[PATCH 0/5] spi: spi-zynqmp-gqspi: fix spi issues

[PATCH 0/5] spi: spi-zynqmp-gqspi: fix spi issues

[quanyang.wang]

From: Quanyang Wang <quanyang.wang@windriver.com>

Hi all,

This series fix some issues that occurs in spi-zynqmp-gqspi.c.

Thanks,
Quanyang

Amit Kumar Mahapatra (1):
  spi: spi-zynqmp-gqspi: Resolved slab-out-of-bounds bug

Quanyang Wang (4):
  spi: spi-zynqmp-gqspi: fix clk_enable/disable imbalance issue
  spi: spi-zynqmp-gqspi: fix hang issue when suspend/resume
  spi: spi-zynqmp-gqspi: fix use-after-free in zynqmp_qspi_exec_op
  spi: spi-zynqmp-gqspi: return -ENOMEM if dma_map_single fails

 drivers/spi/spi-zynqmp-gqspi.c | 115 +++++++++++++++------------------
 1 file changed, 51 insertions(+), 64 deletions(-)

-- 
2.25.1

[PATCH 1/5] spi: spi-zynqmp-gqspi: fix clk_enable/disable imbalance issue

[quanyang.wang]

From: Quanyang Wang <quanyang.wang@windriver.com>

The clks "pclk" and "ref_clk" are enabled twice during the probe. The
first time is in the function zynqmp_qspi_probe and the second time is
in zynqmp_qspi_setup_op which is called by devm_spi_register_controller.
Then calling zynqmp_qspi_remove (rmmod this module) to disable these clks
will trigger a warning as below:

[  309.124604] Unpreparing enabled qspi_ref
[  309.128641] WARNING: CPU: 1 PID: 537 at drivers/clk/clk.c:824 clk_core_unprepare+0x108/0x110

Since pm_runtime works now, clks can be enabled/disabled by calling
zynqmp_runtime_suspend/resume. So we don't need to enable these clks
explicitly in zynqmp_qspi_setup_op. Remove them to fix this issue.

And remove clk enabling/disabling in zynqmp_qspi_resume because there is
no spi transfer operation so enabling ref_clk is redundant meanwhile pclk
is not disabled for it is shared with other peripherals.

Furthermore replace clk_enable/disable with clk_prepare_enable and
clk_disable_unprepare in runtime_suspend/resume functions.

Fixes: 1c26372e5aa9 ("spi: spi-zynqmp-gqspi: Update driver to use spi-mem framework")
Signed-off-by: Quanyang Wang <quanyang.wang@windriver.com>
---
 drivers/spi/spi-zynqmp-gqspi.c | 47 ++++++----------------------------
 1 file changed, 8 insertions(+), 39 deletions(-)

diff --git a/drivers/spi/spi-zynqmp-gqspi.c b/drivers/spi/spi-zynqmp-gqspi.c
index 32e53f379e9b..f9056f0a480c 100644
--- a/drivers/spi/spi-zynqmp-gqspi.c
+++ b/drivers/spi/spi-zynqmp-gqspi.c
@@ -487,24 +487,10 @@ static int zynqmp_qspi_setup_op(struct spi_device *qspi)
 {
 	struct spi_controller *ctlr = qspi->master;
 	struct zynqmp_qspi *xqspi = spi_controller_get_devdata(ctlr);
-	struct device *dev = &ctlr->dev;
-	int ret;
 
 	if (ctlr->busy)
 		return -EBUSY;
 
