Skip to contentSkip to conversation

ARC: Fix ICCM & DCCM runtime size checks

PATCH In review Sasha Levin 2 participants 27 messages May 28, 2020 – May 31, 2020

Subsystems (12)
lore
Share & tools
lore

Sasha Levin

author maintainer
#0 lore raw

PATCH 1/26 ARC: Fix ICCM & DCCM runtime size checks

From: Eugeniy Paltsev <[email protected]>

[ Upstream commit 43900edf67d7ef3ac8909854d75b8a1fba2d570c ]

As of today the ICCM and DCCM size checks are incorrectly using mismatched units (KiB checked against bytes). The CONFIG_ARC_DCCM_SZ and CONFIG_ARC_ICCM_SZ are in KiB, but the size calculated in runtime and stored in cpu->dccm.sz and cpu->iccm.sz is in bytes.

Fix that.

---
 arch/arc/kernel/setup.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
arch/arc/kernel/setup.c +3 -2
index 7ee89dc61f6e..23dc002aa574 100644
@@ -12,6 +12,7 @@
 #include <linux/clocksource.h>
 #include <linux/console.h>
 #include <linux/module.h>
+#include <linux/sizes.h>
 #include <linux/cpu.h>
 #include <linux/of_fdt.h>
 #include <linux/of.h>
@@ -409,12 +410,12 @@ static void arc_chk_core_config(void)
 	if ((unsigned int)__arc_dccm_base != cpu->dccm.base_addr)
 		panic("Linux built with incorrect DCCM Base address\n");
 
-	if (CONFIG_ARC_DCCM_SZ != cpu->dccm.sz)
+	if (CONFIG_ARC_DCCM_SZ * SZ_1K != cpu->dccm.sz)
 		panic("Linux built with incorrect DCCM Size\n");
 #endif
 
 #ifdef CONFIG_ARC_HAS_ICCM
-	if (CONFIG_ARC_ICCM_SZ != cpu->iccm.sz)
+	if (CONFIG_ARC_ICCM_SZ * SZ_1K != cpu->iccm.sz)
 		panic("Linux built with incorrect ICCM Size\n");
 #endif
 
Signature · 7 lines
2.25.1


_______________________________________________
linux-snps-arc mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-snps-arc
author maintainer
#1 lore raw

PATCH 2/26 ARC: [plat-eznps]: Restrict to CONFIG_ISA_ARCOMPACT

From: Vineet Gupta <[email protected]>

[ Upstream commit 799587d5731db9dcdafaac4002463aa7d9cd6cf7 ]

Elide invalid configuration EZNPS + ARCv2, triggered by a make allyesconfig build.

Granted the root cause is in source code (asm/barrier.h) where we check for ARCv2 before PLAT_EZNPS, but it is better to avoid such combinations at onset rather then baking subtle nuances into code.

---
 arch/arc/plat-eznps/Kconfig | 1 +
 1 file changed, 1 insertion(+)
arch/arc/plat-eznps/Kconfig +1 -0
index a931d0a256d0..a645bca5899a 100644
@@ -6,6 +6,7 @@
 
 menuconfig ARC_PLAT_EZNPS
 	bool "\"EZchip\" ARC dev platform"
+	depends on ISA_ARCOMPACT
 	select CPU_BIG_ENDIAN
 	select CLKSRC_NPS if !PHYS_ADDR_T_64BIT
 	select EZNPS_GIC
Signature · 7 lines
2.25.1


_______________________________________________
linux-snps-arc mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-snps-arc
author maintainer
#11 lore raw

PATCH 11/26 drm/edid: Add Oculus Rift S to non-desktop list

In reply to Sasha Levin

From: Jan Schmidt <[email protected]>

[ Upstream commit 5a3f610877e9d08968ea7237551049581f02b163 ]

Add a quirk for the Oculus Rift S OVR0012 display so it shows up as a non-desktop display.

---
 drivers/gpu/drm/drm_edid.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/drm_edid.c +2 -1
index ea2849338d6c..9b69e55ad701 100644
@@ -191,10 +191,11 @@ static const struct edid_quirk {
 	{ "HVR", 0xaa01, EDID_QUIRK_NON_DESKTOP },
 	{ "HVR", 0xaa02, EDID_QUIRK_NON_DESKTOP },
 
-	/* Oculus Rift DK1, DK2, and CV1 VR Headsets */
+	/* Oculus Rift DK1, DK2, CV1 and Rift S VR Headsets */
 	{ "OVR", 0x0001, EDID_QUIRK_NON_DESKTOP },
 	{ "OVR", 0x0003, EDID_QUIRK_NON_DESKTOP },
 	{ "OVR", 0x0004, EDID_QUIRK_NON_DESKTOP },
+	{ "OVR", 0x0012, EDID_QUIRK_NON_DESKTOP },
 
 	/* Windows Mixed Reality Headsets */
 	{ "ACR", 0x7fce, EDID_QUIRK_NON_DESKTOP },
Signature · 6 lines
2.25.1

_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel
author maintainer
#12 lore raw

PATCH 12/26 s390/mm: fix set_huge_pte_at() for empty ptes

In reply to Sasha Levin

From: Gerald Schaefer <[email protected]>

[ Upstream commit ac8372f3b4e41015549b331a4f350224661e7fc6 ]

On s390, the layout of normal and large ptes (i.e. pmds/puds) differs. Therefore, set_huge_pte_at() does a conversion from a normal pte to the corresponding large pmd/pud. So, when converting an empty pte, this should result in an empty pmd/pud, which would return true for pmd/pud_none().

However, after conversion we also mark the pmd/pud as large, and therefore present. For empty ptes, this will result in an empty pmd/pud that is also marked as large, and pmd/pud_none() would not return true.

There is currently no issue with this behaviour, as set_huge_pte_at() does not seem to be called for empty ptes. It would be valid though, so let's fix this by not marking empty ptes as large in set_huge_pte_at().

This was found by testing a patch from from Anshuman Khandual, which is currently discussed on LKML ("mm/debug: Add more arch page table helper tests").

---
 arch/s390/mm/hugetlbpage.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
arch/s390/mm/hugetlbpage.c +6 -3
index 5674710a4841..7dfae86afa47 100644
@@ -159,10 +159,13 @@ void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
 		rste &= ~_SEGMENT_ENTRY_NOEXEC;
 
 	/* Set correct table type for 2G hugepages */
-	if ((pte_val(*ptep) & _REGION_ENTRY_TYPE_MASK) == _REGION_ENTRY_TYPE_R3)
-		rste |= _REGION_ENTRY_TYPE_R3 | _REGION3_ENTRY_LARGE;
-	else
+	if ((pte_val(*ptep) & _REGION_ENTRY_TYPE_MASK) == _REGION_ENTRY_TYPE_R3) {
+		if (likely(pte_present(pte)))
+			rste |= _REGION3_ENTRY_LARGE;
+		rste |= _REGION_ENTRY_TYPE_R3;
+	} else if (likely(pte_present(pte)))
 		rste |= _SEGMENT_ENTRY_LARGE;
+
 	clear_huge_pte_skeys(mm, rste);
 	pte_val(*ptep) = rste;
 }
Signature · 1 line
2.25.1
author maintainer
#6 lore raw

PATCH 7/26 net sched: fix reporting the first-time use timestamp

In reply to Sasha Levin

From: Roman Mashak <[email protected]>

[ Upstream commit b15e62631c5f19fea9895f7632dae9c1b27fe0cd ]

When a new action is installed, firstuse field of 'tcf_t' is explicitly set to 0. Value of zero means "new action, not yet used"; as a packet hits the action, 'firstuse' is stamped with the current jiffies value.

tcf_tm_dump() should return 0 for firstuse if action has not yet been hit.

Fixes 48d8ee1694dd ("net sched actions: aggregate dumping of actions timeinfo")
Cc Jamal Hadi Salim <[email protected]>
---
 include/net/act_api.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
include/net/act_api.h +2 -1
index b18c699681ca..59d05feecfb8 100644
@@ -69,7 +69,8 @@ static inline void tcf_tm_dump(struct tcf_t *dtm, const struct tcf_t *stm)
 {
 	dtm->install = jiffies_to_clock_t(jiffies - stm->install);
 	dtm->lastuse = jiffies_to_clock_t(jiffies - stm->lastuse);
-	dtm->firstuse = jiffies_to_clock_t(jiffies - stm->firstuse);
+	dtm->firstuse = stm->firstuse ?
+		jiffies_to_clock_t(jiffies - stm->firstuse) : 0;
 	dtm->expires = jiffies_to_clock_t(stm->expires);
 }
 
Signature · 1 line
2.25.1
author maintainer
#16 lore raw

PATCH 16/26 net: ethernet: stmmac: Enable interface clocks on probe for IPQ806x

In reply to Sasha Levin

From: Jonathan McDowell <[email protected]>

[ Upstream commit a96ac8a0045e3cbe3e5af6d1b3c78c6c2065dec5 ]

The ipq806x_gmac_probe() function enables the PTP clock but not the appropriate interface clocks. This means that if the bootloader hasn't done so attempting to bring up the interface will fail with an error like:

[   59.028131] ipq806x-gmac-dwmac 37600000.ethernet: Failed to reset the dma
[   59.028196] ipq806x-gmac-dwmac 37600000.ethernet eth1: stmmac_hw_setup: DMA engine initialization failed
[   59.034056] ipq806x-gmac-dwmac 37600000.ethernet eth1: stmmac_open: Hw setup failed

This patch, a slightly cleaned up version of one posted by Sergey Sergeev in:

https://forum.openwrt.org/t/support-for-mikrotik-rb3011uias-rm/4064/257

correctly enables the clock; we have already configured the source just before this.

