Skip to contentSkip to conversation

net: sfp: improve high power module implementation

PATCH Superseded Russell King (Oracle) 5 participants 24 messages Oct 19, 2022 – Oct 21, 2022

Subsystems (5)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore

Russell King (Oracle)

author maintainer
#0 lore raw

cover net: sfp: improve high power module implementation

Hi,

This series aims to improve the power level switching between standard level 1 and the higher power levels.

The first patch updates the DT binding documentation to include the minimum and default of 1W, which is the base level that every SFP cage must support. Hence, it makes sense to document this in the binding.

The second patch enforces a minimum of 1W when parsing the firmware description, and optimises the code for that case; there's no need to check for SFF8472 compliance since we will not need to touch the A2h registers.

Patch 3 validates that the module supports SFF-8472 rev 10.2 before checking for power level 2 - rev 10.2 is where support for power levels was introduced, so if the module doesn't support this revision, it doesn't support power levels. Setting the power level 2 declaration bit is likely to be spurious.

Patch 4 does the same for power level 3, except this was introduced in SFF-8472 rev 11.9. The revision code was never updated, so we use the rev 11.4 to signify this.

Patch 5 cleans up the code - rather than using BIT(0), we now use a properly named value for the power level select bit.

Patch 6 introduces a read-modify-write helper.

Patch 7 gets rid of the DM7052 hack (which sets a power level declaration bit but is not compatible with SFF-8472 rev 10.2, and the module does not implement the A2h I2C address.)

Series tested with my DM7052.

 Documentation/devicetree/bindings/net/sff,sfp.yaml |  2 +
 drivers/net/phy/sfp.c                              | 85 +++++++++++-----------
 include/linux/sfp.h                                |  2 +
 3 files changed, 48 insertions(+), 41 deletions(-)
Signature · 2 lines
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!
#1 lore raw

PATCH 1/7 dt-bindings: net: sff,sfp: update binding

Add a minimum and default for the maximum-power-milliwatt option; module power levels were originally up to 1W, so this is the default and the minimum power level we can have for a functional SFP cage.

---
 Documentation/devicetree/bindings/net/sff,sfp.yaml | 2 ++
 1 file changed, 2 insertions(+)
Documentation/devicetree/bindings/net/sff,sfp.yaml +2 -0
index 06c66ab81c01..20d30cccc95e 100644
@@ -23,6 +23,8 @@ title: Small Form Factor (SFF) Committee Small Form-factor Pluggable (SFP)
 
   maximum-power-milliwatt:
     maxItems: 1
+    minimum: 1000
+    default: 1000
     description:
       Maximum module power consumption Specifies the maximum power consumption
       allowable by a module in the slot, in milli-Watts. Presently, modules can
Signature · 1 line
2.30.2
maintainer
#2 lore raw

Re: [PATCH net-next 1/7] dt-bindings: net: sff,sfp: update binding

Quoted 9 lines · Russell King (Oracle)

Add a minimum and default for the maximum-power-milliwatt option; module power levels were originally up to 1W, so this is the default and the minimum power level we can have for a functional SFP cage.

---
 Documentation/devicetree/bindings/net/sff,sfp.yaml | 2 ++
 1 file changed, 2 insertions(+)

My bot found errors running 'make DT_CHECKER_FLAGS=-m dt_binding_check' on your patch (DT_CHECKER_FLAGS is new in v5.13):

yamllint warnings/errors:

dtschema/dtc warnings/errors:
/builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/net/sff,sfp.yaml: properties:maximum-power-milliwatt: 'minimum' should not be valid under {'enum': ['const', 'enum', 'exclusiveMaximum', 'exclusiveMinimum', 'minimum', 'maximum', 'multipleOf', 'pattern']}
hint: Scalar and array keywords cannot be mixed
from schema $id: http://devicetree.org/meta-schemas/keywords.yaml#

doc reference errors (make refcheckdocs):

See https://patchwork.ozlabs.org/patch/

This check can fail if there are any dependencies. The base for a patch series is generally the most recent rc1.

