| Message ID | 20210902164558.1920849-28-michael@walle.cc |
|---|---|
| State | Superseded |
| Delegated to: | Priyanka Jain |
| Headers | show |
| Series | arm: dts: ls1028a: sync device tree with linux | expand |
On Thu, Sep 02, 2021 at 06:45:56PM +0200, Michael Walle wrote: > The linux device tree hasn't set this property. To be similarly we remove s/similarly/similar/ > it from the u-boot device tree, too. The second controller of the > LS1028A has indeed no card detect pin. Because it is an eMMC controller. > The present state register of the > second controller will always report the card as present. Thus, there > should be no functional change otherwise than one more register access > to read the present state. Tested? > The property should still be introduced in > the linux device tree. How do you feel about dropping this patch? > > Signed-off-by: Michael Walle <michael@walle.cc> > --- > arch/arm/dts/fsl-ls1028a.dtsi | 1 - > 1 file changed, 1 deletion(-) > > diff --git a/arch/arm/dts/fsl-ls1028a.dtsi b/arch/arm/dts/fsl-ls1028a.dtsi > index d2f558d208..af6dabe847 100644 > --- a/arch/arm/dts/fsl-ls1028a.dtsi > +++ b/arch/arm/dts/fsl-ls1028a.dtsi > @@ -210,7 +210,6 @@ > reg = <0x0 0x2150000 0x0 0x10000>; > interrupts = <GIC_SPI 63 IRQ_TYPE_LEVEL_HIGH>; > big-endian; > - non-removable; > bus-width = <4>; > status = "disabled"; > }; > -- > 2.30.2 >
Am 2021-09-15 02:17, schrieb Vladimir Oltean: > On Thu, Sep 02, 2021 at 06:45:56PM +0200, Michael Walle wrote: >> The linux device tree hasn't set this property. To be similarly we >> remove > > s/similarly/similar/ > >> it from the u-boot device tree, too. The second controller of the >> LS1028A has indeed no card detect pin. > > Because it is an eMMC controller. > >> The present state register of the >> second controller will always report the card as present. Thus, there >> should be no functional change otherwise than one more register access >> to read the present state. > > Tested? Yes, I've tested the eMMC (and actually using it this way). I also manually read out the register value and made sure the card detect bit is set. >> The property should still be introduced in >> the linux device tree. > > How do you feel about dropping this patch? If I drop this patch, I'll either have to (1) keep this property the device tree sync patch too or (2) it will be removed without a comment in that patch. I presume you prefer (1), but as I said, I'd keep the device trees between linux and u-boot the same and keep it simple by just copying the files. So yes, I'd like to keep this patch as the point of this patch is to give an explanation why this property is removed (until it might or might not be introduced in the kernel device tree; in fact I thought I had send a patch to the LKML but I must have forgotten it). I was thinking of putting it into the -u-boot.dtsi, but as there is no breakage, I didn't. Thanks for reviewing the patches! -michael >> Signed-off-by: Michael Walle <michael@walle.cc> >> --- >> arch/arm/dts/fsl-ls1028a.dtsi | 1 - >> 1 file changed, 1 deletion(-) >> >> diff --git a/arch/arm/dts/fsl-ls1028a.dtsi >> b/arch/arm/dts/fsl-ls1028a.dtsi >> index d2f558d208..af6dabe847 100644 >> --- a/arch/arm/dts/fsl-ls1028a.dtsi >> +++ b/arch/arm/dts/fsl-ls1028a.dtsi >> @@ -210,7 +210,6 @@ >> reg = <0x0 0x2150000 0x0 0x10000>; >> interrupts = <GIC_SPI 63 IRQ_TYPE_LEVEL_HIGH>; >> big-endian; >> - non-removable; >> bus-width = <4>; >> status = "disabled"; >> }; >> -- >> 2.30.2 >>
On Wed, Sep 15, 2021 at 10:09:45AM +0200, Michael Walle wrote: > Am 2021-09-15 02:17, schrieb Vladimir Oltean: > > On Thu, Sep 02, 2021 at 06:45:56PM +0200, Michael Walle wrote: > > > The linux device tree hasn't set this property. To be similarly we > > > remove > > > > s/similarly/similar/ > > > > > it from the u-boot device tree, too. The second controller of the > > > LS1028A has indeed no card detect pin. > > > > Because it is an eMMC controller. > > > > > The present state register of the > > > second controller will always report the card as present. Thus, there > > > should be no functional change otherwise than one more register access > > > to read the present state. > > > > Tested? > > Yes, I've tested the eMMC (and actually using it this way). I also > manually read out the register value and made sure the card detect > bit is set. > > > > The property should still be introduced in > > > the linux device tree. > > > > How do you feel about dropping this patch? > > If I drop this patch, I'll either have to (1) keep this property > the device tree sync patch too or (2) it will be removed without a > comment in that patch. > > I presume you prefer (1), but as I said, I'd keep the device trees > between linux and u-boot the same and keep it simple by just copying > the files. So yes, I'd like to keep this patch as the point of this > patch is to give an explanation why this property is removed (until > it might or might not be introduced in the kernel device tree; in > fact I thought I had send a patch to the LKML but I must have > forgotten it). > > I was thinking of putting it into the -u-boot.dtsi, but as there > is no breakage, I didn't. Ok, but we should still put "non-removable" in the Linux dtsi soon then.
Am 2021-09-15 12:36, schrieb Vladimir Oltean: > On Wed, Sep 15, 2021 at 10:09:45AM +0200, Michael Walle wrote: >> I was thinking of putting it into the -u-boot.dtsi, but as there >> is no breakage, I didn't. > > Ok, but we should still put "non-removable" in the Linux dtsi soon > then. Just for the reference: https://git.kernel.org/pub/scm/linux/kernel/git/shawnguo/linux.git/commit/?h=for-next&id=8b94aa318aa746fbbc668d6b9b3ad812c835230c -michael
diff --git a/arch/arm/dts/fsl-ls1028a.dtsi b/arch/arm/dts/fsl-ls1028a.dtsi index d2f558d208..af6dabe847 100644 --- a/arch/arm/dts/fsl-ls1028a.dtsi +++ b/arch/arm/dts/fsl-ls1028a.dtsi @@ -210,7 +210,6 @@ reg = <0x0 0x2150000 0x0 0x10000>; interrupts = <GIC_SPI 63 IRQ_TYPE_LEVEL_HIGH>; big-endian; - non-removable; bus-width = <4>; status = "disabled"; };
The linux device tree hasn't set this property. To be similarly we remove it from the u-boot device tree, too. The second controller of the LS1028A has indeed no card detect pin. The present state register of the second controller will always report the card as present. Thus, there should be no functional change otherwise than one more register access to read the present state. The property should still be introduced in the linux device tree. Signed-off-by: Michael Walle <michael@walle.cc> --- arch/arm/dts/fsl-ls1028a.dtsi | 1 - 1 file changed, 1 deletion(-)