Tested on a MikroTik RB3011.

---
 drivers/net/ethernet/stmicro/stmmac/dwmac-ipq806x.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)
drivers/net/ethernet/stmicro/stmmac/dwmac-ipq806x.c +13 -0
index 0d21082ceb93..4d75158c64b2 100644
@@ -318,6 +318,19 @@ static int ipq806x_gmac_probe(struct platform_device *pdev)
 	/* Enable PTP clock */
 	regmap_read(gmac->nss_common, NSS_COMMON_CLK_GATE, &val);
 	val |= NSS_COMMON_CLK_GATE_PTP_EN(gmac->id);
+	switch (gmac->phy_mode) {
+	case PHY_INTERFACE_MODE_RGMII:
+		val |= NSS_COMMON_CLK_GATE_RGMII_RX_EN(gmac->id) |
+			NSS_COMMON_CLK_GATE_RGMII_TX_EN(gmac->id);
+		break;
+	case PHY_INTERFACE_MODE_SGMII:
+		val |= NSS_COMMON_CLK_GATE_GMII_RX_EN(gmac->id) |
+				NSS_COMMON_CLK_GATE_GMII_TX_EN(gmac->id);
+		break;
+	default:
+		/* We don't get here; the switch above will have errored out */
+		unreachable();
+	}
 	regmap_write(gmac->nss_common, NSS_COMMON_CLK_GATE, val);
 
 	if (gmac->phy_mode == PHY_INTERFACE_MODE_SGMII) {
Signature · 1 line
2.25.1
author maintainer
#21 lore raw

PATCH 21/26 net/mlx5: Fix memory leak in mlx5_events_init

In reply to Sasha Levin

From: Moshe Shemesh <[email protected]>

[ Upstream commit df14ad1eccb04a4a28c90389214dbacab085b244 ]

Fix memory leak in mlx5_events_init(), in case
create_single_thread_workqueue() fails, events
struct should be freed.

Fixes 5d3c537f9070 ("net/mlx5: Handle event of power detection in the PCIE slot")
---
 drivers/net/ethernet/mellanox/mlx5/core/events.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx5/core/events.c +3 -1
index 8bcf3426b9c6..3ce17c3d7a00 100644
@@ -346,8 +346,10 @@ int mlx5_events_init(struct mlx5_core_dev *dev)
 	events->dev = dev;
 	dev->priv.events = events;
 	events->wq = create_singlethread_workqueue("mlx5_events");
-	if (!events->wq)
+	if (!events->wq) {
+		kfree(events);
 		return -ENOMEM;
+	}
 	INIT_WORK(&events->pcie_core_work, mlx5_pcie_event);
 
 	return 0;
Signature · 1 line
2.25.1
author maintainer
#24 lore raw

PATCH 24/26 net: ethernet: ti: cpsw: fix ASSERT_RTNL() warning during suspend

In reply to Sasha Levin

From: Grygorii Strashko <[email protected]>

[ Upstream commit 4c64b83d03f4aafcdf710caad994cbc855802e74 ]

vlan_for_each() are required to be called with rtnl_lock taken, otherwise ASSERT_RTNL() warning will be triggered - which happens now during System resume from suspend:
cpsw_suspend()
|- cpsw_ndo_stop()

    |- __hw_addr_ref_unsync_dev()
      |- cpsw_purge_all_mc()
         |- vlan_for_each()
            |- ASSERT_RTNL();

Hence, fix it by surrounding cpsw_ndo_stop() by rtnl_lock/unlock() calls.

Fixes 15180eca569b ("net: ethernet: ti: cpsw: fix vlan mcast")
---
 drivers/net/ethernet/ti/cpsw.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/net/ethernet/ti/cpsw.c +4 -0
index d7a953c647b4..39df8c8feb6c 100644
@@ -2999,11 +2999,15 @@ static int cpsw_suspend(struct device *dev)
 	struct cpsw_common *cpsw = dev_get_drvdata(dev);
 	int i;
 
+	rtnl_lock();
+
 	for (i = 0; i < cpsw->data.slaves; i++)
 		if (cpsw->slaves[i].ndev)
 			if (netif_running(cpsw->slaves[i].ndev))
 				cpsw_ndo_stop(cpsw->slaves[i].ndev);
 
+	rtnl_unlock();
+
 	/* Select sleep pin state */
 	pinctrl_pm_select_sleep_state(dev);
 
Signature · 1 line
2.25.1
author maintainer
#26 lore raw

PATCH 26/26 net: smsc911x: Fix runtime PM imbalance on error

In reply to Sasha Levin

From: Dinghao Liu <[email protected]>

[ Upstream commit 539d39ad0c61b35f69565a037d7586deaf6d6166 ]

Remove runtime PM usage counter decrement when the increment function has not been called to keep the counter balanced.

---
 drivers/net/ethernet/smsc/smsc911x.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)
drivers/net/ethernet/smsc/smsc911x.c +5 -4
index 38068fc34141..c7bdada4d1b9 100644
@@ -2502,20 +2502,20 @@ static int smsc911x_drv_probe(struct platform_device *pdev)
 
 	retval = smsc911x_init(dev);
 	if (retval < 0)
-		goto out_disable_resources;
+		goto out_init_fail;
 
 	netif_carrier_off(dev);
 
 	retval = smsc911x_mii_init(pdev, dev);
 	if (retval) {
 		SMSC_WARN(pdata, probe, "Error %i initialising mii", retval);
-		goto out_disable_resources;
+		goto out_init_fail;
 	}
 
 	retval = register_netdev(dev);
 	if (retval) {
 		SMSC_WARN(pdata, probe, "Error %i registering device", retval);
-		goto out_disable_resources;
+		goto out_init_fail;
 	} else {
 		SMSC_TRACE(pdata, probe,
 			   "Network interface: \"%s\"", dev->name);
@@ -2556,9 +2556,10 @@ static int smsc911x_drv_probe(struct platform_device *pdev)
 
 	return 0;
 
-out_disable_resources:
+out_init_fail:
 	pm_runtime_put(&pdev->dev);
 	pm_runtime_disable(&pdev->dev);
+out_disable_resources:
 	(void)smsc911x_disable_resources(pdev);
 out_enable_resources_fail:
 	smsc911x_free_resources(pdev);
Signature · 1 line
2.25.1
author maintainer
#8 lore raw

PATCH 8/26 x86/mmiotrace: Use cpumask_available() for cpumask_var_t variables

In reply to Sasha Levin

From: Nathan Chancellor <[email protected]>

[ Upstream commit d7110a26e5905ec2fe3fc88bc6a538901accb72b ]

When building with Clang + -Wtautological-compare and CONFIG_CPUMASK_OFFSTACK unset:

arch/x86/mm/mmio-mod.c:375:6: warning: comparison of array 'downed_cpus' equal to a null pointer is always false [-Wtautological-pointer-compare]

          if (downed_cpus == NULL &&
              ^~~~~~~~~~~    ~~~~
  arch/x86/mm/mmio-mod.c:405:6: warning: comparison of array 'downed_cpus'
  equal to a null pointer is always false [-Wtautological-pointer-compare]
          if (downed_cpus == NULL || cpumask_weight(downed_cpus) == 0)
              ^~~~~~~~~~~    ~~~~

2 warnings generated.

Commit

f7e30f01a9e2 ("cpumask: Add helper cpumask_available()")

added cpumask_available() to fix warnings of this nature. Use that here so that clang does not warn regardless of CONFIG_CPUMASK_OFFSTACK's value.

