* [PATCH v3 04/10] dmaengine: edma: Add support for DMA_PAUSE/RESUME operation
From: Peter Ujfalusi @ 2014-04-14 11:41 UTC (permalink / raw)
To: dan.j.williams, vinod.koul, nsekhar, joelf
Cc: dmaengine, davinci-linux-open-source, linux-omap, linux-kernel,
linux-arm-kernel
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
Pause/Resume can be used by the audio stack when the stream is paused/resumed
The edma platform code has support for this and the legacy audio stack used
this.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
drivers/dma/edma.c | 28 ++++++++++++++++++++++++++++
1 file changed, 28 insertions(+)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 2742867fd1e6..7891378a03f0 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -240,6 +240,26 @@ static int edma_slave_config(struct edma_chan *echan,
return 0;
}
+static int edma_dma_pause(struct edma_chan *echan)
+{
+ /* Pause/Resume only allowed with cyclic mode */
+ if (!echan->edesc->cyclic)
+ return -EINVAL;
+
+ edma_pause(echan->ch_num);
+ return 0;
+}
+
+static int edma_dma_resume(struct edma_chan *echan)
+{
+ /* Pause/Resume only allowed with cyclic mode */
+ if (!echan->edesc->cyclic)
+ return -EINVAL;
+
+ edma_resume(echan->ch_num);
+ return 0;
+}
+
static int edma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
unsigned long arg)
{
@@ -255,6 +275,14 @@ static int edma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
config = (struct dma_slave_config *)arg;
ret = edma_slave_config(echan, config);
break;
+ case DMA_PAUSE:
+ ret = edma_dma_pause(echan);
+ break;
+
+ case DMA_RESUME:
+ ret = edma_dma_resume(echan);
+ break;
+
default:
ret = -ENOSYS;
}
--
1.9.2
^ permalink raw reply related
* [PATCH v3 04/10] dmaengine: edma: Add support for DMA_PAUSE/RESUME operation
From: Peter Ujfalusi @ 2014-04-14 11:41 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
Pause/Resume can be used by the audio stack when the stream is paused/resumed
The edma platform code has support for this and the legacy audio stack used
this.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
drivers/dma/edma.c | 28 ++++++++++++++++++++++++++++
1 file changed, 28 insertions(+)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 2742867fd1e6..7891378a03f0 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -240,6 +240,26 @@ static int edma_slave_config(struct edma_chan *echan,
return 0;
}
+static int edma_dma_pause(struct edma_chan *echan)
+{
+ /* Pause/Resume only allowed with cyclic mode */
+ if (!echan->edesc->cyclic)
+ return -EINVAL;
+
+ edma_pause(echan->ch_num);
+ return 0;
+}
+
+static int edma_dma_resume(struct edma_chan *echan)
+{
+ /* Pause/Resume only allowed with cyclic mode */
+ if (!echan->edesc->cyclic)
+ return -EINVAL;
+
+ edma_resume(echan->ch_num);
+ return 0;
+}
+
static int edma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
unsigned long arg)
{
@@ -255,6 +275,14 @@ static int edma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
config = (struct dma_slave_config *)arg;
ret = edma_slave_config(echan, config);
break;
+ case DMA_PAUSE:
+ ret = edma_dma_pause(echan);
+ break;
+
+ case DMA_RESUME:
+ ret = edma_dma_resume(echan);
+ break;
+
default:
ret = -ENOSYS;
}
--
1.9.2
^ permalink raw reply related
* [PATCH v3 05/10] dmaengine: edma: Set DMA_CYCLIC capability flag
From: Peter Ujfalusi @ 2014-04-14 11:42 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
Indicate that the edma dmaengine driver has support for cyclic mode.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
arch/arm/common/edma.c | 1 +
drivers/dma/edma.c | 1 +
2 files changed, 2 insertions(+)
diff --git a/arch/arm/common/edma.c b/arch/arm/common/edma.c
index 999266bf69b9..0b37f7734d0f 100644
--- a/arch/arm/common/edma.c
+++ b/arch/arm/common/edma.c
@@ -1574,6 +1574,7 @@ static struct edma_soc_info *edma_setup_info_from_dt(struct device *dev,
return ERR_PTR(ret);
dma_cap_set(DMA_SLAVE, edma_filter_info.dma_cap);
+ dma_cap_set(DMA_CYCLIC, edma_filter_info.dma_cap);
of_dma_controller_register(dev->of_node, of_dma_simple_xlate,
&edma_filter_info);
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 7891378a03f0..1dd9e8806975 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -891,6 +891,7 @@ static int edma_probe(struct platform_device *pdev)
dma_cap_zero(ecc->dma_slave.cap_mask);
dma_cap_set(DMA_SLAVE, ecc->dma_slave.cap_mask);
+ dma_cap_set(DMA_CYCLIC, ecc->dma_slave.cap_mask);
edma_dma_init(ecc, &ecc->dma_slave, &pdev->dev);
--
1.9.2
^ permalink raw reply related
* [PATCH v3 06/10] dmaengine: edma: Implement device_slave_caps callback
From: Peter Ujfalusi @ 2014-04-14 11:42 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
With the callback implemented omap-dma can provide information to client
drivers regarding to supported address widths, directions, residue
granularity, etc.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
drivers/dma/edma.c | 18 ++++++++++++++++++
1 file changed, 18 insertions(+)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 1dd9e8806975..2f58c04cbcb1 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -852,6 +852,23 @@ static void __init edma_chan_init(struct edma_cc *ecc,
