On Wed, Feb 04, 2015 at 09:38:25AM +0000, Hanjun Guo wrote:

On 2015?02?04? 01:20, Catalin Marinas wrote:

quoted

On Mon, Feb 02, 2015 at 12:45:37PM +0000, Hanjun Guo wrote:

quoted

diff --git a/arch/arm64/kernel/acpi.c b/arch/arm64/kernel/acpi.c
index afe10b4..b9f64ec 100644
--- a/arch/arm64/kernel/acpi.c
+++ b/arch/arm64/kernel/acpi.c

@@ -13,6 +13,8 @@
   *  published by the Free Software Foundation.
   */

+#define pr_fmt(fmt) "ACPI: " fmt
+
  #include <linux/acpi.h>
  #include <linux/bootmem.h>
  #include <linux/cpumask.h>

@@ -49,10 +51,32 @@ void __init __acpi_unmap_table(char *map, unsigned long size)
  	early_memunmap(map, size);
  }

+static int __init acpi_parse_fadt(struct acpi_table_header *table)
+{
+	struct acpi_table_fadt *fadt = (struct acpi_table_fadt *)table;
+
+	/*
+	 * Revision in table header is the FADT Major revision, and there
+	 * is a minor revision of FADT which was introduced by ACPI 5.1,
+	 * we only deal with ACPI 5.1 or newer revision to get GIC and SMP
+	 * boot protocol configuration data, or we will disable ACPI.
+	 */
+	if (table->revision > 5 ||
+	    (table->revision == 5 && fadt->minor_revision >= 1))
+		return 0;
+
+	pr_warn("Unsupported FADT revision %d.%d, should be 5.1+, will disable ACPI\n",
+		table->revision, fadt->minor_revision);
+	disable_acpi();
+
+	return -EINVAL;
+}
+
  /*
   * acpi_boot_table_init() called from setup_arch(), always.
   *	1. find RSDP and get its address, and then find XSDT
   *	2. extract all tables and checksums them all
+ *	3. check ACPI FADT revision
   *
   * We can parse ACPI boot-time tables such as MADT after
   * this function is called.

@@ -64,8 +88,16 @@ void __init acpi_boot_table_init(void)
  		return;

  	/* Initialize the ACPI boot-time table parser. */
-	if (acpi_table_init())
+	if (acpi_table_init()) {
+		disable_acpi();
+		return;
+	}
+
+	if (acpi_table_parse(ACPI_SIG_FADT, acpi_parse_fadt)) {
+		/* disable ACPI if no FADT is found */
  		disable_acpi();
+		pr_err("Can't find FADT\n");
+	}
  }

It looks fine to call disable_acpi() here but a bit weird to call it
again in acpi_parse_fadt(). I guess that's because acpi_table_parse()
ignores the return value of the handler() call. I think it's better to
fix the core code (can be an additional patch on top of this series).

I checked all the code calling acpi_table_parse() and I found that it
will be no functional change if we return the value of handler(), but
I need Rafael's confirm on it.

Are you sure ? All calls to acpi_table_parse() that checks the return
value are affected. I guess that depends on what an error return from
the handler means, from acpi_table_parse():

* Return 0 if table found, -errno if not.

So, if table is found but parsing fails that acpi_table_parse()
signature should be changed if the handler barfs with an error and
it is propagated. Still, I share Catalin's comment.

Have you thought about getting the FADT through:

acpi_get_table_with_size()

and check the revision there instead of going through acpi_table_parse()
for that ?

I wonder if the revision information is not already available without
needing to retrieve the FADT again.

On top of that, this patch should be squashed, I have a feeling that
between patch 4 and 9, there is a window where ACPI versions predating
5.1 are ok on arm64, which is not the case. I do not think that's a
bisectable issue, but keep this in mind please.

Thanks,
Lorenzo