Skip to contentSkip to conversation

staging: wfx: fix potential deadlock in wfx_tx_flush()

PATCH In review Sasha Levin 7 participants 410 messages Jun 18, 2020 – Jun 23, 2020 Page 3 of 3

Subsystems (46)
lore
Share & tools
lore
author maintainer
#318 lore raw

PATCH 298/388 PCI: Avoid FLR for AMD Starship USB 3.0

In reply to Sasha Levin

From: Kevin Buettner <[email protected]>

[ Upstream commit 5727043c73fdfe04597971b5f3f4850d879c1f4f ]

The AMD Starship USB 3.0 host controller advertises Function Level Reset support, but it apparently doesn't work. Add a quirk to prevent use of FLR on this device.

Without this quirk, when attempting to assign (pass through) an AMD Starship USB 3.0 host controller to a guest OS, the system becomes increasingly unresponsive over the course of several minutes, eventually requiring a hard reset. Shortly after attempting to start the guest, I see these messages:

vfio-pci 0000:05:00.3: not ready 1023ms after FLR; waiting vfio-pci 0000:05:00.3: not ready 2047ms after FLR; waiting vfio-pci 0000:05:00.3: not ready 4095ms after FLR; waiting vfio-pci 0000:05:00.3: not ready 8191ms after FLR; waiting

And then eventually:

vfio-pci 0000:05:00.3: not ready 65535ms after FLR; giving up
INFO: NMI handler (perf_event_nmi_handler) took too long to run: 0.000 msecs
perf: interrupt took too long (642744 > 2500), lowering kernel.perf_event_max_sample_rate to 1000
INFO: NMI handler (perf_event_nmi_handler) took too long to run: 82.270 msecs
INFO: NMI handler (perf_event_nmi_handler) took too long to run: 680.608 msecs
INFO: NMI handler (perf_event_nmi_handler) took too long to run: 100.952 msecs
...

  watchdog: BUG: soft lockup - CPU#3 stuck for 22s! [qemu-system-x86:7487]

Tested on a Micro-Star International Co., Ltd. MS-7C59/Creator TRX40 motherboard with an AMD Ryzen Threadripper 3970X.

---
 drivers/pci/quirks.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/pci/quirks.c +2 -0
index 43a0c2ce635e..b1db58d00d2b 100644
@@ -5133,6 +5133,7 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, 0x443, quirk_intel_qat_vf_cap);
  * FLR may cause the following to devices to hang:
  *
  * AMD Starship/Matisse HD Audio Controller 0x1487
+ * AMD Starship USB 3.0 Host Controller 0x148c
  * AMD Matisse USB 3.0 Host Controller 0x149c
  * Intel 82579LM Gigabit Ethernet Controller 0x1502
  * Intel 82579V Gigabit Ethernet Controller 0x1503
@@ -5143,6 +5144,7 @@ static void quirk_no_flr(struct pci_dev *dev)
 	dev->dev_flags |= PCI_DEV_FLAGS_NO_FLR_RESET;
 }
 DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_AMD, 0x1487, quirk_no_flr);
+DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_AMD, 0x148c, quirk_no_flr);
 DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_AMD, 0x149c, quirk_no_flr);
 DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, 0x1502, quirk_no_flr);
 DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, 0x1503, quirk_no_flr);
Signature · 1 line
2.25.1
author maintainer
#319 lore raw

PATCH 299/388 of: Fix a refcounting bug in __of_attach_node_sysfs()

In reply to Sasha Levin

From: Dan Carpenter <[email protected]>

[ Upstream commit 8a325dd06f2358ea0888e4ff1c9ca4bc23bd53f3 ]

The problem in this code is that if kobject_add() fails, then it should call of_node_put(np) to drop the reference count. I've actually moved the of_node_get(np) later in the function to avoid needing to do clean up.

Fixes 5b2c2f5a0ea3 ("of: overlay: add missing of_node_get() in __of_attach_node_sysfs")
---
 drivers/of/kobj.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/of/kobj.c +1 -2
index c72eef988041..a32e60b024b8 100644
@@ -134,8 +134,6 @@ int __of_attach_node_sysfs(struct device_node *np)
 	if (!name)
 		return -ENOMEM;
 
-	of_node_get(np);
-
 	rc = kobject_add(&np->kobj, parent, "%s", name);
 	kfree(name);
 	if (rc)
@@ -144,6 +142,7 @@ int __of_attach_node_sysfs(struct device_node *np)
 	for_each_property_of_node(np, pp)
 		__of_add_property_sysfs(np, pp);
 
+	of_node_get(np);
 	return 0;
 }
 
Signature · 1 line
2.25.1
author maintainer
#320 lore raw

PATCH 300/388 ARM: davinci: fix build failure without I2C

In reply to Sasha Levin

From: Arnd Bergmann <[email protected]>

[ Upstream commit 147922f91965e734664374a3f226adfeaa586d72 ]

The two supplies are referenced outside of #ifdef CONFIG_I2C but defined inside, which breaks the build if that is not built-in:

mach-davinci/board-dm644x-evm.c:861:21: error: use of undeclared identifier 'fixed_supplies_1_8v'

                                     ARRAY_SIZE(fixed_supplies_1_8v), 1800000);
                                                ^

mach-davinci/board-dm644x-evm.c:861:21: error: use of undeclared identifier 'fixed_supplies_1_8v' mach-davinci/board-dm644x-evm.c:861:21: error: use of undeclared identifier 'fixed_supplies_1_8v' mach-davinci/board-dm644x-evm.c:860:49: error: use of undeclared identifier 'fixed_supplies_1_8v'
regulator_register_always_on(0, "fixed-dummy", fixed_supplies_1_8v,

I don't know if the regulators are used anywhere without I2C, but always registering them seems to be the safe choice here.

On a related note, it might be best to also deal with CONFIG_I2C=m across the file, unless this is going to be moved to DT and removed really soon anyway.

Fixes 5e06d19694a4 ("ARM: davinci: dm644x-evm: Add Fixed regulators needed for tlv320aic33")
---
 arch/arm/mach-davinci/board-dm644x-evm.c | 26 ++++++++++++------------
 1 file changed, 13 insertions(+), 13 deletions(-)
arch/arm/mach-davinci/board-dm644x-evm.c +13 -13
index 3461d12bbfc0..a5d3708fedf6 100644
@@ -655,19 +655,6 @@ static struct i2c_board_info __initdata i2c_info[] =  {
 	},
 };
 
-/* Fixed regulator support */
-static struct regulator_consumer_supply fixed_supplies_3_3v[] = {
-	/* Baseboard 3.3V: 5V -> TPS54310PWP -> 3.3V */
-	REGULATOR_SUPPLY("AVDD", "1-001b"),
-	REGULATOR_SUPPLY("DRVDD", "1-001b"),
-};
-
-static struct regulator_consumer_supply fixed_supplies_1_8v[] = {
-	/* Baseboard 1.8V: 5V -> TPS54310PWP -> 1.8V */
-	REGULATOR_SUPPLY("IOVDD", "1-001b"),
-	REGULATOR_SUPPLY("DVDD", "1-001b"),
-};
-
 #define DM644X_I2C_SDA_PIN	GPIO_TO_PIN(2, 12)
 #define DM644X_I2C_SCL_PIN	GPIO_TO_PIN(2, 11)
 
@@ -700,6 +687,19 @@ static void __init evm_init_i2c(void)
 }
 #endif
 
+/* Fixed regulator support */
+static struct regulator_consumer_supply fixed_supplies_3_3v[] = {
+	/* Baseboard 3.3V: 5V -> TPS54310PWP -> 3.3V */
+	REGULATOR_SUPPLY("AVDD", "1-001b"),
+	REGULATOR_SUPPLY("DRVDD", "1-001b"),
+};
+
+static struct regulator_consumer_supply fixed_supplies_1_8v[] = {
+	/* Baseboard 1.8V: 5V -> TPS54310PWP -> 1.8V */
+	REGULATOR_SUPPLY("IOVDD", "1-001b"),
+	REGULATOR_SUPPLY("DVDD", "1-001b"),
+};
+
 #define VENC_STD_ALL	(V4L2_STD_NTSC | V4L2_STD_PAL)
 
 /* venc standard timings */
Signature · 7 lines
2.25.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#321 lore raw

PATCH 301/388 input: i8042 - Remove special PowerPC handling

In reply to Sasha Levin

From: Nathan Chancellor <[email protected]>

[ Upstream commit e4f4ffa8a98c24a4ab482669b1e2b4cfce3f52f4 ]

This causes a build error with CONFIG_WALNUT because kb_cs and kb_data were removed in commit 917f0af9e5a9 ("powerpc: Remove arch/ppc and include/asm-ppc").

ld.lld: error: undefined symbol: kb_cs

Quoted 6 lines

referenced by i8042-ppcio.h:28 (drivers/input/serio/i8042-ppcio.h:28) input/serio/i8042.o:(__i8042_command) in archive drivers/built-in.a referenced by i8042-ppcio.h:28 (drivers/input/serio/i8042-ppcio.h:28) input/serio/i8042.o:(__i8042_command) in archive drivers/built-in.a referenced by i8042-ppcio.h:28 (drivers/input/serio/i8042-ppcio.h:28) input/serio/i8042.o:(__i8042_command) in archive drivers/built-in.a

ld.lld: error: undefined symbol: kb_data

Quoted 7 lines

referenced by i8042.c:309 (drivers/input/serio/i8042.c:309) input/serio/i8042.o:(__i8042_command) in archive drivers/built-in.a referenced by i8042-ppcio.h:33 (drivers/input/serio/i8042-ppcio.h:33) input/serio/i8042.o:(__i8042_command) in archive drivers/built-in.a referenced by i8042.c:319 (drivers/input/serio/i8042.c:319) input/serio/i8042.o:(__i8042_command) in archive drivers/built-in.a referenced 15 more times

Presumably since nobody has noticed this for the last 12 years, there is not anyone actually trying to use this driver so we can just remove this special walnut code and use the generic header so it builds for all configurations.

---
 drivers/input/serio/i8042-ppcio.h | 57 -------------------------------
 drivers/input/serio/i8042.h       |  2 --
 2 files changed, 59 deletions(-)
 delete mode 100644 drivers/input/serio/i8042-ppcio.h
drivers/input/serio/i8042-ppcio.h deleted +0 -57
deleted file mode 100644
index 391f94d9e47d..000000000000
@@ -1,57 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0-only */
-#ifndef _I8042_PPCIO_H
-#define _I8042_PPCIO_H
-
-
-#if defined(CONFIG_WALNUT)
-
-#define I8042_KBD_IRQ 25
-#define I8042_AUX_IRQ 26
-
-#define I8042_KBD_PHYS_DESC "walnutps2/serio0"
-#define I8042_AUX_PHYS_DESC "walnutps2/serio1"
-#define I8042_MUX_PHYS_DESC "walnutps2/serio%d"
-
-extern void *kb_cs;
-extern void *kb_data;
-
-#define I8042_COMMAND_REG (*(int *)kb_cs)
-#define I8042_DATA_REG (*(int *)kb_data)
-
-static inline int i8042_read_data(void)
-{
-	return readb(kb_data);
-}
-
-static inline int i8042_read_status(void)
-{
-	return readb(kb_cs);
-}
-
-static inline void i8042_write_data(int val)
-{
-	writeb(val, kb_data);
-}
-
-static inline void i8042_write_command(int val)
-{
-	writeb(val, kb_cs);
-}
-
-static inline int i8042_platform_init(void)
-{
-	i8042_reset = I8042_RESET_ALWAYS;
-	return 0;
-}
-
-static inline void i8042_platform_exit(void)
-{
-}
-
-#else
-
-#include "i8042-io.h"
-
-#endif
-
-#endif /* _I8042_PPCIO_H */
drivers/input/serio/i8042.h +0 -2
index 38dc27ad3c18..eb376700dfff 100644
@@ -17,8 +17,6 @@
 #include "i8042-ip22io.h"
 #elif defined(CONFIG_SNI_RM)
 #include "i8042-snirm.h"
-#elif defined(CONFIG_PPC)
-#include "i8042-ppcio.h"
 #elif defined(CONFIG_SPARC)
 #include "i8042-sparcio.h"
 #elif defined(CONFIG_X86) || defined(CONFIG_IA64)
Signature · 1 line
2.25.1
author maintainer
#322 lore raw

PATCH 302/388 powerpc/4xx: Don't unmap NULL mbase

In reply to Sasha Levin

From: huhai <[email protected]>

[ Upstream commit bcec081ecc940fc38730b29c743bbee661164161 ]

---
 arch/powerpc/platforms/4xx/pci.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)
arch/powerpc/platforms/4xx/pci.c +1 -3
index e6e2adcc7b64..c13d64c3b019 100644
@@ -1242,7 +1242,7 @@ static void __init ppc460sx_pciex_check_link(struct ppc4xx_pciex_port *port)
 	if (mbase == NULL) {
 		printk(KERN_ERR "%pOF: Can't map internal config space !",
 			port->node);
-		goto done;
+		return;
 	}
 
 	while (attempt && (0 == (in_le32(mbase + PECFG_460SX_DLLSTA)
@@ -1252,9 +1252,7 @@ static void __init ppc460sx_pciex_check_link(struct ppc4xx_pciex_port *port)
 	}
 	if (attempt)
 		port->link = 1;
-done:
 	iounmap(mbase);
-
 }
 
 static struct ppc4xx_pciex_hwops ppc460sx_pcie_hwops __initdata = {
Signature · 1 line
2.25.1
author maintainer
#323 lore raw

PATCH 303/388 powerpc/64s/kuap: Add missing isync to KUAP restore paths

In reply to Sasha Levin

From: Nicholas Piggin <[email protected]>

[ Upstream commit cb2b53cbffe3c388cd676b63f34e54ceb2643ae2 ]

Writing the AMR register is documented to require context synchronizing operations before and after, for it to take effect as expected. The KUAP restore at interrupt exit time deliberately avoids the isync after the AMR update because it only needs to take effect after the context synchronizing RFID that soon follows. Add a comment for this.

The missing isync before the update doesn't have an obvious justification, and seems it could theoretically allow a rogue user access to leak past the AMR update. Add isyncs for these.

---
 arch/powerpc/include/asm/book3s/64/kup-radix.h | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)
arch/powerpc/include/asm/book3s/64/kup-radix.h +10 -1
index 3bcef989a35d..101d60f16d46 100644
@@ -16,7 +16,9 @@
 #ifdef CONFIG_PPC_KUAP
 	BEGIN_MMU_FTR_SECTION_NESTED(67)
 	ld	\gpr, STACK_REGS_KUAP(r1)
+	isync
 	mtspr	SPRN_AMR, \gpr
+	/* No isync required, see kuap_restore_amr() */
 	END_MMU_FTR_SECTION_NESTED_IFSET(MMU_FTR_RADIX_KUAP, 67)
 #endif
 .endm
@@ -62,8 +64,15 @@
 
 static inline void kuap_restore_amr(struct pt_regs *regs)
 {
-	if (mmu_has_feature(MMU_FTR_RADIX_KUAP))
+	if (mmu_has_feature(MMU_FTR_RADIX_KUAP)) {
+		isync();
 		mtspr(SPRN_AMR, regs->kuap);
+		/*
+		 * No isync required here because we are about to RFI back to
+		 * previous context before any user accesses would be made,
+		 * which is a CSI.
+		 */
+	}
 }
 
 static inline void kuap_check_amr(void)
Signature · 1 line
2.25.1
author maintainer
#324 lore raw

PATCH 304/388 extcon: adc-jack: Fix an error handling path in 'adc_jack_probe()'

In reply to Sasha Levin

From: Christophe JAILLET <[email protected]>

[ Upstream commit bc84cff2c92ae5ccb2c37da73756e7174b1b430f ]

In some error handling paths, a call to 'iio_channel_get()' is not balanced by a corresponding call to 'iio_channel_release()'.

This can be achieved easily by using the devm_ variant of 'iio_channel_get()'.

This has the extra benefit to simplify the remove function.

Fixes 19939860dcae ("extcon: adc_jack: adc-jack driver to support 3.5 pi or simliar devices")
---
 drivers/extcon/extcon-adc-jack.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/extcon/extcon-adc-jack.c +1 -2
index ad02dc6747a4..0317b614b680 100644
@@ -124,7 +124,7 @@ static int adc_jack_probe(struct platform_device *pdev)
 	for (i = 0; data->adc_conditions[i].id != EXTCON_NONE; i++);
 	data->num_conditions = i;
 
-	data->chan = iio_channel_get(&pdev->dev, pdata->consumer_channel);
+	data->chan = devm_iio_channel_get(&pdev->dev, pdata->consumer_channel);
 	if (IS_ERR(data->chan))
 		return PTR_ERR(data->chan);
 
@@ -164,7 +164,6 @@ static int adc_jack_remove(struct platform_device *pdev)
 
 	free_irq(data->irq, data);
 	cancel_work_sync(&data->handler.work);
-	iio_channel_release(data->chan);
 
 	return 0;
 }
Signature · 1 line
2.25.1
author maintainer
#325 lore raw

PATCH 305/388 usb: dwc3: meson-g12a: fix error path when fetching the reset line fails

In reply to Sasha Levin

From: Martin Blumenstingl <[email protected]>

[ Upstream commit be8c1001a7e681e8813882a42ed51c8dbffd8800 ]

Disable and unprepare the clocks when devm_reset_control_get_shared() fails. This fixes the error path as this must disable the clocks which were previously enabled.

---
 drivers/usb/dwc3/dwc3-meson-g12a.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/usb/dwc3/dwc3-meson-g12a.c +1 -1
index 2d257bdfe848..eabb3bb6fcaa 100644
@@ -505,7 +505,7 @@ static int dwc3_meson_g12a_probe(struct platform_device *pdev)
 	if (IS_ERR(priv->reset)) {
 		ret = PTR_ERR(priv->reset);
 		dev_err(dev, "failed to get device reset, err=%d\n", ret);
-		return ret;
+		goto err_disable_clks;
 	}
 
 	ret = reset_control_reset(priv->reset);
Signature · 7 lines
2.25.1


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

PATCH 306/388 ASoC: fsl_asrc_dma: Fix dma_chan leak when config DMA channel failed

In reply to Sasha Levin

From: Xiyu Yang <[email protected]>

[ Upstream commit 36124fb19f1ae68a500cd76a76d40c6e81bee346 ]

fsl_asrc_dma_hw_params() invokes dma_request_channel() or fsl_asrc_get_dma_channel(), which returns a reference of the specified dma_chan object to "pair->dma_chan[dir]" with increased refcnt.

The reference counting issue happens in one exception handling path of fsl_asrc_dma_hw_params(). When config DMA channel failed for Back-End, the function forgets to decrease the refcnt increased by dma_request_channel() or fsl_asrc_get_dma_channel(), causing a refcnt leak.

Fix this issue by calling dma_release_channel() when config DMA channel failed.

---
 sound/soc/fsl/fsl_asrc_dma.c | 1 +
 1 file changed, 1 insertion(+)
sound/soc/fsl/fsl_asrc_dma.c +1 -0
index e7178817d7a7..1ee10eafe3e6 100644
@@ -252,6 +252,7 @@ static int fsl_asrc_dma_hw_params(struct snd_soc_component *component,
 	ret = dmaengine_slave_config(pair->dma_chan[dir], &config_be);
 	if (ret) {
 		dev_err(dev, "failed to config DMA channel for Back-End\n");
+		dma_release_channel(pair->dma_chan[dir]);
 		return ret;
 	}
 
Signature · 1 line
2.25.1
author maintainer
#327 lore raw

PATCH 307/388 ASoC: SOF: Intel: hda: fix generic hda codec support

In reply to Sasha Levin

From: Kai Vehmanen <[email protected]>

[ Upstream commit 89d73ccab20a684d8446cea4d8ac6a2608c8d390 ]

Add support for using generic codec driver with SOF. Generic driver is used if:
- snd_sof_intel_hda_common.hda_model="generic" is set, or
- fallback if no other codec driver is found

The implementation is aligned with snd-hda-intel driver, and fixes audio support for systems like Acer Swift 3 SF314-57G, on which this issue was originally reported.

---
 sound/soc/sof/intel/hda-codec.c | 51 +++++++++++++++++++++++++++++----
 1 file changed, 45 insertions(+), 6 deletions(-)
sound/soc/sof/intel/hda-codec.c +45 -6
index 3041fbbb010a..ea021db697b8 100644
@@ -24,19 +24,44 @@
 #define IDISP_VID_INTEL	0x80860000
 
 /* load the legacy HDA codec driver */
-static int hda_codec_load_module(struct hda_codec *codec)
+static int request_codec_module(struct hda_codec *codec)
 {
 #ifdef MODULE
 	char alias[MODULE_NAME_LEN];
-	const char *module = alias;
+	const char *mod = NULL;
 
-	snd_hdac_codec_modalias(&codec->core, alias, sizeof(alias));
-	dev_dbg(&codec->core.dev, "loading codec module: %s\n", module);
-	request_module(module);
+	switch (codec->probe_id) {
+	case HDA_CODEC_ID_GENERIC:
+#if IS_MODULE(CONFIG_SND_HDA_GENERIC)
+		mod = "snd-hda-codec-generic";
 #endif
+		break;
+	default:
+		snd_hdac_codec_modalias(&codec->core, alias, sizeof(alias));
+		mod = alias;
+		break;
+	}
+
+	if (mod) {
+		dev_dbg(&codec->core.dev, "loading codec module: %s\n", mod);
+		request_module(mod);
+	}
+#endif /* MODULE */
 	return device_attach(hda_codec_dev(codec));
 }
 
+static int hda_codec_load_module(struct hda_codec *codec)
+{
+	int ret = request_codec_module(codec);
+
+	if (ret <= 0) {
+		codec->probe_id = HDA_CODEC_ID_GENERIC;
+		ret = request_codec_module(codec);
+	}
+
+	return ret;
+}
+
 /* enable controller wake up event for all codecs with jack connectors */
 void hda_codec_jack_wake_enable(struct snd_sof_dev *sdev)
 {
@@ -78,6 +103,13 @@ void hda_codec_jack_check(struct snd_sof_dev *sdev) {}
 EXPORT_SYMBOL_NS(hda_codec_jack_wake_enable, SND_SOC_SOF_HDA_AUDIO_CODEC);
 EXPORT_SYMBOL_NS(hda_codec_jack_check, SND_SOC_SOF_HDA_AUDIO_CODEC);
 
+#if IS_ENABLED(CONFIG_SND_HDA_GENERIC)
+#define is_generic_config(bus) \
+	((bus)->modelname && !strcmp((bus)->modelname, "generic"))
+#else
+#define is_generic_config(x)	0
+#endif
+
 /* probe individual codec */
 static int hda_codec_probe(struct snd_sof_dev *sdev, int address,
 			   bool hda_codec_use_common_hdmi)
@@ -87,6 +119,7 @@ static int hda_codec_probe(struct snd_sof_dev *sdev, int address,
 #endif
 	struct hda_bus *hbus = sof_to_hbus(sdev);
 	struct hdac_device *hdev;
+	struct hda_codec *codec;
 	u32 hda_cmd = (address << 28) | (AC_NODE_ROOT << 20) |
 		(AC_VERB_PARAMETERS << 8) | AC_PAR_VENDOR_ID;
 	u32 resp = -1;
@@ -108,6 +141,7 @@ static int hda_codec_probe(struct snd_sof_dev *sdev, int address,
 
 	hda_priv->codec.bus = hbus;
 	hdev = &hda_priv->codec.core;
+	codec = &hda_priv->codec;
 
 	ret = snd_hdac_ext_bus_device_init(&hbus->core, address, hdev);
 	if (ret < 0)
@@ -122,6 +156,11 @@ static int hda_codec_probe(struct snd_sof_dev *sdev, int address,
 		hda_priv->need_display_power = true;
 	}
 
+	if (is_generic_config(hbus))
+		codec->probe_id = HDA_CODEC_ID_GENERIC;
+	else
+		codec->probe_id = 0;
+
 	/*
 	 * if common HDMI codec driver is not used, codec load
 	 * is skipped here and hdac_hdmi is used instead
@@ -129,7 +168,7 @@ static int hda_codec_probe(struct snd_sof_dev *sdev, int address,
 	if (hda_codec_use_common_hdmi ||
 	    (resp & 0xFFFF0000) != IDISP_VID_INTEL) {
 		hdev->type = HDA_DEV_LEGACY;
-		ret = hda_codec_load_module(&hda_priv->codec);
+		ret = hda_codec_load_module(codec);
 		/*
 		 * handle ret==0 (no driver bound) as an error, but pass
 		 * other return codes without modification
Signature · 1 line
2.25.1
author maintainer
#328 lore raw

PATCH 308/388 vfio/mdev: Fix reference count leak in add_mdev_supported_type

In reply to Sasha Levin

From: Qiushi Wu <[email protected]>

[ Upstream commit aa8ba13cae3134b8ef1c1b6879f66372531da738 ]

kobject_init_and_add() takes reference even when it fails. If this function returns an error, kobject_put() must be called to properly clean up the memory associated with the object. Thus, replace kfree() by kobject_put() to fix this issue. Previous commit "b8eb718348b8" fixed a similar problem.

Fixes 7b96953bc640 ("vfio: Mediated device Core driver")
---
 drivers/vfio/mdev/mdev_sysfs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/vfio/mdev/mdev_sysfs.c +1 -1
index 8ad14e5c02bf..917fd84c1c6f 100644
@@ -110,7 +110,7 @@ static struct mdev_type *add_mdev_supported_type(struct mdev_parent *parent,
 				   "%s-%s", dev_driver_string(parent->dev),
 				   group->name);
 	if (ret) {
-		kfree(type);
+		kobject_put(&type->kobj);
 		return ERR_PTR(ret);
 	}
 
Signature · 1 line
2.25.1
author maintainer
#329 lore raw

PATCH 309/388 rtc: rv3028: Add missed check for devm_regmap_init_i2c()

In reply to Sasha Levin

From: Chuhong Yuan <[email protected]>

[ Upstream commit c3b29bf6f166f6ed5f04f9c125477358e0e25df8 ]

rv3028_probe() misses a check for devm_regmap_init_i2c(). Add the missed check to fix it.

---
 drivers/rtc/rtc-rv3028.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/rtc/rtc-rv3028.c +2 -0
index a0ddc86c975a..ec84db0b3d7a 100644
@@ -755,6 +755,8 @@ static int rv3028_probe(struct i2c_client *client)
 		return -ENOMEM;
 
 	rv3028->regmap = devm_regmap_init_i2c(client, &regmap_config);
+	if (IS_ERR(rv3028->regmap))
+		return PTR_ERR(rv3028->regmap);
 
 	i2c_set_clientdata(client, rv3028);
 
Signature · 1 line
2.25.1
author maintainer
#330 lore raw

PATCH 310/388 mailbox: imx: Fix return in imx_mu_scu_xlate()

In reply to Sasha Levin

From: Dan Carpenter <[email protected]>

[ Upstream commit 1b3a347b7d56aa637157da1b7df225071af1421f ]

This called from mbox_request_channel(). The caller is expecting error pointers and not NULL so this "return NULL;" will lead to an Oops.

Fixes 0a67003b1985 ("mailbox: imx: add SCU MU support")
---
 drivers/mailbox/imx-mailbox.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/mailbox/imx-mailbox.c +1 -1
index 7906624a731c..9d6f0217077b 100644
@@ -374,7 +374,7 @@ static struct mbox_chan *imx_mu_scu_xlate(struct mbox_controller *mbox,
 		break;
 	default:
 		dev_err(mbox->dev, "Invalid chan type: %d\n", type);
-		return NULL;
+		return ERR_PTR(-EINVAL);
 	}
 
 	if (chan >= mbox->num_chans) {
Signature · 7 lines
2.25.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#331 lore raw

PATCH 311/388 mailbox: zynqmp-ipi: Fix NULL vs IS_ERR() check in zynqmp_ipi_mbox_probe()

In reply to Sasha Levin

From: Wei Yongjun <[email protected]>

[ Upstream commit 445aeeb569f8d7904f8cf80b7c6826bb651ef80e ]

In case of error, the function devm_ioremap() returns NULL pointer not ERR_PTR(). So we should check whether the return value of devm_ioremap() is NULL instead of IS_ERR.

