From: Conor Dooley <conor.dooley@microchip.com>
When the binding was originally written, it was assumed that all
ax45mp-caches had the same properties etc. This has turned out to be
incorrect, as the QiLai SoC has a different number of cache-sets.
Add a specific compatible for the RZ/Five for property enforcement and
in case there turns out to be additional differences between these
implementations of the cache controller.
Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
---
.../devicetree/bindings/cache/andestech,ax45mp-cache.yaml | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/Documentation/devicetree/bindings/cache/andestech,ax45mp-cache.yaml b/Documentation/devicetree/bindings/cache/andestech,ax45mp-cache.yaml
index d2cbe49f4e15f..f533bd178a9f6 100644
--- a/Documentation/devicetree/bindings/cache/andestech,ax45mp-cache.yaml
+++ b/Documentation/devicetree/bindings/cache/andestech,ax45mp-cache.yaml
@@ -28,6 +28,7 @@ select:
properties:
compatible:
items:
+ - const: renesas,r9a07g043f-cache
- const: andestech,ax45mp-cache
- const: cache
@@ -70,7 +71,7 @@ examples:
#include <dt-bindings/interrupt-controller/irq.h>
cache-controller@13400000 {
- compatible = "andestech,ax45mp-cache", "cache";
+ compatible = "renesas,r9a07g043f-cache", "andestech,ax45mp-cache", "cache";
reg = <0x13400000 0x100000>;
interrupts = <508 IRQ_TYPE_LEVEL_HIGH>;
cache-line-size = <64>;
--
2.45.2
Hi Conor,
On Fri, 9 May 2025 at 17:39, Conor Dooley <conor@kernel.org> wrote:
> From: Conor Dooley <conor.dooley@microchip.com>
>
> When the binding was originally written, it was assumed that all
> ax45mp-caches had the same properties etc. This has turned out to be
> incorrect, as the QiLai SoC has a different number of cache-sets.
>
> Add a specific compatible for the RZ/Five for property enforcement and
> in case there turns out to be additional differences between these
> implementations of the cache controller.
>
> Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
Thanks for your patch!
> --- a/Documentation/devicetree/bindings/cache/andestech,ax45mp-cache.yaml
> +++ b/Documentation/devicetree/bindings/cache/andestech,ax45mp-cache.yaml
> @@ -28,6 +28,7 @@ select:
> properties:
> compatible:
> items:
> + - const: renesas,r9a07g043f-cache
This name looks a bit too generic to me, as this is not a generic
cache on the R9A07G043F SoC, but specific to the CPU cores.
> - const: andestech,ax45mp-cache
> - const: cache
>
> @@ -70,7 +71,7 @@ examples:
> #include <dt-bindings/interrupt-controller/irq.h>
>
> cache-controller@13400000 {
> - compatible = "andestech,ax45mp-cache", "cache";
> + compatible = "renesas,r9a07g043f-cache", "andestech,ax45mp-cache", "cache";
> reg = <0x13400000 0x100000>;
> interrupts = <508 IRQ_TYPE_LEVEL_HIGH>;
> cache-line-size = <64>;
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
On Mon, May 12, 2025 at 11:01:26AM +0200, Geert Uytterhoeven wrote: > Hi Conor, > > On Fri, 9 May 2025 at 17:39, Conor Dooley <conor@kernel.org> wrote: > > From: Conor Dooley <conor.dooley@microchip.com> > > > > When the binding was originally written, it was assumed that all > > ax45mp-caches had the same properties etc. This has turned out to be > > incorrect, as the QiLai SoC has a different number of cache-sets. > > > > Add a specific compatible for the RZ/Five for property enforcement and > > in case there turns out to be additional differences between these > > implementations of the cache controller. > > > > Signed-off-by: Conor Dooley <conor.dooley@microchip.com> > > Thanks for your patch! > > > --- a/Documentation/devicetree/bindings/cache/andestech,ax45mp-cache.yaml > > +++ b/Documentation/devicetree/bindings/cache/andestech,ax45mp-cache.yaml > > @@ -28,6 +28,7 @@ select: > > properties: > > compatible: > > items: > > + - const: renesas,r9a07g043f-cache > > This name looks a bit too generic to me, as this is not a generic > cache on the R9A07G043F SoC, but specific to the CPU cores. So "reneasas,r9...-cpu-cache"?
Hi Conor and Geert, On Mon, May 12, 2025 at 10:59 AM Conor Dooley <conor@kernel.org> wrote: > > On Mon, May 12, 2025 at 11:01:26AM +0200, Geert Uytterhoeven wrote: > > Hi Conor, > > > > On Fri, 9 May 2025 at 17:39, Conor Dooley <conor@kernel.org> wrote: > > > From: Conor Dooley <conor.dooley@microchip.com> > > > > > > When the binding was originally written, it was assumed that all > > > ax45mp-caches had the same properties etc. This has turned out to be > > > incorrect, as the QiLai SoC has a different number of cache-sets. > > > > > > Add a specific compatible for the RZ/Five for property enforcement and > > > in case there turns out to be additional differences between these > > > implementations of the cache controller. > > > > > > Signed-off-by: Conor Dooley <conor.dooley@microchip.com> > > > > Thanks for your patch! > > > > > --- a/Documentation/devicetree/bindings/cache/andestech,ax45mp-cache.yaml > > > +++ b/Documentation/devicetree/bindings/cache/andestech,ax45mp-cache.yaml > > > @@ -28,6 +28,7 @@ select: > > > properties: > > > compatible: > > > items: > > > + - const: renesas,r9a07g043f-cache > > > > This name looks a bit too generic to me, as this is not a generic > > cache on the R9A07G043F SoC, but specific to the CPU cores. > > So "reneasas,r9...-cpu-cache"? Maybe "renesas,r9a07g043f-riscv-cache" ? Cheers, Prabhakar
On Mon, 12 May 2025 at 12:05, Lad, Prabhakar <prabhakar.csengg@gmail.com> wrote:
> On Mon, May 12, 2025 at 10:59 AM Conor Dooley <conor@kernel.org> wrote:
> > On Mon, May 12, 2025 at 11:01:26AM +0200, Geert Uytterhoeven wrote:
> > > On Fri, 9 May 2025 at 17:39, Conor Dooley <conor@kernel.org> wrote:
> > > > From: Conor Dooley <conor.dooley@microchip.com>
> > > >
> > > > When the binding was originally written, it was assumed that all
> > > > ax45mp-caches had the same properties etc. This has turned out to be
> > > > incorrect, as the QiLai SoC has a different number of cache-sets.
> > > >
> > > > Add a specific compatible for the RZ/Five for property enforcement and
> > > > in case there turns out to be additional differences between these
> > > > implementations of the cache controller.
> > > >
> > > > Signed-off-by: Conor Dooley <conor.dooley@microchip.com>
> > >
> > > Thanks for your patch!
> > >
> > > > --- a/Documentation/devicetree/bindings/cache/andestech,ax45mp-cache.yaml
> > > > +++ b/Documentation/devicetree/bindings/cache/andestech,ax45mp-cache.yaml
> > > > @@ -28,6 +28,7 @@ select:
> > > > properties:
> > > > compatible:
> > > > items:
> > > > + - const: renesas,r9a07g043f-cache
> > >
> > > This name looks a bit too generic to me, as this is not a generic
> > > cache on the R9A07G043F SoC, but specific to the CPU cores.
> >
> > So "reneasas,r9...-cpu-cache"?
>
> Maybe "renesas,r9a07g043f-riscv-cache" ?
"renesas,r9a07g043f-ax45mp-cache"?
There don't seem to be many vendor-specific derivatives of standardized
caches, except for "brcm,bcm11351-a2-pl310-cache".