---
 arch/x86/mm/mmio-mod.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/x86/mm/mmio-mod.c +2 -2
index b8ef8557d4b3..2a36902d418c 100644
@@ -372,7 +372,7 @@ static void enter_uniprocessor(void)
 	int cpu;
 	int err;
 
-	if (downed_cpus == NULL &&
+	if (!cpumask_available(downed_cpus) &&
 	    !alloc_cpumask_var(&downed_cpus, GFP_KERNEL)) {
 		pr_notice("Failed to allocate mask\n");
 		goto out;
@@ -402,7 +402,7 @@ static void leave_uniprocessor(void)
 	int cpu;
 	int err;
 
-	if (downed_cpus == NULL || cpumask_weight(downed_cpus) == 0)
+	if (!cpumask_available(downed_cpus) || cpumask_weight(downed_cpus) == 0)
 		return;
 	pr_notice("Re-enabling CPUs...\n");
 	for_each_cpu(cpu, downed_cpus) {
Signature · 1 line
2.25.1
author maintainer
#15 lore raw

PATCH 15/26 net/ethernet/freescale: rework quiesce/activate for ucc_geth

In reply to Sasha Levin

From: Valentin Longchamp <[email protected]>

[ Upstream commit 79dde73cf9bcf1dd317a2667f78b758e9fe139ed ]

ugeth_quiesce/activate are used to halt the controller when there is a link change that requires to reconfigure the mac.

The previous implementation called netif_device_detach(). This however causes the initial activation of the netdevice to fail precisely because it's detached. For details, see [1].

A possible workaround was the revert of commit
net: linkwatch: add check for netdevice being present to linkwatch_do_dev However, the check introduced in the above commit is correct and shall be kept.

The netif_device_detach() is thus replaced with netif_tx_stop_all_queues() that prevents any tranmission. This allows to perform mac config change required by the link change, without detaching the corresponding netdevice and thus not preventing its initial activation.

[1] https://lists.openwall.net/netdev/2020/01/08/201

---
 drivers/net/ethernet/freescale/ucc_geth.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)
drivers/net/ethernet/freescale/ucc_geth.c +7 -6
index f839fa94ebdd..d3b8ce734c1b 100644
@@ -42,6 +42,7 @@
 #include <soc/fsl/qe/ucc.h>
 #include <soc/fsl/qe/ucc_fast.h>
 #include <asm/machdep.h>
+#include <net/sch_generic.h>
 
 #include "ucc_geth.h"
 
@@ -1548,11 +1549,8 @@ static int ugeth_disable(struct ucc_geth_private *ugeth, enum comm_dir mode)
 
 static void ugeth_quiesce(struct ucc_geth_private *ugeth)
 {
-	/* Prevent any further xmits, plus detach the device. */
-	netif_device_detach(ugeth->ndev);
-
-	/* Wait for any current xmits to finish. */
-	netif_tx_disable(ugeth->ndev);
+	/* Prevent any further xmits */
+	netif_tx_stop_all_queues(ugeth->ndev);
 
 	/* Disable the interrupt to avoid NAPI rescheduling. */
 	disable_irq(ugeth->ug_info->uf_info.irq);
@@ -1565,7 +1563,10 @@ static void ugeth_activate(struct ucc_geth_private *ugeth)
 {
 	napi_enable(&ugeth->napi);
 	enable_irq(ugeth->ug_info->uf_info.irq);
-	netif_device_attach(ugeth->ndev);
+
+	/* allow to xmit again  */
+	netif_tx_wake_all_queues(ugeth->ndev);
+	__netdev_watchdog_up(ugeth->ndev);
 }
 
 /* Called every time the controller might need to be made
Signature · 1 line
2.25.1
author maintainer
#20 lore raw

PATCH 20/26 net/mlx5e: Fix inner tirs handling

In reply to Sasha Levin

From: Roi Dayan <[email protected]>

[ Upstream commit a16b8e0dcf7043bee46174bed0553cc9e36b63a5 ]

In the cited commit inner_tirs argument was added to create and destroy inner tirs, and no indication was added to mlx5e_modify_tirs_hash() function. In order to have a consistent handling, use inner_indir_tir[0].tirn in tirs destroy/modify function as an indication to whether inner tirs are created.
Inner tirs are not created for representors and before this commit, a call to mlx5e_modify_tirs_hash() was sending HW commands to modify non-existent inner tirs.

Fixes 46dc933cee82 ("net/mlx5e: Provide explicit directive if to create inner indirect tirs")
---
 drivers/net/ethernet/mellanox/mlx5/core/en.h         |  2 +-
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c    | 12 +++++++-----
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c     |  4 ++--
 .../net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c    |  4 ++--
 4 files changed, 12 insertions(+), 10 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en.h +1 -1
index 38aa55638bbe..98304c42e495 100644
@@ -1103,7 +1103,7 @@ void mlx5e_close_drop_rq(struct mlx5e_rq *drop_rq);
 int mlx5e_create_indirect_rqt(struct mlx5e_priv *priv);
 
 int mlx5e_create_indirect_tirs(struct mlx5e_priv *priv, bool inner_ttc);
-void mlx5e_destroy_indirect_tirs(struct mlx5e_priv *priv, bool inner_ttc);
+void mlx5e_destroy_indirect_tirs(struct mlx5e_priv *priv);
 
 int mlx5e_create_direct_rqts(struct mlx5e_priv *priv, struct mlx5e_tir *tirs);
 void mlx5e_destroy_direct_rqts(struct mlx5e_priv *priv, struct mlx5e_tir *tirs);
drivers/net/ethernet/mellanox/mlx5/core/en_main.c +7 -5
index 0e340893ca00..c133beb6a7a5 100644
@@ -2758,7 +2758,8 @@ void mlx5e_modify_tirs_hash(struct mlx5e_priv *priv, void *in, int inlen)
 		mlx5_core_modify_tir(mdev, priv->indir_tir[tt].tirn, in, inlen);
 	}
 
-	if (!mlx5e_tunnel_inner_ft_supported(priv->mdev))
+	/* Verify inner tirs resources allocated */
+	if (!priv->inner_indir_tir[0].tirn)
 		return;
 
 	for (tt = 0; tt < MLX5E_NUM_INDIR_TIRS; tt++) {
@@ -3405,14 +3406,15 @@ out:
 	return err;
 }
 
-void mlx5e_destroy_indirect_tirs(struct mlx5e_priv *priv, bool inner_ttc)
+void mlx5e_destroy_indirect_tirs(struct mlx5e_priv *priv)
 {
 	int i;
 
 	for (i = 0; i < MLX5E_NUM_INDIR_TIRS; i++)
 		mlx5e_destroy_tir(priv->mdev, &priv->indir_tir[i]);
 
-	if (!inner_ttc || !mlx5e_tunnel_inner_ft_supported(priv->mdev))
+	/* Verify inner tirs resources allocated */
+	if (!priv->inner_indir_tir[0].tirn)
 		return;
 
 	for (i = 0; i < MLX5E_NUM_INDIR_TIRS; i++)
@@ -5119,7 +5121,7 @@ err_destroy_xsk_rqts:
 err_destroy_direct_tirs:
 	mlx5e_destroy_direct_tirs(priv, priv->direct_tir);
 err_destroy_indirect_tirs:
-	mlx5e_destroy_indirect_tirs(priv, true);
+	mlx5e_destroy_indirect_tirs(priv);
 err_destroy_direct_rqts:
 	mlx5e_destroy_direct_rqts(priv, priv->direct_tir);
 err_destroy_indirect_rqts:
@@ -5138,7 +5140,7 @@ static void mlx5e_cleanup_nic_rx(struct mlx5e_priv *priv)
 	mlx5e_destroy_direct_tirs(priv, priv->xsk_tir);
 	mlx5e_destroy_direct_rqts(priv, priv->xsk_tir);
 	mlx5e_destroy_direct_tirs(priv, priv->direct_tir);
-	mlx5e_destroy_indirect_tirs(priv, true);
+	mlx5e_destroy_indirect_tirs(priv);
 	mlx5e_destroy_direct_rqts(priv, priv->direct_tir);
 	mlx5e_destroy_rqt(priv, &priv->indir_rqt);
 	mlx5e_close_drop_rq(&priv->drop_rq);
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +2 -2
index fddf644ba349..9b232ef36d53 100644
@@ -1597,7 +1597,7 @@ err_destroy_ttc_table:
 err_destroy_direct_tirs:
 	mlx5e_destroy_direct_tirs(priv, priv->direct_tir);
 err_destroy_indirect_tirs:
-	mlx5e_destroy_indirect_tirs(priv, false);
+	mlx5e_destroy_indirect_tirs(priv);
 err_destroy_direct_rqts:
 	mlx5e_destroy_direct_rqts(priv, priv->direct_tir);
 err_destroy_indirect_rqts:
@@ -1614,7 +1614,7 @@ static void mlx5e_cleanup_rep_rx(struct mlx5e_priv *priv)
 	mlx5_del_flow_rules(rpriv->vport_rx_rule);
 	mlx5e_destroy_ttc_table(priv, &priv->fs.ttc);
 	mlx5e_destroy_direct_tirs(priv, priv->direct_tir);
-	mlx5e_destroy_indirect_tirs(priv, false);
+	mlx5e_destroy_indirect_tirs(priv);
 	mlx5e_destroy_direct_rqts(priv, priv->direct_tir);
 	mlx5e_destroy_rqt(priv, &priv->indir_rqt);
 	mlx5e_close_drop_rq(&priv->drop_rq);
drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c +2 -2
index 3ed8ab2d703d..0fed2419623d 100644
@@ -396,7 +396,7 @@ static int mlx5i_init_rx(struct mlx5e_priv *priv)
 err_destroy_direct_tirs:
 	mlx5e_destroy_direct_tirs(priv, priv->direct_tir);
 err_destroy_indirect_tirs:
-	mlx5e_destroy_indirect_tirs(priv, true);
+	mlx5e_destroy_indirect_tirs(priv);
 err_destroy_direct_rqts:
 	mlx5e_destroy_direct_rqts(priv, priv->direct_tir);
 err_destroy_indirect_rqts:
@@ -412,7 +412,7 @@ static void mlx5i_cleanup_rx(struct mlx5e_priv *priv)
 {
 	mlx5i_destroy_flow_steering(priv);
 	mlx5e_destroy_direct_tirs(priv, priv->direct_tir);
-	mlx5e_destroy_indirect_tirs(priv, true);
+	mlx5e_destroy_indirect_tirs(priv);
 	mlx5e_destroy_direct_rqts(priv, priv->direct_tir);
 	mlx5e_destroy_rqt(priv, &priv->indir_rqt);
 	mlx5e_close_drop_rq(&priv->drop_rq);
Signature · 1 line
2.25.1
author maintainer
#25 lore raw

