Ryan Chen wrote:

quoted

-----Original Message-----
From: openbmc
[off-list ref] On

Behalf

quoted

Of Quan Nguyen
Sent: Wednesday, May 19, 2021 3:50 PM
To: Corey Minyard <redacted>; Rob Herring
[off-list ref]; Joel Stanley [off-list ref]; Andrew
Jeffery [off-list ref]; Brendan Higgins
[off-list ref]; Benjamin Herrenschmidt
[off-list ref]; Wolfram Sang [off-list ref];
Philipp Zabel [off-list ref];
openipmi-developer@lists.sourceforge.net;
devicetree@vger.kernel.org; linux-arm-kernel@lists.infradead.org;
linux-aspeed@lists.ozlabs.org; linux-kernel@vger.kernel.org;
linux-i2c@vger.kernel.org
Cc: Open Source Submission <redacted>;

Thang

quoted

Q .

quoted

Nguyen [off-list ref]; Phong Vo
[off-list ref]; openbmc@lists.ozlabs.org
Subject: [PATCH v3 5/7] i2c: aspeed: Add aspeed_set_slave_busy()

Slave i2c device on AST2500 received a lot of slave irq while it
is busy processing the response. To handle this case, adds and
exports
aspeed_set_slave_busy() for controller to temporary stop slave
irq while slave is handling the response, and re-enable them
again when the

response is ready.

quoted

Signed-off-by: Quan Nguyen <quan@os.amperecomputing.com>
---
v3:
     + First introduce in v3 [Quan]

    drivers/i2c/busses/i2c-aspeed.c | 20 ++++++++++++++++++++
    1 file changed, 20 insertions(+)

diff --git a/drivers/i2c/busses/i2c-aspeed.c

b/drivers/i2c/busses/i2c-aspeed.c index
b2e9c8f0ddf7..9926d04831a2
100644

--- a/drivers/i2c/busses/i2c-aspeed.c
+++ b/drivers/i2c/busses/i2c-aspeed.c

@@ -944,6 +944,26 @@ static int aspeed_i2c_init(struct

aspeed_i2c_bus *bus,
    	return 0;
    }

+#if IS_ENABLED(CONFIG_I2C_SLAVE) void
+aspeed_set_slave_busy(struct i2c_adapter *adap, bool busy) {
+	struct aspeed_i2c_bus *bus = i2c_get_adapdata(adap);
+	unsigned long current_mask, flags;
+
+	spin_lock_irqsave(&bus->lock, flags);
+
+	current_mask = readl(bus->base + ASPEED_I2C_INTR_CTRL_REG);

Hello
	Where the bus->base to be remap?

Hi Ryan,

In "[PATCH v3 6/7] ipmi: ssif_bmc: Add Aspeed SSIF BMC driver", the
->priv is retrieved by calling i2c_get_adapdata(client->adapter).
->And in
aspeed_set_ssif_bmc_status(), call the exported
aspeed_set_slave_busy() using ->priv pointer as code below.

Yes, I see the probe function " ssif_bmc->priv =

i2c_get_adapdata(client->adapter);" to get priv.

quoted

But my question I don’t see the bus->base address be assigned.

Hi Ryan,

In drivers/i2c/busses/i2c-aspeed.c:
struct aspeed_i2c_bus {
          struct i2c_adapter              adap;
          struct device                   *dev;
          void __iomem                    *base;
          struct reset_control            *rst;
          /* Synchronizes I/O mem access to base. */
          spinlock_t                      lock;

So when "struct aspeed_i2c_bus *bus = i2c_get_adapdata(adap);", the
bus->base should point to the base of the aspeed_i2c_bus, which is
already initialized by the aspeed i2c bus driver.

Do I miss something?

Hello Quan,
	After study. I think the ssif_bmc_aspeed.c assume the priv data is the

same struct.

quoted

	That is trick.
	Do we have a better for slave enable/disable call back to implement this?
	If add call back in struct i2c_algorithm; is it workable?

quoted

Hi Ryan,

I dont know which is better, ie: adding callback to struct i2c_algorithm or to
struct i2c_adapter.
I have tried to add generic callback to struct i2c_adapter as below and it
works.

Hello Quan,
	Thanks your feedback.
	Because, if we apply it in callback. It can unify it in ssif_bmc.c to do callback.
	Not have ssif_bmc_aspeed.c, ssif_bmc_xxx.c .....
	Because the priv struct is different in different i2c driver implement. 
Ryan

quoted hunk ↗ jump to hunk

diff --git a/include/linux/i2c.h b/include/linux/i2c.h index

4e7714c88f95..6e9abf2d6abb 100644

--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h

@@ -713,6 +713,10 @@ struct i2c_adapter {
  	const struct i2c_adapter_quirks *quirks;

  	struct irq_domain *host_notify_domain;
+#if IS_ENABLED(CONFIG_I2C_SLAVE)
+	int (*slave_enable)(struct i2c_adapter *adap);
+	int (*slave_disable)(struct i2c_adapter *adap); #endif
  };
  #define to_i2c_adapter(d) container_of(d, struct i2c_adapter, dev)

quoted

+extern void aspeed_set_slave_busy(struct i2c_adapter *adap, bool
+busy); static void aspeed_set_ssif_bmc_status(struct ssif_bmc_ctx
+*ssif_bmc,
unsigned int status)
+{
+	if (status & SSIF_BMC_BUSY)
+		aspeed_set_slave_busy((struct i2c_adapter *)ssif_bmc->priv,

true);

quoted

+	else if (status & SSIF_BMC_READY)
+		aspeed_set_slave_busy((struct i2c_adapter *)ssif_bmc->priv,
+false); }
+
+static int ssif_bmc_probe(struct i2c_client *client, const struct
i2c_device_id *id)
+{
+	struct ssif_bmc_ctx *ssif_bmc;
+
+	ssif_bmc = ssif_bmc_alloc(client, 0);
+	if (IS_ERR(ssif_bmc))
+		return PTR_ERR(ssif_bmc);
+
+	ssif_bmc->priv = i2c_get_adapdata(client->adapter);
+	ssif_bmc->set_ssif_bmc_status = aspeed_set_ssif_bmc_status;
+
+	return 0;
+}

- Quan

`h`	back out one level
`j`	next message in thread
`k`	previous message in thread
`l`	drill in
`Esc`	close help / fold thread tree
`?`	toggle this help