-	ret = clk_enable(xqspi->refclk);
-	if (ret) {
-		dev_err(dev, "Cannot enable device clock.\n");
-		return ret;
-	}
-
-	ret = clk_enable(xqspi->pclk);
-	if (ret) {
-		dev_err(dev, "Cannot enable APB clock.\n");
-		clk_disable(xqspi->refclk);
-		return ret;
-	}
 	zynqmp_gqspi_write(xqspi, GQSPI_EN_OFST, GQSPI_EN_MASK);
 
 	return 0;
@@ -863,26 +849,9 @@ static int __maybe_unused zynqmp_qspi_suspend(struct device *dev)
 static int __maybe_unused zynqmp_qspi_resume(struct device *dev)
 {
 	struct spi_controller *ctlr = dev_get_drvdata(dev);
-	struct zynqmp_qspi *xqspi = spi_controller_get_devdata(ctlr);
-	int ret = 0;
-
-	ret = clk_enable(xqspi->pclk);
-	if (ret) {
-		dev_err(dev, "Cannot enable APB clock.\n");
-		return ret;
-	}
-
-	ret = clk_enable(xqspi->refclk);
-	if (ret) {
-		dev_err(dev, "Cannot enable device clock.\n");
-		clk_disable(xqspi->pclk);
-		return ret;
-	}
 
 	spi_controller_resume(ctlr);
 
-	clk_disable(xqspi->refclk);
-	clk_disable(xqspi->pclk);
 	return 0;
 }
 
@@ -898,8 +867,8 @@ static int __maybe_unused zynqmp_runtime_suspend(struct device *dev)
 {
 	struct zynqmp_qspi *xqspi = (struct zynqmp_qspi *)dev_get_drvdata(dev);
 
-	clk_disable(xqspi->refclk);
-	clk_disable(xqspi->pclk);
+	clk_disable_unprepare(xqspi->refclk);
+	clk_disable_unprepare(xqspi->pclk);
 
 	return 0;
 }
@@ -917,16 +886,16 @@ static int __maybe_unused zynqmp_runtime_resume(struct device *dev)
 	struct zynqmp_qspi *xqspi = (struct zynqmp_qspi *)dev_get_drvdata(dev);
 	int ret;
 
-	ret = clk_enable(xqspi->pclk);
+	ret = clk_prepare_enable(xqspi->pclk);
 	if (ret) {
 		dev_err(dev, "Cannot enable APB clock.\n");
 		return ret;
 	}
 
-	ret = clk_enable(xqspi->refclk);
+	ret = clk_prepare_enable(xqspi->refclk);
 	if (ret) {
 		dev_err(dev, "Cannot enable device clock.\n");
-		clk_disable(xqspi->pclk);
+		clk_disable_unprepare(xqspi->pclk);
 		return ret;
 	}
 
@@ -1136,13 +1105,11 @@ static int zynqmp_qspi_probe(struct platform_device *pdev)
 		goto remove_master;
 	}
 
-	init_completion(&xqspi->data_completion);
-
 	xqspi->refclk = devm_clk_get(&pdev->dev, "ref_clk");
 	if (IS_ERR(xqspi->refclk)) {
 		dev_err(dev, "ref_clk clock not found.\n");
 		ret = PTR_ERR(xqspi->refclk);
-		goto clk_dis_pclk;
+		goto remove_master;
 	}
 
 	ret = clk_prepare_enable(xqspi->pclk);
@@ -1157,6 +1124,8 @@ static int zynqmp_qspi_probe(struct platform_device *pdev)
 		goto clk_dis_pclk;
 	}
 
+	init_completion(&xqspi->data_completion);
+
 	mutex_init(&xqspi->op_lock);
 
 	pm_runtime_use_autosuspend(&pdev->dev);
-- 
2.25.1

Re: [PATCH 1/5] spi: spi-zynqmp-gqspi: fix clk_enable/disable imbalance issue

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 691 bytes --]

On Fri, Apr 16, 2021 at 08:46:48AM +0800, quanyang.wang@windriver.com wrote:

> Since pm_runtime works now, clks can be enabled/disabled by calling
> zynqmp_runtime_suspend/resume. So we don't need to enable these clks
> explicitly in zynqmp_qspi_setup_op. Remove them to fix this issue.