PATCH 25/26 net/mlx4_core: fix a memory leak bug.

In reply to Sasha Levin

From: Qiushi Wu <[email protected]>

[ Upstream commit febfd9d3c7f74063e8e630b15413ca91b567f963 ]

In function mlx4_opreq_action(), pointer "mailbox" is not released, when mlx4_cmd_box() return and error, causing a memory leak bug. Fix this issue by going to "out" label, mlx4_free_cmd_mailbox() can free this pointer.

Fixes fe6f700d6cbb ("net/mlx4_core: Respond to operation request by firmware")
---
 drivers/net/ethernet/mellanox/mlx4/fw.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx4/fw.c +1 -1
index 6e501af0e532..f6ff9620a137 100644
@@ -2734,7 +2734,7 @@ void mlx4_opreq_action(struct work_struct *work)
 		if (err) {
 			mlx4_err(dev, "Failed to retrieve required operation: %d\n",
 				 err);
-			return;
+			goto out;
 		}
 		MLX4_GET(modifier, outbox, GET_OP_REQ_MODIFIER_OFFSET);
 		MLX4_GET(token, outbox, GET_OP_REQ_TOKEN_OFFSET);
Signature · 1 line
2.25.1
author maintainer
#23 lore raw

PATCH 23/26 net/mlx5: Fix error flow in case of function_setup failure

In reply to Sasha Levin

From: Shay Drory <[email protected]>

[ Upstream commit 4f7400d5cbaef676e00cdffb0565bf731c6bb09e ]

Currently, if an error occurred during mlx5_function_setup(), we keep dev->state as DEVICE_STATE_UP.
Fixing it by adding a goto label.

Fixes e161105e58da ("net/mlx5: Function setup/teardown procedures")
---
 drivers/net/ethernet/mellanox/mlx5/core/main.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx5/core/main.c +2 -1
index c96a0e501007..7dcdda9ca351 100644
@@ -1183,7 +1183,7 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, bool boot)
 
 	err = mlx5_function_setup(dev, boot);
 	if (err)
-		goto out;
+		goto err_function;
 
 	if (boot) {
 		err = mlx5_init_once(dev);
@@ -1229,6 +1229,7 @@ err_load:
 		mlx5_cleanup_once(dev);
 function_teardown:
 	mlx5_function_teardown(dev, boot);
+err_function:
 	dev->state = MLX5_DEVICE_STATE_INTERNAL_ERROR;
 	mutex_unlock(&dev->intf_state_mutex);
 
Signature · 1 line
2.25.1
author maintainer
#22 lore raw

PATCH 22/26 net/mlx5e: Update netdev txq on completions during closure

In reply to Sasha Levin

From: Moshe Shemesh <[email protected]>

[ Upstream commit 5e911e2c06bd8c17df29147a5e2d4b17fafda024 ]

On sq closure when we free its descriptors, we should also update netdev txq on completions which would not arrive. Otherwise if we reopen sqs and attach them back, for example on fw fatal recovery flow, we may get tx timeout.

Fixes 29429f3300a3 ("net/mlx5e: Timeout if SQ doesn't flush during close")
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tx.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +6 -3
index dee12f17f9c2..d9e0fc146741 100644
@@ -537,10 +537,9 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget)
 void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq)
 {
 	struct mlx5e_tx_wqe_info *wi;
+	u32 dma_fifo_cc, nbytes = 0;
+	u16 ci, sqcc, npkts = 0;
 	struct sk_buff *skb;
-	u32 dma_fifo_cc;
-	u16 sqcc;
-	u16 ci;
 	int i;
 
 	sqcc = sq->cc;
@@ -565,11 +564,15 @@ void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq)
 		}
 
 		dev_kfree_skb_any(skb);
+		npkts++;
+		nbytes += wi->num_bytes;
 		sqcc += wi->num_wqebbs;
 	}
 
 	sq->dma_fifo_cc = dma_fifo_cc;
 	sq->cc = sqcc;
+
+	netdev_tx_completed_queue(sq->txq, npkts, nbytes);
 }
 
 #ifdef CONFIG_MLX5_CORE_IPOIB
Signature · 1 line
2.25.1
author maintainer
#19 lore raw

PATCH 19/26 net/mlx5: Add command entry handling completion

In reply to Sasha Levin

From: Moshe Shemesh <[email protected]>

[ Upstream commit 17d00e839d3b592da9659c1977d45f85b77f986a ]

When FW response to commands is very slow and all command entries in use are waiting for completion we can have a race where commands can get timeout before they get out of the queue and handled. Timeout completion on uninitialized command will cause releasing command's buffers before accessing it for initialization and then we will get NULL pointer exception while trying access it. It may also cause releasing buffers of another command since we may have timeout completion before even allocating entry index for this command.
Add entry handling completion to avoid this race.

Fixes e126ba97dba9 ("mlx5: Add driver for Mellanox Connect-IB adapters")
---
 drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 14 ++++++++++++++
 include/linux/mlx5/driver.h                   |  1 +
 2 files changed, 15 insertions(+)
drivers/net/ethernet/mellanox/mlx5/core/cmd.c +14 -0
index 08048a2d7259..b6a3370068f1 100644
@@ -861,6 +861,7 @@ static void cmd_work_handler(struct work_struct *work)
 	int alloc_ret;
 	int cmd_mode;
 
+	complete(&ent->handling);
 	sem = ent->page_queue ? &cmd->pages_sem : &cmd->sem;
 	down(sem);
 	if (!ent->page_queue) {
@@ -978,6 +979,11 @@ static int wait_func(struct mlx5_core_dev *dev, struct mlx5_cmd_work_ent *ent)
 	struct mlx5_cmd *cmd = &dev->cmd;
 	int err;
 
+	if (!wait_for_completion_timeout(&ent->handling, timeout) &&
+	    cancel_work_sync(&ent->work)) {
+		ent->ret = -ECANCELED;
+		goto out_err;
+	}
 	if (cmd->mode == CMD_MODE_POLLING || ent->polling) {
 		wait_for_completion(&ent->done);
 	} else if (!wait_for_completion_timeout(&ent->done, timeout)) {
@@ -985,12 +991,17 @@ static int wait_func(struct mlx5_core_dev *dev, struct mlx5_cmd_work_ent *ent)
 		mlx5_cmd_comp_handler(dev, 1UL << ent->idx, true);
 	}
 
+out_err:
 	err = ent->ret;
 
 	if (err == -ETIMEDOUT) {
 		mlx5_core_warn(dev, "%s(0x%x) timeout. Will cause a leak of a command resource\n",
 			       mlx5_command_str(msg_to_opcode(ent->in)),
 			       msg_to_opcode(ent->in));
+	} else if (err == -ECANCELED) {
+		mlx5_core_warn(dev, "%s(0x%x) canceled on out of queue timeout.\n",
+			       mlx5_command_str(msg_to_opcode(ent->in)),
+			       msg_to_opcode(ent->in));
 	}
 	mlx5_core_dbg(dev, "err %d, delivery status %s(%d)\n",
 		      err, deliv_status_to_str(ent->status), ent->status);
@@ -1026,6 +1037,7 @@ static int mlx5_cmd_invoke(struct mlx5_core_dev *dev, struct mlx5_cmd_msg *in,
 	ent->token = token;
 	ent->polling = force_polling;
 
