From 6ca53a75c425cb1cb0929ce07f58d9e41094500c Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Fri, 9 Sep 2022 11:20:21 +0200 Subject: [PATCH 001/138] dt-bindings: hwlock: qcom-hwspinlock: add support for MMIO on older SoCs Older Qualcomm SoCs have TCSR mutex registers with 0x80 stride, instead of 0x1000. Add dedicated compatibles for such case. Unfortunately the binding started using a generic "qcom,tcsr-mutex" compatible without specifying the SoC part, thus it looks now quite inconsistent. Signed-off-by: Krzysztof Kozlowski Acked-by: Rob Herring Signed-off-by: Bjorn Andersson Link: https://lore.kernel.org/r/20220909092035.223915-2-krzysztof.kozlowski@linaro.org --- .../bindings/hwlock/qcom-hwspinlock.yaml | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml b/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml index 1c7149f7d171..de98b961fb38 100644 --- a/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml +++ b/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml @@ -15,9 +15,18 @@ description: properties: compatible: - enum: - - qcom,sfpb-mutex - - qcom,tcsr-mutex + oneOf: + - enum: + - qcom,sfpb-mutex + - qcom,tcsr-mutex + - items: + - enum: + - qcom,apq8084-tcsr-mutex + - qcom,ipq6018-tcsr-mutex + - qcom,msm8226-tcsr-mutex + - qcom,msm8974-tcsr-mutex + - qcom,msm8994-tcsr-mutex + - const: qcom,tcsr-mutex reg: maxItems: 1 From 276a4f1a5fb118bfd6980d42732d530e43f2916a Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Fri, 9 Sep 2022 11:20:22 +0200 Subject: [PATCH 002/138] dt-bindings: hwlock: qcom-hwspinlock: correct example indentation Use some consistent indentation (4-space) for DTS example. Signed-off-by: Krzysztof Kozlowski Acked-by: Rob Herring Signed-off-by: Bjorn Andersson Link: https://lore.kernel.org/r/20220909092035.223915-3-krzysztof.kozlowski@linaro.org --- .../devicetree/bindings/hwlock/qcom-hwspinlock.yaml | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml b/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml index de98b961fb38..1a3adf75934b 100644 --- a/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml +++ b/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml @@ -43,9 +43,9 @@ additionalProperties: false examples: - | - tcsr_mutex: hwlock@1f40000 { - compatible = "qcom,tcsr-mutex"; - reg = <0x01f40000 0x40000>; - #hwlock-cells = <1>; - }; + hwlock@1f40000 { + compatible = "qcom,tcsr-mutex"; + reg = <0x01f40000 0x40000>; + #hwlock-cells = <1>; + }; ... From 90cb380f9ceb811059340d06ff5fd0c0e93ecbe1 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Fri, 9 Sep 2022 11:20:23 +0200 Subject: [PATCH 003/138] hwspinlock: qcom: correct MMIO max register for newer SoCs Newer ARMv8 Qualcomm SoCs using 0x1000 register stride have maximum register 0x20000 (32 mutexes * 0x1000). Fixes: 7a1e6fb1c606 ("hwspinlock: qcom: Allow mmio usage in addition to syscon") Signed-off-by: Krzysztof Kozlowski Reviewed-by: Konrad Dybcio Signed-off-by: Bjorn Andersson Link: https://lore.kernel.org/r/20220909092035.223915-4-krzysztof.kozlowski@linaro.org --- drivers/hwspinlock/qcom_hwspinlock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/hwspinlock/qcom_hwspinlock.c b/drivers/hwspinlock/qcom_hwspinlock.c index 80ea45b3a815..9734e149d981 100644 --- a/drivers/hwspinlock/qcom_hwspinlock.c +++ b/drivers/hwspinlock/qcom_hwspinlock.c @@ -121,7 +121,7 @@ static const struct regmap_config tcsr_mutex_config = { .reg_bits = 32, .reg_stride = 4, .val_bits = 32, - .max_register = 0x40000, + .max_register = 0x20000, .fast_io = true, }; From 5d4753f741d824e04e7ba46f46ec016be120f383 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Fri, 9 Sep 2022 11:20:24 +0200 Subject: [PATCH 004/138] hwspinlock: qcom: add support for MMIO on older SoCs Older Qualcomm SoCs have TCSR mutex registers with 0x80 stride, instead of 0x1000. Add dedicated compatibles and regmap for such case. Signed-off-by: Krzysztof Kozlowski Signed-off-by: Bjorn Andersson Link: https://lore.kernel.org/r/20220909092035.223915-5-krzysztof.kozlowski@linaro.org --- drivers/hwspinlock/qcom_hwspinlock.c | 42 +++++++++++++++++++++------- 1 file changed, 32 insertions(+), 10 deletions(-) diff --git a/drivers/hwspinlock/qcom_hwspinlock.c b/drivers/hwspinlock/qcom_hwspinlock.c index 9734e149d981..9cf186362ae2 100644 --- a/drivers/hwspinlock/qcom_hwspinlock.c +++ b/drivers/hwspinlock/qcom_hwspinlock.c @@ -22,6 +22,7 @@ struct qcom_hwspinlock_of_data { u32 offset; u32 stride; + const struct regmap_config *regmap_config; }; static int qcom_hwspinlock_trylock(struct hwspinlock *lock) @@ -73,15 +74,42 @@ static const struct qcom_hwspinlock_of_data of_sfpb_mutex = { .stride = 0x4, }; -/* All modern platform has offset 0 and stride of 4k */ +static const struct regmap_config tcsr_msm8226_mutex_config = { + .reg_bits = 32, + .reg_stride = 4, + .val_bits = 32, + .max_register = 0x1000, + .fast_io = true, +}; + +static const struct qcom_hwspinlock_of_data of_msm8226_tcsr_mutex = { + .offset = 0, + .stride = 0x80, + .regmap_config = &tcsr_msm8226_mutex_config, +}; + +static const struct regmap_config tcsr_mutex_config = { + .reg_bits = 32, + .reg_stride = 4, + .val_bits = 32, + .max_register = 0x20000, + .fast_io = true, +}; + static const struct qcom_hwspinlock_of_data of_tcsr_mutex = { .offset = 0, .stride = 0x1000, + .regmap_config = &tcsr_mutex_config, }; static const struct of_device_id qcom_hwspinlock_of_match[] = { { .compatible = "qcom,sfpb-mutex", .data = &of_sfpb_mutex }, { .compatible = "qcom,tcsr-mutex", .data = &of_tcsr_mutex }, + { .compatible = "qcom,apq8084-tcsr-mutex", .data = &of_msm8226_tcsr_mutex }, + { .compatible = "qcom,ipq6018-tcsr-mutex", .data = &of_msm8226_tcsr_mutex }, + { .compatible = "qcom,msm8226-tcsr-mutex", .data = &of_msm8226_tcsr_mutex }, + { .compatible = "qcom,msm8974-tcsr-mutex", .data = &of_msm8226_tcsr_mutex }, + { .compatible = "qcom,msm8994-tcsr-mutex", .data = &of_msm8226_tcsr_mutex }, { } }; MODULE_DEVICE_TABLE(of, qcom_hwspinlock_of_match); @@ -117,14 +145,6 @@ static struct regmap *qcom_hwspinlock_probe_syscon(struct platform_device *pdev, return regmap; } -static const struct regmap_config tcsr_mutex_config = { - .reg_bits = 32, - .reg_stride = 4, - .val_bits = 32, - .max_register = 0x20000, - .fast_io = true, -}; - static struct regmap *qcom_hwspinlock_probe_mmio(struct platform_device *pdev, u32 *offset, u32 *stride) { @@ -133,6 +153,8 @@ static struct regmap *qcom_hwspinlock_probe_mmio(struct platform_device *pdev, void __iomem *base; data = of_device_get_match_data(dev); + if (!data->regmap_config) + return ERR_PTR(-EINVAL); *offset = data->offset; *stride = data->stride; @@ -141,7 +163,7 @@ static struct regmap *qcom_hwspinlock_probe_mmio(struct platform_device *pdev, if (IS_ERR(base)) return ERR_CAST(base); - return devm_regmap_init_mmio(dev, base, &tcsr_mutex_config); + return devm_regmap_init_mmio(dev, base, data->regmap_config); } static int qcom_hwspinlock_probe(struct platform_device *pdev) From 76845ba539c3693f572377b38563d19a87266a5b Mon Sep 17 00:00:00 2001 From: Kishon Vijay Abraham I Date: Fri, 14 Oct 2022 11:54:24 +0530 Subject: [PATCH 005/138] MAINTAINERS: Update Kishon's email address in GENERIC PHY FRAMEWORK Update Kishon's email address in GENERIC PHY FRAMEWORK maintainer entry. Cc: Kishon Vijay Abraham I Signed-off-by: Kishon Vijay Abraham I Link: https://lore.kernel.org/r/20221014062424.3327-1-kishon@ti.com Signed-off-by: Vinod Koul --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index cf0f18502372..2526755bf977 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -8595,8 +8595,8 @@ F: include/asm-generic/ F: include/uapi/asm-generic/ GENERIC PHY FRAMEWORK -M: Kishon Vijay Abraham I M: Vinod Koul +M: Kishon Vijay Abraham I L: linux-phy@lists.infradead.org S: Supported Q: https://patchwork.kernel.org/project/linux-phy/list/ From ca1c73628f5bd0c1ef6e46073cc3be2450605b06 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 14 Oct 2022 12:25:06 +0300 Subject: [PATCH 006/138] phy: stm32: fix an error code in probe If "index > usbphyc->nphys" is true then this returns success but it should return -EINVAL. Fixes: 94c358da3a05 ("phy: stm32: add support for STM32 USB PHY Controller (USBPHYC)") Signed-off-by: Dan Carpenter Reviewed-by: Amelie Delaunay Link: https://lore.kernel.org/r/Y0kq8j6S+5nDdMpr@kili Signed-off-by: Vinod Koul --- drivers/phy/st/phy-stm32-usbphyc.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/phy/st/phy-stm32-usbphyc.c b/drivers/phy/st/phy-stm32-usbphyc.c index a98c911cc37a..5bb9647b078f 100644 --- a/drivers/phy/st/phy-stm32-usbphyc.c +++ b/drivers/phy/st/phy-stm32-usbphyc.c @@ -710,6 +710,8 @@ static int stm32_usbphyc_probe(struct platform_device *pdev) ret = of_property_read_u32(child, "reg", &index); if (ret || index > usbphyc->nphys) { dev_err(&phy->dev, "invalid reg property: %d\n", ret); + if (!ret) + ret = -EINVAL; goto put_child; } From 2a4ea83bb8e54986703187edce59c9130a75eb26 Mon Sep 17 00:00:00 2001 From: Jon Hunter Date: Mon, 10 Oct 2022 14:51:32 +0100 Subject: [PATCH 007/138] phy: tegra: xusb: Fix crash during pad power on/down Commit a88520bfc0ec ("usb: gadget: tegra: Reduce pad power") added calls to tegra_phy_xusb_utmi_pad_power_on/down in the Tegra XUDC driver to control the pad power. This change is causing a kernel panic when powering down the pads on entering suspend with the Jetson TX2 platform. The panic occurs because the 'xudc->curr_utmi_phy' is not configured on this platform and we do not check to see if the pointer is valid before attempting to deference the pointer. Fix this by checking to see if the 'phy' pointer passed to tegra_phy_xusb_utmi_pad_power_on/down is valid. Fixes: a88520bfc0ec ("usb: gadget: tegra: Reduce pad power") Signed-off-by: Jon Hunter Link: https://lore.kernel.org/r/20221010135132.30809-1-jonathanh@nvidia.com Signed-off-by: Vinod Koul --- drivers/phy/tegra/xusb.c | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-) diff --git a/drivers/phy/tegra/xusb.c b/drivers/phy/tegra/xusb.c index 95091876c422..dce45fbbd699 100644 --- a/drivers/phy/tegra/xusb.c +++ b/drivers/phy/tegra/xusb.c @@ -1461,8 +1461,14 @@ EXPORT_SYMBOL_GPL(tegra_phy_xusb_utmi_port_reset); void tegra_phy_xusb_utmi_pad_power_on(struct phy *phy) { - struct tegra_xusb_lane *lane = phy_get_drvdata(phy); - struct tegra_xusb_padctl *padctl = lane->pad->padctl; + struct tegra_xusb_lane *lane; + struct tegra_xusb_padctl *padctl; + + if (!phy) + return; + + lane = phy_get_drvdata(phy); + padctl = lane->pad->padctl; if (padctl->soc->ops->utmi_pad_power_on) padctl->soc->ops->utmi_pad_power_on(phy); @@ -1471,8 +1477,14 @@ EXPORT_SYMBOL_GPL(tegra_phy_xusb_utmi_pad_power_on); void tegra_phy_xusb_utmi_pad_power_down(struct phy *phy) { - struct tegra_xusb_lane *lane = phy_get_drvdata(phy); - struct tegra_xusb_padctl *padctl = lane->pad->padctl; + struct tegra_xusb_lane *lane; + struct tegra_xusb_padctl *padctl; + + if (!phy) + return; + + lane = phy_get_drvdata(phy); + padctl = lane->pad->padctl; if (padctl->soc->ops->utmi_pad_power_down) padctl->soc->ops->utmi_pad_power_down(phy); From c29f446108e10bb791220ffbba54d82722d60d45 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Tue, 20 Sep 2022 17:04:12 +0200 Subject: [PATCH 008/138] dt-bindings: hwlock: qcom-hwspinlock: add syscon to MSM8974 The TCSR_MUTEX region contains two set of registers: mutex and halt. Add syscon, so the TCSR mutex device (hwspinlock) can use MMIO based method and in the same time share regmap with other devices for the halt regs. Signed-off-by: Krzysztof Kozlowski Acked-by: Rob Herring Tested-by: Luca Weiss # fairphone-fp2 Signed-off-by: Bjorn Andersson Link: https://lore.kernel.org/r/20220920150414.637634-2-krzysztof.kozlowski@linaro.org --- .../devicetree/bindings/hwlock/qcom-hwspinlock.yaml | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml b/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml index 1a3adf75934b..ee2726149cf3 100644 --- a/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml +++ b/Documentation/devicetree/bindings/hwlock/qcom-hwspinlock.yaml @@ -24,9 +24,13 @@ properties: - qcom,apq8084-tcsr-mutex - qcom,ipq6018-tcsr-mutex - qcom,msm8226-tcsr-mutex - - qcom,msm8974-tcsr-mutex - qcom,msm8994-tcsr-mutex - const: qcom,tcsr-mutex + - items: + - enum: + - qcom,msm8974-tcsr-mutex + - const: qcom,tcsr-mutex + - const: syscon reg: maxItems: 1 From 66063033f77e10b985258126a97573f84bb8d3b4 Mon Sep 17 00:00:00 2001 From: "Jason A. Donenfeld" Date: Wed, 19 Oct 2022 09:55:41 -0600 Subject: [PATCH 009/138] wifi: rt2x00: use explicitly signed or unsigned types On some platforms, `char` is unsigned, but this driver, for the most part, assumed it was signed. In other places, it uses `char` to mean an unsigned number, but only in cases when the values are small. And in still other places, `char` is used as a boolean. Put an end to this confusion by declaring explicit types, depending on the context. Cc: Andrew Morton Cc: Andy Shevchenko Cc: Stanislaw Gruszka Cc: Helmut Schaa Cc: Kalle Valo Signed-off-by: Jason A. Donenfeld Acked-by: Stanislaw Gruszka Signed-off-by: Kalle Valo Link: https://lore.kernel.org/r/20221019155541.3410813-1-Jason@zx2c4.com --- .../net/wireless/ralink/rt2x00/rt2400pci.c | 8 +-- .../net/wireless/ralink/rt2x00/rt2400pci.h | 2 +- .../net/wireless/ralink/rt2x00/rt2500pci.c | 8 +-- .../net/wireless/ralink/rt2x00/rt2500pci.h | 2 +- .../net/wireless/ralink/rt2x00/rt2500usb.c | 8 +-- .../net/wireless/ralink/rt2x00/rt2500usb.h | 2 +- .../net/wireless/ralink/rt2x00/rt2800lib.c | 60 +++++++++---------- .../net/wireless/ralink/rt2x00/rt2800lib.h | 8 +-- .../net/wireless/ralink/rt2x00/rt2x00usb.c | 6 +- drivers/net/wireless/ralink/rt2x00/rt61pci.c | 4 +- drivers/net/wireless/ralink/rt2x00/rt61pci.h | 2 +- drivers/net/wireless/ralink/rt2x00/rt73usb.c | 4 +- drivers/net/wireless/ralink/rt2x00/rt73usb.h | 2 +- 13 files changed, 58 insertions(+), 58 deletions(-) diff --git a/drivers/net/wireless/ralink/rt2x00/rt2400pci.c b/drivers/net/wireless/ralink/rt2x00/rt2400pci.c index 273c5eac3362..ddfc16de1b26 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2400pci.c +++ b/drivers/net/wireless/ralink/rt2x00/rt2400pci.c @@ -1023,9 +1023,9 @@ static int rt2400pci_set_state(struct rt2x00_dev *rt2x00dev, { u32 reg, reg2; unsigned int i; - char put_to_sleep; - char bbp_state; - char rf_state; + bool put_to_sleep; + u8 bbp_state; + u8 rf_state; put_to_sleep = (state != STATE_AWAKE); @@ -1561,7 +1561,7 @@ static int rt2400pci_probe_hw_mode(struct rt2x00_dev *rt2x00dev) { struct hw_mode_spec *spec = &rt2x00dev->spec; struct channel_info *info; - char *tx_power; + u8 *tx_power; unsigned int i; /* diff --git a/drivers/net/wireless/ralink/rt2x00/rt2400pci.h b/drivers/net/wireless/ralink/rt2x00/rt2400pci.h index b8187b6de143..979d5fd8babf 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2400pci.h +++ b/drivers/net/wireless/ralink/rt2x00/rt2400pci.h @@ -939,7 +939,7 @@ #define DEFAULT_TXPOWER 39 #define __CLAMP_TX(__txpower) \ - clamp_t(char, (__txpower), MIN_TXPOWER, MAX_TXPOWER) + clamp_t(u8, (__txpower), MIN_TXPOWER, MAX_TXPOWER) #define TXPOWER_FROM_DEV(__txpower) \ ((__CLAMP_TX(__txpower) - MAX_TXPOWER) + MIN_TXPOWER) diff --git a/drivers/net/wireless/ralink/rt2x00/rt2500pci.c b/drivers/net/wireless/ralink/rt2x00/rt2500pci.c index 8faa0a80e73a..cd6371e25062 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2500pci.c +++ b/drivers/net/wireless/ralink/rt2x00/rt2500pci.c @@ -1176,9 +1176,9 @@ static int rt2500pci_set_state(struct rt2x00_dev *rt2x00dev, { u32 reg, reg2; unsigned int i; - char put_to_sleep; - char bbp_state; - char rf_state; + bool put_to_sleep; + u8 bbp_state; + u8 rf_state; put_to_sleep = (state != STATE_AWAKE); @@ -1856,7 +1856,7 @@ static int rt2500pci_probe_hw_mode(struct rt2x00_dev *rt2x00dev) { struct hw_mode_spec *spec = &rt2x00dev->spec; struct channel_info *info; - char *tx_power; + u8 *tx_power; unsigned int i; /* diff --git a/drivers/net/wireless/ralink/rt2x00/rt2500pci.h b/drivers/net/wireless/ralink/rt2x00/rt2500pci.h index 7e64aee2a172..ba362675c52c 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2500pci.h +++ b/drivers/net/wireless/ralink/rt2x00/rt2500pci.h @@ -1219,6 +1219,6 @@ (((u8)(__txpower)) > MAX_TXPOWER) ? DEFAULT_TXPOWER : (__txpower) #define TXPOWER_TO_DEV(__txpower) \ - clamp_t(char, __txpower, MIN_TXPOWER, MAX_TXPOWER) + clamp_t(u8, __txpower, MIN_TXPOWER, MAX_TXPOWER) #endif /* RT2500PCI_H */ diff --git a/drivers/net/wireless/ralink/rt2x00/rt2500usb.c b/drivers/net/wireless/ralink/rt2x00/rt2500usb.c index bb5ed6630645..4f3b0e6c6256 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2500usb.c +++ b/drivers/net/wireless/ralink/rt2x00/rt2500usb.c @@ -984,9 +984,9 @@ static int rt2500usb_set_state(struct rt2x00_dev *rt2x00dev, u16 reg; u16 reg2; unsigned int i; - char put_to_sleep; - char bbp_state; - char rf_state; + bool put_to_sleep; + u8 bbp_state; + u8 rf_state; put_to_sleep = (state != STATE_AWAKE); @@ -1663,7 +1663,7 @@ static int rt2500usb_probe_hw_mode(struct rt2x00_dev *rt2x00dev) { struct hw_mode_spec *spec = &rt2x00dev->spec; struct channel_info *info; - char *tx_power; + u8 *tx_power; unsigned int i; /* diff --git a/drivers/net/wireless/ralink/rt2x00/rt2500usb.h b/drivers/net/wireless/ralink/rt2x00/rt2500usb.h index 0c070288a140..746f0e950b76 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2500usb.h +++ b/drivers/net/wireless/ralink/rt2x00/rt2500usb.h @@ -839,6 +839,6 @@ (((u8)(__txpower)) > MAX_TXPOWER) ? DEFAULT_TXPOWER : (__txpower) #define TXPOWER_TO_DEV(__txpower) \ - clamp_t(char, __txpower, MIN_TXPOWER, MAX_TXPOWER) + clamp_t(u8, __txpower, MIN_TXPOWER, MAX_TXPOWER) #endif /* RT2500USB_H */ diff --git a/drivers/net/wireless/ralink/rt2x00/rt2800lib.c b/drivers/net/wireless/ralink/rt2x00/rt2800lib.c index cbbb1a4849cf..12b700c7b9c3 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2800lib.c +++ b/drivers/net/wireless/ralink/rt2x00/rt2800lib.c @@ -3372,10 +3372,10 @@ static void rt2800_config_channel_rf53xx(struct rt2x00_dev *rt2x00dev, if (rt2x00_has_cap_bt_coexist(rt2x00dev)) { if (rt2x00_rt_rev_gte(rt2x00dev, RT5390, REV_RT5390F)) { /* r55/r59 value array of channel 1~14 */ - static const char r55_bt_rev[] = {0x83, 0x83, + static const u8 r55_bt_rev[] = {0x83, 0x83, 0x83, 0x73, 0x73, 0x63, 0x53, 0x53, 0x53, 0x43, 0x43, 0x43, 0x43, 0x43}; - static const char r59_bt_rev[] = {0x0e, 0x0e, + static const u8 r59_bt_rev[] = {0x0e, 0x0e, 0x0e, 0x0e, 0x0e, 0x0b, 0x0a, 0x09, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07}; @@ -3384,7 +3384,7 @@ static void rt2800_config_channel_rf53xx(struct rt2x00_dev *rt2x00dev, rt2800_rfcsr_write(rt2x00dev, 59, r59_bt_rev[idx]); } else { - static const char r59_bt[] = {0x8b, 0x8b, 0x8b, + static const u8 r59_bt[] = {0x8b, 0x8b, 0x8b, 0x8b, 0x8b, 0x8b, 0x8b, 0x8a, 0x89, 0x88, 0x88, 0x86, 0x85, 0x84}; @@ -3392,10 +3392,10 @@ static void rt2800_config_channel_rf53xx(struct rt2x00_dev *rt2x00dev, } } else { if (rt2x00_rt_rev_gte(rt2x00dev, RT5390, REV_RT5390F)) { - static const char r55_nonbt_rev[] = {0x23, 0x23, + static const u8 r55_nonbt_rev[] = {0x23, 0x23, 0x23, 0x23, 0x13, 0x13, 0x03, 0x03, 0x03, 0x03, 0x03, 0x03, 0x03, 0x03}; - static const char r59_nonbt_rev[] = {0x07, 0x07, + static const u8 r59_nonbt_rev[] = {0x07, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07, 0x06, 0x05, 0x04, 0x04}; @@ -3406,14 +3406,14 @@ static void rt2800_config_channel_rf53xx(struct rt2x00_dev *rt2x00dev, } else if (rt2x00_rt(rt2x00dev, RT5390) || rt2x00_rt(rt2x00dev, RT5392) || rt2x00_rt(rt2x00dev, RT6352)) { - static const char r59_non_bt[] = {0x8f, 0x8f, + static const u8 r59_non_bt[] = {0x8f, 0x8f, 0x8f, 0x8f, 0x8f, 0x8f, 0x8f, 0x8d, 0x8a, 0x88, 0x88, 0x87, 0x87, 0x86}; rt2800_rfcsr_write(rt2x00dev, 59, r59_non_bt[idx]); } else if (rt2x00_rt(rt2x00dev, RT5350)) { - static const char r59_non_bt[] = {0x0b, 0x0b, + static const u8 r59_non_bt[] = {0x0b, 0x0b, 0x0b, 0x0b, 0x0b, 0x0b, 0x0b, 0x0a, 0x0a, 0x09, 0x08, 0x07, 0x07, 0x06}; @@ -4035,23 +4035,23 @@ static void rt2800_iq_calibrate(struct rt2x00_dev *rt2x00dev, int channel) rt2800_bbp_write(rt2x00dev, 159, cal != 0xff ? cal : 0); } -static char rt2800_txpower_to_dev(struct rt2x00_dev *rt2x00dev, +static s8 rt2800_txpower_to_dev(struct rt2x00_dev *rt2x00dev, unsigned int channel, - char txpower) + s8 txpower) { if (rt2x00_rt(rt2x00dev, RT3593) || rt2x00_rt(rt2x00dev, RT3883)) txpower = rt2x00_get_field8(txpower, EEPROM_TXPOWER_ALC); if (channel <= 14) - return clamp_t(char, txpower, MIN_G_TXPOWER, MAX_G_TXPOWER); + return clamp_t(s8, txpower, MIN_G_TXPOWER, MAX_G_TXPOWER); if (rt2x00_rt(rt2x00dev, RT3593) || rt2x00_rt(rt2x00dev, RT3883)) - return clamp_t(char, txpower, MIN_A_TXPOWER_3593, + return clamp_t(s8, txpower, MIN_A_TXPOWER_3593, MAX_A_TXPOWER_3593); else - return clamp_t(char, txpower, MIN_A_TXPOWER, MAX_A_TXPOWER); + return clamp_t(s8, txpower, MIN_A_TXPOWER, MAX_A_TXPOWER); } static void rt3883_bbp_adjust(struct rt2x00_dev *rt2x00dev, @@ -8530,7 +8530,7 @@ static void rt2800_r_calibration(struct rt2x00_dev *rt2x00dev) u8 bytevalue = 0; int rcalcode; u8 r_cal_code = 0; - char d1 = 0, d2 = 0; + s8 d1 = 0, d2 = 0; u8 rfvalue; u32 MAC_RF_BYPASS0, MAC_RF_CONTROL0, MAC_PWR_PIN_CFG; u32 maccfg; @@ -8591,7 +8591,7 @@ static void rt2800_r_calibration(struct rt2x00_dev *rt2x00dev) if (bytevalue > 128) d1 = bytevalue - 256; else - d1 = (char)bytevalue; + d1 = (s8)bytevalue; rt2800_bbp_write(rt2x00dev, 22, 0x0); rt2800_rfcsr_write_bank(rt2x00dev, 0, 35, 0x01); @@ -8601,7 +8601,7 @@ static void rt2800_r_calibration(struct rt2x00_dev *rt2x00dev) if (bytevalue > 128) d2 = bytevalue - 256; else - d2 = (char)bytevalue; + d2 = (s8)bytevalue; rt2800_bbp_write(rt2x00dev, 22, 0x0); rcalcode = rt2800_calcrcalibrationcode(rt2x00dev, d1, d2); @@ -8703,7 +8703,7 @@ static void rt2800_rxdcoc_calibration(struct rt2x00_dev *rt2x00dev) static u32 rt2800_do_sqrt_accumulation(u32 si) { u32 root, root_pre, bit; - char i; + s8 i; bit = 1 << 15; root = 0; @@ -9330,11 +9330,11 @@ static void rt2800_loft_search(struct rt2x00_dev *rt2x00dev, u8 ch_idx, u8 alc_idx, u8 dc_result[][RF_ALC_NUM][2]) { u32 p0 = 0, p1 = 0, pf = 0; - char idx0 = 0, idx1 = 0; + s8 idx0 = 0, idx1 = 0; u8 idxf[] = {0x00, 0x00}; u8 ibit = 0x20; u8 iorq; - char bidx; + s8 bidx; rt2800_bbp_write(rt2x00dev, 158, 0xb0); rt2800_bbp_write(rt2x00dev, 159, 0x80); @@ -9384,17 +9384,17 @@ static void rt2800_loft_search(struct rt2x00_dev *rt2x00dev, u8 ch_idx, static void rt2800_iq_search(struct rt2x00_dev *rt2x00dev, u8 ch_idx, u8 *ges, u8 *pes) { u32 p0 = 0, p1 = 0, pf = 0; - char perr = 0, gerr = 0, iq_err = 0; - char pef = 0, gef = 0; - char psta, pend; - char gsta, gend; + s8 perr = 0, gerr = 0, iq_err = 0; + s8 pef = 0, gef = 0; + s8 psta, pend; + s8 gsta, gend; u8 ibit = 0x20; u8 first_search = 0x00, touch_neg_max = 0x00; - char idx0 = 0, idx1 = 0; + s8 idx0 = 0, idx1 = 0; u8 gop; u8 bbp = 0; - char bidx; + s8 bidx; for (bidx = 5; bidx >= 1; bidx--) { for (gop = 0; gop < 2; gop++) { @@ -10043,11 +10043,11 @@ static int rt2800_rf_lp_config(struct rt2x00_dev *rt2x00dev, bool btxcal) return 0; } -static char rt2800_lp_tx_filter_bw_cal(struct rt2x00_dev *rt2x00dev) +static s8 rt2800_lp_tx_filter_bw_cal(struct rt2x00_dev *rt2x00dev) { unsigned int cnt; u8 bbp_val; - char cal_val; + s8 cal_val; rt2800_bbp_dcoc_write(rt2x00dev, 0, 0x82); @@ -10079,7 +10079,7 @@ static void rt2800_bw_filter_calibration(struct rt2x00_dev *rt2x00dev, u8 rx_filter_target_20m = 0x27, rx_filter_target_40m = 0x31; int loop = 0, is_ht40, cnt; u8 bbp_val, rf_val; - char cal_r32_init, cal_r32_val, cal_diff; + s8 cal_r32_init, cal_r32_val, cal_diff; u8 saverfb5r00, saverfb5r01, saverfb5r03, saverfb5r04, saverfb5r05; u8 saverfb5r06, saverfb5r07; u8 saverfb5r08, saverfb5r17, saverfb5r18, saverfb5r19, saverfb5r20; @@ -11550,9 +11550,9 @@ static int rt2800_probe_hw_mode(struct rt2x00_dev *rt2x00dev) { struct hw_mode_spec *spec = &rt2x00dev->spec; struct channel_info *info; - char *default_power1; - char *default_power2; - char *default_power3; + s8 *default_power1; + s8 *default_power2; + s8 *default_power3; unsigned int i, tx_chains, rx_chains; u32 reg; diff --git a/drivers/net/wireless/ralink/rt2x00/rt2800lib.h b/drivers/net/wireless/ralink/rt2x00/rt2800lib.h index 3cbef77b4bd3..194de676df8f 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2800lib.h +++ b/drivers/net/wireless/ralink/rt2x00/rt2800lib.h @@ -32,10 +32,10 @@ struct rf_reg_pair { struct rt2800_drv_data { u8 calibration_bw20; u8 calibration_bw40; - char rx_calibration_bw20; - char rx_calibration_bw40; - char tx_calibration_bw20; - char tx_calibration_bw40; + s8 rx_calibration_bw20; + s8 rx_calibration_bw40; + s8 tx_calibration_bw20; + s8 tx_calibration_bw40; u8 bbp25; u8 bbp26; u8 txmixer_gain_24g; diff --git a/drivers/net/wireless/ralink/rt2x00/rt2x00usb.c b/drivers/net/wireless/ralink/rt2x00/rt2x00usb.c index 0827bc860bf8..8fd22c69855f 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2x00usb.c +++ b/drivers/net/wireless/ralink/rt2x00/rt2x00usb.c @@ -117,12 +117,12 @@ int rt2x00usb_vendor_request_buff(struct rt2x00_dev *rt2x00dev, const u16 buffer_length) { int status = 0; - unsigned char *tb; + u8 *tb; u16 off, len, bsize; mutex_lock(&rt2x00dev->csr_mutex); - tb = (char *)buffer; + tb = (u8 *)buffer; off = offset; len = buffer_length; while (len && !status) { @@ -215,7 +215,7 @@ void rt2x00usb_register_read_async(struct rt2x00_dev *rt2x00dev, rd->cr.wLength = cpu_to_le16(sizeof(u32)); usb_fill_control_urb(urb, usb_dev, usb_rcvctrlpipe(usb_dev, 0), - (unsigned char *)(&rd->cr), &rd->reg, sizeof(rd->reg), + (u8 *)(&rd->cr), &rd->reg, sizeof(rd->reg), rt2x00usb_register_read_async_cb, rd); usb_anchor_urb(urb, rt2x00dev->anchor); if (usb_submit_urb(urb, GFP_ATOMIC) < 0) { diff --git a/drivers/net/wireless/ralink/rt2x00/rt61pci.c b/drivers/net/wireless/ralink/rt2x00/rt61pci.c index d92f9eb07dc9..81db7f57c7e4 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt61pci.c +++ b/drivers/net/wireless/ralink/rt2x00/rt61pci.c @@ -1709,7 +1709,7 @@ static int rt61pci_set_state(struct rt2x00_dev *rt2x00dev, enum dev_state state) { u32 reg, reg2; unsigned int i; - char put_to_sleep; + bool put_to_sleep; put_to_sleep = (state != STATE_AWAKE); @@ -2656,7 +2656,7 @@ static int rt61pci_probe_hw_mode(struct rt2x00_dev *rt2x00dev) { struct hw_mode_spec *spec = &rt2x00dev->spec; struct channel_info *info; - char *tx_power; + u8 *tx_power; unsigned int i; /* diff --git a/drivers/net/wireless/ralink/rt2x00/rt61pci.h b/drivers/net/wireless/ralink/rt2x00/rt61pci.h index 5f208ad509bd..d72d0ffd1127 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt61pci.h +++ b/drivers/net/wireless/ralink/rt2x00/rt61pci.h @@ -1484,6 +1484,6 @@ struct hw_pairwise_ta_entry { (((u8)(__txpower)) > MAX_TXPOWER) ? DEFAULT_TXPOWER : (__txpower) #define TXPOWER_TO_DEV(__txpower) \ - clamp_t(char, __txpower, MIN_TXPOWER, MAX_TXPOWER) + clamp_t(u8, __txpower, MIN_TXPOWER, MAX_TXPOWER) #endif /* RT61PCI_H */ diff --git a/drivers/net/wireless/ralink/rt2x00/rt73usb.c b/drivers/net/wireless/ralink/rt2x00/rt73usb.c index e3269fd7c59e..861035444374 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt73usb.c +++ b/drivers/net/wireless/ralink/rt2x00/rt73usb.c @@ -1378,7 +1378,7 @@ static int rt73usb_set_state(struct rt2x00_dev *rt2x00dev, enum dev_state state) { u32 reg, reg2; unsigned int i; - char put_to_sleep; + bool put_to_sleep; put_to_sleep = (state != STATE_AWAKE); @@ -2090,7 +2090,7 @@ static int rt73usb_probe_hw_mode(struct rt2x00_dev *rt2x00dev) { struct hw_mode_spec *spec = &rt2x00dev->spec; struct channel_info *info; - char *tx_power; + u8 *tx_power; unsigned int i; /* diff --git a/drivers/net/wireless/ralink/rt2x00/rt73usb.h b/drivers/net/wireless/ralink/rt2x00/rt73usb.h index 1b56d285c34b..bb0a68516c08 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt73usb.h +++ b/drivers/net/wireless/ralink/rt2x00/rt73usb.h @@ -1063,6 +1063,6 @@ struct hw_pairwise_ta_entry { (((u8)(__txpower)) > MAX_TXPOWER) ? DEFAULT_TXPOWER : (__txpower) #define TXPOWER_TO_DEV(__txpower) \ - clamp_t(char, __txpower, MIN_TXPOWER, MAX_TXPOWER) + clamp_t(u8, __txpower, MIN_TXPOWER, MAX_TXPOWER) #endif /* RT73USB_H */ From 03c0ad4b06c3566de624b4f4b78ac1a5d1e4c8e7 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 13 Oct 2022 19:41:51 +0200 Subject: [PATCH 010/138] wifi: cfg80211: silence a sparse RCU warning All we're going to do with this pointer is assign it to another __rcu pointer, but sparse can't see that, so use rcu_access_pointer() to silence the warning here. Fixes: c90b93b5b782 ("wifi: cfg80211: update hidden BSSes to avoid WARN_ON") Signed-off-by: Johannes Berg --- net/wireless/scan.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/wireless/scan.c b/net/wireless/scan.c index 806a5f1330ff..da752b0cc752 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -1674,7 +1674,9 @@ cfg80211_update_known_bss(struct cfg80211_registered_device *rdev, if (old == rcu_access_pointer(known->pub.ies)) rcu_assign_pointer(known->pub.ies, new->pub.beacon_ies); - cfg80211_update_hidden_bsses(known, new->pub.beacon_ies, old); + cfg80211_update_hidden_bsses(known, + rcu_access_pointer(new->pub.beacon_ies), + old); if (old) kfree_rcu((struct cfg80211_bss_ies *)old, rcu_head); From 50b2e8711462409cd368c41067405aa446dfa2af Mon Sep 17 00:00:00 2001 From: taozhang Date: Sat, 15 Oct 2022 17:38:31 +0800 Subject: [PATCH 011/138] wifi: mac80211: fix memory free error when registering wiphy fail ieee80211_register_hw free the allocated cipher suites when registering wiphy fail, and ieee80211_free_hw will re-free it. set wiphy_ciphers_allocated to false after freeing allocated cipher suites. Signed-off-by: taozhang Signed-off-by: Johannes Berg --- net/mac80211/main.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/net/mac80211/main.c b/net/mac80211/main.c index 46f3eddc2388..02b5abc7326b 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -1439,8 +1439,10 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) ieee80211_led_exit(local); destroy_workqueue(local->workqueue); fail_workqueue: - if (local->wiphy_ciphers_allocated) + if (local->wiphy_ciphers_allocated) { kfree(local->hw.wiphy->cipher_suites); + local->wiphy_ciphers_allocated = false; + } kfree(local->int_scan_req); return result; } @@ -1508,8 +1510,10 @@ void ieee80211_free_hw(struct ieee80211_hw *hw) mutex_destroy(&local->iflist_mtx); mutex_destroy(&local->mtx); - if (local->wiphy_ciphers_allocated) + if (local->wiphy_ciphers_allocated) { kfree(local->hw.wiphy->cipher_suites); + local->wiphy_ciphers_allocated = false; + } idr_for_each(&local->ack_status_frames, ieee80211_free_ack_frame, NULL); From 57b962e627ec0ae53d4d16d7bd1033e27e67677a Mon Sep 17 00:00:00 2001 From: Arend van Spriel Date: Thu, 20 Oct 2022 13:40:40 +0200 Subject: [PATCH 012/138] wifi: cfg80211: fix memory leak in query_regdb_file() In the function query_regdb_file() the alpha2 parameter is duplicated using kmemdup() and subsequently freed in regdb_fw_cb(). However, request_firmware_nowait() can fail without calling regdb_fw_cb() and thus leak memory. Fixes: 007f6c5e6eb4 ("cfg80211: support loading regulatory database as firmware file") Signed-off-by: Arend van Spriel Signed-off-by: Johannes Berg --- net/wireless/reg.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/net/wireless/reg.c b/net/wireless/reg.c index d5c7a5aa6853..c3d950d29432 100644 --- a/net/wireless/reg.c +++ b/net/wireless/reg.c @@ -1084,6 +1084,8 @@ MODULE_FIRMWARE("regulatory.db"); static int query_regdb_file(const char *alpha2) { + int err; + ASSERT_RTNL(); if (regdb) @@ -1093,9 +1095,13 @@ static int query_regdb_file(const char *alpha2) if (!alpha2) return -ENOMEM; - return request_firmware_nowait(THIS_MODULE, true, "regulatory.db", - ®_pdev->dev, GFP_KERNEL, - (void *)alpha2, regdb_fw_cb); + err = request_firmware_nowait(THIS_MODULE, true, "regulatory.db", + ®_pdev->dev, GFP_KERNEL, + (void *)alpha2, regdb_fw_cb); + if (err) + kfree(alpha2); + + return err; } int reg_reload_regdb(void) From 18429c51c7ff6e6bfd627316c54670230967a7e5 Mon Sep 17 00:00:00 2001 From: Paul Zhang Date: Tue, 11 Oct 2022 21:04:28 +0800 Subject: [PATCH 013/138] wifi: cfg80211: Fix bitrates overflow issue MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When invoking function cfg80211_calculate_bitrate_eht about (320 MHz, EHT-MCS 13, EHT-NSS 2, EHT-GI 0), which means the parameters as flags: 0x80, bw: 7, mcs: 13, eht_gi: 0, nss: 2, this formula (result * rate->nss) will overflow and causes the returned bitrate to be 3959 when it should be 57646. Here is the explanation: u64 tmp; u32 result; … /* tmp = result = 4 * rates_996[0] * = 4 * 480388888 = 0x72889c60 */ tmp = result; /* tmp = 0x72889c60 * 6144 = 0xabccea90000 */ tmp *= SCALE; /* tmp = 0xabccea90000 / mcs_divisors[13] * = 0xabccea90000 / 5120 = 0x8970bba6 */ do_div(tmp, mcs_divisors[rate->mcs]); /* result = 0x8970bba6 */ result = tmp; /* normally (result * rate->nss) = 0x8970bba6 * 2 = 0x112e1774c, * but since result is u32, (result * rate->nss) = 0x12e1774c, * overflow happens and it loses the highest bit. * Then result = 0x12e1774c / 8 = 39595753, */ result = (result * rate->nss) / 8; Signed-off-by: Paul Zhang Signed-off-by: Johannes Berg --- net/wireless/util.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/wireless/util.c b/net/wireless/util.c index 1f285b515028..39680e7bad45 100644 --- a/net/wireless/util.c +++ b/net/wireless/util.c @@ -1557,10 +1557,12 @@ static u32 cfg80211_calculate_bitrate_eht(struct rate_info *rate) tmp = result; tmp *= SCALE; do_div(tmp, mcs_divisors[rate->mcs]); - result = tmp; /* and take NSS */ - result = (result * rate->nss) / 8; + tmp *= rate->nss; + do_div(tmp, 8); + + result = tmp; return result / 10000; } From 69188df5f6e4cecc6b76b958979ba363cd5240e8 Mon Sep 17 00:00:00 2001 From: Jonas Jelonek Date: Fri, 14 Oct 2022 16:54:39 +0200 Subject: [PATCH 014/138] wifi: mac80211_hwsim: fix debugfs attribute ps with rc table support Fixes a warning that occurs when rc table support is enabled (IEEE80211_HW_SUPPORTS_RC_TABLE) in mac80211_hwsim and the PS mode is changed via the exported debugfs attribute. When the PS mode is changed, a packet is broadcasted via hwsim_send_nullfunc by creating and transmitting a plain skb with only header initialized. The ieee80211 rate array in the control buffer is zero-initialized. When ratetbl support is enabled, ieee80211_get_tx_rates is called for the skb with sta parameter set to NULL and thus no ratetbl can be used. The final rate array then looks like [-1,0; 0,0; 0,0; 0,0] which causes the warning in ieee80211_get_tx_rate. The issue is fixed by setting the count of the first rate with idx '0' to 1 and hence ieee80211_get_tx_rates won't overwrite it with idx '-1'. Signed-off-by: Jonas Jelonek Signed-off-by: Johannes Berg --- drivers/net/wireless/mac80211_hwsim.