From: John Hubbard <[email protected]>
Hi,
Say, I just ran into this build breakage on today's linux.git, and after
checking the email threads, I do realize that PHY_MODE_SATA is about to be
added as part of [1]. However, I also noticed that it was not identified
as a build fix, nor did anyone notice the potential naming conflict. So
here is a patch for that.
Please feel free to merge, reword, discard, etc, as you see fit. :)
[1] https://lkml.kernel.org/r/[email protected]
John Hubbard (1):
phy: fix build breakage: add PHY_MODE_SATA
drivers/phy/marvell/phy-berlin-sata.c | 5 +++--
include/linux/phy/phy.h | 1 +
2 files changed, 4 insertions(+), 2 deletions(-)
--
2.20.1
From: John Hubbard <[email protected]>
Commit 49e54187ae0b ("ata: libahci_platform: comply to PHY framework") uses
the PHY_MODE_SATA, but that enum had not yet been added. This caused a
build failure for me, with today's linux.git.
Also, there is a potentially conflicting (mis-named) PHY_MODE_SATA, hiding
in the Marvell Berlin SATA PHY driver.
Fix the build by:
1) Renaming Marvell's defined value to a more scoped name,
in order to avoid any potential conflicts: PHY_BERLIN_MODE_SATA.
2) Adding the missing enum, which was going to be added anyway as part
of [1].
[1] https://lkml.kernel.org/r/[email protected]
Fixes: 49e54187ae0b ("ata: libahci_platform: comply to PHY framework")
Cc: Grzegorz Jaszczyk <[email protected]>
Cc: Miquel Raynal <[email protected]>
Cc: Hans de Goede <[email protected]>
Cc: Jens Axboe <[email protected]>
Signed-off-by: John Hubbard <[email protected]>
---
drivers/phy/marvell/phy-berlin-sata.c | 5 +++--
include/linux/phy/phy.h | 1 +
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/phy/marvell/phy-berlin-sata.c b/drivers/phy/marvell/phy-berlin-sata.c
index a91fc67fc4e0..d70ba9bc42d9 100644
--- a/drivers/phy/marvell/phy-berlin-sata.c
+++ b/drivers/phy/marvell/phy-berlin-sata.c
@@ -32,7 +32,7 @@
/* register 0x01 */
#define REF_FREF_SEL_25 BIT(0)
-#define PHY_MODE_SATA (0x0 << 5)
+#define PHY_BERLIN_MODE_SATA (0x0 << 5)
/* register 0x02 */
#define USE_MAX_PLL_RATE BIT(12)
@@ -102,7 +102,8 @@ static int phy_berlin_sata_power_on(struct phy *phy)
/* set PHY mode and ref freq to 25 MHz */
phy_berlin_sata_reg_setbits(ctrl_reg, priv->phy_base, 0x01,
- 0x00ff, REF_FREF_SEL_25 | PHY_MODE_SATA);
+ 0x00ff,
+ REF_FREF_SEL_25 | PHY_BERLIN_MODE_SATA);
/* set PHY up to 6 Gbps */
phy_berlin_sata_reg_setbits(ctrl_reg, priv->phy_base, 0x25,
diff --git a/include/linux/phy/phy.h b/include/linux/phy/phy.h
index e8e118d70fd7..3f350e2749fe 100644
--- a/include/linux/phy/phy.h
+++ b/include/linux/phy/phy.h
@@ -42,6 +42,7 @@ enum phy_mode {
PHY_MODE_PCIE,
PHY_MODE_ETHERNET,
PHY_MODE_MIPI_DPHY,
+ PHY_MODE_SATA
};
/**
--
2.20.1
On 1/12/19 6:29 PM, [email protected] wrote:
> From: John Hubbard <[email protected]>
>
> Commit 49e54187ae0b ("ata: libahci_platform: comply to PHY framework") uses
> the PHY_MODE_SATA, but that enum had not yet been added. This caused a
> build failure for me, with today's linux.git.
