[PATCH for-4.14.y 0/5] Stable commits picked up from lede project

[PATCH for-4.14.y 0/5] Stable commits picked up from lede project

[Amit Pundir]

Hi Greg,

For your consideration, stable commits picked up from lede source
tree https://git.lede-project.org/?p=source.git for 4.14.y.

Cherry-picked and build tested on Linux v4.14.55.

Regards,
Amit Pundir


Christian Lamparter (2):
  crypto: crypto4xx - remove bad list_del
  crypto: crypto4xx - fix crypto4xx_build_pdr, crypto4xx_build_sdr leak

Jaehoon Chung (1):
  PCI: exynos: Fix a potential init_clk_resources NULL pointer
    dereference

Jonas Gorski (2):
  bcm63xx_enet: correct clock usage
  bcm63xx_enet: do not write to random DMA channel on BCM6345

 drivers/crypto/amcc/crypto4xx_core.c         | 23 +++++++++----------
 drivers/net/ethernet/broadcom/bcm63xx_enet.c | 34 ++++++++++++++++++++--------
 drivers/pci/dwc/pci-exynos.c                 |  3 ++-
 3 files changed, 38 insertions(+), 22 deletions(-)

-- 
2.7.4

[PATCH for-4.14.y 1/5] PCI: exynos: Fix a potential init_clk_resources NULL pointer dereference

[Amit Pundir]

From: Jaehoon Chung <jh80.chung@samsung.com>

commit b5d6bc90c9129279d363ccbc02ad11e7b657c0b4 upstream.

In order to avoid triggering a NULL pointer dereference in
exynos_pcie_probe() a check must be put in place to detect if
the init_clk_resources hook is initialized before calling it.

Add the respective function pointer check in exynos_pcie_probe().

Signed-off-by: Jaehoon Chung <jh80.chung@samsung.com>
[lorenzo.pieralisi@arm.com: rewrote the commit log]
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Signed-off-by: Amit Pundir <amit.pundir@linaro.org>
---
Cherry-picked from lede tree https://git.lede-project.org/?p=source.git

 drivers/pci/dwc/pci-exynos.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/pci/dwc/pci-exynos.c b/drivers/pci/dwc/pci-exynos.c
index 5596fdedbb94..ea03f1ec12a4 100644
--- a/drivers/pci/dwc/pci-exynos.c
+++ b/drivers/pci/dwc/pci-exynos.c
@@ -695,7 +695,8 @@ static int __init exynos_pcie_probe(struct platform_device *pdev)
 			return ret;
 	}
 
-	if (ep->ops && ep->ops->get_clk_resources) {
+	if (ep->ops && ep->ops->get_clk_resources &&
+			ep->ops->init_clk_resources) {
 		ret = ep->ops->get_clk_resources(ep);
 		if (ret)
 			return ret;
-- 
2.7.4

[PATCH for-4.14.y 2/5] crypto: crypto4xx - remove bad list_del

[Amit Pundir]

From: Christian Lamparter <chunkeey@googlemail.com>

commit a728a196d253530f17da5c86dc7dfbe58c5f7094 upstream.

alg entries are only added to the list, after the registration
was successful. If the registration failed, it was never added
to the list in the first place.

Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Amit Pundir <amit.pundir@linaro.org>
---
Cherry-picked from lede tree https://git.lede-project.org/?p=source.git

 drivers/crypto/amcc/crypto4xx_core.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/crypto/amcc/crypto4xx_core.c b/drivers/crypto/amcc/crypto4xx_core.c
index 65dc78b91dea..3118cec0d81e 100644
--- a/drivers/crypto/amcc/crypto4xx_core.c
+++ b/drivers/crypto/amcc/crypto4xx_core.c
@@ -1033,12 +1033,10 @@ int crypto4xx_register_alg(struct crypto4xx_device *sec_dev,
 			break;
 		}
 
-		if (rc) {
-			list_del(&alg->entry);
+		if (rc)
 			kfree(alg);
-		} else {
+		else
 			list_add_tail(&alg->entry, &sec_dev->alg_list);
-		}
 	}
 
 	return 0;
-- 
2.7.4

[PATCH for-4.14.y 3/5] crypto: crypto4xx - fix crypto4xx_build_pdr, crypto4xx_build_sdr leak

[Amit Pundir]

From: Christian Lamparter <chunkeey@googlemail.com>

commit 5d59ad6eea82ef8df92b4109615a0dde9d8093e9 upstream.