+	init_completion(&ent->handling);
 	if (!callback)
 		init_completion(&ent->done);
 
@@ -1045,6 +1057,8 @@ static int mlx5_cmd_invoke(struct mlx5_core_dev *dev, struct mlx5_cmd_msg *in,
 	err = wait_func(dev, ent);
 	if (err == -ETIMEDOUT)
 		goto out;
+	if (err == -ECANCELED)
+		goto out_free;
 
 	ds = ent->ts2 - ent->ts1;
 	op = MLX5_GET(mbox_in, in->first.data, opcode);
include/linux/mlx5/driver.h +1 -0
index 3e80f03a387f..897829651204 100644
@@ -756,6 +756,7 @@ struct mlx5_cmd_work_ent {
 	struct delayed_work	cb_timeout_work;
 	void		       *context;
 	int			idx;
+	struct completion	handling;
 	struct completion	done;
 	struct mlx5_cmd        *cmd;
 	struct work_struct	work;
Signature · 1 line
2.25.1
author maintainer
#18 lore raw

PATCH 18/26 net: sun: fix missing release regions in cas_init_one().

In reply to Sasha Levin

From: Qiushi Wu <[email protected]>

[ Upstream commit 5a730153984dd13f82ffae93d7170d76eba204e9 ]

In cas_init_one(), "pdev" is requested by "pci_request_regions", but it was not released after a call of the function “pci_write_config_byte” failed. Thus replace the jump target “err_write_cacheline” by "err_out_free_res".

Fixes 1f26dac32057 ("[NET]: Add Sun Cassini driver.")
---
 drivers/net/ethernet/sun/cassini.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/net/ethernet/sun/cassini.c +1 -2
index c91876f8c536..6e78a33aa5e4 100644
@@ -4971,7 +4971,7 @@ static int cas_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 					  cas_cacheline_size)) {
 			dev_err(&pdev->dev, "Could not set PCI cache "
 			       "line size\n");
-			goto err_write_cacheline;
+			goto err_out_free_res;
 		}
 	}
 #endif
@@ -5144,7 +5144,6 @@ err_out_iounmap:
 err_out_free_res:
 	pci_release_regions(pdev);
 
-err_write_cacheline:
 	/* Try to restore it in case the error occurred after we
 	 * set it.
 	 */
Signature · 1 line
2.25.1
author maintainer
#17 lore raw

PATCH 17/26 selftests: mlxsw: qos_mc_aware: Specify arping timeout as an integer

In reply to Sasha Levin

From: Amit Cohen <[email protected]>

[ Upstream commit 46ca11177ed593f39d534f8d2c74ec5344e90c11 ]

Starting from iputils s20190709 (used in Fedora 31), arping does not support timeout being specified as a decimal:

$ arping -c 1 -I swp1 -b 192.0.2.66 -q -w 0.1
arping: invalid argument: '0.1'

Previously, such timeouts were rounded to an integer.

Fix this by specifying the timeout as an integer.

Fixes a5ee171d087e ("selftests: mlxsw: qos_mc_aware: Add a test for UC awareness")
---
 tools/testing/selftests/drivers/net/mlxsw/qos_mc_aware.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/testing/selftests/drivers/net/mlxsw/qos_mc_aware.sh +1 -1
index 24dd8ed48580..b025daea062d 100755
@@ -300,7 +300,7 @@ test_uc_aware()
 	local i
 
 	for ((i = 0; i < attempts; ++i)); do
-		if $ARPING -c 1 -I $h1 -b 192.0.2.66 -q -w 0.1; then
+		if $ARPING -c 1 -I $h1 -b 192.0.2.66 -q -w 1; then
 			((passes++))
 		fi
 
Signature · 1 line
2.25.1
author maintainer
#14 lore raw

PATCH 14/26 net: mvpp2: fix RX hashing for non-10G ports

In reply to Sasha Levin

From: Russell King <[email protected]>

[ Upstream commit 3138a07ce219acde4c0d7ea0b6d54ba64153328b ]

When rxhash is enabled on any ethernet port except the first in each CP block, traffic flow is prevented. The analysis is below:

I've been investigating this afternoon, and what I've found, comparing a kernel without 895586d5dc32 and with 895586d5dc32 applied is:

- The table programmed into the hardware via mvpp22_rss_fill_table() appears to be identical with or without the commit.

- When rxhash is enabled on eth2, mvpp2_rss_port_c2_enable() reports that c2.attr[0] and c2.attr[2] are written back containing:

- with 895586d5dc32, failing: 00200000 40000000
- without 895586d5dc32, working: 04000000 40000000

- When disabling rxhash, c2.attr[0] and c2.attr[2] are written back as:

04000000 00000000

The second value represents the MVPP22_CLS_C2_ATTR2_RSS_EN bit, the first value is the queue number, which comprises two fields. The high 5 bits are 24:29 and the low three are 21:23 inclusive. This comes from:

       c2.attr[0] = MVPP22_CLS_C2_ATTR0_QHIGH(qh) |
                     MVPP22_CLS_C2_ATTR0_QLOW(ql);

So, the working case gives eth2 a queue id of 4.0, or 32 as per port->first_rxq, and the non-working case a queue id of 0.1, or 1. The allocation of queue IDs seems to be in mvpp2_port_probe():

        if (priv->hw_version == MVPP21)
                port->first_rxq = port->id * port->nrxqs;
        else
                port->first_rxq = port->id * priv->max_port_rxqs;

Where:

        if (priv->hw_version == MVPP21)
                priv->max_port_rxqs = 8;
        else
                priv->max_port_rxqs = 32;

Making the port 0 (eth0 / eth1) have port->first_rxq = 0, and port 1 (eth2) be 32. It seems the idea is that the first 32 queues belong to port 0, the second 32 queues belong to port 1, etc.

mvpp2_rss_port_c2_enable() gets the queue number from it's parameter, 'ctx', which comes from mvpp22_rss_ctx(port, 0). This returns port->rss_ctx[0].

mvpp22_rss_context_create() is responsible for allocating that, which it does by looking for an unallocated priv->rss_tables[] pointer. This table is shared amongst all ports on the CP silicon.

When we write the tables in mvpp22_rss_fill_table(), the RSS table entry is defined by:

                u32 sel = MVPP22_RSS_INDEX_TABLE(rss_ctx) |
                          MVPP22_RSS_INDEX_TABLE_ENTRY(i);

where rss_ctx is the context ID (queue number) and i is the index in the table.

If we look at what is written:

- The first table to be written has "sel" values of 00000000..0000001f, containing values 0..3. This appears to be for eth1. This is table 0, RX queue number 0.
- The second table has "sel" values of 00000100..0000011f, and appears to be for eth2. These contain values 0x20..0x23. This is table 1, RX queue number 0.
- The third table has "sel" values of 00000200..0000021f, and appears to be for eth3. These contain values 0x40..0x43. This is table 2, RX queue number 0.

How do queue numbers translate to the RSS table? There is another table - the RXQ2RSS table, indexed by the MVPP22_RSS_INDEX_QUEUE field of MVPP22_RSS_INDEX and accessed through the MVPP22_RXQ2RSS_TABLE register. Before 895586d5dc32, it was:

       mvpp2_write(priv, MVPP22_RSS_INDEX,
                   MVPP22_RSS_INDEX_QUEUE(port->first_rxq));
       mvpp2_write(priv, MVPP22_RXQ2RSS_TABLE,
                   MVPP22_RSS_TABLE_POINTER(port->id));

and after:

       mvpp2_write(priv, MVPP22_RSS_INDEX, MVPP22_RSS_INDEX_QUEUE(ctx));
       mvpp2_write(priv, MVPP22_RXQ2RSS_TABLE, MVPP22_RSS_TABLE_POINTER(ctx));

Before the commit, for eth2, that would've contained '32' for the index and '1' for the table pointer - mapping queue 32 to table 1. Remember that this is queue-high.queue-low of 4.0.

After the commit, we appear to map queue 1 to table 1. That again looks fine on the face of it.

Section 9.3.1 of the A8040 manual seems indicate the reason that the queue number is separated. queue-low seems to always come from the classifier, whereas queue-high can be from the ingress physical port number or the classifier depending on the MVPP2_CLS_SWFWD_PCTRL_REG.

