[PATCH v2 0/3] ipmi: kcs-bmc: Rework bindings to clean up DT warnings

[PATCH v2 0/3] ipmi: kcs-bmc: Rework bindings to clean up DT warnings

[Andrew Jeffery]

Hello,

This is a short series reworking the devicetree binding and driver for the
ASPEED BMC KCS devices. With the number of supported ASPEED BMC devicetrees the
changes enable removal of more than 100 lines of warning output from dtc.

v1 can be found here:

https://lore.kernel.org/lkml/cover.5630f63168ad5cddf02e9796106f8e086c196907.1575376664.git-series.andrew@aj.id.au/

v2 cleans up the commit message of 2/3 and changes the name of the property
governing the LPC IO address for the KCS devices.

Cheers,

Andrew

Andrew Jeffery (3):
  dt-bindings: ipmi: aspeed: Introduce a v2 binding for KCS
  ipmi: kcs: Finish configuring ASPEED KCS device before enable
  ipmi: kcs: aspeed: Implement v2 bindings

 Documentation/devicetree/bindings/ipmi/aspeed-kcs-bmc.txt |  20 +-
 drivers/char/ipmi/kcs_bmc_aspeed.c                        | 151 +++++--
 2 files changed, 139 insertions(+), 32 deletions(-)

base-commit: 937d6eefc716a9071f0e3bada19200de1bb9d048
-- 
git-series 0.9.1

[PATCH v2 1/3] dt-bindings: ipmi: aspeed: Introduce a v2 binding for KCS

[Andrew Jeffery]

The v2 binding utilises reg and renames some of the v1 properties.

Signed-off-by: Andrew Jeffery <andrew@aj.id.au>
---
v2: Rename slave-reg to aspeed,lpc-io-reg

Rob: After our discussion about the name of 'slave-reg' on v1 I've thought
about it some more and have landed on aspeed,lpc-io-reg. In v1 I argued that
the name should be generic and you suggested that if so it should go in a
generic binding document - I've thought about this some more and concluded that
it was hard to pin down exactly where it should be documented if it were
generic (the generic ASPEED LPC binding is one place, but that would suggest
that the property is still ASPEED-specific; maybe some discussion with
Nuvoton might give some insight).

Regardless, it turns out that the address specification is really
ASPEED-specific in this case: The KCS host interface in the LPC IO space
consists of a data and status register, but the slave controller infers the
address of the second from the address of the first and thus only the address
of the first can be programmed on the BMC-side. ASPEED supply documentation
that maps the LPC-side register layout for given LPC IO base addresses. I think
this is esoteric enough to warrant the aspeed prefix.

 Documentation/devicetree/bindings/ipmi/aspeed-kcs-bmc.txt | 20 +++++---
 1 file changed, 14 insertions(+), 6 deletions(-)

diff --git a/Documentation/devicetree/bindings/ipmi/aspeed-kcs-bmc.txt b/Documentation/devicetree/bindings/ipmi/aspeed-kcs-bmc.txt
index d98a9bf45d6c..193e71ca96b0 100644
--- a/Documentation/devicetree/bindings/ipmi/aspeed-kcs-bmc.txt
+++ b/Documentation/devicetree/bindings/ipmi/aspeed-kcs-bmc.txt
@@ -1,9 +1,10 @@
-* Aspeed KCS (Keyboard Controller Style) IPMI interface
+# Aspeed KCS (Keyboard Controller Style) IPMI interface
 
 The Aspeed SOCs (AST2400 and AST2500) are commonly used as BMCs
 (Baseboard Management Controllers) and the KCS interface can be
 used to perform in-band IPMI communication with their host.
 