If one of the later memory allocations in rypto4xx_build_pdr()
fails: dev->pdr (and/or) dev->pdr_uinfo wouldn't be freed.

crypto4xx_build_sdr() has the same issue with dev->sdr.

Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Amit Pundir <amit.pundir@linaro.org>
---
Cherry-picked from lede tree https://git.lede-project.org/?p=source.git

 drivers/crypto/amcc/crypto4xx_core.c | 17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)

diff --git a/drivers/crypto/amcc/crypto4xx_core.c b/drivers/crypto/amcc/crypto4xx_core.c
index 3118cec0d81e..3f9eee7e555f 100644
--- a/drivers/crypto/amcc/crypto4xx_core.c
+++ b/drivers/crypto/amcc/crypto4xx_core.c
@@ -207,7 +207,7 @@ static u32 crypto4xx_build_pdr(struct crypto4xx_device *dev)
 				  dev->pdr_pa);
 		return -ENOMEM;
 	}
-	memset(dev->pdr, 0,  sizeof(struct ce_pd) * PPC4XX_NUM_PD);
+	memset(dev->pdr, 0, sizeof(struct ce_pd) * PPC4XX_NUM_PD);
 	dev->shadow_sa_pool = dma_alloc_coherent(dev->core_dev->device,
 				   256 * PPC4XX_NUM_PD,
 				   &dev->shadow_sa_pool_pa,
@@ -240,13 +240,15 @@ static u32 crypto4xx_build_pdr(struct crypto4xx_device *dev)
 
 static void crypto4xx_destroy_pdr(struct crypto4xx_device *dev)
 {
-	if (dev->pdr != NULL)
+	if (dev->pdr)
 		dma_free_coherent(dev->core_dev->device,
 				  sizeof(struct ce_pd) * PPC4XX_NUM_PD,
 				  dev->pdr, dev->pdr_pa);
+
 	if (dev->shadow_sa_pool)
 		dma_free_coherent(dev->core_dev->device, 256 * PPC4XX_NUM_PD,
 				  dev->shadow_sa_pool, dev->shadow_sa_pool_pa);
+
 	if (dev->shadow_sr_pool)
 		dma_free_coherent(dev->core_dev->device,
 			sizeof(struct sa_state_record) * PPC4XX_NUM_PD,
@@ -416,12 +418,12 @@ static u32 crypto4xx_build_sdr(struct crypto4xx_device *dev)
 
 static void crypto4xx_destroy_sdr(struct crypto4xx_device *dev)
 {
-	if (dev->sdr != NULL)
+	if (dev->sdr)
 		dma_free_coherent(dev->core_dev->device,
 				  sizeof(struct ce_sd) * PPC4XX_NUM_SD,
 				  dev->sdr, dev->sdr_pa);
 
-	if (dev->scatter_buffer_va != NULL)
+	if (dev->scatter_buffer_va)
 		dma_free_coherent(dev->core_dev->device,
 				  dev->scatter_buffer_size * PPC4XX_NUM_SD,
 				  dev->scatter_buffer_va,
@@ -1191,7 +1193,7 @@ static int crypto4xx_probe(struct platform_device *ofdev)
 
 	rc = crypto4xx_build_gdr(core_dev->dev);
 	if (rc)
-		goto err_build_gdr;
+		goto err_build_pdr;
 
 	rc = crypto4xx_build_sdr(core_dev->dev);
 	if (rc)
@@ -1234,12 +1236,11 @@ static int crypto4xx_probe(struct platform_device *ofdev)
 err_request_irq:
 	irq_dispose_mapping(core_dev->irq);
 	tasklet_kill(&core_dev->tasklet);
-	crypto4xx_destroy_sdr(core_dev->dev);
 err_build_sdr:
+	crypto4xx_destroy_sdr(core_dev->dev);
 	crypto4xx_destroy_gdr(core_dev->dev);
-err_build_gdr:
-	crypto4xx_destroy_pdr(core_dev->dev);
 err_build_pdr:
+	crypto4xx_destroy_pdr(core_dev->dev);
 	kfree(core_dev->dev);
 err_alloc_dev:
 	kfree(core_dev);
-- 
2.7.4

[PATCH for-4.14.y 4/5] bcm63xx_enet: correct clock usage

[Amit Pundir]

From: Jonas Gorski <jonas.gorski@gmail.com>

commit 9c86b846ce02f7e35d7234cf090b80553eba5389 upstream.