> Fixes: 1c26372e5aa9 ("spi: spi-zynqmp-gqspi: Update driver to use spi-mem framework")

Are you *sure* this fixes is accurate?  The patch (and several of the
others that flag the same commit) doesn't apply against for-5.12, though
at this point there's not really enough time to send another pull request
so it doesn't super matter though someone will probably need to help out
with stable backports.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [PATCH 1/5] spi: spi-zynqmp-gqspi: fix clk_enable/disable imbalance issue

[quanyang.wang]

Hi Mark,

On 4/16/21 8:55 PM, Mark Brown wrote:
> On Fri, Apr 16, 2021 at 08:46:48AM +0800, quanyang.wang@windriver.com wrote:
>
>> Since pm_runtime works now, clks can be enabled/disabled by calling
>> zynqmp_runtime_suspend/resume. So we don't need to enable these clks
>> explicitly in zynqmp_qspi_setup_op. Remove them to fix this issue.
>> Fixes: 1c26372e5aa9 ("spi: spi-zynqmp-gqspi: Update driver to use spi-mem framework")
> Are you *sure* this fixes is accurate?  The patch (and several of the
> others that flag the same commit) doesn't apply against for-5.12, though
> at this point there's not really enough time to send another pull request
> so it doesn't super matter though someone will probably need to help out
> with stable backports.

I am sorry. These patches should NOT be with "Fixes" tag since they base 
on the patches

which are not with "Fixes". May I send a V2 patch series which remove 
these "Fixes" tags?

Thanks,

Quanyang

Re: [PATCH 1/5] spi: spi-zynqmp-gqspi: fix clk_enable/disable imbalance issue

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 489 bytes --]

On Fri, Apr 16, 2021 at 10:04:30PM +0800, quanyang.wang wrote:

> I am sorry. These patches should NOT be with "Fixes" tag since they base on
> the patches

> which are not with "Fixes". May I send a V2 patch series which remove these
> "Fixes" tags?

Well, if they're fixing bugs that were present in the named commit then
the the tag makes sense, it's just a question of if they are actually
for that commit or if they are fixing things for other commits like the
runtime PM enablement.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: [PATCH 1/5] spi: spi-zynqmp-gqspi: fix clk_enable/disable imbalance issue

[Quanyang Wang]

Hi Mark,

On 4/16/21 11:12 PM, Mark Brown wrote:
> On Fri, Apr 16, 2021 at 10:04:30PM +0800, quanyang.wang wrote:
>
>> I am sorry. These patches should NOT be with "Fixes" tag since they base on
>> the patches
>> which are not with "Fixes". May I send a V2 patch series which remove these
>> "Fixes" tags?
> Well, if they're fixing bugs that were present in the named commit then
> the the tag makes sense, it's just a question of if they are actually
> for that commit or if they are fixing things for other commits like the
> runtime PM enablement.

Yes,  they are fixing bugs which are introduced by the named commit.

But if only picking they to stable without the patch "spi: 
spi-zynqmp-gqspi: Fix runtime PM imbalance in zynqmp_qspi_probe",

this spi driver will not work. I don't know how to handle this 
condition, so I send a V2 patch series to delete the tags.

Thanks,

Quanyang

[PATCH 2/5] spi: spi-zynqmp-gqspi: fix hang issue when suspend/resume

[quanyang.wang]

From: Quanyang Wang <quanyang.wang@windriver.com>

After calling platform_set_drvdata(pdev, xqspi) in probe, the return
value of dev_get_drvdata(dev) is a pointer to struct zynqmp_qspi but
not struct spi_controller. A wrong structure type passing to the
functions spi_controller_suspend/resume will hang the system.

And we should check the return value of spi_controller_suspend, if
an error is returned, return it to PM subsystem to stop suspend.

Also, GQSPI_EN_MASK should be written to GQSPI_EN_OFST to enable
the spi controller in zynqmp_qspi_resume since it was disabled in
zynqmp_qspi_suspend before.