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/wireless/mac80211_hwsim.c b/drivers/net/wireless/mac80211_hwsim.c index a40636c90ec3..0d81098c7b45 100644 --- a/drivers/net/wireless/mac80211_hwsim.c +++ b/drivers/net/wireless/mac80211_hwsim.c @@ -910,6 +910,7 @@ static void hwsim_send_nullfunc(struct mac80211_hwsim_data *data, u8 *mac, struct hwsim_vif_priv *vp = (void *)vif->drv_priv; struct sk_buff *skb; struct ieee80211_hdr *hdr; + struct ieee80211_tx_info *cb; if (!vp->assoc) return; @@ -931,6 +932,10 @@ static void hwsim_send_nullfunc(struct mac80211_hwsim_data *data, u8 *mac, memcpy(hdr->addr2, mac, ETH_ALEN); memcpy(hdr->addr3, vp->bssid, ETH_ALEN); + cb = IEEE80211_SKB_CB(skb); + cb->control.rates[0].count = 1; + cb->control.rates[1].idx = -1; + rcu_read_lock(); mac80211_hwsim_tx_frame(data->hw, skb, rcu_dereference(vif->bss_conf.chanctx_conf)->def.chan); From 13c30a755847c7e804e1bf755e66e3ff7b7f9367 Mon Sep 17 00:00:00 2001 From: Sjoerd Simons Date: Thu, 20 Oct 2022 09:56:24 +0800 Subject: [PATCH 015/138] soundwire: intel: Initialize clock stop timeout The bus->clk_stop_timeout member is only initialized to a non-zero value during the codec driver probe. This can lead to corner cases where this value remains pegged at zero when the bus suspends, which results in an endless loop in sdw_bus_wait_for_clk_prep_deprep(). Corner cases include configurations with no codecs described in the firmware, or delays in probing codec drivers. Initializing the default timeout to the smallest non-zero value avoid this problem and allows for the existing logic to be preserved: the bus->clk_stop_timeout is set as the maximum required by all codecs connected on the bus. Fixes: 1f2dcf3a154ac ("soundwire: intel: set dev_num_ida_min") Signed-off-by: Sjoerd Simons Reviewed-by: Pierre-Louis Bossart Reviewed-by: Chao Song Signed-off-by: Bard Liao Link: https://lore.kernel.org/r/20221020015624.1703950-1-yung-chuan.liao@linux.intel.com Signed-off-by: Vinod Koul --- drivers/soundwire/intel.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/soundwire/intel.c b/drivers/soundwire/intel.c index 244209358784..8c76541d553f 100644 --- a/drivers/soundwire/intel.c +++ b/drivers/soundwire/intel.c @@ -1513,6 +1513,7 @@ static int intel_link_probe(struct auxiliary_device *auxdev, bus->link_id = auxdev->id; bus->dev_num_ida_min = INTEL_DEV_NUM_IDA_MIN; + bus->clk_stop_timeout = 1; sdw_cdns_probe(cdns); From f936fa7a954b262cb3908bbc8f01ba19dfaf9fbf Mon Sep 17 00:00:00 2001 From: Srinivas Kandagatla Date: Wed, 26 Oct 2022 12:02:05 +0100 Subject: [PATCH 016/138] soundwire: qcom: reinit broadcast completion For some reason we never reinit the broadcast completion, there is a danger that broadcast commands could be treated as completed by driver from previous complete status. Fix this by reinitializing the completion before sending a broadcast command. Fixes: ddea6cf7b619 ("soundwire: qcom: update register read/write routine") Signed-off-by: Srinivas Kandagatla Link: https://lore.kernel.org/r/20221026110210.6575-2-srinivas.kandagatla@linaro.org Signed-off-by: Vinod Koul --- drivers/soundwire/qcom.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/soundwire/qcom.c b/drivers/soundwire/qcom.c index b33d5db494a5..8f1a1eb017a2 100644 --- a/drivers/soundwire/qcom.c +++ b/drivers/soundwire/qcom.c @@ -344,6 +344,9 @@ static int qcom_swrm_cmd_fifo_wr_cmd(struct qcom_swrm_ctrl *swrm, u8 cmd_data, if (swrm_wait_for_wr_fifo_avail(swrm)) return SDW_CMD_FAIL_OTHER; + if (cmd_id == SWR_BROADCAST_CMD_ID) + reinit_completion(&swrm->broadcast); + /* Its assumed that write is okay as we do not get any status back */ swrm->reg_write(swrm, SWRM_CMD_FIFO_WR_CMD, val); From 49a467310dc4fae591a3547860ee04d8730780f4 Mon Sep 17 00:00:00 2001 From: Srinivas Kandagatla Date: Wed, 26 Oct 2022 12:02:06 +0100 Subject: [PATCH 017/138] soundwire: qcom: check for outanding writes before doing a read Reading will increase the fifo count, so check for outstanding cmd wrt. write fifo depth to avoid overflow as read will also increase write fifo cnt. Fixes: a661308c34de ("soundwire: qcom: wait for fifo space to be available before read/write") Signed-off-by: Srinivas Kandagatla Link: https://lore.kernel.org/r/20221026110210.6575-3-srinivas.kandagatla@linaro.org Signed-off-by: Vinod Koul --- drivers/soundwire/qcom.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/soundwire/qcom.c b/drivers/soundwire/qcom.c index 8f1a1eb017a2..cee2b2223141 100644 --- a/drivers/soundwire/qcom.c +++ b/drivers/soundwire/qcom.c @@ -380,6 +380,12 @@ static int qcom_swrm_cmd_fifo_rd_cmd(struct qcom_swrm_ctrl *swrm, val = swrm_get_packed_reg_val(&swrm->rcmd_id, len, dev_addr, reg_addr); + /* + * Check for outstanding cmd wrt. write fifo depth to avoid + * overflow as read will also increase write fifo cnt. + */ + swrm_wait_for_wr_fifo_avail(swrm); + /* wait for FIFO RD to complete to avoid overflow */ usleep_range(100, 105); swrm->reg_write(swrm, SWRM_CMD_FIFO_RD_CMD, val); From 04948e757148f870a31f4887ea2239403f516c3c Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Wed, 26 Oct 2022 18:21:16 +0200 Subject: [PATCH 018/138] phy: qcom-qmp-combo: fix NULL-deref on runtime resume Commit fc64623637da ("phy: qcom-qmp-combo,usb: add support for separate PCS_USB region") started treating the PCS_USB registers as potentially separate from the PCS registers but used the wrong base when no PCS_USB offset has been provided. Fix the PCS_USB base used at runtime resume to prevent dereferencing a NULL pointer on platforms that do not provide a PCS_USB offset (e.g. SC7180). Fixes: fc64623637da ("phy: qcom-qmp-combo,usb: add support for separate PCS_USB region") Cc: stable@vger.kernel.org # 5.20 Signed-off-by: Johan Hovold Reviewed-by: Dmitry Baryshkov Reviewed-by: Andrew Halaney Link: https://lore.kernel.org/r/20221026162116.26462-1-johan+linaro@kernel.org Signed-off-by: Vinod Koul --- drivers/phy/qualcomm/phy-qcom-qmp-combo.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/phy/qualcomm/phy-qcom-qmp-combo.c b/drivers/phy/qualcomm/phy-qcom-qmp-combo.c index 9807c4d935cd..ba9d761ec49a 100644 --- a/drivers/phy/qualcomm/phy-qcom-qmp-combo.c +++ b/drivers/phy/qualcomm/phy-qcom-qmp-combo.c @@ -2240,7 +2240,7 @@ static void qmp_combo_enable_autonomous_mode(struct qmp_phy *qphy) static void qmp_combo_disable_autonomous_mode(struct qmp_phy *qphy) { const struct qmp_phy_cfg *cfg = qphy->cfg; - void __iomem *pcs_usb = qphy->pcs_usb ?: qphy->pcs_usb; + void __iomem *pcs_usb = qphy->pcs_usb ?: qphy->pcs; void __iomem *pcs_misc = qphy->pcs_misc; /* Disable i/o clamp_n on resume for normal mode */ From 986d93f55bdeab1cac858d1e47b41fac10b2d7f6 Mon Sep 17 00:00:00 2001 From: Gaosheng Cui Date: Mon, 31 Oct 2022 10:10:21 +0800 Subject: [PATCH 019/138] audit: fix undefined behavior in bit shift for AUDIT_BIT Shifting signed 32-bit value by 31 bits is undefined, so changing significant bit to unsigned. The UBSAN warning calltrace like below: UBSAN: shift-out-of-bounds in kernel/auditfilter.c:179:23 left shift of 1 by 31 places cannot be represented in type 'int' Call Trace: dump_stack_lvl+0x7d/0xa5 dump_stack+0x15/0x1b ubsan_epilogue+0xe/0x4e __ubsan_handle_shift_out_of_bounds+0x1e7/0x20c audit_register_class+0x9d/0x137 audit_classes_init+0x4d/0xb8 do_one_initcall+0x76/0x430 kernel_init_freeable+0x3b3/0x422 kernel_init+0x24/0x1e0 ret_from_fork+0x1f/0x30 Signed-off-by: Gaosheng Cui [PM: remove bad 'Fixes' tag as issue predates git, added in v2.6.6-rc1] Signed-off-by: Paul Moore --- include/uapi/linux/audit.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/uapi/linux/audit.h b/include/uapi/linux/audit.h index 7c1dc818b1d5..d676ed2b246e 100644 --- a/include/uapi/linux/audit.h +++ b/include/uapi/linux/audit.h @@ -187,7 +187,7 @@ #define AUDIT_MAX_KEY_LEN 256 #define AUDIT_BITMASK_SIZE 64 #define AUDIT_WORD(nr) ((__u32)((nr)/32)) -#define AUDIT_BIT(nr) (1 << ((nr) - AUDIT_WORD(nr)*32)) +#define AUDIT_BIT(nr) (1U << ((nr) - AUDIT_WORD(nr)*32)) #define AUDIT_SYSCALL_CLASSES 16 #define AUDIT_CLASS_DIR_WRITE 0 From e6cb8769452e8236b52134e5cb4a18b8f5986932 Mon Sep 17 00:00:00 2001 From: "Jason A. Donenfeld" Date: Mon, 24 Oct 2022 18:28:43 +0200 Subject: [PATCH 020/138] wifi: airo: do not assign -1 to unsigned char With char becoming unsigned by default, and with `char` alone being ambiguous and based on architecture, we get a warning when assigning the unchecked output of hex_to_bin() to that unsigned char. Mark `key` as a `u8`, which matches the struct's type, and then check each call to hex_to_bin() before casting. Cc: Kalle Valo Cc: linux-wireless@vger.kernel.org Signed-off-by: Jason A. Donenfeld Signed-off-by: Kalle Valo Link: https://lore.kernel.org/r/20221024162843.535921-1-Jason@zx2c4.com --- drivers/net/wireless/cisco/airo.c | 18 ++++++++++++++---- 1 file changed, 14 insertions(+), 4 deletions(-) diff --git a/drivers/net/wireless/cisco/airo.c b/drivers/net/wireless/cisco/airo.c index 10daef81c355..fb2c35bd73bb 100644 --- a/drivers/net/wireless/cisco/airo.c +++ b/drivers/net/wireless/cisco/airo.c @@ -5232,7 +5232,7 @@ static int get_wep_tx_idx(struct airo_info *ai) return -1; } -static int set_wep_key(struct airo_info *ai, u16 index, const char *key, +static int set_wep_key(struct airo_info *ai, u16 index, const u8 *key, u16 keylen, int perm, int lock) { static const unsigned char macaddr[ETH_ALEN] = { 0x01, 0, 0, 0, 0, 0 }; @@ -5283,7 +5283,7 @@ static void proc_wepkey_on_close(struct inode *inode, struct file *file) struct net_device *dev = pde_data(inode); struct airo_info *ai = dev->ml_priv; int i, rc; - char key[16]; + u8 key[16]; u16 index = 0; int j = 0; @@ -5311,12 +5311,22 @@ static void proc_wepkey_on_close(struct inode *inode, struct file *file) } for (i = 0; i < 16*3 && data->wbuffer[i+j]; i++) { + int val; + + if (i % 3 == 2) + continue; + + val = hex_to_bin(data->wbuffer[i+j]); + if (val < 0) { + airo_print_err(ai->dev->name, "WebKey passed invalid key hex"); + return; + } switch(i%3) { case 0: - key[i/3] = hex_to_bin(data->wbuffer[i+j])<<4; + key[i/3] = (u8)val << 4; break; case 1: - key[i/3] |= hex_to_bin(data->wbuffer[i+j]); + key[i/3] |= (u8)val; break; } } From 6788ba8aed4e28e90f72d68a9d794e34eac17295 Mon Sep 17 00:00:00 2001 From: Dokyung Song Date: Fri, 21 Oct 2022 15:13:59 +0900 Subject: [PATCH 021/138] wifi: brcmfmac: Fix potential buffer overflow in brcmf_fweh_event_worker() This patch fixes an intra-object buffer overflow in brcmfmac that occurs when the device provides a 'bsscfgidx' equal to or greater than the buffer size. The patch adds a check that leads to a safe failure if that is the case. This fixes CVE-2022-3628. UBSAN: array-index-out-of-bounds in drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c index 52 is out of range for type 'brcmf_if *[16]' CPU: 0 PID: 1898 Comm: kworker/0:2 Tainted: G O 5.14.0+ #132 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014 Workqueue: events brcmf_fweh_event_worker Call Trace: dump_stack_lvl+0x57/0x7d ubsan_epilogue+0x5/0x40 __ubsan_handle_out_of_bounds+0x69/0x80 ? memcpy+0x39/0x60 brcmf_fweh_event_worker+0xae1/0xc00 ? brcmf_fweh_call_event_handler.isra.0+0x100/0x100 ? rcu_read_lock_sched_held+0xa1/0xd0 ? rcu_read_lock_bh_held+0xb0/0xb0 ? lockdep_hardirqs_on_prepare+0x273/0x3e0 process_one_work+0x873/0x13e0 ? lock_release+0x640/0x640 ? pwq_dec_nr_in_flight+0x320/0x320 ? rwlock_bug.part.0+0x90/0x90 worker_thread+0x8b/0xd10 ? __kthread_parkme+0xd9/0x1d0 ? process_one_work+0x13e0/0x13e0 kthread+0x379/0x450 ? _raw_spin_unlock_irq+0x24/0x30 ? set_kthread_struct+0x100/0x100 ret_from_fork+0x1f/0x30 ================================================================================ general protection fault, probably for non-canonical address 0xe5601c0020023fff: 0000 [#1] SMP KASAN KASAN: maybe wild-memory-access in range [0x2b0100010011fff8-0x2b0100010011ffff] CPU: 0 PID: 1898 Comm: kworker/0:2 Tainted: G O 5.14.0+ #132 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014 Workqueue: events brcmf_fweh_event_worker RIP: 0010:brcmf_fweh_call_event_handler.isra.0+0x42/0x100 