Clean driver, module and FW versions

PATCH In review Leon Romanovsky 8 participants 39 messages Feb 24, 2020 – Apr 14, 2020

Subsystems (22)

Networking Network drivers Arm Rockchip Arm/Allwinner sunXi SoC support +18 more subsystems

lore

Share & tools

lore

Leon Romanovsky

author

2020-02-24 08:52 UTC #0 lore raw

cover Clean driver, module and FW versions

From: Leon Romanovsky <[email protected]>

Change log:
v1:
* Split all FW cleanups patches to separate patches
* Fixed commit message
* Deleted odd DRV_RELDATE
* Added everyone from MAINTAINERS file
v0: https://lore.kernel.org/netdev/[email protected]

----------------------------------------------------------------------------

Hi,

This is first patchset to netdev (already sent RDMA [1] and arch/um [2]) in attempt to unify the version management for in-tree kernel code. The patches follow already accepted ethtool change [3] to set as a default linux kernel version.

It allows us to remove driver version and present to the users unified picture of driver version, which is similar to default MODULE_VERSION().

As part of this series, I deleted various creative attempts to mark absence of FW. There is no need to set "N/A" in ethtool ->fw_version field and it is enough to do not set it.

1.
The code is compile tested and passes 0-day kbuild.
2.
The proposed changes are based on commit:
2bb07f4e1d86 ("tc-testing: updated tdc tests for basic filter")
3.
WIP branch is [4].

[1] https://lore.kernel.org/linux-rdma/[email protected]/
[2] http://lists.infradead.org/pipermail/linux-um/2020-February/002913.html
[3] https://lore.kernel.org/linux-rdma/[email protected]/
[4] https://git.kernel.org/pub/scm/linux/kernel/git/leon/linux-rdma.git/log/?h=ethtool

Thanks

Leon Romanovsky (18):
net/bond: Delete driver and module versions
net/dummy: Ditch driver and module versions
net/3com: Delete driver and module versions from 3com drivers net/adaptec: Clean driver versions
net/aeroflex: Clean ethtool_info struct assignments net/aeroflex: Don't assign FW if it is not available net/agere: Delete unneeded driver version
net/alacritech: Delete driver version
net/allwinner: Remove driver version
net/alteon: Properly report FW version
net/althera: Delete hardcoded driver version
net/amazon: Ensure that driver version is aligned to the linux kernel net/amd: Remove useless driver version
net/apm: Remove useless driver version
net/apm: Properly mark absence of FW
net/aquantia: Delete module version
net/arc: Delete driver version
net/atheros: Clean atheros code from driver version

 drivers/net/bonding/bond_main.c               |  6 +-----
 drivers/net/bonding/bonding_priv.h            |  5 ++---
 drivers/net/dummy.c                           |  3 ---
 drivers/net/ethernet/3com/3c509.c             |  8 +-------
 drivers/net/ethernet/3com/3c515.c             | 16 +---------------
 drivers/net/ethernet/3com/3c589_cs.c          |  2 --
 drivers/net/ethernet/3com/typhoon.c           |  1 -
 drivers/net/ethernet/adaptec/starfire.c       | 19 +------------------
 drivers/net/ethernet/aeroflex/greth.c         |  2 --
 drivers/net/ethernet/agere/et131x.c           |  1 -
 drivers/net/ethernet/agere/et131x.h           |  1 -
 drivers/net/ethernet/alacritech/slicoss.c     |  3 ---
 drivers/net/ethernet/allwinner/sun4i-emac.c   |  2 --
 drivers/net/ethernet/alteon/acenic.c          |  5 ++---
 .../net/ethernet/altera/altera_tse_ethtool.c  |  1 -
 drivers/net/ethernet/amazon/ena/ena_ethtool.c |  1 -
 drivers/net/ethernet/amazon/ena/ena_netdev.c  | 17 ++---------------
 drivers/net/ethernet/amazon/ena/ena_netdev.h  | 11 -----------
 drivers/net/ethernet/amd/amd8111e.c           |  5 +----
 drivers/net/ethernet/amd/au1000_eth.c         |  5 -----
 drivers/net/ethernet/amd/nmclan_cs.c          |  9 +++------
 drivers/net/ethernet/amd/pcnet32.c            |  7 -------
 drivers/net/ethernet/amd/sunlance.c           | 10 ----------
 drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c  |  1 -
 drivers/net/ethernet/amd/xgbe/xgbe-main.c     |  1 -
 drivers/net/ethernet/amd/xgbe/xgbe.h          |  1 -
 drivers/net/ethernet/apm/xgene-v2/ethtool.c   |  2 --
 drivers/net/ethernet/apm/xgene-v2/main.c      |  1 -
 drivers/net/ethernet/apm/xgene-v2/main.h      |  1 -
 .../ethernet/apm/xgene/xgene_enet_ethtool.c   |  2 --
 .../net/ethernet/apm/xgene/xgene_enet_main.c  |  1 -
 .../net/ethernet/apm/xgene/xgene_enet_main.h  |  1 -
 .../net/ethernet/aquantia/atlantic/aq_cfg.h   |  4 ----
 .../ethernet/aquantia/atlantic/aq_common.h    |  1 -
 .../ethernet/aquantia/atlantic/aq_ethtool.c   |  1 -
 .../net/ethernet/aquantia/atlantic/aq_main.c  |  1 -
 drivers/net/ethernet/aquantia/atlantic/ver.h  | 12 ------------
 drivers/net/ethernet/arc/emac.h               |  1 -
 drivers/net/ethernet/arc/emac_arc.c           |  2 --
 drivers/net/ethernet/arc/emac_main.c          |  1 -
 drivers/net/ethernet/arc/emac_rockchip.c      |  2 --
 drivers/net/ethernet/atheros/atl1c/atl1c.h    |  1 -
 .../ethernet/atheros/atl1c/atl1c_ethtool.c    |  2 --
 .../net/ethernet/atheros/atl1c/atl1c_main.c   |  5 -----
 drivers/net/ethernet/atheros/atl1e/atl1e.h    |  1 -
 .../ethernet/atheros/atl1e/atl1e_ethtool.c    |  2 --
 .../net/ethernet/atheros/atl1e/atl1e_main.c   |  4 ----
 drivers/net/ethernet/atheros/atlx/atl1.c      |  6 ------
 drivers/net/ethernet/atheros/atlx/atl2.c      | 10 ----------
 49 files changed, 14 insertions(+), 193 deletions(-)
 delete mode 100644 drivers/net/ethernet/aquantia/atlantic/ver.h

Leon Romanovsky

author

2020-02-24 08:52 UTC #1 lore raw

PATCH 1/18 net/bond: Delete driver and module versions

From: Leon Romanovsky <[email protected]>

The in-kernel code has already unique version, which is based on Linus's tag, update the bond driver to be consistent with that version.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/bonding/bond_main.c    | 6 +-----
 drivers/net/bonding/bonding_priv.h | 5 ++---
 2 files changed, 3 insertions(+), 8 deletions(-)

drivers/net/bonding/bond_main.c +1 -5

index c68c1d1387ee..2e70e43c5df5 100644
@@ -4370,7 +4370,6 @@ static void bond_ethtool_get_drvinfo(struct net_device *bond_dev,
 				     struct ethtool_drvinfo *drvinfo)
 {
 	strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, DRV_VERSION, sizeof(drvinfo->version));
 	snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), "%d",
 		 BOND_ABI_VERSION);
 }

@@ -5008,8 +5007,6 @@ static int __init bonding_init(void)
 	int i;
 	int res;
 
-	pr_info("%s", bond_version);
-
 	res = bond_check_params(&bonding_defaults);
 	if (res)
 		goto out;

@@ -5064,6 +5061,5 @@ static void __exit bonding_exit(void)
 module_init(bonding_init);
 module_exit(bonding_exit);
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
-MODULE_DESCRIPTION(DRV_DESCRIPTION ", v" DRV_VERSION);
+MODULE_DESCRIPTION(DRV_DESCRIPTION);
 MODULE_AUTHOR("Thomas Davis, [email protected] and many others");

drivers/net/bonding/bonding_priv.h +2 -3

index 5a4d81a9437c..45b77bc8c7b3 100644
@@ -14,12 +14,11 @@
 
 #ifndef _BONDING_PRIV_H
 #define _BONDING_PRIV_H
+#include <linux/vermagic.h>
 
-#define DRV_VERSION	"3.7.1"
-#define DRV_RELDATE	"April 27, 2011"
 #define DRV_NAME	"bonding"
 #define DRV_DESCRIPTION	"Ethernet Channel Bonding Driver"
 
-#define bond_version DRV_DESCRIPTION ": v" DRV_VERSION " (" DRV_RELDATE ")\n"
+#define bond_version DRV_DESCRIPTION ": v" UTS_RELEASE "\n"
 
 #endif

Leon Romanovsky

author

2020-02-24 08:52 UTC #2 lore raw

PATCH 2/18 net/dummy: Ditch driver and module versions

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

Delete constant driver and module versions in favor of standard global version which is unique to whole kernel.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/dummy.c | 3 ---
 1 file changed, 3 deletions(-)

drivers/net/dummy.c +0 -3

index 3031a5fc5427..bab3a9bb5e6f 100644
@@ -42,7 +42,6 @@
 #include <linux/u64_stats_sync.h>
 
 #define DRV_NAME	"dummy"
-#define DRV_VERSION	"1.0"
 
 static int numdummies = 1;

@@ -104,7 +103,6 @@ static void dummy_get_drvinfo(struct net_device *dev,
 			      struct ethtool_drvinfo *info)
 {
 	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
 }
 
 static const struct ethtool_ops dummy_ethtool_ops = {

@@ -212,4 +210,3 @@ module_init(dummy_init_module);
 module_exit(dummy_cleanup_module);
 MODULE_LICENSE("GPL");
 MODULE_ALIAS_RTNL_LINK(DRV_NAME);
-MODULE_VERSION(DRV_VERSION);

--
2.24.1

Leon Romanovsky

author

2020-02-24 08:52 UTC #3 lore raw

PATCH 3/18 net/3com: Delete driver and module versions from 3com drivers

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

There is no need to mislead users by providing different versions for driver, ethtool and modules. Delete driver assignments and let use the default one.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/ethernet/3com/3c509.c    |  8 +-------
 drivers/net/ethernet/3com/3c515.c    | 16 +---------------
 drivers/net/ethernet/3com/3c589_cs.c |  2 --
 drivers/net/ethernet/3com/typhoon.c  |  1 -
 4 files changed, 2 insertions(+), 25 deletions(-)

drivers/net/ethernet/3com/3c509.c +1 -7

index 8cafd06ff0c4..b762176a1406 100644
@@ -60,8 +60,6 @@
 */
 
 #define DRV_NAME	"3c509"
-#define DRV_VERSION	"1.20"
-#define DRV_RELDATE	"04Feb2008"
 
 /* A few values that may be tweaked. */

@@ -87,13 +85,12 @@
 #include <linux/device.h>
 #include <linux/eisa.h>
 #include <linux/bitops.h>
+#include <linux/vermagic.h>
 
 #include <linux/uaccess.h>
 #include <asm/io.h>
 #include <asm/irq.h>
 
-static char version[] = DRV_NAME ".c:" DRV_VERSION " " DRV_RELDATE " [email protected]\n";
-
 #ifdef EL3_DEBUG
 static int el3_debug = EL3_DEBUG;
 #else

@@ -547,8 +544,6 @@ static int el3_common_init(struct net_device *dev)
 	       dev->name, dev->base_addr, if_names[(dev->if_port & 0x03)],
 	       dev->dev_addr, dev->irq);
 
-	if (el3_debug > 0)
-		pr_info("%s", version);
 	return 0;
 
 }

@@ -1143,7 +1138,6 @@ el3_netdev_set_ecmd(struct net_device *dev,
 static void el3_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 {
 	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
 }
 
 static int el3_get_link_ksettings(struct net_device *dev,

drivers/net/ethernet/3com/3c515.c +1 ~~-15~~

index 1e233e2f0a5a..90312fcd6319 100644
@@ -22,12 +22,8 @@
 
 */
 
+#include <linux/vermagic.h>
 #define DRV_NAME		"3c515"
-#define DRV_VERSION		"0.99t-ac"
-#define DRV_RELDATE		"28-Oct-2002"
-
-static char *version =
-DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE " [email protected] and others\n";
 
 #define CORKSCREW 1

@@ -84,7 +80,6 @@ static int max_interrupt_work = 20;
 MODULE_AUTHOR("Donald Becker <[email protected]>");
 MODULE_DESCRIPTION("3Com 3c515 Corkscrew driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
 
 /* "Knobs" for adjusting internal parameters. */
 /* Put out somewhat more debugging messages. (0 - no msg, 1 minimal msgs). */

@@ -418,8 +413,6 @@ int init_module(void)
 	int found = 0;
 	if (debug >= 0)
 		corkscrew_debug = debug;
-	if (corkscrew_debug)
-		pr_debug("%s", version);
 	while (corkscrew_scan(-1))
 		found++;
 	return found ? 0 : -ENODEV;

@@ -429,16 +422,10 @@ int init_module(void)
 struct net_device *tc515_probe(int unit)
 {
 	struct net_device *dev = corkscrew_scan(unit);
-	static int printed;
 
 	if (!dev)
 		return ERR_PTR(-ENODEV);
 
-	if (corkscrew_debug > 0 && !printed) {
-		printed = 1;
-		pr_debug("%s", version);
-	}
-
 	return dev;
 }
 #endif				/* not MODULE */

@@ -1540,7 +1527,6 @@ static void netdev_get_drvinfo(struct net_device *dev,
 			       struct ethtool_drvinfo *info)
 {
 	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
 	snprintf(info->bus_info, sizeof(info->bus_info), "ISA 0x%lx",
 		 dev->base_addr);
 }

drivers/net/ethernet/3com/3c589_cs.c +0 -2

index d47cde6c5f08..09816e84314d 100644
@@ -23,7 +23,6 @@
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #define DRV_NAME	"3c589_cs"
-#define DRV_VERSION	"1.162-ac"
 
 #include <linux/module.h>
 #include <linux/kernel.h>

@@ -482,7 +481,6 @@ static void netdev_get_drvinfo(struct net_device *dev,
 			       struct ethtool_drvinfo *info)
 {
 	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
 	snprintf(info->bus_info, sizeof(info->bus_info),
 		"PCMCIA 0x%lx", dev->base_addr);
 }

drivers/net/ethernet/3com/typhoon.c +0 -1

index 14fce6658106..4383ee615793 100644
@@ -127,7 +127,6 @@ static const int multicast_filter_limit = 32;
 #include "typhoon.h"
 
 MODULE_AUTHOR("David Dillow <[email protected]>");
-MODULE_VERSION("1.0");
 MODULE_LICENSE("GPL");
 MODULE_FIRMWARE(FIRMWARE_NAME);
 MODULE_DESCRIPTION("3Com Typhoon Family (3C990, 3CR990, and variants)");

--
2.24.1

Borislav Petkov

2020-04-11 15:56 UTC #4 lore raw

PATCH 1/1 net/3com/3c515: Fix MODULE_ARCH_VERMAGIC redefinition

From: Borislav Petkov <[email protected]>

Change the include order so that MODULE_ARCH_VERMAGIC from the arch header arch/x86/include/asm/module.h gets used instead of the fallback from include/linux/vermagic.h and thus fix:

In file included from ./include/linux/module.h:30,

                   from drivers/net/ethernet/3com/3c515.c:56:

./arch/x86/include/asm/module.h:73: warning: "MODULE_ARCH_VERMAGIC" redefined
73 | # define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY
|
In file included from drivers/net/ethernet/3com/3c515.c:25:
./include/linux/vermagic.h:28: note: this is the location of the previous definition
28 | #define MODULE_ARCH_VERMAGIC ""
|

Fixes 6bba2e89a88c ("net/3com: Delete driver and module versions from 3com drivers")

Signed-off-by Borislav Petkov <[email protected]>

---
 drivers/net/ethernet/3com/3c515.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/net/ethernet/3com/3c515.c +1 -1

index 90312fcd6319..cdceef891dbd 100644
@@ -22,7 +22,6 @@
 
 */
 
-#include <linux/vermagic.h>
 #define DRV_NAME		"3c515"
 
 #define CORKSCREW 1

@@ -67,6 +66,7 @@ static int max_interrupt_work = 20;
 #include <linux/timer.h>
 #include <linux/ethtool.h>
 #include <linux/bitops.h>
+#include <linux/vermagic.h>
 
 #include <linux/uaccess.h>
 #include <asm/io.h>

Signature · 7 lines

2.21.0

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Leon Romanovsky

author

2020-04-11 16:11 UTC #5 lore raw

Re: [PATCH] net/3com/3c515: Fix MODULE_ARCH_VERMAGIC redefinition

Quoted 16 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>
Change the include order so that MODULE_ARCH_VERMAGIC from the arch header arch/x86/include/asm/module.h gets used instead of the fallback from include/linux/vermagic.h and thus fix:
In file included from ./include/linux/module.h:30,
                   from drivers/net/ethernet/3com/3c515.c:56:
./arch/x86/include/asm/module.h:73: warning: "MODULE_ARCH_VERMAGIC" redefined
73 | # define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY
|
In file included from drivers/net/ethernet/3com/3c515.c:25:
./include/linux/vermagic.h:28: note: this is the location of the previous definition
28 | #define MODULE_ARCH_VERMAGIC ""
|

Quoted 5 lines

Fixes 6bba2e89a88c ("net/3com: Delete driver and module versions from 3com drivers")
Signed-off-by Borislav Petkov <[email protected]>
---
 drivers/net/ethernet/3com/3c515.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Probably, this is the right change, but I have a feeling that the right solution will be inside headers itself. It is a little bit strange that both very common kernel headers like module.h and vermagic.h are location dependant.