If you already ran 'make dt_binding_check' and didn't see the above error(s), then make sure 'yamllint' is installed and dt-schema is up to date:

pip3 install dtschema --upgrade

Please check and re-submit.

author maintainer
#3 lore raw

Re: [PATCH net-next 1/7] dt-bindings: net: sff,sfp: update binding

Quoted 14 lines · Rob Herring
Quoted 9 lines · Russell King (Oracle)

Add a minimum and default for the maximum-power-milliwatt option; module power levels were originally up to 1W, so this is the default and the minimum power level we can have for a functional SFP cage.

---
 Documentation/devicetree/bindings/net/sff,sfp.yaml | 2 ++
 1 file changed, 2 insertions(+)

My bot found errors running 'make DT_CHECKER_FLAGS=-m dt_binding_check' on your patch (DT_CHECKER_FLAGS is new in v5.13):

Quoted 6 lines

yamllint warnings/errors:

dtschema/dtc warnings/errors:
/builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/net/sff,sfp.yaml: properties:maximum-power-milliwatt: 'minimum' should not be valid under {'enum': ['const', 'enum', 'exclusiveMaximum', 'exclusiveMinimum', 'minimum', 'maximum', 'multipleOf', 'pattern']}
hint: Scalar and array keywords cannot be mixed
from schema $id: http://devicetree.org/meta-schemas/keywords.yaml#

I'm reading that error message and it means absolutely nothing to me. Please can you explain it (and also re-word it to be clearer)?

Thanks.

Signature · 2 lines
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!
maintainer
#4 lore raw

Re: [PATCH net-next 1/7] dt-bindings: net: sff,sfp: update binding

Quoted 17 lines · Russell King (Oracle)
Quoted 16 lines · Rob Herring
Quoted 9 lines · Russell King (Oracle)

Add a minimum and default for the maximum-power-milliwatt option; module power levels were originally up to 1W, so this is the default and the minimum power level we can have for a functional SFP cage.

---
 Documentation/devicetree/bindings/net/sff,sfp.yaml | 2 ++
 1 file changed, 2 insertions(+)

My bot found errors running 'make DT_CHECKER_FLAGS=-m dt_binding_check' on your patch (DT_CHECKER_FLAGS is new in v5.13):

yamllint warnings/errors:

Quoted 7 lines
Quoted 4 lines

dtschema/dtc warnings/errors:
/builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/net/sff,sfp.yaml: properties:maximum-power-milliwatt: 'minimum' should not be valid under {'enum': ['const', 'enum', 'exclusiveMaximum', 'exclusiveMinimum', 'minimum', 'maximum', 'multipleOf', 'pattern']}
hint: Scalar and array keywords cannot be mixed
from schema $id: http://devicetree.org/meta-schemas/keywords.yaml#

I'm reading that error message and it means absolutely nothing to me. Please can you explain it (and also re-word it to be clearer)?

'maxItems' is a constraint for arrays. 'maximum' is a constraint for scalar values. Mixing them does not make sense.

I have little control over the 1st line as that comes from jsonschema package. 'hint' is what I've added to explain things a bit more.

Rob

maintainer
#5 lore raw

Re: [PATCH net-next 1/7] dt-bindings: net: sff,sfp: update binding

Quoted 24 lines · Rob Herring
Quoted 22 lines · Russell King (Oracle)
Quoted 20 lines · Rob Herring
Quoted 9 lines · Russell King (Oracle)

Add a minimum and default for the maximum-power-milliwatt option; module power levels were originally up to 1W, so this is the default and the minimum power level we can have for a functional SFP cage.