Code: 89 f5 53 48 89 fb 48 83 ec 08 e8 79 0b 38 fe 48 85 ed 74 7e e8 6f 0b 38 fe 48 89 ea 48 b8 00 00 00 00 00 fc ff df 48 c1 ea 03 <80> 3c 02 00 0f 85 8b 00 00 00 4c 8b 7d 00 44 89 e0 48 ba 00 00 00 RSP: 0018:ffffc9000259fbd8 EFLAGS: 00010207 RAX: dffffc0000000000 RBX: ffff888115d8cd50 RCX: 0000000000000000 RDX: 0560200020023fff RSI: ffffffff8304bc91 RDI: ffff888115d8cd50 RBP: 2b0100010011ffff R08: ffff888112340050 R09: ffffed1023549809 R10: ffff88811aa4c047 R11: ffffed1023549808 R12: 0000000000000045 R13: ffffc9000259fca0 R14: ffff888112340050 R15: ffff888112340000 FS: 0000000000000000(0000) GS:ffff88811aa00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 000000004053ccc0 CR3: 0000000112740000 CR4: 0000000000750ef0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: brcmf_fweh_event_worker+0x117/0xc00 ? brcmf_fweh_call_event_handler.isra.0+0x100/0x100 ? rcu_read_lock_sched_held+0xa1/0xd0 ? rcu_read_lock_bh_held+0xb0/0xb0 ? lockdep_hardirqs_on_prepare+0x273/0x3e0 process_one_work+0x873/0x13e0 ? lock_release+0x640/0x640 ? pwq_dec_nr_in_flight+0x320/0x320 ? rwlock_bug.part.0+0x90/0x90 worker_thread+0x8b/0xd10 ? __kthread_parkme+0xd9/0x1d0 ? process_one_work+0x13e0/0x13e0 kthread+0x379/0x450 ? _raw_spin_unlock_irq+0x24/0x30 ? set_kthread_struct+0x100/0x100 ret_from_fork+0x1f/0x30 Modules linked in: 88XXau(O) 88x2bu(O) ---[ end trace 41d302138f3ff55a ]--- RIP: 0010:brcmf_fweh_call_event_handler.isra.0+0x42/0x100 Code: 89 f5 53 48 89 fb 48 83 ec 08 e8 79 0b 38 fe 48 85 ed 74 7e e8 6f 0b 38 fe 48 89 ea 48 b8 00 00 00 00 00 fc ff df 48 c1 ea 03 <80> 3c 02 00 0f 85 8b 00 00 00 4c 8b 7d 00 44 89 e0 48 ba 00 00 00 RSP: 0018:ffffc9000259fbd8 EFLAGS: 00010207 RAX: dffffc0000000000 RBX: ffff888115d8cd50 RCX: 0000000000000000 RDX: 0560200020023fff RSI: ffffffff8304bc91 RDI: ffff888115d8cd50 RBP: 2b0100010011ffff R08: ffff888112340050 R09: ffffed1023549809 R10: ffff88811aa4c047 R11: ffffed1023549808 R12: 0000000000000045 R13: ffffc9000259fca0 R14: ffff888112340050 R15: ffff888112340000 FS: 0000000000000000(0000) GS:ffff88811aa00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 000000004053ccc0 CR3: 0000000112740000 CR4: 0000000000750ef0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Kernel panic - not syncing: Fatal exception Reported-by: Dokyung Song Reported-by: Jisoo Jang Reported-by: Minsuk Kang Reviewed-by: Arend van Spriel Cc: Signed-off-by: Dokyung Song Signed-off-by: Kalle Valo Link: https://lore.kernel.org/r/20221021061359.GA550858@laguna --- drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c index bc3f4e4edcdf..dac7eb77799b 100644 --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c @@ -228,6 +228,10 @@ static void brcmf_fweh_event_worker(struct work_struct *work) brcmf_fweh_event_name(event->code), event->code, event->emsg.ifidx, event->emsg.bsscfgidx, event->emsg.addr); + if (event->emsg.bsscfgidx >= BRCMF_MAX_IFS) { + bphy_err(drvr, "invalid bsscfg index: %u\n", event->emsg.bsscfgidx); + goto event_free; + } /* convert event message */ emsg_be = &event->emsg; From 42378a9ca55347102bbf86708776061d8fe3ece2 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Fri, 28 Oct 2022 19:54:30 -0700 Subject: [PATCH 022/138] bpf, verifier: Fix memory leak in array reallocation for stack state If an error (NULL) is returned by krealloc(), callers of realloc_array() were setting their allocation pointers to NULL, but on error krealloc() does not touch the original allocation. This would result in a memory resource leak. Instead, free the old allocation on the error handling path. The memory leak information is as follows as also reported by Zhengchao: unreferenced object 0xffff888019801800 (size 256): comm "bpf_repo", pid 6490, jiffies 4294959200 (age 17.170s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [<00000000b211474b>] __kmalloc_node_track_caller+0x45/0xc0 [<0000000086712a0b>] krealloc+0x83/0xd0 [<00000000139aab02>] realloc_array+0x82/0xe2 [<00000000b1ca41d1>] grow_stack_state+0xfb/0x186 [<00000000cd6f36d2>] check_mem_access.cold+0x141/0x1341 [<0000000081780455>] do_check_common+0x5358/0xb350 [<0000000015f6b091>] bpf_check.cold+0xc3/0x29d [<000000002973c690>] bpf_prog_load+0x13db/0x2240 [<00000000028d1644>] __sys_bpf+0x1605/0x4ce0 [<00000000053f29bd>] __x64_sys_bpf+0x75/0xb0 [<0000000056fedaf5>] do_syscall_64+0x35/0x80 [<000000002bd58261>] entry_SYSCALL_64_after_hwframe+0x63/0xcd Fixes: c69431aab67a ("bpf: verifier: Improve function state reallocation") Reported-by: Zhengchao Shao Reported-by: Kees Cook Signed-off-by: Kees Cook Signed-off-by: Daniel Borkmann Reviewed-by: Bill Wendling Cc: Lorenz Bauer Link: https://lore.kernel.org/bpf/20221029025433.2533810-1-keescook@chromium.org --- kernel/bpf/verifier.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 7f0a9f6cb889..dd9019c8b0db 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -1027,12 +1027,17 @@ static void *copy_array(void *dst, const void *src, size_t n, size_t size, gfp_t */ static void *realloc_array(void *arr, size_t old_n, size_t new_n, size_t size) { + void *new_arr; + if (!new_n || old_n == new_n) goto out; - arr = krealloc_array(arr, new_n, size, GFP_KERNEL); - if (!arr) + new_arr = krealloc_array(arr, new_n, size, GFP_KERNEL); + if (!new_arr) { + kfree(arr); return NULL; + } + arr = new_arr; if (new_n > old_n) memset(arr + old_n * size, 0, (new_n - old_n) * size); From 8ec95b94716a1e4d126edc3fb2bc426a717e2dba Mon Sep 17 00:00:00 2001 From: Wang Yufen Date: Tue, 1 Nov 2022 09:31:36 +0800 Subject: [PATCH 023/138] bpf, sockmap: Fix the sk->sk_forward_alloc warning of sk_stream_kill_queues When running `test_sockmap` selftests, the following warning appears: WARNING: CPU: 2 PID: 197 at net/core/stream.c:205 sk_stream_kill_queues+0xd3/0xf0 Call Trace: inet_csk_destroy_sock+0x55/0x110 tcp_rcv_state_process+0xd28/0x1380 ? tcp_v4_do_rcv+0x77/0x2c0 tcp_v4_do_rcv+0x77/0x2c0 __release_sock+0x106/0x130 __tcp_close+0x1a7/0x4e0 tcp_close+0x20/0x70 inet_release+0x3c/0x80 __sock_release+0x3a/0xb0 sock_close+0x14/0x20 __fput+0xa3/0x260 task_work_run+0x59/0xb0 exit_to_user_mode_prepare+0x1b3/0x1c0 syscall_exit_to_user_mode+0x19/0x50 do_syscall_64+0x48/0x90 entry_SYSCALL_64_after_hwframe+0x44/0xae The root case is in commit 84472b436e76 ("bpf, sockmap: Fix more uncharged while msg has more_data"), where I used msg->sg.size to replace the tosend, causing breakage: if (msg->apply_bytes && msg->apply_bytes < tosend) tosend = psock->apply_bytes; Fixes: 84472b436e76 ("bpf, sockmap: Fix more uncharged while msg has more_data") Reported-by: Jakub Sitnicki Signed-off-by: Wang Yufen Signed-off-by: Daniel Borkmann Acked-by: John Fastabend Acked-by: Jakub Sitnicki Link: https://lore.kernel.org/bpf/1667266296-8794-1-git-send-email-wangyufen@huawei.com --- net/ipv4/tcp_bpf.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/net/ipv4/tcp_bpf.c b/net/ipv4/tcp_bpf.c index c501c329b1db..cf9c3e8f7ccb 100644 --- a/net/ipv4/tcp_bpf.c +++ b/net/ipv4/tcp_bpf.c @@ -278,7 +278,7 @@ static int tcp_bpf_send_verdict(struct sock *sk, struct sk_psock *psock, { bool cork = false, enospc = sk_msg_full(msg); struct sock *sk_redir; - u32 tosend, delta = 0; + u32 tosend, origsize, sent, delta = 0; u32 eval = __SK_NONE; int ret; @@ -333,10 +333,12 @@ static int tcp_bpf_send_verdict(struct sock *sk, struct sk_psock *psock, cork = true; psock->cork = NULL; } - sk_msg_return(sk, msg, msg->sg.size); + sk_msg_return(sk, msg, tosend); release_sock(sk); + origsize = msg->sg.size; ret = tcp_bpf_sendmsg_redir(sk_redir, msg, tosend, flags); + sent = origsize - msg->sg.size; if (eval == __SK_REDIRECT) sock_put(sk_redir); @@ -375,7 +377,7 @@ static int tcp_bpf_send_verdict(struct sock *sk, struct sk_psock *psock, msg->sg.data[msg->sg.start].page_link && msg->sg.data[msg->sg.start].length) { if (eval == __SK_REDIRECT) - sk_mem_charge(sk, msg->sg.size); + sk_mem_charge(sk, tosend - sent); goto more_data; } } From 780854186946e0de2be192ee7fa5125666533b3a Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Wed, 26 Oct 2022 14:39:59 +0800 Subject: [PATCH 024/138] wifi: mac80211: fix general-protection-fault in ieee80211_subif_start_xmit() When device is running and the interface status is changed, the gpf issue is triggered. The problem triggering process is as follows: Thread A: Thread B ieee80211_runtime_change_iftype() process_one_work() ... ... ieee80211_do_stop() ... ... ... sdata->bss = NULL ... ... ieee80211_subif_start_xmit() ieee80211_multicast_to_unicast //!sdata->bss->multicast_to_unicast cause gpf issue When the interface status is changed, the sending queue continues to send packets. After the bss is set to NULL, the bss is accessed. As a result, this causes a general-protection-fault issue. The following is the stack information: general protection fault, probably for non-canonical address 0xdffffc000000002f: 0000 [#1] PREEMPT SMP KASAN KASAN: null-ptr-deref in range [0x0000000000000178-0x000000000000017f] Workqueue: mld mld_ifc_work RIP: 0010:ieee80211_subif_start_xmit+0x25b/0x1310 Call Trace: dev_hard_start_xmit+0x1be/0x990 __dev_queue_xmit+0x2c9a/0x3b60 ip6_finish_output2+0xf92/0x1520 ip6_finish_output+0x6af/0x11e0 ip6_output+0x1ed/0x540 mld_sendpack+0xa09/0xe70 mld_ifc_work+0x71c/0xdb0 process_one_work+0x9bf/0x1710 worker_thread+0x665/0x1080 kthread+0x2e4/0x3a0 ret_from_fork+0x1f/0x30 Fixes: f856373e2f31 ("wifi: mac80211: do not wake queues on a vif that is being stopped") Reported-by: syzbot+c6e8fca81c294fd5620a@syzkaller.appspotmail.com Signed-off-by: Zhengchao Shao Link: https://lore.kernel.org/r/20221026063959.177813-1-shaozhengchao@huawei.com Signed-off-by: Johannes Berg --- net/mac80211/tx.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index a364148149f9..874f2a4d831d 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -4418,6 +4418,11 @@ netdev_tx_t ieee80211_subif_start_xmit(struct sk_buff *skb, if (likely(!is_multicast_ether_addr(eth->h_dest))) goto normal; + if (unlikely(!ieee80211_sdata_running(sdata))) { + kfree_skb(skb); + return NETDEV_TX_OK; + } + if (unlikely(ieee80211_multicast_to_unicast(skb, dev))) { struct sk_buff_head queue; From 39e7b5de9853bd92ddbfa4b14165babacd7da0ba Mon Sep 17 00:00:00 2001 From: Nicolas Cavallari Date: Thu, 27 Oct 2022 16:01:33 +0200 Subject: [PATCH 025/138] wifi: mac80211: Fix ack frame idr leak when mesh has no route When trying to transmit an data frame with tx_status to a destination that have no route in the mesh, then it is dropped without recrediting the ack_status_frames idr. Once it is exhausted, wpa_supplicant starts failing to do SAE with NL80211_CMD_FRAME and logs "nl80211: Frame command failed". Use ieee80211_free_txskb() instead of kfree_skb() to fix it. Signed-off-by: Nicolas Cavallari Link: https://lore.kernel.org/r/20221027140133.1504-1-nicolas.cavallari@green-communications.fr Signed-off-by: Johannes Berg --- net/mac80211/mesh_pathtbl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/mac80211/mesh_pathtbl.c b/net/mac80211/mesh_pathtbl.c index acc1c299f1ae..69d5e1ec6ede 100644 --- a/net/mac80211/mesh_pathtbl.c +++ b/net/mac80211/mesh_pathtbl.c @@ -710,7 +710,7 @@ int mesh_path_send_to_gates(struct mesh_path *mpath) void mesh_path_discard_frame(struct ieee80211_sub_if_data *sdata, struct sk_buff *skb) { - kfree_skb(skb); + ieee80211_free_txskb(&sdata->local->hw, skb); sdata->u.mesh.mshstats.dropped_frames_no_route++; } From 30ac96f7cc973bb850c718c9bbe1fdcedfbe826b Mon Sep 17 00:00:00 2001 From: Howard Hsu Date: Thu, 27 Oct 2022 09:56:53 +0800 Subject: [PATCH 026/138] wifi: mac80211: Set TWT Information Frame Disabled bit as 1 The TWT Information Frame Disabled bit of control field of TWT Setup frame shall be set to 1 since handling TWT Information frame is not supported by current mac80211 implementation. Fixes: f5a4c24e689f ("mac80211: introduce individual TWT support in AP mode") Signed-off-by: Howard Hsu Link: https://lore.kernel.org/r/20221027015653.1448-1-howard-yh.hsu@mediatek.com Signed-off-by: Johannes Berg --- net/mac80211/s1g.