Re: [PATCH v1 3/8] xen/riscv: introduce stack stuff

[Andrew Cooper <Andrew.Cooper3@xxxxxxxxxx>]

On 06/01/2023 1:14 pm, Oleksii Kurochko wrote:
> diff --git a/xen/arch/riscv/riscv64/head.S b/xen/arch/riscv/riscv64/head.S
> index 990edb70a0..ddc7104701 100644
> --- a/xen/arch/riscv/riscv64/head.S
> +++ b/xen/arch/riscv/riscv64/head.S
> @@ -1,4 +1,10 @@
>          .section .text.header, "ax", %progbits
>  
>  ENTRY(start)
> -        j  start
> +        la      sp, cpu0_boot_stack
> +        li      t0, PAGE_SIZE
> +        add     sp, sp, t0
> +
> +_start_hang:
> +        wfi
> +        j  _start_hang
> diff --git a/xen/arch/riscv/setup.c b/xen/arch/riscv/setup.c
> new file mode 100644
> index 0000000000..2c7dca1daa
> --- /dev/null
> +++ b/xen/arch/riscv/setup.c
> @@ -0,0 +1,6 @@
> +#include <xen/init.h>
> +#include <xen/compile.h>
> +
> +/* Xen stack for bringing up the first CPU. */
> +unsigned char __initdata cpu0_boot_stack[PAGE_SIZE]
> +    __aligned(PAGE_SIZE);

Ah, I didn't spot his when looking at the unified delta of the series.

You want most of patch 7 merged into this, so we end up with

+void __init noreturn start_xen(void)
+{
+    for ( ;; )
+        asm volatile ("wfi");
+
+    unreachable();
+}

in setup.c too.  That means we don't transiently add _start_hang just to
delete it 3 patches later.  (And it will fix Jan's observation about the
misaligned operand.)

Adding the early_printk("Hello from C env\n"); line wants merging into
patch 6.

~Andrew