---
 Documentation/devicetree/bindings/net/sff,sfp.yaml | 2 ++
 1 file changed, 2 insertions(+)

My bot found errors running 'make DT_CHECKER_FLAGS=-m dt_binding_check' on your patch (DT_CHECKER_FLAGS is new in v5.13):

yamllint warnings/errors:

dtschema/dtc warnings/errors:
/builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/net/sff,sfp.yaml: properties:maximum-power-milliwatt: 'minimum' should not be valid under {'enum': ['const', 'enum', 'exclusiveMaximum', 'exclusiveMinimum', 'minimum', 'maximum', 'multipleOf', 'pattern']}
hint: Scalar and array keywords cannot be mixed
from schema $id: http://devicetree.org/meta-schemas/keywords.yaml#

Quoted 5 lines
Quoted 2 lines

I'm reading that error message and it means absolutely nothing to me. Please can you explain it (and also re-word it to be clearer)?

'maxItems' is a constraint for arrays. 'maximum' is a constraint for scalar values. Mixing them does not make sense.

TBC, dropping 'maxItems' is what is needed here.

Rob

author maintainer
#6 lore raw

Re: [PATCH net-next 1/7] dt-bindings: net: sff,sfp: update binding

Quoted 25 lines · Rob Herring
Quoted 24 lines · Rob Herring
Quoted 22 lines · Russell King (Oracle)
Quoted 20 lines · Rob Herring
Quoted 9 lines · Russell King (Oracle)

Add a minimum and default for the maximum-power-milliwatt option; module power levels were originally up to 1W, so this is the default and the minimum power level we can have for a functional SFP cage.

---
 Documentation/devicetree/bindings/net/sff,sfp.yaml | 2 ++
 1 file changed, 2 insertions(+)

My bot found errors running 'make DT_CHECKER_FLAGS=-m dt_binding_check' on your patch (DT_CHECKER_FLAGS is new in v5.13):

yamllint warnings/errors:

dtschema/dtc warnings/errors:
/builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/net/sff,sfp.yaml: properties:maximum-power-milliwatt: 'minimum' should not be valid under {'enum': ['const', 'enum', 'exclusiveMaximum', 'exclusiveMinimum', 'minimum', 'maximum', 'multipleOf', 'pattern']}
hint: Scalar and array keywords cannot be mixed
from schema $id: http://devicetree.org/meta-schemas/keywords.yaml#

Quoted 7 lines
Quoted 5 lines
Quoted 2 lines

I'm reading that error message and it means absolutely nothing to me. Please can you explain it (and also re-word it to be clearer)?

'maxItems' is a constraint for arrays. 'maximum' is a constraint for scalar values. Mixing them does not make sense.

TBC, dropping 'maxItems' is what is needed here.

So how does this work?

maxItems: 1

tells it that there should be an array of one property, which is at the DT level fundamentally the same as a scalar property.

minimum:
default:
maximum:

tells it that this is a scalar property, so there should be exactly one item or the property should not be mentioned?

Signature · 2 lines
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!
maintainer
#7 lore raw

Re: [PATCH net-next 1/7] dt-bindings: net: sff,sfp: update binding

Quoted 30 lines · Russell King (Oracle)
Quoted 28 lines · Rob Herring
Quoted 27 lines · Rob Herring
Quoted 24 lines · Russell King (Oracle)
Quoted 14 lines · Rob Herring
Quoted 9 lines · Russell King (Oracle)

Add a minimum and default for the maximum-power-milliwatt option; module power levels were originally up to 1W, so this is the default and the minimum power level we can have for a functional SFP cage.

---
 Documentation/devicetree/bindings/net/sff,sfp.yaml | 2 ++
 1 file changed, 2 insertions(+)

My bot found errors running 'make DT_CHECKER_FLAGS=-m dt_binding_check' on your patch (DT_CHECKER_FLAGS is new in v5.13):

Quoted 6 lines

yamllint warnings/errors:

dtschema/dtc warnings/errors:
/builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/net/sff,sfp.yaml: properties:maximum-power-milliwatt: 'minimum' should not be valid under {'enum': ['const', 'enum', 'exclusiveMaximum', 'exclusiveMinimum', 'minimum', 'maximum', 'multipleOf', 'pattern']}
hint: Scalar and array keywords cannot be mixed
from schema $id: http://devicetree.org/meta-schemas/keywords.yaml#

I'm reading that error message and it means absolutely nothing to me. Please can you explain it (and also re-word it to be clearer)?