Thanks

Quoted 29 lines

drivers/net/ethernet/3com/3c515.c +1 -2

index 90312fcd6319..cdceef891dbd 100644
@@ -22,7 +22,6 @@
 
 */
 
-#include <linux/vermagic.h>
 #define DRV_NAME		"3c515"
 
 #define CORKSCREW 1

@@ -67,6 +66,7 @@ static int max_interrupt_work = 20;
 #include <linux/timer.h>
 #include <linux/ethtool.h>
 #include <linux/bitops.h>
+#include <linux/vermagic.h>
 
 #include <linux/uaccess.h>
 #include <asm/io.h>
--

2.21.0

--
Regards/Gruss,
Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Borislav Petkov

2020-04-11 17:35 UTC #6 lore raw

Re: [PATCH] net/3com/3c515: Fix MODULE_ARCH_VERMAGIC redefinition

Quoted 4 lines · Leon Romanovsky

Probably, this is the right change, but I have a feeling that the right solution will be inside headers itself. It is a little bit strange that both very common kernel headers like module.h and vermagic.h are location dependant.

Judging by how only a couple of net drivers include vermagic.h directly, doh, of course:

drivers/net/ethernet/3com/3c509.c +0 -1

index b762176a1406..139d0120f511 100644
@@ -85,7 +85,6 @@
 #include <linux/device.h>
 #include <linux/eisa.h>
 #include <linux/bitops.h>
-#include <linux/vermagic.h>
 
 #include <linux/uaccess.h>
 #include <asm/io.h>

drivers/net/ethernet/3com/3c515.c +0 -1

index 90312fcd6319..47b4215bb93b 100644
@@ -22,7 +22,6 @@
 
 */
 
-#include <linux/vermagic.h>
 #define DRV_NAME		"3c515"
 
 #define CORKSCREW 1

---

Drivers include

#include <linux/module.h>

which includes

#include <asm/module.h>

which defines the arch-specific MODULE_ARCH_VERMAGIC.

Why did you need to include vermagic.h directly? i386 builds fine with the vermagic.h includes removed or was it some other arches which needed it?

Thx.

Signature · 4 lines

Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Leon Romanovsky

author

2020-04-11 18:10 UTC #7 lore raw

Re: [PATCH] net/3com/3c515: Fix MODULE_ARCH_VERMAGIC redefinition

Quoted 42 lines · Borislav Petkov

Quoted 4 lines · Leon Romanovsky
Probably, this is the right change, but I have a feeling that the right solution will be inside headers itself. It is a little bit strange that both very common kernel headers like module.h and vermagic.h are location dependant.
Judging by how only a couple of net drivers include vermagic.h directly, doh, of course:
drivers/net/ethernet/3com/3c509.c +0 -1
index b762176a1406..139d0120f511 100644
@@ -85,7 +85,6 @@
 #include <linux/device.h>
 #include <linux/eisa.h>
 #include <linux/bitops.h>
-#include <linux/vermagic.h>
 
 #include <linux/uaccess.h>
 #include <asm/io.h>
drivers/net/ethernet/3com/3c515.c +0 -2
index 90312fcd6319..47b4215bb93b 100644
@@ -22,7 +22,6 @@
 
 */
 
-#include <linux/vermagic.h>
 #define DRV_NAME		"3c515"
 
 #define CORKSCREW 1
 
---
Drivers include
#include <linux/module.h>
which includes

Quoted 7 lines

#include <asm/module.h>
which defines the arch-specific MODULE_ARCH_VERMAGIC.
Why did you need to include vermagic.h directly? i386 builds fine with the vermagic.h includes removed or was it some other arches which needed it?

I want to think that it was an outcome of some 0-day kbuild report, but I am not sure about that anymore [1].

Thanks

[1] https://lists.01.org/hyperkitty/list/[email protected]/

Quoted 8 lines

Thx.
--
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette

Borislav Petkov

2020-04-11 18:41 UTC #8 lore raw

Re: [PATCH] net/3com/3c515: Fix MODULE_ARCH_VERMAGIC redefinition

Quoted 2 lines · Leon Romanovsky

I want to think that it was an outcome of some 0-day kbuild report, but I am not sure about that anymore [1].

I pushed it here:

https://git.kernel.org/pub/scm/linux/kernel/git/bp/bp.git/log/?h=rc0%2b0-3c

The 0day bot will send me a mail soon. :)

Signature · 4 lines

Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

David Miller

maintainer

2020-04-13 04:03 UTC #9 lore raw

Re: [PATCH] net/3com/3c515: Fix MODULE_ARCH_VERMAGIC redefinition

In reply to Borislav Petkov

From: Borislav Petkov <[email protected]>
Date: Sat, 11 Apr 2020 17:56:23 +0200

Quoted 16 lines

From: Borislav Petkov <[email protected]>
Change the include order so that MODULE_ARCH_VERMAGIC from the arch header arch/x86/include/asm/module.h gets used instead of the fallback from include/linux/vermagic.h and thus fix:
In file included from ./include/linux/module.h:30,
                   from drivers/net/ethernet/3com/3c515.c:56:
./arch/x86/include/asm/module.h:73: warning: "MODULE_ARCH_VERMAGIC" redefined
73 | # define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY
|
In file included from drivers/net/ethernet/3com/3c515.c:25:
./include/linux/vermagic.h:28: note: this is the location of the previous definition
28 | #define MODULE_ARCH_VERMAGIC ""
|

Quoted 2 lines

Fixes 6bba2e89a88c ("net/3com: Delete driver and module versions from 3com drivers")
Signed-off-by Borislav Petkov <[email protected]>

I'm so confused, that commit in the Fixes: tag is _removing_ code but adding new #include directives?!?!

Is vermagic.h really needed in these files?

Leon Romanovsky

author

2020-04-13 04:55 UTC #10 lore raw

Re: [PATCH] net/3com/3c515: Fix MODULE_ARCH_VERMAGIC redefinition

Quoted 19 lines · David Miller

From: Borislav Petkov <[email protected]>
Date: Sat, 11 Apr 2020 17:56:23 +0200
Quoted 16 lines
From: Borislav Petkov <[email protected]>
Change the include order so that MODULE_ARCH_VERMAGIC from the arch header arch/x86/include/asm/module.h gets used instead of the fallback from include/linux/vermagic.h and thus fix:
In file included from ./include/linux/module.h:30,
                   from drivers/net/ethernet/3com/3c515.c:56:
./arch/x86/include/asm/module.h:73: warning: "MODULE_ARCH_VERMAGIC" redefined
73 | # define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY
|
In file included from drivers/net/ethernet/3com/3c515.c:25:
./include/linux/vermagic.h:28: note: this is the location of the previous definition
28 | #define MODULE_ARCH_VERMAGIC ""
|

Quoted 7 lines

Quoted 2 lines
Fixes 6bba2e89a88c ("net/3com: Delete driver and module versions from 3com drivers")
Signed-off-by Borislav Petkov <[email protected]>
I'm so confused, that commit in the Fixes: tag is _removing_ code but adding new #include directives?!?!
Is vermagic.h really needed in these files?

You are completely right, it is not needed at all in those files.

Thanks

David Miller

maintainer

2020-04-13 05:07 UTC #11 lore raw

Re: [PATCH] net/3com/3c515: Fix MODULE_ARCH_VERMAGIC redefinition

From: Leon Romanovsky <[email protected]>
Date: Mon, 13 Apr 2020 07:55:55 +0300

Quoted 23 lines · David Miller

Quoted 22 lines
From: Borislav Petkov <[email protected]>
Date: Sat, 11 Apr 2020 17:56:23 +0200
Quoted 18 lines
From: Borislav Petkov <[email protected]>
Change the include order so that MODULE_ARCH_VERMAGIC from the arch header arch/x86/include/asm/module.h gets used instead of the fallback from include/linux/vermagic.h and thus fix:
In file included from ./include/linux/module.h:30,
                   from drivers/net/ethernet/3com/3c515.c:56:
./arch/x86/include/asm/module.h:73: warning: "MODULE_ARCH_VERMAGIC" redefined
73 | # define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY
|
In file included from drivers/net/ethernet/3com/3c515.c:25:
./include/linux/vermagic.h:28: note: this is the location of the previous definition
28 | #define MODULE_ARCH_VERMAGIC ""
|
Fixes 6bba2e89a88c ("net/3com: Delete driver and module versions from 3com drivers")
Signed-off-by Borislav Petkov <[email protected]>

Quoted 6 lines

Quoted 4 lines
I'm so confused, that commit in the Fixes: tag is _removing_ code but adding new #include directives?!?!
Is vermagic.h really needed in these files?
You are completely right, it is not needed at all in those files.

Ok let's just remove it to fix this.

Leon Romanovsky

author

2020-04-13 05:26 UTC #12 lore raw

Re: [PATCH] net/3com/3c515: Fix MODULE_ARCH_VERMAGIC redefinition

Quoted 26 lines · David Miller

From: Leon Romanovsky <[email protected]>
Date: Mon, 13 Apr 2020 07:55:55 +0300
Quoted 23 lines · David Miller
Quoted 22 lines
From: Borislav Petkov <[email protected]>
Date: Sat, 11 Apr 2020 17:56:23 +0200
Quoted 18 lines
From: Borislav Petkov <[email protected]>
Change the include order so that MODULE_ARCH_VERMAGIC from the arch header arch/x86/include/asm/module.h gets used instead of the fallback from include/linux/vermagic.h and thus fix:
In file included from ./include/linux/module.h:30,
                   from drivers/net/ethernet/3com/3c515.c:56:
./arch/x86/include/asm/module.h:73: warning: "MODULE_ARCH_VERMAGIC" redefined
73 | # define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY
|
In file included from drivers/net/ethernet/3com/3c515.c:25:
./include/linux/vermagic.h:28: note: this is the location of the previous definition
28 | #define MODULE_ARCH_VERMAGIC ""
|
Fixes 6bba2e89a88c ("net/3com: Delete driver and module versions from 3com drivers")
Signed-off-by Borislav Petkov <[email protected]>

Quoted 8 lines

Quoted 6 lines
Quoted 4 lines
I'm so confused, that commit in the Fixes: tag is _removing_ code but adding new #include directives?!?!
Is vermagic.h really needed in these files?
You are completely right, it is not needed at all in those files.
Ok let's just remove it to fix this.

Thanks a lot.

How do you want us to handle it? Boris resend, me to send, you to fix?

Thanks

David Miller

maintainer

2020-04-13 05:36 UTC #13 lore raw

Re: [PATCH] net/3com/3c515: Fix MODULE_ARCH_VERMAGIC redefinition

From: Leon Romanovsky <[email protected]>
Date: Mon, 13 Apr 2020 08:26:37 +0300

Quoted 1 line

How do you want us to handle it? Boris resend, me to send, you to fix?

Anyone other than me can do it ;-)

Borislav Petkov

2020-04-13 08:04 UTC #14 lore raw

Re: [PATCH] net/3com/3c515: Fix MODULE_ARCH_VERMAGIC redefinition

Quoted 6 lines · David Miller

From: Leon Romanovsky <[email protected]>
Date: Mon, 13 Apr 2020 08:26:37 +0300
Quoted 1 line
How do you want us to handle it? Boris resend, me to send, you to fix?
Anyone other than me can do it ;-)

Ok, here's what I'm thinking: that vermagic.h is normally automatically included in the *mod.c as part of the module creation, see add_header() in modpost.c.

So then perhaps drivers should not use it directly due to the current inclusion order:

linux/module.h includes asm/module.h and that arch-specific header defines MODULE_VERMAGIC* for the respective arch.

linux/vermagic.h defines all those fallbacks for those MODULE_VERMAGIC* things and if the inclusion order is swapped - we get the redefinition warning.

Yesterday I tried the below - basically get rid of all the remaining includers of linux/vermagic.h but two are left:

drivers/net/ethernet/hisilicon/hns3/hns3_enet.c:18:#include <linux/vermagic.h> drivers/net/ethernet/netronome/nfp/nfp_main.c:17:#include <linux/vermagic.h>

because both use VERMAGIC_STRING directly.

So,

* one could either allow that and sort the inclusion order so that, for example, asm/module.h includes linux/vermagic.h and thus the fallbacks are there present.

* remove all uses of VERMAGIC_STRING from the drivers, add a header guard which prevents people from using it directly and leave VERMAGIC_STRING only to the internal module machinery in the kernel.

Judging by how only a handful of old drivers are even using that, perhaps not too far fetched.

In any case, this needs a maintainer decision.

Leon, if you wanna do it whatever you guys end up agreeing on, just go ahead and submit the patches - it's not like I don't have anything else on the TODO :-) Just add a Reported-by: me and that should be enough.

If you're busy too, lemme know and I'll put it on my todo then.