Fixes 4981b82ba2ff ("mailbox: ZynqMP IPI mailbox controller")
---
 drivers/mailbox/zynqmp-ipi-mailbox.c | 20 ++++++++------------
 1 file changed, 8 insertions(+), 12 deletions(-)
drivers/mailbox/zynqmp-ipi-mailbox.c +8 -12
index 86887c9a349a..f9cc674ba9b7 100644
@@ -504,10 +504,9 @@ static int zynqmp_ipi_mbox_probe(struct zynqmp_ipi_mbox *ipi_mbox,
 		mchan->req_buf_size = resource_size(&res);
 		mchan->req_buf = devm_ioremap(mdev, res.start,
 					      mchan->req_buf_size);
-		if (IS_ERR(mchan->req_buf)) {
+		if (!mchan->req_buf) {
 			dev_err(mdev, "Unable to map IPI buffer I/O memory\n");
-			ret = PTR_ERR(mchan->req_buf);
-			return ret;
+			return -ENOMEM;
 		}
 	} else if (ret != -ENODEV) {
 		dev_err(mdev, "Unmatched resource %s, %d.\n", name, ret);
@@ -520,10 +519,9 @@ static int zynqmp_ipi_mbox_probe(struct zynqmp_ipi_mbox *ipi_mbox,
 		mchan->resp_buf_size = resource_size(&res);
 		mchan->resp_buf = devm_ioremap(mdev, res.start,
 					       mchan->resp_buf_size);
-		if (IS_ERR(mchan->resp_buf)) {
+		if (!mchan->resp_buf) {
 			dev_err(mdev, "Unable to map IPI buffer I/O memory\n");
-			ret = PTR_ERR(mchan->resp_buf);
-			return ret;
+			return -ENOMEM;
 		}
 	} else if (ret != -ENODEV) {
 		dev_err(mdev, "Unmatched resource %s.\n", name);
@@ -543,10 +541,9 @@ static int zynqmp_ipi_mbox_probe(struct zynqmp_ipi_mbox *ipi_mbox,
 		mchan->req_buf_size = resource_size(&res);
 		mchan->req_buf = devm_ioremap(mdev, res.start,
 					      mchan->req_buf_size);
-		if (IS_ERR(mchan->req_buf)) {
+		if (!mchan->req_buf) {
 			dev_err(mdev, "Unable to map IPI buffer I/O memory\n");
-			ret = PTR_ERR(mchan->req_buf);
-			return ret;
+			return -ENOMEM;
 		}
 	} else if (ret != -ENODEV) {
 		dev_err(mdev, "Unmatched resource %s.\n", name);
@@ -559,10 +556,9 @@ static int zynqmp_ipi_mbox_probe(struct zynqmp_ipi_mbox *ipi_mbox,
 		mchan->resp_buf_size = resource_size(&res);
 		mchan->resp_buf = devm_ioremap(mdev, res.start,
 					       mchan->resp_buf_size);
-		if (IS_ERR(mchan->resp_buf)) {
+		if (!mchan->resp_buf) {
 			dev_err(mdev, "Unable to map IPI buffer I/O memory\n");
-			ret = PTR_ERR(mchan->resp_buf);
-			return ret;
+			return -ENOMEM;
 		}
 	} else if (ret != -ENODEV) {
 		dev_err(mdev, "Unmatched resource %s.\n", name);
Signature · 1 line
2.25.1
author maintainer
#332 lore raw

PATCH 312/388 rxrpc: Adjust /proc/net/rxrpc/calls to display call->debug_id not user_ID

In reply to Sasha Levin

From: David Howells <[email protected]>

[ Upstream commit 32f71aa497cfb23d37149c2ef16ad71fce2e45e2 ]

The user ID value isn't actually much use - and leaks a kernel pointer or a userspace value - so replace it with the call debug ID, which appears in trace points.

---
 net/rxrpc/proc.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
net/rxrpc/proc.c +3 -3
index 8b179e3c802a..543afd9bd664 100644
@@ -68,7 +68,7 @@ static int rxrpc_call_seq_show(struct seq_file *seq, void *v)
 			 "Proto Local                                          "
 			 " Remote                                         "
 			 " SvID ConnID   CallID   End Use State    Abort   "
-			 " UserID           TxSeq    TW RxSeq    RW RxSerial RxTimo\n");
+			 " DebugId  TxSeq    TW RxSeq    RW RxSerial RxTimo\n");
 		return 0;
 	}
 
@@ -100,7 +100,7 @@ static int rxrpc_call_seq_show(struct seq_file *seq, void *v)
 	rx_hard_ack = READ_ONCE(call->rx_hard_ack);
 	seq_printf(seq,
 		   "UDP   %-47.47s %-47.47s %4x %08x %08x %s %3u"
-		   " %-8.8s %08x %lx %08x %02x %08x %02x %08x %06lx\n",
+		   " %-8.8s %08x %08x %08x %02x %08x %02x %08x %06lx\n",
 		   lbuff,
 		   rbuff,
 		   call->service_id,
@@ -110,7 +110,7 @@ static int rxrpc_call_seq_show(struct seq_file *seq, void *v)
 		   atomic_read(&call->usage),
 		   rxrpc_call_states[call->state],
 		   call->abort_code,
-		   call->user_call_ID,
+		   call->debug_id,
 		   tx_hard_ack, READ_ONCE(call->tx_top) - tx_hard_ack,
 		   rx_hard_ack, READ_ONCE(call->rx_top) - rx_hard_ack,
 		   call->rx_serial,
Signature · 1 line
2.25.1
author maintainer
#333 lore raw

PATCH 313/388 openrisc: Fix issue with argument clobbering for clone/fork

In reply to Sasha Levin

From: Stafford Horne <[email protected]>

[ Upstream commit 6bd140e14d9aaa734ec37985b8b20a96c0ece948 ]

Working on the OpenRISC glibc port I found that sometimes clone was working strange. That the tls data argument sent in r7 was always wrong. Further investigation revealed that the arguments were getting clobbered in the entry code. This patch removes the code that writes to the argument registers. This was likely due to some old code hanging around.

This patch fixes this up for clone and fork. This fork clobber is harmless but also useless so remove.

---
 arch/openrisc/kernel/entry.S | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/openrisc/kernel/entry.S +2 -2
index e4a78571f883..c6481cfc5220 100644
@@ -1166,13 +1166,13 @@ ENTRY(__sys_clone)
 	l.movhi	r29,hi(sys_clone)
 	l.ori	r29,r29,lo(sys_clone)
 	l.j	_fork_save_extra_regs_and_call
-	 l.addi	r7,r1,0
+	 l.nop
 
 ENTRY(__sys_fork)
 	l.movhi	r29,hi(sys_fork)
 	l.ori	r29,r29,lo(sys_fork)
 	l.j	_fork_save_extra_regs_and_call
-	 l.addi	r3,r1,0
+	 l.nop
 
 ENTRY(sys_rt_sigreturn)
 	l.jal	_sys_rt_sigreturn
Signature · 1 line
2.25.1
author maintainer
#334 lore raw

PATCH 314/388 drm/nouveau/disp/gm200-: fix NV_PDISP_SOR_HDMI2_CTRL(n) selection

In reply to Sasha Levin

From: Ben Skeggs <[email protected]>

[ Upstream commit a1ef8bad506e4ffa0c57ac5f8cb99ab5cbc3b1fc ]

This is a SOR register, and not indexed by the bound head.

Fixes display not coming up on high-bandwidth HDMI displays under a number of configurations.

---
 drivers/gpu/drm/nouveau/nvkm/engine/disp/hdmigm200.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/nouveau/nvkm/engine/disp/hdmigm200.c +2 -2
index 9b16a08eb4d9..bf6d41fb0c9f 100644
@@ -27,10 +27,10 @@ void
 gm200_hdmi_scdc(struct nvkm_ior *ior, int head, u8 scdc)
 {
 	struct nvkm_device *device = ior->disp->engine.subdev.device;
-	const u32 hoff = head * 0x800;
+	const u32 soff = nv50_ior_base(ior);
 	const u32 ctrl = scdc & 0x3;
 
-	nvkm_mask(device, 0x61c5bc + hoff, 0x00000003, ctrl);
+	nvkm_mask(device, 0x61c5bc + soff, 0x00000003, ctrl);
 
 	ior->tmds.high_speed = !!(scdc & 0x2);
 }
Signature · 1 line
2.25.1
author maintainer
#335 lore raw

PATCH 315/388 ceph: don't return -ESTALE if there's still an open file

In reply to Sasha Levin

From: Luis Henriques <[email protected]>

[ Upstream commit 878dabb64117406abd40977b87544d05bb3031fc ]

Similarly to commit 03f219041fdb ("ceph: check i_nlink while converting a file handle to dentry"), this fixes another corner case with name_to_handle_at/open_by_handle_at. The issue has been detected by xfstest generic/467, when doing:

- name_to_handle_at("/cephfs/myfile")
- open("/cephfs/myfile")
- unlink("/cephfs/myfile")
- sync; sync;
- drop caches
- open_by_handle_at()

The call to open_by_handle_at should not fail because the file hasn't been deleted yet (only unlinked) and we do have a valid handle to it. -ESTALE shall be returned only if i_nlink is 0 *and* i_count is 1.

This patch also makes sure we have LINK caps before checking i_nlink.

---
 fs/ceph/export.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
fs/ceph/export.c +8 -1
index 79dc06881e78..e088843a7734 100644
@@ -172,9 +172,16 @@ struct inode *ceph_lookup_inode(struct super_block *sb, u64 ino)
 static struct dentry *__fh_to_dentry(struct super_block *sb, u64 ino)
 {
 	struct inode *inode = __lookup_inode(sb, ino);
+	int err;
+
 	if (IS_ERR(inode))
 		return ERR_CAST(inode);
-	if (inode->i_nlink == 0) {
+	/* We need LINK caps to reliably check i_nlink */
+	err = ceph_do_getattr(inode, CEPH_CAP_LINK_SHARED, false);
+	if (err)
+		return ERR_PTR(err);
+	/* -ESTALE if inode as been unlinked and no file is open */
+	if ((inode->i_nlink == 0) && (atomic_read(&inode->i_count) == 1)) {
 		iput(inode);
 		return ERR_PTR(-ESTALE);
 	}
Signature · 1 line
2.25.1
author maintainer
#336 lore raw

PATCH 316/388 nfsd4: make drc_slab global, not per-net

In reply to Sasha Levin

From: "J. Bruce Fields" <[email protected]>

[ Upstream commit 027690c75e8fd91b60a634d31c4891a6e39d45bd ]

I made every global per-network-namespace instead. But perhaps doing that to this slab was a step too far.

The kmem_cache_create call in our net init method also seems to be responsible for this lockdep warning:

[   45.163710] Unable to find swap-space signature
[   45.375718] trinity-c1 (855): attempted to duplicate a private mapping with mremap.  This is not supported.
[   46.055744] futex_wake_op: trinity-c1 tries to shift op by -209; fix this program
[   51.011723]
[   51.013378] ======================================================
[   51.013875] WARNING: possible circular locking dependency detected
[   51.014378] 5.2.0-rc2 #1 Not tainted
[   51.014672] ------------------------------------------------------
[   51.015182] trinity-c2/886 is trying to acquire lock:
[   51.015593] 000000005405f099 (slab_mutex){+.+.}, at: slab_attr_store+0xa2/0x130
[   51.016190]
[   51.016190] but task is already holding lock:
[   51.016652] 00000000ac662005 (kn->count#43){++++}, at: kernfs_fop_write+0x286/0x500
[   51.017266]
[   51.017266] which lock already depends on the new lock.
[   51.017266]
[   51.017909]
[   51.017909] the existing dependency chain (in reverse order) is:
[   51.018497]
[   51.018497] -> #1 (kn->count#43){++++}:
[   51.018956]        __lock_acquire+0x7cf/0x1a20
[   51.019317]        lock_acquire+0x17d/0x390
[   51.019658]        __kernfs_remove+0x892/0xae0
[   51.020020]        kernfs_remove_by_name_ns+0x78/0x110
[   51.020435]        sysfs_remove_link+0x55/0xb0
[   51.020832]        sysfs_slab_add+0xc1/0x3e0
[   51.021332]        __kmem_cache_create+0x155/0x200
[   51.021720]        create_cache+0xf5/0x320
[   51.022054]        kmem_cache_create_usercopy+0x179/0x320
[   51.022486]        kmem_cache_create+0x1a/0x30
[   51.022867]        nfsd_reply_cache_init+0x278/0x560
[   51.023266]        nfsd_init_net+0x20f/0x5e0
[   51.023623]        ops_init+0xcb/0x4b0
[   51.023928]        setup_net+0x2fe/0x670
[   51.024315]        copy_net_ns+0x30a/0x3f0
[   51.024653]        create_new_namespaces+0x3c5/0x820
[   51.025257]        unshare_nsproxy_namespaces+0xd1/0x240
[   51.025881]        ksys_unshare+0x506/0x9c0
[   51.026381]        __x64_sys_unshare+0x3a/0x50
[   51.026937]        do_syscall_64+0x110/0x10b0
[   51.027509]        entry_SYSCALL_64_after_hwframe+0x49/0xbe
[   51.028175]
[   51.028175] -> #0 (slab_mutex){+.+.}:
[   51.028817]        validate_chain+0x1c51/0x2cc0
[   51.029422]        __lock_acquire+0x7cf/0x1a20
[   51.029947]        lock_acquire+0x17d/0x390
[   51.030438]        __mutex_lock+0x100/0xfa0
[   51.030995]        mutex_lock_nested+0x27/0x30
[   51.031516]        slab_attr_store+0xa2/0x130
[   51.032020]        sysfs_kf_write+0x11d/0x180
[   51.032529]        kernfs_fop_write+0x32a/0x500
[   51.033056]        do_loop_readv_writev+0x21d/0x310
[   51.033627]        do_iter_write+0x2e5/0x380
[   51.034148]        vfs_writev+0x170/0x310
[   51.034616]        do_pwritev+0x13e/0x160
[   51.035100]        __x64_sys_pwritev+0xa3/0x110
[   51.035633]        do_syscall_64+0x110/0x10b0
[   51.036200]        entry_SYSCALL_64_after_hwframe+0x49/0xbe
[   51.036924]
[   51.036924] other info that might help us debug this:
[   51.036924]
[   51.037876]  Possible unsafe locking scenario:
[   51.037876]
[   51.038556]        CPU0                    CPU1
[   51.039130]        ----                    ----
[   51.039676]   lock(kn->count#43);
[   51.040084]                                lock(slab_mutex);
[   51.040597]                                lock(kn->count#43);
[   51.041062]   lock(slab_mutex);
[   51.041320]
[   51.041320]  *** DEADLOCK ***
[   51.041320]
[   51.041793] 3 locks held by trinity-c2/886:
[   51.042128]  #0: 000000001f55e152 (sb_writers#5){.+.+}, at: vfs_writev+0x2b9/0x310
[   51.042739]  #1: 00000000c7d6c034 (&of->mutex){+.+.}, at: kernfs_fop_write+0x25b/0x500
[   51.043400]  #2: 00000000ac662005 (kn->count#43){++++}, at: kernfs_fop_write+0x286/0x500
Fixes 3ba75830ce17 "drc containerization"
---
 fs/nfsd/cache.h    |  2 ++
 fs/nfsd/netns.h    |  1 -
 fs/nfsd/nfscache.c | 29 +++++++++++++++++------------
 fs/nfsd/nfsctl.c   |  6 ++++++
 4 files changed, 25 insertions(+), 13 deletions(-)
fs/nfsd/cache.h +2 -0
index 10ec5ecdf117..65c331f75e9c 100644
@@ -78,6 +78,8 @@ enum {
 /* Checksum this amount of the request */
 #define RC_CSUMLEN		(256U)
 
+int	nfsd_drc_slab_create(void);
+void	nfsd_drc_slab_free(void);
 int	nfsd_reply_cache_init(struct nfsd_net *);
 void	nfsd_reply_cache_shutdown(struct nfsd_net *);
 int	nfsd_cache_lookup(struct svc_rqst *);
fs/nfsd/netns.h +0 -1
index 09aa545825bd..9217cb64bf0e 100644
@@ -139,7 +139,6 @@ struct nfsd_net {
 	 * Duplicate reply cache
 	 */
 	struct nfsd_drc_bucket   *drc_hashtbl;
-	struct kmem_cache        *drc_slab;
 
 	/* max number of entries allowed in the cache */
 	unsigned int             max_drc_entries;
fs/nfsd/nfscache.c +17 -12
index 96352ab7bd81..0c10bfea039e 100644
@@ -36,6 +36,8 @@ struct nfsd_drc_bucket {
 	spinlock_t cache_lock;
 };
 
+static struct kmem_cache	*drc_slab;
+
 static int	nfsd_cache_append(struct svc_rqst *rqstp, struct kvec *vec);
 static unsigned long nfsd_reply_cache_count(struct shrinker *shrink,
 					    struct shrink_control *sc);
@@ -95,7 +97,7 @@ nfsd_reply_cache_alloc(struct svc_rqst *rqstp, __wsum csum,
 {
 	struct svc_cacherep	*rp;
 
-	rp = kmem_cache_alloc(nn->drc_slab, GFP_KERNEL);
+	rp = kmem_cache_alloc(drc_slab, GFP_KERNEL);
 	if (rp) {
 		rp->c_state = RC_UNUSED;
 		rp->c_type = RC_NOCACHE;
@@ -129,7 +131,7 @@ nfsd_reply_cache_free_locked(struct nfsd_drc_bucket *b, struct svc_cacherep *rp,
 		atomic_dec(&nn->num_drc_entries);
 		nn->drc_mem_usage -= sizeof(*rp);
 	}
-	kmem_cache_free(nn->drc_slab, rp);
+	kmem_cache_free(drc_slab, rp);
 }
 
 static void
@@ -141,6 +143,18 @@ nfsd_reply_cache_free(struct nfsd_drc_bucket *b, struct svc_cacherep *rp,
 	spin_unlock(&b->cache_lock);
 }
 
+int nfsd_drc_slab_create(void)
+{
+	drc_slab = kmem_cache_create("nfsd_drc",
+				sizeof(struct svc_cacherep), 0, 0, NULL);
+	return drc_slab ? 0: -ENOMEM;
+}
+
+void nfsd_drc_slab_free(void)
+{
+	kmem_cache_destroy(drc_slab);
+}
+
 int nfsd_reply_cache_init(struct nfsd_net *nn)
 {
 	unsigned int hashsize;
@@ -159,18 +173,13 @@ int nfsd_reply_cache_init(struct nfsd_net *nn)
 	if (status)
 		goto out_nomem;
 
-	nn->drc_slab = kmem_cache_create("nfsd_drc",
-				sizeof(struct svc_cacherep), 0, 0, NULL);
-	if (!nn->drc_slab)
-		goto out_shrinker;
-
 	nn->drc_hashtbl = kcalloc(hashsize,
 				sizeof(*nn->drc_hashtbl), GFP_KERNEL);
 	if (!nn->drc_hashtbl) {
 		nn->drc_hashtbl = vzalloc(array_size(hashsize,
 						 sizeof(*nn->drc_hashtbl)));
 		if (!nn->drc_hashtbl)
-			goto out_slab;
+			goto out_shrinker;
 	}
 
 	for (i = 0; i < hashsize; i++) {
@@ -180,8 +189,6 @@ int nfsd_reply_cache_init(struct nfsd_net *nn)
 	nn->drc_hashsize = hashsize;
 
 	return 0;
-out_slab:
-	kmem_cache_destroy(nn->drc_slab);
 out_shrinker:
 	unregister_shrinker(&nn->nfsd_reply_cache_shrinker);
 out_nomem:
@@ -209,8 +216,6 @@ void nfsd_reply_cache_shutdown(struct nfsd_net *nn)
 	nn->drc_hashtbl = NULL;
 	nn->drc_hashsize = 0;
 
-	kmem_cache_destroy(nn->drc_slab);
-	nn->drc_slab = NULL;
 }
 
 /*
fs/nfsd/nfsctl.c +6 -0
index 3bb2db947d29..71687d99b090 100644
@@ -1533,6 +1533,9 @@ static int __init init_nfsd(void)
 		goto out_free_slabs;
 	nfsd_fault_inject_init(); /* nfsd fault injection controls */
 	nfsd_stat_init();	/* Statistics */
+	retval = nfsd_drc_slab_create();
+	if (retval)
+		goto out_free_stat;
 	nfsd_lockd_init();	/* lockd->nfsd callbacks */
 	retval = create_proc_exports_entry();
 	if (retval)
@@ -1546,6 +1549,8 @@ static int __init init_nfsd(void)
 	remove_proc_entry("fs/nfs", NULL);
 out_free_lockd:
 	nfsd_lockd_shutdown();
+	nfsd_drc_slab_free();
+out_free_stat:
 	nfsd_stat_shutdown();
 	nfsd_fault_inject_cleanup();
 	nfsd4_exit_pnfs();
@@ -1560,6 +1565,7 @@ static int __init init_nfsd(void)
 
 static void __exit exit_nfsd(void)
 {
+	nfsd_drc_slab_free();
 	remove_proc_entry("fs/nfs/exports", NULL);
 	remove_proc_entry("fs/nfs", NULL);
 	nfsd_stat_shutdown();
Signature · 1 line
2.25.1
author maintainer
#337 lore raw

PATCH 317/388 pwm: imx27: Fix rounding behavior

In reply to Sasha Levin

From: Uwe Kleine-König <[email protected]>

[ Upstream commit aef1a3799b5cb3ba4841f6034497b179646ccc70 ]

To not trigger the warnings provided by CONFIG_PWM_DEBUG

- use up-rounding in .get_state()
- don't divide by the result of a division
- don't use the rounded counter value for the period length to calculate the counter value for the duty cycle

---
 drivers/pwm/pwm-imx27.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)
drivers/pwm/pwm-imx27.c +10 -10
index a6e40d4c485f..732a6f3701e8 100644
@@ -150,13 +150,12 @@ static void pwm_imx27_get_state(struct pwm_chip *chip,
 
 	prescaler = MX3_PWMCR_PRESCALER_GET(val);
 	pwm_clk = clk_get_rate(imx->clk_per);
-	pwm_clk = DIV_ROUND_CLOSEST_ULL(pwm_clk, prescaler);
 	val = readl(imx->mmio_base + MX3_PWMPR);
 	period = val >= MX3_PWMPR_MAX ? MX3_PWMPR_MAX : val;
 
 	/* PWMOUT (Hz) = PWMCLK / (PWMPR + 2) */
-	tmp = NSEC_PER_SEC * (u64)(period + 2);
-	state->period = DIV_ROUND_CLOSEST_ULL(tmp, pwm_clk);
+	tmp = NSEC_PER_SEC * (u64)(period + 2) * prescaler;
+	state->period = DIV_ROUND_UP_ULL(tmp, pwm_clk);
 
 	/*
 	 * PWMSAR can be read only if PWM is enabled. If the PWM is disabled,
@@ -167,8 +166,8 @@ static void pwm_imx27_get_state(struct pwm_chip *chip,
 	else
 		val = imx->duty_cycle;
 
-	tmp = NSEC_PER_SEC * (u64)(val);
-	state->duty_cycle = DIV_ROUND_CLOSEST_ULL(tmp, pwm_clk);
+	tmp = NSEC_PER_SEC * (u64)(val) * prescaler;
+	state->duty_cycle = DIV_ROUND_UP_ULL(tmp, pwm_clk);
 
 	pwm_imx27_clk_disable_unprepare(imx);
 }
@@ -220,22 +219,23 @@ static int pwm_imx27_apply(struct pwm_chip *chip, struct pwm_device *pwm,
 	struct pwm_imx27_chip *imx = to_pwm_imx27_chip(chip);
 	struct pwm_state cstate;
 	unsigned long long c;
+	unsigned long long clkrate;
 	int ret;
 	u32 cr;
 
 	pwm_get_state(pwm, &cstate);
 
-	c = clk_get_rate(imx->clk_per);
-	c *= state->period;
+	clkrate = clk_get_rate(imx->clk_per);
+	c = clkrate * state->period;
 
-	do_div(c, 1000000000);
+	do_div(c, NSEC_PER_SEC);
 	period_cycles = c;
 
 	prescale = period_cycles / 0x10000 + 1;
 
 	period_cycles /= prescale;
-	c = (unsigned long long)period_cycles * state->duty_cycle;
-	do_div(c, state->period);
+	c = clkrate * state->duty_cycle;
+	do_div(c, NSEC_PER_SEC * prescale);
 	duty_cycles = c;
 
 	/*
Signature · 7 lines
2.25.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#338 lore raw

PATCH 318/388 gfs2: Allow lock_nolock mount to specify jid=X

In reply to Sasha Levin

From: Bob Peterson <[email protected]>

[ Upstream commit ea22eee4e6027d8927099de344f7fff43c507ef9 ]

Before this patch, a simple typo accidentally added \n to the jid= string for lock_nolock mounts. This made it impossible to mount a gfs2 file system with a journal other than journal0. Thus:

mount -tgfs2 -o hostdata="jid=1" <device> <mount pt>

Resulted in:
mount: wrong fs type, bad option, bad superblock on <device>

In most cases this is not a problem. However, for debugging and testing purposes we sometimes want to test the integrity of other journals. This patch removes the unnecessary \n and thus allows lock_nolock users to specify an alternate journal.

---
 fs/gfs2/ops_fstype.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/gfs2/ops_fstype.c +1 -1
index e2b69ffcc6a8..094f5fe7c009 100644
@@ -880,7 +880,7 @@ static int init_per_node(struct gfs2_sbd *sdp, int undo)
 }
 
 static const match_table_t nolock_tokens = {
-	{ Opt_jid, "jid=%d\n", },
+	{ Opt_jid, "jid=%d", },
 	{ Opt_err, NULL },
 };
 
Signature · 1 line
2.25.1
author maintainer
#339 lore raw

PATCH 319/388 ovl: verify permissions in ovl_path_open()

In reply to Sasha Levin

From: Miklos Szeredi <[email protected]>

[ Upstream commit 56230d956739b9cb1cbde439d76227d77979a04d ]

Check permission before opening a real file.

ovl_path_open() is used by readdir and copy-up routines.

ovl_permission() theoretically already checked copy up permissions, but it doesn't hurt to re-do these checks during the actual copy-up.

For directory reading ovl_permission() only checks access to topmost underlying layer. Readdir on a merged directory accesses layers below the topmost one as well. Permission wasn't checked for these layers.

Note: modifying ovl_permission() to perform this check would be far more complex and hence more bug prone. The result is less precise permissions returned in access(2). If this turns out to be an issue, we can revisit this bug.

---
 fs/overlayfs/util.c | 27 ++++++++++++++++++++++++++-
 1 file changed, 26 insertions(+), 1 deletion(-)
fs/overlayfs/util.c +26 -1
index 36b60788ee47..a0878039332a 100644
@@ -459,7 +459,32 @@ bool ovl_is_whiteout(struct dentry *dentry)
 
 struct file *ovl_path_open(struct path *path, int flags)
 {
-	return dentry_open(path, flags | O_NOATIME, current_cred());
+	struct inode *inode = d_inode(path->dentry);
+	int err, acc_mode;
+
+	if (flags & ~(O_ACCMODE | O_LARGEFILE))
+		BUG();
+
+	switch (flags & O_ACCMODE) {
+	case O_RDONLY:
+		acc_mode = MAY_READ;
+		break;
+	case O_WRONLY:
+		acc_mode = MAY_WRITE;
+		break;
+	default:
+		BUG();
+	}
+
+	err = inode_permission(inode, acc_mode | MAY_OPEN);
+	if (err)
+		return ERR_PTR(err);
+
+	/* O_NOATIME is an optimization, don't fail if not permitted */
+	if (inode_owner_or_capable(inode))
+		flags |= O_NOATIME;
+
+	return dentry_open(path, flags, current_cred());
 }
 
 /* Caller should hold ovl_inode->lock */
Signature · 1 line
2.25.1
#340 lore raw

Re: [PATCH AUTOSEL 5.7 319/388] ovl: verify permissions in ovl_path_open()

Quoted 32 lines · Sasha Levin

From: Miklos Szeredi <[email protected]>

[ Upstream commit 56230d956739b9cb1cbde439d76227d77979a04d ]

Check permission before opening a real file.

ovl_path_open() is used by readdir and copy-up routines.

ovl_permission() theoretically already checked copy up permissions, but it doesn't hurt to re-do these checks during the actual copy-up.

For directory reading ovl_permission() only checks access to topmost underlying layer. Readdir on a merged directory accesses layers below the topmost one as well. Permission wasn't checked for these layers.

Note: modifying ovl_permission() to perform this check would be far more complex and hence more bug prone. The result is less precise permissions returned in access(2). If this turns out to be an issue, we can revisit this bug.

