2022-10-03 06:45:12

by Divya Koppera

[permalink] [raw]
Subject: [PATCH net] net: phy: micrel: Fixes FIELD_GET assertion

FIELD_GET() must only be used with a mask that is a compile-time
constant. Mark the functions as __always_inline to avoid the problem.

Fixes: 21b688dabecb6a ("net: phy: micrel: Cable Diag feature for lan8814 phy")
Reported-by: kernel test robot <[email protected]>
Signed-off-by: Divya Koppera <[email protected]>
---
drivers/net/phy/micrel.c | 9 +++++----
1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c
index 3757e069c486..54a17b576eac 100644
--- a/drivers/net/phy/micrel.c
+++ b/drivers/net/phy/micrel.c
@@ -1838,7 +1838,7 @@ static int ksz886x_cable_test_start(struct phy_device *phydev)
return phy_clear_bits(phydev, MII_BMCR, BMCR_ANENABLE | BMCR_SPEED100);
}

-static int ksz886x_cable_test_result_trans(u16 status, u16 mask)
+static __always_inline int ksz886x_cable_test_result_trans(u16 status, u16 mask)
{
switch (FIELD_GET(mask, status)) {
case KSZ8081_LMD_STAT_NORMAL:
@@ -1854,13 +1854,13 @@ static int ksz886x_cable_test_result_trans(u16 status, u16 mask)
}
}

-static bool ksz886x_cable_test_failed(u16 status, u16 mask)
+static __always_inline bool ksz886x_cable_test_failed(u16 status, u16 mask)
{
return FIELD_GET(mask, status) ==
KSZ8081_LMD_STAT_FAIL;
}

-static bool ksz886x_cable_test_fault_length_valid(u16 status, u16 mask)
+static __always_inline bool ksz886x_cable_test_fault_length_valid(u16 status, u16 mask)
{
switch (FIELD_GET(mask, status)) {
case KSZ8081_LMD_STAT_OPEN:
@@ -1871,7 +1871,8 @@ static bool ksz886x_cable_test_fault_length_valid(u16 status, u16 mask)
return false;
}

-static int ksz886x_cable_test_fault_length(struct phy_device *phydev, u16 status, u16 data_mask)
+static __always_inline int ksz886x_cable_test_fault_length(struct phy_device *phydev,
+ u16 status, u16 data_mask)
{
int dt;

--
2.17.1


2022-10-04 00:56:50

by Jakub Kicinski

[permalink] [raw]
Subject: Re: [PATCH net] net: phy: micrel: Fixes FIELD_GET assertion

On Mon, 3 Oct 2022 12:01:30 +0530 Divya Koppera wrote:
> FIELD_GET() must only be used with a mask that is a compile-time
> constant. Mark the functions as __always_inline to avoid the problem.
>
> Fixes: 21b688dabecb6a ("net: phy: micrel: Cable Diag feature for lan8814 phy")

Does not apply cleanly to net, please rebase & resend.

2022-10-04 07:17:29

by Divya Koppera

[permalink] [raw]
Subject: RE: [PATCH net] net: phy: micrel: Fixes FIELD_GET assertion

Hi Jakub,

> -----Original Message-----
> From: Jakub Kicinski <[email protected]>
> Sent: Tuesday, October 4, 2022 5:39 AM
> To: Divya Koppera - I30481 <[email protected]>
> Cc: [email protected]; [email protected]; [email protected];
> [email protected]; [email protected]; [email protected];
> [email protected]; [email protected]; UNGLinuxDriver
> <[email protected]>
> Subject: Re: [PATCH net] net: phy: micrel: Fixes FIELD_GET assertion
>
> EXTERNAL EMAIL: Do not click links or open attachments unless you know the
> content is safe
>
> On Mon, 3 Oct 2022 12:01:30 +0530 Divya Koppera wrote:
> > FIELD_GET() must only be used with a mask that is a compile-time
> > constant. Mark the functions as __always_inline to avoid the problem.
> >
> > Fixes: 21b688dabecb6a ("net: phy: micrel: Cable Diag feature for
> > lan8814 phy")
>
> Does not apply cleanly to net, please rebase & resend.

I generated patch on net-next, as it is fix I kept for net. When I tried to apply these changes on net, this is getting failed as this main patch(21b688dabecb6a) did not go in net. Shall I resend patch for net-next?

Thanks,
Divya

2022-10-04 20:38:47

by Jakub Kicinski

[permalink] [raw]
Subject: Re: [PATCH net] net: phy: micrel: Fixes FIELD_GET assertion

On Tue, 4 Oct 2022 06:28:44 +0000 [email protected] wrote:
> > Does not apply cleanly to net, please rebase & resend.
>
> I generated patch on net-next, as it is fix I kept for net. When I
> tried to apply these changes on net, this is getting failed as this
> main patch(21b688dabecb6a) did not go in net. Shall I resend patch
> for net-next?

Maybe wait a day or two and rebase + resend in two day or so, we're
waiting to Linus to pull net-next after which point the trees will get
forwarded to the same point.