}
}
+#define EDMA_DMA_BUSWIDTHS (BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) | \
+ BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) | \
+ BIT(DMA_SLAVE_BUSWIDTH_4_BYTES))
+
+static int edma_dma_device_slave_caps(struct dma_chan *dchan,
+ struct dma_slave_caps *caps)
+{
+ caps->src_addr_widths = EDMA_DMA_BUSWIDTHS;
+ caps->dstn_addr_widths = EDMA_DMA_BUSWIDTHS;
+ caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
+ caps->cmd_pause = true;
+ caps->cmd_terminate = true;
+ caps->residue_granularity = DMA_RESIDUE_GRANULARITY_DESCRIPTOR;
+
+ return 0;
+}
+
static void edma_dma_init(struct edma_cc *ecc, struct dma_device *dma,
struct device *dev)
{
@@ -862,6 +879,7 @@ static void edma_dma_init(struct edma_cc *ecc, struct dma_device *dma,
dma->device_issue_pending = edma_issue_pending;
dma->device_tx_status = edma_tx_status;
dma->device_control = edma_control;
+ dma->device_slave_caps = edma_dma_device_slave_caps;
dma->dev = dev;
INIT_LIST_HEAD(&dma->channels);
--
1.9.2
^ permalink raw reply related
* [PATCH v3 07/10] dmaengine: edma: Reduce debug print verbosity for non verbose debugging
From: Peter Ujfalusi @ 2014-04-14 11:42 UTC (permalink / raw)
To: dan.j.williams, vinod.koul, nsekhar, joelf
Cc: dmaengine, davinci-linux-open-source, linux-omap, linux-kernel,
linux-arm-kernel
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
Do not print the paRAM information when verbose debugging is not asked and
also reduce the number of lines printed in edma_prep_dma_cyclic()
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
drivers/dma/edma.c | 11 +++++------
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 2f58c04cbcb1..6d9edc47150d 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -141,7 +141,7 @@ static void edma_execute(struct edma_chan *echan)
for (i = 0; i < nslots; i++) {
j = i + edesc->processed;
edma_write_slot(echan->slot[i], &edesc->pset[j]);
- dev_dbg(echan->vchan.chan.device->dev,
+ dev_vdbg(echan->vchan.chan.device->dev,
"\n pset[%d]:\n"
" chnum\t%d\n"
" slot\t%d\n"
@@ -560,9 +560,8 @@ static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
edesc->cyclic = 1;
edesc->pset_nr = nslots;
- dev_dbg(dev, "%s: nslots=%d\n", __func__, nslots);
- dev_dbg(dev, "%s: period_len=%d\n", __func__, period_len);
- dev_dbg(dev, "%s: buf_len=%d\n", __func__, buf_len);
+ dev_dbg(dev, "%s: channel=%d nslots=%d period_len=%zu buf_len=%zu\n",
+ __func__, echan->ch_num, nslots, period_len, buf_len);
for (i = 0; i < nslots; i++) {
/* Allocate a PaRAM slot, if needed */
@@ -596,8 +595,8 @@ static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
else
src_addr += period_len;
- dev_dbg(dev, "%s: Configure period %d of buf:\n", __func__, i);
- dev_dbg(dev,
+ dev_vdbg(dev, "%s: Configure period %d of buf:\n", __func__, i);
+ dev_vdbg(dev,
"\n pset[%d]:\n"
" chnum\t%d\n"
" slot\t%d\n"
--
1.9.2
^ permalink raw reply related
* [PATCH v3 07/10] dmaengine: edma: Reduce debug print verbosity for non verbose debugging
From: Peter Ujfalusi @ 2014-04-14 11:42 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
Do not print the paRAM information when verbose debugging is not asked and
also reduce the number of lines printed in edma_prep_dma_cyclic()
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
drivers/dma/edma.c | 11 +++++------
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 2f58c04cbcb1..6d9edc47150d 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -141,7 +141,7 @@ static void edma_execute(struct edma_chan *echan)
for (i = 0; i < nslots; i++) {
j = i + edesc->processed;
edma_write_slot(echan->slot[i], &edesc->pset[j]);
- dev_dbg(echan->vchan.chan.device->dev,
+ dev_vdbg(echan->vchan.chan.device->dev,
"\n pset[%d]:\n"
" chnum\t%d\n"
" slot\t%d\n"
@@ -560,9 +560,8 @@ static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
edesc->cyclic = 1;
edesc->pset_nr = nslots;
- dev_dbg(dev, "%s: nslots=%d\n", __func__, nslots);
- dev_dbg(dev, "%s: period_len=%d\n", __func__, period_len);
- dev_dbg(dev, "%s: buf_len=%d\n", __func__, buf_len);
+ dev_dbg(dev, "%s: channel=%d nslots=%d period_len=%zu buf_len=%zu\n",
+ __func__, echan->ch_num, nslots, period_len, buf_len);
for (i = 0; i < nslots; i++) {
/* Allocate a PaRAM slot, if needed */
@@ -596,8 +595,8 @@ static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
else
src_addr += period_len;
- dev_dbg(dev, "%s: Configure period %d of buf:\n", __func__, i);
- dev_dbg(dev,
+ dev_vdbg(dev, "%s: Configure period %d of buf:\n", __func__, i);
+ dev_vdbg(dev,
"\n pset[%d]:\n"
" chnum\t%d\n"
" slot\t%d\n"
--
1.9.2
^ permalink raw reply related
* [PATCH v3 08/10] dmaengine: edma: Prefix debug prints where the text were identical in prep callbacks
From: Peter Ujfalusi @ 2014-04-14 11:42 UTC (permalink / raw)
To: dan.j.williams, vinod.koul, nsekhar, joelf
Cc: dmaengine, davinci-linux-open-source, linux-omap, linux-kernel,
linux-arm-kernel
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
prep_slave_sg and prep_dma_cyclic callbacks have mostly same failure cases
with the same texts printed in case we hit them. It helps when debugging if
we know exactly which callback generated the errors.