---
 fs/overlayfs/util.c | 27 ++++++++++++++++++++++++++-
 1 file changed, 26 insertions(+), 1 deletion(-)
fs/overlayfs/util.c +0 -0
index 36b60788ee47..a0878039332a 100644

fs/overlayfs/util.c bool ovl_is_whiteout(struct dentry *dentry) · line 466

Quoted 9 lines
fs/overlayfs/util.c +5 -1
@@ -459,7 +459,32 @@ bool ovl_is_whiteout(struct dentry *dentry)
 
 struct file *ovl_path_open(struct path *path, int flags)
 {
-       return dentry_open(path, flags | O_NOATIME, current_cred());
+       struct inode *inode = d_inode(path->dentry);
+       int err, acc_mode;
+
+       if (flags & ~(O_ACCMODE | O_LARGEFILE))
+               BUG();

This BUG: triggered on stable-rc 5.7, 5.4, 4.19 and 4.14 on arm64 architecture.

steps to reproduce:
- cd /opt/ltp
- ./runltp -s execveat03

[  564.915016] ------------[ cut here ]------------
[  564.919654] kernel BUG at fs/overlayfs/util.c:466!
[  564.924458] Internal error: Oops - BUG: 0 [#1] PREEMPT SMP
[  564.929954] Modules linked in: overlay rfkill tda998x cec
drm_kms_helper drm crct10dif_ce fuse
[  564.938617] CPU: 1 PID: 15054 Comm: execveat03 Not tainted
5.7.6-rc1-00793-gccc085714c08 #1
[  564.946987] Hardware name: ARM Juno development board (r2) (DT)
[  564.952921] pstate: 60000005 (nZCv daif -PAN -UAO)
[  564.957786] pc : ovl_path_open+0xd0/0xd8 [overlay]
[  564.962642] lr : ovl_path_open+0x30/0xd8 [overlay]
[  564.967443] sp : ffff00090a65f740
[  564.970763] x29: ffff00090a65f740 x28: 0000000000000000
[  564.976091] x27: ffff00090a65f950 x26: ffffa000097506c0
[  564.981419] x25: ffff00090388f000 x24: ffff000912b39a28
[  564.986746] x23: ffff00090a65f950 x22: ffff00090b45e580
[  564.992072] x21: ffff00090a65f950 x20: ffff0008dddbcb10
[  564.997398] x19: 0000000000004000 x18: 0000000000000000
[  565.002723] x17: 0000000000000000 x16: 0000000000000000
[  565.008048] x15: 0000000000000000 x14: ffffa000103c0898
[  565.013374] x13: ffffa00010382fec x12: ffff80011bd530dc
[  565.018700] x11: 1fffe0011bd530db x10: ffff80011bd530db
[  565.024026] x9 : dfffa00000000000 x8 : ffff0008dea986d8
[  565.029351] x7 : 0000000000000001 x6 : 00000000f1f1f1f1
[  565.034677] x5 : 00000000f3000000 x4 : dfffa00000000000
[  565.040003] x3 : ffffa0000973f950 x2 : 0000000000000007
[  565.045328] x1 : 0000000000004000 x0 : ffff00090f79c930
[  565.050653] Call trace:
[  565.053162]  ovl_path_open+0xd0/0xd8 [overlay]
[  565.057670]  ovl_check_d_type_supported+0x9c/0x180 [overlay]
[  565.063396]  ovl_get_workdir.isra.0+0x238/0x5e0 [overlay]
[  565.068861]  ovl_fill_super+0x800/0x1a48 [overlay]
[  565.073673]  mount_nodev+0x50/0xc0
[  565.077135]  ovl_mount+0x1c/0x28 [overlay]
[  565.081245]  legacy_get_tree+0x74/0xc0
[  565.085005]  vfs_get_tree+0x4c/0x160
[  565.088589]  do_mount+0x700/0xb90
[  565.091913]  __arm64_sys_mount+0xc0/0x110
[  565.095938]  el0_svc_common.constprop.0+0xa4/0x1e0
[  565.100742]  do_el0_svc+0x38/0xa0
[  565.104068]  el0_sync_handler+0x11c/0x190
[  565.108087]  el0_sync+0x158/0x180
[  565.111419] Code: a8c37bfd d50323bf d65f03c0 d4210000 (d4210000)
[  565.117529] ---[ end trace 357a3872beec6c53 ]---
[  565.122157] note: execveat03[15054] exited with preempt_count 1
[  565.129476] ------------[ cut here ]------------
[  565.134122] WARNING: CPU: 1 PID: 0 at kernel/rcu/tree.c:569
rcu_idle_enter+0xd0/0xd8
[  565.141879] Modules linked in: overlay rfkill tda998x cec
drm_kms_helper drm crct10dif_ce fuse
[  565.150539] CPU: 1 PID: 0 Comm: swapper/1 Tainted: G      D
  5.7.6-rc1-00793-gccc085714c08 #1
[  565.159866] Hardware name: ARM Juno development board (r2) (DT)
[  565.165799] pstate: 200003c5 (nzCv DAIF -PAN -UAO)
[  565.170603] pc : rcu_idle_enter+0xd0/0xd8
[  565.174623] lr : rcu_idle_enter+0x34/0xd8
[  565.178639] sp : ffff000935577ea0
[  565.181958] x29: ffff000935577ea0 x28: 0000000000000000
[  565.187285] x27: ffff000912d83000 x26: ffffa0001359e2c0
 [  565.192611] x25: ffff000935552b80 x24: 1fffe00126aaefe6
[  565.198098] x23: ffffa000125abff8 x22: ffffa0001359e298
[  565.203424] x21: ffff000936606fd0 x20: ffffa000125b5f00
[  565.208909] x19: ffff000936606f00 x18: 0000000000000000

Broadcast message from systemd-jo[ 565.214234] x17: 0000000000000000
x16: 0000000000000000
urnald@juno (Mon 2020-06-22 19:5[ 565.222417] x15: 0000000000000000
x14: ffffa000103c1370
9:27 UTC):

[  565.230517] x13: ffffa000103c0898 x12: ffff800126cc0d8c
[  565.236959] x11: 1fffe00126cc0d8b x10: ffff800126cc0d8b
[  565.242285] x9 : dfffa00000000000 x8 : ffff000936606c5b
[  565.247683] x7 : 0000000000000001 x6 : ffff000936606c58
[  565.253009] x5 : 00007ffed933f275 x4 : dfffa00000000000
[  565.258493] x3 : ffffa000101d48e4 x2 : 0000000000000007

kernel[300]: [ 564.924458] Inter[ 565.263819] x1 : 4000000000000002 x0 : 4000000000000000
nal error: Oops - BUG: 0 [#1] PR[ 565.272001] Call trace:
EEMPT SMP

[  565.277235]  rcu_idle_enter+0xd0/0xd8
[  565.281938]  do_idle+0x288/0x328
[  565.285176]  cpu_startup_entry+0x28/0x48
[  565.289182]  secondary_start_kernel+0x248/0x2b0
[  565.293721] ---[ end trace 357a3872beec6c54 ]---

https://lkft.validation.linaro.org/scheduler/job/1515478#L3876

Signature · 2 lines
Linaro LKFT
https://lkft.linaro.org
author maintainer
#341 lore raw

PATCH 320/388 scsi: iscsi: Fix reference count leak in iscsi_boot_create_kobj

In reply to Sasha Levin

From: Qiushi Wu <[email protected]>

[ Upstream commit 0267ffce562c8bbf9b57ebe0e38445ad04972890 ]

kobject_init_and_add() takes reference even when it fails. If this function returns an error, kobject_put() must be called to properly clean up the memory associated with the object.

---
 drivers/scsi/iscsi_boot_sysfs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/iscsi_boot_sysfs.c +1 -1
index e4857b728033..a64abe38db2d 100644
@@ -352,7 +352,7 @@ iscsi_boot_create_kobj(struct iscsi_boot_kset *boot_kset,
 	boot_kobj->kobj.kset = boot_kset->kset;
 	if (kobject_init_and_add(&boot_kobj->kobj, &iscsi_boot_ktype,
 				 NULL, name, index)) {
-		kfree(boot_kobj);
+		kobject_put(&boot_kobj->kobj);
 		return NULL;
 	}
 	boot_kobj->data = data;
Signature · 1 line
2.25.1
author maintainer
#342 lore raw

PATCH 321/388 scsi: ufs: Don't update urgent bkops level when toggling auto bkops

In reply to Sasha Levin

From: Can Guo <[email protected]>

[ Upstream commit be32acff43800c87dc5c707f5d47cc607b76b653 ]

Urgent bkops level is used to compare against actual bkops status read from UFS device. Urgent bkops level is set during initialization and might be updated in exception event handler during runtime. But it should not be updated to the actual bkops status every time when auto bkops is toggled. Otherwise, if urgent bkops level is updated to 0, auto bkops shall always be kept enabled.

Fixes 24366c2afbb0 ("scsi: ufs: Recheck bkops level if bkops is disabled")
---
 drivers/scsi/ufs/ufshcd.c | 1 -
 1 file changed, 1 deletion(-)
drivers/scsi/ufs/ufshcd.c +0 -1
index 698e8d20b4ba..52740b60d786 100644
@@ -5098,7 +5098,6 @@ static int ufshcd_bkops_ctrl(struct ufs_hba *hba,
 		err = ufshcd_enable_auto_bkops(hba);
 	else
 		err = ufshcd_disable_auto_bkops(hba);
-	hba->urgent_bkops_lvl = curr_status;
 out:
 	return err;
 }
Signature · 7 lines
2.25.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#343 lore raw

PATCH 322/388 modpost: fix -i (--ignore-errors) MAKEFLAGS detection

In reply to Sasha Levin

From: Masahiro Yamada <[email protected]>

[ Upstream commit 91e6ee581270b8ae970f028b898314d73f16870b ]

$(filter -i,$(MAKEFLAGS)) works only in limited use-cases.

The representation of $(MAKEFLAGS) depends on various factors:
- GNU Make version (version 3.8x or version 4.x)
- The presence of other flags like -j

In my experiments, $(MAKEFLAGS) is expanded as follows:

* GNU Make 3.8x:

* without -j option:
--no-print-directory -Rri

* with -j option:
--no-print-directory -Rr --jobserver-fds=3,4 -j -i

* GNU Make 4.x:

* without -j option:
irR --no-print-directory

* with -j option:
irR -j --jobserver-fds=3,4 --no-print-directory

For GNU Make 4.x, the flags are grouped as 'irR', which does not work.

For the single thread build with GNU Make 3.8x, the flags are grouped as '-Rri', which does not work either.

To make it work for all cases, do likewise as commit 6f0fa58e4596 ("kbuild: simplify silent build (-s) detection").

BTW, since commit ff9b45c55b26 ("kbuild: modpost: read modules.order instead of $(MODVERDIR)/*.mod"), you also need to pass -k option to build final *.ko files. 'make -i -k' ignores compile errors in modules, and build as many remaining *.ko as possible.

Please note this feature is kind of dangerous if other modules depend on the broken module because the generated modules will lack the correct module dependency or CRC. Honestly, I am not a big fan of it, but I am keeping this feature.

Fixes eed380f3f593 ("modpost: Optionally ignore secondary errors seen if a single module build fails")
Cc Guenter Roeck <[email protected]>
---
 scripts/Makefile.modpost | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
scripts/Makefile.modpost +6 -1
index 957eed6a17a5..33aaa572f686 100644
@@ -66,7 +66,7 @@ __modpost:
 
 else
 
-MODPOST += $(subst -i,-n,$(filter -i,$(MAKEFLAGS))) -s -T - \
+MODPOST += -s -T - \
 	$(if $(KBUILD_NSDEPS),-d $(MODULES_NSDEPS))
 
 ifeq ($(KBUILD_EXTMOD),)
@@ -82,6 +82,11 @@ include $(if $(wildcard $(KBUILD_EXTMOD)/Kbuild), \
              $(KBUILD_EXTMOD)/Kbuild, $(KBUILD_EXTMOD)/Makefile)
 endif
 
+# 'make -i -k' ignores compile errors, and builds as many modules as possible.
+ifneq ($(findstring i,$(filter-out --%,$(MAKEFLAGS))),)
+MODPOST += -n
+endif
+
 # find all modules listed in modules.order
 modules := $(sort $(shell cat $(MODORDER)))
 
Signature · 1 line
2.25.1
author maintainer
#344 lore raw

PATCH 323/388 pinctrl: imxl: Fix an error handling path in 'imx1_pinctrl_core_probe()'

In reply to Sasha Levin

From: Christophe JAILLET <[email protected]>

[ Upstream commit 9eb728321286c4b31e964d2377fca2368526d408 ]

When 'pinctrl_register()' has been turned into 'devm_pinctrl_register()', an error handling path has not been updated.

Axe a now unneeded 'pinctrl_unregister()'.

Fixes e55e025d1687 ("pinctrl: imxl: Use devm_pinctrl_register() for pinctrl registration")
---
 drivers/pinctrl/freescale/pinctrl-imx1-core.c | 1 -
 1 file changed, 1 deletion(-)
drivers/pinctrl/freescale/pinctrl-imx1-core.c +0 -1
index c00d0022d311..421f7d1886e5 100644
@@ -638,7 +638,6 @@ int imx1_pinctrl_core_probe(struct platform_device *pdev,
 
 	ret = of_platform_populate(pdev->dev.of_node, NULL, NULL, &pdev->dev);
 	if (ret) {
-		pinctrl_unregister(ipctl->pctl);
 		dev_err(&pdev->dev, "Failed to populate subdevices\n");
 		return ret;
 	}
Signature · 1 line
2.25.1
author maintainer
#345 lore raw

PATCH 324/388 pinctrl: sirf: add missing put_device() call in sirfsoc_gpio_probe()

In reply to Sasha Levin

From: yu kuai <[email protected]>

[ Upstream commit 66339f2fba3fa05b0507e1887e4c6abcfc6228a3 ]

A coccicheck run provided information like the following:

drivers/pinctrl/sirf/pinctrl-sirf.c:798:2-8: ERROR: missing put_device; call of_find_device_by_node on line 792, but without a corresponding object release within this function.

Generated by: scripts/coccinelle/free/put_device.cocci

Thus add a jump target to fix the exception handling for this function implementation.

Fixes 5130216265f6 ("PINCTRL: SiRF: add GPIO and GPIO irq support in CSR SiRFprimaII")
---
 drivers/pinctrl/sirf/pinctrl-sirf.c | 20 ++++++++++++++------
 1 file changed, 14 insertions(+), 6 deletions(-)
drivers/pinctrl/sirf/pinctrl-sirf.c +14 -6
index 1ebcb957c654..63a287d5795f 100644
@@ -794,13 +794,17 @@ static int sirfsoc_gpio_probe(struct device_node *np)
 		return -ENODEV;
 
 	sgpio = devm_kzalloc(&pdev->dev, sizeof(*sgpio), GFP_KERNEL);
-	if (!sgpio)
-		return -ENOMEM;
+	if (!sgpio) {
+		err = -ENOMEM;
+		goto out_put_device;
+	}
 	spin_lock_init(&sgpio->lock);
 
 	regs = of_iomap(np, 0);
-	if (!regs)
-		return -ENOMEM;
+	if (!regs) {
+		err = -ENOMEM;
+		goto out_put_device;
+	}
 
 	sgpio->chip.gc.request = sirfsoc_gpio_request;
 	sgpio->chip.gc.free = sirfsoc_gpio_free;
@@ -824,8 +828,10 @@ static int sirfsoc_gpio_probe(struct device_node *np)
 	girq->parents = devm_kcalloc(&pdev->dev, SIRFSOC_GPIO_NO_OF_BANKS,
 				     sizeof(*girq->parents),
 				     GFP_KERNEL);
-	if (!girq->parents)
-		return -ENOMEM;
+	if (!girq->parents) {
+		err = -ENOMEM;
+		goto out_put_device;
+	}
 	for (i = 0; i < SIRFSOC_GPIO_NO_OF_BANKS; i++) {
 		bank = &sgpio->sgpio_bank[i];
 		spin_lock_init(&bank->lock);
@@ -868,6 +874,8 @@ static int sirfsoc_gpio_probe(struct device_node *np)
 	gpiochip_remove(&sgpio->chip.gc);
 out:
 	iounmap(regs);
+out_put_device:
+	put_device(&pdev->dev);
 	return err;
 }
 
Signature · 1 line
2.25.1
author maintainer
#346 lore raw

PATCH 325/388 pinctrl: freescale: imx: Fix an error handling path in 'imx_pinctrl_probe()'

In reply to Sasha Levin

From: Christophe JAILLET <[email protected]>

[ Upstream commit 11d8da5cabf7c6c3263ba2cd9c00260395867048 ]

'pinctrl_unregister()' should not be called to undo 'devm_pinctrl_register_and_init()', it is already handled by the framework.

This simplifies the error handling paths of the probe function. The 'imx_free_resources()' can be removed as well.

Fixes a51c158bf0f7 ("pinctrl: imx: use radix trees for groups and functions")
---
 drivers/pinctrl/freescale/pinctrl-imx.c | 19 ++-----------------
 1 file changed, 2 insertions(+), 17 deletions(-)
drivers/pinctrl/freescale/pinctrl-imx.c +2 -17
index 9f42036c5fbb..1f81569c7ae3 100644
@@ -774,16 +774,6 @@ static int imx_pinctrl_probe_dt(struct platform_device *pdev,
 	return 0;
 }
 
-/*
- * imx_free_resources() - free memory used by this driver
- * @info: info driver instance
- */
-static void imx_free_resources(struct imx_pinctrl *ipctl)
-{
-	if (ipctl->pctl)
-		pinctrl_unregister(ipctl->pctl);
-}
-
 int imx_pinctrl_probe(struct platform_device *pdev,
 		      const struct imx_pinctrl_soc_info *info)
 {
@@ -874,23 +864,18 @@ int imx_pinctrl_probe(struct platform_device *pdev,
 					     &ipctl->pctl);
 	if (ret) {
 		dev_err(&pdev->dev, "could not register IMX pinctrl driver\n");
-		goto free;
+		return ret;
 	}
 
 	ret = imx_pinctrl_probe_dt(pdev, ipctl);
 	if (ret) {
 		dev_err(&pdev->dev, "fail to probe dt properties\n");
-		goto free;
+		return ret;
 	}
 
 	dev_info(&pdev->dev, "initialized IMX pinctrl driver\n");
 
 	return pinctrl_enable(ipctl->pctl);
-
-free:
-	imx_free_resources(ipctl);
-
-	return ret;
 }
 
 static int __maybe_unused imx_pinctrl_suspend(struct device *dev)
Signature · 1 line
2.25.1
author maintainer
#347 lore raw

PATCH 326/388 pinctrl: freescale: imx: Use 'devm_of_iomap()' to avoid a resource leak in case of error in 'imx_pinctrl_probe()'

In reply to Sasha Levin

From: Christophe JAILLET <[email protected]>

[ Upstream commit ba403242615c2c99e27af7984b1650771a2cc2c9 ]

Use 'devm_of_iomap()' instead 'of_iomap()' to avoid a resource leak in case of error.

Update the error handling code accordingly.

Fixes 26d8cde5260b ("pinctrl: freescale: imx: add shared input select reg support")
---
 drivers/pinctrl/freescale/pinctrl-imx.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/pinctrl/freescale/pinctrl-imx.c +4 -3
index 1f81569c7ae3..cb7e0f08d2cf 100644
@@ -824,12 +824,13 @@ int imx_pinctrl_probe(struct platform_device *pdev,
 				return -EINVAL;
 			}
 
-			ipctl->input_sel_base = of_iomap(np, 0);
+			ipctl->input_sel_base = devm_of_iomap(&pdev->dev, np,
+							      0, NULL);
 			of_node_put(np);
-			if (!ipctl->input_sel_base) {
+			if (IS_ERR(ipctl->input_sel_base)) {
 				dev_err(&pdev->dev,
 					"iomuxc input select base address not found\n");
-				return -ENOMEM;
+				return PTR_ERR(ipctl->input_sel_base);
 			}
 		}
 	}
Signature · 7 lines
2.25.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#348 lore raw

PATCH 327/388 nfsd: safer handling of corrupted c_type

In reply to Sasha Levin

From: "J. Bruce Fields" <[email protected]>

[ Upstream commit c25bf185e57213b54ea0d632ac04907310993433 ]

This can only happen if there's a bug somewhere, so let's make it a WARN not a printk. Also, I think it's safest to ignore the corruption rather than trying to fix it by removing a cache entry.

---
 fs/nfsd/nfscache.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
fs/nfsd/nfscache.c +1 -2
index 0c10bfea039e..4a258065188e 100644
@@ -469,8 +469,7 @@ nfsd_cache_lookup(struct svc_rqst *rqstp)
 		rtn = RC_REPLY;
 		break;
 	default:
-		printk(KERN_WARNING "nfsd: bad repcache type %d\n", rp->c_type);
-		nfsd_reply_cache_free_locked(b, rp, nn);
+		WARN_ONCE(1, "nfsd: bad repcache type %d\n", rp->c_type);
 	}
 
 	goto out;
Signature · 1 line
2.25.1
author maintainer
#349 lore raw

PATCH 328/388 RDMA/cm: Spurious WARNING triggered in cm_destroy_id()

In reply to Sasha Levin

From: Ka-Cheong Poon <[email protected]>

[ Upstream commit fba97dc7fc76b2c9a909fa0b3786d30a9899f5cf ]

If the cm_id state is IB_CM_REP_SENT when cm_destroy_id() is called, it calls cm_send_rej_locked().

In cm_send_rej_locked(), it calls cm_enter_timewait() and the state is changed to IB_CM_TIMEWAIT.

Now back to cm_destroy_id(), it breaks from the switch statement, and the next call is WARN_ON(cm_id->state != IB_CM_IDLE).

This triggers a spurious warning. Instead, the code should goto retest after returning from cm_send_rej_locked() to move the state to IDLE.

Fixes 67b3c8dceac6 ("RDMA/cm: Make sure the cm_id is in the IB_CM_IDLE state in destroy")
---
 drivers/infiniband/core/cm.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/infiniband/core/cm.c +3 -1
index 17f14e0eafe4..1c2bf18cda9f 100644
@@ -1076,7 +1076,9 @@ static void cm_destroy_id(struct ib_cm_id *cm_id, int err)
 	case IB_CM_REP_SENT:
 	case IB_CM_MRA_REP_RCVD:
 		ib_cancel_mad(cm_id_priv->av.port->mad_agent, cm_id_priv->msg);
-		/* Fall through */
+		cm_send_rej_locked(cm_id_priv, IB_CM_REJ_CONSUMER_DEFINED, NULL,
+				   0, NULL, 0);
+		goto retest;
 	case IB_CM_MRA_REQ_SENT:
 	case IB_CM_REP_RCVD:
 	case IB_CM_MRA_REP_SENT:
Signature · 1 line
2.25.1
author maintainer
#350 lore raw

PATCH 329/388 drm/amd/display: Revalidate bandwidth before commiting DC updates

In reply to Sasha Levin

From: Nicholas Kazlauskas <[email protected]>

[ Upstream commit a24eaa5c51255b344d5a321f1eeb3205f2775498 ]

[Why]
Whenever we switch between tiled formats without also switching pixel formats or doing anything else that recreates the DC plane state we can run into underflow or hangs since we're not updating the DML parameters before committing to the hardware.

[How]
If the update type is FULL then call validate_bandwidth again to update the DML parmeters before committing the state.

This is basically just a workaround and protective measure against update types being added DC where we could run into this issue in the future.

We can only fully validate the state in advance before applying it to the hardware if we recreate all the plane and stream states since we can't modify what's currently in use.

The next step is to update DM to ensure that we're creating the plane and stream states for whatever could potentially be a full update in DC to pre-emptively recreate the state for DC global validation.

The workaround can stay until this has been fixed in DM.

---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 6 ++++++
 1 file changed, 6 insertions(+)
drivers/gpu/drm/amd/display/dc/core/dc.c +6 -0
index 47431ca6986d..4a619328101c 100644
@@ -2517,6 +2517,12 @@ void dc_commit_updates_for_stream(struct dc *dc,
 
 	copy_stream_update_to_stream(dc, context, stream, stream_update);
 
+	if (!dc->res_pool->funcs->validate_bandwidth(dc, context, false)) {
+		DC_ERROR("Mode validation failed for stream update!\n");
+		dc_release_state(context);
+		return;
+	}
+
 	commit_planes_for_stream(
 				dc,
 				srf_updates,
Signature · 6 lines
2.25.1

_______________________________________________
amd-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
author maintainer
#351 lore raw

PATCH 330/388 ext4: handle ext4_mark_inode_dirty errors

In reply to Sasha Levin

From: Harshad Shirwadkar <[email protected]>

[ Upstream commit 4209ae12b12265d475bba28634184423149bd14f ]

ext4_mark_inode_dirty() can fail for real reasons. Ignoring its return value may lead ext4 to ignore real failures that would result in corruption / crashes. Harden ext4_mark_inode_dirty error paths to fail as soon as possible and return errors to the caller whenever appropriate.

One of the possible scnearios when this bug could affected is that while creating a new inode, its directory entry gets added successfully but while writing the inode itself mark_inode_dirty returns error which is ignored. This would result in inconsistency that the directory entry points to a non-existent inode.

Ran gce-xfstests smoke tests and verified that there were no regressions.

---
 fs/ext4/acl.c       |  2 +-
 fs/ext4/ext4.h      |  2 +-
 fs/ext4/ext4_jbd2.h |  5 ++-
 fs/ext4/extents.c   | 34 +++++++++++---------
 fs/ext4/file.c      | 11 +++++--
 fs/ext4/indirect.c  |  4 ++-
 fs/ext4/inline.c    |  6 ++--
 fs/ext4/inode.c     | 38 ++++++++++++++++-------
 fs/ext4/migrate.c   | 12 ++++---
 fs/ext4/namei.c     | 76 +++++++++++++++++++++++++++++----------------
 fs/ext4/super.c     | 16 ++++++----
 fs/ext4/xattr.c     |  6 ++--
 12 files changed, 139 insertions(+), 73 deletions(-)
fs/ext4/acl.c +1 -1
index 8c7bbf3e566d..470be69f19aa 100644
@@ -256,7 +256,7 @@ ext4_set_acl(struct inode *inode, struct posix_acl *acl, int type)
 	if (!error && update_mode) {
 		inode->i_mode = mode;
 		inode->i_ctime = current_time(inode);
-		ext4_mark_inode_dirty(handle, inode);
+		error = ext4_mark_inode_dirty(handle, inode);
 	}
 out_stop:
 	ext4_journal_stop(handle);
fs/ext4/ext4.h +1 -1
index ad2dbf6e4924..51a85b50033a 100644
@@ -3354,7 +3354,7 @@ struct ext4_extent;
  */
 #define EXT_MAX_BLOCKS	0xffffffff
 
-extern int ext4_ext_tree_init(handle_t *handle, struct inode *);
+extern void ext4_ext_tree_init(handle_t *handle, struct inode *inode);
 extern int ext4_ext_index_trans_blocks(struct inode *inode, int extents);
 extern int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 			       struct ext4_map_blocks *map, int flags);
fs/ext4/ext4_jbd2.h +4 -1
index 4b9002f0e84c..3bacf76d2609 100644
@@ -222,7 +222,10 @@ ext4_mark_iloc_dirty(handle_t *handle,
 int ext4_reserve_inode_write(handle_t *handle, struct inode *inode,
 			struct ext4_iloc *iloc);
 
-int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode);
+#define ext4_mark_inode_dirty(__h, __i)					\
+		__ext4_mark_inode_dirty((__h), (__i), __func__, __LINE__)
+int __ext4_mark_inode_dirty(handle_t *handle, struct inode *inode,
+				const char *func, unsigned int line);
 
 int ext4_expand_extra_isize(struct inode *inode,
 			    unsigned int new_extra_isize,
fs/ext4/extents.c +20 -14
index 2b4b94542e34..5148695d520d 100644
@@ -816,7 +816,7 @@ ext4_ext_binsearch(struct inode *inode,
 
 }
 
-int ext4_ext_tree_init(handle_t *handle, struct inode *inode)
+void ext4_ext_tree_init(handle_t *handle, struct inode *inode)
 {
 	struct ext4_extent_header *eh;
 
@@ -826,7 +826,6 @@ int ext4_ext_tree_init(handle_t *handle, struct inode *inode)
 	eh->eh_magic = EXT4_EXT_MAGIC;
 	eh->eh_max = cpu_to_le16(ext4_ext_space_root(inode, 0));
 	ext4_mark_inode_dirty(handle, inode);
-	return 0;
 }
 
 struct ext4_ext_path *
@@ -1319,7 +1318,7 @@ static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode,
 		  ext4_idx_pblock(EXT_FIRST_INDEX(neh)));
 
 	le16_add_cpu(&neh->eh_depth, 1);
-	ext4_mark_inode_dirty(handle, inode);
+	err = ext4_mark_inode_dirty(handle, inode);
 out:
 	brelse(bh);
 
@@ -4363,7 +4362,7 @@ static int ext4_alloc_file_blocks(struct file *file, ext4_lblk_t offset,
 	struct inode *inode = file_inode(file);
 	handle_t *handle;
 	int ret = 0;
-	int ret2 = 0;
+	int ret2 = 0, ret3 = 0;
 	int retries = 0;
 	int depth = 0;
 	struct ext4_map_blocks map;
@@ -4423,10 +4422,11 @@ static int ext4_alloc_file_blocks(struct file *file, ext4_lblk_t offset,
 			if (ext4_update_inode_size(inode, epos) & 0x1)
 				inode->i_mtime = inode->i_ctime;
 		}
-		ext4_mark_inode_dirty(handle, inode);
+		ret2 = ext4_mark_inode_dirty(handle, inode);
 		ext4_update_inode_fsync_trans(handle, inode, 1);
-		ret2 = ext4_journal_stop(handle);
-		if (ret2)
+		ret3 = ext4_journal_stop(handle);
+		ret2 = ret3 ? ret3 : ret2;
+		if (unlikely(ret2))
 			break;
 	}
 	if (ret == -ENOSPC &&
@@ -4577,7 +4577,9 @@ static long ext4_zero_range(struct file *file, loff_t offset,
 	inode->i_mtime = inode->i_ctime = current_time(inode);
 	if (new_size)
 		ext4_update_inode_size(inode, new_size);
-	ext4_mark_inode_dirty(handle, inode);
+	ret = ext4_mark_inode_dirty(handle, inode);
+	if (unlikely(ret))
+		goto out_handle;
 
 	/* Zero out partial block at the edges of the range */
 	ret = ext4_zero_partial_blocks(handle, inode, offset, len);
@@ -4587,6 +4589,7 @@ static long ext4_zero_range(struct file *file, loff_t offset,
 	if (file->f_flags & O_SYNC)
 		ext4_handle_sync(handle);
 
