Hi Jonas,
On 1/8/26 12:07 AM, Jonas Karlman wrote:
The RK3582 SoC is a variant of the RK3588S with some IP blocks disabled.
What blocks are disabled/non-working is indicated by ip-state in OTP.
This add initial support for RK3582 by using ft_system_setup() to mark
any cpu, gpu and/or vdec/venc node with status=fail as indicated by
ip-state.
This apply same policy as vendor U-Boot for RK3582, i.e. two big cpu
cores, the gpu and one vdec/venc core is always failed/disabled.
Enable Kconfig option OF_SYSTEM_SETUP in board defconfig to make use of
the required DT fixups for RK3582 board variants.
Signed-off-by: Jonas Karlman <[email protected]>
---
Changes in v4:
- Update policy to always fail gpu for RK3582
- Drop code and comment related to RK3583
- Update rkvdec node name to match latest mainline Linux DT patches
Should we really patch the rkvdec nodes if they aren't stable yet? But I
see this is something I already reported on v3 and said it's up to Kever :)
Should we support booting downstream kernel with (I assume) its
different paths?
Changes in v3:
- Apply same policy for RK3582/RK3583 to match vendor U-Boot
linux-6.1-stan-rkr6 tag, allow use of the GPU and one vdec core.
- Update rkvdec node name to match latest mainline Linux DT patches.
Changes in v2:
- Refactor code to first apply policy to ip-state and then fail cores
based on the updated ip-state
- Add support for failing rkvdec and rkvenc cores
- Append soc compatible to board model
- Not picking up t-b tag due to significant changes to patch
---
arch/arm/mach-rockchip/rk3588/rk3588.c | 215 +++++++++++++++++++++++++
1 file changed, 215 insertions(+)
diff --git a/arch/arm/mach-rockchip/rk3588/rk3588.c
b/arch/arm/mach-rockchip/rk3588/rk3588.c
index 6324c6f12867..c1138ffcb87f 100644
--- a/arch/arm/mach-rockchip/rk3588/rk3588.c
+++ b/arch/arm/mach-rockchip/rk3588/rk3588.c
@@ -7,6 +7,7 @@
#define LOG_CATEGORY LOGC_ARCH
#include <dm.h>
+#include <fdt_support.h>
#include <misc.h>
#include <spl.h>
#include <asm/armv8/mmu.h>
@@ -240,6 +241,16 @@ int rockchip_dram_init_banksize_fixup(struct bd_info *bd)
#define RK3588_OTP_CPU_CODE_OFFSET 0x02
#define RK3588_OTP_SPECIFICATION_OFFSET 0x06
+#define RK3588_OTP_IP_STATE_OFFSET 0x1d
+
+#define FAIL_CPU_CLUSTER0 GENMASK(3, 0)
+#define FAIL_CPU_CLUSTER1 GENMASK(5, 4)
+#define FAIL_CPU_CLUSTER2 GENMASK(7, 6)
+#define FAIL_GPU GENMASK(4, 1)
+#define FAIL_RKVDEC0 BIT(6)
+#define FAIL_RKVDEC1 BIT(7)
+#define FAIL_RKVENC0 BIT(0)
+#define FAIL_RKVENC1 BIT(2)
int checkboard(void)
{
@@ -285,3 +296,207 @@ int checkboard(void)
return 0;
}
+
+static int fdt_path_del_node(void *fdt, const char *path)
+{
+ int nodeoffset;
+
+ nodeoffset = fdt_path_offset(fdt, path);
+ if (nodeoffset < 0)
+ return nodeoffset;
+
+ return fdt_del_node(fdt, nodeoffset);
+}
+
+static int fdt_path_set_name(void *fdt, const char *path, const char *name)
+{
+ int nodeoffset;
+
+ nodeoffset = fdt_path_offset(fdt, path);
+ if (nodeoffset < 0)
+ return nodeoffset;
+
+ return fdt_set_name(fdt, nodeoffset, name);
+}
+
Wondering if we shouldn't have a name that is a bit less close to what
you would find as API in include/fdt_support.h or libfdt? SO there's no
confusion when reading the code where there are callers to this function?
+/*
+ * RK3582 is a variant of the RK3588S with some IP blocks disabled. What blocks
+ * are disabled/non-working is indicated by ip-state in OTP. ft_system_setup()
+ * is used to mark any cpu, gpu and/or vdec/venc node with status=fail as
+ * indicated by ip-state. Apply same policy as vendor U-Boot for RK3582, i.e.