Quoted 6 lines
Quoted 4 lines
Quoted 2 lines

'maxItems' is a constraint for arrays. 'maximum' is a constraint for scalar values. Mixing them does not make sense.

TBC, dropping 'maxItems' is what is needed here.

So how does this work?

Do you really want to know? ;)

Quoted 2 lines

maxItems: 1

json-schema happily ignores any keywords that it doesn't understand or don't make sense for a specific context. The DT meta-schema tries to prevent that.

Quoted 2 lines

tells it that there should be an array of one property, which is at the DT level fundamentally the same as a scalar property.

Yes, it is true that the YAML encoded DT and (currently) the internal encoding used by the tools encode everything as matrices simply because dtc doing the YAML encoding doesn't know the types beyond what DTS source level provides, so everything has to be the same encoding. Now we use the type information in the schemas to decode the DTBs directly and don't have that limitation. Once I remove the YAML encoding, we can stop encoding everything as a matrix and having to fixup the schemas from scalar -> array -> matrix.

Quoted 6 lines

minimum:
default:
maximum:

tells it that this is a scalar property, so there should be exactly one item or the property should not be mentioned?

Not sure I follow the question. As the property is defined as a scalar, it only needs scalar keywords. Internally, the schema gets expanded to:

prop:
minItems: 1
maxItems: 1
items:
- maxItems: 1
minItems: 1
items:
- maximum: ???
minimum: ???
default: ???

This is what processed-schemas.json will contain if you just have the scalar keywords.

It's a bit more messy now with the unit suffixes as initially they were all scalars, but over time we've had to allow for arrays. So it's really they default to scalars unless you need an array in which you can define:

prop:
maxItems: 2
items:

    maximum: ???

Could you do 'maxItems: 1' here? Yes, that would be a valid schema, but IIRC we'll still complain because it is redundant.

Rob

author maintainer
#8 lore raw

Re: [PATCH net-next 1/7] dt-bindings: net: sff,sfp: update binding

In reply to Rob Herring
Quoted 23 lines · Rob Herring
Quoted 22 lines · Russell King (Oracle)
Quoted 20 lines · Rob Herring
Quoted 9 lines · Russell King (Oracle)

Add a minimum and default for the maximum-power-milliwatt option; module power levels were originally up to 1W, so this is the default and the minimum power level we can have for a functional SFP cage.

---
 Documentation/devicetree/bindings/net/sff,sfp.yaml | 2 ++
 1 file changed, 2 insertions(+)

My bot found errors running 'make DT_CHECKER_FLAGS=-m dt_binding_check' on your patch (DT_CHECKER_FLAGS is new in v5.13):

yamllint warnings/errors:

dtschema/dtc warnings/errors:
/builds/robherring/dt-review-ci/linux/Documentation/devicetree/bindings/net/sff,sfp.yaml: properties:maximum-power-milliwatt: 'minimum' should not be valid under {'enum': ['const', 'enum', 'exclusiveMaximum', 'exclusiveMinimum', 'minimum', 'maximum', 'multipleOf', 'pattern']}
hint: Scalar and array keywords cannot be mixed
from schema $id: http://devicetree.org/meta-schemas/keywords.yaml#

Quoted 8 lines
Quoted 2 lines

I'm reading that error message and it means absolutely nothing to me. Please can you explain it (and also re-word it to be clearer)?

'maxItems' is a constraint for arrays. 'maximum' is a constraint for scalar values. Mixing them does not make sense.

I have little control over the 1st line as that comes from jsonschema package. 'hint' is what I've added to explain things a bit more.

Given that maximum-power-milliwatt is a single value and has never been an array, it seems then that the original conversion to yaml was wrong. What should it have been?

(I'm clueless what the difference is between an array and scalar in this yaml stuff.)

Signature · 2 lines
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!
#9 lore raw

PATCH 2/7 net: sfp: check firmware provided max power

In reply to Russell King (Oracle)

Check that the firmware provided maximum power is at least 1W, which iis the minimum power level for any SFP module.