Check the return code of prepare_enable and change one last instance of
enable only to prepare_enable. Also properly disable and release the
clock in error paths and on remove for enetsw.

Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Amit Pundir <amit.pundir@linaro.org>
---
Cherry-picked from lede tree https://git.lede-project.org/?p=source.git

 drivers/net/ethernet/broadcom/bcm63xx_enet.c | 31 +++++++++++++++++++++-------
 1 file changed, 23 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/broadcom/bcm63xx_enet.c b/drivers/net/ethernet/broadcom/bcm63xx_enet.c
index 4f3845a58126..b7d8b6600be6 100644
--- a/drivers/net/ethernet/broadcom/bcm63xx_enet.c
+++ b/drivers/net/ethernet/broadcom/bcm63xx_enet.c
@@ -1773,7 +1773,9 @@ static int bcm_enet_probe(struct platform_device *pdev)
 		ret = PTR_ERR(priv->mac_clk);
 		goto out;
 	}
-	clk_prepare_enable(priv->mac_clk);
+	ret = clk_prepare_enable(priv->mac_clk);
+	if (ret)
+		goto out_put_clk_mac;
 
 	/* initialize default and fetch platform data */
 	priv->rx_ring_size = BCMENET_DEF_RX_DESC;
@@ -1805,9 +1807,11 @@ static int bcm_enet_probe(struct platform_device *pdev)
 		if (IS_ERR(priv->phy_clk)) {
 			ret = PTR_ERR(priv->phy_clk);
 			priv->phy_clk = NULL;
-			goto out_put_clk_mac;
+			goto out_disable_clk_mac;
 		}
-		clk_prepare_enable(priv->phy_clk);
+		ret = clk_prepare_enable(priv->phy_clk);
+		if (ret)
+			goto out_put_clk_phy;
 	}
 
 	/* do minimal hardware init to be able to probe mii bus */
@@ -1901,13 +1905,16 @@ static int bcm_enet_probe(struct platform_device *pdev)
 out_uninit_hw:
 	/* turn off mdc clock */
 	enet_writel(priv, 0, ENET_MIISC_REG);
-	if (priv->phy_clk) {
+	if (priv->phy_clk)
 		clk_disable_unprepare(priv->phy_clk);
+
+out_put_clk_phy:
+	if (priv->phy_clk)
 		clk_put(priv->phy_clk);
-	}
 
-out_put_clk_mac:
+out_disable_clk_mac:
 	clk_disable_unprepare(priv->mac_clk);
+out_put_clk_mac:
 	clk_put(priv->mac_clk);
 out:
 	free_netdev(dev);
@@ -2752,7 +2759,9 @@ static int bcm_enetsw_probe(struct platform_device *pdev)
 		ret = PTR_ERR(priv->mac_clk);
 		goto out_unmap;
 	}
-	clk_enable(priv->mac_clk);
+	ret = clk_prepare_enable(priv->mac_clk);
+	if (ret)
+		goto out_put_clk;
 
 	priv->rx_chan = 0;
 	priv->tx_chan = 1;
@@ -2773,7 +2782,7 @@ static int bcm_enetsw_probe(struct platform_device *pdev)
 
 	ret = register_netdev(dev);
 	if (ret)
-		goto out_put_clk;
+		goto out_disable_clk;
 
 	netif_carrier_off(dev);
 	platform_set_drvdata(pdev, dev);
@@ -2782,6 +2791,9 @@ static int bcm_enetsw_probe(struct platform_device *pdev)
 
 	return 0;
 
+out_disable_clk:
+	clk_disable_unprepare(priv->mac_clk);
+
 out_put_clk:
 	clk_put(priv->mac_clk);
 
@@ -2813,6 +2825,9 @@ static int bcm_enetsw_remove(struct platform_device *pdev)
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	release_mem_region(res->start, resource_size(res));
 
+	clk_disable_unprepare(priv->mac_clk);
+	clk_put(priv->mac_clk);
+
 	free_netdev(dev);
 	return 0;
 }
-- 
2.7.4

[PATCH for-4.14.y 5/5] bcm63xx_enet: do not write to random DMA channel on BCM6345

[Amit Pundir]

From: Jonas Gorski <jonas.gorski@gmail.com>

commit d6213c1f2ad54a964b77471690264ed685718928 upstream.