+ * two big cpu cores, the gpu and one vdec/venc core is always failed. Enable
+ * OF_SYSTEM_SETUP to use the required DT fixups for RK3582 board variants.
+ */
+int ft_system_setup(void *blob, struct bd_info *bd)
+{
+ static const char * const cpu_node_names[] = {
+ "cpu@0", "cpu@100", "cpu@200", "cpu@300",
+ "cpu@400", "cpu@500", "cpu@600", "cpu@700",
+ };
+ int parent, node, i, comp_len, len, ret;
+ bool cluster1_removed = false;
+ u8 cpu_code[2], ip_state[3];
+ struct udevice *dev;
+ char soc_comp[16];
+ const char *comp;
+ void *data;
+
+ if (!IS_ENABLED(CONFIG_OF_SYSTEM_SETUP))
+ return 0;
+
Isn't that guaranteed to be the case as ft_system_setup is only called
when CONFIG_OF_SYSTEM_SETUP is set? What are you trying to prevent (e.g.
compilation warnings?).
+ if (!IS_ENABLED(CONFIG_ROCKCHIP_OTP) || !CONFIG_IS_ENABLED(MISC))
+ return -ENOSYS;
+
This means that if one enables OF_SYSTEM_SETUP but not ROCKCHIP_OTP or
MISC, then this will fail the boot even on RK3588(s), correct?
Should we add a select/depends on in Kconfig somewhere so this cannot
happen?
+ ret = uclass_get_device_by_driver(UCLASS_MISC,
+ DM_DRIVER_GET(rockchip_otp), &dev);
+ if (ret) {
+ log_debug("Could not find otp device, ret=%d\n", ret);
+ return ret;
+ }
+
+ /* cpu-code: SoC model, e.g. 0x35 0x82 or 0x35 0x88 */
+ ret = misc_read(dev, RK3588_OTP_CPU_CODE_OFFSET, cpu_code, 2);
+ if (ret < 0) {
+ log_debug("Could not read cpu-code, ret=%d\n", ret);
+ return ret;
+ }
+
+ log_debug("cpu-code: %02x %02x\n", cpu_code[0], cpu_code[1]);
+
+ /* only fail cores on rk3582 */
+ if (!(cpu_code[0] == 0x35 && cpu_code[1] == 0x82))
+ return 0;
+
+ ret = misc_read(dev, RK3588_OTP_IP_STATE_OFFSET, &ip_state, 3);
+ if (ret < 0) {
+ log_err("Could not read ip-state, ret=%d\n", ret);
+ return ret;
+ }
+
+ log_debug("ip-state: %02x %02x %02x (otp)\n",
+ ip_state[0], ip_state[1], ip_state[2]);
+
+ /* policy: fail entire big core cluster when one or more core is bad */
+ if (ip_state[0] & FAIL_CPU_CLUSTER1)
+ ip_state[0] |= FAIL_CPU_CLUSTER1;
+ if (ip_state[0] & FAIL_CPU_CLUSTER2)
+ ip_state[0] |= FAIL_CPU_CLUSTER2;
+
+ /* policy: always fail one big core cluster on rk3582 */
+ if (!(ip_state[0] & (FAIL_CPU_CLUSTER1 | FAIL_CPU_CLUSTER2)))
+ ip_state[0] |= FAIL_CPU_CLUSTER2;
+
+ /* policy: always fail gpu on rk3582 */
+ ip_state[1] |= FAIL_GPU;
+
+ /* policy: always fail one rkvdec core on rk3582 */
+ if (!(ip_state[1] & (FAIL_RKVDEC0 | FAIL_RKVDEC1)))
+ ip_state[1] |= FAIL_RKVDEC1;
+
+ /* policy: always fail one rkvenc core on rk3582 */
+ if (!(ip_state[2] & (FAIL_RKVENC0 | FAIL_RKVENC1)))
+ ip_state[2] |= FAIL_RKVENC1;
+
+ log_debug("ip-state: %02x %02x %02x (policy)\n",
+ ip_state[0], ip_state[1], ip_state[2]);
+
+ /* cpu cluster1: ip_state[0]: bit4~5 */
+ if ((ip_state[0] & FAIL_CPU_CLUSTER1) == FAIL_CPU_CLUSTER1) {
+ log_debug("remove cpu-map cluster1\n");
+ fdt_path_del_node(blob, "/cpus/cpu-map/cluster1");
Check the return code and log (+error out?) if there's an error.
