linux-kernel - [PATCH 4.4 46/88] spi: bcm2835: Fix race on DMA termination

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:   Fri, 11 Jan 2019 15:08:15 +0100
From:   Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:     linux-kernel@...r.kernel.org
Cc:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        stable@...r.kernel.org, Lukas Wunner <lukas@...ner.de>,
        Mathias Duckeck <m.duckeck@...bus.de>,
        Frank Pavlic <f.pavlic@...bus.de>,
        Martin Sperl <kernel@...tin.sperl.org>,
        Noralf Trønnes <noralf@...nnes.org>,
        Mark Brown <broonie@...nel.org>
Subject: [PATCH 4.4 46/88] spi: bcm2835: Fix race on DMA termination

4.4-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Lukas Wunner <lukas@...ner.de>

commit e82b0b3828451c1cd331d9f304c6078fcd43b62e upstream.

If a DMA transfer finishes orderly right when spi_transfer_one_message()
determines that it has timed out, the callbacks bcm2835_spi_dma_done()
and bcm2835_spi_handle_err() race to call dmaengine_terminate_all(),
potentially leading to double termination.

Prevent by atomically changing the dma_pending flag before calling
dmaengine_terminate_all().

Signed-off-by: Lukas Wunner <lukas@...ner.de>
Fixes: 3ecd37edaa2a ("spi: bcm2835: enable dma modes for transfers meeting certain conditions")
Cc: stable@...r.kernel.org # v4.2+
Cc: Mathias Duckeck <m.duckeck@...bus.de>
Cc: Frank Pavlic <f.pavlic@...bus.de>
Cc: Martin Sperl <kernel@...tin.sperl.org>
Cc: Noralf Trønnes <noralf@...nnes.org>
Signed-off-by: Mark Brown <broonie@...nel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>

---
 drivers/spi/spi-bcm2835.c |   10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

--- a/drivers/spi/spi-bcm2835.c
+++ b/drivers/spi/spi-bcm2835.c
@@ -233,10 +233,9 @@ static void bcm2835_spi_dma_done(void *d
 	 * is called the tx-dma must have finished - can't get to this
 	 * situation otherwise...
 	 */
-	dmaengine_terminate_all(master->dma_tx);
-
-	/* mark as no longer pending */
-	bs->dma_pending = 0;
+	if (cmpxchg(&bs->dma_pending, true, false)) {
+		dmaengine_terminate_all(master->dma_tx);
+	}
 
 	/* and mark as completed */;
 	complete(&master->xfer_completion);
@@ -617,10 +616,9 @@ static void bcm2835_spi_handle_err(struc
 	struct bcm2835_spi *bs = spi_master_get_devdata(master);
 
 	/* if an error occurred and we have an active dma, then terminate */
-	if (bs->dma_pending) {
+	if (cmpxchg(&bs->dma_pending, true, false)) {
 		dmaengine_terminate_all(master->dma_tx);
 		dmaengine_terminate_all(master->dma_rx);
-		bs->dma_pending = 0;
 	}
 	/* and reset */
 	bcm2835_spi_reset_hw(master);