Thx.

drivers/net/bonding/bonding_priv.h +1 -1

index 45b77bc8c7b3..48cdf3a49a7d 100644
@@ -14,7 +14,7 @@
 
 #ifndef _BONDING_PRIV_H
 #define _BONDING_PRIV_H
-#include <linux/vermagic.h>
+#include <generated/utsrelease.h>
 
 #define DRV_NAME	"bonding"
 #define DRV_DESCRIPTION	"Ethernet Channel Bonding Driver"

drivers/net/ethernet/3com/3c509.c +0 -1

index b762176a1406..139d0120f511 100644
@@ -85,7 +85,6 @@
 #include <linux/device.h>
 #include <linux/eisa.h>
 #include <linux/bitops.h>
-#include <linux/vermagic.h>
 
 #include <linux/uaccess.h>
 #include <asm/io.h>

drivers/net/ethernet/3com/3c515.c +0 -1

index 90312fcd6319..47b4215bb93b 100644
@@ -22,7 +22,6 @@
 
 */
 
-#include <linux/vermagic.h>
 #define DRV_NAME		"3c515"
 
 #define CORKSCREW 1

drivers/net/ethernet/adaptec/starfire.c +0 -1

index 2db42211329f..a64191fc2af9 100644
@@ -45,7 +45,6 @@
 #include <asm/processor.h>		/* Processor type for cache alignment. */
 #include <linux/uaccess.h>
 #include <asm/io.h>
-#include <linux/vermagic.h>
 
 /*
  * The current frame processor firmware fails to checksum a fragment

drivers/net/ethernet/pensando/ionic/ionic_main.c +1 -1

index 588c62e9add7..3ed150512091 100644
@@ -6,7 +6,7 @@
 #include <linux/module.h>
 #include <linux/netdevice.h>
 #include <linux/utsname.h>
-#include <linux/vermagic.h>
+#include <generated/utsrelease.h>
 
 #include "ionic.h"
 #include "ionic_bus.h"

drivers/power/supply/test_power.c +1 -1

index 65c23ef6408d..b3c05ff05783 100644
@@ -16,7 +16,7 @@
 #include <linux/power_supply.h>
 #include <linux/errno.h>
 #include <linux/delay.h>
-#include <linux/vermagic.h>
+#include <generated/utsrelease.h>
 
 enum test_power_id {
 	TEST_AC,

net/ethtool/ioctl.c +2 -1

index 89d0b1827aaf..adab97e500cf 100644
@@ -17,7 +17,6 @@
 #include <linux/phy.h>
 #include <linux/bitops.h>
 #include <linux/uaccess.h>
-#include <linux/vermagic.h>
 #include <linux/vmalloc.h>
 #include <linux/sfp.h>
 #include <linux/slab.h>

@@ -29,6 +28,8 @@
 #include <net/flow_offload.h>
 #include <linux/ethtool_netlink.h>
 
+#include <generated/utsrelease.h>
+
 #include "common.h"
 
 /*

Signature · 4 lines

Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Leon Romanovsky

author

2020-04-13 08:40 UTC #15 lore raw

Re: [PATCH] net/3com/3c515: Fix MODULE_ARCH_VERMAGIC redefinition

Quoted 42 lines · Borislav Petkov

Quoted 6 lines · David Miller
From: Leon Romanovsky <[email protected]>
Date: Mon, 13 Apr 2020 08:26:37 +0300
Quoted 1 line
How do you want us to handle it? Boris resend, me to send, you to fix?
Anyone other than me can do it ;-)
Ok, here's what I'm thinking: that vermagic.h is normally automatically included in the *mod.c as part of the module creation, see add_header() in modpost.c.
So then perhaps drivers should not use it directly due to the current inclusion order:
linux/module.h includes asm/module.h and that arch-specific header defines MODULE_VERMAGIC* for the respective arch.
linux/vermagic.h defines all those fallbacks for those MODULE_VERMAGIC* things and if the inclusion order is swapped - we get the redefinition warning.
Yesterday I tried the below - basically get rid of all the remaining includers of linux/vermagic.h but two are left:
drivers/net/ethernet/hisilicon/hns3/hns3_enet.c:18:#include <linux/vermagic.h> drivers/net/ethernet/netronome/nfp/nfp_main.c:17:#include <linux/vermagic.h>
because both use VERMAGIC_STRING directly.
So,
* one could either allow that and sort the inclusion order so that, for example, asm/module.h includes linux/vermagic.h and thus the fallbacks are there present.
or
* remove all uses of VERMAGIC_STRING from the drivers, add a header guard which prevents people from using it directly and leave VERMAGIC_STRING only to the internal module machinery in the kernel.

Quoted 8 lines

Judging by how only a handful of old drivers are even using that, perhaps not too far fetched.
In any case, this needs a maintainer decision.
Leon, if you wanna do it whatever you guys end up agreeing on, just go ahead and submit the patches - it's not like I don't have anything else on the TODO :-) Just add a Reported-by: me and that should be enough.

I broke it so I should fix and will send a patch today/tomorrow.

Thanks

Quoted 108 lines

If you're busy too, lemme know and I'll put it on my todo then.

Thx.

drivers/net/bonding/bonding_priv.h +1 -1

index 45b77bc8c7b3..48cdf3a49a7d 100644
@@ -14,7 +14,7 @@
 
 #ifndef _BONDING_PRIV_H
 #define _BONDING_PRIV_H
-#include <linux/vermagic.h>
+#include <generated/utsrelease.h>
 
 #define DRV_NAME	"bonding"
 #define DRV_DESCRIPTION	"Ethernet Channel Bonding Driver"

drivers/net/ethernet/3com/3c509.c +0 -1

index b762176a1406..139d0120f511 100644
@@ -85,7 +85,6 @@
 #include <linux/device.h>
 #include <linux/eisa.h>
 #include <linux/bitops.h>
-#include <linux/vermagic.h>
 
 #include <linux/uaccess.h>
 #include <asm/io.h>

drivers/net/ethernet/3com/3c515.c +0 -1

index 90312fcd6319..47b4215bb93b 100644
@@ -22,7 +22,6 @@
 
 */
 
-#include <linux/vermagic.h>
 #define DRV_NAME		"3c515"
 
 #define CORKSCREW 1

drivers/net/ethernet/adaptec/starfire.c +0 -1

index 2db42211329f..a64191fc2af9 100644
@@ -45,7 +45,6 @@
 #include <asm/processor.h>		/* Processor type for cache alignment. */
 #include <linux/uaccess.h>
 #include <asm/io.h>
-#include <linux/vermagic.h>
 
 /*
  * The current frame processor firmware fails to checksum a fragment

drivers/net/ethernet/pensando/ionic/ionic_main.c +1 -1

index 588c62e9add7..3ed150512091 100644
@@ -6,7 +6,7 @@
 #include <linux/module.h>
 #include <linux/netdevice.h>
 #include <linux/utsname.h>
-#include <linux/vermagic.h>
+#include <generated/utsrelease.h>
 
 #include "ionic.h"
 #include "ionic_bus.h"

drivers/power/supply/test_power.c +1 -1

index 65c23ef6408d..b3c05ff05783 100644
@@ -16,7 +16,7 @@
 #include <linux/power_supply.h>
 #include <linux/errno.h>
 #include <linux/delay.h>
-#include <linux/vermagic.h>
+#include <generated/utsrelease.h>
 
 enum test_power_id {
 	TEST_AC,

net/ethtool/ioctl.c +2 -2

index 89d0b1827aaf..adab97e500cf 100644
@@ -17,7 +17,6 @@
 #include <linux/phy.h>
 #include <linux/bitops.h>
 #include <linux/uaccess.h>
-#include <linux/vermagic.h>
 #include <linux/vmalloc.h>
 #include <linux/sfp.h>
 #include <linux/slab.h>

@@ -29,6 +28,8 @@
 #include <net/flow_offload.h>
 #include <linux/ethtool_netlink.h>
 
+#include <generated/utsrelease.h>
+
 #include "common.h"
 
 /*
 
 
--

Regards/Gruss,
Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Leon Romanovsky

author

2020-04-14 12:23 UTC #16 lore raw

Re: [PATCH] net/3com/3c515: Fix MODULE_ARCH_VERMAGIC redefinition

Quoted 46 lines · Leon Romanovsky

Quoted 45 lines · Borislav Petkov
Quoted 6 lines · David Miller
From: Leon Romanovsky <[email protected]>
Date: Mon, 13 Apr 2020 08:26:37 +0300
Quoted 1 line
How do you want us to handle it? Boris resend, me to send, you to fix?
Anyone other than me can do it ;-)
Ok, here's what I'm thinking: that vermagic.h is normally automatically included in the *mod.c as part of the module creation, see add_header() in modpost.c.
So then perhaps drivers should not use it directly due to the current inclusion order:
linux/module.h includes asm/module.h and that arch-specific header defines MODULE_VERMAGIC* for the respective arch.
linux/vermagic.h defines all those fallbacks for those MODULE_VERMAGIC* things and if the inclusion order is swapped - we get the redefinition warning.
Yesterday I tried the below - basically get rid of all the remaining includers of linux/vermagic.h but two are left:
drivers/net/ethernet/hisilicon/hns3/hns3_enet.c:18:#include <linux/vermagic.h> drivers/net/ethernet/netronome/nfp/nfp_main.c:17:#include <linux/vermagic.h>
because both use VERMAGIC_STRING directly.
So,
* one could either allow that and sort the inclusion order so that, for example, asm/module.h includes linux/vermagic.h and thus the fallbacks are there present.
or
* remove all uses of VERMAGIC_STRING from the drivers, add a header guard which prevents people from using it directly and leave VERMAGIC_STRING only to the internal module machinery in the kernel.
Judging by how only a handful of old drivers are even using that, perhaps not too far fetched.

Quoted 7 lines

Quoted 5 lines
In any case, this needs a maintainer decision.
Leon, if you wanna do it whatever you guys end up agreeing on, just go ahead and submit the patches - it's not like I don't have anything else on the TODO :-) Just add a Reported-by: me and that should be enough.
I broke it so I should fix and will send a patch today/tomorrow.

ok, the patches are here, will wait for the kbuild results and will send.
https://git.kernel.org/pub/scm/linux/kernel/git/leon/linux-rdma.git/log/?h=vermagic

Thanks

Quoted 111 lines

Thanks

Quoted 108 lines

If you're busy too, lemme know and I'll put it on my todo then.

Thx.

drivers/net/bonding/bonding_priv.h +1 -1

index 45b77bc8c7b3..48cdf3a49a7d 100644
@@ -14,7 +14,7 @@
 
 #ifndef _BONDING_PRIV_H
 #define _BONDING_PRIV_H
-#include <linux/vermagic.h>
+#include <generated/utsrelease.h>
 
 #define DRV_NAME	"bonding"
 #define DRV_DESCRIPTION	"Ethernet Channel Bonding Driver"

drivers/net/ethernet/3com/3c509.c +0 -1

index b762176a1406..139d0120f511 100644
@@ -85,7 +85,6 @@
 #include <linux/device.h>
 #include <linux/eisa.h>
 #include <linux/bitops.h>
-#include <linux/vermagic.h>
 
 #include <linux/uaccess.h>
 #include <asm/io.h>

drivers/net/ethernet/3com/3c515.c +0 -1

index 90312fcd6319..47b4215bb93b 100644
@@ -22,7 +22,6 @@
 
 */
 
-#include <linux/vermagic.h>
 #define DRV_NAME		"3c515"
 
 #define CORKSCREW 1

drivers/net/ethernet/adaptec/starfire.c +0 -1

index 2db42211329f..a64191fc2af9 100644
@@ -45,7 +45,6 @@
 #include <asm/processor.h>		/* Processor type for cache alignment. */
 #include <linux/uaccess.h>
 #include <asm/io.h>
-#include <linux/vermagic.h>
 
 /*
  * The current frame processor firmware fails to checksum a fragment

drivers/net/ethernet/pensando/ionic/ionic_main.c +1 -1

index 588c62e9add7..3ed150512091 100644
@@ -6,7 +6,7 @@
 #include <linux/module.h>
 #include <linux/netdevice.h>
 #include <linux/utsname.h>
-#include <linux/vermagic.h>
+#include <generated/utsrelease.h>
 
 #include "ionic.h"
 #include "ionic_bus.h"

drivers/power/supply/test_power.c +1 -1

index 65c23ef6408d..b3c05ff05783 100644
@@ -16,7 +16,7 @@
 #include <linux/power_supply.h>
 #include <linux/errno.h>
 #include <linux/delay.h>
-#include <linux/vermagic.h>
+#include <generated/utsrelease.h>
 
 enum test_power_id {
 	TEST_AC,

net/ethtool/ioctl.c +2 -2

index 89d0b1827aaf..adab97e500cf 100644
@@ -17,7 +17,6 @@
 #include <linux/phy.h>
 #include <linux/bitops.h>
 #include <linux/uaccess.h>
-#include <linux/vermagic.h>
 #include <linux/vmalloc.h>
 #include <linux/sfp.h>
 #include <linux/slab.h>

@@ -29,6 +28,8 @@
 #include <net/flow_offload.h>
 #include <linux/ethtool_netlink.h>
 
+#include <generated/utsrelease.h>
+
 #include "common.h"
 
 /*
 
 
--

Regards/Gruss,
Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Leon Romanovsky

author

2020-02-24 08:52 UTC #17 lore raw

PATCH 4/18 net/adaptec: Clean driver versions

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

Delete useless driver version in favor of default ones.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/ethernet/adaptec/starfire.c | 19 +------------------
 1 file changed, 1 insertion(+), 18 deletions(-)

drivers/net/ethernet/adaptec/starfire.c +1 ~~-18~~

index 165d18405b0c..2db42211329f 100644
@@ -27,8 +27,6 @@
 */
 
 #define DRV_NAME	"starfire"
-#define DRV_VERSION	"2.1"
-#define DRV_RELDATE	"July  6, 2008"
 
 #include <linux/interrupt.h>
 #include <linux/module.h>

@@ -47,6 +45,7 @@
 #include <asm/processor.h>		/* Processor type for cache alignment. */
 #include <linux/uaccess.h>
 #include <asm/io.h>
+#include <linux/vermagic.h>
 
 /*
  * The current frame processor firmware fails to checksum a fragment

@@ -165,15 +164,9 @@ static int rx_copybreak /* = 0 */;
 #define FIRMWARE_RX	"adaptec/starfire_rx.bin"
 #define FIRMWARE_TX	"adaptec/starfire_tx.bin"
 
-/* These identify the driver base version and may not be removed. */
-static const char version[] =
-KERN_INFO "starfire.c:v1.03 7/26/2000  Written by Donald Becker <[email protected]>\n"
-" (unofficial 2.2/2.4 kernel port, version " DRV_VERSION ", " DRV_RELDATE ")\n";
-
 MODULE_AUTHOR("Donald Becker <[email protected]>");
 MODULE_DESCRIPTION("Adaptec Starfire Ethernet driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
 MODULE_FIRMWARE(FIRMWARE_RX);
 MODULE_FIRMWARE(FIRMWARE_TX);

@@ -654,13 +647,6 @@ static int starfire_init_one(struct pci_dev *pdev,
 	int drv_flags, io_size;
 	int boguscnt;
 
-/* when built into the kernel, we only print version if device is found */
-#ifndef MODULE
-	static int printed_version;
-	if (!printed_version++)
-		printk(version);
-#endif
-
 	if (pci_enable_device (pdev))
 		return -EIO;

@@ -1853,7 +1839,6 @@ static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 {
 	struct netdev_private *np = netdev_priv(dev);
 	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
 	strlcpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info));
 }

@@ -2073,8 +2058,6 @@ static int __init starfire_init (void)
 {
 /* when a module, this is printed whether or not devices are found in probe */
 #ifdef MODULE
-	printk(version);
-
 	printk(KERN_INFO DRV_NAME ": polling (NAPI) enabled\n");
 #endif

--
2.24.1

Leon Romanovsky

author

2020-02-24 08:52 UTC #18 lore raw

PATCH 5/18 net/aeroflex: Clean ethtool_info struct assignments

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

If FW version is not available, it is enough to leave that field as empty, there is no need to write N/A.