+## v1
 Required properties:
 - compatible : should be one of
     "aspeed,ast2400-kcs-bmc"
@@ -12,14 +13,21 @@ Required properties:
 - kcs_chan : The LPC channel number in the controller
 - kcs_addr : The host CPU IO map address
 
+## v2
+Required properties:
+- compatible : should be one of
+    "aspeed,ast2400-kcs-bmc-v2"
+    "aspeed,ast2500-kcs-bmc-v2"
+- reg : The address and size of the IDR, ODR and STR registers
+- interrupts : interrupt generated by the controller
+- aspeed,lpc-io-reg : The host CPU LPC IO address for the device
 
 Example:
 
-    kcs3: kcs3@0 {
-        compatible = "aspeed,ast2500-kcs-bmc";
-        reg = <0x0 0x80>;
+    kcs3: kcs@24 {
+        compatible = "aspeed,ast2500-kcs-bmc-v2";
+        reg = <0x24 0x1>, <0x30 0x1>, <0x3c 0x1>;
+        aspeed,lpc-reg = <0xca2>;
         interrupts = <8>;
-        kcs_chan = <3>;
-        kcs_addr = <0xCA2>;
         status = "okay";
     };
-- 
git-series 0.9.1

[PATCH v2 2/3] ipmi: kcs: Finish configuring ASPEED KCS device before enable

[Andrew Jeffery]

The interrupts were configured after the channel was enabled. Configure
them beforehand so they will work.

Signed-off-by: Andrew Jeffery <andrew@aj.id.au>
Reviewed-by: Joel Stanley <joel@jms.id.au>
Reviewed-by: Haiyue Wang <haiyue.wang@linux.intel.com>
---
v2: Update the commit message based on Corey's suggestion

 drivers/char/ipmi/kcs_bmc_aspeed.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/char/ipmi/kcs_bmc_aspeed.c b/drivers/char/ipmi/kcs_bmc_aspeed.c
index 3c955946e647..e3dd09022589 100644
--- a/drivers/char/ipmi/kcs_bmc_aspeed.c
+++ b/drivers/char/ipmi/kcs_bmc_aspeed.c
@@ -268,13 +268,14 @@ static int aspeed_kcs_probe(struct platform_device *pdev)
 	kcs_bmc->io_inputb = aspeed_kcs_inb;
 	kcs_bmc->io_outputb = aspeed_kcs_outb;
 
+	rc = aspeed_kcs_config_irq(kcs_bmc, pdev);
+	if (rc)
+		return rc;
+
 	dev_set_drvdata(dev, kcs_bmc);
 
 	aspeed_kcs_set_address(kcs_bmc, addr);
 	aspeed_kcs_enable_channel(kcs_bmc, true);
-	rc = aspeed_kcs_config_irq(kcs_bmc, pdev);
-	if (rc)
-		return rc;
 
 	rc = misc_register(&kcs_bmc->miscdev);
 	if (rc) {
-- 
git-series 0.9.1

[PATCH v2 3/3] ipmi: kcs: aspeed: Implement v2 bindings

[Andrew Jeffery]

The v2 bindings allow us to extract the resources from the devicetree.
The table in the driver is retained to derive the channel index, which
removes the need for kcs_chan property from the v1 bindings. The v2
bindings allow us to reduce the number of warnings generated by the
existing devicetree nodes.

Signed-off-by: Andrew Jeffery <andrew@aj.id.au>
Reviewed-by: Joel Stanley <joel@jms.id.au>
Reviewed-by: Haiyue Wang <haiyue.wang@linux.intel.com>
---
v2: Use aspeed,lpc-io-reg based on the updated binding

 drivers/char/ipmi/kcs_bmc_aspeed.c | 144 +++++++++++++++++++++++++-----
 1 file changed, 121 insertions(+), 23 deletions(-)

diff --git a/drivers/char/ipmi/kcs_bmc_aspeed.c b/drivers/char/ipmi/kcs_bmc_aspeed.c
index e3dd09022589..9422d55a0476 100644
--- a/drivers/char/ipmi/kcs_bmc_aspeed.c
+++ b/drivers/char/ipmi/kcs_bmc_aspeed.c
@@ -12,6 +12,7 @@
 #include <linux/mfd/syscon.h>
 #include <linux/module.h>
 #include <linux/of.h>
+#include <linux/of_address.h>
 #include <linux/platform_device.h>
 #include <linux/poll.h>
 #include <linux/regmap.h>
@@ -233,38 +234,133 @@ static const struct kcs_ioreg ast_kcs_bmc_ioregs[KCS_CHANNEL_MAX] = {
 	{ .idr = LPC_IDR4, .odr = LPC_ODR4, .str = LPC_STR4 },
 };
 