Fixes: 1c26372e5aa9 ("spi: spi-zynqmp-gqspi: Update driver to use spi-mem framework")
Signed-off-by: Quanyang Wang <quanyang.wang@windriver.com>
---
 drivers/spi/spi-zynqmp-gqspi.c | 20 ++++++++++++++------
 1 file changed, 14 insertions(+), 6 deletions(-)

diff --git a/drivers/spi/spi-zynqmp-gqspi.c b/drivers/spi/spi-zynqmp-gqspi.c
index f9056f0a480c..1146359528b9 100644
--- a/drivers/spi/spi-zynqmp-gqspi.c
+++ b/drivers/spi/spi-zynqmp-gqspi.c
@@ -157,6 +157,7 @@ enum mode_type {GQSPI_MODE_IO, GQSPI_MODE_DMA};
  * @data_completion:	completion structure
  */
 struct zynqmp_qspi {
+	struct spi_controller *ctlr;
 	void __iomem *regs;
 	struct clk *refclk;
 	struct clk *pclk;
@@ -827,10 +828,13 @@ static void zynqmp_qspi_read_op(struct zynqmp_qspi *xqspi, u8 rx_nbits,
  */
 static int __maybe_unused zynqmp_qspi_suspend(struct device *dev)
 {
-	struct spi_controller *ctlr = dev_get_drvdata(dev);
-	struct zynqmp_qspi *xqspi = spi_controller_get_devdata(ctlr);
+	struct zynqmp_qspi *xqspi = dev_get_drvdata(dev);
+	struct spi_controller *ctlr = xqspi->ctlr;
+	int ret;
 
-	spi_controller_suspend(ctlr);
+	ret = spi_controller_suspend(ctlr);
+	if (ret)
+		return ret;
 
 	zynqmp_gqspi_write(xqspi, GQSPI_EN_OFST, 0x0);
 
@@ -848,7 +852,10 @@ static int __maybe_unused zynqmp_qspi_suspend(struct device *dev)
  */
 static int __maybe_unused zynqmp_qspi_resume(struct device *dev)
 {
-	struct spi_controller *ctlr = dev_get_drvdata(dev);
+	struct zynqmp_qspi *xqspi = dev_get_drvdata(dev);
+	struct spi_controller *ctlr = xqspi->ctlr;
+
+	zynqmp_gqspi_write(xqspi, GQSPI_EN_OFST, GQSPI_EN_MASK);
 
 	spi_controller_resume(ctlr);
 
@@ -865,7 +872,7 @@ static int __maybe_unused zynqmp_qspi_resume(struct device *dev)
  */
 static int __maybe_unused zynqmp_runtime_suspend(struct device *dev)
 {
-	struct zynqmp_qspi *xqspi = (struct zynqmp_qspi *)dev_get_drvdata(dev);
+	struct zynqmp_qspi *xqspi = dev_get_drvdata(dev);
 
 	clk_disable_unprepare(xqspi->refclk);
 	clk_disable_unprepare(xqspi->pclk);
@@ -883,7 +890,7 @@ static int __maybe_unused zynqmp_runtime_suspend(struct device *dev)
  */
 static int __maybe_unused zynqmp_runtime_resume(struct device *dev)
 {
-	struct zynqmp_qspi *xqspi = (struct zynqmp_qspi *)dev_get_drvdata(dev);
+	struct zynqmp_qspi *xqspi = dev_get_drvdata(dev);
 	int ret;
 
 	ret = clk_prepare_enable(xqspi->pclk);
@@ -1090,6 +1097,7 @@ static int zynqmp_qspi_probe(struct platform_device *pdev)
 
 	xqspi = spi_controller_get_devdata(ctlr);
 	xqspi->dev = dev;
+	xqspi->ctlr = ctlr;
 	platform_set_drvdata(pdev, xqspi);
 
 	xqspi->regs = devm_platform_ioremap_resource(pdev, 0);
-- 
2.25.1

[PATCH 3/5] spi: spi-zynqmp-gqspi: Resolved slab-out-of-bounds bug

[quanyang.wang]

From: Amit Kumar Mahapatra <amit.kumar-mahapatra@xilinx.com>

During a transfer the driver filled the fifo with 4bytes,
even if the data that needs to be transfer is less that 4bytes.
This resulted in slab-out-of-bounds bug in KernelAddressSanitizer.