The driver version is replaced in favor of generally available in-tree variant.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/ethernet/aeroflex/greth.c | 1 -
 1 file changed, 1 deletion(-)

drivers/net/ethernet/aeroflex/greth.c +0 -1

index 2a9f8643629c..198f1544e271 100644
@@ -1114,7 +1114,6 @@ static void greth_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *in
 
 	strlcpy(info->driver, dev_driver_string(greth->dev),
 		sizeof(info->driver));
-	strlcpy(info->version, "revision: 1.0", sizeof(info->version));
 	strlcpy(info->bus_info, greth->dev->bus->name, sizeof(info->bus_info));
 	strlcpy(info->fw_version, "N/A", sizeof(info->fw_version));
 }

--
2.24.1

Leon Romanovsky

author

2020-02-24 08:52 UTC #19 lore raw

PATCH 6/18 net/aeroflex: Don't assign FW if it is not available

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

If FW version is not available, it is enough to leave that field as empty, there is no need to write N/A.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/ethernet/aeroflex/greth.c | 1 -
 1 file changed, 1 deletion(-)

drivers/net/ethernet/aeroflex/greth.c +0 -1

index 198f1544e271..bf546118dbc6 100644
@@ -1115,7 +1115,6 @@ static void greth_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *in
 	strlcpy(info->driver, dev_driver_string(greth->dev),
 		sizeof(info->driver));
 	strlcpy(info->bus_info, greth->dev->bus->name, sizeof(info->bus_info));
-	strlcpy(info->fw_version, "N/A", sizeof(info->fw_version));
 }
 
 static void greth_get_regs(struct net_device *dev, struct ethtool_regs *regs, void *p)

Leon Romanovsky

author

2020-02-24 08:53 UTC #20 lore raw

PATCH 7/18 net/agere: Delete unneeded driver version

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

There is no need in driver version for in-tree kernel code.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/ethernet/agere/et131x.c | 1 -
 drivers/net/ethernet/agere/et131x.h | 1 -
 2 files changed, 2 deletions(-)

drivers/net/ethernet/agere/et131x.c +0 -1

index cb6a761d5c11..1b19385ad8a9 100644
@@ -2958,7 +2958,6 @@ static void et131x_get_drvinfo(struct net_device *netdev,
 	struct et131x_adapter *adapter = netdev_priv(netdev);
 
 	strlcpy(info->driver, DRIVER_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRIVER_VERSION, sizeof(info->version));
 	strlcpy(info->bus_info, pci_name(adapter->pdev),
 		sizeof(info->bus_info));
 }

drivers/net/ethernet/agere/et131x.h +0 -1

index be9a11c02526..d0e922584d8a 100644
@@ -46,7 +46,6 @@
  */
 
 #define DRIVER_NAME "et131x"
-#define DRIVER_VERSION "v2.0"
 
 /* EEPROM registers */

--
2.24.1

Leon Romanovsky

author

2020-02-24 08:53 UTC #21 lore raw

PATCH 8/18 net/alacritech: Delete driver version

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

Use standard variant of the driver version.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/ethernet/alacritech/slicoss.c | 3 ---
 1 file changed, 3 deletions(-)

drivers/net/ethernet/alacritech/slicoss.c +0 -3

index 9daef4c8feef..6234fcd844ee 100644
@@ -26,7 +26,6 @@
 #include "slic.h"
 
 #define DRV_NAME			"slicoss"
-#define DRV_VERSION			"1.0"
 
 static const struct pci_device_id slic_id_tbl[] = {
 	{ PCI_DEVICE(PCI_VENDOR_ID_ALACRITECH,

@@ -1533,7 +1532,6 @@ static void slic_get_drvinfo(struct net_device *dev,
 	struct slic_device *sdev = netdev_priv(dev);
 
 	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
 	strlcpy(info->bus_info, pci_name(sdev->pdev), sizeof(info->bus_info));
 }

@@ -1852,4 +1850,3 @@ module_pci_driver(slic_driver);
 MODULE_DESCRIPTION("Alacritech non-accelerated SLIC driver");
 MODULE_AUTHOR("Lino Sanfilippo <[email protected]>");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);

Leon Romanovsky

author

2020-02-24 08:53 UTC #22 lore raw

PATCH 9/18 net/allwinner: Remove driver version

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

There is no need in custom driver version for in-tree code.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/ethernet/allwinner/sun4i-emac.c | 2 --
 1 file changed, 2 deletions(-)

drivers/net/ethernet/allwinner/sun4i-emac.c +0 -2

index 22cadfbeedfb..18d3b4340bd4 100644
@@ -33,7 +33,6 @@
 #include "sun4i-emac.h"
 
 #define DRV_NAME		"sun4i-emac"
-#define DRV_VERSION		"1.02"
 
 #define EMAC_MAX_FRAME_LEN	0x0600

@@ -212,7 +211,6 @@ static void emac_get_drvinfo(struct net_device *dev,
 			      struct ethtool_drvinfo *info)
 {
 	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
 	strlcpy(info->bus_info, dev_name(&dev->dev), sizeof(info->bus_info));
 }

Maxime Ripard

Acked-by on v1 9/18

2020-02-24 09:43 UTC #23 lore raw

Re: [PATCH net-next v1 09/18] net/allwinner: Remove driver version

Quoted 5 lines · Leon Romanovsky

From: Leon Romanovsky <[email protected]>
There is no need in custom driver version for in-tree code.
Signed-off-by Leon Romanovsky <[email protected]>

Acked-by Maxime Ripard <[email protected]>

Thanks!
Maxime

Leon Romanovsky

author

2020-02-24 08:53 UTC #24 lore raw

PATCH 10/18 net/alteon: Properly report FW version

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

The acenic driver assigns FW version in driver version field, as part of cleanup driver version, set FW version properly.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/ethernet/alteon/acenic.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

drivers/net/ethernet/alteon/acenic.c +2 -3

index f366faf88eee..5d192d551623 100644
@@ -2699,9 +2699,8 @@ static void ace_get_drvinfo(struct net_device *dev,
 	struct ace_private *ap = netdev_priv(dev);
 
 	strlcpy(info->driver, "acenic", sizeof(info->driver));
-	snprintf(info->version, sizeof(info->version), "%i.%i.%i",
-		 ap->firmware_major, ap->firmware_minor,
-		 ap->firmware_fix);
+	snprintf(info->fw_version, sizeof(info->version), "%i.%i.%i",
+		 ap->firmware_major, ap->firmware_minor, ap->firmware_fix);
 
 	if (ap->pdev)
 		strlcpy(info->bus_info, pci_name(ap->pdev),

--
2.24.1

Leon Romanovsky

author

2020-02-24 08:53 UTC #25 lore raw

PATCH 11/18 net/althera: Delete hardcoded driver version

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

Convert to use default version provided by ethtool.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/ethernet/altera/altera_tse_ethtool.c | 1 -
 1 file changed, 1 deletion(-)

drivers/net/ethernet/altera/altera_tse_ethtool.c +0 -1

index 23823464f2e7..4299f1301149 100644
@@ -67,7 +67,6 @@ static void tse_get_drvinfo(struct net_device *dev,
 	u32 rev = ioread32(&priv->mac_dev->megacore_revision);
 
 	strcpy(info->driver, "altera_tse");
-	strcpy(info->version, "v8.0");
 	snprintf(info->fw_version, ETHTOOL_FWVERS_LEN, "v%d.%d",
 		 rev & 0xFFFF, (rev & 0xFFFF0000) >> 16);
 	sprintf(info->bus_info, "platform");

Leon Romanovsky

author

2020-02-24 08:53 UTC #26 lore raw

PATCH 12/18 net/amazon: Ensure that driver version is aligned to the linux kernel

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

Upstream drivers are managed inside global repository and released all together, this ensure that driver version is the same as linux kernel, so update amazon drivers to properly reflect it.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/ethernet/amazon/ena/ena_ethtool.c |  1 -
 drivers/net/ethernet/amazon/ena/ena_netdev.c  | 17 ++---------------
 drivers/net/ethernet/amazon/ena/ena_netdev.h  | 11 -----------
 3 files changed, 2 insertions(+), 27 deletions(-)

drivers/net/ethernet/amazon/ena/ena_ethtool.c +0 -1

index ced1d577b62a..19262f37db84 100644
@@ -404,7 +404,6 @@ static void ena_get_drvinfo(struct net_device *dev,
 	struct ena_adapter *adapter = netdev_priv(dev);
 
 	strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version));
 	strlcpy(info->bus_info, pci_name(adapter->pdev),
 		sizeof(info->bus_info));
 }

drivers/net/ethernet/amazon/ena/ena_netdev.c +2 ~~-15~~

index 0b2fd96b93d7..4faf81c456d8 100644
@@ -49,12 +49,9 @@
 #include <linux/bpf_trace.h>
 #include "ena_pci_id_tbl.h"
 
-static char version[] = DEVICE_NAME " v" DRV_MODULE_VERSION "\n";
-
 MODULE_AUTHOR("Amazon.com, Inc. or its affiliates");
 MODULE_DESCRIPTION(DEVICE_NAME);
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_MODULE_VERSION);
 
 /* Time in jiffies before concluding the transmitter is hung. */
 #define TX_TIMEOUT  (5 * HZ)

@@ -3093,11 +3090,7 @@ static void ena_config_host_info(struct ena_com_dev *ena_dev,
 	host_info->os_dist = 0;
 	strncpy(host_info->os_dist_str, utsname()->release,
 		sizeof(host_info->os_dist_str) - 1);
-	host_info->driver_version =
-		(DRV_MODULE_VER_MAJOR) |
-		(DRV_MODULE_VER_MINOR << ENA_ADMIN_HOST_INFO_MINOR_SHIFT) |
-		(DRV_MODULE_VER_SUBMINOR << ENA_ADMIN_HOST_INFO_SUB_MINOR_SHIFT) |
-		("K"[0] << ENA_ADMIN_HOST_INFO_MODULE_TYPE_SHIFT);
+	host_info->driver_version = LINUX_VERSION_CODE;
 	host_info->num_cpus = num_online_cpus();
 
 	host_info->driver_supported_features =

@@ -3476,9 +3469,7 @@ static int ena_restore_device(struct ena_adapter *adapter)
 		netif_carrier_on(adapter->netdev);
 
 	mod_timer(&adapter->timer_service, round_jiffies(jiffies + HZ));
-	dev_err(&pdev->dev,
-		"Device reset completed successfully, Driver info: %s\n",
-		version);
+	dev_err(&pdev->dev, "Device reset completed successfully\n");
 
 	return rc;
 err_disable_msix:

@@ -4116,8 +4107,6 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	dev_dbg(&pdev->dev, "%s\n", __func__);
 
-	dev_info_once(&pdev->dev, "%s", version);
-
 	rc = pci_enable_device_mem(pdev);
 	if (rc) {
 		dev_err(&pdev->dev, "pci_enable_device_mem() failed!\n");

@@ -4429,8 +4418,6 @@ static struct pci_driver ena_pci_driver = {
 
 static int __init ena_init(void)
 {
-	pr_info("%s", version);
-
 	ena_wq = create_singlethread_workqueue(DRV_MODULE_NAME);
 	if (!ena_wq) {
 		pr_err("Failed to create workqueue\n");

drivers/net/ethernet/amazon/ena/ena_netdev.h +0 ~~-11~~

index 8795e0b1dc3c..74c7f10b60dd 100644
@@ -45,18 +45,7 @@
 #include "ena_com.h"
 #include "ena_eth_com.h"
 
-#define DRV_MODULE_VER_MAJOR	2
-#define DRV_MODULE_VER_MINOR	1
-#define DRV_MODULE_VER_SUBMINOR 0
-
 #define DRV_MODULE_NAME		"ena"
-#ifndef DRV_MODULE_VERSION
-#define DRV_MODULE_VERSION \
-	__stringify(DRV_MODULE_VER_MAJOR) "."	\
-	__stringify(DRV_MODULE_VER_MINOR) "."	\
-	__stringify(DRV_MODULE_VER_SUBMINOR) "K"
-#endif
-
 #define DEVICE_NAME	"Elastic Network Adapter (ENA)"
 
 /* 1 for AENQ + ADMIN */

--
2.24.1

Kiyanovski, Arthur

maintainer

2020-02-24 09:03 UTC #27 lore raw

RE: [PATCH net-next v1 12/18] net/amazon: Ensure that driver version is aligned to the linux kernel

Quoted 140 lines

-----Original Message-----
From: Leon Romanovsky <[email protected]>
Sent: Monday, February 24, 2020 10:53 AM
To: David S. Miller <[email protected]>; Jakub Kicinski <[email protected]>
Cc: Leon Romanovsky <[email protected]>; Tom Lendacky <[email protected]>; Keyur Chudgar <[email protected]>; Don Fry <[email protected]>; Veaceslav Falico <[email protected]>; Jay Vosburgh <[email protected]>; [email protected]; Maxime Ripard <[email protected]>; Heiko Stuebner <[email protected]>; Mark Einon <[email protected]>; Chris Snook <[email protected]>; [email protected]; Iyappan Subramanian <[email protected]>; Igor Russkikh <[email protected]>; David Dillow <[email protected]>; Belgazal, Netanel <[email protected]>; Quan Nguyen <[email protected]>; Jay Cliburn <[email protected]>; Lino Sanfilippo <[email protected]>; [email protected]; Andreas Larsson <[email protected]>; Andy Gospodarek <[email protected]>; [email protected]; Thor Thayer <[email protected]>; [email protected]; Ion Badulescu <[email protected]>; Kiyanovski, Arthur <[email protected]>; Jes Sorensen <[email protected]>; [email protected]; Chen-Yu Tsai <[email protected]>
Subject: [PATCH net-next v1 12/18] net/amazon: Ensure that driver version is aligned to the linux kernel
From: Leon Romanovsky <[email protected]>
Upstream drivers are managed inside global repository and released all together, this ensure that driver version is the same as linux kernel, so update amazon drivers to properly reflect it.
Signed-off-by Leon Romanovsky <[email protected]>
---
 drivers/net/ethernet/amazon/ena/ena_ethtool.c |  1 -
drivers/net/ethernet/amazon/ena/ena_netdev.c  | 17 ++---------------
drivers/net/ethernet/amazon/ena/ena_netdev.h  | 11 -----------
 3 files changed, 2 insertions(+), 27 deletions(-)
drivers/net/ethernet/amazon/ena/ena_ethtool.c +0 -1
index ced1d577b62a..19262f37db84 100644
@@ -404,7 +404,6 @@ static void ena_get_drvinfo(struct net_device *dev,
 	struct ena_adapter *adapter = netdev_priv(dev);
 
 	strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version));
 	strlcpy(info->bus_info, pci_name(adapter->pdev),
 		sizeof(info->bus_info));
 }
drivers/net/ethernet/amazon/ena/ena_netdev.c +0 -2
index 0b2fd96b93d7..4faf81c456d8 100644
@@ -49,12 +49,9 @@
 #include <linux/bpf_trace.h>
 #include "ena_pci_id_tbl.h"
 
-static char version[] = DEVICE_NAME " v" DRV_MODULE_VERSION "\n";
-
 MODULE_AUTHOR("Amazon.com, Inc. or its affiliates");
MODULE_DESCRIPTION(DEVICE_NAME); MODULE_LICENSE("GPL"); - MODULE_VERSION(DRV_MODULE_VERSION);
/* Time in jiffies before concluding the transmitter is hung. */ #define
TX_TIMEOUT (5 * HZ) @@ -3093,11 +3090,7 @@ static void ena_config_host_info(struct ena_com_dev *ena_dev,
drivers/net/ethernet/amazon/ena/ena_netdev.c +0 -3
 	host_info->os_dist = 0;
 	strncpy(host_info->os_dist_str, utsname()->release,
 		sizeof(host_info->os_dist_str) - 1);
-	host_info->driver_version =
-		(DRV_MODULE_VER_MAJOR) |
-		(DRV_MODULE_VER_MINOR <<
ENA_ADMIN_HOST_INFO_MINOR_SHIFT) |
drivers/net/ethernet/amazon/ena/ena_netdev.c +0 -1
-		(DRV_MODULE_VER_SUBMINOR <<
ENA_ADMIN_HOST_INFO_SUB_MINOR_SHIFT) |
drivers/net/ethernet/amazon/ena/ena_netdev.c +1 -1
-		("K"[0] << ENA_ADMIN_HOST_INFO_MODULE_TYPE_SHIFT);
+	host_info->driver_version = LINUX_VERSION_CODE;
 	host_info->num_cpus = num_online_cpus();
 
 	host_info->driver_supported_features = @@ -3476,9 +3469,7 @@
static int ena_restore_device(struct ena_adapter *adapter)
drivers/net/ethernet/amazon/ena/ena_netdev.c +1 -5
 		netif_carrier_on(adapter->netdev);
 
 	mod_timer(&adapter->timer_service, round_jiffies(jiffies + HZ));
-	dev_err(&pdev->dev,
-		"Device reset completed successfully, Driver info: %s\n",
-		version);
+	dev_err(&pdev->dev, "Device reset completed successfully\n");
 
 	return rc;
 err_disable_msix:
@@ -4116,8 +4107,6 @@ static int ena_probe(struct pci_dev *pdev, const
 struct pci_device_id *ent)
 
 	dev_dbg(&pdev->dev, "%s\n", __func__);
 