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/mac80211/s1g.c b/net/mac80211/s1g.c index 8ca7d45d6daa..c1f964e9991c 100644 --- a/net/mac80211/s1g.c +++ b/net/mac80211/s1g.c @@ -112,6 +112,9 @@ ieee80211_s1g_rx_twt_setup(struct ieee80211_sub_if_data *sdata, goto out; } + /* TWT Information not supported yet */ + twt->control |= IEEE80211_TWT_CONTROL_RX_DISABLED; + drv_add_twt_setup(sdata->local, sdata, &sta->sta, twt); out: ieee80211_s1g_send_twt_setup(sdata, mgmt->sa, sdata->vif.addr, twt); From 34de8e6e0e1f66e431abf4123934a2581cb5f133 Mon Sep 17 00:00:00 2001 From: Pu Lehui Date: Wed, 2 Nov 2022 16:40:34 +0800 Subject: [PATCH 027/138] bpftool: Fix NULL pointer dereference when pin {PROG, MAP, LINK} without FILE When using bpftool to pin {PROG, MAP, LINK} without FILE, segmentation fault will occur. The reson is that the lack of FILE will cause strlen to trigger NULL pointer dereference. The corresponding stacktrace is shown below: do_pin do_pin_any do_pin_fd mount_bpffs_for_pin strlen(name) <- NULL pointer dereference Fix it by adding validation to the common process. Fixes: 75a1e792c335 ("tools: bpftool: Allow all prog/map handles for pinning objects") Signed-off-by: Pu Lehui Signed-off-by: Daniel Borkmann Reviewed-by: Quentin Monnet Link: https://lore.kernel.org/bpf/20221102084034.3342995-1-pulehui@huaweicloud.com --- tools/bpf/bpftool/common.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/tools/bpf/bpftool/common.c b/tools/bpf/bpftool/common.c index 8727765add88..0cdb4f711510 100644 --- a/tools/bpf/bpftool/common.c +++ b/tools/bpf/bpftool/common.c @@ -300,6 +300,9 @@ int do_pin_any(int argc, char **argv, int (*get_fd)(int *, char ***)) int err; int fd; + if (!REQ_ARGS(3)) + return -EINVAL; + fd = get_fd(&argc, &argv); if (fd < 0) return fd; From 3a89b6dec9920026eaa90fe8457f4348d3388a98 Mon Sep 17 00:00:00 2001 From: "Tyler J. Stachecki" Date: Wed, 2 Nov 2022 18:56:39 +0200 Subject: [PATCH 028/138] wifi: ath11k: Fix QCN9074 firmware boot on x86 The 2.7.0 series of QCN9074's firmware requests 5 segments of memory instead of 3 (as in the 2.5.0 series). The first segment (11M) is too large to be kalloc'd in one go on x86 and requires piecemeal 1MB allocations, as was the case with the prior public firmware (2.5.0, 15M). Since f6f92968e1e5, ath11k will break the memory requests, but only if there were fewer than 3 segments requested by the firmware. It seems that 5 segments works fine and allows QCN9074 to boot on x86 with firmware 2.7.0, so change things accordingly. Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1 Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.5.0.1-01208-QCAHKSWPL_SILICONZ-1 Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.16 Signed-off-by: Tyler J. Stachecki Signed-off-by: Kalle Valo Link: https://lore.kernel.org/r/20221022042728.43015-1-stachecki.tyler@gmail.com --- drivers/net/wireless/ath/ath11k/qmi.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/ath/ath11k/qmi.h b/drivers/net/wireless/ath/ath11k/qmi.h index 2ec56a34fa81..0909d53cefeb 100644 --- a/drivers/net/wireless/ath/ath11k/qmi.h +++ b/drivers/net/wireless/ath/ath11k/qmi.h @@ -27,7 +27,7 @@ #define ATH11K_QMI_WLANFW_MAX_NUM_MEM_SEG_V01 52 #define ATH11K_QMI_CALDB_SIZE 0x480000 #define ATH11K_QMI_BDF_EXT_STR_LENGTH 0x20 -#define ATH11K_QMI_FW_MEM_REQ_SEGMENT_CNT 3 +#define ATH11K_QMI_FW_MEM_REQ_SEGMENT_CNT 5 #define QMI_WLFW_REQUEST_MEM_IND_V01 0x0035 #define QMI_WLFW_FW_MEM_READY_IND_V01 0x0037 From f45cb6b29cd36514e13f7519770873d8c0457008 Mon Sep 17 00:00:00 2001 From: Wen Gong Date: Wed, 2 Nov 2022 13:48:03 +0200 Subject: [PATCH 029/138] wifi: ath11k: avoid deadlock during regulatory update in ath11k_regd_update() (cherry picked from commit d99884ad9e3673a12879bc2830f6e5a66cccbd78 in ath-next as users are seeing this bug more now, also cc stable) Running this test in a loop it is easy to reproduce an rtnl deadlock: iw reg set FI ifconfig wlan0 down What happens is that thread A (workqueue) tries to update the regulatory: try to acquire the rtnl_lock of ar->regd_update_work rtnl_lock+0x17/0x20 ath11k_regd_update+0x15a/0x260 [ath11k] ath11k_regd_update_work+0x15/0x20 [ath11k] process_one_work+0x228/0x670 worker_thread+0x4d/0x440 kthread+0x16d/0x1b0 ret_from_fork+0x22/0x30 And thread B (ifconfig) tries to stop the interface: try to cancel_work_sync(&ar->regd_update_work) in ath11k_mac_op_stop(). ifconfig 3109 [003] 2414.232506: probe: ath11k_mac_op_stop: (ffffffffc14187a0) drv_stop+0x30 ([mac80211]) ieee80211_do_stop+0x5d2 ([mac80211]) ieee80211_stop+0x3e ([mac80211]) __dev_close_many+0x9e ([kernel.kallsyms]) __dev_change_flags+0xbe ([kernel.kallsyms]) dev_change_flags+0x23 ([kernel.kallsyms]) devinet_ioctl+0x5e3 ([kernel.kallsyms]) inet_ioctl+0x197 ([kernel.kallsyms]) sock_do_ioctl+0x4d ([kernel.kallsyms]) sock_ioctl+0x264 ([kernel.kallsyms]) __x64_sys_ioctl+0x92 ([kernel.kallsyms]) do_syscall_64+0x3a ([kernel.kallsyms]) entry_SYSCALL_64_after_hwframe+0x63 ([kernel.kallsyms]) __GI___ioctl+0x7 (/lib/x86_64-linux-gnu/libc-2.23.so) The sequence of deadlock is: 1. Thread B calls rtnl_lock(). 2. Thread A starts to run and calls rtnl_lock() from within ath11k_regd_update_work(), then enters wait state because the lock is owned by thread B. 3. Thread B continues to run and tries to call cancel_work_sync(&ar->regd_update_work), but thread A is in ath11k_regd_update_work() waiting for rtnl_lock(). So cancel_work_sync() forever waits for ath11k_regd_update_work() to finish and we have a deadlock. Fix this by switching from using regulatory_set_wiphy_regd_sync() to regulatory_set_wiphy_regd(). Now cfg80211 will schedule another workqueue which handles the locking on it's own. So the ath11k workqueue can simply exit without taking any locks, avoiding the deadlock. Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3 Cc: Signed-off-by: Wen Gong [kvalo: improve commit log] Signed-off-by: Kalle Valo --- drivers/net/wireless/ath/ath11k/reg.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/drivers/net/wireless/ath/ath11k/reg.c b/drivers/net/wireless/ath/ath11k/reg.c index 7ee3ff69dfc8..6fae4e61ede7 100644 --- a/drivers/net/wireless/ath/ath11k/reg.c +++ b/drivers/net/wireless/ath/ath11k/reg.c @@ -287,11 +287,7 @@ int ath11k_regd_update(struct ath11k *ar) goto err; } - rtnl_lock(); - wiphy_lock(ar->hw->wiphy); - ret = regulatory_set_wiphy_regd_sync(ar->hw->wiphy, regd_copy); - wiphy_unlock(ar->hw->wiphy); - rtnl_unlock(); + ret = regulatory_set_wiphy_regd(ar->hw->wiphy, regd_copy); kfree(regd_copy); From aec1dc972d27c837d1406310dab5170189eb01e5 Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Wed, 2 Nov 2022 11:25:16 -0700 Subject: [PATCH 030/138] net/ipv4: Fix linux/in.h header dependencies __DECLARE_FLEX_ARRAY is defined in include/uapi/linux/stddef.h but doesn't seem to be explicitly included from include/uapi/linux/in.h, which breaks BPF selftests builds (once we sync linux/stddef.h into tools/include directory in the next patch). Fix this by explicitly including linux/stddef.h. Given this affects BPF CI and bpf tree, targeting this for bpf tree. Fixes: 5854a09b4957 ("net/ipv4: Use __DECLARE_FLEX_ARRAY() helper") Signed-off-by: Andrii Nakryiko Signed-off-by: Daniel Borkmann Cc: Gustavo A. R. Silva Cc: Jakub Kicinski Link: https://lore.kernel.org/bpf/20221102182517.2675301-1-andrii@kernel.org --- include/uapi/linux/in.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/uapi/linux/in.h b/include/uapi/linux/in.h index f243ce665f74..07a4cb149305 100644 --- a/include/uapi/linux/in.h +++ b/include/uapi/linux/in.h @@ -20,6 +20,7 @@ #define _UAPI_LINUX_IN_H #include +#include #include #include From a778f5d46b6287ebe26e24b48f3e8079c2db8ed2 Mon Sep 17 00:00:00 2001 From: Andrii Nakryiko Date: Wed, 2 Nov 2022 11:25:17 -0700 Subject: [PATCH 031/138] tools/headers: Pull in stddef.h to uapi to fix BPF selftests build in CI With recent sync of linux/in.h tools/include headers are now relying on __DECLARE_FLEX_ARRAY macro, which isn't itself defined inside tools/include headers anywhere and is instead assumed to be present in system-wide UAPI header. This breaks isolated environments that don't have kernel UAPI headers installed system-wide, like BPF CI ([0]). To fix this, bring in include/uapi/linux/stddef.h into tools/include. We can't just copy/paste it, though, it has to be processed with scripts/headers_install.sh, which has a dependency on scripts/unifdef. So the full command to (re-)generate stddef.h for inclusion into tools/include directory is: $ make scripts_unifdef && \ cp $KBUILD_OUTPUT/scripts/unifdef scripts/ && \ scripts/headers_install.sh include/uapi/linux/stddef.h tools/include/uapi/linux/stddef.h This assumes KBUILD_OUTPUT envvar is set and used for out-of-tree builds. [0] https://github.com/kernel-patches/bpf/actions/runs/3379432493/jobs/5610982609 Fixes: 036b8f5b8970 ("tools headers uapi: Update linux/in.h copy") Signed-off-by: Andrii Nakryiko Signed-off-by: Daniel Borkmann Cc: Jakub Kicinski Cc: Arnaldo Carvalho de Melo Link: https://lore.kernel.org/bpf/20221102182517.2675301-2-andrii@kernel.org --- tools/include/uapi/linux/in.h | 1 + tools/include/uapi/linux/stddef.h | 47 +++++++++++++++++++++++++++++++ 2 files changed, 48 insertions(+) create mode 100644 tools/include/uapi/linux/stddef.h diff --git a/tools/include/uapi/linux/in.h b/tools/include/uapi/linux/in.h index f243ce665f74..07a4cb149305 100644 --- a/tools/include/uapi/linux/in.h +++ b/tools/include/uapi/linux/in.h @@ -20,6 +20,7 @@ #define _UAPI_LINUX_IN_H #include +#include #include #include diff --git a/tools/include/uapi/linux/stddef.h b/tools/include/uapi/linux/stddef.h new file mode 100644 index 000000000000..bb6ea517efb5 --- /dev/null +++ b/tools/include/uapi/linux/stddef.h @@ -0,0 +1,47 @@ +/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ +#ifndef _LINUX_STDDEF_H +#define _LINUX_STDDEF_H + + + +#ifndef __always_inline +#define __always_inline __inline__ +#endif + +/** + * __struct_group() - Create a mirrored named and anonyomous struct + * + * @TAG: The tag name for the named sub-struct (usually empty) + * @NAME: The identifier name of the mirrored sub-struct + * @ATTRS: Any struct attributes (usually empty) + * @MEMBERS: The member declarations for the mirrored structs + * + * Used to create an anonymous union of two structs with identical layout + * and size: one anonymous and one named. The former's members can be used + * normally without sub-struct naming, and the latter can be used to + * reason about the start, end, and size of the group of struct members. + * The named struct can also be explicitly tagged for layer reuse, as well + * as both having struct attributes appended. + */ +#define __struct_group(TAG, NAME, ATTRS, MEMBERS...) \ + union { \ + struct { MEMBERS } ATTRS; \ + struct TAG { MEMBERS } ATTRS NAME; \ + } + +/** + * __DECLARE_FLEX_ARRAY() - Declare a flexible array usable in a union + * + * @TYPE: The type of each flexible array element + * @NAME: The name of the flexible array member + * + * In order to have a flexible array member in a union or alone in a + * struct, it needs to be wrapped in an anonymous struct with at least 1 + * named member, but that member can be empty. + */ +#define __DECLARE_FLEX_ARRAY(TYPE, NAME) \ + struct { \ + struct { } __empty_ ## NAME; \ + TYPE NAME[]; \ + } +#endif From 8bbabb3fddcd0f858be69ed5abc9b470a239d6f2 Mon Sep 17 00:00:00 2001 From: Cong Wang Date: Tue, 1 Nov 2022 