+out_handle:
 	ext4_journal_stop(handle);
 out_mutex:
 	inode_unlock(inode);
@@ -4700,8 +4703,7 @@ int ext4_convert_unwritten_extents(handle_t *handle, struct inode *inode,
 				   loff_t offset, ssize_t len)
 {
 	unsigned int max_blocks;
-	int ret = 0;
-	int ret2 = 0;
+	int ret = 0, ret2 = 0, ret3 = 0;
 	struct ext4_map_blocks map;
 	unsigned int blkbits = inode->i_blkbits;
 	unsigned int credits = 0;
@@ -4734,9 +4736,13 @@ int ext4_convert_unwritten_extents(handle_t *handle, struct inode *inode,
 				     "ext4_ext_map_blocks returned %d",
 				     inode->i_ino, map.m_lblk,
 				     map.m_len, ret);
-		ext4_mark_inode_dirty(handle, inode);
-		if (credits)
-			ret2 = ext4_journal_stop(handle);
+		ret2 = ext4_mark_inode_dirty(handle, inode);
+		if (credits) {
+			ret3 = ext4_journal_stop(handle);
+			if (unlikely(ret3))
+				ret2 = ret3;
+		}
+
 		if (ret <= 0 || ret2)
 			break;
 	}
@@ -5304,7 +5310,7 @@ static int ext4_collapse_range(struct inode *inode, loff_t offset, loff_t len)
 	if (IS_SYNC(inode))
 		ext4_handle_sync(handle);
 	inode->i_mtime = inode->i_ctime = current_time(inode);
-	ext4_mark_inode_dirty(handle, inode);
+	ret = ext4_mark_inode_dirty(handle, inode);
 	ext4_update_inode_fsync_trans(handle, inode, 1);
 
 out_stop:
fs/ext4/file.c +9 -2
index 0d624250a62b..b8e69f9e3858 100644
@@ -287,6 +287,7 @@ static ssize_t ext4_handle_inode_extension(struct inode *inode, loff_t offset,
 	bool truncate = false;
 	u8 blkbits = inode->i_blkbits;
 	ext4_lblk_t written_blk, end_blk;
+	int ret;
 
 	/*
 	 * Note that EXT4_I(inode)->i_disksize can get extended up to
@@ -327,8 +328,14 @@ static ssize_t ext4_handle_inode_extension(struct inode *inode, loff_t offset,
 		goto truncate;
 	}
 
-	if (ext4_update_inode_size(inode, offset + written))
-		ext4_mark_inode_dirty(handle, inode);
+	if (ext4_update_inode_size(inode, offset + written)) {
+		ret = ext4_mark_inode_dirty(handle, inode);
+		if (unlikely(ret)) {
+			written = ret;
+			ext4_journal_stop(handle);
+			goto truncate;
+		}
+	}
 
 	/*
 	 * We may need to truncate allocated but not written blocks beyond EOF.
fs/ext4/indirect.c +3 -1
index 107f0043f67f..be2b66eb65f7 100644
@@ -467,7 +467,9 @@ static int ext4_splice_branch(handle_t *handle,
 		/*
 		 * OK, we spliced it into the inode itself on a direct block.
 		 */
-		ext4_mark_inode_dirty(handle, ar->inode);
+		err = ext4_mark_inode_dirty(handle, ar->inode);
+		if (unlikely(err))
+			goto err_out;
 		jbd_debug(5, "splicing direct\n");
 	}
 	return err;
fs/ext4/inline.c +4 -2
index f35e289e17aa..c3a1ad2db122 100644
@@ -1260,7 +1260,7 @@ static int ext4_convert_inline_data_nolock(handle_t *handle,
 int ext4_try_add_inline_entry(handle_t *handle, struct ext4_filename *fname,
 			      struct inode *dir, struct inode *inode)
 {
-	int ret, inline_size, no_expand;
+	int ret, ret2, inline_size, no_expand;
 	void *inline_start;
 	struct ext4_iloc iloc;
 
@@ -1314,7 +1314,9 @@ int ext4_try_add_inline_entry(handle_t *handle, struct ext4_filename *fname,
 
 out:
 	ext4_write_unlock_xattr(dir, &no_expand);
-	ext4_mark_inode_dirty(handle, dir);
+	ret2 = ext4_mark_inode_dirty(handle, dir);
+	if (unlikely(ret2 && !ret))
+		ret = ret2;
 	brelse(iloc.bh);
 	return ret;
 }
fs/ext4/inode.c +26 -12
index 2a4aae6acdcb..87430d276bcc 100644
@@ -1296,7 +1296,7 @@ static int ext4_write_end(struct file *file,
 	 * filesystems.
 	 */
 	if (i_size_changed || inline_data)
-		ext4_mark_inode_dirty(handle, inode);
+		ret = ext4_mark_inode_dirty(handle, inode);
 
 	if (pos + len > inode->i_size && !verity && ext4_can_truncate(inode))
 		/* if we have allocated more blocks and copied
@@ -3077,7 +3077,7 @@ static int ext4_da_write_end(struct file *file,
 			 * new_i_size is less that inode->i_size
 			 * bu greater than i_disksize.(hint delalloc)
 			 */
-			ext4_mark_inode_dirty(handle, inode);
+			ret = ext4_mark_inode_dirty(handle, inode);
 		}
 	}
 
@@ -3094,7 +3094,7 @@ static int ext4_da_write_end(struct file *file,
 	if (ret2 < 0)
 		ret = ret2;
 	ret2 = ext4_journal_stop(handle);
-	if (!ret)
+	if (unlikely(ret2 && !ret))
 		ret = ret2;
 
 	return ret ? ret : copied;
@@ -3886,6 +3886,8 @@ int ext4_update_disksize_before_punch(struct inode *inode, loff_t offset,
 				      loff_t len)
 {
 	handle_t *handle;
+	int ret;
+
 	loff_t size = i_size_read(inode);
 
 	WARN_ON(!inode_is_locked(inode));
@@ -3899,10 +3901,10 @@ int ext4_update_disksize_before_punch(struct inode *inode, loff_t offset,
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 	ext4_update_i_disksize(inode, size);
-	ext4_mark_inode_dirty(handle, inode);
+	ret = ext4_mark_inode_dirty(handle, inode);
 	ext4_journal_stop(handle);
 
-	return 0;
+	return ret;
 }
 
 static void ext4_wait_dax_page(struct ext4_inode_info *ei)
@@ -3954,7 +3956,7 @@ int ext4_punch_hole(struct inode *inode, loff_t offset, loff_t length)
 	loff_t first_block_offset, last_block_offset;
 	handle_t *handle;
 	unsigned int credits;
-	int ret = 0;
+	int ret = 0, ret2 = 0;
 
 	trace_ext4_punch_hole(inode, offset, length, 0);
 
@@ -4077,7 +4079,9 @@ int ext4_punch_hole(struct inode *inode, loff_t offset, loff_t length)
 		ext4_handle_sync(handle);
 
 	inode->i_mtime = inode->i_ctime = current_time(inode);
-	ext4_mark_inode_dirty(handle, inode);
+	ret2 = ext4_mark_inode_dirty(handle, inode);
+	if (unlikely(ret2))
+		ret = ret2;
 	if (ret >= 0)
 		ext4_update_inode_fsync_trans(handle, inode, 1);
 out_stop:
@@ -4146,7 +4150,7 @@ int ext4_truncate(struct inode *inode)
 {
 	struct ext4_inode_info *ei = EXT4_I(inode);
 	unsigned int credits;
-	int err = 0;
+	int err = 0, err2;
 	handle_t *handle;
 	struct address_space *mapping = inode->i_mapping;
 
@@ -4234,7 +4238,9 @@ int ext4_truncate(struct inode *inode)
 		ext4_orphan_del(handle, inode);
 
 	inode->i_mtime = inode->i_ctime = current_time(inode);
-	ext4_mark_inode_dirty(handle, inode);
+	err2 = ext4_mark_inode_dirty(handle, inode);
+	if (unlikely(err2 && !err))
+		err = err2;
 	ext4_journal_stop(handle);
 
 	trace_ext4_truncate_exit(inode);
@@ -5292,6 +5298,8 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr)
 			inode->i_gid = attr->ia_gid;
 		error = ext4_mark_inode_dirty(handle, inode);
 		ext4_journal_stop(handle);
+		if (unlikely(error))
+			return error;
 	}
 
 	if (attr->ia_valid & ATTR_SIZE) {
@@ -5777,7 +5785,8 @@ int ext4_expand_extra_isize(struct inode *inode,
  * Whenever the user wants stuff synced (sys_sync, sys_msync, sys_fsync)
  * we start and wait on commits.
  */
-int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode)
+int __ext4_mark_inode_dirty(handle_t *handle, struct inode *inode,
+				const char *func, unsigned int line)
 {
 	struct ext4_iloc iloc;
 	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
@@ -5787,13 +5796,18 @@ int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode)
 	trace_ext4_mark_inode_dirty(inode, _RET_IP_);
 	err = ext4_reserve_inode_write(handle, inode, &iloc);
 	if (err)
-		return err;
+		goto out;
 
 	if (EXT4_I(inode)->i_extra_isize < sbi->s_want_extra_isize)
 		ext4_try_to_expand_extra_isize(inode, sbi->s_want_extra_isize,
 					       iloc, handle);
 
-	return ext4_mark_iloc_dirty(handle, inode, &iloc);
+	err = ext4_mark_iloc_dirty(handle, inode, &iloc);
+out:
+	if (unlikely(err))
+		ext4_error_inode_err(inode, func, line, 0, err,
+					"mark_inode_dirty error");
+	return err;
 }
 
 /*
fs/ext4/migrate.c +8 -4
index fb6520f37135..c5e3fc998211 100644
@@ -287,7 +287,7 @@ static int free_ind_block(handle_t *handle, struct inode *inode, __le32 *i_data)
 static int ext4_ext_swap_inode_data(handle_t *handle, struct inode *inode,
 						struct inode *tmp_inode)
 {
-	int retval;
+	int retval, retval2 = 0;
 	__le32	i_data[3];
 	struct ext4_inode_info *ei = EXT4_I(inode);
 	struct ext4_inode_info *tmp_ei = EXT4_I(tmp_inode);
@@ -342,7 +342,9 @@ static int ext4_ext_swap_inode_data(handle_t *handle, struct inode *inode,
 	 * i_blocks when freeing the indirect meta-data blocks
 	 */
 	retval = free_ind_block(handle, inode, i_data);
-	ext4_mark_inode_dirty(handle, inode);
+	retval2 = ext4_mark_inode_dirty(handle, inode);
+	if (unlikely(retval2 && !retval))
+		retval = retval2;
 
 err_out:
 	return retval;
@@ -601,7 +603,7 @@ int ext4_ind_migrate(struct inode *inode)
 	ext4_lblk_t			start, end;
 	ext4_fsblk_t			blk;
 	handle_t			*handle;
-	int				ret;
+	int				ret, ret2 = 0;
 
 	if (!ext4_has_feature_extents(inode->i_sb) ||
 	    (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)))
@@ -655,7 +657,9 @@ int ext4_ind_migrate(struct inode *inode)
 	memset(ei->i_data, 0, sizeof(ei->i_data));
 	for (i = start; i <= end; i++)
 		ei->i_data[i] = cpu_to_le32(blk++);
-	ext4_mark_inode_dirty(handle, inode);
+	ret2 = ext4_mark_inode_dirty(handle, inode);
+	if (unlikely(ret2 && !ret))
+		ret = ret2;
 errout:
 	ext4_journal_stop(handle);
 	up_write(&EXT4_I(inode)->i_data_sem);
fs/ext4/namei.c +49 -27

192 lines not on this page · raw message

fs/ext4/super.c +10 -6
index bf5fcb477f66..6135e187e3ed 100644
@@ -5885,7 +5885,7 @@ static int ext4_quota_on(struct super_block *sb, int type, int format_id,
 		EXT4_I(inode)->i_flags |= EXT4_NOATIME_FL | EXT4_IMMUTABLE_FL;
 		inode_set_flags(inode, S_NOATIME | S_IMMUTABLE,
 				S_NOATIME | S_IMMUTABLE);
-		ext4_mark_inode_dirty(handle, inode);
+		err = ext4_mark_inode_dirty(handle, inode);
 		ext4_journal_stop(handle);
 	unlock_inode:
 		inode_unlock(inode);
@@ -5987,12 +5987,14 @@ static int ext4_quota_off(struct super_block *sb, int type)
 	 * this is not a hard failure and quotas are already disabled.
 	 */
 	handle = ext4_journal_start(inode, EXT4_HT_QUOTA, 1);
-	if (IS_ERR(handle))
+	if (IS_ERR(handle)) {
+		err = PTR_ERR(handle);
 		goto out_unlock;
+	}
 	EXT4_I(inode)->i_flags &= ~(EXT4_NOATIME_FL | EXT4_IMMUTABLE_FL);
 	inode_set_flags(inode, 0, S_NOATIME | S_IMMUTABLE);
 	inode->i_mtime = inode->i_ctime = current_time(inode);
-	ext4_mark_inode_dirty(handle, inode);
+	err = ext4_mark_inode_dirty(handle, inode);
 	ext4_journal_stop(handle);
 out_unlock:
 	inode_unlock(inode);
@@ -6050,7 +6052,7 @@ static ssize_t ext4_quota_write(struct super_block *sb, int type,
 {
 	struct inode *inode = sb_dqopt(sb)->files[type];
 	ext4_lblk_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
-	int err, offset = off & (sb->s_blocksize - 1);
+	int err = 0, err2 = 0, offset = off & (sb->s_blocksize - 1);
 	int retries = 0;
 	struct buffer_head *bh;
 	handle_t *handle = journal_current_handle();
@@ -6098,9 +6100,11 @@ static ssize_t ext4_quota_write(struct super_block *sb, int type,
 	if (inode->i_size < off + len) {
 		i_size_write(inode, off + len);
 		EXT4_I(inode)->i_disksize = inode->i_size;
-		ext4_mark_inode_dirty(handle, inode);
+		err2 = ext4_mark_inode_dirty(handle, inode);
+		if (unlikely(err2 && !err))
+			err = err2;
 	}
-	return len;
+	return err ? err : len;
 }
 #endif
 
fs/ext4/xattr.c +4 -2
index 21df43a25328..b4847aad5a8f 100644
@@ -1327,7 +1327,7 @@ static int ext4_xattr_inode_write(handle_t *handle, struct inode *ea_inode,
 	int blocksize = ea_inode->i_sb->s_blocksize;
 	int max_blocks = (bufsize + blocksize - 1) >> ea_inode->i_blkbits;
 	int csize, wsize = 0;
-	int ret = 0;
+	int ret = 0, ret2 = 0;
 	int retries = 0;
 
 retry:
@@ -1385,7 +1385,9 @@ static int ext4_xattr_inode_write(handle_t *handle, struct inode *ea_inode,
 	ext4_update_i_disksize(ea_inode, wsize);
 	inode_unlock(ea_inode);
 
-	ext4_mark_inode_dirty(handle, ea_inode);
+	ret2 = ext4_mark_inode_dirty(handle, ea_inode);
+	if (unlikely(ret2 && !ret))
+		ret = ret2;
 
 out:
 	brelse(bh);
Signature · 1 line
2.25.1
author maintainer
#352 lore raw

PATCH 331/388 ext4: don't block for O_DIRECT if IOCB_NOWAIT is set

In reply to Sasha Levin

From: Jens Axboe <[email protected]>

[ Upstream commit 6e014c621e7271649f0d51e54dbe1db4c10486c8 ]

Running with some debug patches to detect illegal blocking triggered the extend/unaligned condition in ext4. If ext4 needs to extend the file (and hence go to buffered IO), or if the app is doing unaligned IO, then ext4 asks the iomap code to wait for IO completion. If the caller asked for no-wait semantics by setting IOCB_NOWAIT, then ext4 should return -EAGAIN instead.

---
 fs/ext4/file.c | 6 ++++++
 1 file changed, 6 insertions(+)
fs/ext4/file.c +6 -0
index b8e69f9e3858..2a01e31a032c 100644
@@ -502,6 +502,12 @@ static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from)
 	if (ret <= 0)
 		return ret;
 
+	/* if we're going to block and IOCB_NOWAIT is set, return -EAGAIN */
+	if ((iocb->ki_flags & IOCB_NOWAIT) && (unaligned_io || extend)) {
+		ret = -EAGAIN;
+		goto out;
+	}
+
 	offset = iocb->ki_pos;
 	count = ret;
 
Signature · 1 line
2.25.1
author maintainer
#353 lore raw

PATCH 332/388 crypto: omap-sham - add proper load balancing support for multicore

In reply to Sasha Levin

From: Tero Kristo <[email protected]>

[ Upstream commit 281c377872ff5d15d80df25fc4df02d2676c7cde ]

The current implementation of the multiple accelerator core support for OMAP SHA does not work properly. It always picks up the first probed accelerator core if this is available, and rest of the book keeping also gets confused if there are two cores available. Add proper load balancing support for SHA, and also fix any bugs related to the multicore support while doing it.

---
 drivers/crypto/omap-sham.c | 64 ++++++++++++++++++--------------------
 1 file changed, 31 insertions(+), 33 deletions(-)
drivers/crypto/omap-sham.c +31 -33
index 0cbf9c932a0f..a82a3596dca3 100644
@@ -169,8 +169,6 @@ struct omap_sham_hmac_ctx {
 };
 
 struct omap_sham_ctx {
-	struct omap_sham_dev	*dd;
-
 	unsigned long		flags;
 
 	/* fallback stuff */
@@ -933,27 +931,35 @@ static int omap_sham_update_dma_stop(struct omap_sham_dev *dd)
 	return 0;
 }
 
+struct omap_sham_dev *omap_sham_find_dev(struct omap_sham_reqctx *ctx)
+{
+	struct omap_sham_dev *dd;
+
+	if (ctx->dd)
+		return ctx->dd;
+
+	spin_lock_bh(&sham.lock);
+	dd = list_first_entry(&sham.dev_list, struct omap_sham_dev, list);
+	list_move_tail(&dd->list, &sham.dev_list);
+	ctx->dd = dd;
+	spin_unlock_bh(&sham.lock);
+
+	return dd;
+}
+
 static int omap_sham_init(struct ahash_request *req)
 {
 	struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
 	struct omap_sham_ctx *tctx = crypto_ahash_ctx(tfm);
 	struct omap_sham_reqctx *ctx = ahash_request_ctx(req);
-	struct omap_sham_dev *dd = NULL, *tmp;
+	struct omap_sham_dev *dd;
 	int bs = 0;
 
-	spin_lock_bh(&sham.lock);
-	if (!tctx->dd) {
-		list_for_each_entry(tmp, &sham.dev_list, list) {
-			dd = tmp;
-			break;
-		}
-		tctx->dd = dd;
-	} else {
-		dd = tctx->dd;
-	}
-	spin_unlock_bh(&sham.lock);
+	ctx->dd = NULL;
 
-	ctx->dd = dd;
+	dd = omap_sham_find_dev(ctx);
+	if (!dd)
+		return -ENODEV;
 
 	ctx->flags = 0;
 
@@ -1223,8 +1229,7 @@ static int omap_sham_handle_queue(struct omap_sham_dev *dd,
 static int omap_sham_enqueue(struct ahash_request *req, unsigned int op)
 {
 	struct omap_sham_reqctx *ctx = ahash_request_ctx(req);
-	struct omap_sham_ctx *tctx = crypto_tfm_ctx(req->base.tfm);
-	struct omap_sham_dev *dd = tctx->dd;
+	struct omap_sham_dev *dd = ctx->dd;
 
 	ctx->op = op;
 
@@ -1234,7 +1239,7 @@ static int omap_sham_enqueue(struct ahash_request *req, unsigned int op)
 static int omap_sham_update(struct ahash_request *req)
 {
 	struct omap_sham_reqctx *ctx = ahash_request_ctx(req);
-	struct omap_sham_dev *dd = ctx->dd;
+	struct omap_sham_dev *dd = omap_sham_find_dev(ctx);
 
 	if (!req->nbytes)
 		return 0;
@@ -1338,21 +1343,8 @@ static int omap_sham_setkey(struct crypto_ahash *tfm, const u8 *key,
 	struct omap_sham_hmac_ctx *bctx = tctx->base;
 	int bs = crypto_shash_blocksize(bctx->shash);
 	int ds = crypto_shash_digestsize(bctx->shash);
-	struct omap_sham_dev *dd = NULL, *tmp;
 	int err, i;
 
-	spin_lock_bh(&sham.lock);
-	if (!tctx->dd) {
-		list_for_each_entry(tmp, &sham.dev_list, list) {
-			dd = tmp;
-			break;
-		}
-		tctx->dd = dd;
-	} else {
-		dd = tctx->dd;
-	}
-	spin_unlock_bh(&sham.lock);
-
 	err = crypto_shash_setkey(tctx->fallback, key, keylen);
 	if (err)
 		return err;
@@ -1370,7 +1362,7 @@ static int omap_sham_setkey(struct crypto_ahash *tfm, const u8 *key,
 
 	memset(bctx->ipad + keylen, 0, bs - keylen);
 
-	if (!test_bit(FLAGS_AUTO_XOR, &dd->flags)) {
+	if (!test_bit(FLAGS_AUTO_XOR, &sham.flags)) {
 		memcpy(bctx->opad, bctx->ipad, bs);
 
 		for (i = 0; i < bs; i++) {
@@ -2174,6 +2166,7 @@ static int omap_sham_probe(struct platform_device *pdev)
 	}
 
 	dd->flags |= dd->pdata->flags;
+	sham.flags |= dd->pdata->flags;
 
 	pm_runtime_use_autosuspend(dev);
 	pm_runtime_set_autosuspend_delay(dev, DEFAULT_AUTOSUSPEND_DELAY);
@@ -2201,6 +2194,9 @@ static int omap_sham_probe(struct platform_device *pdev)
 	spin_unlock(&sham.lock);
 
 	for (i = 0; i < dd->pdata->algs_info_size; i++) {
+		if (dd->pdata->algs_info[i].registered)
+			break;
+
 		for (j = 0; j < dd->pdata->algs_info[i].size; j++) {
 			struct ahash_alg *alg;
 
@@ -2252,9 +2248,11 @@ static int omap_sham_remove(struct platform_device *pdev)
 	list_del(&dd->list);
 	spin_unlock(&sham.lock);
 	for (i = dd->pdata->algs_info_size - 1; i >= 0; i--)
-		for (j = dd->pdata->algs_info[i].registered - 1; j >= 0; j--)
+		for (j = dd->pdata->algs_info[i].registered - 1; j >= 0; j--) {
 			crypto_unregister_ahash(
 					&dd->pdata->algs_info[i].algs_list[j]);
+			dd->pdata->algs_info[i].registered--;
+		}
 	tasklet_kill(&dd->done_task);
 	pm_runtime_disable(&pdev->dev);
 
Signature · 1 line
2.25.1
author maintainer
#354 lore raw

PATCH 333/388 pwm: Add missing "CONFIG_" prefix

In reply to Sasha Levin

From: Kees Cook <[email protected]>

[ Upstream commit f5641d053d46a9a18fe13f2ecb4a7b4a66d9cdf7 ]

The IS_ENABLED() use was missing the CONFIG_ prefix which would have lead to skipping this code.

Fixes 3ad1f3a33286 ("pwm: Implement some checks for lowlevel drivers")
---
 drivers/pwm/core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/pwm/core.c +1 -1
index 9973c442b455..6b3cbc0490c6 100644
@@ -121,7 +121,7 @@ static int pwm_device_request(struct pwm_device *pwm, const char *label)
 		pwm->chip->ops->get_state(pwm->chip, pwm, &pwm->state);
 		trace_pwm_get(pwm, &pwm->state);
 
-		if (IS_ENABLED(PWM_DEBUG))
+		if (IS_ENABLED(CONFIG_PWM_DEBUG))
 			pwm->last = pwm->state;
 	}
 
Signature · 1 line
2.25.1
author maintainer
#355 lore raw

PATCH 334/388 bpf: Fix an error code in check_btf_func()

In reply to Sasha Levin

From: Dan Carpenter <[email protected]>

[ Upstream commit e7ed83d6fa1a00d0f2ad0327e73d3ea9e7ea8de1 ]

This code returns success if the "info_aux" allocation fails but it should return -ENOMEM.

Fixes 8c1b6e69dcc1 ("bpf: Compare BTF types of functions arguments with actual types")
---
 kernel/bpf/verifier.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/bpf/verifier.c +1 -1
index efe14cf24bc6..739d9ba3ba6b 100644
@@ -7366,7 +7366,7 @@ static int check_btf_func(struct bpf_verifier_env *env,
 	const struct btf *btf;
 	void __user *urecord;
 	u32 prev_offset = 0;
-	int ret = 0;
+	int ret = -ENOMEM;
 
 	nfuncs = attr->func_info_cnt;
 	if (!nfuncs)
Signature · 1 line
2.25.1
author maintainer
#356 lore raw

PATCH 335/388 geneve: change from tx_error to tx_dropped on missing metadata

In reply to Sasha Levin

From: Jiri Benc <[email protected]>

[ Upstream commit 9d149045b3c0e44c049cdbce8a64e19415290017 ]

If the geneve interface is in collect_md (external) mode, it can't send any packets submitted directly to its net interface, as such packets won't have metadata attached. This is expected.

However, the kernel itself sends some packets to the interface, most notably, IPv6 DAD, IPv6 multicast listener reports, etc. This is not wrong, as tunnel metadata can be specified in routing table (although technically, that has never worked for IPv6, but hopefully will be fixed eventually) and then the interface must correctly participate in IPv6 housekeeping.

The problem is that any such attempt increases the tx_error counter. Just bringing up a geneve interface with IPv6 enabled is enough to see a number of tx_errors. That causes confusion among users, prompting them to find a network error where there is none.

Change the counter used to tx_dropped. That better conveys the meaning (there's nothing wrong going on, just some packets are getting dropped) and hopefully will make admins panic less.

---
 drivers/net/geneve.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/net/geneve.c +4 -3
index 6b461be1820b..75266580b586 100644
@@ -987,9 +987,10 @@ static netdev_tx_t geneve_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (geneve->collect_md) {
 		info = skb_tunnel_info(skb);
 		if (unlikely(!info || !(info->mode & IP_TUNNEL_INFO_TX))) {
-			err = -EINVAL;
 			netdev_dbg(dev, "no tunnel metadata\n");
-			goto tx_error;
+			dev_kfree_skb(skb);
+			dev->stats.tx_dropped++;
+			return NETDEV_TX_OK;
 		}
 	} else {
 		info = &geneve->info;
@@ -1006,7 +1007,7 @@ static netdev_tx_t geneve_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	if (likely(!err))
 		return NETDEV_TX_OK;
-tx_error:
+
 	dev_kfree_skb(skb);
 
 	if (err == -ELOOP)
Signature · 1 line
2.25.1
author maintainer
#357 lore raw

PATCH 336/388 lib/zlib: remove outdated and incorrect pre-increment optimization

In reply to Sasha Levin

From: Jann Horn <[email protected]>

[ Upstream commit acaab7335bd6f0c0b54ce3a00bd7f18222ce0f5f ]

The zlib inflate code has an old micro-optimization based on the assumption that for pre-increment memory accesses, the compiler will generate code that fits better into the processor's pipeline than what would be generated for post-increment memory accesses.

This optimization was already removed in upstream zlib in 2016:
https://github.com/madler/zlib/commit/9aaec95e8211

This optimization causes UB according to C99, which says in section 6.5.6 "Additive operators": "If both the pointer operand and the result point to elements of the same array object, or one past the last element of the array object, the evaluation shall not produce an overflow; otherwise, the behavior is undefined".

This UB is not only a theoretical concern, but can also cause trouble for future work on compiler-based sanitizers.

According to the zlib commit, this optimization also is not optimal anymore with modern compilers.

Replace uses of OFF, PUP and UP_UNALIGNED with their definitions in the POSTINC case, and remove the macro definitions, just like in the upstream patch.

