riscv: Move early dtb mapping into the fixmap region

commit ef69d2559fe91f23d27a3d6fd640b5641787d22e upstream.

riscv establishes 2 virtual mappings:

- early_pg_dir maps the kernel which allows to discover the system
  memory
- swapper_pg_dir installs the final mapping (linear mapping included)

We used to map the dtb in early_pg_dir using DTB_EARLY_BASE_VA, and this
mapping was not carried over in swapper_pg_dir. It happens that
early_init_fdt_scan_reserved_mem() must be called before swapper_pg_dir is
setup otherwise we could allocate reserved memory defined in the dtb.
And this function initializes reserved_mem variable with addresses that
lie in the early_pg_dir dtb mapping: when those addresses are reused
with swapper_pg_dir, this mapping does not exist and then we trap.

The previous "fix" was incorrect as early_init_fdt_scan_reserved_mem()
must be called before swapper_pg_dir is set up otherwise we could
allocate in reserved memory defined in the dtb.

So move the dtb mapping in the fixmap region which is established in
early_pg_dir and handed over to swapper_pg_dir.

This patch had to be backported because:
- the documentation for sv57 is not present here (as sv48/57 are not
  present)

Fixes: 922b0375fc93 ("riscv: Fix memblock reservation for device tree blob")
Fixes: 8f3a2b4a96dc ("RISC-V: Move DT mapping outof fixmap")
Fixes: 50e63dd8ed92 ("riscv: fix reserved memory setup")
Reported-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/all/f8e67f82-103d-156c-deb0-d6d6e2756f5e@microchip.com/
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Tested-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20230329081932.79831-2-alexghiti@rivosinc.com
Cc: stable@vger.kernel.org # 6.1.x
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

1 files changed, 34 insertions, 27 deletions

diff --git a/arch/riscv/mm/init.c b/arch/riscv/mm/init.c
index 50a1b6edd491..5570c52deb0b 100644
--- a/arch/riscv/mm/init.c
+++ b/arch/riscv/mm/init.c
@@ -57,7 +57,6 @@ unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)]
 EXPORT_SYMBOL(empty_zero_page);
 
 extern char _start[];
-#define DTB_EARLY_BASE_VA      PGDIR_SIZE
 void *_dtb_early_va __initdata;
 uintptr_t _dtb_early_pa __initdata;
 
@@ -236,6 +235,14 @@ static void __init setup_bootmem(void)
 	set_max_mapnr(max_low_pfn - ARCH_PFN_OFFSET);
 
 	reserve_initrd_mem();
+
+	/*
+	 * No allocation should be done before reserving the memory as defined
+	 * in the device tree, otherwise the allocation could end up in a
+	 * reserved region.
+	 */
+	early_init_fdt_scan_reserved_mem();
+
 	/*
 	 * If DTB is built in, no need to reserve its memblock.
 	 * Otherwise, do reserve it but avoid using
@@ -279,9 +286,6 @@ pgd_t trampoline_pg_dir[PTRS_PER_PGD] __page_aligned_bss;
 static pte_t fixmap_pte[PTRS_PER_PTE] __page_aligned_bss;
 
 pgd_t early_pg_dir[PTRS_PER_PGD] __initdata __aligned(PAGE_SIZE);
-static p4d_t __maybe_unused early_dtb_p4d[PTRS_PER_P4D] __initdata __aligned(PAGE_SIZE);
-static pud_t __maybe_unused early_dtb_pud[PTRS_PER_PUD] __initdata __aligned(PAGE_SIZE);
-static pmd_t __maybe_unused early_dtb_pmd[PTRS_PER_PMD] __initdata __aligned(PAGE_SIZE);
 
 #ifdef CONFIG_XIP_KERNEL
 #define pt_ops			(*(struct pt_alloc_ops *)XIP_FIXUP(&pt_ops))
@@ -626,9 +630,6 @@ static void __init create_p4d_mapping(p4d_t *p4dp,
 #define trampoline_pgd_next	(pgtable_l5_enabled ?			\
 		(uintptr_t)trampoline_p4d : (pgtable_l4_enabled ?	\
 		(uintptr_t)trampoline_pud : (uintptr_t)trampoline_pmd))
-#define early_dtb_pgd_next	(pgtable_l5_enabled ?			\
-		(uintptr_t)early_dtb_p4d : (pgtable_l4_enabled ?	\
-		(uintptr_t)early_dtb_pud : (uintptr_t)early_dtb_pmd))
 #else
 #define pgd_next_t		pte_t
 #define alloc_pgd_next(__va)	pt_ops.alloc_pte(__va)
@@ -636,7 +637,6 @@ static void __init create_p4d_mapping(p4d_t *p4dp,
 #define create_pgd_next_mapping(__nextp, __va, __pa, __sz, __prot)	\
 	create_pte_mapping(__nextp, __va, __pa, __sz, __prot)
 #define fixmap_pgd_next		((uintptr_t)fixmap_pte)
-#define early_dtb_pgd_next	((uintptr_t)early_dtb_pmd)
 #define create_p4d_mapping(__pmdp, __va, __pa, __sz, __prot) do {} while(0)
 #define create_pud_mapping(__pmdp, __va, __pa, __sz, __prot) do {} while(0)
 #define create_pmd_mapping(__pmdp, __va, __pa, __sz, __prot) do {} while(0)
@@ -859,32 +859,28 @@ static void __init create_kernel_page_table(pgd_t *pgdir, bool early)
  * this means 2 PMD entries whereas for 32-bit kernel, this is only 1 PGDIR
  * entry.
  */
-static void __init create_fdt_early_page_table(pgd_t *pgdir, uintptr_t dtb_pa)
+static void __init create_fdt_early_page_table(pgd_t *pgdir,
+					       uintptr_t fix_fdt_va,
+					       uintptr_t dtb_pa)
 {
-#ifndef CONFIG_BUILTIN_DTB
 	uintptr_t pa = dtb_pa & ~(PMD_SIZE - 1);
 