We set the port bit in MVPP2_CLS_SWFWD_PCTRL_REG, meaning that queue-high comes from the MVPP2_CLS_SWFWD_P2HQ_REG() register... and this seems to be where our bug comes from.

mvpp2_cls_oversize_rxq_set() sets this up as:

        mvpp2_write(port->priv, MVPP2_CLS_SWFWD_P2HQ_REG(port->id),
                    (port->first_rxq >> MVPP2_CLS_OVERSIZE_RXQ_LOW_BITS));

        val = mvpp2_read(port->priv, MVPP2_CLS_SWFWD_PCTRL_REG);
        val |= MVPP2_CLS_SWFWD_PCTRL_MASK(port->id);
        mvpp2_write(port->priv, MVPP2_CLS_SWFWD_PCTRL_REG, val);

Setting the MVPP2_CLS_SWFWD_PCTRL_MASK bit means that the queue-high for eth2 is _always_ 4, so only queues 32 through 39 inclusive are available to eth2. Yet, we're trying to tell the classifier to set queue-high, which will be ignored, to zero. Hence, the queue-high field (MVPP22_CLS_C2_ATTR0_QHIGH()) from the classifier will be ignored.

This means we end up directing traffic from eth2 not to queue 1, but to queue 33, and then we tell it to look up queue 33 in the RSS table. However, RSS table has not been programmed for queue 33, and so it ends up (presumably) dropping the packets.

It seems that mvpp22_rss_context_create() doesn't take account of the fact that the upper 5 bits of the queue ID can't actually be changed due to the settings in mvpp2_cls_oversize_rxq_set(), _or_ it seems that mvpp2_cls_oversize_rxq_set() has been missed in this commit. Either way, these two functions mutually disagree with what queue number should be used.

Looking deeper into what mvpp2_cls_oversize_rxq_set() and the MTU validation is doing, it seems that MVPP2_CLS_SWFWD_P2HQ_REG() is used for over-sized packets attempting to egress through this port. With the classifier having had RSS enabled and directing eth2 traffic to queue 1, we may still have packets appearing on queue 32 for this port.

However, the only way we may end up with over-sized packets attempting to egress through eth2 - is if the A8040 forwards frames between its ports. From what I can see, we don't support that feature, and the kernel restricts the egress packet size to the MTU. In any case, if we were to attempt to transmit an oversized packet, we have no support in the kernel to deal with that appearing in the port's receive queue.

So, this patch attempts to solve the issue by clearing the MVPP2_CLS_SWFWD_PCTRL_MASK() bit, allowing MVPP22_CLS_C2_ATTR0_QHIGH() from the classifier to define the queue-high field of the queue number.

My testing seems to confirm my findings above - clearing this bit means that if I enable rxhash on eth2, the interface can then pass traffic, as we are now directing traffic to RX queue 1 rather than queue 33. Traffic still seems to work with rxhash off as well.

Fixes 895586d5dc32 ("net: mvpp2: cls: Use RSS contexts to handle RSS tables")
---
 drivers/net/ethernet/marvell/mvpp2/mvpp2_cls.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/marvell/mvpp2/mvpp2_cls.c +1 -1
index 4344a59c823f..6122057d60c0 100644
@@ -1070,7 +1070,7 @@ void mvpp2_cls_oversize_rxq_set(struct mvpp2_port *port)
 		    (port->first_rxq >> MVPP2_CLS_OVERSIZE_RXQ_LOW_BITS));
 
 	val = mvpp2_read(port->priv, MVPP2_CLS_SWFWD_PCTRL_REG);
-	val |= MVPP2_CLS_SWFWD_PCTRL_MASK(port->id);
+	val &= ~MVPP2_CLS_SWFWD_PCTRL_MASK(port->id);
 	mvpp2_write(port->priv, MVPP2_CLS_SWFWD_PCTRL_REG, val);
 }
 
Signature · 1 line
2.25.1
author maintainer
#13 lore raw

PATCH 13/26 null_blk: return error for invalid zone size

In reply to Sasha Levin

From: Chaitanya Kulkarni <[email protected]>

[ Upstream commit e274832590211c4b1b1e807ca66fad8b5bb8b328 ]

In null_init_zone_dev() check if the zone size is larger than device capacity, return error if needed.

This also fixes the following oops :-

null_blk: changed the number of conventional zones to 4294967295

BUG: kernel NULL pointer dereference, address: 0000000000000010
PGD 7d76c5067 P4D 7d76c5067 PUD 7d240c067 PMD 0
Oops: 0002 [#1] SMP NOPTI
CPU: 4 PID: 5508 Comm: nullbtests.sh Tainted: G OE 5.7.0-rc4lblk-fnext0
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.0-59-gc9ba5276e4
RIP: 0010:null_init_zoned_dev+0x17a/0x27f [null_blk]
RSP: 0018:ffffc90007007e00 EFLAGS: 00010246
RAX: 0000000000000020 RBX: ffff8887fb3f3c00 RCX: 0000000000000007
RDX: 0000000000000000 RSI: ffff8887ca09d688 RDI: ffff888810fea510
RBP: 0000000000000010 R08: ffff8887ca09d688 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000000 R12: ffff8887c26e8000
R13: ffffffffa05e9390 R14: 0000000000000000 R15: 0000000000000001
FS:  00007fcb5256f740(0000) GS:ffff888810e00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000010 CR3: 000000081e8fe000 CR4: 00000000003406e0
Call Trace:
 null_add_dev+0x534/0x71b [null_blk]
 nullb_device_power_store.cold.41+0x8/0x2e [null_blk]
 configfs_write_file+0xe6/0x150
 vfs_write+0xba/0x1e0
 ksys_write+0x5f/0xe0
 do_syscall_64+0x60/0x250
 entry_SYSCALL_64_after_hwframe+0x49/0xb3
RIP: 0033:0x7fcb51c71840
---
 drivers/block/null_blk_zoned.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/block/null_blk_zoned.c +4 -0
index 3d7fdea872f8..2553e05e0725 100644
@@ -20,6 +20,10 @@ int null_zone_init(struct nullb_device *dev)
 		pr_err("zone_size must be power-of-two\n");
 		return -EINVAL;
 	}
+	if (dev->zone_size > dev->size) {
+		pr_err("Zone size larger than device capacity\n");
+		return -EINVAL;
+	}
 
 	dev->zone_size_sects = dev->zone_size << ZONE_SIZE_SHIFT;
 	dev->nr_zones = dev_size >>
Signature · 1 line
2.25.1
author maintainer
#10 lore raw

PATCH 10/26 r8152: support additional Microsoft Surface Ethernet Adapter variant

In reply to Sasha Levin

From: Marc Payne <[email protected]>

[ Upstream commit c27a204383616efba5a4194075e90819961ff66a ]

Device id 0927 is the RTL8153B-based component of the 'Surface USB-C to Ethernet and USB Adapter' and may be used as a component of other devices in future. Tested and working with the r8152 driver.

Update the cdc_ether blacklist due to the RTL8153 'network jam on suspend' issue which this device will cause (personally confirmed).

---
 drivers/net/usb/cdc_ether.c | 11 +++++++++--
 drivers/net/usb/r8152.c     |  1 +
 2 files changed, 10 insertions(+), 2 deletions(-)
drivers/net/usb/cdc_ether.c +9 -2
index fe630438f67b..9bb37ac99a10 100644
@@ -808,14 +808,21 @@ static const struct usb_device_id	products[] = {
 	.driver_info = 0,
 },
 
-/* Microsoft Surface 3 dock (based on Realtek RTL8153) */
+/* Microsoft Surface Ethernet Adapter (based on Realtek RTL8153) */
 {
 	USB_DEVICE_AND_INTERFACE_INFO(MICROSOFT_VENDOR_ID, 0x07c6, USB_CLASS_COMM,
 			USB_CDC_SUBCLASS_ETHERNET, USB_CDC_PROTO_NONE),
 	.driver_info = 0,
 },
 