Now that we enforce the minimum of 1W, we can exit early from sfp_module_parse_power() if the module power is 1W or less.

---
 drivers/net/phy/sfp.c | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)
drivers/net/phy/sfp.c +11 -7
index 40c9a64c5e30..f7ad4d5d9041 100644
@@ -1766,6 +1766,12 @@ static int sfp_module_parse_power(struct sfp *sfp)
 	if (sfp->id.ext.options & cpu_to_be16(SFP_OPTIONS_HIGH_POWER_LEVEL))
 		power_mW = 2000;
 
+	/* Power level 1 modules (max. 1W) are always supported. */
+	if (power_mW <= 1000) {
+		sfp->module_power_mW = power_mW;
+		return 0;
+	}
+
 	supports_a2 = sfp->id.ext.sff8472_compliance !=
 				SFP_SFF8472_COMPLIANCE_NONE ||
 		      sfp->id.ext.diagmon & SFP_DIAGMON_DDM;
@@ -1789,12 +1795,6 @@ static int sfp_module_parse_power(struct sfp *sfp)
 		}
 	}
 
-	if (power_mW <= 1000) {
-		/* Modules below 1W do not require a power change sequence */
-		sfp->module_power_mW = power_mW;
-		return 0;
-	}
-
 	if (!supports_a2) {
 		/* The module power level is below the host maximum and the
 		 * module appears not to implement bus address 0xa2, so assume
@@ -2729,8 +2729,12 @@ static int sfp_probe(struct platform_device *pdev)
 
 	device_property_read_u32(&pdev->dev, "maximum-power-milliwatt",
 				 &sfp->max_power_mW);
-	if (!sfp->max_power_mW)
+	if (sfp->max_power_mW < 1000) {
+		if (sfp->max_power_mW)
+			dev_warn(sfp->dev,
+				 "Firmware bug: host maximum power should be at least 1W\n");
 		sfp->max_power_mW = 1000;
+	}
 
 	dev_info(sfp->dev, "Host maximum power %u.%uW\n",
 		 sfp->max_power_mW / 1000, (sfp->max_power_mW / 100) % 10);
Signature · 1 line
2.30.2
maintainer Reviewed-by on v1 2/7
#10 lore raw

Re: [PATCH net-next 2/7] net: sfp: check firmware provided max power

Quoted 2 lines · Russell King (Oracle)

Check that the firmware provided maximum power is at least 1W, which iis the minimum power level for any SFP module.

is

Quoted 5 lines

Now that we enforce the minimum of 1W, we can exit early from sfp_module_parse_power() if the module power is 1W or less.

Andrew

#11 lore raw

PATCH 3/7 net: sfp: ignore power level 2 prior to SFF-8472 Rev 10.2

In reply to Russell King (Oracle)

Power level 2 was introduced by SFF-8472 revision 10.2. Ignore the power declaration bit for modules that are not compliant with at least this revision.

This should remove any spurious indication of 1.5W modules.

---
 drivers/net/phy/sfp.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/phy/sfp.c +2 -1
index f7ad4d5d9041..a7635b02524a 100644
@@ -1761,7 +1761,8 @@ static int sfp_module_parse_power(struct sfp *sfp)
 	u32 power_mW = 1000;
 	bool supports_a2;
 
-	if (sfp->id.ext.options & cpu_to_be16(SFP_OPTIONS_POWER_DECL))
+	if (sfp->id.ext.sff8472_compliance >= SFP_SFF8472_COMPLIANCE_REV10_2 &&
+	    sfp->id.ext.options & cpu_to_be16(SFP_OPTIONS_POWER_DECL))
 		power_mW = 1500;
 	if (sfp->id.ext.options & cpu_to_be16(SFP_OPTIONS_HIGH_POWER_LEVEL))
 		power_mW = 2000;
Signature · 1 line
2.30.2
maintainer Reviewed-by on v1 3/7
#12 lore raw

Re: [PATCH net-next 3/7] net: sfp: ignore power level 2 prior to SFF-8472 Rev 10.2

Quoted 7 lines · Russell King (Oracle)

Power level 2 was introduced by SFF-8472 revision 10.2. Ignore the power declaration bit for modules that are not compliant with at least this revision.