>
> Also, there is a potentially conflicting (mis-named) PHY_MODE_SATA, hiding
> in the Marvell Berlin SATA PHY driver.
>
> Fix the build by:
>
> 1) Renaming Marvell's defined value to a more scoped name,
> in order to avoid any potential conflicts: PHY_BERLIN_MODE_SATA.
>
> 2) Adding the missing enum, which was going to be added anyway as part
> of [1].
>
> [1] https://lkml.kernel.org/r/[email protected]
>
> Fixes: 49e54187ae0b ("ata: libahci_platform: comply to PHY framework")
Linus, this is probably a better option in terms of what should go in to
fix that commit.
>
> Cc: Grzegorz Jaszczyk <[email protected]>
> Cc: Miquel Raynal <[email protected]>
> Cc: Hans de Goede <[email protected]>
> Cc: Jens Axboe <[email protected]>
> Signed-off-by: John Hubbard <[email protected]>
> ---
> drivers/phy/marvell/phy-berlin-sata.c | 5 +++--
> include/linux/phy/phy.h | 1 +
> 2 files changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/phy/marvell/phy-berlin-sata.c b/drivers/phy/marvell/phy-berlin-sata.c
> index a91fc67fc4e0..d70ba9bc42d9 100644
> --- a/drivers/phy/marvell/phy-berlin-sata.c
> +++ b/drivers/phy/marvell/phy-berlin-sata.c
> @@ -32,7 +32,7 @@
>
> /* register 0x01 */
> #define REF_FREF_SEL_25 BIT(0)
> -#define PHY_MODE_SATA (0x0 << 5)
> +#define PHY_BERLIN_MODE_SATA (0x0 << 5)
>
> /* register 0x02 */
> #define USE_MAX_PLL_RATE BIT(12)
> @@ -102,7 +102,8 @@ static int phy_berlin_sata_power_on(struct phy *phy)
>
> /* set PHY mode and ref freq to 25 MHz */
> phy_berlin_sata_reg_setbits(ctrl_reg, priv->phy_base, 0x01,
> - 0x00ff, REF_FREF_SEL_25 | PHY_MODE_SATA);
> + 0x00ff,
> + REF_FREF_SEL_25 | PHY_BERLIN_MODE_SATA);
>
> /* set PHY up to 6 Gbps */
> phy_berlin_sata_reg_setbits(ctrl_reg, priv->phy_base, 0x25,
> diff --git a/include/linux/phy/phy.h b/include/linux/phy/phy.h
> index e8e118d70fd7..3f350e2749fe 100644
> --- a/include/linux/phy/phy.h
> +++ b/include/linux/phy/phy.h
> @@ -42,6 +42,7 @@ enum phy_mode {
> PHY_MODE_PCIE,
> PHY_MODE_ETHERNET,
> PHY_MODE_MIPI_DPHY,
> + PHY_MODE_SATA
> };
>
> /**
>
--
Jens Axboe
On Sat, Jan 12, 2019 at 6:05 PM Jens Axboe <[email protected]> wrote:
>
> On 1/12/19 6:29 PM, [email protected] wrote:
> > From: John Hubbard <[email protected]>
> >
> > Commit 49e54187ae0b ("ata: libahci_platform: comply to PHY framework") uses
> > the PHY_MODE_SATA, but that enum had not yet been added. This caused a
> > build failure for me, with today's linux.git.
> >
> > Also, there is a potentially conflicting (mis-named) PHY_MODE_SATA, hiding
> > in the Marvell Berlin SATA PHY driver.
> >
> > Fix the build by:
> >
> > 1) Renaming Marvell's defined value to a more scoped name,
> > in order to avoid any potential conflicts: PHY_BERLIN_MODE_SATA.
> >
> > 2) Adding the missing enum, which was going to be added anyway as part
> > of [1].
> >
> > [1] https://lkml.kernel.org/r/[email protected]
> >
> > Fixes: 49e54187ae0b ("ata: libahci_platform: comply to PHY framework")
>
> Linus, this is probably a better option in terms of what should go in to
> fix that commit.