-	create_pgd_mapping(early_pg_dir, DTB_EARLY_BASE_VA,
-			   IS_ENABLED(CONFIG_64BIT) ? early_dtb_pgd_next : pa,
-			   PGDIR_SIZE,
-			   IS_ENABLED(CONFIG_64BIT) ? PAGE_TABLE : PAGE_KERNEL);
-
-	if (pgtable_l5_enabled)
-		create_p4d_mapping(early_dtb_p4d, DTB_EARLY_BASE_VA,
-				   (uintptr_t)early_dtb_pud, P4D_SIZE, PAGE_TABLE);
-
-	if (pgtable_l4_enabled)
-		create_pud_mapping(early_dtb_pud, DTB_EARLY_BASE_VA,
-				   (uintptr_t)early_dtb_pmd, PUD_SIZE, PAGE_TABLE);
+#ifndef CONFIG_BUILTIN_DTB
+	/* Make sure the fdt fixmap address is always aligned on PMD size */
+	BUILD_BUG_ON(FIX_FDT % (PMD_SIZE / PAGE_SIZE));
 
-	if (IS_ENABLED(CONFIG_64BIT)) {
-		create_pmd_mapping(early_dtb_pmd, DTB_EARLY_BASE_VA,
+	/* In 32-bit only, the fdt lies in its own PGD */
+	if (!IS_ENABLED(CONFIG_64BIT)) {
+		create_pgd_mapping(early_pg_dir, fix_fdt_va,
+				   pa, MAX_FDT_SIZE, PAGE_KERNEL);
+	} else {
+		create_pmd_mapping(fixmap_pmd, fix_fdt_va,
 				   pa, PMD_SIZE, PAGE_KERNEL);
-		create_pmd_mapping(early_dtb_pmd, DTB_EARLY_BASE_VA + PMD_SIZE,
+		create_pmd_mapping(fixmap_pmd, fix_fdt_va + PMD_SIZE,
 				   pa + PMD_SIZE, PMD_SIZE, PAGE_KERNEL);
 	}
 
-	dtb_early_va = (void *)DTB_EARLY_BASE_VA + (dtb_pa & (PMD_SIZE - 1));
+	dtb_early_va = (void *)fix_fdt_va + (dtb_pa & (PMD_SIZE - 1));
 #else
 	/*
 	 * For 64-bit kernel, __va can't be used since it would return a linear
@@ -1054,7 +1050,8 @@ asmlinkage void __init setup_vm(uintptr_t dtb_pa)
 	create_kernel_page_table(early_pg_dir, true);
 
 	/* Setup early mapping for FDT early scan */
-	create_fdt_early_page_table(early_pg_dir, dtb_pa);
+	create_fdt_early_page_table(early_pg_dir,
+				    __fix_to_virt(FIX_FDT), dtb_pa);
 
 	/*
 	 * Bootime fixmap only can handle PMD_SIZE mapping. Thus, boot-ioremap
@@ -1096,6 +1093,16 @@ static void __init setup_vm_final(void)
 	u64 i;
 
 	/* Setup swapper PGD for fixmap */
+#if !defined(CONFIG_64BIT)
+	/*
+	 * In 32-bit, the device tree lies in a pgd entry, so it must be copied
+	 * directly in swapper_pg_dir in addition to the pgd entry that points
+	 * to fixmap_pte.
+	 */
+	unsigned long idx = pgd_index(__fix_to_virt(FIX_FDT));
+
+	set_pgd(&swapper_pg_dir[idx], early_pg_dir[idx]);
+#endif
 	create_pgd_mapping(swapper_pg_dir, FIXADDR_START,
 			   __pa_symbol(fixmap_pgd_next),
 			   PGDIR_SIZE, PAGE_TABLE);