[PATCH V2 ethtool-next 0/2] ethtool: add two link extended substates of bad signal

STALE1813d

6 messages, 2 authors, 2021-08-26 · open the first message on its own page

[PATCH V2 ethtool-next 0/2] ethtool: add two link extended substates of bad signal

From: Guangbin Huang <hidden>
Date: 2021-08-25 07:49:20

This series adds two link extended substates of bad signal.

change log:
V1 -> V2:
1. Split the uapi header update into a separate patch according to
   Michal's opinion as follow link:
   https://lore.kernel.org/netdev/20210824173614.mkv5i72sutxtdvrk@lion.mk-sys.cz/

Guangbin Huang (2):
  update UAPI header copies
  netlink: settings: add two link extended substates of bad signal
    integrity

 netlink/settings.c   | 4 ++++
 uapi/linux/ethtool.h | 2 ++
 2 files changed, 6 insertions(+)

-- 
2.8.1

[PATCH V2 ethtool-next 1/2] update UAPI header copies

From: Guangbin Huang <hidden>
Date: 2021-08-25 07:49:16

Update to kernel commit 5b4ecc3d4c4a.

Signed-off-by: Guangbin Huang <redacted>
---
 uapi/linux/ethtool.h | 2 ++
 1 file changed, 2 insertions(+)
diff --git a/uapi/linux/ethtool.h b/uapi/linux/ethtool.h
index c6ec1111ffa3..bd1f09b23cf5 100644
--- a/uapi/linux/ethtool.h
+++ b/uapi/linux/ethtool.h
@@ -637,6 +637,8 @@ enum ethtool_link_ext_substate_link_logical_mismatch {
 enum ethtool_link_ext_substate_bad_signal_integrity {
 	ETHTOOL_LINK_EXT_SUBSTATE_BSI_LARGE_NUMBER_OF_PHYSICAL_ERRORS = 1,
 	ETHTOOL_LINK_EXT_SUBSTATE_BSI_UNSUPPORTED_RATE,
+	ETHTOOL_LINK_EXT_SUBSTATE_BSI_SERDES_REFERENCE_CLOCK_LOST,
+	ETHTOOL_LINK_EXT_SUBSTATE_BSI_SERDES_ALOS,
 };
 
 /* More information in addition to ETHTOOL_LINK_EXT_STATE_CABLE_ISSUE. */
-- 
2.8.1

[PATCH V2 ethtool-next 2/2] netlink: settings: add two link extended substates of bad signal integrity

From: Guangbin Huang <hidden>
Date: 2021-08-25 07:49:26

Add two link extended substates of bad signal integrity available in the
kernel.

ETHTOOL_LINK_EXT_SUBSTATE_BSI_SERDES_REFERENCE_CLOCK_LOST means the input
external clock signal for SerDes is too weak or lost.

ETHTOOL_LINK_EXT_SUBSTATE_BSI_SERDES_ALOS means the received signal for
SerDes is too weak because analog loss of signal.

Signed-off-by: Guangbin Huang <redacted>
---
 netlink/settings.c | 4 ++++
 1 file changed, 4 insertions(+)
diff --git a/netlink/settings.c b/netlink/settings.c
index e47a38f3058f..6d10a0703861 100644
--- a/netlink/settings.c
+++ b/netlink/settings.c
@@ -639,6 +639,10 @@ static const char *const names_bad_signal_integrity_link_ext_substate[] = {
 		"Large number of physical errors",
 	[ETHTOOL_LINK_EXT_SUBSTATE_BSI_UNSUPPORTED_RATE]		=
 		"Unsupported rate",
+	[ETHTOOL_LINK_EXT_SUBSTATE_BSI_SERDES_REFERENCE_CLOCK_LOST]	=
+		"Serdes reference clock lost",
+	[ETHTOOL_LINK_EXT_SUBSTATE_BSI_SERDES_ALOS]			=
+		"Serdes ALOS",
 };
 
 static const char *const names_cable_issue_link_ext_substate[] = {
-- 
2.8.1

Re: [PATCH V2 ethtool-next 2/2] netlink: settings: add two link extended substates of bad signal integrity

From: Michal Kubecek <hidden>
Date: 2021-08-26 09:34:20

On Wed, Aug 25, 2021 at 03:45:13PM +0800, Guangbin Huang wrote:
Add two link extended substates of bad signal integrity available in the
kernel.