This patch resolves slab-out-of-bounds bug by filling the fifo
with the number of bytes that needs to transferred.

Signed-off-by: Amit Kumar Mahapatra <amit.kumar-mahapatra@xilinx.com>
---
 drivers/spi/spi-zynqmp-gqspi.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/spi/spi-zynqmp-gqspi.c b/drivers/spi/spi-zynqmp-gqspi.c
index 1146359528b9..2e2607b5dee9 100644
--- a/drivers/spi/spi-zynqmp-gqspi.c
+++ b/drivers/spi/spi-zynqmp-gqspi.c
@@ -509,17 +509,19 @@ static void zynqmp_qspi_filltxfifo(struct zynqmp_qspi *xqspi, int size)
 	u32 count = 0, intermediate;
 
 	while ((xqspi->bytes_to_transfer > 0) && (count < size) && (xqspi->txbuf)) {
-		memcpy(&intermediate, xqspi->txbuf, 4);
-		zynqmp_gqspi_write(xqspi, GQSPI_TXD_OFST, intermediate);
-
 		if (xqspi->bytes_to_transfer >= 4) {
+			memcpy(&intermediate, xqspi->txbuf, 4);
 			xqspi->txbuf += 4;
 			xqspi->bytes_to_transfer -= 4;
+			count += 4;
 		} else {
+			memcpy(&intermediate, xqspi->txbuf,
+			       xqspi->bytes_to_transfer);
 			xqspi->txbuf += xqspi->bytes_to_transfer;
 			xqspi->bytes_to_transfer = 0;
+			count += xqspi->bytes_to_transfer;
 		}
-		count++;
+		zynqmp_gqspi_write(xqspi, GQSPI_TXD_OFST, intermediate);
 	}
 }
 
-- 
2.25.1

[PATCH 4/5] spi: spi-zynqmp-gqspi: fix use-after-free in zynqmp_qspi_exec_op

[quanyang.wang]

From: Quanyang Wang <quanyang.wang@windriver.com>

When handling op->addr, it is using the buffer "tmpbuf" which has been
freed. This will trigger a use-after-free KASAN warning. Let's use
temporary variables to store op->addr.val and op->cmd.opcode to fix
this issue.

Fixes: 1c26372e5aa9 ("spi: spi-zynqmp-gqspi: Update driver to use spi-mem framework")
Signed-off-by: Quanyang Wang <quanyang.wang@windriver.com>
---
 drivers/spi/spi-zynqmp-gqspi.c | 14 ++++----------
 1 file changed, 4 insertions(+), 10 deletions(-)

diff --git a/drivers/spi/spi-zynqmp-gqspi.c b/drivers/spi/spi-zynqmp-gqspi.c
index 2e2607b5dee9..419bc1e6358b 100644
--- a/drivers/spi/spi-zynqmp-gqspi.c
+++ b/drivers/spi/spi-zynqmp-gqspi.c
@@ -928,8 +928,9 @@ static int zynqmp_qspi_exec_op(struct spi_mem *mem,
 	struct zynqmp_qspi *xqspi = spi_controller_get_devdata
 				    (mem->spi->master);
 	int err = 0, i;
-	u8 *tmpbuf;
 	u32 genfifoentry = 0;
+	u16 opcode = op->cmd.opcode;
+	u64 opaddr;
 
 	dev_dbg(xqspi->dev, "cmd:%#x mode:%d.%d.%d.%d\n",
 		op->cmd.opcode, op->cmd.buswidth, op->addr.buswidth,
@@ -942,14 +943,8 @@ static int zynqmp_qspi_exec_op(struct spi_mem *mem,
 	genfifoentry |= xqspi->genfifobus;
 