This should remove any spurious indication of 1.5W modules.

Andrew

#13 lore raw

PATCH 4/7 net: sfp: ignore power level 3 prior to SFF-8472 Rev 11.4

In reply to Russell King (Oracle)

Power level 3 was included in SFF-8472 revision 11.9, but this does not have a compliance code. Use revision 11.4 as the minimum compliance level instead.

This should avoid any spurious indication of 2W modules.

---
 drivers/net/phy/sfp.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/net/phy/sfp.c +3 -1
index a7635b02524a..af676e28ba6a 100644
@@ -1764,7 +1764,9 @@ static int sfp_module_parse_power(struct sfp *sfp)
 	if (sfp->id.ext.sff8472_compliance >= SFP_SFF8472_COMPLIANCE_REV10_2 &&
 	    sfp->id.ext.options & cpu_to_be16(SFP_OPTIONS_POWER_DECL))
 		power_mW = 1500;
-	if (sfp->id.ext.options & cpu_to_be16(SFP_OPTIONS_HIGH_POWER_LEVEL))
+	/* Added in Rev 11.9, but there is no compliance code for this */
+	if (sfp->id.ext.sff8472_compliance >= SFP_SFF8472_COMPLIANCE_REV11_4 &&
+	    sfp->id.ext.options & cpu_to_be16(SFP_OPTIONS_HIGH_POWER_LEVEL))
 		power_mW = 2000;
 
 	/* Power level 1 modules (max. 1W) are always supported. */
Signature · 1 line
2.30.2
maintainer Reviewed-by on v1 4/7
#14 lore raw

Re: [PATCH net-next 4/7] net: sfp: ignore power level 3 prior to SFF-8472 Rev 11.4

Quoted 7 lines · Russell King (Oracle)

Power level 3 was included in SFF-8472 revision 11.9, but this does not have a compliance code. Use revision 11.4 as the minimum compliance level instead.

This should avoid any spurious indication of 2W modules.

Andrew

#15 lore raw

PATCH 5/7 net: sfp: provide a definition for the power level select bit

In reply to Russell King (Oracle)

Provide a named definition for the power level select bit in the extended status register, rather than using BIT(0) in the code.

---
 drivers/net/phy/sfp.c | 6 +++---
 include/linux/sfp.h   | 2 ++
 2 files changed, 5 insertions(+), 3 deletions(-)
drivers/net/phy/sfp.c +3 -3
index af676e28ba6a..16bce0ea68d9 100644
@@ -1837,13 +1837,13 @@ static int sfp_sm_mod_hpower(struct sfp *sfp, bool enable)
 	 * all bytes 0xff) at 0x51 but does not accept writes.  In any case,
 	 * if the bit is already set, we're already in high power mode.
 	 */
-	if (!!(val & BIT(0)) == enable)
+	if (!!(val & SFP_EXT_STATUS_PWRLVL_SELECT) == enable)
 		return 0;
 
 	if (enable)
-		val |= BIT(0);
+		val |= SFP_EXT_STATUS_PWRLVL_SELECT;
 	else
-		val &= ~BIT(0);
+		val &= ~SFP_EXT_STATUS_PWRLVL_SELECT;
 
 	err = sfp_write(sfp, true, SFP_EXT_STATUS, &val, sizeof(val));
 	if (err != sizeof(val)) {
include/linux/sfp.h +2 -0
index d1f343853b6c..01ae9f1dd2ad 100644
@@ -489,6 +489,8 @@ enum {
 	SFP_WARN1_RXPWR_LOW		= BIT(6),
 
 	SFP_EXT_STATUS			= 0x76,
+	SFP_EXT_STATUS_PWRLVL_SELECT	= BIT(0),
+
 	SFP_VSL				= 0x78,
 	SFP_PAGE			= 0x7f,
 };
Signature · 1 line
2.30.2
#17 lore raw

PATCH 6/7 net: sfp: add sfp_modify_u8() helper

In reply to Russell King (Oracle)

Add a helper to modify bits in a single byte in memory space, and use it when updating the soft tx-disable flag in the module.