-	dev_info_once(&pdev->dev, "%s", version);
-
 	rc = pci_enable_device_mem(pdev);
 	if (rc) {
drivers/net/ethernet/amazon/ena/ena_netdev.c +0 -2
 		dev_err(&pdev->dev, "pci_enable_device_mem() failed!\n");
@@ -4429,8 +4418,6 @@ static struct pci_driver ena_pci_driver = {
 
 static int __init ena_init(void)
 {
-	pr_info("%s", version);
-
 	ena_wq = create_singlethread_workqueue(DRV_MODULE_NAME);
 	if (!ena_wq) {
 		pr_err("Failed to create workqueue\n"); diff --git
a/drivers/net/ethernet/amazon/ena/ena_netdev.h
b/drivers/net/ethernet/amazon/ena/ena_netdev.h
index 8795e0b1dc3c..74c7f10b60dd 100644
drivers/net/ethernet/amazon/ena/ena_netdev.h +0 ~~-12~~
@@ -45,18 +45,7 @@
 #include "ena_com.h"
 #include "ena_eth_com.h"
 
-#define DRV_MODULE_VER_MAJOR	2
-#define DRV_MODULE_VER_MINOR	1
-#define DRV_MODULE_VER_SUBMINOR 0
-
 #define DRV_MODULE_NAME		"ena"
-#ifndef DRV_MODULE_VERSION
-#define DRV_MODULE_VERSION \
-	__stringify(DRV_MODULE_VER_MAJOR) "."	\
-	__stringify(DRV_MODULE_VER_MINOR) "."	\
-	__stringify(DRV_MODULE_VER_SUBMINOR) "K"
-#endif
-
 #define DEVICE_NAME	"Elastic Network Adapter (ENA)"
 
 /* 1 for AENQ + ADMIN */
--

Quoted 1 line

2.24.1

Hi Leon, David,

This patch is not good for the ENA driver as it breaks the interface with the FW of the ENA device in ena_config_host_info(), host_info is later reported to the FW. Please do not merge it yet.
We are now working on altering your patch so that it won't break this interface and will send it to you in the next few hours.

Thanks,
Arthur

Leon Romanovsky

author

2020-02-24 09:41 UTC #28 lore raw

Re: [PATCH net-next v1 12/18] net/amazon: Ensure that driver version is aligned to the linux kernel

Quoted 138 lines · Arthur

Quoted 138 lines
-----Original Message-----
From: Leon Romanovsky <[email protected]>
Sent: Monday, February 24, 2020 10:53 AM
To: David S. Miller <[email protected]>; Jakub Kicinski <[email protected]>
Cc: Leon Romanovsky <[email protected]>; Tom Lendacky <[email protected]>; Keyur Chudgar <[email protected]>; Don Fry <[email protected]>; Veaceslav Falico <[email protected]>; Jay Vosburgh <[email protected]>; [email protected]; Maxime Ripard <[email protected]>; Heiko Stuebner <[email protected]>; Mark Einon <[email protected]>; Chris Snook <[email protected]>; [email protected]; Iyappan Subramanian <[email protected]>; Igor Russkikh <[email protected]>; David Dillow <[email protected]>; Belgazal, Netanel <[email protected]>; Quan Nguyen <[email protected]>; Jay Cliburn <[email protected]>; Lino Sanfilippo <[email protected]>; [email protected]; Andreas Larsson <[email protected]>; Andy Gospodarek <[email protected]>; [email protected]; Thor Thayer <[email protected]>; [email protected]; Ion Badulescu <[email protected]>; Kiyanovski, Arthur <[email protected]>; Jes Sorensen <[email protected]>; [email protected]; Chen-Yu Tsai <[email protected]>
Subject: [PATCH net-next v1 12/18] net/amazon: Ensure that driver version is aligned to the linux kernel
From: Leon Romanovsky <[email protected]>
Upstream drivers are managed inside global repository and released all together, this ensure that driver version is the same as linux kernel, so update amazon drivers to properly reflect it.
Signed-off-by Leon Romanovsky <[email protected]>
---
 drivers/net/ethernet/amazon/ena/ena_ethtool.c |  1 -
drivers/net/ethernet/amazon/ena/ena_netdev.c  | 17 ++---------------
drivers/net/ethernet/amazon/ena/ena_netdev.h  | 11 -----------
 3 files changed, 2 insertions(+), 27 deletions(-)
drivers/net/ethernet/amazon/ena/ena_ethtool.c +0 -1
index ced1d577b62a..19262f37db84 100644
@@ -404,7 +404,6 @@ static void ena_get_drvinfo(struct net_device *dev,
 	struct ena_adapter *adapter = netdev_priv(dev);
 
 	strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version));
 	strlcpy(info->bus_info, pci_name(adapter->pdev),
 		sizeof(info->bus_info));
 }
drivers/net/ethernet/amazon/ena/ena_netdev.c +0 -2
index 0b2fd96b93d7..4faf81c456d8 100644
@@ -49,12 +49,9 @@
 #include <linux/bpf_trace.h>
 #include "ena_pci_id_tbl.h"
 
-static char version[] = DEVICE_NAME " v" DRV_MODULE_VERSION "\n";
-
 MODULE_AUTHOR("Amazon.com, Inc. or its affiliates");
MODULE_DESCRIPTION(DEVICE_NAME); MODULE_LICENSE("GPL"); - MODULE_VERSION(DRV_MODULE_VERSION);
/* Time in jiffies before concluding the transmitter is hung. */ #define
TX_TIMEOUT (5 * HZ) @@ -3093,11 +3090,7 @@ static void ena_config_host_info(struct ena_com_dev *ena_dev,
drivers/net/ethernet/amazon/ena/ena_netdev.c +0 -3
 	host_info->os_dist = 0;
 	strncpy(host_info->os_dist_str, utsname()->release,
 		sizeof(host_info->os_dist_str) - 1);
-	host_info->driver_version =
-		(DRV_MODULE_VER_MAJOR) |
-		(DRV_MODULE_VER_MINOR <<
ENA_ADMIN_HOST_INFO_MINOR_SHIFT) |
drivers/net/ethernet/amazon/ena/ena_netdev.c +0 -1
-		(DRV_MODULE_VER_SUBMINOR <<
ENA_ADMIN_HOST_INFO_SUB_MINOR_SHIFT) |
drivers/net/ethernet/amazon/ena/ena_netdev.c +1 -1
-		("K"[0] << ENA_ADMIN_HOST_INFO_MODULE_TYPE_SHIFT);
+	host_info->driver_version = LINUX_VERSION_CODE;
 	host_info->num_cpus = num_online_cpus();
 
 	host_info->driver_supported_features = @@ -3476,9 +3469,7 @@
static int ena_restore_device(struct ena_adapter *adapter)
drivers/net/ethernet/amazon/ena/ena_netdev.c +1 -5
 		netif_carrier_on(adapter->netdev);
 
 	mod_timer(&adapter->timer_service, round_jiffies(jiffies + HZ));
-	dev_err(&pdev->dev,
-		"Device reset completed successfully, Driver info: %s\n",
-		version);
+	dev_err(&pdev->dev, "Device reset completed successfully\n");
 
 	return rc;
 err_disable_msix:
@@ -4116,8 +4107,6 @@ static int ena_probe(struct pci_dev *pdev, const
 struct pci_device_id *ent)
 
 	dev_dbg(&pdev->dev, "%s\n", __func__);
 
-	dev_info_once(&pdev->dev, "%s", version);
-
 	rc = pci_enable_device_mem(pdev);
 	if (rc) {
drivers/net/ethernet/amazon/ena/ena_netdev.c +0 -2
 		dev_err(&pdev->dev, "pci_enable_device_mem() failed!\n");
@@ -4429,8 +4418,6 @@ static struct pci_driver ena_pci_driver = {
 
 static int __init ena_init(void)
 {
-	pr_info("%s", version);
-
 	ena_wq = create_singlethread_workqueue(DRV_MODULE_NAME);
 	if (!ena_wq) {
 		pr_err("Failed to create workqueue\n"); diff --git
a/drivers/net/ethernet/amazon/ena/ena_netdev.h
b/drivers/net/ethernet/amazon/ena/ena_netdev.h
index 8795e0b1dc3c..74c7f10b60dd 100644
drivers/net/ethernet/amazon/ena/ena_netdev.h +0 ~~-11~~
@@ -45,18 +45,7 @@
 #include "ena_com.h"
 #include "ena_eth_com.h"
 
-#define DRV_MODULE_VER_MAJOR	2
-#define DRV_MODULE_VER_MINOR	1
-#define DRV_MODULE_VER_SUBMINOR 0
-
 #define DRV_MODULE_NAME		"ena"
-#ifndef DRV_MODULE_VERSION
-#define DRV_MODULE_VERSION \
-	__stringify(DRV_MODULE_VER_MAJOR) "."	\
-	__stringify(DRV_MODULE_VER_MINOR) "."	\
-	__stringify(DRV_MODULE_VER_SUBMINOR) "K"
-#endif
-
 #define DEVICE_NAME	"Elastic Network Adapter (ENA)"
 

Quoted 8 lines

Quoted 3 lines
drivers/net/ethernet/amazon/ena/ena_netdev.h +0 -1
 /* 1 for AENQ + ADMIN */
--
2.24.1
Hi Leon, David,
This patch is not good for the ENA driver as it breaks the interface with the FW of the ENA device in ena_config_host_info(), host_info is later reported to the FW. Please do not merge it yet.

As I wrote here [1], I tried to avoid any changes in SW<->FW interfaces. Can you please show me the dump stack of how is such info forwarded to FW? How do you distinguish between different distro versions and driver releases?

Quoted 1 line

We are now working on altering your patch so that it won't break this interface and will send it to you in the next few hours.

It is good, thanks.

[1] https://lore.kernel.org/netdev/20200223091031.GA422704@unreal

Quoted 4 lines

Thanks,
Arthur

[email protected]

2020-02-24 15:41 UTC #29 lore raw

PATCH 1/1 net/amazon: Ensure that driver version is aligned to the linux kernel

From: Arthur Kiyanovski <[email protected]>

Upstream drivers are managed inside global repository and released all together, this ensure that driver version is the same as linux kernel, so update amazon drivers to properly reflect it.

Also rename current driver version constants used in interface with ENA device FW for code clarity.

Signed-off-by Leon Romanovsky <[email protected]>

Signed-off-by Arthur Kiyanovski <[email protected]>

---
 drivers/net/ethernet/amazon/ena/ena_ethtool.c |  1 -
 drivers/net/ethernet/amazon/ena/ena_netdev.c  | 17 ++++-------------
 drivers/net/ethernet/amazon/ena/ena_netdev.h  | 16 ++++++++--------
 3 files changed, 12 insertions(+), 22 deletions(-)

drivers/net/ethernet/amazon/ena/ena_ethtool.c +0 -1

index 971f02ea5..ad2148b86 100644
@@ -404,7 +404,6 @@ static void ena_get_drvinfo(struct net_device *dev,
 	struct ena_adapter *adapter = netdev_priv(dev);
 
 	strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version));
 	strlcpy(info->bus_info, pci_name(adapter->pdev),
 		sizeof(info->bus_info));
 }

drivers/net/ethernet/amazon/ena/ena_netdev.c +4 ~~-13~~

index 1c1a41bd1..b14772812 100644
@@ -49,12 +49,9 @@
 #include "ena_netdev.h"
 #include "ena_pci_id_tbl.h"
 
-static char version[] = DEVICE_NAME " v" DRV_MODULE_VERSION "\n";
-
 MODULE_AUTHOR("Amazon.com, Inc. or its affiliates");
 MODULE_DESCRIPTION(DEVICE_NAME);
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_MODULE_VERSION);
 
 /* Time in jiffies before concluding the transmitter is hung. */
 #define TX_TIMEOUT  (5 * HZ)

@@ -2441,9 +2438,9 @@ static void ena_config_host_info(struct ena_com_dev *ena_dev,
 	strncpy(host_info->os_dist_str, utsname()->release,
 		sizeof(host_info->os_dist_str) - 1);
 	host_info->driver_version =
-		(DRV_MODULE_VER_MAJOR) |
-		(DRV_MODULE_VER_MINOR << ENA_ADMIN_HOST_INFO_MINOR_SHIFT) |
-		(DRV_MODULE_VER_SUBMINOR << ENA_ADMIN_HOST_INFO_SUB_MINOR_SHIFT) |
+		(DRV_MODULE_GEN_MAJOR) |
+		(DRV_MODULE_GEN_MINOR << ENA_ADMIN_HOST_INFO_MINOR_SHIFT) |
+		(DRV_MODULE_GEN_SUBMINOR << ENA_ADMIN_HOST_INFO_SUB_MINOR_SHIFT) |
 		("K"[0] << ENA_ADMIN_HOST_INFO_MODULE_TYPE_SHIFT);
 	host_info->num_cpus = num_online_cpus();

@@ -2822,9 +2819,7 @@ static int ena_restore_device(struct ena_adapter *adapter)
 		netif_carrier_on(adapter->netdev);
 
 	mod_timer(&adapter->timer_service, round_jiffies(jiffies + HZ));
-	dev_err(&pdev->dev,
-		"Device reset completed successfully, Driver info: %s\n",
-		version);
+	dev_err(&pdev->dev, "Device reset completed successfully\n");
 
 	return rc;
 err_disable_msix:

@@ -3459,8 +3454,6 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	dev_dbg(&pdev->dev, "%s\n", __func__);
 
-	dev_info_once(&pdev->dev, "%s", version);
-
 	rc = pci_enable_device_mem(pdev);
 	if (rc) {
 		dev_err(&pdev->dev, "pci_enable_device_mem() failed!\n");

@@ -3766,8 +3759,6 @@ static struct pci_driver ena_pci_driver = {
 
 static int __init ena_init(void)
 {
-	pr_info("%s", version);
-
 	ena_wq = create_singlethread_workqueue(DRV_MODULE_NAME);
 	if (!ena_wq) {
 		pr_err("Failed to create workqueue\n");

drivers/net/ethernet/amazon/ena/ena_netdev.h +8 -8

index 2fe5eeea6..ac3f481b7 100644
@@ -44,16 +44,16 @@
 #include "ena_com.h"
 #include "ena_eth_com.h"
 
-#define DRV_MODULE_VER_MAJOR	2
-#define DRV_MODULE_VER_MINOR	1
-#define DRV_MODULE_VER_SUBMINOR 0
+#define DRV_MODULE_GEN_MAJOR	2
+#define DRV_MODULE_GEN_MINOR	1
+#define DRV_MODULE_GEN_SUBMINOR 0
 
 #define DRV_MODULE_NAME		"ena"
-#ifndef DRV_MODULE_VERSION
-#define DRV_MODULE_VERSION \
-	__stringify(DRV_MODULE_VER_MAJOR) "."	\
-	__stringify(DRV_MODULE_VER_MINOR) "."	\
-	__stringify(DRV_MODULE_VER_SUBMINOR) "K"
+#ifndef DRV_MODULE_GENERATION
+#define DRV_MODULE_GENERATION \
+	__stringify(DRV_MODULE_GEN_MAJOR) "."	\
+	__stringify(DRV_MODULE_GEN_MINOR) "."	\
+	__stringify(DRV_MODULE_GEN_SUBMINOR) "K"
 #endif
 
 #define DEVICE_NAME	"Elastic Network Adapter (ENA)"

Signature · 1 line

2.24.1.AMZN

Leon Romanovsky

2020-02-24 16:26 UTC #30 lore raw

Re: [PATCH] net/amazon: Ensure that driver version is aligned to the linux kernel

Quoted 93 lines · On Mon, Feb 24, 2020 at 03:40:27PM +0000, [email protected] wrote:

From: Arthur Kiyanovski <[email protected]>

Upstream drivers are managed inside global repository and released all together, this ensure that driver version is the same as linux kernel, so update amazon drivers to properly reflect it.

