[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20230203121702.15725-1-aman.kumar@intel.com>
Date: Fri, 3 Feb 2023 17:47:02 +0530
From: aman.kumar@...el.com
To: vkoul@...nel.org, dmaengine@...r.kernel.org
Cc: andriy.shevchenko@...ux.intel.com, srikanth.thokala@...el.com,
aman.kumar@...el.com, mallikarjunappa.sangannavar@...el.com,
linux-kernel@...r.kernel.org
Subject: [PATCH v1 1/1] dmaengine: idma64: Update bytes_transferred field
From: Aman Kumar <aman.kumar@...el.com>
Currently when 8250 data transfer is done, bytes_tranferred always returns
0 at /sys/devices/pci0000\:\:**.*/dma/dma*chan*/bytes_transferred.
In many cases it gives false impression that data is not being
trasferred via DMA.
So, updating the bytes_transferred field to count the bytes
whenever there is data transfer using idma64.
Co-developed-by: Srikanth Thokala <srikanth.thokala@...el.com>
Signed-off-by: Srikanth Thokala <srikanth.thokala@...el.com>
Signed-off-by: Aman Kumar <aman.kumar@...el.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
---
drivers/dma/idma64.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/drivers/dma/idma64.c b/drivers/dma/idma64.c
index c33087c5cd02..8880d2b91bf5 100644
--- a/drivers/dma/idma64.c
+++ b/drivers/dma/idma64.c
@@ -137,8 +137,11 @@ static void idma64_chan_irq(struct idma64 *idma64, unsigned short c,
u32 status_err, u32 status_xfer)
{
struct idma64_chan *idma64c = &idma64->chan[c];
+ struct dma_chan_percpu *stat;
struct idma64_desc *desc;
+ stat = this_cpu_ptr(idma64c->vchan.chan.local);
+
spin_lock(&idma64c->vchan.lock);
desc = idma64c->desc;
if (desc) {
@@ -149,6 +152,7 @@ static void idma64_chan_irq(struct idma64 *idma64, unsigned short c,
dma_writel(idma64, CLEAR(XFER), idma64c->mask);
desc->status = DMA_COMPLETE;
vchan_cookie_complete(&desc->vdesc);
+ stat->bytes_transferred += desc->length;
idma64_start_transfer(idma64c);
}
--
2.17.1
Powered by blists - more mailing lists