Skip to contentSkip to conversation

platform/x86: hp-wmi: fix heap OOB read and memory corruption in hp_wmi_perform_query()

PATCH In review Muhammad Bilal 2 participants 3 messages Aug 26 – Oct 5

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 platform/x86: hp-wmi: fix heap OOB read and memory corruption in hp_wmi_perform_query()

hp_wmi_perform_query() evaluates an ACPI WMI method and expects a struct bios_return header in the returned buffer. However, it only checks that the ACPI object type is ACPI_TYPE_BUFFER without verifying that obj->buffer.length is at least sizeof(*bios_return) (8 bytes).

If the firmware or ACPI method returns a buffer shorter than 8 bytes:

1. bios_return->return_code at offset 4 reads out of bounds from the allocated ACPI buffer slab.

2. If return_code evaluates to 0 and outsize is non-zero, (int)(obj->buffer.length - sizeof(*bios_return)) evaluates to a negative integer (e.g. -8 for a 0-length buffer). actual_outsize is set to this negative value and passed to memcpy(buffer, ..., actual_outsize), where the signed negative value is cast to size_t (e.g. 0xfffffffffffffff8), attempting to copy ~18 Exabytes and triggering an immediate kernel crash. Additionally, memset(buffer + actual_outsize, ...) performs an out-of-bounds write before buffer.

Fix this by ensuring obj->buffer.length is at least sizeof(*bios_return) before accessing the return code, and compute actual_outsize safely using min_t(u32, ...), which is sound specifically because the guard above already rules out obj->buffer.length being smaller than sizeof(*bios_return) by the time that subtraction runs.

Fixes c3021ea1beee ("hp-wmi: allow setting input and output buffer sizes separately")

---
v2: Rebased onto current platform-drivers-x86.git 'fixes' (confirmed
matching current mainline, blob b2773fc1aca414792b99406315496a2c3a64dd06). v1 was diffed against a stale local copy of hp-wmi.c that predated a batch of unrelated upstream changes (GPU MUX switch support, new OMEN/Victus board IDs), so its hunks no longer applied. The fix itself is unchanged, only the base and line numbers moved. Also correcting the recipient list: Hans de Goede's address is [email protected], not [email protected] as v1 had it.

 drivers/platform/x86/hp/hp-wmi.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/platform/x86/hp/hp-wmi.c +4 -3
index b2773fc..d7ad206 100644
@@ -691,8 +691,9 @@ static int hp_wmi_perform_query(int query, enum hp_wmi_command command,
 		goto out_free;
 	}
 
-	if (obj->type != ACPI_TYPE_BUFFER) {
-		pr_warn("query 0x%x returned an invalid object 0x%x\n", query, ret);
+	if (obj->type != ACPI_TYPE_BUFFER ||
+	    obj->buffer.length < sizeof(*bios_return)) {
+		pr_warn("query 0x%x returned wrong type or too small buffer\n", query);
 		ret = -EINVAL;
 		goto out_free;
 	}
@@ -711,7 +712,7 @@ static int hp_wmi_perform_query(int query, enum hp_wmi_command command,
 	if (!outsize)
 		goto out_free;
 
-	actual_outsize = min(outsize, (int)(obj->buffer.length - sizeof(*bios_return)));
+	actual_outsize = min_t(u32, outsize, obj->buffer.length - sizeof(*bios_return));
 	memcpy(buffer, obj->buffer.pointer + sizeof(*bios_return), actual_outsize);
 	memset(buffer + actual_outsize, 0, outsize - actual_outsize);
 
Signature · 1 line
2.55.0
#1 lore raw

ping.

Quoted 71 lines · Muhammad Bilal

hp_wmi_perform_query() evaluates an ACPI WMI method and expects a struct bios_return header in the returned buffer. However, it only checks that the ACPI object type is ACPI_TYPE_BUFFER without verifying that obj->buffer.length is at least sizeof(*bios_return) (8 bytes).