Also rename current driver version constants used in interface with ENA device FW for code clarity.

Signed-off-by Leon Romanovsky <[email protected]>

Signed-off-by Arthur Kiyanovski <[email protected]>

---
 drivers/net/ethernet/amazon/ena/ena_ethtool.c |  1 -
 drivers/net/ethernet/amazon/ena/ena_netdev.c  | 17 ++++-------------
 drivers/net/ethernet/amazon/ena/ena_netdev.h  | 16 ++++++++--------
 3 files changed, 12 insertions(+), 22 deletions(-)

drivers/net/ethernet/amazon/ena/ena_ethtool.c +0 -1

index 971f02ea5..ad2148b86 100644
@@ -404,7 +404,6 @@ static void ena_get_drvinfo(struct net_device *dev,
 	struct ena_adapter *adapter = netdev_priv(dev);
 
 	strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version));
 	strlcpy(info->bus_info, pci_name(adapter->pdev),
 		sizeof(info->bus_info));
 }

drivers/net/ethernet/amazon/ena/ena_netdev.c +4 ~~-13~~

index 1c1a41bd1..b14772812 100644
@@ -49,12 +49,9 @@
 #include "ena_netdev.h"
 #include "ena_pci_id_tbl.h"
 
-static char version[] = DEVICE_NAME " v" DRV_MODULE_VERSION "\n";
-
 MODULE_AUTHOR("Amazon.com, Inc. or its affiliates");
 MODULE_DESCRIPTION(DEVICE_NAME);
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_MODULE_VERSION);
 
 /* Time in jiffies before concluding the transmitter is hung. */
 #define TX_TIMEOUT  (5 * HZ)

@@ -2441,9 +2438,9 @@ static void ena_config_host_info(struct ena_com_dev *ena_dev,
 	strncpy(host_info->os_dist_str, utsname()->release,
 		sizeof(host_info->os_dist_str) - 1);
 	host_info->driver_version =
-		(DRV_MODULE_VER_MAJOR) |
-		(DRV_MODULE_VER_MINOR << ENA_ADMIN_HOST_INFO_MINOR_SHIFT) |
-		(DRV_MODULE_VER_SUBMINOR << ENA_ADMIN_HOST_INFO_SUB_MINOR_SHIFT) |
+		(DRV_MODULE_GEN_MAJOR) |
+		(DRV_MODULE_GEN_MINOR << ENA_ADMIN_HOST_INFO_MINOR_SHIFT) |
+		(DRV_MODULE_GEN_SUBMINOR << ENA_ADMIN_HOST_INFO_SUB_MINOR_SHIFT) |
 		("K"[0] << ENA_ADMIN_HOST_INFO_MODULE_TYPE_SHIFT);
 	host_info->num_cpus = num_online_cpus();

@@ -2822,9 +2819,7 @@ static int ena_restore_device(struct ena_adapter *adapter)
 		netif_carrier_on(adapter->netdev);
 
 	mod_timer(&adapter->timer_service, round_jiffies(jiffies + HZ));
-	dev_err(&pdev->dev,
-		"Device reset completed successfully, Driver info: %s\n",
-		version);
+	dev_err(&pdev->dev, "Device reset completed successfully\n");
 
 	return rc;
 err_disable_msix:

@@ -3459,8 +3454,6 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	dev_dbg(&pdev->dev, "%s\n", __func__);
 
-	dev_info_once(&pdev->dev, "%s", version);
-
 	rc = pci_enable_device_mem(pdev);
 	if (rc) {
 		dev_err(&pdev->dev, "pci_enable_device_mem() failed!\n");

@@ -3766,8 +3759,6 @@ static struct pci_driver ena_pci_driver = {
 
 static int __init ena_init(void)
 {
-	pr_info("%s", version);
-
 	ena_wq = create_singlethread_workqueue(DRV_MODULE_NAME);
 	if (!ena_wq) {
 		pr_err("Failed to create workqueue\n");

drivers/net/ethernet/amazon/ena/ena_netdev.h +0 -0

index 2fe5eeea6..ac3f481b7 100644

drivers/net/ethernet/amazon/ena/ena_netdev.h line 49

Quoted 9 lines

drivers/net/ethernet/amazon/ena/ena_netdev.h +3 -3

@@ -44,16 +44,16 @@
 #include "ena_com.h"
 #include "ena_eth_com.h"
 
-#define DRV_MODULE_VER_MAJOR	2
-#define DRV_MODULE_VER_MINOR	1
-#define DRV_MODULE_VER_SUBMINOR 0
+#define DRV_MODULE_GEN_MAJOR	2
+#define DRV_MODULE_GEN_MINOR	1
+#define DRV_MODULE_GEN_SUBMINOR 0

I'm completely fine with this variant and did very similar thing for broadcom driver where authors forwarded similar data to FW, but please add the following line above those defines:
/* DO NOT CHANGE DRV_MODULE_GEN_* values in in-tree code */

Thanks

Quoted 18 lines

drivers/net/ethernet/amazon/ena/ena_netdev.h +5 -6

 #define DRV_MODULE_NAME		"ena"
-#ifndef DRV_MODULE_VERSION
-#define DRV_MODULE_VERSION \
-	__stringify(DRV_MODULE_VER_MAJOR) "."	\
-	__stringify(DRV_MODULE_VER_MINOR) "."	\
-	__stringify(DRV_MODULE_VER_SUBMINOR) "K"
+#ifndef DRV_MODULE_GENERATION
+#define DRV_MODULE_GENERATION \
+	__stringify(DRV_MODULE_GEN_MAJOR) "."	\
+	__stringify(DRV_MODULE_GEN_MINOR) "."	\
+	__stringify(DRV_MODULE_GEN_SUBMINOR) "K"
 #endif
 
 #define DEVICE_NAME	"Elastic Network Adapter (ENA)"
--

2.24.1.AMZN

Leon Romanovsky

author

2020-02-24 08:53 UTC #31 lore raw

PATCH 13/18 net/amd: Remove useless driver version

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

Convert AMD drivers to respect universal linux kernel version.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/ethernet/amd/amd8111e.c          |  5 +----
 drivers/net/ethernet/amd/au1000_eth.c        |  5 -----
 drivers/net/ethernet/amd/nmclan_cs.c         |  9 +++------
 drivers/net/ethernet/amd/pcnet32.c           |  7 -------
 drivers/net/ethernet/amd/sunlance.c          | 10 ----------
 drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c |  1 -
 drivers/net/ethernet/amd/xgbe/xgbe-main.c    |  1 -
 drivers/net/ethernet/amd/xgbe/xgbe.h         |  1 -
 8 files changed, 4 insertions(+), 35 deletions(-)

drivers/net/ethernet/amd/amd8111e.c +1 -4

index 0f3b743425e8..7a1286f8e983 100644
@@ -84,9 +84,8 @@ Revision History:
 
 #include "amd8111e.h"
 #define MODULE_NAME	"amd8111e"
-#define MODULE_VERS	"3.0.7"
 MODULE_AUTHOR("Advanced Micro Devices, Inc.");
-MODULE_DESCRIPTION ("AMD8111 based 10/100 Ethernet Controller. Driver Version "MODULE_VERS);
+MODULE_DESCRIPTION("AMD8111 based 10/100 Ethernet Controller.");
 MODULE_LICENSE("GPL");
 module_param_array(speed_duplex, int, NULL, 0);
 MODULE_PARM_DESC(speed_duplex, "Set device speed and duplex modes, 0: Auto Negotiate, 1: 10Mbps Half Duplex, 2: 10Mbps Full Duplex, 3: 100Mbps Half Duplex, 4: 100Mbps Full Duplex");

@@ -1366,7 +1365,6 @@ static void amd8111e_get_drvinfo(struct net_device *dev,
 	struct amd8111e_priv *lp = netdev_priv(dev);
 	struct pci_dev *pci_dev = lp->pci_dev;
 	strlcpy(info->driver, MODULE_NAME, sizeof(info->driver));
-	strlcpy(info->version, MODULE_VERS, sizeof(info->version));
 	snprintf(info->fw_version, sizeof(info->fw_version),
 		"%u", chip_version);
 	strlcpy(info->bus_info, pci_name(pci_dev), sizeof(info->bus_info));

@@ -1875,7 +1873,6 @@ static int amd8111e_probe_one(struct pci_dev *pdev,
 
 	/*  display driver and device information */
     	chip_version = (readl(lp->mmio + CHIPID) & 0xf0000000)>>28;
-	dev_info(&pdev->dev, "AMD-8111e Driver Version: %s\n", MODULE_VERS);
 	dev_info(&pdev->dev, "[ Rev %x ] PCI 10/100BaseT Ethernet %pM\n",
 		 chip_version, dev->dev_addr);
 	if (lp->ext_phy_id)

drivers/net/ethernet/amd/au1000_eth.c +0 -5

index 089a4fbc61a0..9f6e3cc2ce80 100644
@@ -63,14 +63,12 @@ static int au1000_debug = 3;
 				NETIF_MSG_LINK)
 
 #define DRV_NAME	"au1000_eth"
-#define DRV_VERSION	"1.7"
 #define DRV_AUTHOR	"Pete Popov <[email protected]>"
 #define DRV_DESC	"Au1xxx on-chip Ethernet driver"
 
 MODULE_AUTHOR(DRV_AUTHOR);
 MODULE_DESCRIPTION(DRV_DESC);
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
 
 /* AU1000 MAC registers and bits */
 #define MAC_CONTROL		0x0

@@ -656,7 +654,6 @@ au1000_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 	struct au1000_private *aup = netdev_priv(dev);
 
 	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
 	snprintf(info->bus_info, sizeof(info->bus_info), "%s %d", DRV_NAME,
 		 aup->mac_id);
 }

@@ -1290,8 +1287,6 @@ static int au1000_probe(struct platform_device *pdev)
 	netdev_info(dev, "Au1xx0 Ethernet found at 0x%lx, irq %d\n",
 			(unsigned long)base->start, irq);
 
-	pr_info_once("%s version %s %s\n", DRV_NAME, DRV_VERSION, DRV_AUTHOR);
-
 	return 0;
 
 err_out:

drivers/net/ethernet/amd/nmclan_cs.c +3 -6

index 023aecf6ab30..11c0b13edd30 100644
@@ -114,8 +114,6 @@ Log: nmclan_cs.c,v
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #define DRV_NAME	"nmclan_cs"
-#define DRV_VERSION	"0.16"
-
 
 /* ----------------------------------------------------------------------------
 Conditional Compilation Options

@@ -367,7 +365,7 @@ typedef struct _mace_private {
 
     char tx_free_frames; /* Number of free transmit frame buffers */
     char tx_irq_disabled; /* MACE TX interrupt disabled */
-
+
     spinlock_t bank_lock; /* Must be held if you step off bank 0 */
 } mace_private;

@@ -444,7 +442,7 @@ static int nmclan_probe(struct pcmcia_device *link)
     lp = netdev_priv(dev);
     lp->p_dev = link;
     link->priv = dev;
-
+
     spin_lock_init(&lp->bank_lock);
     link->resource[0]->end = 32;
     link->resource[0]->flags |= IO_DATA_PATH_WIDTH_AUTO;

@@ -817,7 +815,6 @@ static void netdev_get_drvinfo(struct net_device *dev,
 			       struct ethtool_drvinfo *info)
 {
 	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
 	snprintf(info->bus_info, sizeof(info->bus_info),
 		"PCMCIA 0x%lx", dev->base_addr);
 }

@@ -1110,7 +1107,7 @@ static int mace_rx(struct net_device *dev, unsigned char RxCnt)
 	if (pkt_len & 1)
 	    *(skb_tail_pointer(skb) - 1) = inb(ioaddr + AM2150_RCV);
 	skb->protocol = eth_type_trans(skb, dev);
-
+
 	netif_rx(skb); /* Send the packet to the upper (protocol) layers. */
 
 	dev->stats.rx_packets++;

drivers/net/ethernet/amd/pcnet32.c +0 -7

index dc7d88227e76..07e8211eea51 100644
@@ -24,13 +24,9 @@
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #define DRV_NAME	"pcnet32"
-#define DRV_VERSION	"1.35"
 #define DRV_RELDATE	"21.Apr.2008"
 #define PFX		DRV_NAME ": "
 
-static const char *const version =
-    DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE " [email protected]\n";
-
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/sched.h>

@@ -809,7 +805,6 @@ static void pcnet32_get_drvinfo(struct net_device *dev,
 	struct pcnet32_private *lp = netdev_priv(dev);
 
 	strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
-	strlcpy(info->version, DRV_VERSION, sizeof(info->version));
 	if (lp->pci_dev)
 		strlcpy(info->bus_info, pci_name(lp->pci_dev),
 			sizeof(info->bus_info));

@@ -3006,8 +3001,6 @@ MODULE_LICENSE("GPL");
 
 static int __init pcnet32_init_module(void)
 {
-	pr_info("%s", version);
-
 	pcnet32_debug = netif_msg_init(debug, PCNET32_MSG_DEFAULT);
 
 	if ((tx_start_pt >= 0) && (tx_start_pt <= 3))

drivers/net/ethernet/amd/sunlance.c +0 ~~-10~~

index b00e00881253..a21b2e60157e 100644
@@ -105,14 +105,9 @@ static char lancestr[] = "LANCE";
 #include <asm/irq.h>
 
 #define DRV_NAME	"sunlance"
-#define DRV_VERSION	"2.02"
 #define DRV_RELDATE	"8/24/03"
 #define DRV_AUTHOR	"Miguel de Icaza ([email protected])"
 
-static char version[] =
-	DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE " " DRV_AUTHOR "\n";
-
-MODULE_VERSION(DRV_VERSION);
 MODULE_AUTHOR(DRV_AUTHOR);
 MODULE_DESCRIPTION("Sun Lance ethernet driver");
 MODULE_LICENSE("GPL");

@@ -1282,7 +1277,6 @@ static void lance_free_hwresources(struct lance_private *lp)
 static void sparc_lance_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
 {
 	strlcpy(info->driver, "sunlance", sizeof(info->driver));
-	strlcpy(info->version, "2.02", sizeof(info->version));
 }
 
 static const struct ethtool_ops sparc_lance_ethtool_ops = {

@@ -1305,7 +1299,6 @@ static int sparc_lance_probe_one(struct platform_device *op,
 				 struct platform_device *lebuffer)
 {
 	struct device_node *dp = op->dev.of_node;
-	static unsigned version_printed;
 	struct lance_private *lp;
 	struct net_device *dev;
 	int    i;

@@ -1316,9 +1309,6 @@ static int sparc_lance_probe_one(struct platform_device *op,
 
 	lp = netdev_priv(dev);
 