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
Hi Geert, On Mon, May 12, 2025 at 12:05 PM Geert Uytterhoeven <geert@linux-m68k.org> wrote: > > On Mon, 12 May 2025 at 12:05, Lad, Prabhakar <prabhakar.csengg@gmail.com> wrote: > > On Mon, May 12, 2025 at 10:59 AM Conor Dooley <conor@kernel.org> wrote: > > > On Mon, May 12, 2025 at 11:01:26AM +0200, Geert Uytterhoeven wrote: > > > > On Fri, 9 May 2025 at 17:39, Conor Dooley <conor@kernel.org> wrote: > > > > > From: Conor Dooley <conor.dooley@microchip.com> > > > > > > > > > > When the binding was originally written, it was assumed that all > > > > > ax45mp-caches had the same properties etc. This has turned out to be > > > > > incorrect, as the QiLai SoC has a different number of cache-sets. > > > > > > > > > > Add a specific compatible for the RZ/Five for property enforcement and > > > > > in case there turns out to be additional differences between these > > > > > implementations of the cache controller. > > > > > > > > > > Signed-off-by: Conor Dooley <conor.dooley@microchip.com> > > > > > > > > Thanks for your patch! > > > > > > > > > --- a/Documentation/devicetree/bindings/cache/andestech,ax45mp-cache.yaml > > > > > +++ b/Documentation/devicetree/bindings/cache/andestech,ax45mp-cache.yaml > > > > > @@ -28,6 +28,7 @@ select: > > > > > properties: > > > > > compatible: > > > > > items: > > > > > + - const: renesas,r9a07g043f-cache > > > > > > > > This name looks a bit too generic to me, as this is not a generic > > > > cache on the R9A07G043F SoC, but specific to the CPU cores. > > > > > > So "reneasas,r9...-cpu-cache"? > > > > Maybe "renesas,r9a07g043f-riscv-cache" ? > > "renesas,r9a07g043f-ax45mp-cache"? > Sounds good to me. Cheers, Prabhakar
On Mon, May 12, 2025 at 01:05:13PM +0200, Geert Uytterhoeven wrote: > On Mon, 12 May 2025 at 12:05, Lad, Prabhakar <prabhakar.csengg@gmail.com> wrote: > > On Mon, May 12, 2025 at 10:59 AM Conor Dooley <conor@kernel.org> wrote: > > > On Mon, May 12, 2025 at 11:01:26AM +0200, Geert Uytterhoeven wrote: > > > > On Fri, 9 May 2025 at 17:39, Conor Dooley <conor@kernel.org> wrote: > > > > > From: Conor Dooley <conor.dooley@microchip.com> > > > > > > > > > > When the binding was originally written, it was assumed that all > > > > > ax45mp-caches had the same properties etc. This has turned out to be > > > > > incorrect, as the QiLai SoC has a different number of cache-sets. > > > > > > > > > > Add a specific compatible for the RZ/Five for property enforcement and > > > > > in case there turns out to be additional differences between these > > > > > implementations of the cache controller. > > > > > > > > > > Signed-off-by: Conor Dooley <conor.dooley@microchip.com> > > > > > > > > Thanks for your patch! > > > > > > > > > --- a/Documentation/devicetree/bindings/cache/andestech,ax45mp-cache.yaml > > > > > +++ b/Documentation/devicetree/bindings/cache/andestech,ax45mp-cache.yaml > > > > > @@ -28,6 +28,7 @@ select: > > > > > properties: > > > > > compatible: > > > > > items: > > > > > + - const: renesas,r9a07g043f-cache > > > > > > > > This name looks a bit too generic to me, as this is not a generic > > > > cache on the R9A07G043F SoC, but specific to the CPU cores. > > > > > > So "reneasas,r9...-cpu-cache"? > > > > Maybe "renesas,r9a07g043f-riscv-cache" ? > > "renesas,r9a07g043f-ax45mp-cache"? > > There don't seem to be many vendor-specific derivatives of standardized > caches, except for "brcm,bcm11351-a2-pl310-cache". The sifive stuff is all "vendor,soc-cache" into "sifive,ccache" but there's little ambiguity about there being an arm version of the same soc there. I don't mind the "renesas,r9...-ax45mp-cache" one you suggested, feels better than "riscv" to me.
On Fri, May 09, 2025 at 04:37:57PM +0100, Conor Dooley wrote: > [EXTERNAL MAIL] > > From: Conor Dooley <conor.dooley@microchip.com> > > When the binding was originally written, it was assumed that all > ax45mp-caches had the same properties etc. This has turned out to be > incorrect, as the QiLai SoC has a different number of cache-sets. > > Add a specific compatible for the RZ/Five for property enforcement and > in case there turns out to be additional differences between these > implementations of the cache controller. > > Signed-off-by: Conor Dooley <conor.dooley@microchip.com> Acked-by: Ben Zong-You Xie <ben717@andestech.com>
© 2016 - 2025 Red Hat, Inc.