diff mbox series

[5/6] spi: rockchip: terminate dma transmission when slave abort

Message ID 20220211034344.4130-6-jon.lin@rock-chips.com (mailing list archive)
State New, archived
Headers show
Series None | expand

Commit Message

Jon Lin Feb. 11, 2022, 3:43 a.m. UTC
After slave abort, all DMA should be stopped, or it will affect the
next transmission:

[   31.693877] Unable to handle kernel paging request at virtual address ffffff8105a2a7c0
[   31.694643] Mem abort info:
[   31.694898]   ESR = 0x96000045
[   31.695179]   EC = 0x25: DABT (current EL), IL = 32 bits
[   31.695653]   SET = 0, FnV = 0
[   31.695931]   EA = 0, S1PTW = 0
[   31.696218] Data abort info:
[   31.696485]   ISV = 0, ISS = 0x00000045
[   31.696832]   CM = 0, WnR = 1
[   31.697112] swapper pgtable: 4k pages, 39-bit VAs, pgdp=000000000142f000
[   31.697713] [ffffff8105a2a7c0] pgd=0000000000000000, p4d=0000000000000000, pud=0000000000000000
[   31.698502] Internal error: Oops: 96000045 [#1] SMP
[   31.698943] Modules linked in:
[   31.699235] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G        W         5.10.43 #8
[   31.699895] Hardware name: Rockchip RK3588 EVB1 LP4 V10 Board (DT)
[   31.700455] pstate: 60400089 (nZCv daIf +PAN -UAO -TCO BTYPE=--)
[   31.701000] pc : rockchip_spi_slave_abort+0x150/0x18c
[   31.701456] lr : rockchip_spi_slave_abort+0x78/0x18c

Signed-off-by: Jon Lin <jon.lin@rock-chips.com>
---

 drivers/spi/spi-rockchip.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

Comments

Mark Brown Feb. 11, 2022, 11:49 a.m. UTC | #1
On Fri, Feb 11, 2022 at 11:43:42AM +0800, Jon Lin wrote:

> After slave abort, all DMA should be stopped, or it will affect the
> next transmission:

Again, this is a fix and should be at the start of the series.

> 
> [   31.693877] Unable to handle kernel paging request at virtual address ffffff8105a2a7c0
> [   31.694643] Mem abort info:
> [   31.694898]   ESR = 0x96000045
> [   31.695179]   EC = 0x25: DABT (current EL), IL = 32 bits
> [   31.695653]   SET = 0, FnV = 0
> [   31.695931]   EA = 0, S1PTW = 0

Please think hard before including complete backtraces in upstream
reports, they are very large and contain almost no useful information
relative to their size so often obscure the relevant content in your
message. If part of the backtrace is usefully illustrative (it often is
for search engines if nothing else) then it's usually better to pull out
the relevant sections.
diff mbox series

Patch

diff --git a/drivers/spi/spi-rockchip.c b/drivers/spi/spi-rockchip.c
index 5457af616a44..3d4e95eda4a6 100644
--- a/drivers/spi/spi-rockchip.c
+++ b/drivers/spi/spi-rockchip.c
@@ -617,8 +617,6 @@  static int rockchip_spi_slave_abort(struct spi_controller *ctlr)
 	if (atomic_read(&rs->state) & RXDMA) {
 		dmaengine_pause(ctlr->dma_rx);
 		status = dmaengine_tx_status(ctlr->dma_rx, ctlr->dma_rx->cookie, &state);
-		dmaengine_terminate_sync(ctlr->dma_rx);
-		atomic_set(&rs->state, 0);
 		if (status == DMA_ERROR) {
 			rs->rx = rs->xfer->rx_buf;
 			rs->xfer->len = 0;
@@ -648,6 +646,11 @@  static int rockchip_spi_slave_abort(struct spi_controller *ctlr)
 	}
 
 out:
+	if (atomic_read(&rs->state) & RXDMA)
+		dmaengine_terminate_sync(ctlr->dma_rx);
+	if (atomic_read(&rs->state) & TXDMA)
+		dmaengine_terminate_sync(ctlr->dma_tx);
+	atomic_set(&rs->state, 0);
 	spi_enable_chip(rs, false);
 	rs->slave_abort = true;
 	spi_finalize_current_transfer(ctlr);