-	if (sparc_lance_debug && version_printed++ == 0)
-		printk (KERN_INFO "%s", version);
-
 	spin_lock_init(&lp->lock);
 
 	/* Copy the IDPROM ethernet address to the device structure, later we

drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c +0 -1

index 8083173f1a8f..b23c8ee24ee3 100644
@@ -405,7 +405,6 @@ static void xgbe_get_drvinfo(struct net_device *netdev,
 	struct xgbe_hw_features *hw_feat = &pdata->hw_feat;
 
 	strlcpy(drvinfo->driver, XGBE_DRV_NAME, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, XGBE_DRV_VERSION, sizeof(drvinfo->version));
 	strlcpy(drvinfo->bus_info, dev_name(pdata->dev),
 		sizeof(drvinfo->bus_info));
 	snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), "%d.%d.%d",

drivers/net/ethernet/amd/xgbe/xgbe-main.c +0 -1

index 7ce9c69e9c44..2a70714a791d 100644
@@ -127,7 +127,6 @@
 
 MODULE_AUTHOR("Tom Lendacky <[email protected]>");
 MODULE_LICENSE("Dual BSD/GPL");
-MODULE_VERSION(XGBE_DRV_VERSION);
 MODULE_DESCRIPTION(XGBE_DRV_DESC);
 
 static int debug = -1;

drivers/net/ethernet/amd/xgbe/xgbe.h +0 -1

index 47bcbcf58048..5897e46faca5 100644
@@ -135,7 +135,6 @@
 #include <linux/list.h>
 
 #define XGBE_DRV_NAME		"amd-xgbe"
-#define XGBE_DRV_VERSION	"1.0.3"
 #define XGBE_DRV_DESC		"AMD 10 Gigabit Ethernet Driver"
 
 /* Descriptor related defines */

Leon Romanovsky

author

2020-02-24 08:53 UTC #32 lore raw

PATCH 14/18 net/apm: Remove useless driver version

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

Delete module version in favor of global and unique linux kernel.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/ethernet/apm/xgene-v2/ethtool.c         | 1 -
 drivers/net/ethernet/apm/xgene-v2/main.c            | 1 -
 drivers/net/ethernet/apm/xgene-v2/main.h            | 1 -
 drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c | 1 -
 drivers/net/ethernet/apm/xgene/xgene_enet_main.c    | 1 -
 drivers/net/ethernet/apm/xgene/xgene_enet_main.h    | 1 -
 6 files changed, 6 deletions(-)

drivers/net/ethernet/apm/xgene-v2/ethtool.c +0 -1

index a58250c1b57a..da748beb7047 100644
@@ -89,7 +89,6 @@ static void xge_get_drvinfo(struct net_device *ndev,
 	struct platform_device *pdev = pdata->pdev;
 
 	strcpy(info->driver, "xgene-enet-v2");
-	strcpy(info->version, XGENE_ENET_V2_VERSION);
 	snprintf(info->fw_version, ETHTOOL_FWVERS_LEN, "N/A");
 	sprintf(info->bus_info, "%s", pdev->name);
 }

drivers/net/ethernet/apm/xgene-v2/main.c +0 -1

index c48f60996761..860c18fb7aae 100644
@@ -741,5 +741,4 @@ module_platform_driver(xge_driver);
 
 MODULE_DESCRIPTION("APM X-Gene SoC Ethernet v2 driver");
 MODULE_AUTHOR("Iyappan Subramanian <[email protected]>");
-MODULE_VERSION(XGENE_ENET_V2_VERSION);
 MODULE_LICENSE("GPL");

drivers/net/ethernet/apm/xgene-v2/main.h +0 -1

index d41439d2709d..b3985a7be59d 100644
@@ -28,7 +28,6 @@
 #include "ring.h"
 #include "ethtool.h"
 
-#define XGENE_ENET_V2_VERSION	"v1.0"
 #define XGENE_ENET_STD_MTU	1536
 #define XGENE_ENET_MIN_FRAME	60
 #define IRQ_ID_SIZE             16

drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c +0 -1

index 246dec27140d..4e7a95bd83d7 100644
@@ -103,7 +103,6 @@ static void xgene_get_drvinfo(struct net_device *ndev,
 	struct platform_device *pdev = pdata->pdev;
 
 	strcpy(info->driver, "xgene_enet");
-	strcpy(info->version, XGENE_DRV_VERSION);
 	snprintf(info->fw_version, ETHTOOL_FWVERS_LEN, "N/A");
 	sprintf(info->bus_info, "%s", pdev->name);
 }

drivers/net/ethernet/apm/xgene/xgene_enet_main.c +0 -1

index 6aee2f0fc0db..5f1fc6582d74 100644
@@ -2179,7 +2179,6 @@ static struct platform_driver xgene_enet_driver = {
 module_platform_driver(xgene_enet_driver);
 
 MODULE_DESCRIPTION("APM X-Gene SoC Ethernet driver");
-MODULE_VERSION(XGENE_DRV_VERSION);
 MODULE_AUTHOR("Iyappan Subramanian <[email protected]>");
 MODULE_AUTHOR("Keyur Chudgar <[email protected]>");
 MODULE_LICENSE("GPL");

drivers/net/ethernet/apm/xgene/xgene_enet_main.h +0 -1

index 18f4923b1723..d35a338120cf 100644
@@ -28,7 +28,6 @@
 #include "xgene_enet_ring2.h"
 #include "../../../phy/mdio-xgene.h"
 
-#define XGENE_DRV_VERSION	"v1.0"
 #define ETHER_MIN_PACKET	64
 #define ETHER_STD_PACKET	1518
 #define XGENE_ENET_STD_MTU	1536

--
2.24.1

Leon Romanovsky

author

2020-02-24 08:53 UTC #33 lore raw

PATCH 15/18 net/apm: Properly mark absence of FW

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

There is no need to set "N/A" if FW is not available.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/ethernet/apm/xgene-v2/ethtool.c         | 1 -
 drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c | 1 -
 2 files changed, 2 deletions(-)

drivers/net/ethernet/apm/xgene-v2/ethtool.c +0 -1

index da748beb7047..b78d1a99fe81 100644
@@ -89,7 +89,6 @@ static void xge_get_drvinfo(struct net_device *ndev,
 	struct platform_device *pdev = pdata->pdev;
 
 	strcpy(info->driver, "xgene-enet-v2");
-	snprintf(info->fw_version, ETHTOOL_FWVERS_LEN, "N/A");
 	sprintf(info->bus_info, "%s", pdev->name);
 }

drivers/net/ethernet/apm/xgene/xgene_enet_ethtool.c +0 -1

index 4e7a95bd83d7..ada70425b48c 100644
@@ -103,7 +103,6 @@ static void xgene_get_drvinfo(struct net_device *ndev,
 	struct platform_device *pdev = pdata->pdev;
 
 	strcpy(info->driver, "xgene_enet");
-	snprintf(info->fw_version, ETHTOOL_FWVERS_LEN, "N/A");
 	sprintf(info->bus_info, "%s", pdev->name);
 }

--
2.24.1

Leon Romanovsky

author

2020-02-24 08:53 UTC #34 lore raw

PATCH 16/18 net/aquantia: Delete module version

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

There is no need to keep module and driver versions in in-tree kernel code.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/ethernet/aquantia/atlantic/aq_cfg.h     |  4 ----
 drivers/net/ethernet/aquantia/atlantic/aq_common.h  |  1 -
 drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c |  1 -
 drivers/net/ethernet/aquantia/atlantic/aq_main.c    |  1 -
 drivers/net/ethernet/aquantia/atlantic/ver.h        | 12 ------------
 5 files changed, 19 deletions(-)
 delete mode 100644 drivers/net/ethernet/aquantia/atlantic/ver.h

drivers/net/ethernet/aquantia/atlantic/aq_cfg.h +0 -4

index f0c41f7408e5..7560f5506e55 100644
@@ -9,8 +9,6 @@
 #ifndef AQ_CFG_H
 #define AQ_CFG_H
 
-#include <generated/utsrelease.h>
-
 #define AQ_CFG_VECS_DEF   8U
 #define AQ_CFG_TCS_DEF    1U

@@ -85,7 +83,5 @@
 #define AQ_CFG_DRV_AUTHOR      "aQuantia"
 #define AQ_CFG_DRV_DESC        "aQuantia Corporation(R) Network Driver"
 #define AQ_CFG_DRV_NAME        "atlantic"
-#define AQ_CFG_DRV_VERSION	UTS_RELEASE \
-				AQ_CFG_DRV_VERSION_SUFFIX
 
 #endif /* AQ_CFG_H */

drivers/net/ethernet/aquantia/atlantic/aq_common.h +0 -1

index 42ea8d8daa46..c8c402b013bb 100644
@@ -12,7 +12,6 @@
 #include <linux/etherdevice.h>
 #include <linux/pci.h>
 #include <linux/if_vlan.h>
-#include "ver.h"
 #include "aq_cfg.h"
 #include "aq_utils.h"

drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c +0 -1

index 7b55633d2cb9..0bdaa0d785b7 100644
@@ -132,7 +132,6 @@ static void aq_ethtool_get_drvinfo(struct net_device *ndev,
 	regs_count = aq_nic_get_regs_count(aq_nic);
 
 	strlcat(drvinfo->driver, AQ_CFG_DRV_NAME, sizeof(drvinfo->driver));
-	strlcat(drvinfo->version, AQ_CFG_DRV_VERSION, sizeof(drvinfo->version));
 
 	snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version),
 		 "%u.%u.%u", firmware_version >> 24,

drivers/net/ethernet/aquantia/atlantic/aq_main.c +0 -1

index 538f460a3da7..9fcab646cbd5 100644
@@ -19,7 +19,6 @@
 #include <linux/udp.h>
 
 MODULE_LICENSE("GPL v2");
-MODULE_VERSION(AQ_CFG_DRV_VERSION);
 MODULE_AUTHOR(AQ_CFG_DRV_AUTHOR);
 MODULE_DESCRIPTION(AQ_CFG_DRV_DESC);

drivers/net/ethernet/aquantia/atlantic/ver.h deleted +0 ~~-12~~

deleted file mode 100644
index 597654b51e01..000000000000
@@ -1,12 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0-only */
-/*
- * aQuantia Corporation Network Driver
- * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved
- */
-
-#ifndef VER_H
-#define VER_H
-
-#define AQ_CFG_DRV_VERSION_SUFFIX "-kern"
-
-#endif /* VER_H */

Leon Romanovsky

author

2020-02-24 08:53 UTC #35 lore raw

PATCH 17/18 net/arc: Delete driver version

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

Drop constant driver version in favour of global linux kernel.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/ethernet/arc/emac.h          | 1 -
 drivers/net/ethernet/arc/emac_arc.c      | 2 --
 drivers/net/ethernet/arc/emac_main.c     | 1 -
 drivers/net/ethernet/arc/emac_rockchip.c | 2 --
 4 files changed, 6 deletions(-)

drivers/net/ethernet/arc/emac.h +0 -1

index d9efbc8d783b..d820ae03a966 100644
@@ -130,7 +130,6 @@ struct arc_emac_mdio_bus_data {
  */
 struct arc_emac_priv {
 	const char *drv_name;
-	const char *drv_version;
 	void (*set_mac_speed)(void *priv, unsigned int speed);
 
 	/* Devices */

drivers/net/ethernet/arc/emac_arc.c +0 -2

index 539166112993..1c7736b7eaf7 100644
@@ -15,7 +15,6 @@
 #include "emac.h"
 
 #define DRV_NAME    "emac_arc"
-#define DRV_VERSION "1.0"
 
 static int emac_arc_probe(struct platform_device *pdev)
 {

@@ -36,7 +35,6 @@ static int emac_arc_probe(struct platform_device *pdev)
 
 	priv = netdev_priv(ndev);
 	priv->drv_name = DRV_NAME;
-	priv->drv_version = DRV_VERSION;
 
 	err = of_get_phy_mode(dev->of_node, &interface);
 	if (err) {

drivers/net/ethernet/arc/emac_main.c +0 -1

index 17bda4e8cc45..38cd968b6a3b 100644
@@ -92,7 +92,6 @@ static void arc_emac_get_drvinfo(struct net_device *ndev,
 	struct arc_emac_priv *priv = netdev_priv(ndev);
 
 	strlcpy(info->driver, priv->drv_name, sizeof(info->driver));
-	strlcpy(info->version, priv->drv_version, sizeof(info->version));
 }
 
 static const struct ethtool_ops arc_emac_ethtool_ops = {

drivers/net/ethernet/arc/emac_rockchip.c +0 -2

index aae231c5224f..48ecdf15eddc 100644
@@ -16,7 +16,6 @@
 #include "emac.h"
 
 #define DRV_NAME        "rockchip_emac"
-#define DRV_VERSION     "1.1"
 
 struct emac_rockchip_soc_data {
 	unsigned int grf_offset;

@@ -112,7 +111,6 @@ static int emac_rockchip_probe(struct platform_device *pdev)
 
 	priv = netdev_priv(ndev);
 	priv->emac.drv_name = DRV_NAME;
-	priv->emac.drv_version = DRV_VERSION;
 	priv->emac.set_mac_speed = emac_rockchip_set_mac_speed;
 
 	err = of_get_phy_mode(dev->of_node, &interface);

Leon Romanovsky

author

2020-02-24 08:53 UTC #36 lore raw

PATCH 18/18 net/atheros: Clean atheros code from driver version

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>

Use linux kernel version for ethtool and module versions.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/ethernet/atheros/atl1c/atl1c.h         |  1 -
 drivers/net/ethernet/atheros/atl1c/atl1c_ethtool.c |  2 --
 drivers/net/ethernet/atheros/atl1c/atl1c_main.c    |  5 -----
 drivers/net/ethernet/atheros/atl1e/atl1e.h         |  1 -
 drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c |  2 --
 drivers/net/ethernet/atheros/atl1e/atl1e_main.c    |  4 ----
 drivers/net/ethernet/atheros/atlx/atl1.c           |  6 ------
 drivers/net/ethernet/atheros/atlx/atl2.c           | 10 ----------
 8 files changed, 31 deletions(-)

drivers/net/ethernet/atheros/atl1c/atl1c.h +0 -1

index 60b2febd7315..a0562a90fb6d 100644
@@ -583,7 +583,6 @@ struct atl1c_adapter {
 		readl(((a)->hw_addr + reg) + ((offset) << 2)))
 
 extern char atl1c_driver_name[];
-extern char atl1c_driver_version[];
 
 void atl1c_reinit_locked(struct atl1c_adapter *adapter);
 s32 atl1c_reset_hw(struct atl1c_hw *hw);

drivers/net/ethernet/atheros/atl1c/atl1c_ethtool.c +0 -2

index b5a70a36fa04..e2eb7b8c63a0 100644
@@ -221,8 +221,6 @@ static void atl1c_get_drvinfo(struct net_device *netdev,
 	struct atl1c_adapter *adapter = netdev_priv(netdev);
 
 	strlcpy(drvinfo->driver,  atl1c_driver_name, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, atl1c_driver_version,
-		sizeof(drvinfo->version));
 	strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
 		sizeof(drvinfo->bus_info));
 }

drivers/net/ethernet/atheros/atl1c/atl1c_main.c +0 -5

index 0d67b951c0b2..00bd7bd55794 100644
@@ -8,9 +8,7 @@
 
 #include "atl1c.h"
 
-#define ATL1C_DRV_VERSION "1.0.1.1-NAPI"
 char atl1c_driver_name[] = "atl1c";
-char atl1c_driver_version[] = ATL1C_DRV_VERSION;
 
