Hi,
On 2023/2/23 02:32, Krzysztof Kozlowski wrote:
On 22/02/2023 17:55, suijingfeng wrote:
The display controller is a pci device, it's pci vendor id is
0x0014, it's pci device id is 0x7a06.
Signed-off-by: suijingfeng <suijingfeng@xxxxxxxxxxx>
---
.../boot/dts/loongson/loongson64-2k1000.dtsi | 21 +++++++++++++++++++
1 file changed, 21 insertions(+)
diff --git a/arch/mips/boot/dts/loongson/loongson64-2k1000.dtsi b/arch/mips/boot/dts/loongson/loongson64-2k1000.dtsi
index 8143a61111e3..a528af3977d9 100644
--- a/arch/mips/boot/dts/loongson/loongson64-2k1000.dtsi
+++ b/arch/mips/boot/dts/loongson/loongson64-2k1000.dtsi
@@ -31,6 +31,18 @@ memory@200000 {
<0x00000001 0x10000000 0x00000001 0xb0000000>; /* 6912 MB at 4352MB */
};
+ reserved-memory {
+ #address-cells = <2>;
+ #size-cells = <2>;
+ ranges;
+
+ display_reserved: framebuffer@30000000 {
+ compatible = "shared-dma-pool";
+ reg = <0x0 0x30000000 0x0 0x04000000>; /* 64M */
+ linux,cma-default;
+ };
+ };
+
cpu_clk: cpu_clk {
#clock-cells = <0>;
compatible = "fixed-clock";
@@ -198,6 +210,15 @@ sata@8,0 {
interrupt-parent = <&liointc0>;
};
+ display-controller@6,0 {
+ compatible = "loongson,ls2k1000-dc";
+
+ reg = <0x3000 0x0 0x0 0x0 0x0>;
+ interrupts = <28 IRQ_TYPE_LEVEL_LOW>;
+ interrupt-parent = <&liointc0>;
+ memory-region = <&display_reserved>;
NAK.
Err :(, please give me a chance to explain
Test your code against the bindings you send.
I can guarantee to you that I test may code more than twice. The code
used to testing is listed at link [1].
This patchset mainly used to illustrate how we made the driver in [1]
usable on our SoC platform.
It's the same
patchset. You basically send something which the same moment is incorrect.
Loongson display controller IP has been integrated in both Loongson
North Bridge chipset(ls7a1000 and ls7a2000) and Loongson SoCs(ls2k1000
and ls2k2000 etc), it even has been included in Loongson BMC(ls2k0500 bmc)
products.
When use this driver on Loongson embedded platform(say ls2k2000,
ls2k1000 and ls2k0500) ,
the PMON/Uboot firmware(my company using pmon most of time) will pass a
DT to the kernel.
Different boards will pass different DTs. But when using this driver on
Loongson server and
PC platform( ls3c5000/ls3a5000+ls7a1000/ls7a2000), there will no DT
supplied. The firmware
and kernel side developer of Loongson choose ACPI+UEFI for such
platform, more discussion
can be found at [2]. Therefore, on such a situation we decide to send
the patch at separate patchset.
It is not like the arm and risc-v, as the binding would not be always
exits. If we put those patches
into a same patchset, some reviewers would suggest us to revise our code.
To a form that the code *ALWAYS* probed from the DT, this is not desired.
Besides, the driver code + dt support is petty large, separate it is
more easy to review and manage.
Finally, Thanks your kindly guiding and valuable reviews.
[1] https://patchwork.freedesktop.org/patch/523409/?series=113566&rev=4
[2] https://lkml.org/lkml/2022/7/15/135
Best regards,
Krzysztof
