linux-kernel - [Patch v2 16/20] media: ti-vpe: cal: Properly calculate max resolution boundary

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]

Date:   Mon, 4 Nov 2019 13:31:36 -0600
From:   Benoit Parrot <bparrot@...com>
To:     Hans Verkuil <hverkuil@...all.nl>
CC:     <linux-media@...r.kernel.org>, Rob Herring <robh+dt@...nel.org>,
        <devicetree@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
        Benoit Parrot <bparrot@...com>
Subject: [Patch v2 16/20] media: ti-vpe: cal: Properly calculate max resolution boundary

Currently we were using an arbitrarily small maximum resolution mostly
based on available sensor capabilities. However the hardware DMA limits
are much higher than the statically define maximum resolution we were
using.

There we rework the boundary check code to handle the maximum width and
height based on the maximum line width in bytes and re-calculating the
pixel width based on the given pixel format.

Signed-off-by: Benoit Parrot <bparrot@...com>
---
 drivers/media/platform/ti-vpe/cal.c | 16 +++++++++++-----
 1 file changed, 11 insertions(+), 5 deletions(-)

diff --git a/drivers/media/platform/ti-vpe/cal.c b/drivers/media/platform/ti-vpe/cal.c
index f0210a743bdc..20160845014e 100644
--- a/drivers/media/platform/ti-vpe/cal.c
+++ b/drivers/media/platform/ti-vpe/cal.c
@@ -34,8 +34,8 @@
 
 #define CAL_MODULE_NAME "cal"
 
-#define MAX_WIDTH 1920
-#define MAX_HEIGHT 1200
+#define MAX_WIDTH_BYTES (8192 * 8)
+#define MAX_HEIGHT_LINES 16383
 
 #define CAL_VERSION "0.1.0"
 
@@ -1337,15 +1337,21 @@ static int cal_calc_format_size(struct cal_ctx *ctx,
 				const struct cal_fmt *fmt,
 				struct v4l2_format *f)
 {
-	u32 bpl;
+	u32 bpl, max_width;
 
 	if (!fmt) {
 		ctx_dbg(3, ctx, "No cal_fmt provided!\n");
 		return -EINVAL;
 	}
 
-	v4l_bound_align_image(&f->fmt.pix.width, 48, MAX_WIDTH, 2,
-			      &f->fmt.pix.height, 32, MAX_HEIGHT, 0, 0);
+	/*
+	 * Maximum width is bound by the DMA max width in bytes.
+	 * We need to recalculate the actual maxi width depending on the
+	 * number of bytes per pixels required.
+	 */
+	max_width = MAX_WIDTH_BYTES / (ALIGN(fmt->bpp, 8) >> 3);
+	v4l_bound_align_image(&f->fmt.pix.width, 48, max_width, 2,
+			      &f->fmt.pix.height, 32, MAX_HEIGHT_LINES, 0, 0);
 
 	bpl = (f->fmt.pix.width * ALIGN(fmt->bpp, 8)) >> 3;
 	f->fmt.pix.bytesperline = ALIGN(bpl, 16);
-- 
2.17.1