+ cluster1_removed = true;
+ }
+
+ /* cpu cluster2: ip_state[0]: bit6~7 */
+ if ((ip_state[0] & FAIL_CPU_CLUSTER2) == FAIL_CPU_CLUSTER2) {
+ log_debug("remove cpu-map cluster2\n");
+ fdt_path_del_node(blob, "/cpus/cpu-map/cluster2");
Check the return code and log (+error out?) if there's an error.
+ } else if (cluster1_removed) {
+ /* cluster nodes must be named in a continuous series */
+ log_debug("rename cpu-map cluster2\n");
+ fdt_path_set_name(blob, "/cpus/cpu-map/cluster2", "cluster1");
Check the return code and log (+error out?) if there's an error.
+ }
+
+ /* gpu: ip_state[1]: bit1~4 */
+ if (ip_state[1] & FAIL_GPU) {
+ log_debug("fail gpu\n");
+ fdt_status_fail_by_pathf(blob, "/gpu@fb000000");
Check the return code and log (+error out?) if there's an error.
+ }
+
+ /* rkvdec: ip_state[1]: bit6,7 */
+ if (ip_state[1] & FAIL_RKVDEC0) {
+ log_debug("fail rkvdec0\n");
+ fdt_status_fail_by_pathf(blob, "/video-codec@fdc38000");
+ fdt_status_fail_by_pathf(blob, "/iommu@fdc38700");
Check the return code and log (+error out?) if there's an error.
+ }
+ if (ip_state[1] & FAIL_RKVDEC1) {
+ log_debug("fail rkvdec1\n");
+ fdt_status_fail_by_pathf(blob, "/video-codec@fdc40000");
+ fdt_status_fail_by_pathf(blob, "/iommu@fdc40700");
Check the return code and log (+error out?) if there's an error.
+ }
+
+ /* rkvenc: ip_state[2]: bit0,2 */
+ if (ip_state[2] & FAIL_RKVENC0) {
+ log_debug("fail rkvenc0\n");
+ fdt_status_fail_by_pathf(blob, "/video-codec@fdbd0000");
+ fdt_status_fail_by_pathf(blob, "/iommu@fdbdf000");
Check the return code and log (+error out?) if there's an error.
+ }
+ if (ip_state[2] & FAIL_RKVENC1) {
+ log_debug("fail rkvenc1\n");
+ fdt_status_fail_by_pathf(blob, "/video-codec@fdbe0000");
+ fdt_status_fail_by_pathf(blob, "/iommu@fdbef000");
Check the return code and log (+error out?) if there's an error.
+ }
+
+ parent = fdt_path_offset(blob, "/cpus");
+ if (parent < 0) {
+ log_err("Could not find /cpus, parent=%d\n", parent);
+ return parent;
+ }
+
+ /* cpu: ip_state[0]: bit0~7 */
+ for (i = 0; i < 8; i++) {
+ /* fail any bad cpu core */
+ if (!(ip_state[0] & BIT(i)))
Is this guaranteed to be properly set in sync with FAIL_CPU_CLUSTERx?
E.g. can I have a FAIL_CPU_CLUSTER2 but none (or only one) of the big
cores in the cluster not marked as bad in the ip-state OTP?
Shouldn't we check also for FAIL_CPU_CLUSTERx here and disable all cores
in bad clusters since we are removing the clusters anyway?
+ continue;
+
+ node = fdt_subnode_offset(blob, parent, cpu_node_names[i]);
+ if (node >= 0) {
+ log_debug("fail cpu %s\n", cpu_node_names[i]);
+ fdt_status_fail(blob, node);
Check the return code and log (+error out?) if there's an error.
+ } else {
+ log_err("Could not find %s, node=%d\n",
+ cpu_node_names[i], node);
+ return node;
+ }
suggestion: Flatten this into
if (node < 0) {
log_err();
return node;
}
log_debug();
fdt_status_fail(blob, node);
+ }
+
Please add a big comment here explaining what the below is for, so we
don't have to guess what the intent is based on the code :)
Cheers,
Quentin