---
 lib/zlib_inflate/inffast.c | 91 +++++++++++++++-----------------------
 1 file changed, 35 insertions(+), 56 deletions(-)
lib/zlib_inflate/inffast.c +35 -56
index 2c13ecc5bb2c..ed1f3df27260 100644
@@ -10,17 +10,6 @@
 
 #ifndef ASMINF
 
-/* Allow machine dependent optimization for post-increment or pre-increment.
-   Based on testing to date,
-   Pre-increment preferred for:
-   - PowerPC G3 (Adler)
-   - MIPS R5000 (Randers-Pehrson)
-   Post-increment preferred for:
-   - none
-   No measurable difference:
-   - Pentium III (Anderson)
-   - M68060 (Nikl)
- */
 union uu {
 	unsigned short us;
 	unsigned char b[2];
@@ -38,16 +27,6 @@ get_unaligned16(const unsigned short *p)
 	return mm.us;
 }
 
-#ifdef POSTINC
-#  define OFF 0
-#  define PUP(a) *(a)++
-#  define UP_UNALIGNED(a) get_unaligned16((a)++)
-#else
-#  define OFF 1
-#  define PUP(a) *++(a)
-#  define UP_UNALIGNED(a) get_unaligned16(++(a))
-#endif
-
 /*
    Decode literal, length, and distance codes and write out the resulting
    literal and match bytes until either not enough input or output is
@@ -115,9 +94,9 @@ void inflate_fast(z_streamp strm, unsigned start)
 
     /* copy state to local variables */
     state = (struct inflate_state *)strm->state;
-    in = strm->next_in - OFF;
+    in = strm->next_in;
     last = in + (strm->avail_in - 5);
-    out = strm->next_out - OFF;
+    out = strm->next_out;
     beg = out - (start - strm->avail_out);
     end = out + (strm->avail_out - 257);
 #ifdef INFLATE_STRICT
@@ -138,9 +117,9 @@ void inflate_fast(z_streamp strm, unsigned start)
        input data or output space */
     do {
         if (bits < 15) {
-            hold += (unsigned long)(PUP(in)) << bits;
+            hold += (unsigned long)(*in++) << bits;
             bits += 8;
-            hold += (unsigned long)(PUP(in)) << bits;
+            hold += (unsigned long)(*in++) << bits;
             bits += 8;
         }
         this = lcode[hold & lmask];
@@ -150,14 +129,14 @@ void inflate_fast(z_streamp strm, unsigned start)
         bits -= op;
         op = (unsigned)(this.op);
         if (op == 0) {                          /* literal */
-            PUP(out) = (unsigned char)(this.val);
+            *out++ = (unsigned char)(this.val);
         }
         else if (op & 16) {                     /* length base */
             len = (unsigned)(this.val);
             op &= 15;                           /* number of extra bits */
             if (op) {
                 if (bits < op) {
-                    hold += (unsigned long)(PUP(in)) << bits;
+                    hold += (unsigned long)(*in++) << bits;
                     bits += 8;
                 }
                 len += (unsigned)hold & ((1U << op) - 1);
@@ -165,9 +144,9 @@ void inflate_fast(z_streamp strm, unsigned start)
                 bits -= op;
             }
             if (bits < 15) {
-                hold += (unsigned long)(PUP(in)) << bits;
+                hold += (unsigned long)(*in++) << bits;
                 bits += 8;
-                hold += (unsigned long)(PUP(in)) << bits;
+                hold += (unsigned long)(*in++) << bits;
                 bits += 8;
             }
             this = dcode[hold & dmask];
@@ -180,10 +159,10 @@ void inflate_fast(z_streamp strm, unsigned start)
                 dist = (unsigned)(this.val);
                 op &= 15;                       /* number of extra bits */
                 if (bits < op) {
-                    hold += (unsigned long)(PUP(in)) << bits;
+                    hold += (unsigned long)(*in++) << bits;
                     bits += 8;
                     if (bits < op) {
-                        hold += (unsigned long)(PUP(in)) << bits;
+                        hold += (unsigned long)(*in++) << bits;
                         bits += 8;
                     }
                 }
@@ -205,13 +184,13 @@ void inflate_fast(z_streamp strm, unsigned start)
                         state->mode = BAD;
                         break;
                     }
-                    from = window - OFF;
+                    from = window;
                     if (write == 0) {           /* very common case */
                         from += wsize - op;
                         if (op < len) {         /* some from window */
                             len -= op;
                             do {
-                                PUP(out) = PUP(from);
+                                *out++ = *from++;
                             } while (--op);
                             from = out - dist;  /* rest from output */
                         }
@@ -222,14 +201,14 @@ void inflate_fast(z_streamp strm, unsigned start)
                         if (op < len) {         /* some from end of window */
                             len -= op;
                             do {
-                                PUP(out) = PUP(from);
+                                *out++ = *from++;
                             } while (--op);
-                            from = window - OFF;
+                            from = window;
                             if (write < len) {  /* some from start of window */
                                 op = write;
                                 len -= op;
                                 do {
-                                    PUP(out) = PUP(from);
+                                    *out++ = *from++;
                                 } while (--op);
                                 from = out - dist;      /* rest from output */
                             }
@@ -240,21 +219,21 @@ void inflate_fast(z_streamp strm, unsigned start)
                         if (op < len) {         /* some from window */
                             len -= op;
                             do {
-                                PUP(out) = PUP(from);
+                                *out++ = *from++;
                             } while (--op);
                             from = out - dist;  /* rest from output */
                         }
                     }
                     while (len > 2) {
-                        PUP(out) = PUP(from);
-                        PUP(out) = PUP(from);
-                        PUP(out) = PUP(from);
+                        *out++ = *from++;
+                        *out++ = *from++;
+                        *out++ = *from++;
                         len -= 3;
                     }
                     if (len) {
-                        PUP(out) = PUP(from);
+                        *out++ = *from++;
                         if (len > 1)
-                            PUP(out) = PUP(from);
+                            *out++ = *from++;
                     }
                 }
                 else {
@@ -264,29 +243,29 @@ void inflate_fast(z_streamp strm, unsigned start)
                     from = out - dist;          /* copy direct from output */
 		    /* minimum length is three */
 		    /* Align out addr */
-		    if (!((long)(out - 1 + OFF) & 1)) {
-			PUP(out) = PUP(from);
+		    if (!((long)(out - 1) & 1)) {
+			*out++ = *from++;
 			len--;
 		    }
-		    sout = (unsigned short *)(out - OFF);
+		    sout = (unsigned short *)(out);
 		    if (dist > 2) {
 			unsigned short *sfrom;
 
-			sfrom = (unsigned short *)(from - OFF);
+			sfrom = (unsigned short *)(from);
 			loops = len >> 1;
 			do
 #ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
-			    PUP(sout) = PUP(sfrom);
+			    *sout++ = *sfrom++;
 #else
-			    PUP(sout) = UP_UNALIGNED(sfrom);
+			    *sout++ = get_unaligned16(sfrom++);
 #endif
 			while (--loops);
-			out = (unsigned char *)sout + OFF;
-			from = (unsigned char *)sfrom + OFF;
+			out = (unsigned char *)sout;
+			from = (unsigned char *)sfrom;
 		    } else { /* dist == 1 or dist == 2 */
 			unsigned short pat16;
 
-			pat16 = *(sout-1+OFF);
+			pat16 = *(sout-1);
 			if (dist == 1) {
 				union uu mm;
 				/* copy one char pattern to both bytes */
@@ -296,12 +275,12 @@ void inflate_fast(z_streamp strm, unsigned start)
 			}
 			loops = len >> 1;
 			do
-			    PUP(sout) = pat16;
+			    *sout++ = pat16;
 			while (--loops);
-			out = (unsigned char *)sout + OFF;
+			out = (unsigned char *)sout;
 		    }
 		    if (len & 1)
-			PUP(out) = PUP(from);
+			*out++ = *from++;
                 }
             }
             else if ((op & 64) == 0) {          /* 2nd level distance code */
@@ -336,8 +315,8 @@ void inflate_fast(z_streamp strm, unsigned start)
     hold &= (1U << bits) - 1;
 
     /* update state and return */
-    strm->next_in = in + OFF;
-    strm->next_out = out + OFF;
+    strm->next_in = in;
+    strm->next_out = out;
     strm->avail_in = (unsigned)(in < last ? 5 + (last - in) : 5 - (in - last));
     strm->avail_out = (unsigned)(out < end ?
                                  257 + (end - out) : 257 - (out - end));
Signature · 1 line
2.25.1
author maintainer
#358 lore raw

PATCH 337/388 include/linux/bitops.h: avoid clang shift-count-overflow warnings

In reply to Sasha Levin

From: Arnd Bergmann <[email protected]>

[ Upstream commit bd93f003b7462ae39a43c531abca37fe7073b866 ]

Clang normally does not warn about certain issues in inline functions when it only happens in an eliminated code path. However if something else goes wrong, it does tend to complain about the definition of hweight_long() on 32-bit targets:

  include/linux/bitops.h:75:41: error: shift count >= width of type [-Werror,-Wshift-count-overflow]
          return sizeof(w) == 4 ? hweight32(w) : hweight64(w);
                                                 ^~~~~~~~~~~~
  include/asm-generic/bitops/const_hweight.h:29:49: note: expanded from macro 'hweight64'
   define hweight64(w) (__builtin_constant_p(w) ? __const_hweight64(w) : __arch_hweight64(w))
                                                  ^~~~~~~~~~~~~~~~~~~~
  include/asm-generic/bitops/const_hweight.h:21:76: note: expanded from macro '__const_hweight64'
   define __const_hweight64(w) (__const_hweight32(w) + __const_hweight32((w) >> 32))
                                                                             ^  ~~
  include/asm-generic/bitops/const_hweight.h:20:49: note: expanded from macro '__const_hweight32'
   define __const_hweight32(w) (__const_hweight16(w) + __const_hweight16((w) >> 16))
                                                  ^
  include/asm-generic/bitops/const_hweight.h:19:72: note: expanded from macro '__const_hweight16'
   define __const_hweight16(w) (__const_hweight8(w)  + __const_hweight8((w)  >> 8 ))
                                                                         ^
  include/asm-generic/bitops/const_hweight.h:12:9: note: expanded from macro '__const_hweight8'
            (!!((w) & (1ULL << 2))) +     \

Adding an explicit cast to __u64 avoids that warning and makes it easier to read other output.

---
 include/linux/bitops.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/bitops.h +1 -1
index 9acf654f0b19..99f2ac30b1d9 100644
@@ -72,7 +72,7 @@ static inline int get_bitmask_order(unsigned int count)
 
 static __always_inline unsigned long hweight_long(unsigned long w)
 {
-	return sizeof(w) == 4 ? hweight32(w) : hweight64(w);
+	return sizeof(w) == 4 ? hweight32(w) : hweight64((__u64)w);
 }
 
 /**
Signature · 1 line
2.25.1
author maintainer
#359 lore raw

PATCH 338/388 selftests/vm/pkeys: fix alloc_random_pkey() to make it really random

In reply to Sasha Levin

From: Ram Pai <[email protected]>

[ Upstream commit 6e373263ce07eeaa6410843179535fbdf561fc31 ]

alloc_random_pkey() was allocating the same pkey every time. Not all pkeys were geting tested. This fixes it.

---
 tools/testing/selftests/x86/protection_keys.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
tools/testing/selftests/x86/protection_keys.c +2 -1
index 480995bceefa..47191af46617 100644
@@ -24,6 +24,7 @@
 #define _GNU_SOURCE
 #include <errno.h>
 #include <linux/futex.h>
+#include <time.h>
 #include <sys/time.h>
 #include <sys/syscall.h>
 #include <string.h>
@@ -612,10 +613,10 @@ int alloc_random_pkey(void)
 	int nr_alloced = 0;
 	int random_index;
 	memset(alloced_pkeys, 0, sizeof(alloced_pkeys));
+	srand((unsigned int)time(NULL));
 
 	/* allocate every possible key and make a note of which ones we got */
 	max_nr_pkey_allocs = NR_PKEYS;
-	max_nr_pkey_allocs = 1;
 	for (i = 0; i < max_nr_pkey_allocs; i++) {
 		int new_pkey = alloc_pkey();
 		if (new_pkey < 0)
Signature · 1 line
2.25.1
author maintainer
#360 lore raw

PATCH 339/388 blktrace: use errno instead of bi_status

In reply to Sasha Levin

From: Chaitanya Kulkarni <[email protected]>

[ Upstream commit 48bc3cd3e07a1486f45d9971c75d6090976c3b1b ]

In blk_add_trace_spliti() blk_add_trace_bio_remap() use blk_status_to_errno() to pass the error instead of pasing the bi_status. This fixes the sparse warning.

---
 kernel/trace/blktrace.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
kernel/trace/blktrace.c +6 -3
index ca39dc3230cb..c6d59a457f50 100644
@@ -995,8 +995,10 @@ static void blk_add_trace_split(void *ignore,
 
 		__blk_add_trace(bt, bio->bi_iter.bi_sector,
 				bio->bi_iter.bi_size, bio_op(bio), bio->bi_opf,
-				BLK_TA_SPLIT, bio->bi_status, sizeof(rpdu),
-				&rpdu, blk_trace_bio_get_cgid(q, bio));
+				BLK_TA_SPLIT,
+				blk_status_to_errno(bio->bi_status),
+				sizeof(rpdu), &rpdu,
+				blk_trace_bio_get_cgid(q, bio));
 	}
 	rcu_read_unlock();
 }
@@ -1033,7 +1035,8 @@ static void blk_add_trace_bio_remap(void *ignore,
 	r.sector_from = cpu_to_be64(from);
 
 	__blk_add_trace(bt, bio->bi_iter.bi_sector, bio->bi_iter.bi_size,
-			bio_op(bio), bio->bi_opf, BLK_TA_REMAP, bio->bi_status,
+			bio_op(bio), bio->bi_opf, BLK_TA_REMAP,
+			blk_status_to_errno(bio->bi_status),
 			sizeof(r), &r, blk_trace_bio_get_cgid(q, bio));
 	rcu_read_unlock();
 }
Signature · 1 line
2.25.1
author maintainer
#361 lore raw

PATCH 340/388 blktrace: fix endianness in get_pdu_int()

In reply to Sasha Levin

From: Chaitanya Kulkarni <[email protected]>

[ Upstream commit 71df3fd82e7cccec7b749a8607a4662d9f7febdd ]

In function get_pdu_len() replace variable type from __u64 to __be64. This fixes sparse warning.

---
 kernel/trace/blktrace.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/trace/blktrace.c +1 -1
index c6d59a457f50..cba2093edee2 100644
@@ -1256,7 +1256,7 @@ static inline __u16 t_error(const struct trace_entry *ent)
 
 static __u64 get_pdu_int(const struct trace_entry *ent, bool has_cg)
 {
-	const __u64 *val = pdu_start(ent, has_cg);
+	const __be64 *val = pdu_start(ent, has_cg);
 	return be64_to_cpu(*val);
 }
 
Signature · 1 line
2.25.1
author maintainer
#362 lore raw

PATCH 341/388 blktrace: fix endianness for blk_log_remap()

In reply to Sasha Levin

From: Chaitanya Kulkarni <[email protected]>

[ Upstream commit 5aec598c456fe3c1b71a1202cbb42bdc2a643277 ]

The function blk_log_remap() can be simplified by removing the call to get_pdu_remap() that copies the values into extra variable to print the data, which also fixes the endiannness warning reported by sparse.

---
 kernel/trace/blktrace.c | 19 ++++---------------
 1 file changed, 4 insertions(+), 15 deletions(-)
kernel/trace/blktrace.c +4 -15
index cba2093edee2..35610a4be4a9 100644
@@ -1260,17 +1260,6 @@ static __u64 get_pdu_int(const struct trace_entry *ent, bool has_cg)
 	return be64_to_cpu(*val);
 }
 
-static void get_pdu_remap(const struct trace_entry *ent,
-			  struct blk_io_trace_remap *r, bool has_cg)
-{
-	const struct blk_io_trace_remap *__r = pdu_start(ent, has_cg);
-	__u64 sector_from = __r->sector_from;
-
-	r->device_from = be32_to_cpu(__r->device_from);
-	r->device_to   = be32_to_cpu(__r->device_to);
-	r->sector_from = be64_to_cpu(sector_from);
-}
-
 typedef void (blk_log_action_t) (struct trace_iterator *iter, const char *act,
 	bool has_cg);
 
@@ -1410,13 +1399,13 @@ static void blk_log_with_error(struct trace_seq *s,
 
 static void blk_log_remap(struct trace_seq *s, const struct trace_entry *ent, bool has_cg)
 {
-	struct blk_io_trace_remap r = { .device_from = 0, };
+	const struct blk_io_trace_remap *__r = pdu_start(ent, has_cg);
 
-	get_pdu_remap(ent, &r, has_cg);
 	trace_seq_printf(s, "%llu + %u <- (%d,%d) %llu\n",
 			 t_sector(ent), t_sec(ent),
-			 MAJOR(r.device_from), MINOR(r.device_from),
-			 (unsigned long long)r.sector_from);
+			 MAJOR(be32_to_cpu(__r->device_from)),
+			 MINOR(be32_to_cpu(__r->device_from)),
+			 be64_to_cpu(__r->sector_from));
 }
 
 static void blk_log_plug(struct trace_seq *s, const struct trace_entry *ent, bool has_cg)
Signature · 1 line
2.25.1
author maintainer
#363 lore raw

PATCH 342/388 KVM: selftests: Fix build with "make ARCH=x86_64"

In reply to Sasha Levin

From: Vitaly Kuznetsov <[email protected]>

[ Upstream commit b80db73dc8be7022adae1b4414a1bebce50fe915 ]

Marcelo reports that kvm selftests fail to build with "make ARCH=x86_64":

gcc -Wall -Wstrict-prototypes -Wuninitialized -O2 -g -std=gnu99
-fno-stack-protector -fno-PIE -I../../../../tools/include -I../../../../tools/arch/x86_64/include -I../../../../usr/include/ -Iinclude -Ilib -Iinclude/x86_64 -I.. -c lib/kvm_util.c -o /var/tmp/20200604202744-bin/lib/kvm_util.o

In file included from lib/kvm_util.c:11:
include/x86_64/processor.h:14:10: fatal error: asm/msr-index.h: No such
file or directory

 #include <asm/msr-index.h>
          ^~~~~~~~~~~~~~~~~

compilation terminated.

"make ARCH=x86", however, works. The problem is that arch specific headers for x86_64 live in 'tools/arch/x86/include', not in 'tools/arch/x86_64/include'.

Fixes 66d69e081b52 ("selftests: fix kvm relocatable native/cross builds and installs")

Message-Id: <[email protected]>

---
 tools/testing/selftests/kvm/Makefile | 4 ++++
 1 file changed, 4 insertions(+)
tools/testing/selftests/kvm/Makefile +4 -0
index 42f4f49f2a48..2c85b9dd86f5 100644
@@ -80,7 +80,11 @@ LIBKVM += $(LIBKVM_$(UNAME_M))
 INSTALL_HDR_PATH = $(top_srcdir)/usr
 LINUX_HDR_PATH = $(INSTALL_HDR_PATH)/include/
 LINUX_TOOL_INCLUDE = $(top_srcdir)/tools/include
+ifeq ($(ARCH),x86_64)
+LINUX_TOOL_ARCH_INCLUDE = $(top_srcdir)/tools/arch/x86/include
+else
 LINUX_TOOL_ARCH_INCLUDE = $(top_srcdir)/tools/arch/$(ARCH)/include
+endif
 CFLAGS += -Wall -Wstrict-prototypes -Wuninitialized -O2 -g -std=gnu99 \
 	-fno-stack-protector -fno-PIE -I$(LINUX_TOOL_INCLUDE) \
 	-I$(LINUX_TOOL_ARCH_INCLUDE) -I$(LINUX_HDR_PATH) -Iinclude \
Signature · 1 line
2.25.1
author maintainer
#364 lore raw

PATCH 343/388 gfs2: fix use-after-free on transaction ail lists

In reply to Sasha Levin

From: Bob Peterson <[email protected]>

[ Upstream commit 83d060ca8d90fa1e3feac227f995c013100862d3 ]

Before this patch, transactions could be merged into the system transaction by function gfs2_merge_trans(), but the transaction ail lists were never merged. Because the ail flushing mechanism can run separately, bd elements can be attached to the transaction's buffer list during the transaction (trans_add_meta, etc) but quickly moved to its ail lists. Later, in function gfs2_trans_end, the transaction can be freed (by gfs2_trans_end) while it still has bd elements queued to its ail lists, which can cause it to either lose track of the bd elements altogether (memory leak) or worse, reference the bd elements after the parent transaction has been freed.

Although I've not seen any serious consequences, the problem becomes apparent with the previous patch's addition of:

	gfs2_assert_warn(sdp, list_empty(&tr->tr_ail1_list));

to function gfs2_trans_free().

This patch adds logic into gfs2_merge_trans() to move the merged transaction's ail lists to the sdp transaction. This prevents the use-after-free. To do this properly, we need to hold the ail lock, so we pass sdp into the function instead of the transaction itself.

---
 fs/gfs2/log.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)
fs/gfs2/log.c +9 -2
index 0644e58c6191..b7a5221bea7d 100644
@@ -1003,8 +1003,10 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl, u32 flags)
  * @new: New transaction to be merged
  */
 
-static void gfs2_merge_trans(struct gfs2_trans *old, struct gfs2_trans *new)
+static void gfs2_merge_trans(struct gfs2_sbd *sdp, struct gfs2_trans *new)
 {
+	struct gfs2_trans *old = sdp->sd_log_tr;
+
 	WARN_ON_ONCE(!test_bit(TR_ATTACHED, &old->tr_flags));
 
 	old->tr_num_buf_new	+= new->tr_num_buf_new;
@@ -1016,6 +1018,11 @@ static void gfs2_merge_trans(struct gfs2_trans *old, struct gfs2_trans *new)
 
 	list_splice_tail_init(&new->tr_databuf, &old->tr_databuf);
 	list_splice_tail_init(&new->tr_buf, &old->tr_buf);
+
+	spin_lock(&sdp->sd_ail_lock);
+	list_splice_tail_init(&new->tr_ail1_list, &old->tr_ail1_list);
+	list_splice_tail_init(&new->tr_ail2_list, &old->tr_ail2_list);
+	spin_unlock(&sdp->sd_ail_lock);
 }
 
 static void log_refund(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
@@ -1027,7 +1034,7 @@ static void log_refund(struct gfs2_sbd *sdp, struct gfs2_trans *tr)
 	gfs2_log_lock(sdp);
 
 	if (sdp->sd_log_tr) {
-		gfs2_merge_trans(sdp->sd_log_tr, tr);
+		gfs2_merge_trans(sdp, tr);
 	} else if (tr->tr_num_buf_new || tr->tr_num_databuf_new) {
 		gfs2_assert_withdraw(sdp, test_bit(TR_ALLOCED, &tr->tr_flags));
 		sdp->sd_log_tr = tr;
Signature · 1 line
2.25.1
author maintainer
#365 lore raw

PATCH 344/388 net: dp83867: Fix OF_MDIO config check

In reply to Sasha Levin

From: Dan Murphy <[email protected]>

[ Upstream commit 506de00677b84dfc6718cbbd3495b1d90df5d098 ]

When CONFIG_OF_MDIO is set to be a module the code block is not compiled. Use the IS_ENABLED macro that checks for both built in as well as module.

Fixes 2a10154abcb75 ("net: phy: dp83867: Add TI dp83867 phy")
---
 drivers/net/phy/dp83867.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/phy/dp83867.c +1 -1
index b55e3c0403ed..ddac79960ea7 100644
@@ -488,7 +488,7 @@ static int dp83867_verify_rgmii_cfg(struct phy_device *phydev)
 	return 0;
 }
 
-#ifdef CONFIG_OF_MDIO
+#if IS_ENABLED(CONFIG_OF_MDIO)
 static int dp83867_of_init(struct phy_device *phydev)
 {
 	struct dp83867_private *dp83867 = phydev->priv;
Signature · 1 line
2.25.1
author maintainer
#366 lore raw

PATCH 345/388 net: marvell: Fix OF_MDIO config check

In reply to Sasha Levin

From: Dan Murphy <[email protected]>

[ Upstream commit 5cd119d9a05f1c1a08778a7305b4ca0f16bc1e20 ]

When CONFIG_OF_MDIO is set to be a module the code block is not compiled. Use the IS_ENABLED macro that checks for both built in as well as module.

Fixes cf41a51db8985 ("of/phylib: Use device tree properties to initialize Marvell PHYs.")
---
 drivers/net/phy/marvell.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/phy/marvell.c +1 -1
index 7fc8e10c5f33..a435f7352cfb 100644
@@ -337,7 +337,7 @@ static int m88e1101_config_aneg(struct phy_device *phydev)
 	return marvell_config_aneg(phydev);
 }
 
-#ifdef CONFIG_OF_MDIO
+#if IS_ENABLED(CONFIG_OF_MDIO)
 /* Set and/or override some configuration registers based on the
  * marvell,reg-init property stored in the of_node for the phydev.
  *
Signature · 1 line
2.25.1
author maintainer
#367 lore raw

PATCH 346/388 net: mscc: Fix OF_MDIO config check

In reply to Sasha Levin

From: Dan Murphy <[email protected]>

[ Upstream commit ae602786407fef34e1d66b3c8f278a10ed37197e ]

When CONFIG_OF_MDIO is set to be a module the code block is not compiled. Use the IS_ENABLED macro that checks for both built in as well as module.

Fixes 4f58e6dceb0e4 ("net: phy: Cleanup the Edge-Rate feature in Microsemi PHYs.")
---
 drivers/net/phy/mscc/mscc.h      | 2 +-
 drivers/net/phy/mscc/mscc_main.c | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)
drivers/net/phy/mscc/mscc.h +1 -1
index 414e3b31bb1f..132f9bf49198 100644
@@ -375,7 +375,7 @@ struct vsc8531_private {
 #endif
 };
 
-#ifdef CONFIG_OF_MDIO
+#if IS_ENABLED(CONFIG_OF_MDIO)
 struct vsc8531_edge_rate_table {
 	u32 vddmac;
 	u32 slowdown[8];
drivers/net/phy/mscc/mscc_main.c +2 -2
index c8aa6d905d8e..485a4f8a6a9a 100644
@@ -98,7 +98,7 @@ static const struct vsc85xx_hw_stat vsc8584_hw_stats[] = {
 	},
 };
 
-#ifdef CONFIG_OF_MDIO
+#if IS_ENABLED(CONFIG_OF_MDIO)
 static const struct vsc8531_edge_rate_table edge_table[] = {
 	{MSCC_VDDMAC_3300, { 0, 2,  4,  7, 10, 17, 29, 53} },
 	{MSCC_VDDMAC_2500, { 0, 3,  6, 10, 14, 23, 37, 63} },
@@ -382,7 +382,7 @@ static void vsc85xx_wol_get(struct phy_device *phydev,
 	mutex_unlock(&phydev->lock);
 }
 
-#ifdef CONFIG_OF_MDIO
+#if IS_ENABLED(CONFIG_OF_MDIO)
 static int vsc85xx_edge_rate_magic_get(struct phy_device *phydev)
 {
 	u32 vdd, sd;
Signature · 1 line
2.25.1
author maintainer
#368 lore raw

PATCH 347/388 ntb_perf: pass correct struct device to dma_alloc_coherent

In reply to Sasha Levin

From: Sanjay R Mehta <[email protected]>

[ Upstream commit 98f4e140264eeb52f22ff05be6b6dd48237255ac ]

Currently, ntb->dev is passed to dma_alloc_coherent and dma_free_coherent calls. The returned dma_addr_t is the CPU physical address. This works fine as long as IOMMU is disabled. But when IOMMU is enabled, we need to make sure that IOVA is returned for dma_addr_t. So the correct way to achieve this is by changing the first parameter of dma_alloc_coherent() as ntb->pdev->dev instead.

Fixes 5648e56d03fa ("NTB: ntb_perf: Add full multi-port NTB API support")
---
 drivers/ntb/test/ntb_perf.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/ntb/test/ntb_perf.c +4 -4
index 972f6d984f6d..1c93b9f5c73b 100644
@@ -557,7 +557,7 @@ static void perf_free_inbuf(struct perf_peer *peer)
 		return;
 
 	(void)ntb_mw_clear_trans(peer->perf->ntb, peer->pidx, peer->gidx);
-	dma_free_coherent(&peer->perf->ntb->dev, peer->inbuf_size,
+	dma_free_coherent(&peer->perf->ntb->pdev->dev, peer->inbuf_size,
 			  peer->inbuf, peer->inbuf_xlat);
 	peer->inbuf = NULL;
 }
@@ -586,8 +586,9 @@ static int perf_setup_inbuf(struct perf_peer *peer)
 
 	perf_free_inbuf(peer);
 
-	peer->inbuf = dma_alloc_coherent(&perf->ntb->dev, peer->inbuf_size,
-					 &peer->inbuf_xlat, GFP_KERNEL);
+	peer->inbuf = dma_alloc_coherent(&perf->ntb->pdev->dev,
+					 peer->inbuf_size, &peer->inbuf_xlat,
+					 GFP_KERNEL);
 	if (!peer->inbuf) {
 		dev_err(&perf->ntb->dev, "Failed to alloc inbuf of %pa\n",
 			&peer->inbuf_size);
@@ -1554,4 +1555,3 @@ static void __exit perf_exit(void)
 	destroy_workqueue(perf_wq);
 }
 module_exit(perf_exit);
-
Signature · 1 line
2.25.1
author maintainer
#369 lore raw

PATCH 348/388 ntb_tool: pass correct struct device to dma_alloc_coherent

In reply to Sasha Levin

From: Sanjay R Mehta <[email protected]>

[ Upstream commit 433efe720674efd9fdbcef78be75793393cf05db ]

Currently, ntb->dev is passed to dma_alloc_coherent and dma_free_coherent calls. The returned dma_addr_t is the CPU physical address. This works fine as long as IOMMU is disabled. But when IOMMU is enabled, we need to make sure that IOVA is returned for dma_addr_t. So the correct way to achieve this is by changing the first parameter of dma_alloc_coherent() as ntb->pdev->dev instead.