At the same time change the debug level for descriptor allocation failure
from dbg to err since all other error cases are dev_err and this failure is
similarly fatal as the other ones.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
drivers/dma/edma.c | 14 ++++++++------
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 6d9edc47150d..bc8175c92e0c 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -436,14 +436,14 @@ static struct dma_async_tx_descriptor *edma_prep_slave_sg(
}
if (dev_width == DMA_SLAVE_BUSWIDTH_UNDEFINED) {
- dev_err(dev, "Undefined slave buswidth\n");
+ dev_err(dev, "%s: Undefined slave buswidth\n", __func__);
return NULL;
}
edesc = kzalloc(sizeof(*edesc) + sg_len *
sizeof(edesc->pset[0]), GFP_ATOMIC);
if (!edesc) {
- dev_dbg(dev, "Failed to allocate a descriptor\n");
+ dev_err(dev, "%s: Failed to allocate a descriptor\n", __func__);
return NULL;
}
@@ -459,7 +459,8 @@ static struct dma_async_tx_descriptor *edma_prep_slave_sg(
EDMA_SLOT_ANY);
if (echan->slot[i] < 0) {
kfree(edesc);
- dev_err(dev, "Failed to allocate slot\n");
+ dev_err(dev, "%s: Failed to allocate slot\n",
+ __func__);
return NULL;
}
}
@@ -528,7 +529,7 @@ static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
}
if (dev_width == DMA_SLAVE_BUSWIDTH_UNDEFINED) {
- dev_err(dev, "Undefined slave buswidth\n");
+ dev_err(dev, "%s: Undefined slave buswidth\n", __func__);
return NULL;
}
@@ -553,7 +554,7 @@ static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
edesc = kzalloc(sizeof(*edesc) + nslots *
sizeof(edesc->pset[0]), GFP_ATOMIC);
if (!edesc) {
- dev_dbg(dev, "Failed to allocate a descriptor\n");
+ dev_err(dev, "%s: Failed to allocate a descriptor\n", __func__);
return NULL;
}
@@ -571,7 +572,8 @@ static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
EDMA_SLOT_ANY);
if (echan->slot[i] < 0) {
kfree(edesc);
- dev_err(dev, "Failed to allocate slot\n");
+ dev_err(dev, "%s: Failed to allocate slot\n",
+ __func__);
return NULL;
}
}
--
1.9.2
^ permalink raw reply related
* [PATCH v3 08/10] dmaengine: edma: Prefix debug prints where the text were identical in prep callbacks
From: Peter Ujfalusi @ 2014-04-14 11:42 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
prep_slave_sg and prep_dma_cyclic callbacks have mostly same failure cases
with the same texts printed in case we hit them. It helps when debugging if
we know exactly which callback generated the errors.
At the same time change the debug level for descriptor allocation failure
from dbg to err since all other error cases are dev_err and this failure is
similarly fatal as the other ones.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
drivers/dma/edma.c | 14 ++++++++------
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 6d9edc47150d..bc8175c92e0c 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -436,14 +436,14 @@ static struct dma_async_tx_descriptor *edma_prep_slave_sg(
}
if (dev_width == DMA_SLAVE_BUSWIDTH_UNDEFINED) {
- dev_err(dev, "Undefined slave buswidth\n");
+ dev_err(dev, "%s: Undefined slave buswidth\n", __func__);
return NULL;
}
edesc = kzalloc(sizeof(*edesc) + sg_len *
sizeof(edesc->pset[0]), GFP_ATOMIC);
if (!edesc) {
- dev_dbg(dev, "Failed to allocate a descriptor\n");
+ dev_err(dev, "%s: Failed to allocate a descriptor\n", __func__);
return NULL;
}
@@ -459,7 +459,8 @@ static struct dma_async_tx_descriptor *edma_prep_slave_sg(
EDMA_SLOT_ANY);
if (echan->slot[i] < 0) {
kfree(edesc);
- dev_err(dev, "Failed to allocate slot\n");
+ dev_err(dev, "%s: Failed to allocate slot\n",
+ __func__);
return NULL;
}
}
@@ -528,7 +529,7 @@ static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
}
if (dev_width == DMA_SLAVE_BUSWIDTH_UNDEFINED) {
- dev_err(dev, "Undefined slave buswidth\n");
+ dev_err(dev, "%s: Undefined slave buswidth\n", __func__);
return NULL;
}
@@ -553,7 +554,7 @@ static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
edesc = kzalloc(sizeof(*edesc) + nslots *
sizeof(edesc->pset[0]), GFP_ATOMIC);
if (!edesc) {
- dev_dbg(dev, "Failed to allocate a descriptor\n");
+ dev_err(dev, "%s: Failed to allocate a descriptor\n", __func__);
return NULL;
}
@@ -571,7 +572,8 @@ static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
EDMA_SLOT_ANY);
if (echan->slot[i] < 0) {
kfree(edesc);
- dev_err(dev, "Failed to allocate slot\n");
+ dev_err(dev, "%s: Failed to allocate slot\n",
+ __func__);
return NULL;
}
}
--
1.9.2
^ permalink raw reply related
* [PATCH v3 09/10] dmaengine: edma: Add channel number to debug prints
From: Peter Ujfalusi @ 2014-04-14 11:42 UTC (permalink / raw)
To: dan.j.williams, vinod.koul, nsekhar, joelf
Cc: dmaengine, linux-kernel, linux-arm-kernel, linux-omap,
davinci-linux-open-source
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
It helps to identify issues if we have some information regarding to the
channel which the event is associated.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
drivers/dma/edma.c | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index bc8175c92e0c..4aa5eb005b5c 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -185,7 +185,8 @@ static void edma_execute(struct edma_chan *echan)
edma_resume(echan->ch_num);
if (edesc->processed <= MAX_NR_SG) {
