[PATCH] ACPI / tables: Fix a NULL pointer dereference in acpi_table_initrd_scan()

[PATCH] ACPI / tables: Fix a NULL pointer dereference in acpi_table_initrd_scan()

[Zhou Qingyang]

In acpi_table_initrd_scan(), the return value of acpi_os_map_memory()
is assigned to table and there is a dereference of it after that.
acpi_os_map_memory() will return NULL on failure, which may lead to NULL
pointer dereference.

Fix this bug by adding a NULL check of table.

This bug was found by a static analyzer.

Builds with 'make allyesconfig' show no new warnings,
and our static analyzer no longer warns about this code.

Fixes: 5ae74f2cc2f1 ("ACPI / tables: Move table override mechanisms to tables.c")
Signed-off-by: Zhou Qingyang <zhou1615@umn.edu>
---
The analysis employs differential checking to identify inconsistent 
security operations (e.g., checks or kfrees) between two code paths 
and confirms that the inconsistent operations are not recovered in the
current function or the callers, so they constitute bugs. 

Note that, as a bug found by static analysis, it can be a false
positive or hard to trigger. Multiple researchers have cross-reviewed
the bug.

 drivers/acpi/tables.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/acpi/tables.c b/drivers/acpi/tables.c
index 8b10c192ed32..356e08c4015b 100644
--- a/drivers/acpi/tables.c
+++ b/drivers/acpi/tables.c
@@ -755,6 +755,9 @@ static void __init acpi_table_initrd_scan(void)
 	while (table_offset + ACPI_HEADER_SIZE <= all_tables_size) {
 		table = acpi_os_map_memory(acpi_tables_addr + table_offset,
 					   ACPI_HEADER_SIZE);
+		if (!table)
+			return;
+
 		if (table_offset + table->length > all_tables_size) {
 			acpi_os_unmap_memory(table, ACPI_HEADER_SIZE);
 			WARN_ON(1);
-- 
2.25.1

Re: [PATCH] ACPI / tables: Fix a NULL pointer dereference in acpi_table_initrd_scan()

[Greg KH]

On Tue, Jan 25, 2022 at 01:02:37AM +0800, Zhou Qingyang wrote:
> In acpi_table_initrd_scan(), the return value of acpi_os_map_memory()
> is assigned to table and there is a dereference of it after that.
> acpi_os_map_memory() will return NULL on failure, which may lead to NULL
> pointer dereference.
> 
> Fix this bug by adding a NULL check of table.
> 
> This bug was found by a static analyzer.
> 
> Builds with 'make allyesconfig' show no new warnings,
> and our static analyzer no longer warns about this code.
> 
> Fixes: 5ae74f2cc2f1 ("ACPI / tables: Move table override mechanisms to tables.c")
> Signed-off-by: Zhou Qingyang <zhou1615@umn.edu>
> ---
> The analysis employs differential checking to identify inconsistent 
> security operations (e.g., checks or kfrees) between two code paths 
> and confirms that the inconsistent operations are not recovered in the
> current function or the callers, so they constitute bugs. 
> 
> Note that, as a bug found by static analysis, it can be a false
> positive or hard to trigger. Multiple researchers have cross-reviewed
> the bug.
> 
>  drivers/acpi/tables.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/drivers/acpi/tables.c b/drivers/acpi/tables.c
> index 8b10c192ed32..356e08c4015b 100644
> --- a/drivers/acpi/tables.c
> +++ b/drivers/acpi/tables.c
> @@ -755,6 +755,9 @@ static void __init acpi_table_initrd_scan(void)
>  	while (table_offset + ACPI_HEADER_SIZE <= all_tables_size) {
>  		table = acpi_os_map_memory(acpi_tables_addr + table_offset,
>  					   ACPI_HEADER_SIZE);
> +		if (!table)
> +			return;
> +
>  		if (table_offset + table->length > all_tables_size) {
>  			acpi_os_unmap_memory(table, ACPI_HEADER_SIZE);
>  			WARN_ON(1);
> -- 
> 2.25.1
> 

As stated before, umn.edu is still not allowed to contribute to the
Linux kernel.  Please work with your administration to resolve this
issue.