-static int aspeed_kcs_probe(struct platform_device *pdev)
+static struct kcs_bmc *aspeed_kcs_probe_of_v1(struct platform_device *pdev)
 {
-	struct device *dev = &pdev->dev;
 	struct aspeed_kcs_bmc *priv;
-	struct kcs_bmc *kcs_bmc;
-	u32 chan, addr;
+	struct device_node *np;
+	struct kcs_bmc *kcs;
+	u32 channel;
+	u32 slave;
 	int rc;
 
-	rc = of_property_read_u32(dev->of_node, "kcs_chan", &chan);
-	if ((rc != 0) || (chan == 0 || chan > KCS_CHANNEL_MAX)) {
-		dev_err(dev, "no valid 'kcs_chan' configured\n");
-		return -ENODEV;
+	np = pdev->dev.of_node;
+
+	rc = of_property_read_u32(np, "kcs_chan", &channel);
+	if ((rc != 0) || (channel == 0 || channel > KCS_CHANNEL_MAX)) {
+		dev_err(&pdev->dev, "no valid 'kcs_chan' configured\n");
+		return ERR_PTR(-EINVAL);
 	}
 
-	rc = of_property_read_u32(dev->of_node, "kcs_addr", &addr);
+	kcs = kcs_bmc_alloc(&pdev->dev, sizeof(struct aspeed_kcs_bmc), channel);
+	if (!kcs)
+		return ERR_PTR(-ENOMEM);
+
+	priv = kcs_bmc_priv(kcs);
+	priv->map = syscon_node_to_regmap(pdev->dev.parent->of_node);
+	if (IS_ERR(priv->map)) {
+		dev_err(&pdev->dev, "Couldn't get regmap\n");
+		return ERR_PTR(-ENODEV);
+	}
+
+	rc = of_property_read_u32(np, "kcs_addr", &slave);
 	if (rc) {
-		dev_err(dev, "no valid 'kcs_addr' configured\n");
-		return -ENODEV;
+		dev_err(&pdev->dev, "no valid 'kcs_addr' configured\n");
+		return ERR_PTR(-EINVAL);
 	}
 
-	kcs_bmc = kcs_bmc_alloc(dev, sizeof(*priv), chan);
-	if (!kcs_bmc)
-		return -ENOMEM;
+	kcs->ioreg = ast_kcs_bmc_ioregs[channel - 1];
+	aspeed_kcs_set_address(kcs, slave);
+
+	return 0;
+}
+
+static int aspeed_kcs_calculate_channel(const struct kcs_ioreg *regs)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(ast_kcs_bmc_ioregs); i++) {
+		if (!memcmp(&ast_kcs_bmc_ioregs[i], regs, sizeof(*regs)))
+			return i + 1;
+	}
+
+	return -EINVAL;
+}
+
+static struct kcs_bmc *aspeed_kcs_probe_of_v2(struct platform_device *pdev)
+{
+	struct aspeed_kcs_bmc *priv;
+	struct device_node *np;
+	struct kcs_ioreg ioreg;
+	struct kcs_bmc *kcs;
+	const __be32 *reg;
+	int channel;
+	u32 slave;
+	int rc;
 
-	priv = kcs_bmc_priv(kcs_bmc);
-	priv->map = syscon_node_to_regmap(dev->parent->of_node);
+	np = pdev->dev.of_node;
+
+	/* Don't translate addresses, we want offsets for the regmaps */
+	reg = of_get_address(np, 0, NULL, NULL);
+	if (!reg)
+		return ERR_PTR(-EINVAL);
+	ioreg.idr = be32_to_cpup(reg);
+
+	reg = of_get_address(np, 1, NULL, NULL);
+	if (!reg)
+		return ERR_PTR(-EINVAL);
+	ioreg.odr = be32_to_cpup(reg);
+
+	reg = of_get_address(np, 2, NULL, NULL);
+	if (!reg)
+		return ERR_PTR(-EINVAL);
+	ioreg.str = be32_to_cpup(reg);
+
+	channel = aspeed_kcs_calculate_channel(&ioreg);
+	if (channel < 0)
+		return ERR_PTR(channel);
+
+	kcs = kcs_bmc_alloc(&pdev->dev, sizeof(struct aspeed_kcs_bmc), channel);
+	if (!kcs)
+		return ERR_PTR(-ENOMEM);
+
+	kcs->ioreg = ioreg;
+
+	priv = kcs_bmc_priv(kcs);
+	priv->map = syscon_node_to_regmap(pdev->dev.parent->of_node);
 	if (IS_ERR(priv->map)) {
-		dev_err(dev, "Couldn't get regmap\n");
-		return -ENODEV;
+		dev_err(&pdev->dev, "Couldn't get regmap\n");
+		return ERR_PTR(-ENODEV);
 	}
 
