ARMv8: Allow dynamic early stack pointer

[u-boot] / arch / arm / lib / crt0_64.S

diff --git a/arch/arm/lib/crt0_64.S b/arch/arm/lib/crt0_64.S
index 77563967e517902863a530967f4c2ea02e891305..a181283e0fa9008162cfad63b5ddfd9adf98ed39 100644 (file)
--- a/arch/arm/lib/crt0_64.S
+++ b/arch/arm/lib/crt0_64.S
@@ -27,7 +27,8 @@
  *    the GD ('global data') structure, both located in some readily
  *    available RAM (SRAM, locked cache...). In this context, VARIABLE
  *    global data, initialized or not (BSS), are UNAVAILABLE; only
- *    CONSTANT initialized data are available.
+ *    CONSTANT initialized data are available. GD should be zeroed
+ *    before board_init_f() is called.
  *
  * 2. Call board_init_f(). This function prepares the hardware for
  *    execution from system RAM (DRAM, DDR...) As system RAM may not
@@ -36,24 +37,31 @@
  *    data include the relocation destination, the future stack, and
  *    the future GD location.
  *
- * (the following applies only to non-SPL builds)
- *
  * 3. Set up intermediate environment where the stack and GD are the
  *    ones allocated by board_init_f() in system RAM, but BSS and
  *    initialized non-const data are still not available.
  *
- * 4. Call relocate_code(). This function relocates U-Boot from its
- *    current location into the relocation destination computed by
- *    board_init_f().
+ * 4a.For U-Boot proper (not SPL), call relocate_code(). This function
+ *    relocates U-Boot from its current location into the relocation
+ *    destination computed by board_init_f().
+ *
+ * 4b.For SPL, board_init_f() just returns (to crt0). There is no
+ *    code relocation in SPL.
  *
  * 5. Set up final environment for calling board_init_r(). This
  *    environment has BSS (initialized to 0), initialized non-const
  *    data (initialized to their intended value), and stack in system
- *    RAM. GD has retained values set by board_init_f(). Some CPUs
- *    have some work left to do at this point regarding memory, so
- *    call c_runtime_cpu_setup.
+ *    RAM (for SPL moving the stack and GD into RAM is optional - see
+ *    CONFIG_SPL_STACK_R). GD has retained values set by board_init_f().
+ *
+ * TODO: For SPL, implement stack relocation on AArch64.
  *
- * 6. Branch to board_init_r().
+ * 6. For U-Boot proper (not SPL), some CPUs have some work left to do
+ *    at this point regarding memory, so call c_runtime_cpu_setup.
+ *
+ * 7. Branch to board_init_r().
+ *
+ * For more information see 'Board Initialisation Flow in README.
  */
 
 ENTRY(_main)
@@ -61,13 +69,28 @@ ENTRY(_main)
 /*
  * Set up initial C runtime environment and call board_init_f(0).
  */
+#if defined(CONFIG_TPL_BUILD) && defined(CONFIG_TPL_NEEDS_SEPARATE_STACK)
+       ldr     x0, =(CONFIG_TPL_STACK)
+#elif defined(CONFIG_SPL_BUILD) && defined(CONFIG_SPL_STACK)
+       ldr     x0, =(CONFIG_SPL_STACK)
+#elif defined(CONFIG_SYS_INIT_SP_BSS_OFFSET)
+       adr     x0, __bss_start
+       add     x0, x0, #CONFIG_SYS_INIT_SP_BSS_OFFSET
+#else
        ldr     x0, =(CONFIG_SYS_INIT_SP_ADDR)
-       sub     x0, x0, #GD_SIZE        /* allocate one GD above SP */
+#endif
        bic     sp, x0, #0xf    /* 16-byte alignment for ABI compliance */
-       mov     x18, sp                 /* GD is above SP */
+       mov     x0, sp
+       bl      board_init_f_alloc_reserve
+       mov     sp, x0
+       /* set up gd here, outside any C code */
+       mov     x18, x0
+       bl      board_init_f_init_reserve
+
        mov     x0, #0
        bl      board_init_f
 
+#if !defined(CONFIG_SPL_BUILD)
 /*
  * Set up intermediate environment (new sp and gd) and call
  * relocate_code(addr_moni). Trick here is that we'll return
@@ -75,10 +98,17 @@ ENTRY(_main)
  */
        ldr     x0, [x18, #GD_START_ADDR_SP]    /* x0 <- gd->start_addr_sp */
        bic     sp, x0, #0xf    /* 16-byte alignment for ABI compliance */
-       ldr     x18, [x18, #GD_BD]              /* x18 <- gd->bd */
-       sub     x18, x18, #GD_SIZE              /* new GD is below bd */
+       ldr     x18, [x18, #GD_NEW_GD]          /* x18 <- gd->new_gd */
 
        adr     lr, relocation_return
+#if CONFIG_POSITION_INDEPENDENT
+       /* Add in link-vs-runtime offset */
+       adr     x0, _start              /* x0 <- Runtime value of _start */
+       ldr     x9, _TEXT_BASE          /* x9 <- Linked value of _start */
+       sub     x9, x9, x0              /* x9 <- Run-vs-link offset */
+       add     lr, lr, x9
+#endif
+       /* Add in link-vs-relocation offset */
        ldr     x9, [x18, #GD_RELOC_OFF]        /* x9 <- gd->reloc_off */
        add     lr, lr, x9      /* new return address after relocation */
        ldr     x0, [x18, #GD_RELOCADDR]        /* x0 <- gd->relocaddr */
@@ -90,16 +120,30 @@ relocation_return:
  * Set up final (full) environment
  */
        bl      c_runtime_cpu_setup             /* still call old routine */
+#endif /* !CONFIG_SPL_BUILD */
+#if defined(CONFIG_SPL_BUILD)
+       bl      spl_relocate_stack_gd           /* may return NULL */
+       /* set up gd here, outside any C code, if new stack is returned */
+       cmp     x0, #0
+       csel    x18, x0, x18, ne
+       /*
+        * Perform 'sp = (x0 != NULL) ? x0 : sp' while working
+        * around the constraint that conditional moves can not
+        * have 'sp' as an operand
+        */
+       mov     x1, sp
+       cmp     x0, #0
+       csel    x0, x0, x1, ne
+       mov     sp, x0
+#endif
 
 /*
  * Clear BSS section
  */
        ldr     x0, =__bss_start                /* this is auto-relocated! */
        ldr     x1, =__bss_end                  /* this is auto-relocated! */
-       mov     x2, #0
 clear_loop:
-       str     x2, [x0]
-       add     x0, x0, #8
+       str     xzr, [x0], #8
        cmp     x0, x1
        b.lo    clear_loop