[PATCH] arm64: Relax the kernel cache requirements for boot

[Catalin Marinas]

With system caches for the host OS or architected caches for guest OS we
cannot easily guarantee that there are no dirty or stale cache lines for
the areas of memory written by the kernel during boot with the MMU off
(therefore non-cacheable accesses).

This patch adds the necessary cache maintenance during boot and relaxes
the booting requirements.

Signed-off-by: Catalin Marinas <catalin.marinas at arm.com>
---

This patch is required at least for EFI_STUB on platforms with a system
cache (and I really hope it's one that handles standard cache ops by VA)
and for guests running UEFI or some boot-loader (with or without system
caches) that disable the MMU during boot. Later on, probably useful for
kexec as well.

In addition, commit 82b2f495fba3 (arm64: virt: ensure visibility of
__boot_cpu_mode) will be reverted as it is no longer needed after this
patch.

 Documentation/arm64/booting.txt | 10 ++++++++--
 arch/arm64/kernel/head.S        | 30 ++++++++++++++++++++++++++++--
 arch/arm64/mm/cache.S           |  9 +++++++++
 3 files changed, 45 insertions(+), 4 deletions(-)

diff --git a/Documentation/arm64/booting.txt b/Documentation/arm64/booting.txt
index a9691cc48fe3..beb754e87c65 100644
--- a/Documentation/arm64/booting.txt
+++ b/Documentation/arm64/booting.txt
@@ -111,8 +111,14 @@ Before jumping into the kernel, the following conditions must be met:
 - Caches, MMUs
   The MMU must be off.
   Instruction cache may be on or off.
-  Data cache must be off and invalidated.
-  External caches (if present) must be configured and disabled.
+  The address range corresponding to the loaded kernel image must be
+  cleaned to the PoC. In the presence of a system cache or other
+  coherent masters with caches enabled, this will typically require
+  cache maintenance by VA rather than set/way operations.
+  System caches which respect the architected cache maintenance by VA
+  operations must be configured and may be enabled.
+  System caches which do not respect architected cache maintenance by VA
+  operations (not recommended) must be configured and disabled.
 
 - Architected timers
   CNTFRQ must be programmed with the timer frequency and CNTVOFF must
diff --git a/arch/arm64/kernel/head.S b/arch/arm64/kernel/head.S
index 61035d6814cb..26109682d2fa 100644
--- a/arch/arm64/kernel/head.S
+++ b/arch/arm64/kernel/head.S
@@ -26,6 +26,7 @@
 #include <asm/assembler.h>
 #include <asm/ptrace.h>
 #include <asm/asm-offsets.h>
+#include <asm/cache.h>
 #include <asm/cputype.h>
 #include <asm/memory.h>
 #include <asm/thread_info.h>
@@ -229,7 +230,11 @@ ENTRY(set_cpu_boot_mode_flag)
 	cmp	w20, #BOOT_CPU_MODE_EL2
 	b.ne	1f
 	add	x1, x1, #4
-1:	str	w20, [x1]			// This CPU has booted in EL1
+1:	dc	cvac, x1			// Clean potentially dirty cache line
+	dsb	sy
+	str	w20, [x1]			// This CPU has booted in EL1
+	dc	civac, x1			// Clean&invalidate potentially stale cache line
+	dsb	sy
 	ret
 ENDPROC(set_cpu_boot_mode_flag)
 
@@ -240,8 +245,9 @@ ENDPROC(set_cpu_boot_mode_flag)
  * This is not in .bss, because we set it sufficiently early that the boot-time
  * zeroing of .bss would clobber it.
  */
-	.pushsection	.data
+	.pushsection	.data..cacheline_aligned
 ENTRY(__boot_cpu_mode)
+	.align	L1_CACHE_SHIFT
 	.long	BOOT_CPU_MODE_EL2
 	.long	0
 	.popsection
@@ -408,6 +414,15 @@ ENDPROC(__calc_phys_offset)
  */
 __create_page_tables:
 	pgtbl	x25, x26, x24			// idmap_pg_dir and swapper_pg_dir addresses
+	mov	x27, lr
+
+	/*
+	 * Invalidate the idmap and swapper page tables to avoid potential
+	 * dirty cache lines being evicted.
+	 */
+	mov	x0, x25
+	add	x1, x26, #SWAPPER_DIR_SIZE
+	bl	__inval_cache_range
 
 	/*
 	 * Clear the idmap and swapper page tables.
@@ -470,6 +485,17 @@ __create_page_tables:
 	add	x0, x26, #2 * PAGE_SIZE		// section table address
 	create_pgd_entry x26, x0, x5, x6, x7
 #endif
+
+	/*
+	 * Since the page tables have been populated with non-cacheable
+	 * accesses (MMU disabled), invalidate the idmap and swapper page
+	 * tables again to remove any speculatively loaded cache lines.
+	 */
+	mov	x0, x25
+	add	x1, x26, #SWAPPER_DIR_SIZE
+	bl	__inval_cache_range
+
+	mov	lr, x27
 	ret
 ENDPROC(__create_page_tables)
 	.ltorg
diff --git a/arch/arm64/mm/cache.S b/arch/arm64/mm/cache.S
index c46f48b33c14..e803a62e0e45 100644
--- a/arch/arm64/mm/cache.S
+++ b/arch/arm64/mm/cache.S
@@ -168,6 +168,14 @@ ENTRY(__flush_dcache_area)
 ENDPROC(__flush_dcache_area)
 
 /*
+ *	__inval_cache_range(start, end)
+ *	- start   - start address of region
+ *	- end     - end address of region
+ */
+ENTRY(__inval_cache_range)
+	/* FALLTHROUGH */
+
+/*
  *	__dma_inv_range(start, end)
  *	- start   - virtual start address of region
  *	- end     - virtual end address of region
@@ -183,6 +191,7 @@ __dma_inv_range:
 	b.lo	1b
 	dsb	sy
 	ret
+ENDPROC(__inval_cache_range)
 ENDPROC(__dma_inv_range)
 
 /*