These patches take steps towards removing sk98lin from the upstream
kernel.
-- Remove sk98lin's MODULE_DEVICE_TABLE to avoid
confusing userland tools about which driver to load;
-- Mark sk98lin as Obsolete in the MAINTAINERS file; and,
-- Add sk98lin to the feature-removal-schedule.txt file in the
Documentation directory.
I accept the possibility that I may be jumping the gun on this.
However, I think it is worth opening this discussion.
Patches to follow...
The skge driver claims support for the identical list of hardware
supported by the in-kernel sk98lin driver. This can confuse userland
tools which pick modules based on the PCI ID lists exported through
MODULE_DEVICE_TABLE.
This patch removes the MODULE_DEVICE_TABLE line from sk98lin. The
driver can still be loaded manually if necessary or desireable.
Signed-off-by: John W. Linville <[email protected]>
---
drivers/net/sk98lin/skge.c | 2 --
1 files changed, 2 deletions(-)
diff --git a/drivers/net/sk98lin/skge.c b/drivers/net/sk98lin/skge.c
--- a/drivers/net/sk98lin/skge.c
+++ b/drivers/net/sk98lin/skge.c
@@ -5227,8 +5227,6 @@ static struct pci_device_id skge_pci_tbl
{ 0 }
};
-MODULE_DEVICE_TABLE(pci, skge_pci_tbl);
-
static struct pci_driver skge_driver = {
.name = "sk98lin",
.id_table = skge_pci_tbl,
The sk98lin and skge drivers cover the same set of hardware, and the
skge driver enjoys community support. This patch marks the sk98lin
driver as obsolete.
Signed-off-by: John W. Linville <[email protected]>
---
MAINTAINERS | 2 +-
1 files changed, 1 insertion(+), 1 deletion(-)
diff --git a/MAINTAINERS b/MAINTAINERS
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1592,7 +1592,7 @@ M: [email protected]
P: Ralph Roesler
M: [email protected]
W: http://www.syskonnect.com
-S: Supported
+S: Obsolete
MAESTRO PCI SOUND DRIVERS
P: Zach Brown
The sk98lin and skge drivers cover the same set of hardware, yet the
sk98lin driver is barely maintained. This patch schedules sk98lin
for removal early next year.
Signed-off-by: John W. Linville <[email protected]>
---
Documentation/feature-removal-schedule.txt | 9 +++++++++
1 files changed, 9 insertions(+)
diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -95,3 +95,12 @@ Why: This interface has been obsoleted b
to link against API-compatible library on top of libnfnetlink_queue
instead of the current 'libipq'.
Who: Harald Welte <[email protected]>
+
+---------------------------
+
+What: sk98lin driver
+When: April 2006
+Why: This driver has been essentially unmaintained for some time.
+ Internally it is a mess, and the original authors are unwilling to
+ correct it. Now the skge driver covers all the same hardware.
+Who: John W. Linville <[email protected]>
On Tue, Oct 18, 2005 at 06:38:24PM -0700, Stephen Hemminger wrote:
> John W. Linville wrote:
>
> >These patches take steps towards removing sk98lin from the upstream
> >kernel.
> >
> > -- Remove sk98lin's MODULE_DEVICE_TABLE to avoid
> > confusing userland tools about which driver to load;
> I applaud the initiative, but this it is too premature to obsolete the
> existing driver. There may be lots of chip versions and other variables
> that make
> the existing driver a better choice. Maybe eepro100 is a better target
> for removal right now.
That's cool...but I still think the first one (removing the
MODULE_DEVICE_TABLE) is worthwhile. At least that gets more testers
for skge.
What do you think?
John
--
John W. Linville
[email protected]
John W. Linville wrote:
>These patches take steps towards removing sk98lin from the upstream
>kernel.
>
> -- Remove sk98lin's MODULE_DEVICE_TABLE to avoid
> confusing userland tools about which driver to load;
>
> -- Mark sk98lin as Obsolete in the MAINTAINERS file; and,
>
> -- Add sk98lin to the feature-removal-schedule.txt file in the
> Documentation directory.
>
>I accept the possibility that I may be jumping the gun on this.
>However, I think it is worth opening this discussion.
>
>Patches to follow...
>
>
I applaud the initiative, but this it is too premature to obsolete the
existing driver. There may be lots of chip versions and other variables
that make
the existing driver a better choice. Maybe eepro100 is a better target
for removal right now.
John W. Linville wrote:
> These patches take steps towards removing sk98lin from the upstream
> kernel.
>
> -- Remove sk98lin's MODULE_DEVICE_TABLE to avoid
> confusing userland tools about which driver to load;
>
> -- Mark sk98lin as Obsolete in the MAINTAINERS file; and,
>
> -- Add sk98lin to the feature-removal-schedule.txt file in the
> Documentation directory.
>
> I accept the possibility that I may be jumping the gun on this.
> However, I think it is worth opening this discussion.
I'll let Stephen make the call on this one (as he did...).
Jeff