-	/* TP-LINK UE300 USB 3.0 Ethernet Adapters (based on Realtek RTL8153) */
+/* Microsoft Surface Ethernet Adapter (based on Realtek RTL8153B) */
+{
+	USB_DEVICE_AND_INTERFACE_INFO(MICROSOFT_VENDOR_ID, 0x0927, USB_CLASS_COMM,
+			USB_CDC_SUBCLASS_ETHERNET, USB_CDC_PROTO_NONE),
+	.driver_info = 0,
+},
+
+/* TP-LINK UE300 USB 3.0 Ethernet Adapters (based on Realtek RTL8153) */
 {
 	USB_DEVICE_AND_INTERFACE_INFO(TPLINK_VENDOR_ID, 0x0601, USB_CLASS_COMM,
 			USB_CDC_SUBCLASS_ETHERNET, USB_CDC_PROTO_NONE),
drivers/net/usb/r8152.c +1 -0
index 44ea5dcc43fd..cd1a07175e11 100644
@@ -5837,6 +5837,7 @@ static const struct usb_device_id rtl8152_table[] = {
 	{REALTEK_USB_DEVICE(VENDOR_ID_REALTEK, 0x8153)},
 	{REALTEK_USB_DEVICE(VENDOR_ID_MICROSOFT, 0x07ab)},
 	{REALTEK_USB_DEVICE(VENDOR_ID_MICROSOFT, 0x07c6)},
+	{REALTEK_USB_DEVICE(VENDOR_ID_MICROSOFT, 0x0927)},
 	{REALTEK_USB_DEVICE(VENDOR_ID_SAMSUNG, 0xa101)},
 	{REALTEK_USB_DEVICE(VENDOR_ID_LENOVO,  0x304f)},
 	{REALTEK_USB_DEVICE(VENDOR_ID_LENOVO,  0x3062)},
Signature · 1 line
2.25.1
author maintainer
#9 lore raw

PATCH 9/26 net: bmac: Fix read of MAC address from ROM

In reply to Sasha Levin

From: Jeremy Kerr <[email protected]>

[ Upstream commit ef01cee2ee1b369c57a936166483d40942bcc3e3 ]

In bmac_get_station_address, We're reading two bytes at a time from ROM, but we do that six times, resulting in 12 bytes of read & writes. This means we will write off the end of the six-byte destination buffer.

This change fixes the for-loop to only read/write six bytes.

Based on a proposed fix from Finn Thain <[email protected]>.

---
 drivers/net/ethernet/apple/bmac.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/apple/bmac.c +1 -1
index a58185b1d8bf..3e3711b60d01 100644
@@ -1182,7 +1182,7 @@ bmac_get_station_address(struct net_device *dev, unsigned char *ea)
 	int i;
 	unsigned short data;
 
-	for (i = 0; i < 6; i++)
+	for (i = 0; i < 3; i++)
 		{
 			reset_and_select_srom(dev);
 			data = read_srom(dev, i + EnetAddressOffset/2, SROMAddressBits);
Signature · 1 line
2.25.1
author maintainer
#5 lore raw

PATCH 6/26 io_uring: initialize ctx->sqo_wait earlier

In reply to Sasha Levin

From: Jens Axboe <[email protected]>

[ Upstream commit 583863ed918136412ddf14de2e12534f17cfdc6f ]

Ensure that ctx->sqo_wait is initialized as soon as the ctx is allocated, instead of deferring it to the offload setup. This fixes a syzbot reported lockdep complaint, which is really due to trying to wake_up on an uninitialized wait queue:

RSP: 002b:00007fffb1fb9aa8 EFLAGS: 00000246 ORIG_RAX: 00000000000001a9
RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 0000000000441319
RDX: 0000000000000001 RSI: 0000000020000140 RDI: 000000000000047b
RBP: 0000000000010475 R08: 0000000000000001 R09: 00000000004002c8
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000402260
R13: 00000000004022f0 R14: 0000000000000000 R15: 0000000000000000
INFO: trying to register non-static key.
the code is fine but needs lockdep annotation.
turning off the locking correctness validator.

CPU: 1 PID: 7090 Comm: syz-executor222 Not tainted 5.7.0-rc1-next-20200415-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
 __dump_stack lib/dump_stack.c:77 [inline]
 dump_stack+0x188/0x20d lib/dump_stack.c:118
 assign_lock_key kernel/locking/lockdep.c:913 [inline]
 register_lock_class+0x1664/0x1760 kernel/locking/lockdep.c:1225
 __lock_acquire+0x104/0x4c50 kernel/locking/lockdep.c:4234
 lock_acquire+0x1f2/0x8f0 kernel/locking/lockdep.c:4934
 __raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
 _raw_spin_lock_irqsave+0x8c/0xbf kernel/locking/spinlock.c:159
 __wake_up_common_lock+0xb4/0x130 kernel/sched/wait.c:122
 io_cqring_ev_posted+0xa5/0x1e0 fs/io_uring.c:1160
 io_poll_remove_all fs/io_uring.c:4357 [inline]
 io_ring_ctx_wait_and_kill+0x2bc/0x5a0 fs/io_uring.c:7305
 io_uring_create fs/io_uring.c:7843 [inline]
 io_uring_setup+0x115e/0x22b0 fs/io_uring.c:7870
 do_syscall_64+0xf6/0x7d0 arch/x86/entry/common.c:295
 entry_SYSCALL_64_after_hwframe+0x49/0xb3
RIP: 0033:0x441319
Code: e8 5c ae 02 00 48 83 c4 18 c3 0f 1f 80 00 00 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 0f 83 bb 0a fc ff c3 66 2e 0f 1f 84 00 00 00 00
RSP: 002b:00007fffb1fb9aa8 EFLAGS: 00000246 ORIG_RAX: 00000000000001a9
---
 fs/io_uring.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/io_uring.c +1 -1
index b2ccb908f6b6..2050100e6e84 100644
@@ -409,6 +409,7 @@ static struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p)
 	}
 
 	ctx->flags = p->flags;
+	init_waitqueue_head(&ctx->sqo_wait);
 	init_waitqueue_head(&ctx->cq_wait);
 	init_completion(&ctx->ctx_done);
 	init_completion(&ctx->sqo_thread_started);
@@ -3237,7 +3238,6 @@ static int io_sq_offload_start(struct io_ring_ctx *ctx,
 {
 	int ret;
 
-	init_waitqueue_head(&ctx->sqo_wait);
 	mmgrab(current->mm);
 	ctx->sqo_mm = current->mm;
 
Signature · 1 line
2.25.1
author maintainer
#4 lore raw

PATCH 5/26 i2c: altera: Fix race between xfer_msg and isr thread

In reply to Sasha Levin

From: Atsushi Nemoto <[email protected]>

[ Upstream commit 5d4c7977499a736f3f80826bdc9744344ad55589 ]

Use a mutex to protect access to idev->msg_len, idev->buf, etc. which are modified by both altr_i2c_xfer_msg() and altr_i2c_isr().

This is the minimal fix for easy backporting. A cleanup to remove the spinlock will be added later.

---
 drivers/i2c/busses/i2c-altera.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)
drivers/i2c/busses/i2c-altera.c +9 -1
index 92d2c706c2a7..a60042431370 100644
@@ -70,6 +70,7 @@
  * @isr_mask: cached copy of local ISR enables.
  * @isr_status: cached copy of local ISR status.
  * @lock: spinlock for IRQ synchronization.
+ * @isr_mutex: mutex for IRQ thread.
  */
 struct altr_i2c_dev {
 	void __iomem *base;
@@ -86,6 +87,7 @@ struct altr_i2c_dev {
 	u32 isr_mask;
 	u32 isr_status;
 	spinlock_t lock;	/* IRQ synchronization */
+	struct mutex isr_mutex;
 };
 
 static void
@@ -245,10 +247,11 @@ static irqreturn_t altr_i2c_isr(int irq, void *_dev)
 	struct altr_i2c_dev *idev = _dev;
 	u32 status = idev->isr_status;
 
+	mutex_lock(&idev->isr_mutex);
 	if (!idev->msg) {
 		dev_warn(idev->dev, "unexpected interrupt\n");
 		altr_i2c_int_clear(idev, ALTR_I2C_ALL_IRQ);
-		return IRQ_HANDLED;
+		goto out;
 	}
 	read = (idev->msg->flags & I2C_M_RD) != 0;
 
@@ -301,6 +304,8 @@ static irqreturn_t altr_i2c_isr(int irq, void *_dev)
 		complete(&idev->msg_complete);
 		dev_dbg(idev->dev, "Message Complete\n");
 	}
+out:
+	mutex_unlock(&idev->isr_mutex);
 
 	return IRQ_HANDLED;
 }
@@ -312,6 +317,7 @@ static int altr_i2c_xfer_msg(struct altr_i2c_dev *idev, struct i2c_msg *msg)
 	u32 value;
 	u8 addr = i2c_8bit_addr_from_msg(msg);
 
+	mutex_lock(&idev->isr_mutex);
 	idev->msg = msg;
 	idev->msg_len = msg->len;
 	idev->buf = msg->buf;
@@ -336,6 +342,7 @@ static int altr_i2c_xfer_msg(struct altr_i2c_dev *idev, struct i2c_msg *msg)
 		altr_i2c_int_enable(idev, imask, true);
 		altr_i2c_fill_tx_fifo(idev);
 	}
+	mutex_unlock(&idev->isr_mutex);
 
 	time_left = wait_for_completion_timeout(&idev->msg_complete,
 						ALTR_I2C_XFER_TIMEOUT);
@@ -409,6 +416,7 @@ static int altr_i2c_probe(struct platform_device *pdev)
 	idev->dev = &pdev->dev;
 	init_completion(&idev->msg_complete);
 	spin_lock_init(&idev->lock);
+	mutex_init(&idev->isr_mutex);
 
 	ret = device_property_read_u32(idev->dev, "fifo-size",
 				       &idev->fifo_size);
Signature · 1 line
2.25.1
author maintainer
#2 lore raw

PATCH 3/26 evm: Fix RCU list related warnings

In reply to Sasha Levin

From: Madhuparna Bhowmik <[email protected]>

[ Upstream commit 770f60586d2af0590be263f55fd079226313922c ]

This patch fixes the following warning and few other instances of traversal of evm_config_xattrnames list:

[   32.848432] =============================
[   32.848707] WARNING: suspicious RCU usage
[   32.848966] 5.7.0-rc1-00006-ga8d5875ce5f0b #1 Not tainted
[   32.849308] -----------------------------
[   32.849567] security/integrity/evm/evm_main.c:231 RCU-list traversed in non-reader section!!