21:34:17 -0700 Subject: [PATCH 032/138] bpf, sock_map: Move cancel_work_sync() out of sock lock Stanislav reported a lockdep warning, which is caused by the cancel_work_sync() called inside sock_map_close(), as analyzed below by Jakub: psock->work.func = sk_psock_backlog() ACQUIRE psock->work_mutex sk_psock_handle_skb() skb_send_sock() __skb_send_sock() sendpage_unlocked() kernel_sendpage() sock->ops->sendpage = inet_sendpage() sk->sk_prot->sendpage = tcp_sendpage() ACQUIRE sk->sk_lock tcp_sendpage_locked() RELEASE sk->sk_lock RELEASE psock->work_mutex sock_map_close() ACQUIRE sk->sk_lock sk_psock_stop() sk_psock_clear_state(psock, SK_PSOCK_TX_ENABLED) cancel_work_sync() __cancel_work_timer() __flush_work() // wait for psock->work to finish RELEASE sk->sk_lock We can move the cancel_work_sync() out of the sock lock protection, but still before saved_close() was called. Fixes: 799aa7f98d53 ("skmsg: Avoid lock_sock() in sk_psock_backlog()") Reported-by: Stanislav Fomichev Signed-off-by: Cong Wang Signed-off-by: Daniel Borkmann Tested-by: Jakub Sitnicki Acked-by: John Fastabend Acked-by: Jakub Sitnicki Link: https://lore.kernel.org/bpf/20221102043417.279409-1-xiyou.wangcong@gmail.com --- include/linux/skmsg.h | 2 +- net/core/skmsg.c | 7 ++----- net/core/sock_map.c | 7 ++++--- 3 files changed, 7 insertions(+), 9 deletions(-) diff --git a/include/linux/skmsg.h b/include/linux/skmsg.h index 48f4b645193b..70d6cb94e580 100644 --- a/include/linux/skmsg.h +++ b/include/linux/skmsg.h @@ -376,7 +376,7 @@ static inline void sk_psock_report_error(struct sk_psock *psock, int err) } struct sk_psock *sk_psock_init(struct sock *sk, int node); -void sk_psock_stop(struct sk_psock *psock, bool wait); +void sk_psock_stop(struct sk_psock *psock); #if IS_ENABLED(CONFIG_BPF_STREAM_PARSER) int sk_psock_init_strp(struct sock *sk, struct sk_psock *psock); diff --git a/net/core/skmsg.c b/net/core/skmsg.c index 1efdc47a999b..e6b9ced3eda8 100644 --- a/net/core/skmsg.c +++ b/net/core/skmsg.c @@ -803,16 +803,13 @@ static void sk_psock_link_destroy(struct sk_psock *psock) } } -void sk_psock_stop(struct sk_psock *psock, bool wait) +void sk_psock_stop(struct sk_psock *psock) { spin_lock_bh(&psock->ingress_lock); sk_psock_clear_state(psock, SK_PSOCK_TX_ENABLED); sk_psock_cork_free(psock); __sk_psock_zap_ingress(psock); spin_unlock_bh(&psock->ingress_lock); - - if (wait) - cancel_work_sync(&psock->work); } static void sk_psock_done_strp(struct sk_psock *psock); @@ -850,7 +847,7 @@ void sk_psock_drop(struct sock *sk, struct sk_psock *psock) sk_psock_stop_verdict(sk, psock); write_unlock_bh(&sk->sk_callback_lock); - sk_psock_stop(psock, false); + sk_psock_stop(psock); INIT_RCU_WORK(&psock->rwork, sk_psock_destroy); queue_rcu_work(system_wq, &psock->rwork); diff --git a/net/core/sock_map.c b/net/core/sock_map.c index a660baedd9e7..81beb16ab1eb 100644 --- a/net/core/sock_map.c +++ b/net/core/sock_map.c @@ -1596,7 +1596,7 @@ void sock_map_destroy(struct sock *sk) saved_destroy = psock->saved_destroy; sock_map_remove_links(sk, psock); rcu_read_unlock(); - sk_psock_stop(psock, false); + sk_psock_stop(psock); sk_psock_put(sk, psock); saved_destroy(sk); } @@ -1619,9 +1619,10 @@ void sock_map_close(struct sock *sk, long timeout) saved_close = psock->saved_close; sock_map_remove_links(sk, psock); rcu_read_unlock(); - sk_psock_stop(psock, true); - sk_psock_put(sk, psock); + sk_psock_stop(psock); release_sock(sk); + cancel_work_sync(&psock->work); + sk_psock_put(sk, psock); saved_close(sk, timeout); } EXPORT_SYMBOL_GPL(sock_map_close); From a207620123f27e6f63bab13b46b20e03a3deec8d Mon Sep 17 00:00:00 2001 From: Lukas Bulwahn Date: Mon, 31 Oct 2022 10:29:20 +0100 Subject: [PATCH 033/138] mm/slab_common: repair kernel-doc for __ksize() Commit 445d41d7a7c1 ("Merge branch 'slab/for-6.1/kmalloc_size_roundup' into slab/for-next") resolved a conflict of two concurrent changes to __ksize(). However, it did not adjust the kernel-doc comment of __ksize(), while the name of the argument to __ksize() was renamed. Hence, ./scripts/ kernel-doc -none mm/slab_common.c warns about it. Adjust the kernel-doc comment for __ksize() for make W=1 happiness. Signed-off-by: Lukas Bulwahn Acked-by: Hyeonggon Yoo <42.hyeyoo@gmail.com> Signed-off-by: Vlastimil Babka --- mm/slab_common.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/mm/slab_common.c b/mm/slab_common.c index 33b1886b06eb..74a991fd9d31 100644 --- a/mm/slab_common.c +++ b/mm/slab_common.c @@ -1010,7 +1010,7 @@ EXPORT_SYMBOL(kfree); /** * __ksize -- Report full size of underlying allocation - * @objp: pointer to the object + * @object: pointer to the object * * This should only be used internally to query the true size of allocations. * It is not meant to be a way to discover the usable size of an allocation @@ -1018,7 +1018,7 @@ EXPORT_SYMBOL(kfree); * the originally requested allocation size may trigger KASAN, UBSAN_BOUNDS, * and/or FORTIFY_SOURCE. * - * Return: size of the actual memory used by @objp in bytes + * Return: size of the actual memory used by @object in bytes */ size_t __ksize(const void *object) { From f1db20814af532f85e091231223e5e4818e8464b Mon Sep 17 00:00:00 2001 From: Youlin Li Date: Thu, 3 Nov 2022 17:34:39 +0800 Subject: [PATCH 034/138] bpf: Fix wrong reg type conversion in release_reference() Some helper functions will allocate memory. To avoid memory leaks, the verifier requires the eBPF program to release these memories by calling the corresponding helper functions. When a resource is released, all pointer registers corresponding to the resource should be invalidated. The verifier use release_references() to do this job, by apply __mark_reg_unknown() to each relevant register. It will give these registers the type of SCALAR_VALUE. A register that will contain a pointer value at runtime, but of type SCALAR_VALUE, which may allow the unprivileged user to get a kernel pointer by storing this register into a map. Using __mark_reg_not_init() while NOT allow_ptr_leaks can mitigate this problem. Fixes: fd978bf7fd31 ("bpf: Add reference tracking to verifier") Signed-off-by: Youlin Li Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20221103093440.3161-1-liulin063@gmail.com --- kernel/bpf/verifier.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index dd9019c8b0db..225666307bba 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -6623,8 +6623,12 @@ static int release_reference(struct bpf_verifier_env *env, return err; bpf_for_each_reg_in_vstate(env->cur_state, state, reg, ({ - if (reg->ref_obj_id == ref_obj_id) - __mark_reg_unknown(env, reg); + if (reg->ref_obj_id == ref_obj_id) { + if (!env->allow_ptr_leaks) + __mark_reg_not_init(env, reg); + else + __mark_reg_unknown(env, reg); + } })); return 0; From 475244f5e06beeda7b557d9dde46a5f439bf3379 Mon Sep 17 00:00:00 2001 From: Youlin Li Date: Thu, 3 Nov 2022 17:34:40 +0800 Subject: [PATCH 035/138] selftests/bpf: Add verifier test for release_reference() Add a test case to ensure that released pointer registers will not be leaked into the map. Before fix: ./test_verifier 984 984/u reference tracking: try to leak released ptr reg FAIL Unexpected success to load! verification time 67 usec stack depth 4 processed 23 insns (limit 1000000) max_states_per_insn 0 total_states 2 peak_states 2 mark_read 1 984/p reference tracking: try to leak released ptr reg OK Summary: 1 PASSED, 0 SKIPPED, 1 FAILED After fix: ./test_verifier 984 984/u reference tracking: try to leak released ptr reg OK 984/p reference tracking: try to leak released ptr reg OK Summary: 2 PASSED, 0 SKIPPED, 0 FAILED Signed-off-by: Youlin Li Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20221103093440.3161-2-liulin063@gmail.com --- .../selftests/bpf/verifier/ref_tracking.c | 36 +++++++++++++++++++ 1 file changed, 36 insertions(+) diff --git a/tools/testing/selftests/bpf/verifier/ref_tracking.c b/tools/testing/selftests/bpf/verifier/ref_tracking.c index f18ce867271f..fd683a32a276 100644 --- a/tools/testing/selftests/bpf/verifier/ref_tracking.c +++ b/tools/testing/selftests/bpf/verifier/ref_tracking.c @@ -1044,3 +1044,39 @@ .result_unpriv = REJECT, .errstr_unpriv = "unknown func", }, +{ + "reference tracking: try to leak released ptr reg", + .insns = { + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_0, -4), + BPF_MOV64_REG(BPF_REG_2, BPF_REG_10), + BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4), + BPF_LD_MAP_FD(BPF_REG_1, 0), + BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem), + BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1), + BPF_EXIT_INSN(), + BPF_MOV64_REG(BPF_REG_9, BPF_REG_0), + + BPF_MOV64_IMM(BPF_REG_0, 0), + BPF_LD_MAP_FD(BPF_REG_1, 0), + BPF_MOV64_IMM(BPF_REG_2, 8), + BPF_MOV64_IMM(BPF_REG_3, 0), + BPF_EMIT_CALL(BPF_FUNC_ringbuf_reserve), + BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1), + BPF_EXIT_INSN(), + BPF_MOV64_REG(BPF_REG_8, BPF_REG_0), + + BPF_MOV64_REG(BPF_REG_1, BPF_REG_8), + BPF_MOV64_IMM(BPF_REG_2, 0), + BPF_EMIT_CALL(BPF_FUNC_ringbuf_discard), + BPF_MOV64_IMM(BPF_REG_0, 0), + + BPF_STX_MEM(BPF_DW, BPF_REG_9, BPF_REG_8, 0), + BPF_EXIT_INSN() + }, + .fixup_map_array_48b = { 4 }, + .fixup_map_ringbuf = { 11 }, + .result = ACCEPT, + .result_unpriv = REJECT, + .errstr_unpriv = "R8 !read_ok" +}, From 9e4b7a99a03aefd37ba7bb1f022c8efab5019165 Mon Sep 17 00:00:00 2001 From: Jiri Benc Date: Wed, 2 Nov 2022 17:53:25 +0100 Subject: [PATCH 036/138] net: gso: fix panic on frag_list with mixed head alloc types Since commit 3dcbdb134f32 ("net: gso: Fix skb_segment splat when splitting gso_size mangled skb having linear-headed frag_list"), it is allowed to change gso_size of a GRO packet. However, that commit assumes that "checking the first list_skb member suffices; i.e if either of the list_skb members have non head_frag head, then the first one has too". It turns out this assumption does not hold. We've seen BUG_ON being hit in skb_segment when skbs on the frag_list had differing head_frag with the vmxnet3 driver. This happens because __netdev_alloc_skb and __napi_alloc_skb can return a skb that is page backed or kmalloced depending on the requested size. As the result, the last small skb in the GRO packet can be kmalloced. There are three different locations where this can be fixed: (1) We could check head_frag in GRO and not allow GROing skbs with different head_frag. However, that would lead to performance regression on normal forward paths with unmodified gso_size, where !head_frag in the last packet is not a problem. (2) Set a flag in bpf_skb_net_grow and bpf_skb_net_shrink indicating that NETIF_F_SG is undesirable. That would need to eat a bit in sk_buff. Furthermore, that flag can be unset when all skbs on the frag_list are page backed. To retain good performance, bpf_skb_net_grow/shrink would have to walk the frag_list. (3) Walk the frag_list in skb_segment when determining whether NETIF_F_SG should be cleared. This of course slows things down. This patch implements (3). To limit the performance impact in skb_segment, the list is walked only for skbs with SKB_GSO_DODGY set that have gso_size changed. Normal paths thus will not hit it. We could check only the last skb but since we need to walk the whole list anyway, let's stay on the safe side. Fixes: 3dcbdb134f32 ("net: gso: Fix skb_segment splat when splitting gso_size mangled skb having linear-headed frag_list") Signed-off-by: Jiri Benc Reviewed-by: Willem de Bruijn Link: https://lore.kernel.org/r/e04426a6a91baf4d1081e1b478c82b5de25fdf21.1667407944.git.jbenc@redhat.