Fixes 5648e56d03fa ("NTB: ntb_perf: Add full multi-port NTB API support")
---
 drivers/ntb/test/ntb_tool.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/ntb/test/ntb_tool.c +3 -3
index 69da758fe64c..9eaeb221d980 100644
@@ -590,7 +590,7 @@ static int tool_setup_mw(struct tool_ctx *tc, int pidx, int widx,
 	inmw->size = min_t(resource_size_t, req_size, size);
 	inmw->size = round_up(inmw->size, addr_align);
 	inmw->size = round_up(inmw->size, size_align);
-	inmw->mm_base = dma_alloc_coherent(&tc->ntb->dev, inmw->size,
+	inmw->mm_base = dma_alloc_coherent(&tc->ntb->pdev->dev, inmw->size,
 					   &inmw->dma_base, GFP_KERNEL);
 	if (!inmw->mm_base)
 		return -ENOMEM;
@@ -612,7 +612,7 @@ static int tool_setup_mw(struct tool_ctx *tc, int pidx, int widx,
 	return 0;
 
 err_free_dma:
-	dma_free_coherent(&tc->ntb->dev, inmw->size, inmw->mm_base,
+	dma_free_coherent(&tc->ntb->pdev->dev, inmw->size, inmw->mm_base,
 			  inmw->dma_base);
 	inmw->mm_base = NULL;
 	inmw->dma_base = 0;
@@ -629,7 +629,7 @@ static void tool_free_mw(struct tool_ctx *tc, int pidx, int widx)
 
 	if (inmw->mm_base != NULL) {
 		ntb_mw_clear_trans(tc->ntb, pidx, widx);
-		dma_free_coherent(&tc->ntb->dev, inmw->size,
+		dma_free_coherent(&tc->ntb->pdev->dev, inmw->size,
 				  inmw->mm_base, inmw->dma_base);
 	}
 
Signature · 1 line
2.25.1
author maintainer
#370 lore raw

PATCH 349/388 NTB: ntb_tool: reading the link file should not end in a NULL byte

In reply to Sasha Levin

From: Logan Gunthorpe <[email protected]>

[ Upstream commit 912e12813dd03c602e4922fc34709ec4d4380cf0 ]

When running ntb_test this warning is issued:

./ntb_test.sh: line 200: warning: command substitution: ignored null byte in input

This is caused by the kernel returning one more byte than is necessary when reading the link file.

Reduce the number of bytes read back to 2 as it was before the commit that regressed this.

Fixes 7f46c8b3a552 ("NTB: ntb_tool: Add full multi-port NTB API support")
---
 drivers/ntb/test/ntb_tool.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/ntb/test/ntb_tool.c +1 -2
index 9eaeb221d980..b7bf3f863d79 100644
@@ -504,7 +504,7 @@ static ssize_t tool_peer_link_read(struct file *filep, char __user *ubuf,
 	buf[1] = '\n';
 	buf[2] = '\0';
 
-	return simple_read_from_buffer(ubuf, size, offp, buf, 3);
+	return simple_read_from_buffer(ubuf, size, offp, buf, 2);
 }
 
 static TOOL_FOPS_RDWR(tool_peer_link_fops,
@@ -1690,4 +1690,3 @@ static void __exit tool_exit(void)
 	debugfs_remove_recursive(tool_dbgfs_topdir);
 }
 module_exit(tool_exit);
-
Signature · 1 line
2.25.1
author maintainer
#371 lore raw

PATCH 350/388 NTB: Revert the change to use the NTB device dev for DMA allocations

In reply to Sasha Levin

From: Logan Gunthorpe <[email protected]>

[ Upstream commit 40da7d9a93c8941737ef4a1208d32c13ce017fe1 ]

Commit 417cf39cfea9 ("NTB: Set dma mask and dma coherent mask to NTB devices") started using the NTB device for DMA allocations which was turns out was wrong. If the IOMMU is enabled, such alloctanions will always fail with messages such as:

DMAR: Allocating domain for 0000:02:00.1 failed

This is because the IOMMU has not setup the device for such use.

Change the tools back to using the PCI device for allocations seeing it doesn't make sense to add an IOMMU group for the non-physical NTB device. Also remove the code that sets the DMA mask as it no longer makes sense to do this.

Fixes 7f46c8b3a552 ("NTB: ntb_tool: Add full multi-port NTB API support")
---
 drivers/ntb/core.c | 1 -
 1 file changed, 1 deletion(-)
drivers/ntb/core.c +0 -1
index c9a0912b175f..f8f75a504a58 100644
@@ -311,4 +311,3 @@ static void __exit ntb_driver_exit(void)
 	bus_unregister(&ntb_bus);
 }
 module_exit(ntb_driver_exit);
-
Signature · 1 line
2.25.1
author maintainer
#372 lore raw

PATCH 351/388 NTB: perf: Don't require one more memory window than number of peers

In reply to Sasha Levin

From: Logan Gunthorpe <[email protected]>

[ Upstream commit a9c4211ac918ade1522aced6b5acfbe824722f7d ]

ntb_perf should not require more than one memory window per peer. This was probably an off-by-one error.

Fixes 5648e56d03fa ("NTB: ntb_perf: Add full multi-port NTB API support")
---
 drivers/ntb/test/ntb_perf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/ntb/test/ntb_perf.c +1 -1
index 1c93b9f5c73b..21c8a3bac1e0 100644
@@ -654,7 +654,7 @@ static int perf_init_service(struct perf_ctx *perf)
 {
 	u64 mask;
 
-	if (ntb_peer_mw_count(perf->ntb) < perf->pcnt + 1) {
+	if (ntb_peer_mw_count(perf->ntb) < perf->pcnt) {
 		dev_err(&perf->ntb->dev, "Not enough memory windows\n");
 		return -EINVAL;
 	}
Signature · 1 line
2.25.1
author maintainer
#373 lore raw

PATCH 352/388 NTB: perf: Fix support for hardware that doesn't have port numbers

In reply to Sasha Levin

From: Logan Gunthorpe <[email protected]>

[ Upstream commit b54369a248c2e033bfcf5d6917e08cf9d73d54a6 ]

Legacy drivers do not have port numbers (but is reliably only two ports) and was broken by the recent commit that added mult-port support to ntb_perf. This is especially important to support the cross link topology which is perfectly symmetric and cannot assign unique port numbers easily.

Hardware that returns zero for both the local port and the peer should just always use gidx=0 for the only peer.

Fixes 5648e56d03fa ("NTB: ntb_perf: Add full multi-port NTB API support")
---
 drivers/ntb/test/ntb_perf.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
drivers/ntb/test/ntb_perf.c +10 -0
index 21c8a3bac1e0..0b1eae07b133 100644
@@ -1460,6 +1460,16 @@ static int perf_init_peers(struct perf_ctx *perf)
 	if (perf->gidx == -1)
 		perf->gidx = pidx;
 
+	/*
+	 * Hardware with only two ports may not have unique port
+	 * numbers. In this case, the gidxs should all be zero.
+	 */
+	if (perf->pcnt == 1 &&  ntb_port_number(perf->ntb) == 0 &&
+	    ntb_peer_port_number(perf->ntb, 0) == 0) {
+		perf->gidx = 0;
+		perf->peers[0].gidx = 0;
+	}
+
 	for (pidx = 0; pidx < perf->pcnt; pidx++) {
 		ret = perf_setup_peer_mw(&perf->peers[pidx]);
 		if (ret)
Signature · 1 line
2.25.1
author maintainer
#374 lore raw

PATCH 353/388 NTB: perf: Fix race condition when run with ntb_test

In reply to Sasha Levin

From: Logan Gunthorpe <[email protected]>

[ Upstream commit 34d8673a01b053b6231a995a4eec9341163d63be ]

When running ntb_test, the script tries to run the ntb_perf test immediately after probing the modules. Since adding multi-port support, this fails seeing the new initialization procedure in ntb_perf can not complete instantly.

To fix this we add a completion which is waited on when a test is started. In this way, run can be written any time after the module is loaded and it will wait for the initialization to complete instead of sending an error.

Fixes 5648e56d03fa ("NTB: ntb_perf: Add full multi-port NTB API support")
---
 drivers/ntb/test/ntb_perf.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
drivers/ntb/test/ntb_perf.c +8 -2
index 0b1eae07b133..528751803419 100644
@@ -159,6 +159,8 @@ struct perf_peer {
 	/* NTB connection setup service */
 	struct work_struct	service;
 	unsigned long		sts;
+
+	struct completion init_comp;
 };
 #define to_peer_service(__work) \
 	container_of(__work, struct perf_peer, service)
@@ -547,6 +549,7 @@ static int perf_setup_outbuf(struct perf_peer *peer)
 
 	/* Initialization is finally done */
 	set_bit(PERF_STS_DONE, &peer->sts);
+	complete_all(&peer->init_comp);
 
 	return 0;
 }
@@ -638,6 +641,7 @@ static void perf_service_work(struct work_struct *work)
 		perf_setup_outbuf(peer);
 
 	if (test_and_clear_bit(PERF_CMD_CLEAR, &peer->sts)) {
+		init_completion(&peer->init_comp);
 		clear_bit(PERF_STS_DONE, &peer->sts);
 		if (test_bit(0, &peer->perf->busy_flag) &&
 		    peer == peer->perf->test_peer) {
@@ -1084,8 +1088,9 @@ static int perf_submit_test(struct perf_peer *peer)
 	struct perf_thread *pthr;
 	int tidx, ret;
 
-	if (!test_bit(PERF_STS_DONE, &peer->sts))
-		return -ENOLINK;
+	ret = wait_for_completion_interruptible(&peer->init_comp);
+	if (ret < 0)
+		return ret;
 
 	if (test_and_set_bit_lock(0, &perf->busy_flag))
 		return -EBUSY;
@@ -1456,6 +1461,7 @@ static int perf_init_peers(struct perf_ctx *perf)
 			peer->gidx = pidx;
 		}
 		INIT_WORK(&peer->service, perf_service_work);
+		init_completion(&peer->init_comp);
 	}
 	if (perf->gidx == -1)
 		perf->gidx = pidx;
Signature · 1 line
2.25.1
author maintainer
#375 lore raw

PATCH 354/388 NTB: ntb_test: Fix bug when counting remote files

In reply to Sasha Levin

From: Logan Gunthorpe <[email protected]>

[ Upstream commit 2130c0ba69d69bb21f5c52787f2587db00d13d8a ]

When remote files are counted in get_files_count, without using SSH, the code returns 0 because there is a colon prepended to $LOC. $VPATH should have been used instead of $LOC.

Fixes 06bd0407d06c ("NTB: ntb_test: Update ntb_tool Scratchpad tests")
---
 tools/testing/selftests/ntb/ntb_test.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/testing/selftests/ntb/ntb_test.sh +1 -1
index 9c60337317c6..020137b61407 100755
@@ -241,7 +241,7 @@ function get_files_count()
 	split_remote $LOC
 
 	if [[ "$REMOTE" == "" ]]; then
-		echo $(ls -1 "$LOC"/${NAME}* 2>/dev/null | wc -l)
+		echo $(ls -1 "$VPATH"/${NAME}* 2>/dev/null | wc -l)
 	else
 		echo $(ssh "$REMOTE" "ls -1 \"$VPATH\"/${NAME}* | \
 		       wc -l" 2> /dev/null)
Signature · 1 line
2.25.1
author maintainer
#376 lore raw

PATCH 355/388 i2c: icy: Fix build with CONFIG_AMIGA_PCMCIA=n

In reply to Sasha Levin

From: Max Staudt <[email protected]>

[ Upstream commit cdb555397f438592bab00599037c347b700cf397 ]

This has been found by the Kernel Test Robot:
http://lkml.iu.edu/hypermail/linux/kernel/2006.0/06862.html

With CONFIG_AMIGA_PCMCIA=n, io_mm.h does not pull in amigahw.h and ZTWO_VADDR is undefined. Add forgotten include to i2c-icy.c

Fixes 4768e90ecaec ("i2c: Add i2c-icy for I2C on m68k/Amiga")
---
 drivers/i2c/busses/i2c-icy.c | 1 +
 1 file changed, 1 insertion(+)
drivers/i2c/busses/i2c-icy.c +1 -0
index 271470f4d8a9..66c9923fc766 100644
@@ -43,6 +43,7 @@
 #include <linux/i2c.h>
 #include <linux/i2c-algo-pcf.h>
 
+#include <asm/amigahw.h>
 #include <asm/amigaints.h>
 #include <linux/zorro.h>
 
Signature · 1 line
2.25.1
author maintainer
#377 lore raw

PATCH 356/388 mailbox: imx: Add context save/restore for suspend/resume

In reply to Sasha Levin

From: Dong Aisheng <[email protected]>

[ Upstream commit ba5f9fa0ca85a6137fa35efd3a1256d8bb6bc5ff ]

For "mem" mode suspend on i.MX8 SoCs, MU settings could be lost because its power is off, so save/restore is needed for MU settings during suspend/resume. However, the restore can ONLY be done when MU settings are actually lost, for the scenario of settings NOT lost in "freeze" mode suspend, since there could be still IPC going on multiple CPUs, restoring the MU settings could overwrite the TIE by mistake and cause system freeze, so need to make sure ONLY restore the MU settings when it is powered off, Anson fixes this by checking whether restore is actually needed when resume.

---
 drivers/mailbox/imx-mailbox.c | 35 +++++++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)
drivers/mailbox/imx-mailbox.c +35 -0
index 9d6f0217077b..478308fb82cc 100644
@@ -66,6 +66,8 @@ struct imx_mu_priv {
 	struct clk		*clk;
 	int			irq;
 
+	u32 xcr;
+
 	bool			side_b;
 };
 
@@ -558,12 +560,45 @@ static const struct of_device_id imx_mu_dt_ids[] = {
 };
 MODULE_DEVICE_TABLE(of, imx_mu_dt_ids);
 
+static int imx_mu_suspend_noirq(struct device *dev)
+{
+	struct imx_mu_priv *priv = dev_get_drvdata(dev);
+
+	priv->xcr = imx_mu_read(priv, priv->dcfg->xCR);
+
+	return 0;
+}
+
+static int imx_mu_resume_noirq(struct device *dev)
+{
+	struct imx_mu_priv *priv = dev_get_drvdata(dev);
+
+	/*
+	 * ONLY restore MU when context lost, the TIE could
+	 * be set during noirq resume as there is MU data
+	 * communication going on, and restore the saved
+	 * value will overwrite the TIE and cause MU data
+	 * send failed, may lead to system freeze. This issue
+	 * is observed by testing freeze mode suspend.
+	 */
+	if (!imx_mu_read(priv, priv->dcfg->xCR))
+		imx_mu_write(priv, priv->xcr, priv->dcfg->xCR);
+
+	return 0;
+}
+
+static const struct dev_pm_ops imx_mu_pm_ops = {
+	SET_NOIRQ_SYSTEM_SLEEP_PM_OPS(imx_mu_suspend_noirq,
+				      imx_mu_resume_noirq)
+};
+
 static struct platform_driver imx_mu_driver = {
 	.probe		= imx_mu_probe,
 	.remove		= imx_mu_remove,
 	.driver = {
 		.name	= "imx_mu",
 		.of_match_table = imx_mu_dt_ids,
+		.pm = &imx_mu_pm_ops,
 	},
 };
 module_platform_driver(imx_mu_driver);
Signature · 7 lines
2.25.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#378 lore raw

PATCH 357/388 arm64: ftrace: Change CONFIG_FTRACE_WITH_REGS to CONFIG_DYNAMIC_FTRACE_WITH_REGS

In reply to Sasha Levin

From: Joe Perches <[email protected]>

[ Upstream commit 91970bef48d68d06b2bb3f464b572ad50941f6a9 ]

CONFIG_FTRACE_WITH_REGS does not exist as a Kconfig symbol.

---
 arch/arm64/kernel/ftrace.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
arch/arm64/kernel/ftrace.c +2 -1
index 8618faa82e6d..86a5cf9bc19a 100644
@@ -69,7 +69,8 @@ static struct plt_entry *get_ftrace_plt(struct module *mod, unsigned long addr)
 
 	if (addr == FTRACE_ADDR)
 		return &plt[FTRACE_PLT_IDX];
-	if (addr == FTRACE_REGS_ADDR && IS_ENABLED(CONFIG_FTRACE_WITH_REGS))
+	if (addr == FTRACE_REGS_ADDR &&
+	    IS_ENABLED(CONFIG_DYNAMIC_FTRACE_WITH_REGS))
 		return &plt[FTRACE_REGS_PLT_IDX];
 #endif
 	return NULL;
Signature · 7 lines
2.25.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#379 lore raw

PATCH 358/388 drivers/perf: hisi: Fix wrong value for all counters enable

In reply to Sasha Levin

From: Shaokun Zhang <[email protected]>

[ Upstream commit 961abd78adcb4c72c343fcd9f9dc5e2ebbe9b448 ]

In L3C uncore PMU drivers, bit16 is used to control all counters enable & disable. Wrong value is given in the driver and its default value is 1'b1, it can work because each PMU counter has its own control bits too. Let's fix the wrong value.

Fixes 2940bc433370 ("perf: hisi: Add support for HiSilicon SoC L3C PMU driver")
Cc Will Deacon <[email protected]>
Cc Mark Rutland <[email protected]>
---
 drivers/perf/hisilicon/hisi_uncore_l3c_pmu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/perf/hisilicon/hisi_uncore_l3c_pmu.c +1 -1
index 1151e99b241c..479de4be99eb 100644
@@ -35,7 +35,7 @@
 /* L3C has 8-counters */
 #define L3C_NR_COUNTERS		0x8
 
-#define L3C_PERF_CTRL_EN	0x20000
+#define L3C_PERF_CTRL_EN	0x10000
 #define L3C_EVTYPE_NONE		0xff
 
 /*
Signature · 1 line
2.25.1
author maintainer
#380 lore raw

PATCH 359/388 selftests/net: in timestamping, strncpy needs to preserve null byte

In reply to Sasha Levin

From: tannerlove <[email protected]>

[ Upstream commit 8027bc0307ce59759b90679fa5d8b22949586d20 ]

If user passed an interface option longer than 15 characters, then device.ifr_name and hwtstamp.ifr_name became non-null-terminated strings. The compiler warned about this:

timestamping.c:353:2: warning: ‘strncpy’ specified bound 16 equals \
destination size [-Wstringop-truncation]
353 | strncpy(device.ifr_name, interface, sizeof(device.ifr_name));

Fixes cb9eff097831 ("net: new user space API for time stamping of incoming and outgoing packets")
---
 tools/testing/selftests/net/timestamping.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
tools/testing/selftests/net/timestamping.c +8 -2
index aca3491174a1..f4bb4fef0f39 100644
@@ -313,10 +313,16 @@ int main(int argc, char **argv)
 	int val;
 	socklen_t len;
 	struct timeval next;
+	size_t if_len;
 
 	if (argc < 2)
 		usage(0);
 	interface = argv[1];
+	if_len = strlen(interface);
+	if (if_len >= IFNAMSIZ) {
+		printf("interface name exceeds IFNAMSIZ\n");
+		exit(1);
+	}
 
 	for (i = 2; i < argc; i++) {
 		if (!strcasecmp(argv[i], "SO_TIMESTAMP"))
@@ -350,12 +356,12 @@ int main(int argc, char **argv)
 		bail("socket");
 
 	memset(&device, 0, sizeof(device));
-	strncpy(device.ifr_name, interface, sizeof(device.ifr_name));
+	memcpy(device.ifr_name, interface, if_len + 1);
 	if (ioctl(sock, SIOCGIFADDR, &device) < 0)
 		bail("getting interface IP address");
 
 	memset(&hwtstamp, 0, sizeof(hwtstamp));
-	strncpy(hwtstamp.ifr_name, interface, sizeof(hwtstamp.ifr_name));
+	memcpy(hwtstamp.ifr_name, interface, if_len + 1);
 	hwtstamp.ifr_data = (void *)&hwconfig;
 	memset(&hwconfig, 0, sizeof(hwconfig));
 	hwconfig.tx_type =
Signature · 1 line
2.25.1
author maintainer
#381 lore raw

PATCH 360/388 f2fs: don't return vmalloc() memory from f2fs_kmalloc()

In reply to Sasha Levin

From: Eric Biggers <[email protected]>

[ Upstream commit 0b6d4ca04a86b9dababbb76e58d33c437e127b77 ]

kmalloc() returns kmalloc'ed memory, and kvmalloc() returns either kmalloc'ed or vmalloc'ed memory. But the f2fs wrappers, f2fs_kmalloc() and f2fs_kvmalloc(), both return both kinds of memory.

It's redundant to have two functions that do the same thing, and also breaking the standard naming convention is causing bugs since people assume it's safe to kfree() memory allocated by f2fs_kmalloc(). See e.g. the various allocations in fs/f2fs/compress.c.

Fix this by making f2fs_kmalloc() just use kmalloc(). And to avoid re-introducing the allocation failures that the vmalloc fallback was intended to fix, convert the largest allocations to use f2fs_kvmalloc().

---
 fs/f2fs/checkpoint.c | 4 ++--
 fs/f2fs/f2fs.h       | 8 +-------
 fs/f2fs/node.c       | 8 ++++----
 fs/f2fs/super.c      | 2 +-
 4 files changed, 8 insertions(+), 14 deletions(-)
fs/f2fs/checkpoint.c +2 -2
index 852890b72d6a..448b3dc6f925 100644
@@ -889,8 +889,8 @@ int f2fs_get_valid_checkpoint(struct f2fs_sb_info *sbi)
 	int i;
 	int err;
 
-	sbi->ckpt = f2fs_kzalloc(sbi, array_size(blk_size, cp_blks),
-				 GFP_KERNEL);
+	sbi->ckpt = f2fs_kvzalloc(sbi, array_size(blk_size, cp_blks),
+				  GFP_KERNEL);
 	if (!sbi->ckpt)
 		return -ENOMEM;
 	/*
fs/f2fs/f2fs.h +1 -7
index 1f8ae28d7ccf..6579f43f8519 100644
@@ -2935,18 +2935,12 @@ static inline bool f2fs_may_extent_tree(struct inode *inode)
 static inline void *f2fs_kmalloc(struct f2fs_sb_info *sbi,
 					size_t size, gfp_t flags)
 {
-	void *ret;
-
 	if (time_to_inject(sbi, FAULT_KMALLOC)) {
 		f2fs_show_injection_info(sbi, FAULT_KMALLOC);
 		return NULL;
 	}
 
-	ret = kmalloc(size, flags);
-	if (ret)
-		return ret;
-
-	return kvmalloc(size, flags);
+	return kmalloc(size, flags);
 }
 
 static inline void *f2fs_kzalloc(struct f2fs_sb_info *sbi,
fs/f2fs/node.c +4 -4
index ecbd6bd14a49..daf531e69b67 100644
@@ -2928,7 +2928,7 @@ static int __get_nat_bitmaps(struct f2fs_sb_info *sbi)
 		return 0;
 
 	nm_i->nat_bits_blocks = F2FS_BLK_ALIGN((nat_bits_bytes << 1) + 8);
-	nm_i->nat_bits = f2fs_kzalloc(sbi,
+	nm_i->nat_bits = f2fs_kvzalloc(sbi,
 			nm_i->nat_bits_blocks << F2FS_BLKSIZE_BITS, GFP_KERNEL);
 	if (!nm_i->nat_bits)
 		return -ENOMEM;
@@ -3061,9 +3061,9 @@ static int init_free_nid_cache(struct f2fs_sb_info *sbi)
 	int i;
 
 	nm_i->free_nid_bitmap =
-		f2fs_kzalloc(sbi, array_size(sizeof(unsigned char *),
-					     nm_i->nat_blocks),
-			     GFP_KERNEL);
+		f2fs_kvzalloc(sbi, array_size(sizeof(unsigned char *),
+					      nm_i->nat_blocks),
+			      GFP_KERNEL);
 	if (!nm_i->free_nid_bitmap)
 		return -ENOMEM;
 
fs/f2fs/super.c +1 -1
index c5e8cb31626f..b7d14ddcd469 100644
@@ -3027,7 +3027,7 @@ static int init_blkz_info(struct f2fs_sb_info *sbi, int devi)
 	if (nr_sectors & (bdev_zone_sectors(bdev) - 1))
 		FDEV(devi).nr_blkz++;
 
-	FDEV(devi).blkz_seq = f2fs_kzalloc(sbi,
+	FDEV(devi).blkz_seq = f2fs_kvzalloc(sbi,
 					BITS_TO_LONGS(FDEV(devi).nr_blkz)
 					* sizeof(unsigned long),
 					GFP_KERNEL);
Signature · 1 line
2.25.1
author maintainer
#382 lore raw

PATCH 361/388 afs: Fix memory leak in afs_put_sysnames()

In reply to Sasha Levin

From: Zhihao Cheng <[email protected]>

[ Upstream commit 2ca068be09bf8e285036603823696140026dcbe7 ]

Fix afs_put_sysnames() to actually free the specified afs_sysnames object after its reference count has been decreased to zero and its contents have been released.

Fixes 6f8880d8e681557 ("afs: Implement @sys substitution handling")
---
 fs/afs/proc.c | 1 +
 1 file changed, 1 insertion(+)
fs/afs/proc.c +1 -0
index 468e1713bce1..6f34c84a0fd0 100644
@@ -563,6 +563,7 @@ void afs_put_sysnames(struct afs_sysnames *sysnames)
 			if (sysnames->subs[i] != afs_init_sysname &&
 			    sysnames->subs[i] != sysnames->blank)
 				kfree(sysnames->subs[i]);
+		kfree(sysnames);
 	}
 }
 
Signature · 1 line
2.25.1
author maintainer
#383 lore raw

PATCH 362/388 ASoC: soc-pcm: dpcm: fix playback/capture checks

In reply to Sasha Levin

From: Pierre-Louis Bossart <[email protected]>

[ Upstream commit b73287f0b0745961b14e5ebcce92cc8ed24d4d52 ]

Recent changes in the ASoC core prevent multi-cpu BE dailinks from being used. DPCM does support multi-cpu DAIs for BE Dailinks, but not for FE.

Handle the FE checks first, and make sure all DAIs support the same capabilities within the same dailink.

---
 sound/soc/soc-pcm.c | 44 ++++++++++++++++++++++++++++++++++----------
 1 file changed, 34 insertions(+), 10 deletions(-)
sound/soc/soc-pcm.c +34 -10
index 1f302de44052..39ce61c5b874 100644
@@ -2908,20 +2908,44 @@ int soc_new_pcm(struct snd_soc_pcm_runtime *rtd, int num)
 	struct snd_pcm *pcm;
 	char new_name[64];
 	int ret = 0, playback = 0, capture = 0;
+	int stream;
 	int i;
 
+	if (rtd->dai_link->dynamic && rtd->num_cpus > 1) {
+		dev_err(rtd->dev,
+			"DPCM doesn't support Multi CPU for Front-Ends yet\n");
+		return -EINVAL;
+	}
+
 	if (rtd->dai_link->dynamic || rtd->dai_link->no_pcm) {
-		cpu_dai = asoc_rtd_to_cpu(rtd, 0);
-		if (rtd->num_cpus > 1) {
-			dev_err(rtd->dev,
-				"DPCM doesn't support Multi CPU yet\n");
-			return -EINVAL;
+		if (rtd->dai_link->dpcm_playback) {
+			stream = SNDRV_PCM_STREAM_PLAYBACK;
+
+			for_each_rtd_cpu_dais(rtd, i, cpu_dai)
+				if (!snd_soc_dai_stream_valid(cpu_dai,
+							      stream)) {
+					dev_err(rtd->card->dev,
+						"CPU DAI %s for rtd %s does not support playback\n",
+						cpu_dai->name,
+						rtd->dai_link->stream_name);
+					return -EINVAL;
+				}
+			playback = 1;
+		}
+		if (rtd->dai_link->dpcm_capture) {
+			stream = SNDRV_PCM_STREAM_CAPTURE;
+
+			for_each_rtd_cpu_dais(rtd, i, cpu_dai)
+				if (!snd_soc_dai_stream_valid(cpu_dai,
+							      stream)) {
+					dev_err(rtd->card->dev,
+						"CPU DAI %s for rtd %s does not support capture\n",
+						cpu_dai->name,
+						rtd->dai_link->stream_name);
+					return -EINVAL;
+				}
+			capture = 1;
 		}
-
-		playback = rtd->dai_link->dpcm_playback &&
-			   snd_soc_dai_stream_valid(cpu_dai, SNDRV_PCM_STREAM_PLAYBACK);
-		capture = rtd->dai_link->dpcm_capture &&
-			  snd_soc_dai_stream_valid(cpu_dai, SNDRV_PCM_STREAM_CAPTURE);
 	} else {
 		/* Adapt stream for codec2codec links */
 		int cpu_capture = rtd->dai_link->params ?
Signature · 1 line
2.25.1
author maintainer
#384 lore raw

PATCH 363/388 ASoC: core: only convert non DPCM link to DPCM link

In reply to Sasha Levin

From: Bard Liao <[email protected]>

[ Upstream commit 607fa205a7e4dfad28b8a67ab1c985756ddbccb0 ]

Additional checks for valid DAIs expose a corner case, where existing BE dailinks get modified, e.g. HDMI links are tagged with dpcm_capture=1 even if the DAIs are for playback.

This patch makes those changes conditional and flags configuration issues when a BE dailink is has no_pcm=0 but dpcm_playback or dpcm_capture=1 (which makes no sense).

As discussed on the alsa-devel mailing list, there are redundant flags for dpcm_playback, dpcm_capture, playback_only, capture_only. This will have to be cleaned-up in a future update. For now only correct and flag problematic configurations.