The DMA controller regs actually point to DMA channel 0, so the write to
ENETDMA_CFG_REG will actually modify a random DMA channel.

Since DMA controller registers do not exist on BCM6345, guard the write
with the usual check for dma_has_sram.

Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Amit Pundir <amit.pundir@linaro.org>
---
Cherry-picked from lede tree https://git.lede-project.org/?p=source.git

 drivers/net/ethernet/broadcom/bcm63xx_enet.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/broadcom/bcm63xx_enet.c b/drivers/net/ethernet/broadcom/bcm63xx_enet.c
index b7d8b6600be6..68470c7c630a 100644
--- a/drivers/net/ethernet/broadcom/bcm63xx_enet.c
+++ b/drivers/net/ethernet/broadcom/bcm63xx_enet.c
@@ -1062,7 +1062,8 @@ static int bcm_enet_open(struct net_device *dev)
 	val = enet_readl(priv, ENET_CTL_REG);
 	val |= ENET_CTL_ENABLE_MASK;
 	enet_writel(priv, val, ENET_CTL_REG);
-	enet_dma_writel(priv, ENETDMA_CFG_EN_MASK, ENETDMA_CFG_REG);
+	if (priv->dma_has_sram)
+		enet_dma_writel(priv, ENETDMA_CFG_EN_MASK, ENETDMA_CFG_REG);
 	enet_dmac_writel(priv, priv->dma_chan_en_mask,
 			 ENETDMAC_CHANCFG, priv->rx_chan);
 
-- 
2.7.4

Re: [PATCH for-4.14.y 0/5] Stable commits picked up from lede project

[Greg KH]

On Tue, Jul 17, 2018 at 01:38:29PM +0530, Amit Pundir wrote:
> Hi Greg,

Welcome back!  :)

> For your consideration, stable commits picked up from lede source
> tree https://git.lede-project.org/?p=source.git for 4.14.y.
> 
> Cherry-picked and build tested on Linux v4.14.55.

Why only 4.14?  These aren't needed further back?  Just checking...

thanks,

greg k-h

Re: [PATCH for-4.14.y 0/5] Stable commits picked up from lede project

[Amit Pundir]

On Tue, 17 Jul 2018 at 13:43, Greg KH <gregkh@linuxfoundation.org> wrote:
>
> On Tue, Jul 17, 2018 at 01:38:29PM +0530, Amit Pundir wrote:
> > Hi Greg,
>
> Welcome back!  :)

Sorry I've been sitting on few of these for a long time. Got stuck in
some other work items.

>
> > For your consideration, stable commits picked up from lede source
> > tree https://git.lede-project.org/?p=source.git for 4.14.y.
> >
> > Cherry-picked and build tested on Linux v4.14.55.
>
> Why only 4.14?  These aren't needed further back?  Just checking...

Sent few for 4.9 as well. Couple of these 4.14 ones are cleanly
applied on 4.9.y as well but they are not present on Lede's 4.9 tree,
so I don't know if they are useful for 4.9.y as such, so I skipped
them over.

Regards,
Amit Pundir

>
> thanks,
>
> greg k-h

Re: [PATCH for-4.14.y 0/5] Stable commits picked up from lede project

[Greg Kroah-Hartman]

On Tue, Jul 17, 2018 at 01:54:49PM +0530, Amit Pundir wrote:
> On Tue, 17 Jul 2018 at 13:43, Greg KH <gregkh@linuxfoundation.org> wrote:
> >
> > On Tue, Jul 17, 2018 at 01:38:29PM +0530, Amit Pundir wrote:
> > > Hi Greg,
> >
> > Welcome back!  :)
> 
> Sorry I've been sitting on few of these for a long time. Got stuck in
> some other work items.
> 
> >
> > > For your consideration, stable commits picked up from lede source
> > > tree https://git.lede-project.org/?p=source.git for 4.14.y.
> > >
> > > Cherry-picked and build tested on Linux v4.14.55.
> >
> > Why only 4.14?  These aren't needed further back?  Just checking...
> 
> Sent few for 4.9 as well. Couple of these 4.14 ones are cleanly
> applied on 4.9.y as well but they are not present on Lede's 4.9 tree,
> so I don't know if they are useful for 4.9.y as such, so I skipped
> them over.

Well, it looks like they are relevant there, as well as for 4.4.y, so
I've applied them all over the place.  Please be more "complete" if at
all possible next time.

thanks,

greg k-h