I'm OK with this, but it does beg the question how the patch was
tested before submitting, if it didn't build.
Is there functional breakage behind it? I currently lack online
hardware to test myself, unfortunately.
-Olof
Hi Olof,
Olof Johansson <[email protected]> wrote on Sat, 12 Jan 2019 19:57:12
-0800:
> On Sat, Jan 12, 2019 at 6:05 PM Jens Axboe <[email protected]> wrote:
> >
> > On 1/12/19 6:29 PM, [email protected] wrote:
> > > From: John Hubbard <[email protected]>
> > >
> > > Commit 49e54187ae0b ("ata: libahci_platform: comply to PHY framework") uses
> > > the PHY_MODE_SATA, but that enum had not yet been added. This caused a
> > > build failure for me, with today's linux.git.
> > >
> > > Also, there is a potentially conflicting (mis-named) PHY_MODE_SATA, hiding
> > > in the Marvell Berlin SATA PHY driver.
> > >
> > > Fix the build by:
> > >
> > > 1) Renaming Marvell's defined value to a more scoped name,
> > > in order to avoid any potential conflicts: PHY_BERLIN_MODE_SATA.
> > >
> > > 2) Adding the missing enum, which was going to be added anyway as part
> > > of [1].
> > >
> > > [1] https://lkml.kernel.org/r/[email protected]
> > >
> > > Fixes: 49e54187ae0b ("ata: libahci_platform: comply to PHY framework")
> >
> > Linus, this is probably a better option in terms of what should go in to
> > fix that commit.
>
> I'm OK with this, but it does beg the question how the patch was
> tested before submitting, if it didn't build.
>
> Is there functional breakage behind it? I currently lack online
> hardware to test myself, unfortunately.
This is my mistake, I forgot to tell Jens about this dependency,
I am very sorry about that. As reported by John, this patch depends on
the addition of PHY_MODE_SATA in the PHY type enumeration. This series
([1]) has been delayed and I should have warned Jens about it. I'm fine
with the above fix though.
Kishon, will you be able to base phy-next on top of this fix? It will
be needed for the addition of the COMPHY driver.
Thanks and again, sorry for the troubles.
Miquèl
On 14/01/19 2:27 PM, Miquel Raynal wrote:
> Hi Olof,
>
> Olof Johansson <[email protected]> wrote on Sat, 12 Jan 2019 19:57:12
> -0800:
>
>> On Sat, Jan 12, 2019 at 6:05 PM Jens Axboe <[email protected]> wrote:
>>>
>>> On 1/12/19 6:29 PM, [email protected] wrote:
>>>> From: John Hubbard <[email protected]>
>>>>
>>>> Commit 49e54187ae0b ("ata: libahci_platform: comply to PHY framework") uses
>>>> the PHY_MODE_SATA, but that enum had not yet been added. This caused a
>>>> build failure for me, with today's linux.git.
>>>>
>>>> Also, there is a potentially conflicting (mis-named) PHY_MODE_SATA, hiding
>>>> in the Marvell Berlin SATA PHY driver.
>>>>
>>>> Fix the build by:
>>>>
>>>> 1) Renaming Marvell's defined value to a more scoped name,
>>>> in order to avoid any potential conflicts: PHY_BERLIN_MODE_SATA.
>>>>
>>>> 2) Adding the missing enum, which was going to be added anyway as part
>>>> of [1].
>>>>
>>>> [1] https://lkml.kernel.org/r/[email protected]
>>>>
>>>> Fixes: 49e54187ae0b ("ata: libahci_platform: comply to PHY framework")
>>>
>>> Linus, this is probably a better option in terms of what should go in to
>>> fix that commit.
>>
>> I'm OK with this, but it does beg the question how the patch was
>> tested before submitting, if it didn't build.
>>
>> Is there functional breakage behind it? I currently lack online
>> hardware to test myself, unfortunately.
