linux-kernel - [PATCH] hda - fail ELD reading early

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-next>] [day] [month] [year] [list]

Message-ID: <20111122084623.GA30050@localhost>
Date:	Tue, 22 Nov 2011 16:46:23 +0800
From:	Wu Fengguang <fengguang.wu@...el.com>
To:	Takashi Iwai <tiwai@...e.de>
Cc:	"alsa-devel@...a-project.org" <alsa-devel@...a-project.org>,
	LKML <linux-kernel@...r.kernel.org>
Subject: [PATCH] hda - fail ELD reading early

With the ELD repoll mechanism, we can (and should) fail the ELD reading
immediately when find something obviously wrong and let the caller retry
after some delay.

Signed-off-by: Wu Fengguang <fengguang.wu@...el.com>
---
 sound/pci/hda/hda_eld.c |   28 +++++++++++++++++++---------
 1 file changed, 19 insertions(+), 9 deletions(-)

--- linux.orig/sound/pci/hda/hda_eld.c	2011-11-22 16:02:58.000000000 +0800
+++ linux/sound/pci/hda/hda_eld.c	2011-11-22 16:36:10.000000000 +0800
@@ -347,18 +347,28 @@ int snd_hdmi_get_eld(struct hdmi_eld *el
 
 	for (i = 0; i < size; i++) {
 		unsigned int val = hdmi_get_eld_data(codec, nid, i);
+		/*
+		 * Graphics driver might be writing to ELD buffer right now.
+		 * Just abort. The caller will repoll after a while.
+		 */
 		if (!(val & AC_ELDD_ELD_VALID)) {
-			if (!i) {
-				snd_printd(KERN_INFO
-					   "HDMI: invalid ELD data\n");
-				ret = -EINVAL;
-				goto error;
-			}
 			snd_printd(KERN_INFO
 				  "HDMI: invalid ELD data byte %d\n", i);
-			val = 0;
-		} else
-			val &= AC_ELDD_ELD_DATA;
+			ret = -EINVAL;
+			goto error;
+		}
+		val &= AC_ELDD_ELD_DATA;
+		/*
+		 * The first byte cannot be zero. This can happen on some DVI
+		 * connections. Some Intel chips may also need some 250ms delay
+		 * to return non-zero ELD data, even when the graphics driver
+		 * correctly writes ELD content before setting ELD_valid bit.
+		 */
+		if (!val && !i) {
+			snd_printdd(KERN_INFO "HDMI: 0 ELD data\n");
+			ret = -EINVAL;
+			goto error;
+		}
 		buf[i] = val;
 	}
 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/