 /*
  * atl1c_pci_tbl - PCI Device ID Table

@@ -37,7 +35,6 @@ MODULE_AUTHOR("Jie Yang");
 MODULE_AUTHOR("Qualcomm Atheros Inc., <[email protected]>");
 MODULE_DESCRIPTION("Qualcomm Atheros 100/1000M Ethernet Network Driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(ATL1C_DRV_VERSION);
 
 static int atl1c_stop_mac(struct atl1c_hw *hw);
 static void atl1c_disable_l0s_l1(struct atl1c_hw *hw);

@@ -2642,8 +2639,6 @@ static int atl1c_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		goto err_register;
 	}
 
-	if (netif_msg_probe(adapter))
-		dev_info(&pdev->dev, "version %s\n", ATL1C_DRV_VERSION);
 	cards_found++;
 	return 0;

drivers/net/ethernet/atheros/atl1e/atl1e.h +0 -1

index e9893da50995..9fcad783c939 100644
@@ -482,7 +482,6 @@ struct atl1e_adapter {
 		readl(((a)->hw_addr + reg) + ((offset) << 2)))
 
 extern char atl1e_driver_name[];
-extern char atl1e_driver_version[];
 
 void atl1e_check_options(struct atl1e_adapter *adapter);
 int atl1e_up(struct atl1e_adapter *adapter);

drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c +0 -2

index c6b9e7ea8e38..0cbde352d1ba 100644
@@ -307,8 +307,6 @@ static void atl1e_get_drvinfo(struct net_device *netdev,
 	struct atl1e_adapter *adapter = netdev_priv(netdev);
 
 	strlcpy(drvinfo->driver,  atl1e_driver_name, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, atl1e_driver_version,
-		sizeof(drvinfo->version));
 	strlcpy(drvinfo->fw_version, "L1e", sizeof(drvinfo->fw_version));
 	strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
 		sizeof(drvinfo->bus_info));

drivers/net/ethernet/atheros/atl1e/atl1e_main.c +0 -4

index e0d89942d537..223ef846123e 100644
@@ -8,10 +8,7 @@
 
 #include "atl1e.h"
 
-#define DRV_VERSION "1.0.0.7-NAPI"
-
 char atl1e_driver_name[] = "ATL1E";
-char atl1e_driver_version[] = DRV_VERSION;
 #define PCI_DEVICE_ID_ATTANSIC_L1E      0x1026
 /*
  * atl1e_pci_tbl - PCI Device ID Table

@@ -33,7 +30,6 @@ MODULE_DEVICE_TABLE(pci, atl1e_pci_tbl);
 MODULE_AUTHOR("Atheros Corporation, <[email protected]>, Jie Yang <[email protected]>");
 MODULE_DESCRIPTION("Atheros 1000M Ethernet Network Driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
 
 static void atl1e_setup_mac_ctrl(struct atl1e_adapter *adapter);

drivers/net/ethernet/atheros/atlx/atl1.c +0 -6

index b498fd6a47d0..271e7034fa70 100644
@@ -65,12 +65,10 @@
 
 #include "atl1.h"
 
-#define ATLX_DRIVER_VERSION "2.1.3"
 MODULE_AUTHOR("Xiong Huang <[email protected]>, "
 	      "Chris Snook <[email protected]>, "
 	      "Jay Cliburn <[email protected]>");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(ATLX_DRIVER_VERSION);
 
 /* Temporary hack for merging atl1 and atl2 */
 #include "atlx.c"

@@ -2965,8 +2963,6 @@ static int atl1_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	/* get device revision number */
 	adapter->hw.dev_rev = ioread16(adapter->hw.hw_addr +
 		(REG_MASTER_CTRL + 2));
-	if (netif_msg_probe(adapter))
-		dev_info(&pdev->dev, "version %s\n", ATLX_DRIVER_VERSION);
 
 	/* set default ring resource counts */
 	adapter->rfd_ring.count = adapter->rrd_ring.count = ATL1_DEFAULT_RFD;

@@ -3344,8 +3340,6 @@ static void atl1_get_drvinfo(struct net_device *netdev,
 	struct atl1_adapter *adapter = netdev_priv(netdev);
 
 	strlcpy(drvinfo->driver, ATLX_DRIVER_NAME, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, ATLX_DRIVER_VERSION,
-		sizeof(drvinfo->version));
 	strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
 		sizeof(drvinfo->bus_info));
 }

drivers/net/ethernet/atheros/atlx/atl2.c +0 ~~-10~~

index b81a4e0c5b57..7c52b92b599d 100644
@@ -36,18 +36,13 @@
 
 #include "atl2.h"
 
-#define ATL2_DRV_VERSION "2.2.3"
-
 static const char atl2_driver_name[] = "atl2";
 static const char atl2_driver_string[] = "Atheros(R) L2 Ethernet Driver";
-static const char atl2_copyright[] = "Copyright (c) 2007 Atheros Corporation.";
-static const char atl2_driver_version[] = ATL2_DRV_VERSION;
 static const struct ethtool_ops atl2_ethtool_ops;
 
 MODULE_AUTHOR("Atheros Corporation <[email protected]>, Chris Snook <[email protected]>");
 MODULE_DESCRIPTION("Atheros Fast Ethernet Network Driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(ATL2_DRV_VERSION);
 
 /*
  * atl2_pci_tbl - PCI Device ID Table

@@ -1688,9 +1683,6 @@ static struct pci_driver atl2_driver = {
  */
 static int __init atl2_init_module(void)
 {
-	printk(KERN_INFO "%s - version %s\n", atl2_driver_string,
-		atl2_driver_version);
-	printk(KERN_INFO "%s\n", atl2_copyright);
 	return pci_register_driver(&atl2_driver);
 }
 module_init(atl2_init_module);

@@ -2011,8 +2003,6 @@ static void atl2_get_drvinfo(struct net_device *netdev,
 	struct atl2_adapter *adapter = netdev_priv(netdev);
 
 	strlcpy(drvinfo->driver,  atl2_driver_name, sizeof(drvinfo->driver));
-	strlcpy(drvinfo->version, atl2_driver_version,
-		sizeof(drvinfo->version));
 	strlcpy(drvinfo->fw_version, "L2", sizeof(drvinfo->fw_version));
 	strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
 		sizeof(drvinfo->bus_info));

Chris Snook

maintainer Acked-by on v1 18/18

2020-02-24 18:15 UTC #37 lore raw

Re: [PATCH net-next v1 18/18] net/atheros: Clean atheros code from driver version

Acked-by Chris Snook <[email protected]>

Quoted 201 lines · Leon Romanovsky

From: Leon Romanovsky <[email protected]>

Use linux kernel version for ethtool and module versions.

Signed-off-by Leon Romanovsky <[email protected]>

---
 drivers/net/ethernet/atheros/atl1c/atl1c.h         |  1 -
 drivers/net/ethernet/atheros/atl1c/atl1c_ethtool.c |  2 --
 drivers/net/ethernet/atheros/atl1c/atl1c_main.c    |  5 -----
 drivers/net/ethernet/atheros/atl1e/atl1e.h         |  1 -
 drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c |  2 --
 drivers/net/ethernet/atheros/atl1e/atl1e_main.c    |  4 ----
 drivers/net/ethernet/atheros/atlx/atl1.c           |  6 ------
 drivers/net/ethernet/atheros/atlx/atl2.c           | 10 ----------
 8 files changed, 31 deletions(-)

drivers/net/ethernet/atheros/atl1c/atl1c.h +0 -1

index 60b2febd7315..a0562a90fb6d 100644
@@ -583,7 +583,6 @@ struct atl1c_adapter {
                readl(((a)->hw_addr + reg) + ((offset) << 2)))
 
 extern char atl1c_driver_name[];
-extern char atl1c_driver_version[];
 
 void atl1c_reinit_locked(struct atl1c_adapter *adapter);
 s32 atl1c_reset_hw(struct atl1c_hw *hw);

drivers/net/ethernet/atheros/atl1c/atl1c_ethtool.c +0 -2

index b5a70a36fa04..e2eb7b8c63a0 100644
@@ -221,8 +221,6 @@ static void atl1c_get_drvinfo(struct net_device *netdev,
        struct atl1c_adapter *adapter = netdev_priv(netdev);
 
        strlcpy(drvinfo->driver,  atl1c_driver_name, sizeof(drvinfo->driver));
-       strlcpy(drvinfo->version, atl1c_driver_version,
-               sizeof(drvinfo->version));
        strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
                sizeof(drvinfo->bus_info));
 }

drivers/net/ethernet/atheros/atl1c/atl1c_main.c +0 -5

index 0d67b951c0b2..00bd7bd55794 100644
@@ -8,9 +8,7 @@
 
 #include "atl1c.h"
 
-#define ATL1C_DRV_VERSION "1.0.1.1-NAPI"
 char atl1c_driver_name[] = "atl1c";
-char atl1c_driver_version[] = ATL1C_DRV_VERSION;
 
 /*
  * atl1c_pci_tbl - PCI Device ID Table

@@ -37,7 +35,6 @@ MODULE_AUTHOR("Jie Yang");
 MODULE_AUTHOR("Qualcomm Atheros Inc., <[email protected]>");
 MODULE_DESCRIPTION("Qualcomm Atheros 100/1000M Ethernet Network Driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(ATL1C_DRV_VERSION);
 
 static int atl1c_stop_mac(struct atl1c_hw *hw);
 static void atl1c_disable_l0s_l1(struct atl1c_hw *hw);

@@ -2642,8 +2639,6 @@ static int atl1c_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
                goto err_register;
        }
 
-       if (netif_msg_probe(adapter))
-               dev_info(&pdev->dev, "version %s\n", ATL1C_DRV_VERSION);
        cards_found++;
        return 0;

drivers/net/ethernet/atheros/atl1e/atl1e.h +0 -1

index e9893da50995..9fcad783c939 100644
@@ -482,7 +482,6 @@ struct atl1e_adapter {
                readl(((a)->hw_addr + reg) + ((offset) << 2)))
 
 extern char atl1e_driver_name[];
-extern char atl1e_driver_version[];
 
 void atl1e_check_options(struct atl1e_adapter *adapter);
 int atl1e_up(struct atl1e_adapter *adapter);

drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c +0 -2

index c6b9e7ea8e38..0cbde352d1ba 100644
@@ -307,8 +307,6 @@ static void atl1e_get_drvinfo(struct net_device *netdev,
        struct atl1e_adapter *adapter = netdev_priv(netdev);
 
        strlcpy(drvinfo->driver,  atl1e_driver_name, sizeof(drvinfo->driver));
-       strlcpy(drvinfo->version, atl1e_driver_version,
-               sizeof(drvinfo->version));
        strlcpy(drvinfo->fw_version, "L1e", sizeof(drvinfo->fw_version));
        strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
                sizeof(drvinfo->bus_info));

drivers/net/ethernet/atheros/atl1e/atl1e_main.c +0 -4

index e0d89942d537..223ef846123e 100644
@@ -8,10 +8,7 @@
 
 #include "atl1e.h"
 
-#define DRV_VERSION "1.0.0.7-NAPI"
-
 char atl1e_driver_name[] = "ATL1E";
-char atl1e_driver_version[] = DRV_VERSION;
 #define PCI_DEVICE_ID_ATTANSIC_L1E      0x1026
 /*
  * atl1e_pci_tbl - PCI Device ID Table

@@ -33,7 +30,6 @@ MODULE_DEVICE_TABLE(pci, atl1e_pci_tbl);
 MODULE_AUTHOR("Atheros Corporation, <[email protected]>, Jie Yang <[email protected]>");
 MODULE_DESCRIPTION("Atheros 1000M Ethernet Network Driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(DRV_VERSION);
 
 static void atl1e_setup_mac_ctrl(struct atl1e_adapter *adapter);

drivers/net/ethernet/atheros/atlx/atl1.c +0 -6

index b498fd6a47d0..271e7034fa70 100644
@@ -65,12 +65,10 @@
 
 #include "atl1.h"
 
-#define ATLX_DRIVER_VERSION "2.1.3"
 MODULE_AUTHOR("Xiong Huang <[email protected]>, "
              "Chris Snook <[email protected]>, "
              "Jay Cliburn <[email protected]>");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(ATLX_DRIVER_VERSION);
 
 /* Temporary hack for merging atl1 and atl2 */
 #include "atlx.c"

@@ -2965,8 +2963,6 @@ static int atl1_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
        /* get device revision number */
        adapter->hw.dev_rev = ioread16(adapter->hw.hw_addr +
                (REG_MASTER_CTRL + 2));
-       if (netif_msg_probe(adapter))
-               dev_info(&pdev->dev, "version %s\n", ATLX_DRIVER_VERSION);
 
        /* set default ring resource counts */
        adapter->rfd_ring.count = adapter->rrd_ring.count = ATL1_DEFAULT_RFD;

@@ -3344,8 +3340,6 @@ static void atl1_get_drvinfo(struct net_device *netdev,
        struct atl1_adapter *adapter = netdev_priv(netdev);
 
        strlcpy(drvinfo->driver, ATLX_DRIVER_NAME, sizeof(drvinfo->driver));
-       strlcpy(drvinfo->version, ATLX_DRIVER_VERSION,
-               sizeof(drvinfo->version));
        strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
                sizeof(drvinfo->bus_info));
 }

drivers/net/ethernet/atheros/atlx/atl2.c +0 ~~-11~~

index b81a4e0c5b57..7c52b92b599d 100644
@@ -36,18 +36,13 @@
 
 #include "atl2.h"
 
-#define ATL2_DRV_VERSION "2.2.3"
-
 static const char atl2_driver_name[] = "atl2";
 static const char atl2_driver_string[] = "Atheros(R) L2 Ethernet Driver";
-static const char atl2_copyright[] = "Copyright (c) 2007 Atheros Corporation.";
-static const char atl2_driver_version[] = ATL2_DRV_VERSION;
 static const struct ethtool_ops atl2_ethtool_ops;
 
 MODULE_AUTHOR("Atheros Corporation <[email protected]>, Chris Snook <[email protected]>");
 MODULE_DESCRIPTION("Atheros Fast Ethernet Network Driver");
 MODULE_LICENSE("GPL");
-MODULE_VERSION(ATL2_DRV_VERSION);
 
 /*
  * atl2_pci_tbl - PCI Device ID Table

@@ -1688,9 +1683,6 @@ static struct pci_driver atl2_driver = {
  */
 static int __init atl2_init_module(void)
 {
-       printk(KERN_INFO "%s - version %s\n", atl2_driver_string,
-               atl2_driver_version);
-       printk(KERN_INFO "%s\n", atl2_copyright);
        return pci_register_driver(&atl2_driver);
 }
 module_init(atl2_init_module);

@@ -2011,8 +2003,6 @@ static void atl2_get_drvinfo(struct net_device *netdev,
        struct atl2_adapter *adapter = netdev_priv(netdev);
 
        strlcpy(drvinfo->driver,  atl2_driver_name, sizeof(drvinfo->driver));
-       strlcpy(drvinfo->version, atl2_driver_version,
-               sizeof(drvinfo->version));
        strlcpy(drvinfo->fw_version, "L2", sizeof(drvinfo->fw_version));
        strlcpy(drvinfo->bus_info, pci_name(adapter->pdev),
                sizeof(drvinfo->bus_info));
--

2.24.1

David Miller

maintainer

2020-02-24 19:49 UTC #38 lore raw

In reply to Leon Romanovsky

From: Leon Romanovsky <[email protected]>
Date: Mon, 24 Feb 2020 10:52:53 +0200

Quoted 9 lines

From: Leon Romanovsky <[email protected]>
Change log:
v1:
* Split all FW cleanups patches to separate patches
* Fixed commit message
* Deleted odd DRV_RELDATE
* Added everyone from MAINTAINERS file
v0: https://lore.kernel.org/netdev/[email protected]

Series applied, thanks Leon.