- dev_dbg(dev, "first transfer starting %d\n", echan->ch_num);
+ dev_dbg(dev, "first transfer starting on channel %d\n",
+ echan->ch_num);
edma_start(echan->ch_num);
}
@@ -195,7 +196,7 @@ static void edma_execute(struct edma_chan *echan)
* MAX_NR_SG
*/
if (echan->missed) {
- dev_dbg(dev, "missed event in execute detected\n");
+ dev_dbg(dev, "missed event on channel %d\n", echan->ch_num);
edma_clean_channel(echan->ch_num);
edma_stop(echan->ch_num);
edma_start(echan->ch_num);
@@ -735,7 +736,7 @@ static int edma_alloc_chan_resources(struct dma_chan *chan)
echan->alloced = true;
echan->slot[0] = echan->ch_num;
- dev_dbg(dev, "allocated channel for %u:%u\n",
+ dev_dbg(dev, "allocated channel %d for %u:%u\n", echan->ch_num,
EDMA_CTLR(echan->ch_num), EDMA_CHAN_SLOT(echan->ch_num));
return 0;
--
1.9.2
^ permalink raw reply related
* [PATCH v3 09/10] dmaengine: edma: Add channel number to debug prints
From: Peter Ujfalusi @ 2014-04-14 11:42 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
It helps to identify issues if we have some information regarding to the
channel which the event is associated.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
drivers/dma/edma.c | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index bc8175c92e0c..4aa5eb005b5c 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -185,7 +185,8 @@ static void edma_execute(struct edma_chan *echan)
edma_resume(echan->ch_num);
if (edesc->processed <= MAX_NR_SG) {
- dev_dbg(dev, "first transfer starting %d\n", echan->ch_num);
+ dev_dbg(dev, "first transfer starting on channel %d\n",
+ echan->ch_num);
edma_start(echan->ch_num);
}
@@ -195,7 +196,7 @@ static void edma_execute(struct edma_chan *echan)
* MAX_NR_SG
*/
if (echan->missed) {
- dev_dbg(dev, "missed event in execute detected\n");
+ dev_dbg(dev, "missed event on channel %d\n", echan->ch_num);
edma_clean_channel(echan->ch_num);
edma_stop(echan->ch_num);
edma_start(echan->ch_num);
@@ -735,7 +736,7 @@ static int edma_alloc_chan_resources(struct dma_chan *chan)
echan->alloced = true;
echan->slot[0] = echan->ch_num;
- dev_dbg(dev, "allocated channel for %u:%u\n",
+ dev_dbg(dev, "allocated channel %d for %u:%u\n", echan->ch_num,
EDMA_CTLR(echan->ch_num), EDMA_CHAN_SLOT(echan->ch_num));
return 0;
--
1.9.2
^ permalink raw reply related
* [PATCH v3 10/10] dmaengine: edma: Print the direction value as well when it is not supported
From: Peter Ujfalusi @ 2014-04-14 11:42 UTC (permalink / raw)
To: dan.j.williams, vinod.koul, nsekhar, joelf
Cc: dmaengine, linux-kernel, linux-arm-kernel, linux-omap,
davinci-linux-open-source
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
In case of not supported direction it is better to print the direction also.
It is unlikely, but in such an event it helps with the debugging.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
drivers/dma/edma.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 4aa5eb005b5c..91849aa50de1 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -432,7 +432,7 @@ static struct dma_async_tx_descriptor *edma_prep_slave_sg(
dev_width = echan->cfg.dst_addr_width;
burst = echan->cfg.dst_maxburst;
} else {
- dev_err(dev, "%s: bad direction?\n", __func__);
+ dev_err(dev, "%s: bad direction: %d\n", __func__, direction);
return NULL;
}
@@ -525,7 +525,7 @@ static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
dev_width = echan->cfg.dst_addr_width;
burst = echan->cfg.dst_maxburst;
} else {
- dev_err(dev, "%s: bad direction?\n", __func__);
+ dev_err(dev, "%s: bad direction: %d\n", __func__, direction);
return NULL;
}
--
1.9.2
^ permalink raw reply related
* [PATCH v3 10/10] dmaengine: edma: Print the direction value as well when it is not supported
From: Peter Ujfalusi @ 2014-04-14 11:42 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
In case of not supported direction it is better to print the direction also.
It is unlikely, but in such an event it helps with the debugging.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
drivers/dma/edma.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 4aa5eb005b5c..91849aa50de1 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -432,7 +432,7 @@ static struct dma_async_tx_descriptor *edma_prep_slave_sg(
dev_width = echan->cfg.dst_addr_width;
burst = echan->cfg.dst_maxburst;
} else {
- dev_err(dev, "%s: bad direction?\n", __func__);
+ dev_err(dev, "%s: bad direction: %d\n", __func__, direction);
return NULL;
}
@@ -525,7 +525,7 @@ static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
dev_width = echan->cfg.dst_addr_width;
burst = echan->cfg.dst_maxburst;
} else {
- dev_err(dev, "%s: bad direction?\n", __func__);
+ dev_err(dev, "%s: bad direction: %d\n", __func__, direction);
return NULL;
}
--
1.9.2
^ permalink raw reply related
* Re: [Query]: tick-sched: why don't we stop tick when we are running idle task?
From: Viresh Kumar @ 2014-04-14 11:42 UTC (permalink / raw)
To: Peter Zijlstra
Cc: Frederic Weisbecker, Thomas Gleixner, Linux Kernel Mailing List,
Lists linaro-kernel
In-Reply-To: <20140414110245.GG11096@twins.programming.kicks-ass.net>
On 14 April 2014 16:32, Peter Zijlstra <peterz@infradead.org> wrote:
> I'm still not sure _what_ you're trying to solve here. What are you
> doing and why?
