Devicetree
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Wandun Chen" <chenwandun1@gmail.com>
Cc: robh@kernel.org, conor+dt@kernel.org, devicetree@vger.kernel.org
Subject: Re: [PATCH v3 4/5] of: reserved_mem: reject static regions overlapping no-map memory
Date: Sun, 20 Sep 2026 10:03:47 +0000	[thread overview]
Message-ID: <20260920100347.9D08A1F000FF@smtp.kernel.org> (raw)
In-Reply-To: <20260920092852.614973-5-chenwandun1@gmail.com>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] of: reserved_mem: integer overflow in rmem_overlaps_check bypasses overlap detection
--

commit bab4bcea3e71f149872f45d815c003c14a811a0a
Author: Wandun Chen <chenwandun@lixiang.com>

of: reserved_mem: reject static regions overlapping no-map memory

This commit updates the reserved memory initialization to reject static no-map
regions that overlap with existing reserved no-map memory. It also skips
no-map regions during the late scan if they overlap with already reserved
ones, preventing multiple drivers from mapping the same physical memory.

Link: https://sashiko.dev/#/message/20260814084718.29C341F000E9%40smtp.kernel.org [1]

> diff --git a/drivers/of/of_reserved_mem.c b/drivers/of/of_reserved_mem.c
> index 3a6c6dbfd7b11..8d2057f2ac12e 100644
> --- a/drivers/of/of_reserved_mem.c
> +++ b/drivers/of/of_reserved_mem.c

[ ... ]

> @@ -266,6 +267,20 @@ static void __init __rmem_check_for_overlap(void)
>  	}
>  }
>  
> +static bool __init rmem_overlaps_check(phys_addr_t base, phys_addr_t size,
> +				       int start)
> +{
> +	int i;
> +
> +	for (i = start; i < reserved_mem_count; i++) {
> +		struct reserved_mem *r = &reserved_mem[i];
> +
> +		if (memblock_addrs_overlap(base, size, r->base, r->size))
> +			return true;

[Severity: High]
Does this code allow an integer wrap-around to bypass the overlap check?

The size variable read from the device tree is used without being bounded by
memblock_cap_size() before calling memblock_addrs_overlap().

If a crafted device tree or overlay defines a static reserved memory node
with a size large enough to cause base + size to wrap around phys_addr_t,
the internal addition in memblock_addrs_overlap() could wrap. This would
cause it to incorrectly return false, bypassing this overlap check.

The region then reaches memblock_is_region_rsrv_rmem(), which does cap the
size and finds the valid overlap marked by an earlier valid region, allowing
multiple reserved memory regions to be mapped to the exact same physical
memory.

> +	}
> +	return false;
> +}
> +

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260920092852.614973-1-chenwandun1@gmail.com?part=4

  reply	other threads:[~2026-09-20 10:03 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-20  9:28 [PATCH v3 0/5] of: reserved_mem: several fixes about reserved memory Wandun Chen
2026-09-20  9:28 ` [PATCH v3 1/5] of: reserved_mem: release dynamically allocated no-map region on init failure Wandun Chen
2026-09-20  9:28 ` [PATCH v3 2/5] of: reserved_mem: retain static no-map memory " Wandun Chen
2026-09-20  9:28 ` [PATCH v3 3/5] of: reserved_mem: skip init for regions whose early reservation failed Wandun Chen
2026-09-20  9:28 ` [PATCH v3 4/5] of: reserved_mem: reject static regions overlapping no-map memory Wandun Chen
2026-09-20 10:03   ` sashiko-bot [this message]
2026-09-20  9:28 ` [PATCH v3 5/5] of: reserved_mem: reject static mapped regions overlapping existing reservations Wandun Chen
2026-09-20 10:13   ` sashiko-bot
2026-09-22  8:48 ` [PATCH v3 0/5] of: reserved_mem: several fixes about reserved memory Mike Rapoport
2026-09-22  9:24   ` Wandun
2026-10-03  8:04     ` Mike Rapoport
2026-10-06 13:26       ` Marek Szyprowski
2026-10-08  6:38         ` Mike Rapoport
2026-10-08  9:43           ` Marek Szyprowski

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260920100347.9D08A1F000FF@smtp.kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=chenwandun1@gmail.com \
    --cc=conor+dt@kernel.org \
    --cc=devicetree@vger.kernel.org \
    --cc=robh@kernel.org \
    --cc=sashiko-reviews@lists.linux.dev \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox