linux-kernel - [PATCH V1] accel/amdxdna: Fix race where send ring appears full due to delayed head update

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: <20251211045125.1724604-1-lizhi.hou@amd.com>
Date: Wed, 10 Dec 2025 20:51:25 -0800
From: Lizhi Hou <lizhi.hou@....com>
To: <ogabbay@...nel.org>, <quic_jhugo@...cinc.com>,
	<dri-devel@...ts.freedesktop.org>, <maciej.falkowski@...ux.intel.com>
CC: Lizhi Hou <lizhi.hou@....com>, <linux-kernel@...r.kernel.org>,
	<max.zhen@....com>, <sonal.santan@....com>, <mario.limonciello@....com>
Subject: [PATCH V1] accel/amdxdna: Fix race where send ring appears full due to delayed head update

The firmware sends a response and interrupts the driver before advancing
the mailbox send ring head pointer. As a result, the driver may observe
the response and attempt to send a new request before the firmware has
updated the head pointer. In this window, the send ring still appears
full, causing the driver to incorrectly fail the send operation.

This race can be triggered more easily in a multithreaded environment,
leading to unexpected and spurious "send ring full" failures.

To address this, poll the send ring head pointer for up to 100us before
returning a full-ring condition. This allows the firmware time to update
the head pointer.

Fixes: b87f920b9344 ("accel/amdxdna: Support hardware mailbox")
Signed-off-by: Lizhi Hou <lizhi.hou@....com>
---
 drivers/accel/amdxdna/amdxdna_mailbox.c | 27 +++++++++++++++----------
 1 file changed, 16 insertions(+), 11 deletions(-)

diff --git a/drivers/accel/amdxdna/amdxdna_mailbox.c b/drivers/accel/amdxdna/amdxdna_mailbox.c
index a60a85ce564c..469242ed8224 100644
--- a/drivers/accel/amdxdna/amdxdna_mailbox.c
+++ b/drivers/accel/amdxdna/amdxdna_mailbox.c
@@ -191,26 +191,34 @@ mailbox_send_msg(struct mailbox_channel *mb_chann, struct mailbox_msg *mb_msg)
 	u32 head, tail;
 	u32 start_addr;
 	u32 tmp_tail;
+	int ret;
 
 	head = mailbox_get_headptr(mb_chann, CHAN_RES_X2I);
 	tail = mb_chann->x2i_tail;
-	ringbuf_size = mailbox_get_ringbuf_size(mb_chann, CHAN_RES_X2I);
+	ringbuf_size = mailbox_get_ringbuf_size(mb_chann, CHAN_RES_X2I) - sizeof(u32);
 	start_addr = mb_chann->res[CHAN_RES_X2I].rb_start_addr;
 	tmp_tail = tail + mb_msg->pkg_size;
 
-	if (tail < head && tmp_tail >= head)
-		goto no_space;
-
-	if (tail >= head && (tmp_tail > ringbuf_size - sizeof(u32) &&
-			     mb_msg->pkg_size >= head))
-		goto no_space;
 
-	if (tail >= head && tmp_tail > ringbuf_size - sizeof(u32)) {
+check_again:
+	if (tail >= head && tmp_tail > ringbuf_size) {
 		write_addr = mb_chann->mb->res.ringbuf_base + start_addr + tail;
 		writel(TOMBSTONE, write_addr);
 
 		/* tombstone is set. Write from the start of the ringbuf */
 		tail = 0;
+		tmp_tail = tail + mb_msg->pkg_size;
+	}
+
+	if (tail < head && tmp_tail >= head) {
+		ret = read_poll_timeout(mailbox_get_headptr, head,
+					tmp_tail < head || tail >= head,
+					1, 100, false, mb_chann, CHAN_RES_X2I);
+		if (ret)
+			return ret;
+
+		if (tail >= head)
+			goto check_again;
 	}
 
 	write_addr = mb_chann->mb->res.ringbuf_base + start_addr + tail;
@@ -222,9 +230,6 @@ mailbox_send_msg(struct mailbox_channel *mb_chann, struct mailbox_msg *mb_msg)
 			    mb_msg->pkg.header.id);
 
 	return 0;
-
-no_space:
-	return -ENOSPC;
 }
 
 static int
-- 
2.34.1