Hi Peter,
We are working building ARM Networking machines. Networking Data
plane is handled completely at user space. At run time we may fix
any number of CPUs for data plane activities. There will be a single
user space thread per CPU for these data plane packet processing.
Due to timing constraints these cores can't allow any interruption
from kernel. These include interruption from:
- other tasks: Fixed with cpusets
- timers/hrtimers: Implemented cpuset.quiesce as you suggested:
Waiting for reviews
- workqueues: Probably would be fixed by Frederic's work.
- Tick: Even with NO_HZ_FULL we get a tick every second. This is
what I am trying to address here. Frederic earlier suggested to
offload this accounting to other CPUs and so was my initial proposal.
Please let me know what's the right way to get this fixed and I will
try it that way.
Thanks for your inputs.
^ permalink raw reply
* [PATCH v3 00/10] dma: edma: Fixes for cyclic (audio) operation
From: Peter Ujfalusi @ 2014-04-14 11:41 UTC (permalink / raw)
To: dan.j.williams, vinod.koul, nsekhar, joelf
Cc: dmaengine, linux-kernel, linux-arm-kernel, linux-omap,
davinci-linux-open-source
Hi,
Changes since v2:
- Dropped patch 10 from v2 (simplify direction configuration...)
- Dropped the channel priority related patches since we are going to go via
different route for configuring the priority.
- Added ACK from Joel for the patches since they are not changed since v2
Changes since v1:
- ASoC patches removed
- Comments from Andriy Shevchenko addressed
- patch added to fix cases when src/dst_maxburst is set to 0
The series contains now only:
Support for DMA pause/resume in cyclic mode
device_slave_caps callback and DMA_CYCLIC flag correction.
While debugging the edma to get things sorted out I noticed that the debug was
too verbose and the important information was hidden even when the we did not
asked for verbose dmaengine debug.
I have included some debug cleanups for the edma dmaengine driver also.
Regards,
Peter
---
Peter Ujfalusi (10):
platform_data: edma: Be precise with the paRAM struct
arm: common: edma: Save the number of event queues/TCs
dmaengine: edma: Correct the handling of src/dst_maxburst == 0
dmaengine: edma: Add support for DMA_PAUSE/RESUME operation
dmaengine: edma: Set DMA_CYCLIC capability flag
dmaengine: edma: Implement device_slave_caps callback
dmaengine: edma: Reduce debug print verbosity for non verbose
debugging
dmaengine: edma: Prefix debug prints where the text were identical in
prep callbacks
dmaengine: edma: Add channel number to debug prints
dmaengine: edma: Print the direction value as well when it is not
supported
arch/arm/common/edma.c | 4 ++
drivers/dma/edma.c | 87 ++++++++++++++++++++++++++++++--------
include/linux/platform_data/edma.h | 18 ++++----
3 files changed, 83 insertions(+), 26 deletions(-)
--
1.9.2
^ permalink raw reply
* [PATCH v3 01/10] platform_data: edma: Be precise with the paRAM struct
From: Peter Ujfalusi @ 2014-04-14 11:41 UTC (permalink / raw)
To: dan.j.williams, vinod.koul, nsekhar, joelf
Cc: dmaengine, linux-kernel, linux-arm-kernel, linux-omap,
davinci-linux-open-source
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
The edmacc_param struct should follow the layout of the paRAM area in the
HW. Be explicit on the size of the fields (u32) and also mark the struct
as packed to avoid any padding on non 32bit architectures.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
include/linux/platform_data/edma.h | 18 +++++++++---------
1 file changed, 9 insertions(+), 9 deletions(-)
diff --git a/include/linux/platform_data/edma.h b/include/linux/platform_data/edma.h
index f50821cb64be..923f8a3e4ce0 100644
--- a/include/linux/platform_data/edma.h
+++ b/include/linux/platform_data/edma.h
@@ -43,15 +43,15 @@
/* PaRAM slots are laid out like this */
struct edmacc_param {
- unsigned int opt;
- unsigned int src;
- unsigned int a_b_cnt;
- unsigned int dst;
- unsigned int src_dst_bidx;
- unsigned int link_bcntrld;
- unsigned int src_dst_cidx;
- unsigned int ccnt;
-};
+ u32 opt;
+ u32 src;
+ u32 a_b_cnt;
+ u32 dst;
+ u32 src_dst_bidx;
+ u32 link_bcntrld;
+ u32 src_dst_cidx;
+ u32 ccnt;
+} __packed;
/* fields in edmacc_param.opt */
#define SAM BIT(0)
--
1.9.2
^ permalink raw reply related
* Re: TUN device performance regression
From: Martin Filip @ 2014-04-14 11:42 UTC (permalink / raw)
To: netdev
In-Reply-To: <1397475118.11864.5.camel@interceptor.in.smoula.net>
> Hello,
>
> I've noticed significant TUN device performance drop between 3.13 and
> 3.14 kernels, problem seems to persist until current git version.
>
> With good kernel I can achieve full 100mbps through openvpn without any
> problem, with problematic kernel TX performance drops down to ~50kbps.
>
> According to git bisect it seems that root of all evil is commit
> 53d6471cef17262d3ad1c7ce8982a234244f68ec.
>
> I can confirm that performance is great again with current kernel when I
> revert this one.
>
I've forgot to mention kernel bug:
https://bugzilla.kernel.org/show_bug.cgi?id=74051
--
Martin Filip
e-mail: nexus@smoula.net
jabberID: nexus@smoula.net
http://www.smoula.net
^ permalink raw reply
* [PATCH v3 05/10] dmaengine: edma: Set DMA_CYCLIC capability flag
From: Peter Ujfalusi @ 2014-04-14 11:42 UTC (permalink / raw)
To: dan.j.williams, vinod.koul, nsekhar, joelf
Cc: dmaengine, linux-kernel, linux-arm-kernel, linux-omap,
davinci-linux-open-source
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
Indicate that the edma dmaengine driver has support for cyclic mode.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
arch/arm/common/edma.c | 1 +
drivers/dma/edma.c | 1 +
2 files changed, 2 insertions(+)
diff --git a/arch/arm/common/edma.c b/arch/arm/common/edma.c
index 999266bf69b9..0b37f7734d0f 100644
--- a/arch/arm/common/edma.c
+++ b/arch/arm/common/edma.c
@@ -1574,6 +1574,7 @@ static struct edma_soc_info *edma_setup_info_from_dt(struct device *dev,
return ERR_PTR(ret);
dma_cap_set(DMA_SLAVE, edma_filter_info.dma_cap);
+ dma_cap_set(DMA_CYCLIC, edma_filter_info.dma_cap);
of_dma_controller_register(dev->of_node, of_dma_simple_xlate,
&edma_filter_info);
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 7891378a03f0..1dd9e8806975 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -891,6 +891,7 @@ static int edma_probe(struct platform_device *pdev)
dma_cap_zero(ecc->dma_slave.cap_mask);
dma_cap_set(DMA_SLAVE, ecc->dma_slave.cap_mask);
+ dma_cap_set(DMA_CYCLIC, ecc->dma_slave.cap_mask);
edma_dma_init(ecc, &ecc->dma_slave, &pdev->dev);
--
1.9.2
^ permalink raw reply related
* [PATCH v3 06/10] dmaengine: edma: Implement device_slave_caps callback
From: Peter Ujfalusi @ 2014-04-14 11:42 UTC (permalink / raw)
To: dan.j.williams, vinod.koul, nsekhar, joelf
Cc: dmaengine, linux-kernel, linux-arm-kernel, linux-omap,
davinci-linux-open-source
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
With the callback implemented omap-dma can provide information to client
drivers regarding to supported address widths, directions, residue
granularity, etc.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
drivers/dma/edma.c | 18 ++++++++++++++++++
1 file changed, 18 insertions(+)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 1dd9e8806975..2f58c04cbcb1 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -852,6 +852,23 @@ static void __init edma_chan_init(struct edma_cc *ecc,
}
}
+#define EDMA_DMA_BUSWIDTHS (BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) | \
+ BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) | \
+ BIT(DMA_SLAVE_BUSWIDTH_4_BYTES))
+
+static int edma_dma_device_slave_caps(struct dma_chan *dchan,
+ struct dma_slave_caps *caps)
+{
+ caps->src_addr_widths = EDMA_DMA_BUSWIDTHS;
+ caps->dstn_addr_widths = EDMA_DMA_BUSWIDTHS;
+ caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
+ caps->cmd_pause = true;
+ caps->cmd_terminate = true;
+ caps->residue_granularity = DMA_RESIDUE_GRANULARITY_DESCRIPTOR;
+
+ return 0;
+}
+
static void edma_dma_init(struct edma_cc *ecc, struct dma_device *dma,
struct device *dev)
{
@@ -862,6 +879,7 @@ static void edma_dma_init(struct edma_cc *ecc, struct dma_device *dma,
dma->device_issue_pending = edma_issue_pending;
dma->device_tx_status = edma_tx_status;
dma->device_control = edma_control;
+ dma->device_slave_caps = edma_dma_device_slave_caps;
dma->dev = dev;
INIT_LIST_HEAD(&dma->channels);
--
1.9.2
^ permalink raw reply related
* [PATCH v3 04/10] dmaengine: edma: Add support for DMA_PAUSE/RESUME operation
From: Peter Ujfalusi @ 2014-04-14 11:41 UTC (permalink / raw)
To: dan.j.williams, vinod.koul, nsekhar, joelf
Cc: dmaengine, linux-kernel, linux-arm-kernel, linux-omap,
davinci-linux-open-source
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
Pause/Resume can be used by the audio stack when the stream is paused/resumed
The edma platform code has support for this and the legacy audio stack used
this.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
drivers/dma/edma.c | 28 ++++++++++++++++++++++++++++
1 file changed, 28 insertions(+)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 2742867fd1e6..7891378a03f0 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -240,6 +240,26 @@ static int edma_slave_config(struct edma_chan *echan,
return 0;
}
+static int edma_dma_pause(struct edma_chan *echan)
+{
+ /* Pause/Resume only allowed with cyclic mode */
+ if (!echan->edesc->cyclic)
+ return -EINVAL;
+
+ edma_pause(echan->ch_num);
+ return 0;
+}
+
+static int edma_dma_resume(struct edma_chan *echan)
+{
+ /* Pause/Resume only allowed with cyclic mode */
+ if (!echan->edesc->cyclic)
+ return -EINVAL;
+
+ edma_resume(echan->ch_num);
+ return 0;
+}
+
static int edma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
unsigned long arg)
{
@@ -255,6 +275,14 @@ static int edma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
config = (struct dma_slave_config *)arg;
ret = edma_slave_config(echan, config);
break;
+ case DMA_PAUSE:
+ ret = edma_dma_pause(echan);
+ break;
+
+ case DMA_RESUME:
+ ret = edma_dma_resume(echan);
+ break;
+
default:
ret = -ENOSYS;
}
--
1.9.2
^ permalink raw reply related
* [PATCH v3 06/10] dmaengine: edma: Implement device_slave_caps callback