>
> This is my mistake, I forgot to tell Jens about this dependency,
> I am very sorry about that. As reported by John, this patch depends on
> the addition of PHY_MODE_SATA in the PHY type enumeration. This series
> ([1]) has been delayed and I should have warned Jens about it. I'm fine
> with the above fix though.
>
> Kishon, will you be able to base phy-next on top of this fix? It will
> be needed for the addition of the COMPHY driver.
Sure, I'll include this in my tree.
Thanks
Kishon
On 13/01/19 7:35 AM, Jens Axboe wrote:
> On 1/12/19 6:29 PM, [email protected] wrote:
>> From: John Hubbard <[email protected]>
>>
>> Commit 49e54187ae0b ("ata: libahci_platform: comply to PHY framework") uses
>> the PHY_MODE_SATA, but that enum had not yet been added. This caused a
>> build failure for me, with today's linux.git.
>>
>> Also, there is a potentially conflicting (mis-named) PHY_MODE_SATA, hiding
>> in the Marvell Berlin SATA PHY driver.
>>
>> Fix the build by:
>>
>> 1) Renaming Marvell's defined value to a more scoped name,
>> in order to avoid any potential conflicts: PHY_BERLIN_MODE_SATA.
>>
>> 2) Adding the missing enum, which was going to be added anyway as part
>> of [1].
>>
>> [1] https://lkml.kernel.org/r/[email protected]
>>
>> Fixes: 49e54187ae0b ("ata: libahci_platform: comply to PHY framework")
>
> Linus, this is probably a better option in terms of what should go in to
> fix that commit.
>
>>
>> Cc: Grzegorz Jaszczyk <[email protected]>
>> Cc: Miquel Raynal <[email protected]>
>> Cc: Hans de Goede <[email protected]>
>> Cc: Jens Axboe <[email protected]>
>> Signed-off-by: John Hubbard <[email protected]>
Acked-by: Kishon Vijay Abraham I <[email protected]>
>> ---
>> drivers/phy/marvell/phy-berlin-sata.c | 5 +++--
>> include/linux/phy/phy.h | 1 +
>> 2 files changed, 4 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/phy/marvell/phy-berlin-sata.c b/drivers/phy/marvell/phy-berlin-sata.c
>> index a91fc67fc4e0..d70ba9bc42d9 100644
>> --- a/drivers/phy/marvell/phy-berlin-sata.c
>> +++ b/drivers/phy/marvell/phy-berlin-sata.c
>> @@ -32,7 +32,7 @@
>>
>> /* register 0x01 */
>> #define REF_FREF_SEL_25 BIT(0)
>> -#define PHY_MODE_SATA (0x0 << 5)
>> +#define PHY_BERLIN_MODE_SATA (0x0 << 5)
>>
>> /* register 0x02 */
>> #define USE_MAX_PLL_RATE BIT(12)
>> @@ -102,7 +102,8 @@ static int phy_berlin_sata_power_on(struct phy *phy)
>>
>> /* set PHY mode and ref freq to 25 MHz */
>> phy_berlin_sata_reg_setbits(ctrl_reg, priv->phy_base, 0x01,
>> - 0x00ff, REF_FREF_SEL_25 | PHY_MODE_SATA);
>> + 0x00ff,
>> + REF_FREF_SEL_25 | PHY_BERLIN_MODE_SATA);
>>
>> /* set PHY up to 6 Gbps */
>> phy_berlin_sata_reg_setbits(ctrl_reg, priv->phy_base, 0x25,
>> diff --git a/include/linux/phy/phy.h b/include/linux/phy/phy.h
>> index e8e118d70fd7..3f350e2749fe 100644
>> --- a/include/linux/phy/phy.h
>> +++ b/include/linux/phy/phy.h
>> @@ -42,6 +42,7 @@ enum phy_mode {
>> PHY_MODE_PCIE,
>> PHY_MODE_ETHERNET,
>> PHY_MODE_MIPI_DPHY,
>> + PHY_MODE_SATA
>> };
>>
>> /**
>>
>
>