-	kcs_bmc->ioreg = ast_kcs_bmc_ioregs[chan - 1];
+	rc = of_property_read_u32(np, "aspeed,lpc-io-reg", &slave);
+	if (rc)
+		return ERR_PTR(rc);
+
+	aspeed_kcs_set_address(kcs, slave);
+
+	return kcs;
+}
+
+static int aspeed_kcs_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct kcs_bmc *kcs_bmc;
+	struct device_node *np;
+	int rc;
+
+	np = pdev->dev.of_node;
+	if (of_device_is_compatible(np, "aspeed,ast2400-kcs-bmc") ||
+			of_device_is_compatible(np, "aspeed,ast2500-kcs-bmc"))
+		kcs_bmc = aspeed_kcs_probe_of_v1(pdev);
+	else if (of_device_is_compatible(np, "aspeed,ast2400-kcs-bmc-v2") ||
+			of_device_is_compatible(np, "aspeed,ast2500-kcs-bmc-v2"))
+		kcs_bmc = aspeed_kcs_probe_of_v2(pdev);
+	else
+		return -EINVAL;
+
+	if (IS_ERR(kcs_bmc))
+		return PTR_ERR(kcs_bmc);
+
 	kcs_bmc->io_inputb = aspeed_kcs_inb;
 	kcs_bmc->io_outputb = aspeed_kcs_outb;
 
@@ -274,7 +370,6 @@ static int aspeed_kcs_probe(struct platform_device *pdev)
 
 	dev_set_drvdata(dev, kcs_bmc);
 
-	aspeed_kcs_set_address(kcs_bmc, addr);
 	aspeed_kcs_enable_channel(kcs_bmc, true);
 
 	rc = misc_register(&kcs_bmc->miscdev);
@@ -283,9 +378,10 @@ static int aspeed_kcs_probe(struct platform_device *pdev)
 		return rc;
 	}
 
-	pr_info("channel=%u addr=0x%x idr=0x%x odr=0x%x str=0x%x\n",
-		chan, addr,
-		kcs_bmc->ioreg.idr, kcs_bmc->ioreg.odr, kcs_bmc->ioreg.str);
+	dev_dbg(&pdev->dev,
+		"Probed KCS device %d (IDR=0x%x, ODR=0x%x, STR=0x%x)\n",
+		kcs_bmc->channel, kcs_bmc->ioreg.idr, kcs_bmc->ioreg.odr,
+		kcs_bmc->ioreg.str);
 
 	return 0;
 }
@@ -302,6 +398,8 @@ static int aspeed_kcs_remove(struct platform_device *pdev)
 static const struct of_device_id ast_kcs_bmc_match[] = {
 	{ .compatible = "aspeed,ast2400-kcs-bmc" },
 	{ .compatible = "aspeed,ast2500-kcs-bmc" },
+	{ .compatible = "aspeed,ast2400-kcs-bmc-v2" },
+	{ .compatible = "aspeed,ast2500-kcs-bmc-v2" },
 	{ }
 };
 MODULE_DEVICE_TABLE(of, ast_kcs_bmc_match);
-- 
git-series 0.9.1

Re: [PATCH v2 1/3] dt-bindings: ipmi: aspeed: Introduce a v2 binding for KCS

[Rob Herring]

