linux-kernel - Re: [PATCH] media: qcom: camss: vfe: Fix out-of-bounds access in vfe_isr_reg

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <8d260d93-38cc-459d-b8f5-40bec5497277@linaro.org>
Date: Mon, 29 Dec 2025 00:12:44 +0000
From: Bryan O'Donoghue <bryan.odonoghue@...aro.org>
To: Alper Ak <alperyasinak1@...il.com>, hverkuil@...nel.org,
 mchehab@...nel.org
Cc: Robert Foss <rfoss@...nel.org>, Todor Tomov <todor.too@...il.com>,
 Vladimir Zapolskiy <vladimir.zapolskiy@...aro.org>,
 Jonathan Marek <jonathan@...ek.ca>, linux-media@...r.kernel.org,
 linux-arm-msm@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] media: qcom: camss: vfe: Fix out-of-bounds access in
 vfe_isr_reg_update()

On 27/12/2025 10:24, Alper Ak wrote:
> vfe_isr() iterates using MSM_VFE_IMAGE_MASTERS_NUM(7) as the loop
> bound and passes the index to vfe_isr_reg_update(). However,
> vfe->line[] array is defined with VFE_LINE_NUM_MAX(4):
> 
>      struct vfe_line line[VFE_LINE_NUM_MAX];
> 
> When index is 4, 5, 6, the access to vfe->line[line_id] exceeds
> the array bounds and resulting in out-of-bounds memory access.
> 
> Add a bounds check to ensure line_id is within valid range before
> accessing the array.
> 
> Fixes: 4edc8eae715c ("media: camss: Add initial support for VFE hardware version Titan 480")
> Signed-off-by: Alper Ak <alperyasinak1@...il.com>
> ---
>   drivers/media/platform/qcom/camss/camss-vfe-480.c | 3 +++
>   1 file changed, 3 insertions(+)
> 
> diff --git a/drivers/media/platform/qcom/camss/camss-vfe-480.c b/drivers/media/platform/qcom/camss/camss-vfe-480.c
> index 4feea590a47b..fac111135c78 100644
> --- a/drivers/media/platform/qcom/camss/camss-vfe-480.c
> +++ b/drivers/media/platform/qcom/camss/camss-vfe-480.c
> @@ -237,6 +237,9 @@ static void vfe_isr_reg_update(struct vfe_device *vfe, enum vfe_line_id line_id)
>   	struct vfe_output *output;
>   	unsigned long flags;
>   
> +	if (line_id >= VFE_LINE_NUM_MAX)
> +		return;
> +
>   	spin_lock_irqsave(&vfe->output_lock, flags);
>   	vfe_reg_update_clear(vfe, line_id);
>   

A logical bug.

I think we shouldn't be calling this function at all with a index that 
is too large though.

Something like this instead. An explicit loop for the output lines and 
another loop for the write-master completions.

diff --git a/drivers/media/platform/qcom/camss/camss-vfe-480.c 
b/drivers/media/platform/qcom/camss/camss-vfe-480.c
index 4feea590a47bc..d73f733fde045 100644
--- a/drivers/media/platform/qcom/camss/camss-vfe-480.c
+++ b/drivers/media/platform/qcom/camss/camss-vfe-480.c
@@ -202,11 +202,13 @@ static irqreturn_t vfe_isr(int irq, void *dev)
                 writel_relaxed(status, vfe->base + VFE_BUS_IRQ_CLEAR(0));
                 writel_relaxed(1, vfe->base + VFE_BUS_IRQ_CLEAR_GLOBAL);

-               /* Loop through all WMs IRQs */
-               for (i = 0; i < MSM_VFE_IMAGE_MASTERS_NUM; i++) {
+               for (i = 0; i < MAX_VFE_OUTPUT_LINES; i++) {
                         if (status & BUS_IRQ_MASK_0_RDI_RUP(vfe, i))
                                 vfe_isr_reg_update(vfe, i);
+               }

+               /* Loop through all WMs IRQs */
+               for (i = 0; i < MSM_VFE_IMAGE_MASTERS_NUM; i++) {
                         if (status & BUS_IRQ_MASK_0_COMP_DONE(vfe, 
RDI_COMP_GROUP(i)))
                                 vfe_buf_done(vfe, i);
                 }

---
bod