Since entries are only added to the list and never deleted, use list_for_each_entry_lockless() instead of list_for_each_entry_rcu for traversing the list. Also, add a relevant comment in evm_secfs.c to indicate this fact.

---
 security/integrity/evm/evm_crypto.c | 2 +-
 security/integrity/evm/evm_main.c   | 4 ++--
 security/integrity/evm/evm_secfs.c  | 9 ++++++++-
 3 files changed, 11 insertions(+), 4 deletions(-)
security/integrity/evm/evm_crypto.c +1 -1
index cc826c2767a3..fbc2ee6d46fc 100644
@@ -209,7 +209,7 @@ static int evm_calc_hmac_or_hash(struct dentry *dentry,
 	data->hdr.length = crypto_shash_digestsize(desc->tfm);
 
 	error = -ENODATA;
-	list_for_each_entry_rcu(xattr, &evm_config_xattrnames, list) {
+	list_for_each_entry_lockless(xattr, &evm_config_xattrnames, list) {
 		bool is_ima = false;
 
 		if (strcmp(xattr->name, XATTR_NAME_IMA) == 0)
security/integrity/evm/evm_main.c +2 -2
index f9a81b187fae..a2c393385db0 100644
@@ -99,7 +99,7 @@ static int evm_find_protected_xattrs(struct dentry *dentry)
 	if (!(inode->i_opflags & IOP_XATTR))
 		return -EOPNOTSUPP;
 
-	list_for_each_entry_rcu(xattr, &evm_config_xattrnames, list) {
+	list_for_each_entry_lockless(xattr, &evm_config_xattrnames, list) {
 		error = __vfs_getxattr(dentry, inode, xattr->name, NULL, 0);
 		if (error < 0) {
 			if (error == -ENODATA)
@@ -230,7 +230,7 @@ static int evm_protected_xattr(const char *req_xattr_name)
 	struct xattr_list *xattr;
 
 	namelen = strlen(req_xattr_name);
-	list_for_each_entry_rcu(xattr, &evm_config_xattrnames, list) {
+	list_for_each_entry_lockless(xattr, &evm_config_xattrnames, list) {
 		if ((strlen(xattr->name) == namelen)
 		    && (strncmp(req_xattr_name, xattr->name, namelen) == 0)) {
 			found = 1;
security/integrity/evm/evm_secfs.c +8 -1
index c11c1f7b3ddd..0f37ef27268d 100644
@@ -234,7 +234,14 @@ static ssize_t evm_write_xattrs(struct file *file, const char __user *buf,
 		goto out;
 	}
 
-	/* Guard against races in evm_read_xattrs */
+	/*
+	 * xattr_list_mutex guards against races in evm_read_xattrs().
+	 * Entries are only added to the evm_config_xattrnames list
+	 * and never deleted. Therefore, the list is traversed
+	 * using list_for_each_entry_lockless() without holding
+	 * the mutex in evm_calc_hmac_or_hash(), evm_find_protected_xattrs()
+	 * and evm_protected_xattr().
+	 */
 	mutex_lock(&xattr_list_mutex);
 	list_for_each_entry(tmp, &evm_config_xattrnames, list) {
 		if (strcmp(xattr->name, tmp->name) == 0) {
Signature · 1 line
2.25.1
author maintainer
#3 lore raw

PATCH 4/26 scsi: pm: Balance pm_only counter of request queue during system resume

In reply to Sasha Levin

From: Can Guo <[email protected]>

[ Upstream commit 05d18ae1cc8a0308b12f37b4ab94afce3535fac9 ]

During system resume, scsi_resume_device() decreases a request queue's pm_only counter if the scsi device was quiesced before. But after that, if the scsi device's RPM status is RPM_SUSPENDED, the pm_only counter is still held (non-zero). Current SCSI resume hook only sets the RPM status of the scsi_device and its request queue to RPM_ACTIVE, but leaves the pm_only counter unchanged. This may make the request queue's pm_only counter remain non-zero after resume hook returns, hence those who are waiting on the mq_freeze_wq would never be woken up. Fix this by calling blk_post_runtime_resume() if a sdev's RPM status was RPM_SUSPENDED.

(struct request_queue)0xFFFFFF815B69E938

	pm_only = (counter = 2),
	rpm_status = 0,
	dev = 0xFFFFFF815B0511A0,

((struct device)0xFFFFFF815B0511A0)).power
is_suspended = FALSE,
runtime_status = RPM_ACTIVE,

(struct scsi_device)0xffffff815b051000

	request_queue = 0xFFFFFF815B69E938,
	sdev_state = SDEV_RUNNING,
	quiesced_by = 0x0,

B::v.f_/task_0xFFFFFF810C246940

-000|__switch_to(prev = 0xFFFFFF810C246940, next = 0xFFFFFF80A49357C0)
-001|context_switch(inline)
-001|__schedule(?)
-002|schedule()
-003|blk_queue_enter(q = 0xFFFFFF815B69E938, flags = 0)
-004|generic_make_request(?)
-005|submit_bio(bio = 0xFFFFFF80A8195B80)
---
 drivers/scsi/scsi_pm.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
drivers/scsi/scsi_pm.c +8 -2
index 3717eea37ecb..5f0ad8b32e3a 100644
@@ -80,6 +80,10 @@ static int scsi_dev_type_resume(struct device *dev,
 	dev_dbg(dev, "scsi resume: %d\n", err);
 
 	if (err == 0) {
+		bool was_runtime_suspended;
+
+		was_runtime_suspended = pm_runtime_suspended(dev);
+
 		pm_runtime_disable(dev);
 		err = pm_runtime_set_active(dev);
 		pm_runtime_enable(dev);
@@ -93,8 +97,10 @@ static int scsi_dev_type_resume(struct device *dev,
 		 */
 		if (!err && scsi_is_sdev_device(dev)) {
 			struct scsi_device *sdev = to_scsi_device(dev);
-
-			blk_set_runtime_active(sdev->request_queue);
+			if (was_runtime_suspended)
+				blk_post_runtime_resume(sdev->request_queue, 0);
+			else
+				blk_set_runtime_active(sdev->request_queue);
 		}
 	}
 
Signature · 1 line
2.25.1
#7 lore raw

Re: [PATCH AUTOSEL 5.4 07/26] net sched: fix reporting the first-time use timestamp

In reply to Sasha Levin

Hi!

Quoted 11 lines

From: Roman Mashak <[email protected]>

[ Upstream commit b15e62631c5f19fea9895f7632dae9c1b27fe0cd ]

When a new action is installed, firstuse field of 'tcf_t' is explicitly set to 0. Value of zero means "new action, not yet used"; as a packet hits the action, 'firstuse' is stamped with the current jiffies value.

tcf_tm_dump() should return 0 for firstuse if action has not yet been hit.

include/net/act_api.h static inline void tcf_tm_dump(struct tcf_t *dtm, const struct tcf_t *stm) · line 75

Quoted 8 lines
include/net/act_api.h +2 -1
@@ -69,7 +69,8 @@ static inline void tcf_tm_dump(struct tcf_t *dtm, const struct tcf_t *stm)
 {
 	dtm->install = jiffies_to_clock_t(jiffies - stm->install);
 	dtm->lastuse = jiffies_to_clock_t(jiffies - stm->lastuse);
-	dtm->firstuse = jiffies_to_clock_t(jiffies - stm->firstuse);
+	dtm->firstuse = stm->firstuse ?
+		jiffies_to_clock_t(jiffies - stm->firstuse) : 0;
 	dtm->expires = jiffies_to_clock_t(stm->expires);
 }

Jiffies start at small negative value (and are expected to overflow periodically). This code will not work correctly in that case.

Best regards,
Pavel

Signature · 2 lines
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html