Thread (9 messages) 9 messages, 2 authors, 2025-12-24

Re: [PATCH v2 4/4] arm64: dts: amlogic: Add S7D Reset Controller

From: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Date: 2025-12-24 23:34:43
Also in: linux-amlogic, linux-devicetree, lkml

On Tue, Dec 23, 2025 at 6:37 AM Xianwei Zhao via B4 Relay
[off-list ref] wrote:
[...]
+#define RESET_BRG_A55_PIPE0            166
For S7 (without D suffix) this is called RESET_BRG_A53_PIPE0 - which
one is correct (looking at amlogic-s7.dtsi and amlogic-s7d.dtsi
possibly A55 is correct)?
unrelated side-note while I was checking amlogic-s7.dtsi and
amlogic-s7d.dtsi: why do we have D/I and L2 caches defined for one but
not the other?

[...]
+#define RESET_BRG_NIC_EMMC             183
+/*                                     164     */
this needs to state 184

On a related side-note: while reviewing this patch I'm wondering how
S7 differs from S7D.
Is S7 a cut-down version of the S7D SoC (e.g. some - possibly faulty -
IP blocks fused off), is S7D a slightly updated version (like rev 2.0)
of S7, ...?
What I'm missing is a "big picture" (for older SoCs we typically had
some SBC vendor publish a datasheet - for these newer SoCs there's no
public datasheets).

That said, I think the approach chosen for the reset controller (copy
& paste, except 13 reset lines) is fine because there seem to be
actual differences in the reset lines. For other
drivers/implementations this approach would likely get very hard to
review/maintain (e.g. if we end up with a 3000 line driver and there's
only 250 lines difference between the two SoCs).


Best regards,
Martin
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help