On Mon, 16 Dec 2019 12:57:40 +1030, Andrew Jeffery wrote:
> The v2 binding utilises reg and renames some of the v1 properties.
> 
> Signed-off-by: Andrew Jeffery <andrew@aj.id.au>
> ---
> v2: Rename slave-reg to aspeed,lpc-io-reg
> 
> Rob: After our discussion about the name of 'slave-reg' on v1 I've thought
> about it some more and have landed on aspeed,lpc-io-reg. In v1 I argued that
> the name should be generic and you suggested that if so it should go in a
> generic binding document - I've thought about this some more and concluded that
> it was hard to pin down exactly where it should be documented if it were
> generic (the generic ASPEED LPC binding is one place, but that would suggest
> that the property is still ASPEED-specific; maybe some discussion with
> Nuvoton might give some insight).
> 
> Regardless, it turns out that the address specification is really
> ASPEED-specific in this case: The KCS host interface in the LPC IO space
> consists of a data and status register, but the slave controller infers the
> address of the second from the address of the first and thus only the address
> of the first can be programmed on the BMC-side. ASPEED supply documentation
> that maps the LPC-side register layout for given LPC IO base addresses. I think
> this is esoteric enough to warrant the aspeed prefix.
> 
>  Documentation/devicetree/bindings/ipmi/aspeed-kcs-bmc.txt | 20 +++++---
>  1 file changed, 14 insertions(+), 6 deletions(-)
> 

Reviewed-by: Rob Herring <robh@kernel.org>

Re: [PATCH v2 0/3] ipmi: kcs-bmc: Rework bindings to clean up DT warnings

[Andrew Jeffery]

On Mon, 16 Dec 2019, at 12:57, Andrew Jeffery wrote:
> Hello,
> 
> This is a short series reworking the devicetree binding and driver for the
> ASPEED BMC KCS devices. With the number of supported ASPEED BMC devicetrees the
> changes enable removal of more than 100 lines of warning output from dtc.
> 
> v1 can be found here:
> 
> https://lore.kernel.org/lkml/cover.5630f63168ad5cddf02e9796106f8e086c196907.1575376664.git-series.andrew@aj.id.au/
> 
> v2 cleans up the commit message of 2/3 and changes the name of the property
> governing the LPC IO address for the KCS devices.

Ping?

Re: [PATCH v2 0/3] ipmi: kcs-bmc: Rework bindings to clean up DT warnings

[Corey Minyard]

On Fri, Apr 03, 2020 at 02:20:21PM +1030, Andrew Jeffery wrote:
> 
> 
> On Mon, 16 Dec 2019, at 12:57, Andrew Jeffery wrote:
> > Hello,
> > 
> > This is a short series reworking the devicetree binding and driver for the
> > ASPEED BMC KCS devices. With the number of supported ASPEED BMC devicetrees the
> > changes enable removal of more than 100 lines of warning output from dtc.
> > 
> > v1 can be found here:
> > 
> > https://lore.kernel.org/lkml/cover.5630f63168ad5cddf02e9796106f8e086c196907.1575376664.git-series.andrew@aj.id.au/
> > 
> > v2 cleans up the commit message of 2/3 and changes the name of the property
> > governing the LPC IO address for the KCS devices.
> 
> Ping?

Sorry, I've been busy.  I've looked this over and it seems ok, and it's
in my next tree.

Thanks,

-corey

Re: [PATCH v2 0/3] ipmi: kcs-bmc: Rework bindings to clean up DT warnings

[Andrew Jeffery]

On Fri, 3 Apr 2020, at 23:38, Corey Minyard wrote:
> On Fri, Apr 03, 2020 at 02:20:21PM +1030, Andrew Jeffery wrote:
> > 
> > 
> > On Mon, 16 Dec 2019, at 12:57, Andrew Jeffery wrote:
> > > Hello,
> > > 
> > > This is a short series reworking the devicetree binding and driver for the
> > > ASPEED BMC KCS devices. With the number of supported ASPEED BMC devicetrees the
> > > changes enable removal of more than 100 lines of warning output from dtc.
> > > 
> > > v1 can be found here:
> > > 
> > > https://lore.kernel.org/lkml/cover.5630f63168ad5cddf02e9796106f8e086c196907.1575376664.git-series.andrew@aj.id.au/
> > > 
> > > v2 cleans up the commit message of 2/3 and changes the name of the property
> > > governing the LPC IO address for the KCS devices.
> > 
> > Ping?
> 
> Sorry, I've been busy.

No worries, so have I. I was surprised when I found it had been 4 months :)

>  I've looked this over and it seems ok, and it's
> in my next tree.

Thanks.

Andrew