ETHTOOL_LINK_EXT_SUBSTATE_BSI_SERDES_REFERENCE_CLOCK_LOST means the input
external clock signal for SerDes is too weak or lost.

ETHTOOL_LINK_EXT_SUBSTATE_BSI_SERDES_ALOS means the received signal for
SerDes is too weak because analog loss of signal.

Signed-off-by: Guangbin Huang <redacted>
Applied, thank you.

Michal

Re: [PATCH V2 ethtool-next 1/2] update UAPI header copies

From: Michal Kubecek <hidden>
Date: 2021-08-26 09:45:36

On Wed, Aug 25, 2021 at 03:45:12PM +0800, Guangbin Huang wrote:
quoted hunk
Update to kernel commit 5b4ecc3d4c4a.

Signed-off-by: Guangbin Huang <redacted>
---
 uapi/linux/ethtool.h | 2 ++
 1 file changed, 2 insertions(+)
diff --git a/uapi/linux/ethtool.h b/uapi/linux/ethtool.h
index c6ec1111ffa3..bd1f09b23cf5 100644
--- a/uapi/linux/ethtool.h
+++ b/uapi/linux/ethtool.h
@@ -637,6 +637,8 @@ enum ethtool_link_ext_substate_link_logical_mismatch {
 enum ethtool_link_ext_substate_bad_signal_integrity {
 	ETHTOOL_LINK_EXT_SUBSTATE_BSI_LARGE_NUMBER_OF_PHYSICAL_ERRORS = 1,
 	ETHTOOL_LINK_EXT_SUBSTATE_BSI_UNSUPPORTED_RATE,
+	ETHTOOL_LINK_EXT_SUBSTATE_BSI_SERDES_REFERENCE_CLOCK_LOST,
+	ETHTOOL_LINK_EXT_SUBSTATE_BSI_SERDES_ALOS,
 };
 
 /* More information in addition to ETHTOOL_LINK_EXT_STATE_CABLE_ISSUE. */
-- 
2.8.1
I replaced this commit with a full update of uapi headers. The point is
that if we keep cherry picking only specific changes in the headers, it
will become harder and harder to check if something is missing or if we
diverged from kernel. This is why an update of uapi headers should
always update all of them to the state of the same kernel commit
(usually current master or net-next tree).

I added the link to ethtool-import-uapi script to devel documentation on
the ethtool web page make things easier.

Michal

Re: [PATCH V2 ethtool-next 1/2] update UAPI header copies

From: huangguangbin (A) <hidden>
Date: 2021-08-26 11:22:08


On 2021/8/26 17:45, Michal Kubecek wrote:
On Wed, Aug 25, 2021 at 03:45:12PM +0800, Guangbin Huang wrote:
quoted
Update to kernel commit 5b4ecc3d4c4a.

Signed-off-by: Guangbin Huang <redacted>
---
  uapi/linux/ethtool.h | 2 ++
  1 file changed, 2 insertions(+)
diff --git a/uapi/linux/ethtool.h b/uapi/linux/ethtool.h
index c6ec1111ffa3..bd1f09b23cf5 100644
--- a/uapi/linux/ethtool.h
+++ b/uapi/linux/ethtool.h
@@ -637,6 +637,8 @@ enum ethtool_link_ext_substate_link_logical_mismatch {
  enum ethtool_link_ext_substate_bad_signal_integrity {
  	ETHTOOL_LINK_EXT_SUBSTATE_BSI_LARGE_NUMBER_OF_PHYSICAL_ERRORS = 1,
  	ETHTOOL_LINK_EXT_SUBSTATE_BSI_UNSUPPORTED_RATE,
+	ETHTOOL_LINK_EXT_SUBSTATE_BSI_SERDES_REFERENCE_CLOCK_LOST,
+	ETHTOOL_LINK_EXT_SUBSTATE_BSI_SERDES_ALOS,
  };
  
  /* More information in addition to ETHTOOL_LINK_EXT_STATE_CABLE_ISSUE. */
-- 
2.8.1
I replaced this commit with a full update of uapi headers. The point is
that if we keep cherry picking only specific changes in the headers, it
will become harder and harder to check if something is missing or if we
diverged from kernel. This is why an update of uapi headers should
always update all of them to the state of the same kernel commit
(usually current master or net-next tree).

I added the link to ethtool-import-uapi script to devel documentation on
the ethtool web page make things easier.

Michal
Ok, thanks!
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help