---
 sound/soc/soc-core.c | 22 +++++++++++++++++++---
 1 file changed, 19 insertions(+), 3 deletions(-)
sound/soc/soc-core.c +19 -3
index 843b8b1c89d4..e5433e8fcf19 100644
@@ -1720,9 +1720,25 @@ static void soc_check_tplg_fes(struct snd_soc_card *card)
 			dai_link->platforms->name = component->name;
 
 			/* convert non BE into BE */
-			dai_link->no_pcm = 1;
-			dai_link->dpcm_playback = 1;
-			dai_link->dpcm_capture = 1;
+			if (!dai_link->no_pcm) {
+				dai_link->no_pcm = 1;
+
+				if (dai_link->dpcm_playback)
+					dev_warn(card->dev,
+						 "invalid configuration, dailink %s has flags no_pcm=0 and dpcm_playback=1\n",
+						 dai_link->name);
+				if (dai_link->dpcm_capture)
+					dev_warn(card->dev,
+						 "invalid configuration, dailink %s has flags no_pcm=0 and dpcm_capture=1\n",
+						 dai_link->name);
+
+				/* convert normal link into DPCM one */
+				if (!(dai_link->dpcm_playback ||
+				      dai_link->dpcm_capture)) {
+					dai_link->dpcm_playback = !dai_link->capture_only;
+					dai_link->dpcm_capture = !dai_link->playback_only;
+				}
+			}
 
 			/* override any BE fixups */
 			dai_link->be_hw_params_fixup =
Signature · 1 line
2.25.1
author maintainer
#385 lore raw

PATCH 364/388 ASoC: SOF: nocodec: conditionally set dpcm_capture/dpcm_playback flags

In reply to Sasha Levin

From: Pierre-Louis Bossart <[email protected]>

[ Upstream commit ba4e5abc6c4e173af7c941c03c067263b686665d ]

With additional checks on dailinks, we see errors such as

[ 3.000418] sof-nocodec sof-nocodec: CPU DAI DMIC01 Pin for rtd

NoCodec-6 does not support playback

It's not clear why we set the dpcm_playback and dpcm_capture flags unconditionally, add a check on number of channels for each direction to avoid invalid configurations.

---
 sound/soc/sof/nocodec.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
sound/soc/sof/nocodec.c +4 -2
index 2233146386cc..71cf5f9db79d 100644
@@ -52,8 +52,10 @@ static int sof_nocodec_bes_setup(struct device *dev,
 		links[i].platforms->name = dev_name(dev);
 		links[i].codecs->dai_name = "snd-soc-dummy-dai";
 		links[i].codecs->name = "snd-soc-dummy";
-		links[i].dpcm_playback = 1;
-		links[i].dpcm_capture = 1;
+		if (ops->drv[i].playback.channels_min)
+			links[i].dpcm_playback = 1;
+		if (ops->drv[i].capture.channels_min)
+			links[i].dpcm_capture = 1;
 	}
 
 	card->dai_link = links;
Signature · 1 line
2.25.1
author maintainer
#386 lore raw

PATCH 365/388 ASoC: Intel: bytcr_rt5640: Add quirk for Toshiba Encore WT10-A tablet

In reply to Sasha Levin

From: Hans de Goede <[email protected]>

[ Upstream commit 199a5e8fda54ab3c8c6f6bf980c004e97ebf5ccb ]

The Toshiba Encore WT10-A tablet almost fully works with the default settings for Bay Trail CR devices. The only issue is that it uses a digital mic. connected the the DMIC1 input instead of an analog mic.

Add a quirk for this model using the default settings with the input-map replaced with BYT_RT5640_DMIC1_MAP.

---
 sound/soc/intel/boards/bytcr_rt5640.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)
sound/soc/intel/boards/bytcr_rt5640.c +12 -0
index fbfd53874b47..5c1a5e2aff6f 100644
@@ -754,6 +754,18 @@ static const struct dmi_system_id byt_rt5640_quirk_table[] = {
 					BYT_RT5640_JD_NOT_INV |
 					BYT_RT5640_MCLK_EN),
 	},
+	{	/* Toshiba Encore WT10-A */
+		.matches = {
+			DMI_EXACT_MATCH(DMI_SYS_VENDOR, "TOSHIBA"),
+			DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "TOSHIBA WT10-A-103"),
+		},
+		.driver_data = (void *)(BYT_RT5640_DMIC1_MAP |
+					BYT_RT5640_JD_SRC_JD1_IN4P |
+					BYT_RT5640_OVCD_TH_2000UA |
+					BYT_RT5640_OVCD_SF_0P75 |
+					BYT_RT5640_SSP0_AIF2 |
+					BYT_RT5640_MCLK_EN),
+	},
 	{	/* Catch-all for generic Insyde tablets, must be last */
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "Insyde"),
Signature · 1 line
2.25.1
author maintainer
#387 lore raw

PATCH 366/388 ASoC: rt5645: Add platform-data for Asus T101HA

In reply to Sasha Levin

From: Hans de Goede <[email protected]>

[ Upstream commit 79d4f823a06796656289f97b922493da5690e46c ]

The Asus T101HA uses the default jack-detect mode 3, but instead of using an analog microphone it is using a DMIC on dmic-data-pin 1, like the Asus T100HA. Note unlike the T100HA its jack-detect is not inverted.

Add a DMI quirk with the correct settings for this model.

---
 sound/soc/codecs/rt5645.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)
sound/soc/codecs/rt5645.c +14 -0
index 6ba1849a77b0..e2e1d5b03b38 100644
@@ -3625,6 +3625,12 @@ static const struct rt5645_platform_data asus_t100ha_platform_data = {
 	.inv_jd1_1 = true,
 };
 
+static const struct rt5645_platform_data asus_t101ha_platform_data = {
+	.dmic1_data_pin = RT5645_DMIC_DATA_IN2N,
+	.dmic2_data_pin = RT5645_DMIC2_DISABLE,
+	.jd_mode = 3,
+};
+
 static const struct rt5645_platform_data lenovo_ideapad_miix_310_pdata = {
 	.jd_mode = 3,
 	.in2_diff = true,
@@ -3708,6 +3714,14 @@ static const struct dmi_system_id dmi_platform_data[] = {
 		},
 		.driver_data = (void *)&asus_t100ha_platform_data,
 	},
+	{
+		.ident = "ASUS T101HA",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "T101HA"),
+		},
+		.driver_data = (void *)&asus_t101ha_platform_data,
+	},
 	{
 		.ident = "MINIX Z83-4",
 		.matches = {
Signature · 1 line
2.25.1
author maintainer
#388 lore raw

PATCH 367/388 bpf/sockmap: Fix kernel panic at __tcp_bpf_recvmsg

In reply to Sasha Levin

From: dihu <[email protected]>

[ Upstream commit 487082fb7bd2a32b66927d2b22e3a81b072b44f0 ]

When user application calls read() with MSG_PEEK flag to read data of bpf sockmap socket, kernel panic happens at __tcp_bpf_recvmsg+0x12c/0x350. sk_msg is not removed from ingress_msg queue after read out under MSG_PEEK flag is set. Because it's not judged whether sk_msg is the last msg of ingress_msg queue, the next sk_msg may be the head of ingress_msg queue, whose memory address of sg page is invalid. So it's necessary to add check codes to prevent this problem.

[20759.125457] BUG: kernel NULL pointer dereference, address:
0000000000000008
[20759.132118] CPU: 53 PID: 51378 Comm: envoy Tainted: G            E
5.4.32 #1
[20759.140890] Hardware name: Inspur SA5212M4/YZMB-00370-109, BIOS
4.1.12 06/18/2017
[20759.149734] RIP: 0010:copy_page_to_iter+0xad/0x300
[20759.270877] __tcp_bpf_recvmsg+0x12c/0x350
[20759.276099] tcp_bpf_recvmsg+0x113/0x370
[20759.281137] inet_recvmsg+0x55/0xc0
[20759.285734] __sys_recvfrom+0xc8/0x130
[20759.290566] ? __audit_syscall_entry+0x103/0x130
[20759.296227] ? syscall_trace_enter+0x1d2/0x2d0
[20759.301700] ? __audit_syscall_exit+0x1e4/0x290
[20759.307235] __x64_sys_recvfrom+0x24/0x30
[20759.312226] do_syscall_64+0x55/0x1b0
[20759.316852] entry_SYSCALL_64_after_hwframe+0x44/0xa9
---
 net/ipv4/tcp_bpf.c | 3 +++
 1 file changed, 3 insertions(+)
net/ipv4/tcp_bpf.c +3 -0
index 9c5540887fbe..7aa68f4aae6c 100644
@@ -64,6 +64,9 @@ int __tcp_bpf_recvmsg(struct sock *sk, struct sk_psock *psock,
 		} while (i != msg_rx->sg.end);
 
 		if (unlikely(peek)) {
+			if (msg_rx == list_last_entry(&psock->ingress_msg,
+						      struct sk_msg, list))
+				break;
 			msg_rx = list_next_entry(msg_rx, list);
 			continue;
 		}
Signature · 1 line
2.25.1
author maintainer
#389 lore raw

PATCH 368/388 bpf, sockhash: Synchronize delete from bucket list on map free

In reply to Sasha Levin

From: Jakub Sitnicki <[email protected]>

[ Upstream commit 75e68e5bf2c7fa9d3e874099139df03d5952a3e1 ]

We can end up modifying the sockhash bucket list from two CPUs when a sockhash is being destroyed (sock_hash_free) on one CPU, while a socket that is in the sockhash is unlinking itself from it on another CPU it (sock_hash_delete_from_link).

This results in accessing a list element that is in an undefined state as reported by KASAN:

| ==================================================================
| BUG: KASAN: wild-memory-access in sock_hash_free+0x13c/0x280
| Write of size 8 at addr dead000000000122 by task kworker/2:1/95
|
| CPU: 2 PID: 95 Comm: kworker/2:1 Not tainted 5.7.0-rc7-02961-ge22c35ab0038-dirty #691
| Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS ?-20190727_073836-buildvm-ppc64le-16.ppc.fedoraproject.org-3.fc31 04/01/2014
| Workqueue: events bpf_map_free_deferred
| Call Trace:
|  dump_stack+0x97/0xe0
|  ? sock_hash_free+0x13c/0x280
|  __kasan_report.cold+0x5/0x40
|  ? mark_lock+0xbc1/0xc00
|  ? sock_hash_free+0x13c/0x280
|  kasan_report+0x38/0x50
|  ? sock_hash_free+0x152/0x280
|  sock_hash_free+0x13c/0x280
|  bpf_map_free_deferred+0xb2/0xd0
|  ? bpf_map_charge_finish+0x50/0x50
|  ? rcu_read_lock_sched_held+0x81/0xb0
|  ? rcu_read_lock_bh_held+0x90/0x90
|  process_one_work+0x59a/0xac0
|  ? lock_release+0x3b0/0x3b0
|  ? pwq_dec_nr_in_flight+0x110/0x110
|  ? rwlock_bug.part.0+0x60/0x60
|  worker_thread+0x7a/0x680
|  ? _raw_spin_unlock_irqrestore+0x4c/0x60
|  kthread+0x1cc/0x220
|  ? process_one_work+0xac0/0xac0
|  ? kthread_create_on_node+0xa0/0xa0
|  ret_from_fork+0x24/0x30
| ==================================================================

Fix it by reintroducing spin-lock protected critical section around the code that removes the elements from the bucket on sockhash free.

To do that we also need to defer processing of removed elements, until out of atomic context so that we can unlink the socket from the map when holding the sock lock.

---
 net/core/sock_map.c | 23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)
net/core/sock_map.c +21 -2
index 7edbf1e92457..050bfac97cfb 100644
@@ -1006,6 +1006,7 @@ static void sock_hash_free(struct bpf_map *map)
 {
 	struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
 	struct bpf_htab_bucket *bucket;
+	struct hlist_head unlink_list;
 	struct bpf_htab_elem *elem;
 	struct hlist_node *node;
 	int i;
@@ -1017,13 +1018,31 @@ static void sock_hash_free(struct bpf_map *map)
 	synchronize_rcu();
 	for (i = 0; i < htab->buckets_num; i++) {
 		bucket = sock_hash_select_bucket(htab, i);
-		hlist_for_each_entry_safe(elem, node, &bucket->head, node) {
-			hlist_del_rcu(&elem->node);
+
+		/* We are racing with sock_hash_delete_from_link to
+		 * enter the spin-lock critical section. Every socket on
+		 * the list is still linked to sockhash. Since link
+		 * exists, psock exists and holds a ref to socket. That
+		 * lets us to grab a socket ref too.
+		 */
+		raw_spin_lock_bh(&bucket->lock);
+		hlist_for_each_entry(elem, &bucket->head, node)
+			sock_hold(elem->sk);
+		hlist_move_list(&bucket->head, &unlink_list);
+		raw_spin_unlock_bh(&bucket->lock);
+
+		/* Process removed entries out of atomic context to
+		 * block for socket lock before deleting the psock's
+		 * link to sockhash.
+		 */
+		hlist_for_each_entry_safe(elem, node, &unlink_list, node) {
+			hlist_del(&elem->node);
 			lock_sock(elem->sk);
 			rcu_read_lock();
 			sock_map_unref(elem->sk, elem);
 			rcu_read_unlock();
 			release_sock(elem->sk);
+			sock_put(elem->sk);
 			sock_hash_free_elem(htab, elem);
 		}
 	}
Signature · 1 line
2.25.1
author maintainer
#390 lore raw

PATCH 369/388 tracing/probe: Fix bpf_task_fd_query() for kprobes and uprobes

In reply to Sasha Levin

From: Jean-Philippe Brucker <[email protected]>

[ Upstream commit 22d5bd6867364b41576a712755271a7d6161abd6 ]

Commit 60d53e2c3b75 ("tracing/probe: Split trace_event related data from trace_probe") removed the trace_[ku]probe structure from the trace_event_call->data pointer. As bpf_get_[ku]probe_info() were forgotten in that change, fix them now. These functions are currently only used by the bpf_task_fd_query() syscall handler to collect information about a perf event.

---
 kernel/trace/trace_kprobe.c | 2 +-
 kernel/trace/trace_uprobe.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
kernel/trace/trace_kprobe.c +1 -1
index 35989383ae11..8eeb95e04bf5 100644
@@ -1629,7 +1629,7 @@ int bpf_get_kprobe_info(const struct perf_event *event, u32 *fd_type,
 	if (perf_type_tracepoint)
 		tk = find_trace_kprobe(pevent, group);
 	else
-		tk = event->tp_event->data;
+		tk = trace_kprobe_primary_from_call(event->tp_event);
 	if (!tk)
 		return -EINVAL;
 
kernel/trace/trace_uprobe.c +1 -1
index 2a8e8e9c1c75..fdd47f99b18f 100644
@@ -1412,7 +1412,7 @@ int bpf_get_uprobe_info(const struct perf_event *event, u32 *fd_type,
 	if (perf_type_tracepoint)
 		tu = find_probe_event(pevent, group);
 	else
-		tu = event->tp_event->data;
+		tu = trace_uprobe_primary_from_call(event->tp_event);
 	if (!tu)
 		return -EINVAL;
 
Signature · 1 line
2.25.1
author maintainer
#391 lore raw

PATCH 370/388 drm/sun4i: hdmi ddc clk: Fix size of m divider

In reply to Sasha Levin

From: Jernej Skrabec <[email protected]>

[ Upstream commit 54e1e06bcf1cf6e7ac3f86daa5f7454add24b494 ]

m divider in DDC clock register is 4 bits wide. Fix that.

---
 drivers/gpu/drm/sun4i/sun4i_hdmi.h         | 2 +-
 drivers/gpu/drm/sun4i/sun4i_hdmi_ddc_clk.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/sun4i/sun4i_hdmi.h +1 -1
index 7ad3f06c127e..00ca35f07ba5 100644
@@ -148,7 +148,7 @@
 #define SUN4I_HDMI_DDC_CMD_IMPLICIT_WRITE	3
 
 #define SUN4I_HDMI_DDC_CLK_REG		0x528
-#define SUN4I_HDMI_DDC_CLK_M(m)			(((m) & 0x7) << 3)
+#define SUN4I_HDMI_DDC_CLK_M(m)			(((m) & 0xf) << 3)
 #define SUN4I_HDMI_DDC_CLK_N(n)			((n) & 0x7)
 
 #define SUN4I_HDMI_DDC_LINE_CTRL_REG	0x540
drivers/gpu/drm/sun4i/sun4i_hdmi_ddc_clk.c +1 -1
index 2ff780114106..12430b9d4e93 100644
@@ -33,7 +33,7 @@ static unsigned long sun4i_ddc_calc_divider(unsigned long rate,
 	unsigned long best_rate = 0;
 	u8 best_m = 0, best_n = 0, _m, _n;
 
-	for (_m = 0; _m < 8; _m++) {
+	for (_m = 0; _m < 16; _m++) {
 		for (_n = 0; _n < 8; _n++) {
 			unsigned long tmp_rate;
 
Signature · 6 lines
2.25.1

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

PATCH 371/388 libbpf: Handle GCC noreturn-turned-volatile quirk

In reply to Sasha Levin

From: Andrii Nakryiko <[email protected]>

[ Upstream commit 32022fd97ed34f6812802bf1288db27c313576f4 ]

Handle a GCC quirk of emitting extra volatile modifier in DWARF (and subsequently preserved in BTF by pahole) for function pointers marked as __attribute__((noreturn)). This was the way to mark such functions before GCC 2.5 added noreturn attribute. Drop such func_proto modifiers, similarly to how it's done for array (also to handle GCC quirk/bug).

Such volatile attribute is emitted by GCC only, so existing selftests can't express such test. Simple repro is like this (compiled with GCC + BTF generated by pahole):

  struct my_struct {
      void __attribute__((noreturn)) (*fn)(int);
  };
  struct my_struct a;

Without this fix, output will be:

struct my_struct {
    voidvolatile  (*fn)(int);
};

With the fix:

struct my_struct {
    void (*fn)(int);
};
---
 tools/lib/bpf/btf_dump.c | 33 ++++++++++++++++++++++++---------
 1 file changed, 24 insertions(+), 9 deletions(-)
tools/lib/bpf/btf_dump.c +24 -9
index 0c28ee82834b..653dbbe2e366 100644
@@ -1137,6 +1137,20 @@ static void btf_dump_emit_mods(struct btf_dump *d, struct id_stack *decl_stack)
 	}
 }
 
+static void btf_dump_drop_mods(struct btf_dump *d, struct id_stack *decl_stack)
+{
+	const struct btf_type *t;
+	__u32 id;
+
+	while (decl_stack->cnt) {
+		id = decl_stack->ids[decl_stack->cnt - 1];
+		t = btf__type_by_id(d->btf, id);
+		if (!btf_is_mod(t))
+			return;
+		decl_stack->cnt--;
+	}
+}
+
 static void btf_dump_emit_name(const struct btf_dump *d,
 			       const char *name, bool last_was_ptr)
 {
@@ -1235,14 +1249,7 @@ static void btf_dump_emit_type_chain(struct btf_dump *d,
 			 * a const/volatile modifier for array, so we are
 			 * going to silently skip them here.
 			 */
-			while (decls->cnt) {
-				next_id = decls->ids[decls->cnt - 1];
-				next_t = btf__type_by_id(d->btf, next_id);
-				if (btf_is_mod(next_t))
-					decls->cnt--;
-				else
-					break;
-			}
+			btf_dump_drop_mods(d, decls);
 
 			if (decls->cnt == 0) {
 				btf_dump_emit_name(d, fname, last_was_ptr);
@@ -1270,7 +1277,15 @@ static void btf_dump_emit_type_chain(struct btf_dump *d,
 			__u16 vlen = btf_vlen(t);
 			int i;
 
-			btf_dump_emit_mods(d, decls);
+			/*
+			 * GCC emits extra volatile qualifier for
+			 * __attribute__((noreturn)) function pointers. Clang
+			 * doesn't do it. It's a GCC quirk for backwards
+			 * compatibility with code written for GCC <2.5. So,
+			 * similarly to extra qualifiers for array, just drop
+			 * them, instead of handling them.
+			 */
+			btf_dump_drop_mods(d, decls);
 			if (decls->cnt) {
 				btf_dump_printf(d, " (");
 				btf_dump_emit_type_chain(d, decls, fname, lvl);
Signature · 1 line
2.25.1
author maintainer
#393 lore raw

PATCH 372/388 scsi: acornscsi: Fix an error handling path in acornscsi_probe()

In reply to Sasha Levin

From: Christophe JAILLET <[email protected]>

[ Upstream commit 42c76c9848e13dbe0538d7ae0147a269dfa859cb ]

'ret' is known to be 0 at this point. Explicitly return -ENOMEM if one of the 'ecardm_iomap()' calls fail.

Fixes e95a1b656a98 ("[ARM] rpc: acornscsi: update to new style ecard driver")
---
 drivers/scsi/arm/acornscsi.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/scsi/arm/acornscsi.c +3 -1
index ddb52e7ba622..9a912fd0f70b 100644
@@ -2911,8 +2911,10 @@ static int acornscsi_probe(struct expansion_card *ec, const struct ecard_id *id)
 
 	ashost->base = ecardm_iomap(ec, ECARD_RES_MEMC, 0, 0);
 	ashost->fast = ecardm_iomap(ec, ECARD_RES_IOCFAST, 0, 0);
-	if (!ashost->base || !ashost->fast)
+	if (!ashost->base || !ashost->fast) {
+		ret = -ENOMEM;
 		goto out_put;
+	}
 
 	host->irq = ec->irq;
 	ashost->host = host;
Signature · 7 lines
2.25.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#394 lore raw

PATCH 373/388 drm/ast: fix missing break in switch statement for format->cpp[0] case 4

In reply to Sasha Levin

From: Colin Ian King <[email protected]>

[ Upstream commit 291ddeb621e4a9f1ced8302a777fbd7fbda058c6 ]

Currently the switch statement for format->cpp[0] value 4 assigns color_index which is never read again and then falls through to the default case and returns. This looks like a missing break statement bug. Fix this by adding a break statement.

Addresses-Coverity: ("Unused value")

---
 drivers/gpu/drm/ast/ast_mode.c | 1 +
 1 file changed, 1 insertion(+)
drivers/gpu/drm/ast/ast_mode.c +1 -0
index cdd6c46d6557..479fc60b6c93 100644
@@ -226,6 +226,7 @@ static void ast_set_vbios_color_reg(struct ast_private *ast,
 	case 3:
 	case 4:
 		color_index = TrueCModeIndex;
+		break;
 	default:
 		return;
 	}
Signature · 6 lines
2.25.1

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

PATCH 374/388 x86/idt: Keep spurious entries unset in system_vectors

In reply to Sasha Levin

From: Vitaly Kuznetsov <[email protected]>

[ Upstream commit 1f1fbc70c10e81f70e9fbe2102d439c883269811 ]

With commit dc20b2d52653 ("x86/idt: Move interrupt gate initialization to IDT code") non assigned system vectors are also marked as used in 'used_vectors' (now 'system_vectors') bitmap. This makes checks in arch_show_interrupts() whether a particular system vector is allocated to always pass and e.g. 'Hyper-V reenlightenment interrupts' entry always shows up in /proc/interrupts.

Another side effect of having all unassigned system vectors marked as used is that irq_matrix_debug_show() will wrongly count them among 'System' vectors.

As it is now ensured that alloc_intr_gate() is not called after init, it is possible to leave unused entries in 'system_vectors' unset to fix these issues.

---
 arch/x86/kernel/idt.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
arch/x86/kernel/idt.c +5 -1
index 87ef69a72c52..7bb4c3cbf4dc 100644
@@ -318,7 +318,11 @@ void __init idt_setup_apic_and_irq_gates(void)
 
 #ifdef CONFIG_X86_LOCAL_APIC
 	for_each_clear_bit_from(i, system_vectors, NR_VECTORS) {
-		set_bit(i, system_vectors);
+		/*
+		 * Don't set the non assigned system vectors in the
+		 * system_vectors bitmap. Otherwise they show up in
+		 * /proc/interrupts.
+		 */
 		entry = spurious_entries_start + 8 * (i - FIRST_SYSTEM_VECTOR);
 		set_intr_gate(i, entry);
 	}
Signature · 1 line
2.25.1
author maintainer
#396 lore raw

PATCH 375/388 x86/mce/dev-mcelog: Fix -Wstringop-truncation warning about strncpy()

In reply to Sasha Levin

From: Tony Luck <[email protected]>

[ Upstream commit 7ccddc4613db446dc3cbb69a3763ba60ec651d13 ]

The kbuild test robot reported this warning:

  arch/x86/kernel/cpu/mce/dev-mcelog.c: In function 'dev_mcelog_init_device':
  arch/x86/kernel/cpu/mce/dev-mcelog.c:346:2: warning: 'strncpy' output \
    truncated before terminating nul copying 12 bytes from a string of the \
    same length [-Wstringop-truncation]

This is accurate, but I don't care that the trailing NUL character isn't copied. The string being copied is just a magic number signature so that crash dump tools can be sure they are decoding the right blob of memory.

Use memcpy() instead of strncpy().

Fixes d8ecca4043f2 ("x86/mce/dev-mcelog: Dynamically allocate space for machine check records")
---
 arch/x86/kernel/cpu/mce/dev-mcelog.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/kernel/cpu/mce/dev-mcelog.c +1 -1
index d089567a9ce8..bcb379b2fd42 100644
@@ -343,7 +343,7 @@ static __init int dev_mcelog_init_device(void)
 	if (!mcelog)
 		return -ENOMEM;
 
-	strncpy(mcelog->signature, MCE_LOG_SIGNATURE, sizeof(mcelog->signature));
+	memcpy(mcelog->signature, MCE_LOG_SIGNATURE, sizeof(mcelog->signature));
 	mcelog->len = mce_log_len;
 	mcelog->recordlen = sizeof(struct mce);
 
Signature · 1 line
2.25.1
author maintainer
#397 lore raw

PATCH 376/388 net/filter: Permit reading NET in load_bytes_relative when MAC not set

In reply to Sasha Levin

From: YiFei Zhu <[email protected]>

[ Upstream commit 0f5d82f187e1beda3fe7295dfc500af266a5bd80 ]

Added a check in the switch case on start_header that checks for the existence of the header, and in the case that MAC is not set and the caller requests for MAC, -EFAULT. If the caller requests for NET then MAC's existence is completely ignored.

There is no function to check NET header's existence and as far as cgroup_skb/egress is concerned it should always be set.

Removed for ptr >= the start of header, considering offset is bounded unsigned and should always be true. len <= end - mac is redundant to ptr + len <= end.

---
 net/core/filter.c | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)
net/core/filter.c +9 -7
index 5cc9276f1023..9080d01d9638 100644
@@ -1766,25 +1766,27 @@ BPF_CALL_5(bpf_skb_load_bytes_relative, const struct sk_buff *, skb,
 	   u32, offset, void *, to, u32, len, u32, start_header)
 {
 	u8 *end = skb_tail_pointer(skb);
-	u8 *net = skb_network_header(skb);
-	u8 *mac = skb_mac_header(skb);
-	u8 *ptr;
+	u8 *start, *ptr;
 
-	if (unlikely(offset > 0xffff || len > (end - mac)))
+	if (unlikely(offset > 0xffff))
 		goto err_clear;
 
 	switch (start_header) {
 	case BPF_HDR_START_MAC:
-		ptr = mac + offset;
+		if (unlikely(!skb_mac_header_was_set(skb)))
+			goto err_clear;
+		start = skb_mac_header(skb);
 		break;
 	case BPF_HDR_START_NET:
-		ptr = net + offset;
+		start = skb_network_header(skb);
 		break;
 	default:
 		goto err_clear;
 	}
 
-	if (likely(ptr >= mac && ptr + len <= end)) {
+	ptr = start + offset;
+
+	if (likely(ptr + len <= end)) {
 		memcpy(to, ptr, len);
 		return 0;
 	}
Signature · 1 line
2.25.1
author maintainer
#398 lore raw

PATCH 377/388 tools, bpftool: Fix memory leak in codegen error cases

In reply to Sasha Levin

From: Tobias Klauser <[email protected]>

[ Upstream commit d4060ac969563113101c79433f2ae005feca1c29 ]

Free the memory allocated for the template on error paths in function codegen.

---
 tools/bpf/bpftool/gen.c | 2 ++
 1 file changed, 2 insertions(+)
tools/bpf/bpftool/gen.c +2 -0
index f8113b3646f5..f5960b48c861 100644
@@ -225,6 +225,7 @@ static int codegen(const char *template, ...)
 		} else {
 			p_err("unrecognized character at pos %td in template '%s'",
 			      src - template - 1, template);
+			free(s);
 			return -EINVAL;
 		}
 	}
@@ -235,6 +236,7 @@ static int codegen(const char *template, ...)
 			if (*src != '\t') {
 				p_err("not enough tabs at pos %td in template '%s'",
 				      src - template - 1, template);
+				free(s);
 				return -EINVAL;
 			}
 		}
Signature · 1 line
2.25.1
author maintainer
#399 lore raw

PATCH 378/388 nvme-fc: don't call nvme_cleanup_cmd() for AENs

In reply to Sasha Levin

From: Daniel Wagner <[email protected]>

[ Upstream commit c9c12e51b82b2bd0c59ac4e27ee5427f382a503f ]

Asynchronous event notifications do not have an associated request. When fcp_io() fails we unconditionally call nvme_cleanup_cmd() which leads to a crash.

