Skip to contentSkip to conversation

dmaengine: mediatek-cqdma: sleeping in atomic context

PATCH Applied Dan Carpenter 2 participants 2 messages May 9, 2019 – May 21, 2019

Subsystems (4)
Kernel janitors Arm MediaTek dmaengine MediaTek DMA driver
lore
Share & tools
lore

Dan Carpenter

author
#0 lore raw

PATCH 1/1 dmaengine: mediatek-cqdma: sleeping in atomic context

The mtk_cqdma_poll_engine_done() function takes a true/false parameter where true means it's called from atomic context. There are a couple places where it was set to false but it's actually in atomic context so it should be true.

All the callers for mtk_cqdma_hard_reset() are holding a spin_lock and in mtk_cqdma_free_chan_resources() we take a spin_lock before calling the mtk_cqdma_poll_engine_done() function.

Fixes b1f01e48df5a ("dmaengine: mediatek: Add MediaTek Command-Queue DMA controller for MT6765 SoC")
Signed-off-by Dan Carpenter <[email protected]>

---
The "atomic" parameter is always true so the temptation was to just remove it entirely.

 drivers/dma/mediatek/mtk-cqdma.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/dma/mediatek/mtk-cqdma.c +2 -2
index 814853842e29..723b11c190b3 100644
@@ -225,7 +225,7 @@ static int mtk_cqdma_hard_reset(struct mtk_cqdma_pchan *pc)
 	mtk_dma_set(pc, MTK_CQDMA_RESET, MTK_CQDMA_HARD_RST_BIT);
 	mtk_dma_clr(pc, MTK_CQDMA_RESET, MTK_CQDMA_HARD_RST_BIT);
 
-	return mtk_cqdma_poll_engine_done(pc, false);
+	return mtk_cqdma_poll_engine_done(pc, true);
 }
 
 static void mtk_cqdma_start(struct mtk_cqdma_pchan *pc,

@@ -671,7 +671,7 @@ static void mtk_cqdma_free_chan_resources(struct dma_chan *c)
 		mtk_dma_set(cvc->pc, MTK_CQDMA_FLUSH, MTK_CQDMA_FLUSH_BIT);
 
 		/* wait for the completion of flush operation */
-		if (mtk_cqdma_poll_engine_done(cvc->pc, false) < 0)
+		if (mtk_cqdma_poll_engine_done(cvc->pc, true) < 0)
 			dev_err(cqdma2dev(to_cqdma_dev(c)), "cqdma flush timeout\n");
 
 		/* clear the flush bit and interrupt flag */
Signature · 1 line
2.18.0
maintainer Applied
#1 lore raw
Quoted 8 lines · Dan Carpenter

The mtk_cqdma_poll_engine_done() function takes a true/false parameter where true means it's called from atomic context. There are a couple places where it was set to false but it's actually in atomic context so it should be true.

All the callers for mtk_cqdma_hard_reset() are holding a spin_lock and in mtk_cqdma_free_chan_resources() we take a spin_lock before calling the mtk_cqdma_poll_engine_done() function.

Applied, thanks

Quoted 6 lines
Fixes b1f01e48df5a ("dmaengine: mediatek: Add MediaTek Command-Queue DMA controller for MT6765 SoC")
Signed-off-by Dan Carpenter <[email protected]>

---
The "atomic" parameter is always true so the temptation was to just remove it entirely.

a patch is welcome :)

Signature · 1 line
~Vinod