ARM: tegra: Always lock the CPU reset vector
authorThierry Reding <treding@nvidia.com>
Fri, 11 Jul 2014 09:06:20 +0000 (11:06 +0200)
committerThierry Reding <treding@nvidia.com>
Thu, 17 Jul 2014 12:58:42 +0000 (14:58 +0200)
Currently the reset vector is not locked on Tegra20 because the hardware
doesn't support it. However in order not to depend on the chip ID, which
becomes available only later in the boot process, we set the bit anyway.

Signed-off-by: Thierry Reding <treding@nvidia.com>
arch/arm/mach-tegra/reset.c

index 5377495d41b86e6afd7fbbc7e09748bb4980da02..894c5c472184f9cf9c08f34966b83ff982766939 100644 (file)
@@ -54,12 +54,10 @@ static void __init tegra_cpu_reset_handler_set(const u32 reset_address)
         * Prevent further modifications to the physical reset vector.
         *  NOTE: Has no effect on chips prior to Tegra30.
         */
-       if (tegra_get_chip_id() != TEGRA20) {
-               reg = readl(sb_ctrl);
-               reg |= 2;
-               writel(reg, sb_ctrl);
-               wmb();
-       }
+       reg = readl(sb_ctrl);
+       reg |= 2;
+       writel(reg, sb_ctrl);
+       wmb();
 }
 
 static void __init tegra_cpu_reset_handler_enable(void)
This page took 0.037771 seconds and 5 git commands to generate.