Re: Atmel UART with dma enabled does not work on branch 5.4.Y from version 5.4.174.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Yes sorry,

enclosed signed patch.


Le 18/02/2022 à 09:46, Greg KH a écrit :
On Fri, Feb 18, 2022 at 09:40:55AM +0100, Mickael GARDET wrote:
Hi,

we observed a regression on our atmel platforms on branch 5.4.Y from version
5.4.174.
uarts are no longer functional if DMA is enabled.

the following patch
Commit e6af9b05bec63cd4d1de2a33968cd0be2a91282a dmaengine: at_xdmac: Start
transfer for cyclic channels in issue_pending
Link:
https://lore.kernel.org/r/20211215110115.191749-3-tudor.ambarus@xxxxxxxxxxxxx
has not been backported but is needed if the patch
Commit 4f4b9b5895614eb2e2b5f4cab7858f44bd113e1b tty: serial: atmel: Call
dma_async_issue_pending()
Link:
https://lore.kernel.org/r/20211125090028.786832-4-tudor.ambarus@xxxxxxxxxxxxx
is applied.

enclosed commit fix this issue and work for me.

Best regards,
Mickael GARDET

 From 9877f93c066be8c2f4e33a6edd4dfa3d6d6974d9 Mon Sep 17 00:00:00 2001
From: Tudor Ambarus <tudor.ambarus@xxxxxxxxxxxxx>
Date: Wed, 15 Dec 2021 13:01:05 +0200
Subject: [PATCH] dmaengine: at_xdmac: Start transfer for cyclic channels in
issue_pending

commit e6af9b05bec63cd4d1de2a33968cd0be2a91282a upstream.

Cyclic channels must too call issue_pending in order to start a transfer.
Start the transfer in issue_pending regardless of the type of channel.
This wrongly worked before, because in the past the transfer was started
at tx_submit level when only a desc in the transfer list.

Fixes: e1f7c9eee707 ("dmaengine: at_xdmac: creation of the atmel eXtended
DMA Controller driver")
Change-Id: If1bf3e13329cebb9904ae40620f6cf2b7f06fe9f
Signed-off-by: Tudor Ambarus <tudor.ambarus@xxxxxxxxxxxxx>
Link:
https://lore.kernel.org/r/20211215110115.191749-3-tudor.ambarus@xxxxxxxxxxxxx
Signed-off-by: Vinod Koul <vkoul@xxxxxxxxxx>
---
drivers/dma/at_xdmac.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c
index f63d141481a3..9aae6b3da356 100644
--- a/drivers/dma/at_xdmac.c
+++ b/drivers/dma/at_xdmac.c
@@ -1726,11 +1726,13 @@ static irqreturn_t at_xdmac_interrupt(int irq,
void *dev_id)
static void at_xdmac_issue_pending(struct dma_chan *chan)
{
struct at_xdmac_chan *atchan = to_at_xdmac_chan(chan);
+    unsigned long flags;

dev_dbg(chan2dev(&atchan->chan), "%s\n", __func__);

-    if (!at_xdmac_chan_is_cyclic(atchan))
-        at_xdmac_advance_work(atchan);
+    spin_lock_irqsave(&atchan->lock, flags);
+    at_xdmac_advance_work(atchan);
+    spin_unlock_irqrestore(&atchan->lock, flags);

return;
}

base-commit: 7b3eb66d0daf61e91cccdb2fe5d271ae5adc5a76

--
2.35.1


The patch is whitespace damaged and can not be applied at all :(

Can you try fixing your email client up, or worst case, attach the
patch, and we can take it from there?

Also be sure to sign-off on the patch, as you have modified it to work
properly on this branch.

thanks,

greg k-h
From 1310f5ba2f2d199e6527d2a289192ff99d8c6c83 Mon Sep 17 00:00:00 2001
From: Tudor Ambarus <tudor.ambarus@xxxxxxxxxxxxx>
Date: Wed, 15 Dec 2021 13:01:05 +0200
Subject: [PATCH] dmaengine: at_xdmac: Start transfer for cyclic channels in
 issue_pending

commit e6af9b05bec63cd4d1de2a33968cd0be2a91282a upstream.

Cyclic channels must too call issue_pending in order to start a transfer.
Start the transfer in issue_pending regardless of the type of channel.
This wrongly worked before, because in the past the transfer was started
at tx_submit level when only a desc in the transfer list.

Fixes: e1f7c9eee707 ("dmaengine: at_xdmac: creation of the atmel eXtended DMA Controller driver")
Change-Id: If1bf3e13329cebb9904ae40620f6cf2b7f06fe9f
Signed-off-by: Tudor Ambarus <tudor.ambarus@xxxxxxxxxxxxx>
Link: https://lore.kernel.org/r/20211215110115.191749-3-tudor.ambarus@xxxxxxxxxxxxx
Signed-off-by: Vinod Koul <vkoul@xxxxxxxxxx>
Signed-off-by: Mickael GARDET <m.gardet@xxxxxxxxxxx>
---
 drivers/dma/at_xdmac.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c
index f63d141481a3..9aae6b3da356 100644
--- a/drivers/dma/at_xdmac.c
+++ b/drivers/dma/at_xdmac.c
@@ -1726,11 +1726,13 @@ static irqreturn_t at_xdmac_interrupt(int irq, void *dev_id)
 static void at_xdmac_issue_pending(struct dma_chan *chan)
 {
 	struct at_xdmac_chan *atchan = to_at_xdmac_chan(chan);
+	unsigned long flags;
 
 	dev_dbg(chan2dev(&atchan->chan), "%s\n", __func__);
 
-	if (!at_xdmac_chan_is_cyclic(atchan))
-		at_xdmac_advance_work(atchan);
+	spin_lock_irqsave(&atchan->lock, flags);
+	at_xdmac_advance_work(atchan);
+	spin_unlock_irqrestore(&atchan->lock, flags);
 
 	return;
 }
-- 
2.35.1


[Index of Archives]     [Linux Kernel]     [Kernel Development Newbies]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite Hiking]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux