diff mbox series

[master,v2,2/2] mvebu: armada-37xx: add patch to forbid cpufreq for 1.2 GHz

Message ID 20210727112043.2864485-2-pepe.schlehofer@gmail.com
State Accepted
Delegated to: Hauke Mehrtens
Headers show
Series Revert "mvebu: 5.10 fix DVFS caused random boot crashes" | expand

Commit Message

Josef Schlehofer July 27, 2021, 11:20 a.m. UTC 
This patch is backported from linux-arm-kernel [1] to improve situation, when
it was reported that 1.2 GHz variant is unstable with DFS.
It waits to be accepted upstream, however, it waits for Marvell people to respond.

[1] https://patchwork.kernel.org/project/linux-arm-kernel/patch/20210630225601.6372-1-kabel@kernel.org/

Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
---
 ...x-forbid-cpufreq-for-1.2-GHz-variant.patch | 39 +++++++++++++++++++
 ...x-forbid-cpufreq-for-1.2-GHz-variant.patch | 39 +++++++++++++++++++
 2 files changed, 78 insertions(+)
 create mode 100644 target/linux/mvebu/patches-5.10/101-cpufreq-armada-37xx-forbid-cpufreq-for-1.2-GHz-variant.patch
 create mode 100644 target/linux/mvebu/patches-5.4/100-cpufreq-armada-37xx-forbid-cpufreq-for-1.2-GHz-variant.patch
diff mbox series

Patch

diff --git a/target/linux/mvebu/patches-5.10/101-cpufreq-armada-37xx-forbid-cpufreq-for-1.2-GHz-variant.patch b/target/linux/mvebu/patches-5.10/101-cpufreq-armada-37xx-forbid-cpufreq-for-1.2-GHz-variant.patch
new file mode 100644
index 0000000000..6a5b8220f8
--- /dev/null
+++ b/target/linux/mvebu/patches-5.10/101-cpufreq-armada-37xx-forbid-cpufreq-for-1.2-GHz-variant.patch
@@ -0,0 +1,39 @@ 
+From: =?utf-8?q?Marek_Beh=C3=BAn?= <kabel@kernel.org>
+Subject: [PATCH v2] cpufreq: armada-37xx: forbid cpufreq for 1.2 GHz variant
+Date: Thu,  1 Jul 2021 00:56:01 +0200
+
+The 1.2 GHz variant of the Armada 3720 SOC is unstable with DVFS: when
+the SOC boots, the WTMI firmware sets clocks and AVS values that work
+correctly with 1.2 GHz CPU frequency, but random crashes occur once
+cpufreq driver starts scaling.
+
+We do not know currently what is the reason:
+- it may be that the voltage value for L0 for 1.2 GHz variant provided
+  by the vendor in the OTP is simply incorrect when scaling is used,
+- it may be that some delay is needed somewhere,
+- it may be something else.
+
+The most sane solution now seems to be to simply forbid the cpufreq
+driver on 1.2 GHz variant.
+
+Signed-off-by: Marek Behún <kabel@kernel.org>
+Fixes: 92ce45fb875d ("cpufreq: Add DVFS support for Armada 37xx")
+---
+ drivers/cpufreq/armada-37xx-cpufreq.c | 6 +++++-
+ 1 file changed, 5 insertions(+), 1 deletion(-)
+
+--- a/drivers/cpufreq/armada-37xx-cpufreq.c
++++ b/drivers/cpufreq/armada-37xx-cpufreq.c
+@@ -102,7 +102,11 @@ struct armada_37xx_dvfs {
+ };
+ 
+ static struct armada_37xx_dvfs armada_37xx_dvfs[] = {
+-	{.cpu_freq_max = 1200*1000*1000, .divider = {1, 2, 4, 6} },
++	/*
++	 * The cpufreq scaling for 1.2 GHz variant of the SOC is currently
++	 * unstable because we do not know how to configure it properly.
++	 */
++	/* {.cpu_freq_max = 1200*1000*1000, .divider = {1, 2, 4, 6} }, */
+ 	{.cpu_freq_max = 1000*1000*1000, .divider = {1, 2, 4, 5} },
+ 	{.cpu_freq_max = 800*1000*1000,  .divider = {1, 2, 3, 4} },
+ 	{.cpu_freq_max = 600*1000*1000,  .divider = {2, 4, 5, 6} },
diff --git a/target/linux/mvebu/patches-5.4/100-cpufreq-armada-37xx-forbid-cpufreq-for-1.2-GHz-variant.patch b/target/linux/mvebu/patches-5.4/100-cpufreq-armada-37xx-forbid-cpufreq-for-1.2-GHz-variant.patch
new file mode 100644
index 0000000000..6a5b8220f8
--- /dev/null
+++ b/target/linux/mvebu/patches-5.4/100-cpufreq-armada-37xx-forbid-cpufreq-for-1.2-GHz-variant.patch
@@ -0,0 +1,39 @@ 
+From: =?utf-8?q?Marek_Beh=C3=BAn?= <kabel@kernel.org>
+Subject: [PATCH v2] cpufreq: armada-37xx: forbid cpufreq for 1.2 GHz variant
+Date: Thu,  1 Jul 2021 00:56:01 +0200
+
+The 1.2 GHz variant of the Armada 3720 SOC is unstable with DVFS: when
+the SOC boots, the WTMI firmware sets clocks and AVS values that work
+correctly with 1.2 GHz CPU frequency, but random crashes occur once
+cpufreq driver starts scaling.
+
+We do not know currently what is the reason:
+- it may be that the voltage value for L0 for 1.2 GHz variant provided
+  by the vendor in the OTP is simply incorrect when scaling is used,
+- it may be that some delay is needed somewhere,
+- it may be something else.
+
+The most sane solution now seems to be to simply forbid the cpufreq
+driver on 1.2 GHz variant.
+
+Signed-off-by: Marek Behún <kabel@kernel.org>
+Fixes: 92ce45fb875d ("cpufreq: Add DVFS support for Armada 37xx")
+---
+ drivers/cpufreq/armada-37xx-cpufreq.c | 6 +++++-
+ 1 file changed, 5 insertions(+), 1 deletion(-)
+
+--- a/drivers/cpufreq/armada-37xx-cpufreq.c
++++ b/drivers/cpufreq/armada-37xx-cpufreq.c
+@@ -102,7 +102,11 @@ struct armada_37xx_dvfs {
+ };
+ 
+ static struct armada_37xx_dvfs armada_37xx_dvfs[] = {
+-	{.cpu_freq_max = 1200*1000*1000, .divider = {1, 2, 4, 6} },
++	/*
++	 * The cpufreq scaling for 1.2 GHz variant of the SOC is currently
++	 * unstable because we do not know how to configure it properly.
++	 */
++	/* {.cpu_freq_max = 1200*1000*1000, .divider = {1, 2, 4, 6} }, */
+ 	{.cpu_freq_max = 1000*1000*1000, .divider = {1, 2, 4, 5} },
+ 	{.cpu_freq_max = 800*1000*1000,  .divider = {1, 2, 3, 4} },
+ 	{.cpu_freq_max = 600*1000*1000,  .divider = {2, 4, 5, 6} },