 	if (op->cmd.opcode) {
-		tmpbuf = kzalloc(op->cmd.nbytes, GFP_KERNEL | GFP_DMA);
-		if (!tmpbuf) {
-			mutex_unlock(&xqspi->op_lock);
-			return -ENOMEM;
-		}
-		tmpbuf[0] = op->cmd.opcode;
 		reinit_completion(&xqspi->data_completion);
-		xqspi->txbuf = tmpbuf;
+		xqspi->txbuf = &opcode;
 		xqspi->rxbuf = NULL;
 		xqspi->bytes_to_transfer = op->cmd.nbytes;
 		xqspi->bytes_to_receive = 0;
@@ -963,13 +958,12 @@ static int zynqmp_qspi_exec_op(struct spi_mem *mem,
 		if (!wait_for_completion_timeout
 		    (&xqspi->data_completion, msecs_to_jiffies(1000))) {
 			err = -ETIMEDOUT;
-			kfree(tmpbuf);
 			goto return_err;
 		}
-		kfree(tmpbuf);
 	}
 
 	if (op->addr.nbytes) {
+		xqspi->txbuf = &opaddr;
 		for (i = 0; i < op->addr.nbytes; i++) {
 			*(((u8 *)xqspi->txbuf) + i) = op->addr.val >>
 					(8 * (op->addr.nbytes - i - 1));
-- 
2.25.1

[PATCH 5/5] spi: spi-zynqmp-gqspi: return -ENOMEM if dma_map_single fails

[quanyang.wang]

From: Quanyang Wang <quanyang.wang@windriver.com>

The spi controller supports 44-bit address space on AXI in DMA mode,
so set dma_addr_t width to 44-bit to avoid using a swiotlb mapping.
In addition, if dma_map_single fails, it should return immediately
instead of continuing doing the DMA operation which bases on invalid
address.

This fixes the following crash which occurs in reading a big block
from flash:

[  123.633577] zynqmp-qspi ff0f0000.spi: swiotlb buffer is full (sz: 4194304 bytes), total 32768 (slots), used 0 (slots)
[  123.644230] zynqmp-qspi ff0f0000.spi: ERR:rxdma:memory not mapped
[  123.784625] Unable to handle kernel paging request at virtual address 00000000003fffc0
[  123.792536] Mem abort info:
[  123.795313]   ESR = 0x96000145
[  123.798351]   EC = 0x25: DABT (current EL), IL = 32 bits
[  123.803655]   SET = 0, FnV = 0
[  123.806693]   EA = 0, S1PTW = 0
[  123.809818] Data abort info:
[  123.812683]   ISV = 0, ISS = 0x00000145
[  123.816503]   CM = 1, WnR = 1
[  123.819455] user pgtable: 4k pages, 48-bit VAs, pgdp=0000000805047000
[  123.825887] [00000000003fffc0] pgd=0000000803b45003, p4d=0000000803b45003, pud=0000000000000000
[  123.834586] Internal error: Oops: 96000145 [#1] PREEMPT SMP

Fixes: 1c26372e5aa9 ("spi: spi-zynqmp-gqspi: Update driver to use spi-mem framework")
Signed-off-by: Quanyang Wang <quanyang.wang@windriver.com>
---
 drivers/spi/spi-zynqmp-gqspi.c | 26 ++++++++++++++++++++------
 1 file changed, 20 insertions(+), 6 deletions(-)