Fixes 16686f3a6c3c ("nvme: move common call to nvme_cleanup_cmd to core layer")
---
 drivers/nvme/host/fc.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/nvme/host/fc.c +3 -2
index 7dfc4a2ecf1e..287a3e8ea317 100644
@@ -2300,10 +2300,11 @@ nvme_fc_start_fcp_op(struct nvme_fc_ctrl *ctrl, struct nvme_fc_queue *queue,
 		opstate = atomic_xchg(&op->state, FCPOP_STATE_COMPLETE);
 		__nvme_fc_fcpop_chk_teardowns(ctrl, op, opstate);
 
-		if (!(op->flags & FCOP_FLAGS_AEN))
+		if (!(op->flags & FCOP_FLAGS_AEN)) {
 			nvme_fc_unmap_data(ctrl, op->rq, op);
+			nvme_cleanup_cmd(op->rq);
+		}
 
-		nvme_cleanup_cmd(op->rq);
 		nvme_fc_ctrl_put(ctrl);
 
 		if (ctrl->rport->remoteport.port_state == FC_OBJSTATE_ONLINE &&
Signature · 7 lines
2.25.1


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

PATCH 379/388 nvme-pci: use simple suspend when a HMB is enabled

In reply to Sasha Levin

From: Christoph Hellwig <[email protected]>

[ Upstream commit b97120b15ebd3de51325084136d3b9c3cce656d6 ]

While the NVMe specification allows the device to access the host memory buffer in host DRAM from all power states, hosts will fail access to DRAM during S3 and similar power states.

Fixes d916b1be94b6 ("nvme-pci: use host managed power state for suspend")
---
 drivers/nvme/host/pci.c | 6 ++++++
 1 file changed, 6 insertions(+)
drivers/nvme/host/pci.c +6 -0
index cc46e250fcac..340bb9c8995e 100644
@@ -2933,9 +2933,15 @@ static int nvme_suspend(struct device *dev)
 	 * the PCI bus layer to put it into D3 in order to take the PCIe link
 	 * down, so as to allow the platform to achieve its minimum low-power
 	 * state (which may not be possible if the link is up).
+	 *
+	 * If a host memory buffer is enabled, shut down the device as the NVMe
+	 * specification allows the device to access the host memory buffer in
+	 * host DRAM from all power states, but hosts will fail access to DRAM
+	 * during S3.
 	 */
 	if (pm_suspend_via_firmware() || !ctrl->npss ||
 	    !pcie_aspm_enabled(pdev) ||
+	    ndev->nr_host_mem_descs ||
 	    (ndev->ctrl.quirks & NVME_QUIRK_SIMPLE_SUSPEND))
 		return nvme_disable_prepare_reset(ndev, true);
 
Signature · 7 lines
2.25.1


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

PATCH 380/388 nfs: set invalid blocks after NFSv4 writes

In reply to Sasha Levin

From: Zheng Bin <[email protected]>

[ Upstream commit 3a39e778690500066b31fe982d18e2e394d3bce2 ]

Use the following command to test nfsv4(size of file1M is 1MB):
mount -t nfs -o vers=4.0,actimeo=60 127.0.0.1/dir1 /mnt cp file1M /mnt
du -h /mnt/file1M -->0 within 60s, then 1M

When write is done(cp file1M /mnt), will call this:
nfs_writeback_done
nfs4_write_done
nfs4_write_done_cb
nfs_writeback_update_inode
nfs_post_op_update_inode_force_wcc_locked(change, ctime, mtime
nfs_post_op_update_inode_force_wcc_locked
nfs_set_cache_invalid
nfs_refresh_inode_locked
nfs_update_inode

nfsd write response contains change, ctime, mtime, the flag will be clear after nfs_update_inode. Howerver, write response does not contain space_used, previous open response contains space_used whose value is 0, so inode->i_blocks is still 0.

nfs_getattr -->called by "du -h"
do_update |= force_sync || nfs_attribute_cache_expired -->false in 60s cache_validity = READ_ONCE(NFS_I(inode)->cache_validity)
do_update |= cache_validity & (NFS_INO_INVALID_ATTR -->false
if (do_update) {
__nfs_revalidate_inode
}

Within 60s, does not send getattr request to nfsd, thus "du -h /mnt/file1M" is 0.

Add a NFS_INO_INVALID_BLOCKS flag, set it when nfsv4 write is done.

Fixes 16e143751727 ("NFS: More fine grained attribute tracking")
---
 fs/nfs/inode.c         | 14 +++++++++++---
 include/linux/nfs_fs.h |  1 +
 2 files changed, 12 insertions(+), 3 deletions(-)
fs/nfs/inode.c +11 -3
index b9d0921cb4fe..0bf1f835de01 100644
@@ -833,6 +833,8 @@ int nfs_getattr(const struct path *path, struct kstat *stat,
 		do_update |= cache_validity & NFS_INO_INVALID_ATIME;
 	if (request_mask & (STATX_CTIME|STATX_MTIME))
 		do_update |= cache_validity & NFS_INO_REVAL_PAGECACHE;
+	if (request_mask & STATX_BLOCKS)
+		do_update |= cache_validity & NFS_INO_INVALID_BLOCKS;
 	if (do_update) {
 		/* Update the attribute cache */
 		if (!(server->flags & NFS_MOUNT_NOAC))
@@ -1764,7 +1766,8 @@ int nfs_post_op_update_inode_force_wcc_locked(struct inode *inode, struct nfs_fa
 	status = nfs_post_op_update_inode_locked(inode, fattr,
 			NFS_INO_INVALID_CHANGE
 			| NFS_INO_INVALID_CTIME
-			| NFS_INO_INVALID_MTIME);
+			| NFS_INO_INVALID_MTIME
+			| NFS_INO_INVALID_BLOCKS);
 	return status;
 }
 
@@ -1871,7 +1874,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 	nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR
 			| NFS_INO_INVALID_ATIME
 			| NFS_INO_REVAL_FORCED
-			| NFS_INO_REVAL_PAGECACHE);
+			| NFS_INO_REVAL_PAGECACHE
+			| NFS_INO_INVALID_BLOCKS);
 
 	/* Do atomic weak cache consistency updates */
 	nfs_wcc_update_inode(inode, fattr);
@@ -2033,8 +2037,12 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 		inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used);
 	} else if (fattr->valid & NFS_ATTR_FATTR_BLOCKS_USED)
 		inode->i_blocks = fattr->du.nfs2.blocks;
-	else
+	else {
+		nfsi->cache_validity |= save_cache_validity &
+				(NFS_INO_INVALID_BLOCKS
+				| NFS_INO_REVAL_FORCED);
 		cache_revalidated = false;
+	}
 
 	/* Update attrtimeo value if we're out of the unstable period */
 	if (attr_changed) {
include/linux/nfs_fs.h +1 -0
index 73eda45f1cfd..6ee9119acc5d 100644
@@ -230,6 +230,7 @@ struct nfs4_copy_state {
 #define NFS_INO_INVALID_OTHER	BIT(12)		/* other attrs are invalid */
 #define NFS_INO_DATA_INVAL_DEFER	\
 				BIT(13)		/* Deferred cache invalidation */
+#define NFS_INO_INVALID_BLOCKS	BIT(14)         /* cached blocks are invalid */
 
 #define NFS_INO_INVALID_ATTR	(NFS_INO_INVALID_CHANGE \
 		| NFS_INO_INVALID_CTIME \
Signature · 1 line
2.25.1
author maintainer
#402 lore raw

PATCH 381/388 NFS: Fix direct WRITE throughput regression

In reply to Sasha Levin

From: Chuck Lever <[email protected]>

[ Upstream commit ba838a75e73f55a780f1ee896b8e3ecb032dba0f ]

I measured a 50% throughput regression for large direct writes.

The observed on-the-wire behavior is that the client sends every NFS WRITE twice: once as an UNSTABLE WRITE plus a COMMIT, and once as a FILE_SYNC WRITE.

This is because the nfs_write_match_verf() check in nfs_direct_commit_complete() fails for every WRITE.

Buffered writes use nfs_write_completion(), which sets req->wb_verf correctly. Direct writes use nfs_direct_write_completion(), which does not set req->wb_verf at all. This leaves req->wb_verf set to all zeroes for every direct WRITE, and thus nfs_direct_commit_completion() always sets NFS_ODIRECT_RESCHED_WRITES.

This fix appears to restore nearly all of the lost performance.

Fixes 1f28476dcb98 ("NFS: Fix O_DIRECT commit verifier handling")
---
 fs/nfs/direct.c | 2 ++
 1 file changed, 2 insertions(+)
fs/nfs/direct.c +2 -0
index a57e7c72c7f4..d49b1d197908 100644
@@ -731,6 +731,8 @@ static void nfs_direct_write_completion(struct nfs_pgio_header *hdr)
 		nfs_list_remove_request(req);
 		if (request_commit) {
 			kref_get(&req->wb_kref);
+			memcpy(&req->wb_verf, &hdr->verf.verifier,
+			       sizeof(req->wb_verf));
 			nfs_mark_request_commit(req, hdr->lseg, &cinfo,
 				hdr->ds_commit_idx);
 		}
Signature · 1 line
2.25.1
author maintainer
#403 lore raw

PATCH 382/388 xdp: Fix xsk_generic_xmit errno

In reply to Sasha Levin

From: Li RongQing <[email protected]>

[ Upstream commit aa2cad0600ed2ca6a0ab39948d4db1666b6c962b ]

Propagate sock_alloc_send_skb error code, not set it to EAGAIN unconditionally, when fail to allocate skb, which might cause that user space unnecessary loops.

---
 net/xdp/xsk.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)
net/xdp/xsk.c +1 -3
index c350108aa38d..a4676107fad0 100644
@@ -397,10 +397,8 @@ static int xsk_generic_xmit(struct sock *sk)
 
 		len = desc.len;
 		skb = sock_alloc_send_skb(sk, len, 1, &err);
-		if (unlikely(!skb)) {
-			err = -EAGAIN;
+		if (unlikely(!skb))
 			goto out;
-		}
 
 		skb_put(skb, len);
 		addr = desc.addr;
Signature · 1 line
2.25.1
author maintainer
#404 lore raw

PATCH 383/388 iavf: fix speed reporting over virtchnl

In reply to Sasha Levin

From: Brett Creeley <[email protected]>

[ Upstream commit e0ef26fbe2b0c62f42ba7667076dc38b693b6fb8 ]

Link speeds are communicated over virtchnl using an enum virtchnl_link_speed. Currently, the highest link speed is 40Gbps which leaves us unable to reflect some speeds that an ice VF is capable of. This causes link speed to be misreported on the iavf driver.

Allow for communicating link speeds using Mbps so that the proper speed can be reported for an ice VF. Moving away from the enum allows us to communicate future speed changes without requiring a new enum to be added.

In order to support communicating link speeds over virtchnl in Mbps the following functionality was added:
- Added u32 link_speed_mbps in the iavf_adapter structure.
- Added the macro ADV_LINK_SUPPORT(_a) to determine if the VF driver supports communicating link speeds in Mbps.
- Added the function iavf_get_vpe_link_status() to fill the correct link_status in the event_data union based on the ADV_LINK_SUPPORT(_a) macro.
- Added the function iavf_set_adapter_link_speed_from_vpe() to determine whether or not to fill the u32 link_speed_mbps or enum virtchnl_link_speed link_speed field in the iavf_adapter structure based on the ADV_LINK_SUPPORT(_a) macro.
- Do not free vf_res in iavf_init_get_resources() as vf_res will be accessed in iavf_get_link_ksettings(); memset to 0 instead. This memory is subsequently freed in iavf_remove().

Fixes 7c710869d64e ("ice: Add handlers for VF netdevice operations")
---
 drivers/net/ethernet/intel/iavf/iavf.h        | 14 +++
 .../net/ethernet/intel/iavf/iavf_ethtool.c    | 14 ++-
 drivers/net/ethernet/intel/iavf/iavf_main.c   | 25 ++++--
 .../net/ethernet/intel/iavf/iavf_virtchnl.c   | 88 ++++++++++++++++---
 4 files changed, 120 insertions(+), 21 deletions(-)
drivers/net/ethernet/intel/iavf/iavf.h +14 -0
index bcd11b4b29df..2d4ce6fdba1a 100644
@@ -87,6 +87,10 @@ struct iavf_vsi {
 #define IAVF_HLUT_ARRAY_SIZE ((IAVF_VFQF_HLUT_MAX_INDEX + 1) * 4)
 #define IAVF_MBPS_DIVISOR	125000 /* divisor to convert to Mbps */
 
+#define IAVF_VIRTCHNL_VF_RESOURCE_SIZE (sizeof(struct virtchnl_vf_resource) + \
+					(IAVF_MAX_VF_VSI * \
+					 sizeof(struct virtchnl_vsi_resource)))
+
 /* MAX_MSIX_Q_VECTORS of these are allocated,
  * but we only use one per queue-specific vector.
  */
@@ -306,6 +310,14 @@ struct iavf_adapter {
 	bool netdev_registered;
 	bool link_up;
 	enum virtchnl_link_speed link_speed;
+	/* This is only populated if the VIRTCHNL_VF_CAP_ADV_LINK_SPEED is set
+	 * in vf_res->vf_cap_flags. Use ADV_LINK_SUPPORT macro to determine if
+	 * this field is valid. This field should be used going forward and the
+	 * enum virtchnl_link_speed above should be considered the legacy way of
+	 * storing/communicating link speeds.
+	 */
+	u32 link_speed_mbps;
+
 	enum virtchnl_ops current_op;
 #define CLIENT_ALLOWED(_a) ((_a)->vf_res ? \
 			    (_a)->vf_res->vf_cap_flags & \
@@ -322,6 +334,8 @@ struct iavf_adapter {
 			VIRTCHNL_VF_OFFLOAD_RSS_PF)))
 #define VLAN_ALLOWED(_a) ((_a)->vf_res->vf_cap_flags & \
 			  VIRTCHNL_VF_OFFLOAD_VLAN)
+#define ADV_LINK_SUPPORT(_a) ((_a)->vf_res->vf_cap_flags & \
+			      VIRTCHNL_VF_CAP_ADV_LINK_SPEED)
 	struct virtchnl_vf_resource *vf_res; /* incl. all VSIs */
 	struct virtchnl_vsi_resource *vsi_res; /* our LAN VSI */
 	struct virtchnl_version_info pf_version;
drivers/net/ethernet/intel/iavf/iavf_ethtool.c +12 -2
index 2c39d46b6138..40a3fc7c5ea5 100644
@@ -278,7 +278,18 @@ static int iavf_get_link_ksettings(struct net_device *netdev,
 	ethtool_link_ksettings_zero_link_mode(cmd, supported);
 	cmd->base.autoneg = AUTONEG_DISABLE;
 	cmd->base.port = PORT_NONE;
-	/* Set speed and duplex */
+	cmd->base.duplex = DUPLEX_FULL;
+
+	if (ADV_LINK_SUPPORT(adapter)) {
+		if (adapter->link_speed_mbps &&
+		    adapter->link_speed_mbps < U32_MAX)
+			cmd->base.speed = adapter->link_speed_mbps;
+		else
+			cmd->base.speed = SPEED_UNKNOWN;
+
+		return 0;
+	}
+
 	switch (adapter->link_speed) {
 	case IAVF_LINK_SPEED_40GB:
 		cmd->base.speed = SPEED_40000;
@@ -306,7 +317,6 @@ static int iavf_get_link_ksettings(struct net_device *netdev,
 	default:
 		break;
 	}
-	cmd->base.duplex = DUPLEX_FULL;
 
 	return 0;
 }
drivers/net/ethernet/intel/iavf/iavf_main.c +18 -7
index 2050649848ba..a21ae74bcd1b 100644
@@ -1756,17 +1756,17 @@ static int iavf_init_get_resources(struct iavf_adapter *adapter)
 	struct net_device *netdev = adapter->netdev;
 	struct pci_dev *pdev = adapter->pdev;
 	struct iavf_hw *hw = &adapter->hw;
-	int err = 0, bufsz;
+	int err;
 
 	WARN_ON(adapter->state != __IAVF_INIT_GET_RESOURCES);
 	/* aq msg sent, awaiting reply */
 	if (!adapter->vf_res) {
-		bufsz = sizeof(struct virtchnl_vf_resource) +
-			(IAVF_MAX_VF_VSI *
-			sizeof(struct virtchnl_vsi_resource));
-		adapter->vf_res = kzalloc(bufsz, GFP_KERNEL);
-		if (!adapter->vf_res)
+		adapter->vf_res = kzalloc(IAVF_VIRTCHNL_VF_RESOURCE_SIZE,
+					  GFP_KERNEL);
+		if (!adapter->vf_res) {
+			err = -ENOMEM;
 			goto err;
+		}
 	}
 	err = iavf_get_vf_config(adapter);
 	if (err == IAVF_ERR_ADMIN_QUEUE_NO_WORK) {
@@ -2036,7 +2036,7 @@ static void iavf_disable_vf(struct iavf_adapter *adapter)
 	iavf_reset_interrupt_capability(adapter);
 	iavf_free_queues(adapter);
 	iavf_free_q_vectors(adapter);
-	kfree(adapter->vf_res);
+	memset(adapter->vf_res, 0, IAVF_VIRTCHNL_VF_RESOURCE_SIZE);
 	iavf_shutdown_adminq(&adapter->hw);
 	adapter->netdev->flags &= ~IFF_UP;
 	clear_bit(__IAVF_IN_CRITICAL_TASK, &adapter->crit_section);
@@ -2487,6 +2487,16 @@ static int iavf_validate_tx_bandwidth(struct iavf_adapter *adapter,
 {
 	int speed = 0, ret = 0;
 
+	if (ADV_LINK_SUPPORT(adapter)) {
+		if (adapter->link_speed_mbps < U32_MAX) {
+			speed = adapter->link_speed_mbps;
+			goto validate_bw;
+		} else {
+			dev_err(&adapter->pdev->dev, "Unknown link speed\n");
+			return -EINVAL;
+		}
+	}
+
 	switch (adapter->link_speed) {
 	case IAVF_LINK_SPEED_40GB:
 		speed = 40000;
@@ -2510,6 +2520,7 @@ static int iavf_validate_tx_bandwidth(struct iavf_adapter *adapter,
 		break;
 	}
 
+validate_bw:
 	if (max_tx_rate > speed) {
 		dev_err(&adapter->pdev->dev,
 			"Invalid tx rate specified\n");
drivers/net/ethernet/intel/iavf/iavf_virtchnl.c +76 -12

138 lines not on this page · raw message

Signature · 1 line
2.25.1
author maintainer
#405 lore raw

PATCH 384/388 net: ipa: program upper nibbles of sequencer type

In reply to Sasha Levin

From: Alex Elder <[email protected]>

[ Upstream commit 636edeaad5577b6023f0de2b98a010d1cea73607 ]

The upper two nibbles of the sequencer type were not used for SDM845, and were assumed to be 0. But for SC7180 they are used, and so they must be programmed by ipa_endpoint_init_seq(). Fix this bug.

IPA_SEQ_PKT_PROCESS_NO_DEC_NO_UCP_DMAP doesn't have a descriptive comment, so add one.

---
 drivers/net/ipa/ipa_endpoint.c | 6 ++++--
 drivers/net/ipa/ipa_reg.h      | 2 ++
 2 files changed, 6 insertions(+), 2 deletions(-)
drivers/net/ipa/ipa_endpoint.c +4 -2
index a21534f1462f..1d823ac0f6d6 100644
@@ -669,10 +669,12 @@ static void ipa_endpoint_init_seq(struct ipa_endpoint *endpoint)
 	u32 seq_type = endpoint->seq_type;
 	u32 val = 0;
 
+	/* Sequencer type is made up of four nibbles */
 	val |= u32_encode_bits(seq_type & 0xf, HPS_SEQ_TYPE_FMASK);
 	val |= u32_encode_bits((seq_type >> 4) & 0xf, DPS_SEQ_TYPE_FMASK);
-	/* HPS_REP_SEQ_TYPE is 0 */
-	/* DPS_REP_SEQ_TYPE is 0 */
+	/* The second two apply to replicated packets */
+	val |= u32_encode_bits((seq_type >> 8) & 0xf, HPS_REP_SEQ_TYPE_FMASK);
+	val |= u32_encode_bits((seq_type >> 12) & 0xf, DPS_REP_SEQ_TYPE_FMASK);
 
 	iowrite32(val, endpoint->ipa->reg_virt + offset);
 }
drivers/net/ipa/ipa_reg.h +2 -0
index 3b8106aa277a..0a688d8c1d7c 100644
@@ -455,6 +455,8 @@ enum ipa_mode {
  *	second packet processing pass + no decipher + microcontroller
  * @IPA_SEQ_DMA_DEC:		DMA + cipher/decipher
  * @IPA_SEQ_DMA_COMP_DECOMP:	DMA + compression/decompression
+ * @IPA_SEQ_PKT_PROCESS_NO_DEC_NO_UCP_DMAP:
+ *	packet processing + no decipher + no uCP + HPS REP DMA parser
  * @IPA_SEQ_INVALID:		invalid sequencer type
  *
  * The values defined here are broken into 4-bit nibbles that are written
Signature · 1 line
2.25.1
author maintainer
#406 lore raw

PATCH 385/388 bpf: sockmap: Don't attach programs to UDP sockets

In reply to Sasha Levin

From: Lorenz Bauer <[email protected]>

[ Upstream commit f6fede8569689dd31e7b0ed15024b25e5ce2e2e5 ]

The stream parser infrastructure isn't set up to deal with UDP sockets, so we mustn't try to attach programs to them.

I remember making this change at some point, but I must have lost it while rebasing or something similar.

---
 net/core/sock_map.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
net/core/sock_map.c +6 -4
index 050bfac97cfb..7e858c1dd711 100644
@@ -417,10 +417,7 @@ static int sock_map_get_next_key(struct bpf_map *map, void *key, void *next)
 	return 0;
 }
 
-static bool sock_map_redirect_allowed(const struct sock *sk)
-{
-	return sk->sk_state != TCP_LISTEN;
-}
+static bool sock_map_redirect_allowed(const struct sock *sk);
 
 static int sock_map_update_common(struct bpf_map *map, u32 idx,
 				  struct sock *sk, u64 flags)
@@ -501,6 +498,11 @@ static bool sk_is_udp(const struct sock *sk)
 	       sk->sk_protocol == IPPROTO_UDP;
 }
 
+static bool sock_map_redirect_allowed(const struct sock *sk)
+{
+	return sk_is_tcp(sk) && sk->sk_state != TCP_LISTEN;
+}
+
 static bool sock_map_sk_is_suitable(const struct sock *sk)
 {
 	return sk_is_tcp(sk) || sk_is_udp(sk);
Signature · 1 line
2.25.1
author maintainer
#407 lore raw

PATCH 386/388 bpf: Fix memlock accounting for sock_hash

In reply to Sasha Levin

From: Andrey Ignatov <[email protected]>

[ Upstream commit 60e5ca8a64bad8f3e2e20a1e57846e497361c700 ]

Add missed bpf_map_charge_init() in sock_hash_alloc() and correspondingly bpf_map_charge_finish() on ENOMEM.

It was found accidentally while working on unrelated selftest that checks "map->memory.pages > 0" is true for all map types.

Before:

	# bpftool m l
	...
	3692: sockhash  name m_sockhash  flags 0x0
		key 4B  value 4B  max_entries 8  memlock 0B

After:

	# bpftool m l
	...
	84: sockmap  name m_sockmap  flags 0x0
		key 4B  value 4B  max_entries 8  memlock 4096B
Fixes 604326b41a6f ("bpf, sockmap: convert to generic sk_msg interface")
---
 net/core/sock_map.c | 4 ++++
 1 file changed, 4 insertions(+)
net/core/sock_map.c +4 -0
index 7e858c1dd711..591457fcbd02 100644
@@ -984,11 +984,15 @@ static struct bpf_map *sock_hash_alloc(union bpf_attr *attr)
 		err = -EINVAL;
 		goto free_htab;
 	}
+	err = bpf_map_charge_init(&htab->map.memory, cost);
+	if (err)
+		goto free_htab;
 
 	htab->buckets = bpf_map_area_alloc(htab->buckets_num *
 					   sizeof(struct bpf_htab_bucket),
 					   htab->map.numa_node);
 	if (!htab->buckets) {
+		bpf_map_charge_finish(&htab->map.memory);
 		err = -ENOMEM;
 		goto free_htab;
 	}
Signature · 1 line
2.25.1
author maintainer
#408 lore raw

PATCH 387/388 libbpf: Support pre-initializing .bss global variables

In reply to Sasha Levin

From: Andrii Nakryiko <[email protected]>

[ Upstream commit caf62492f479585296e9d636c798d5ac256b7b04 ]

Remove invalid assumption in libbpf that .bss map doesn't have to be updated in kernel. With addition of skeleton and memory-mapped initialization image, .bss doesn't have to be all zeroes when BPF map is created, because user-code might have initialized those variables from user-space.

Fixes eba9c5f498a1 ("libbpf: Refactor global data map initialization")
---
 tools/lib/bpf/libbpf.c                        |  4 --
 .../selftests/bpf/prog_tests/skeleton.c       | 45 ++++++++++++++++---
 .../selftests/bpf/progs/test_skeleton.c       | 19 ++++++--
 3 files changed, 55 insertions(+), 13 deletions(-)
tools/lib/bpf/libbpf.c +0 -4
index 8f480e29a6b0..7d80bb1fea4c 100644
@@ -3455,10 +3455,6 @@ bpf_object__populate_internal_map(struct bpf_object *obj, struct bpf_map *map)
 	char *cp, errmsg[STRERR_BUFSIZE];
 	int err, zero = 0;
 
-	/* kernel already zero-initializes .bss map. */
-	if (map_type == LIBBPF_MAP_BSS)
-		return 0;
-
 	err = bpf_map_update_elem(map->fd, &zero, map->mmaped, 0);
 	if (err) {
 		err = -errno;
tools/testing/selftests/bpf/prog_tests/skeleton.c +40 -5

73 lines not on this page · raw message

tools/testing/selftests/bpf/progs/test_skeleton.c +15 -4
index de03a90f78ca..77ae86f44db5 100644
@@ -10,16 +10,26 @@ struct s {
 	long long b;
 } __attribute__((packed));
 
-int in1 = 0;
-long long in2 = 0;
+/* .data section */
+int in1 = -1;
+long long in2 = -1;
+
+/* .bss section */
 char in3 = '\0';
 long long in4 __attribute__((aligned(64))) = 0;
 struct s in5 = {};
 
-long long out2 = 0;
+/* .rodata section */
+const volatile int in6 = 0;
+
+/* .data section */
+int out1 = -1;
+long long out2 = -1;
+
+/* .bss section */
 char out3 = 0;
 long long out4 = 0;
-int out1 = 0;
+int out6 = 0;
 
 extern bool CONFIG_BPF_SYSCALL __kconfig;
 extern int LINUX_KERNEL_VERSION __kconfig;
@@ -36,6 +46,7 @@ int handler(const void *ctx)
 	out3 = in3;
 	out4 = in4;
 	out5 = in5;
+	out6 = in6;
 
 	bpf_syscall = CONFIG_BPF_SYSCALL;
 	kern_ver = LINUX_KERNEL_VERSION;
Signature · 1 line
2.25.1
author maintainer
#409 lore raw

PATCH 388/388 bpf: Undo internal BPF_PROBE_MEM in BPF insns dump

In reply to Sasha Levin

From: Andrii Nakryiko <[email protected]>

[ Upstream commit 29fcb05bbf1a7008900bb9bee347bdbfc7171036 ]

BPF_PROBE_MEM is kernel-internal implmementation details. When dumping BPF instructions to user-space, it needs to be replaced back with BPF_MEM mode.

Fixes 2a02759ef5f8 ("bpf: Add support for BTF pointers to interpreter")
---
 kernel/bpf/syscall.c | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)
kernel/bpf/syscall.c +12 -5
index 4e6dee19a668..5d1d24f56d53 100644
@@ -2923,6 +2923,7 @@ static struct bpf_insn *bpf_insn_prepare_dump(const struct bpf_prog *prog)
 	struct bpf_insn *insns;
 	u32 off, type;
 	u64 imm;
+	u8 code;
 	int i;
 
 	insns = kmemdup(prog->insnsi, bpf_prog_insn_size(prog),
@@ -2931,21 +2932,27 @@ static struct bpf_insn *bpf_insn_prepare_dump(const struct bpf_prog *prog)
 		return insns;
 
 	for (i = 0; i < prog->len; i++) {
-		if (insns[i].code == (BPF_JMP | BPF_TAIL_CALL)) {
+		code = insns[i].code;
+
+		if (code == (BPF_JMP | BPF_TAIL_CALL)) {
 			insns[i].code = BPF_JMP | BPF_CALL;
 			insns[i].imm = BPF_FUNC_tail_call;
 			/* fall-through */
 		}
-		if (insns[i].code == (BPF_JMP | BPF_CALL) ||
-		    insns[i].code == (BPF_JMP | BPF_CALL_ARGS)) {
-			if (insns[i].code == (BPF_JMP | BPF_CALL_ARGS))
+		if (code == (BPF_JMP | BPF_CALL) ||
+		    code == (BPF_JMP | BPF_CALL_ARGS)) {
+			if (code == (BPF_JMP | BPF_CALL_ARGS))
 				insns[i].code = BPF_JMP | BPF_CALL;
 			if (!bpf_dump_raw_ok())
 				insns[i].imm = 0;
 			continue;
 		}
+		if (BPF_CLASS(code) == BPF_LDX && BPF_MODE(code) == BPF_PROBE_MEM) {
+			insns[i].code = BPF_LDX | BPF_SIZE(code) | BPF_MEM;
+			continue;
+		}
 
-		if (insns[i].code != (BPF_LD | BPF_IMM | BPF_DW))
+		if (code != (BPF_LD | BPF_IMM | BPF_DW))
 			continue;
 
 		imm = ((u64)insns[i + 1].imm << 32) | (u32)insns[i].imm;
Signature · 1 line
2.25.1