---
 drivers/net/phy/sfp.c | 31 ++++++++++++++++++++++---------
 1 file changed, 22 insertions(+), 9 deletions(-)
drivers/net/phy/sfp.c +22 -9
index 16bce0ea68d9..921bbedd9b22 100644
@@ -608,6 +608,22 @@ static int sfp_write(struct sfp *sfp, bool a2, u8 addr, void *buf, size_t len)
 	return sfp->write(sfp, a2, addr, buf, len);
 }
 
+static int sfp_modify_u8(struct sfp *sfp, bool a2, u8 addr, u8 mask, u8 val)
+{
+	int ret;
+	u8 old, v;
+
+	ret = sfp_read(sfp, a2, addr, &old, sizeof(old));
+	if (ret != sizeof(old))
+		return ret;
+
+	v = (old & ~mask) | (val & mask);
+	if (v == old)
+		return sizeof(v);
+
+	return sfp_write(sfp, a2, addr, &v, sizeof(v));
+}
+
 static unsigned int sfp_soft_get_state(struct sfp *sfp)
 {
 	unsigned int state = 0;
@@ -633,17 +649,14 @@ static unsigned int sfp_soft_get_state(struct sfp *sfp)
 
 static void sfp_soft_set_state(struct sfp *sfp, unsigned int state)
 {
-	u8 status;
+	u8 mask = SFP_STATUS_TX_DISABLE_FORCE;
+	u8 val = 0;
 
-	if (sfp_read(sfp, true, SFP_STATUS, &status, sizeof(status)) ==
-		     sizeof(status)) {
-		if (state & SFP_F_TX_DISABLE)
-			status |= SFP_STATUS_TX_DISABLE_FORCE;
-		else
-			status &= ~SFP_STATUS_TX_DISABLE_FORCE;
+	if (state & SFP_F_TX_DISABLE)
+		val |= SFP_STATUS_TX_DISABLE_FORCE;
 
-		sfp_write(sfp, true, SFP_STATUS, &status, sizeof(status));
-	}
+
+	sfp_modify_u8(sfp, true, SFP_STATUS, mask, val);
 }
 
 static void sfp_soft_start_poll(struct sfp *sfp)
Signature · 1 line
2.30.2
#19 lore raw

PATCH 7/7 net: sfp: get rid of DM7052 hack when enabling high power

In reply to Russell King (Oracle)

Since we no longer mis-detect high-power mode with the DM7052 module, we no longer need the hack in sfp_module_enable_high_power(), and can now switch this to use sfp_modify_u8().

---
 drivers/net/phy/sfp.c | 29 ++++++-----------------------
 1 file changed, 6 insertions(+), 23 deletions(-)
drivers/net/phy/sfp.c +6 -23
index 921bbedd9b22..39fd1811375c 100644
@@ -1837,31 +1837,14 @@ static int sfp_module_parse_power(struct sfp *sfp)
 
 static int sfp_sm_mod_hpower(struct sfp *sfp, bool enable)
 {
-	u8 val;
 	int err;
 