If the firmware or ACPI method returns a buffer shorter than 8 bytes:

1. bios_return->return_code at offset 4 reads out of bounds from the allocated ACPI buffer slab.

2. If return_code evaluates to 0 and outsize is non-zero, (int)(obj->buffer.length - sizeof(*bios_return)) evaluates to a negative integer (e.g. -8 for a 0-length buffer). actual_outsize is set to this negative value and passed to memcpy(buffer, ..., actual_outsize), where the signed negative value is cast to size_t (e.g. 0xfffffffffffffff8), attempting to copy ~18 Exabytes and triggering an immediate kernel crash. Additionally, memset(buffer + actual_outsize, ...) performs an out-of-bounds write before buffer.

Fix this by ensuring obj->buffer.length is at least sizeof(*bios_return) before accessing the return code, and compute actual_outsize safely using min_t(u32, ...), which is sound specifically because the guard above already rules out obj->buffer.length being smaller than sizeof(*bios_return) by the time that subtraction runs.

Fixes c3021ea1beee ("hp-wmi: allow setting input and output buffer sizes separately")

---
v2: Rebased onto current platform-drivers-x86.git 'fixes' (confirmed
matching current mainline, blob b2773fc1aca414792b99406315496a2c3a64dd06). v1 was diffed against a stale local copy of hp-wmi.c that predated a batch of unrelated upstream changes (GPU MUX switch support, new OMEN/Victus board IDs), so its hunks no longer applied. The fix itself is unchanged, only the base and line numbers moved. Also correcting the recipient list: Hans de Goede's address is [email protected], not [email protected] as v1 had it.

 drivers/platform/x86/hp/hp-wmi.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/platform/x86/hp/hp-wmi.c +4 -4
index b2773fc..d7ad206 100644
@@ -691,8 +691,9 @@ static int hp_wmi_perform_query(int query, enum hp_wmi_command command,
                goto out_free;
        }
 
-       if (obj->type != ACPI_TYPE_BUFFER) {
-               pr_warn("query 0x%x returned an invalid object 0x%x\n", query, ret);
+       if (obj->type != ACPI_TYPE_BUFFER ||
+           obj->buffer.length < sizeof(*bios_return)) {
+               pr_warn("query 0x%x returned wrong type or too small buffer\n", query);
                ret = -EINVAL;
                goto out_free;
        }
@@ -711,7 +712,7 @@ static int hp_wmi_perform_query(int query, enum hp_wmi_command command,
        if (!outsize)
                goto out_free;
 
-       actual_outsize = min(outsize, (int)(obj->buffer.length - sizeof(*bios_return)));
+       actual_outsize = min_t(u32, outsize, obj->buffer.length - sizeof(*bios_return));
        memcpy(buffer, obj->buffer.pointer + sizeof(*bios_return), actual_outsize);
        memset(buffer + actual_outsize, 0, outsize - actual_outsize);
 
--

2.55.0

maintainer
#2 lore raw
In reply to Muhammad Bilal
Quoted 8 lines · Muhammad Bilal

hp_wmi_perform_query() evaluates an ACPI WMI method and expects a struct bios_return header in the returned buffer. However, it only checks that the ACPI object type is ACPI_TYPE_BUFFER without verifying that obj->buffer.length is at least sizeof(*bios_return) (8 bytes).

If the firmware or ACPI method returns a buffer shorter than 8 bytes:

[...]

Thank you for your contribution, it has been applied to my local review-ilpo-next branch. Note it will show up in the public platform-drivers-x86/review-ilpo-next branch only once I've pushed my local branch there, which might take a while.

FYI [if applicable to your patch], as per Linus' policy change, also fixes are mostly routed through for-next unless the fix is for a commit introduced in the most recent cycle or is clearly a regression fix.

The list of commits applied:
[1/1] platform/x86: hp-wmi: fix heap OOB read and memory corruption in hp_wmi_perform_query()

      commit: c59252fa6260423ec3c4aa90b144fcad66851c76

--
i.