Prev: linux-next: manual merge of the tegra tree with the arm tree
Next: [PATCH 02/18] Fix SVM VMCB reset
From: Andrew Morton on 12 Jul 2010 21:40
On Fri, 09 Jul 2010 12:37:36 +0800 Axel Lin <axel.lin(a)gmail.com> wrote:

> When acpi_evaluate_object() is passed ACPI_ALLOCATE_BUFFER,
> the caller must kfree the returned buffer if AE_OK is returned.
>
> Call Trace:
> wmab_execute
> -> wmi_evaluate_method
> -> acpi_evaluate_object
>
> Thus if callers of wmab_execute() pass ACPI_ALLOCATE_BUFFER,
> the return buffer must be kfreed if wmab_execute return AE_OK.
>
> Signed-off-by: Axel Lin <axel.lin(a)gmail.com>
> ---
> drivers/platform/x86/acer-wmi.c | 11 ++++++++++-
> 1 files changed, 10 insertions(+), 1 deletions(-)
>
> diff --git a/drivers/platform/x86/acer-wmi.c b/drivers/platform/x86/acer-wmi.c
> index 1ea6c43..3f44446 100644
> --- a/drivers/platform/x86/acer-wmi.c
> +++ b/drivers/platform/x86/acer-wmi.c
> @@ -555,6 +555,7 @@ static acpi_status AMW0_find_mailled(void)
> obj->buffer.length == sizeof(struct wmab_ret)) {
> ret = *((struct wmab_ret *) obj->buffer.pointer);
> } else {
> + kfree(out.pointer);
> return AE_ERROR;
> }
>
> @@ -598,6 +599,7 @@ static acpi_status AMW0_set_capabilities(void)
> obj->buffer.length == sizeof(struct wmab_ret)) {
> ret = *((struct wmab_ret *) obj->buffer.pointer);
> } else {
> + kfree(out.pointer);
> return AE_ERROR;
> }
>
> @@ -607,15 +609,22 @@ static acpi_status AMW0_set_capabilities(void)
> args.ebx = 2 << 8;
> args.ebx |= ACER_AMW0_BLUETOOTH_MASK;
>
> + /*
> + * It's ok to use existing buffer for next wmab_execute call.
> + * But we need to kfree(out.pointer) if next wmab_execute fail.
> + */
> status = wmab_execute(&args, &out);
> - if (ACPI_FAILURE(status))
> + if (ACPI_FAILURE(status)) {
> + kfree(out.pointer);
> return status;
> + }
>
> obj = (union acpi_object *) out.pointer;
> if (obj && obj->type == ACPI_TYPE_BUFFER
> && obj->buffer.length == sizeof(struct wmab_ret)) {
> ret = *((struct wmab_ret *) obj->buffer.pointer);
> } else {
> + kfree(out.pointer);
> return AE_ERROR;
> }

I think it's best to remove the multiple-return-points approach in
favour of the "goto single return point which undoes things" approach.
that way we reduce the risk of later introducing more leaks.

Please review:

From: Andrew Morton <akpm(a)linux-foundation.org>

avoid multiple return points
remove unneeded cast
remove unneeded initialisation of `status'

Cc: Alan Jenkins <alan-jenkins(a)tuffmail.co.uk>
Cc: Axel Lin <axel.lin(a)gmail.com>
Cc: Carlos Corbacho <carlos(a)strangeworlds.co.uk>
Cc: Matthew Garrett <mjg(a)redhat.com>
Cc: Thomas Renninger <trenn(a)suse.de>
Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org>
---

drivers/platform/x86/acer-wmi.c | 25 ++++++++++++-------------
1 file changed, 12 insertions(+), 13 deletions(-)

diff -puN drivers/platform/x86/acer-wmi.c~acer-wmi-fix-memory-leaks-in-wmab_execute-error-path-v2-fix drivers/platform/x86/acer-wmi.c
--- a/drivers/platform/x86/acer-wmi.c~acer-wmi-fix-memory-leaks-in-wmab_execute-error-path-v2-fix
+++ a/drivers/platform/x86/acer-wmi.c
@@ -571,7 +571,7 @@ static acpi_status AMW0_set_capabilities
{
struct wmab_args args;
struct wmab_ret ret;
- acpi_status status = AE_OK;
+ acpi_status status;
struct acpi_buffer out = { ACPI_ALLOCATE_BUFFER, NULL };
union acpi_object *obj;

@@ -594,13 +594,13 @@ static acpi_status AMW0_set_capabilities
if (ACPI_FAILURE(status))
return status;

- obj = (union acpi_object *) out.pointer;
+ obj = out.pointer;
if (obj && obj->type == ACPI_TYPE_BUFFER &&
obj->buffer.length == sizeof(struct wmab_ret)) {
ret = *((struct wmab_ret *) obj->buffer.pointer);
} else {
- kfree(out.pointer);
- return AE_ERROR;
+ status = AE_ERROR;
+ goto out;
}

if (ret.eax & 0x1)
@@ -614,25 +614,21 @@ static acpi_status AMW0_set_capabilities
* But we need to kfree(out.pointer) if next wmab_execute fail.
*/
status = wmab_execute(&args, &out);
- if (ACPI_FAILURE(status)) {
- kfree(out.pointer);
- return status;
- }
+ if (ACPI_FAILURE(status))
+ goto out;

obj = (union acpi_object *) out.pointer;
if (obj && obj->type == ACPI_TYPE_BUFFER
&& obj->buffer.length == sizeof(struct wmab_ret)) {
ret = *((struct wmab_ret *) obj->buffer.pointer);
} else {
- kfree(out.pointer);
- return AE_ERROR;
+ status = AE_ERROR;
+ goto out;
}

if (ret.eax & 0x1)
interface->capability |= ACER_CAP_BLUETOOTH;

- kfree(out.pointer);
-
/*
* This appears to be safe to enable, since all Wistron based laptops
* appear to use the same EC register for brightness, even if they
@@ -641,7 +637,10 @@ static acpi_status AMW0_set_capabilities
if (quirks->brightness >= 0)
interface->capability |= ACER_CAP_BRIGHTNESS;

- return AE_OK;
+ status = AE_OK;
+out:
+ kfree(out.pointer);
+ return status;
}

static struct wmi_interface AMW0_interface = {
_

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
 | 
Pages: 1
Prev: linux-next: manual merge of the tegra tree with the arm tree
Next: [PATCH 02/18] Fix SVM VMCB reset