-	err = sfp_read(sfp, true, SFP_EXT_STATUS, &val, sizeof(val));
-	if (err != sizeof(val)) {
-		dev_err(sfp->dev, "Failed to read EEPROM: %pe\n", ERR_PTR(err));
-		return -EAGAIN;
-	}
-
-	/* DM7052 reports as a high power module, responds to reads (with
-	 * all bytes 0xff) at 0x51 but does not accept writes.  In any case,
-	 * if the bit is already set, we're already in high power mode.
-	 */
-	if (!!(val & SFP_EXT_STATUS_PWRLVL_SELECT) == enable)
-		return 0;
-
-	if (enable)
-		val |= SFP_EXT_STATUS_PWRLVL_SELECT;
-	else
-		val &= ~SFP_EXT_STATUS_PWRLVL_SELECT;
-
-	err = sfp_write(sfp, true, SFP_EXT_STATUS, &val, sizeof(val));
-	if (err != sizeof(val)) {
-		dev_err(sfp->dev, "Failed to write EEPROM: %pe\n",
-			ERR_PTR(err));
+	err = sfp_modify_u8(sfp, true, SFP_EXT_STATUS,
+			    SFP_EXT_STATUS_PWRLVL_SELECT,
+			    enable ? SFP_EXT_STATUS_PWRLVL_SELECT : 0);
+	if (err != sizeof(u8)) {
+		dev_err(sfp->dev, "failed to %sable high power: %pe\n",
+			enable ? "en" : "dis", ERR_PTR(err));
 		return -EAGAIN;
 	}
 
Signature · 1 line
2.30.2
maintainer Reviewed-by on v1 7/7
#20 lore raw

Re: [PATCH net-next 7/7] net: sfp: get rid of DM7052 hack when enabling high power

Quoted 5 lines · Russell King (Oracle)

Since we no longer mis-detect high-power mode with the DM7052 module, we no longer need the hack in sfp_module_enable_high_power(), and can now switch this to use sfp_modify_u8().

Andrew

maintainer
#21 lore raw

Re: [PATCH net-next 7/7] net: sfp: get rid of DM7052 hack when enabling high power

Quoted 8 lines · On Fri, 21 Oct 2022 17:52:52 +0200 Andrew Lunn wrote:
Quoted 5 lines · Russell King (Oracle)

Since we no longer mis-detect high-power mode with the DM7052 module, we no longer need the hack in sfp_module_enable_high_power(), and can now switch this to use sfp_modify_u8().

FWIW there is a v2 of this, somewhat mis-subjected (pw-bot's auto-Superseding logic missed it for example):
https://lore.kernel.org/all/[email protected]/

author maintainer
#22 lore raw

Re: [PATCH net-next 7/7] net: sfp: get rid of DM7052 hack when enabling high power

Quoted 6 lines · Jakub Kicinski
Quoted 5 lines · On Fri, 21 Oct 2022 17:52:52 +0200 Andrew Lunn wrote:
Quoted 4 lines · Russell King (Oracle)

Since we no longer mis-detect high-power mode with the DM7052 module, we no longer need the hack in sfp_module_enable_high_power(), and can now switch this to use sfp_modify_u8().

Quoted 7 lines
Quoted 3 lines
Quoted 1 line

FWIW there is a v2 of this, somewhat mis-subjected (pw-bot's auto-Superseding logic missed it for example):
https://lore.kernel.org/all/[email protected]/

Yes, sadly I forgot to update the subject line prefix. Getting everything correct every time is quite difficult.

Signature · 2 lines
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!
maintainer
#23 lore raw
In reply to Russell King (Oracle)
Quoted 14 lines · Russell King (Oracle)

Hi,

This series aims to improve the power level switching between standard level 1 and the higher power levels.

The first patch updates the DT binding documentation to include the minimum and default of 1W, which is the base level that every SFP cage must support. Hence, it makes sense to document this in the binding.

The second patch enforces a minimum of 1W when parsing the firmware description, and optimises the code for that case; there's no need to check for SFF8472 compliance since we will not need to touch the A2h registers.

Quoted 5 lines

Patch 3 validates that the module supports SFF-8472 rev 10.2 before checking for power level 2 - rev 10.2 is where support for power levels was introduced, so if the module doesn't support this revision, it doesn't support power levels. Setting the power level 2 declaration bit is likely to be spurious.

Or it is yet another case of violating the standard. The bit is valid, the revision is wrong in the EEPROM. How long do you think it will be before we see a quirk like this?

Quoted 3 lines

Patch 4 does the same for power level 3, except this was introduced in SFF-8472 rev 11.9. The revision code was never updated, so we use the rev 11.4 to signify this.

Great, the standard itself is broken.

Andrew