From: Peter Ujfalusi @ 2014-04-14 11:42 UTC (permalink / raw)
To: dan.j.williams, vinod.koul, nsekhar, joelf
Cc: dmaengine, linux-kernel, linux-arm-kernel, linux-omap,
davinci-linux-open-source
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
With the callback implemented omap-dma can provide information to client
drivers regarding to supported address widths, directions, residue
granularity, etc.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
drivers/dma/edma.c | 18 ++++++++++++++++++
1 file changed, 18 insertions(+)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 1dd9e8806975..2f58c04cbcb1 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -852,6 +852,23 @@ static void __init edma_chan_init(struct edma_cc *ecc,
}
}
+#define EDMA_DMA_BUSWIDTHS (BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) | \
+ BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) | \
+ BIT(DMA_SLAVE_BUSWIDTH_4_BYTES))
+
+static int edma_dma_device_slave_caps(struct dma_chan *dchan,
+ struct dma_slave_caps *caps)
+{
+ caps->src_addr_widths = EDMA_DMA_BUSWIDTHS;
+ caps->dstn_addr_widths = EDMA_DMA_BUSWIDTHS;
+ caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
+ caps->cmd_pause = true;
+ caps->cmd_terminate = true;
+ caps->residue_granularity = DMA_RESIDUE_GRANULARITY_DESCRIPTOR;
+
+ return 0;
+}
+
static void edma_dma_init(struct edma_cc *ecc, struct dma_device *dma,
struct device *dev)
{
@@ -862,6 +879,7 @@ static void edma_dma_init(struct edma_cc *ecc, struct dma_device *dma,
dma->device_issue_pending = edma_issue_pending;
dma->device_tx_status = edma_tx_status;
dma->device_control = edma_control;
+ dma->device_slave_caps = edma_dma_device_slave_caps;
dma->dev = dev;
INIT_LIST_HEAD(&dma->channels);
--
1.9.2
^ permalink raw reply related
* [PATCH v3 05/10] dmaengine: edma: Set DMA_CYCLIC capability flag
From: Peter Ujfalusi @ 2014-04-14 11:42 UTC (permalink / raw)
To: dan.j.williams, vinod.koul, nsekhar, joelf
Cc: dmaengine, linux-kernel, linux-arm-kernel, linux-omap,
davinci-linux-open-source
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
Indicate that the edma dmaengine driver has support for cyclic mode.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
arch/arm/common/edma.c | 1 +
drivers/dma/edma.c | 1 +
2 files changed, 2 insertions(+)
diff --git a/arch/arm/common/edma.c b/arch/arm/common/edma.c
index 999266bf69b9..0b37f7734d0f 100644
--- a/arch/arm/common/edma.c
+++ b/arch/arm/common/edma.c
@@ -1574,6 +1574,7 @@ static struct edma_soc_info *edma_setup_info_from_dt(struct device *dev,
return ERR_PTR(ret);
dma_cap_set(DMA_SLAVE, edma_filter_info.dma_cap);
+ dma_cap_set(DMA_CYCLIC, edma_filter_info.dma_cap);
of_dma_controller_register(dev->of_node, of_dma_simple_xlate,
&edma_filter_info);
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 7891378a03f0..1dd9e8806975 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -891,6 +891,7 @@ static int edma_probe(struct platform_device *pdev)
dma_cap_zero(ecc->dma_slave.cap_mask);
dma_cap_set(DMA_SLAVE, ecc->dma_slave.cap_mask);
+ dma_cap_set(DMA_CYCLIC, ecc->dma_slave.cap_mask);
edma_dma_init(ecc, &ecc->dma_slave, &pdev->dev);
--
1.9.2
^ permalink raw reply related
* [PATCH v3 09/10] dmaengine: edma: Add channel number to debug prints
From: Peter Ujfalusi @ 2014-04-14 11:42 UTC (permalink / raw)
To: dan.j.williams, vinod.koul, nsekhar, joelf
Cc: dmaengine, linux-kernel, linux-arm-kernel, linux-omap,
davinci-linux-open-source
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
It helps to identify issues if we have some information regarding to the
channel which the event is associated.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
drivers/dma/edma.c | 7 ++++---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index bc8175c92e0c..4aa5eb005b5c 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -185,7 +185,8 @@ static void edma_execute(struct edma_chan *echan)
edma_resume(echan->ch_num);
if (edesc->processed <= MAX_NR_SG) {
- dev_dbg(dev, "first transfer starting %d\n", echan->ch_num);
+ dev_dbg(dev, "first transfer starting on channel %d\n",
+ echan->ch_num);
edma_start(echan->ch_num);
}
@@ -195,7 +196,7 @@ static void edma_execute(struct edma_chan *echan)
* MAX_NR_SG
*/
if (echan->missed) {
- dev_dbg(dev, "missed event in execute detected\n");
+ dev_dbg(dev, "missed event on channel %d\n", echan->ch_num);
edma_clean_channel(echan->ch_num);
edma_stop(echan->ch_num);
edma_start(echan->ch_num);
@@ -735,7 +736,7 @@ static int edma_alloc_chan_resources(struct dma_chan *chan)
echan->alloced = true;
echan->slot[0] = echan->ch_num;
- dev_dbg(dev, "allocated channel for %u:%u\n",
+ dev_dbg(dev, "allocated channel %d for %u:%u\n", echan->ch_num,
EDMA_CTLR(echan->ch_num), EDMA_CHAN_SLOT(echan->ch_num));
return 0;
--
1.9.2
^ permalink raw reply related
* [PATCH v3 10/10] dmaengine: edma: Print the direction value as well when it is not supported
From: Peter Ujfalusi @ 2014-04-14 11:42 UTC (permalink / raw)
To: dan.j.williams, vinod.koul, nsekhar, joelf
Cc: dmaengine, linux-kernel, linux-arm-kernel, linux-omap,
davinci-linux-open-source
In-Reply-To: <1397475725-5036-1-git-send-email-peter.ujfalusi@ti.com>
In case of not supported direction it is better to print the direction also.