diff --git a/drivers/spi/spi-zynqmp-gqspi.c b/drivers/spi/spi-zynqmp-gqspi.c
index 419bc1e6358b..328b6559bb19 100644
--- a/drivers/spi/spi-zynqmp-gqspi.c
+++ b/drivers/spi/spi-zynqmp-gqspi.c
@@ -733,7 +733,7 @@ static irqreturn_t zynqmp_qspi_irq(int irq, void *dev_id)
  * zynqmp_qspi_setuprxdma - This function sets up the RX DMA operation
  * @xqspi:	xqspi is a pointer to the GQSPI instance.
  */
-static void zynqmp_qspi_setuprxdma(struct zynqmp_qspi *xqspi)
+static int zynqmp_qspi_setuprxdma(struct zynqmp_qspi *xqspi)
 {
 	u32 rx_bytes, rx_rem, config_reg;
 	dma_addr_t addr;
@@ -747,7 +747,7 @@ static void zynqmp_qspi_setuprxdma(struct zynqmp_qspi *xqspi)
 		zynqmp_gqspi_write(xqspi, GQSPI_CONFIG_OFST, config_reg);
 		xqspi->mode = GQSPI_MODE_IO;
 		xqspi->dma_rx_bytes = 0;
-		return;
+		return 0;
 	}
 
 	rx_rem = xqspi->bytes_to_receive % 4;
@@ -755,8 +755,10 @@ static void zynqmp_qspi_setuprxdma(struct zynqmp_qspi *xqspi)
 
 	addr = dma_map_single(xqspi->dev, (void *)xqspi->rxbuf,
 			      rx_bytes, DMA_FROM_DEVICE);
-	if (dma_mapping_error(xqspi->dev, addr))
+	if (dma_mapping_error(xqspi->dev, addr)) {
 		dev_err(xqspi->dev, "ERR:rxdma:memory not mapped\n");
+		return -ENOMEM;
+	}
 
 	xqspi->dma_rx_bytes = rx_bytes;
 	xqspi->dma_addr = addr;
@@ -777,6 +779,8 @@ static void zynqmp_qspi_setuprxdma(struct zynqmp_qspi *xqspi)
 
 	/* Write the number of bytes to transfer */
 	zynqmp_gqspi_write(xqspi, GQSPI_QSPIDMA_DST_SIZE_OFST, rx_bytes);
+
+	return 0;
 }
 
 /**
@@ -813,11 +817,17 @@ static void zynqmp_qspi_write_op(struct zynqmp_qspi *xqspi, u8 tx_nbits,
  * @genfifoentry:	genfifoentry is pointer to the variable in which
  *			GENFIFO	mask is returned to calling function
  */
-static void zynqmp_qspi_read_op(struct zynqmp_qspi *xqspi, u8 rx_nbits,
+static int zynqmp_qspi_read_op(struct zynqmp_qspi *xqspi, u8 rx_nbits,
 				u32 genfifoentry)
 {
-	zynqmp_qspi_setuprxdma(xqspi);
+	int ret;
+
+	ret = zynqmp_qspi_setuprxdma(xqspi);
+	if (ret)
+		return ret;
 	zynqmp_qspi_fillgenfifo(xqspi, rx_nbits, genfifoentry);
+
+	return 0;
 }
 
 /**
@@ -1031,8 +1041,11 @@ static int zynqmp_qspi_exec_op(struct spi_mem *mem,
 			xqspi->rxbuf = (u8 *)op->data.buf.in;
 			xqspi->bytes_to_receive = op->data.nbytes;
 			xqspi->bytes_to_transfer = 0;
-			zynqmp_qspi_read_op(xqspi, op->data.buswidth,
+			err = zynqmp_qspi_read_op(xqspi, op->data.buswidth,
 					    genfifoentry);
+			if (err)
+				goto return_err;
+
 			zynqmp_gqspi_write(xqspi, GQSPI_CONFIG_OFST,
 					   zynqmp_gqspi_read
 					   (xqspi, GQSPI_CONFIG_OFST) |
@@ -1159,6 +1172,7 @@ static int zynqmp_qspi_probe(struct platform_device *pdev)
 		goto clk_dis_all;
 	}
 
+	dma_set_mask(&pdev->dev, DMA_BIT_MASK(44));
 	ctlr->bits_per_word_mask = SPI_BPW_MASK(8);
 	ctlr->num_chipselect = GQSPI_DEFAULT_NUM_CS;
 	ctlr->mem_ops = &zynqmp_qspi_mem_ops;
-- 
2.25.1