It is unlikely, but in such an event it helps with the debugging.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Joel Fernandes <joelf@ti.com>
---
drivers/dma/edma.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 4aa5eb005b5c..91849aa50de1 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -432,7 +432,7 @@ static struct dma_async_tx_descriptor *edma_prep_slave_sg(
dev_width = echan->cfg.dst_addr_width;
burst = echan->cfg.dst_maxburst;
} else {
- dev_err(dev, "%s: bad direction?\n", __func__);
+ dev_err(dev, "%s: bad direction: %d\n", __func__, direction);
return NULL;
}
@@ -525,7 +525,7 @@ static struct dma_async_tx_descriptor *edma_prep_dma_cyclic(
dev_width = echan->cfg.dst_addr_width;
burst = echan->cfg.dst_maxburst;
} else {
- dev_err(dev, "%s: bad direction?\n", __func__);
+ dev_err(dev, "%s: bad direction: %d\n", __func__, direction);
return NULL;
}
--
1.9.2
^ permalink raw reply related
* Re: Statistics at the host for network nic that was PCI-Passthrough
From: Shimon Zadok @ 2014-04-14 11:43 UTC (permalink / raw)
To: Ian Campbell; +Cc: Xen-Developement
In-Reply-To: <1397468101.9862.228.camel@kazak.uk.xensource.com>
[-- Attachment #1.1: Type: text/plain, Size: 1397 bytes --]
The guest is not in my control. Use case is: user uploads a Guest VM he made by himself...
It could be the case that I give him an sr-iov (vf) by pci-passthrough or a pci device (not vf). I would still like to get a sense of the pci at the host level...
On April 14, 2014 12:35:01 PM GMT+02:00, Ian Campbell <Ian.Campbell@citrix.com> wrote:
>On Mon, 2014-04-14 at 12:32 +0300, Shimon Zadok wrote:
>>
>> -------- Original Message --------
>> Subject: Re: [Xen-devel] Statistics at the host for network nic that
>> was PCI-Passthrough
>> Date: 14-04-2014 12:30
>> From: Shimon Zadok <shimonz@startpo.co.il>
>> To: Ian Campbell <Ian.Campbell@citrix.com>
>>
>> Is it possible to write an extension to pciback to load a module
>> specific to that h/w?
>> It could be a new driver for that hardware that operates by itself
>for
>> reading these statistics from the nic...
>
>Then you would have uncoordinated accesses to the hardware from two
>domains. That might work for some hardware I suppose but it certainly
>isn't a good idea in general.
>
>You'd be better off building something in the domu to export the stats
>e.g. via xenstore. It could be entirely userspace I expect.
>
>Ian.
>
>
>
>_______________________________________________
>Xen-devel mailing list
>Xen-devel@lists.xen.org
>http://lists.xen.org/xen-devel
--
Sent from my Android phone with K-9 Mail. Please excuse my brevity.
[-- Attachment #1.2: Type: text/html, Size: 1913 bytes --]
[-- Attachment #2: Type: text/plain, Size: 126 bytes --]
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
^ permalink raw reply
* Re: [PATCH] [PATCH V2] tty: memleak in alloc_pid
From: Greg Kroah-Hartman @ 2014-04-14 11:47 UTC (permalink / raw)
To: Chen Tingjie; +Cc: Jiri Slaby, linux-kernel, Zhang Jun
In-Reply-To: <1397460675-17983-1-git-send-email-tingjie.chen@intel.com>
On Mon, Apr 14, 2014 at 03:31:15PM +0800, Chen Tingjie wrote:
> There is memleak in alloc_pid:
> ------------------------------
> unreferenced object 0xd3453a80 (size 64):
> comm "adbd", pid 1730, jiffies 66363 (age 6586.950s)
> hex dump (first 32 bytes):
> 01 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
> 00 00 00 00 40 c2 f6 d5 00 d3 25 c1 59 28 00 00 ....@.....%.Y(..
> backtrace:
> [<c1a6f15c>] kmemleak_alloc+0x3c/0xa0
> [<c1320546>] kmem_cache_alloc+0xc6/0x190
> [<c125d51e>] alloc_pid+0x1e/0x400
> [<c123d344>] copy_process.part.39+0xad4/0x1120
> [<c123da59>] do_fork+0x99/0x330
> [<c123dd58>] sys_fork+0x28/0x30
> [<c1a89a08>] syscall_call+0x7/0xb
> [<ffffffff>] 0xffffffff
>
> the leak is due to unreleased pid->count, which execute in function:
> get_pid()(pid->count++) and put_pid()(pid->count--).
>
> The race condition as following:
> task[dumpsys] task[adbd]
> in disassociate_ctty() in tty_signal_session_leader()
> ----------------------- -------------------------
> tty = get_current_tty();
> // tty is not NULL
> ...
> spin_lock_irq(¤t->sighand->siglock);
> put_pid(current->signal->tty_old_pgrp);
> current->signal->tty_old_pgrp = NULL;
> spin_unlock_irq(¤t->sighand->siglock);
>
> spin_lock_irq(&p->sighand->siglock);
> ...
> p->signal->tty = NULL;
> ...
> spin_unlock_irq(&p->sighand->siglock);
>
> tty = get_current_tty();
> // tty NULL, goto else branch by accident.
> if (tty) {
> ...
> put_pid(tty_session);
> put_pid(tty_pgrp);
> ...
> } else {
> print msg
> }
>
> in task[dumpsys], in disassociate_ctty(), tty is set NULL by task[adbd],
> tty_signal_session_leader(), then it goto else branch and lack of
> put_pid(), cause memleak.
>
> move spin_unlock(sighand->siglock) after get_current_tty() can avoid
> the race and fix the memleak.
>
> Change-Id: Ic960dda039c8f99aad3e0f4d176489a966c62f6a
Why is this line here?
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.