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 36 +++++++++++++++++++----------------- 1 file changed, 19 insertions(+), 17 deletions(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index d1a3fa6f3f12..88fa40571d0c 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -4134,23 +4134,25 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb, int i = 0; int pos; - if (list_skb && !list_skb->head_frag && skb_headlen(list_skb) && - (skb_shinfo(head_skb)->gso_type & SKB_GSO_DODGY)) { - /* gso_size is untrusted, and we have a frag_list with a linear - * non head_frag head. - * - * (we assume checking the first list_skb member suffices; - * i.e if either of the list_skb members have non head_frag - * head, then the first one has too). - * - * If head_skb's headlen does not fit requested gso_size, it - * means that the frag_list members do NOT terminate on exact - * gso_size boundaries. Hence we cannot perform skb_frag_t page - * sharing. Therefore we must fallback to copying the frag_list - * skbs; we do so by disabling SG. - */ - if (mss != GSO_BY_FRAGS && mss != skb_headlen(head_skb)) - features &= ~NETIF_F_SG; + if ((skb_shinfo(head_skb)->gso_type & SKB_GSO_DODGY) && + mss != GSO_BY_FRAGS && mss != skb_headlen(head_skb)) { + struct sk_buff *check_skb; + + for (check_skb = list_skb; check_skb; check_skb = check_skb->next) { + if (skb_headlen(check_skb) && !check_skb->head_frag) { + /* gso_size is untrusted, and we have a frag_list with + * a linear non head_frag item. + * + * If head_skb's headlen does not fit requested gso_size, + * it means that the frag_list members do NOT terminate + * on exact gso_size boundaries. Hence we cannot perform + * skb_frag_t page sharing. Therefore we must fallback to + * copying the frag_list skbs; we do so by disabling SG. + */ + features &= ~NETIF_F_SG; + break; + } + } } __skb_push(head_skb, doffset); From 4581dd480c9e42a1ad21dd8b9c110abe41878ce5 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Mon, 31 Oct 2022 22:08:13 -0700 Subject: [PATCH 037/138] net: octeontx2-pf: mcs: consider MACSEC setting Fix build errors when MACSEC=m and OCTEONTX2_PF=y by having OCTEONTX2_PF depend on MACSEC if it is enabled. By adding "|| !MACSEC", this means that MACSEC is not required -- it can be disabled for this driver. drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.o: in function `otx2_remove': ../drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c:(.text+0x2fd0): undefined reference to `cn10k_mcs_free' mips64-linux-ld: drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.o: in function `otx2_mbox_up_handler_mcs_intr_notify': ../drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c:(.text+0x4610): undefined reference to `cn10k_handle_mcs_event' Reported-by: kernel test robot Fixes: c54ffc73601c ("octeontx2-pf: mcs: Introduce MACSEC hardware offloading") Signed-off-by: Randy Dunlap Cc: Subbaraya Sundeep Cc: Sunil Goutham Cc: Geetha sowjanya Cc: hariprasad Cc: "David S. Miller" Cc: Eric Dumazet Cc: Jakub Kicinski Cc: Paolo Abeni Signed-off-by: David S. Miller --- drivers/net/ethernet/marvell/octeontx2/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/marvell/octeontx2/Kconfig b/drivers/net/ethernet/marvell/octeontx2/Kconfig index e1036b0eb6b1..993ac180a5db 100644 --- a/drivers/net/ethernet/marvell/octeontx2/Kconfig +++ b/drivers/net/ethernet/marvell/octeontx2/Kconfig @@ -36,6 +36,7 @@ config OCTEONTX2_PF select DIMLIB depends on PCI depends on PTP_1588_CLOCK_OPTIONAL + depends on MACSEC || !MACSEC help This driver supports Marvell's OcteonTX2 NIC physical function. From cfdcb075048c1e886c45a9c9e681ed222f74ecb9 Mon Sep 17 00:00:00 2001 From: Guangbin Huang Date: Tue, 1 Nov 2022 15:48:38 +0800 Subject: [PATCH 038/138] net: hns3: fix get wrong value of function hclge_get_dscp_prio() As the argument struct hnae3_handle *h of function hclge_get_dscp_prio() can be other client registered in hnae3 layer, we need to transform it into hnae3_handle of local nic client to get right dscp settings for other clients. Fixes: dfea275e06c2 ("net: hns3: optimize converting dscp to priority process of hns3_nic_select_queue()") Signed-off-by: Guangbin Huang Signed-off-by: David S. Miller --- drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index 6962a9d69cf8..987271da6e9b 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -12984,14 +12984,16 @@ static void hclge_clean_vport_config(struct hnae3_ae_dev *ae_dev, int num_vfs) static int hclge_get_dscp_prio(struct hnae3_handle *h, u8 dscp, u8 *tc_mode, u8 *priority) { + struct hclge_vport *vport = hclge_get_vport(h); + if (dscp >= HNAE3_MAX_DSCP) return -EINVAL; if (tc_mode) - *tc_mode = h->kinfo.tc_map_mode; + *tc_mode = vport->nic.kinfo.tc_map_mode; if (priority) - *priority = h->kinfo.dscp_prio[dscp] == HNAE3_PRIO_ID_INVALID ? 0 : - h->kinfo.dscp_prio[dscp]; + *priority = vport->nic.kinfo.dscp_prio[dscp] == HNAE3_PRIO_ID_INVALID ? 0 : + vport->nic.kinfo.dscp_prio[dscp]; return 0; } From cdb525ca92b196f8916102b62431aa0d9a644ff2 Mon Sep 17 00:00:00 2001 From: Adrien Thierry Date: Tue, 1 Nov 2022 14:48:08 -0400 Subject: [PATCH 039/138] selftests/net: give more time to udpgro bg processes to complete startup In some conditions, background processes in udpgro don't have enough time to set up the sockets. When foreground processes start, this results in the test failing with "./udpgso_bench_tx: sendmsg: Connection refused". For instance, this happens from time to time on a Qualcomm SA8540P SoC running CentOS Stream 9. To fix this, increase the time given to background processes to complete the startup before foreground processes start. Signed-off-by: Adrien Thierry Signed-off-by: David S. Miller --- tools/testing/selftests/net/udpgro.sh | 4 ++-- tools/testing/selftests/net/udpgro_bench.sh | 2 +- tools/testing/selftests/net/udpgro_frglist.sh | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) diff --git a/tools/testing/selftests/net/udpgro.sh b/tools/testing/selftests/net/udpgro.sh index ebbd0b282432..6a443ca3cd3a 100755 --- a/tools/testing/selftests/net/udpgro.sh +++ b/tools/testing/selftests/net/udpgro.sh @@ -50,7 +50,7 @@ run_one() { echo "failed" & # Hack: let bg programs complete the startup - sleep 0.1 + sleep 0.2 ./udpgso_bench_tx ${tx_args} ret=$? wait $(jobs -p) @@ -117,7 +117,7 @@ run_one_2sock() { echo "failed" & # Hack: let bg programs complete the startup - sleep 0.1 + sleep 0.2 ./udpgso_bench_tx ${tx_args} -p 12345 sleep 0.1 # first UDP GSO socket should be closed at this point diff --git a/tools/testing/selftests/net/udpgro_bench.sh b/tools/testing/selftests/net/udpgro_bench.sh index fad2d1a71cac..8a1109a545db 100755 --- a/tools/testing/selftests/net/udpgro_bench.sh +++ b/tools/testing/selftests/net/udpgro_bench.sh @@ -39,7 +39,7 @@ run_one() { ip netns exec "${PEER_NS}" ./udpgso_bench_rx -t ${rx_args} -r & # Hack: let bg programs complete the startup - sleep 0.1 + sleep 0.2 ./udpgso_bench_tx ${tx_args} } diff --git a/tools/testing/selftests/net/udpgro_frglist.sh b/tools/testing/selftests/net/udpgro_frglist.sh index 832c738cc3c2..7fe85ba51075 100755 --- a/tools/testing/selftests/net/udpgro_frglist.sh +++ b/tools/testing/selftests/net/udpgro_frglist.sh @@ -44,7 +44,7 @@ run_one() { ip netns exec "${PEER_NS}" ./udpgso_bench_rx ${rx_args} -r & # Hack: let bg programs complete the startup - sleep 0.1 + sleep 0.2 ./udpgso_bench_tx ${tx_args} } From 8bcd560ae8784da57c610d857118c5d6576b1a8f Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Wed, 2 Nov 2022 22:33:12 +0100 Subject: [PATCH 040/138] Revert "net: macsec: report real_dev features when HW offloading is enabled" This reverts commit c850240b6c4132574a00f2da439277ab94265b66. That commit tried to improve the performance of macsec offload by taking advantage of some of the NIC's features, but in doing so, broke macsec offload when the lower device supports both macsec and ipsec offload, as the ipsec offload feature flags (mainly NETIF_F_HW_ESP) were copied from the real device. Since the macsec device doesn't provide xdo_* ops, the XFRM core rejects the registration of the new macsec device in xfrm_api_check. Example perf trace when running ip link add link eni1np1 type macsec port 4 offload mac ip 737 [003] 795.477676: probe:xfrm_dev_event__REGISTER name="macsec0" features=0x1c000080014869 xfrm_dev_event+0x3a notifier_call_chain+0x47 register_netdevice+0x846 macsec_newlink+0x25a ip 737 [003] 795.477687: probe:xfrm_dev_event__return ret=0x8002 (NOTIFY_BAD) notifier_call_chain+0x47 register_netdevice+0x846 macsec_newlink+0x25a dev->features includes NETIF_F_HW_ESP (0x04000000000000), so xfrm_api_check returns NOTIFY_BAD because we don't have dev->xfrmdev_ops on the macsec device. We could probably propagate GSO and a few other features from the lower device, similar to macvlan. This will be done in a future patch. Signed-off-by: Sabrina Dubroca Reviewed-by: Antoine Tenart Reviewed-by: Leon Romanovsky Signed-off-by: David S. Miller --- drivers/net/macsec.c | 27 ++++----------------------- 1 file changed, 4 insertions(+), 23 deletions(-) diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index c891b60937a7..b3f76e8071f2 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -2654,11 +2654,6 @@ static int macsec_upd_offload(struct sk_buff *skb, struct genl_info *info) if (ret) goto rollback; - /* Force features update, since they are different for SW MACSec and - * HW offloading cases. - */ - netdev_update_features(dev); - rtnl_unlock(); return 0; @@ -3432,16 +3427,9 @@ static netdev_tx_t macsec_start_xmit(struct sk_buff *skb, return ret; } -#define SW_MACSEC_FEATURES \ +#define MACSEC_FEATURES \ (NETIF_F_SG | NETIF_F_HIGHDMA | NETIF_F_FRAGLIST) -/* If h/w offloading is enabled, use real device features save for - * VLAN_FEATURES - they require additional ops - * HW_MACSEC - no reason to report it - */ -#define REAL_DEV_FEATURES(dev) \ - ((dev)->features & ~(NETIF_F_VLAN_FEATURES | NETIF_F_HW_MACSEC)) - static int macsec_dev_init(struct net_device *dev) { struct macsec_dev *macsec = macsec_priv(dev); @@ -3458,12 +3446,8 @@ static int macsec_dev_init(struct net_device *dev) return err; } - if (macsec_is_offloaded(macsec)) { - dev->features = REAL_DEV_FEATURES(real_dev); - } else { - dev->features = real_dev->features & SW_MACSEC_FEATURES; - dev->features |= NETIF_F_LLTX | NETIF_F_GSO_SOFTWARE; - } + dev->features = real_dev->features & MACSEC_FEATURES; + dev->features |= NETIF_F_LLTX | NETIF_F_GSO_SOFTWARE; dev->needed_headroom = real_dev->needed_headroom + MACSEC_NEEDED_HEADROOM; @@ -3495,10 +3479,7 @@ static netdev_features_t macsec_fix_features(struct net_device *dev, struct macsec_dev *macsec = macsec_priv(dev); struct net_device *real_dev = macsec->real_dev; - if (macsec_is_offloaded(macsec)) - return REAL_DEV_FEATURES(real_dev); - - features &= (real_dev->features & SW_MACSEC_FEATURES) | + features &= (real_dev->features & MACSEC_FEATURES) | NETIF_F_GSO_SOFTWARE | NETIF_F_SOFT_FEATURES; features |= NETIF_F_LLTX; From 93a30947821c203d08865c4e17ea181c9668ce52 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Wed, 2 Nov 2022 22:33:13 +0100 Subject: [PATCH 041/138] macsec: delete new rxsc when offload fails Currently we get an inconsistent state: - netlink returns the error to userspace - the RXSC is installed but not offloaded Then the device could get confused when we try to add an RXSA, because the RXSC isn't supposed to exist. Fixes: 3cf3227a21d1 ("net: macsec: hardware offloading infrastructure") Signed-off-by: Sabrina Dubroca Reviewed-by: Antoine Tenart Reviewed-by: Leon Romanovsky Signed-off-by: David S. Miller --- drivers/net/macsec.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index b3f76e8071f2..0d6fe34b91ae 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -1876,7 +1876,6 @@ static int macsec_add_rxsc(struct sk_buff *skb, struct genl_info *info) struct macsec_rx_sc *rx_sc; struct nlattr *tb_rxsc[MACSEC_RXSC_ATTR_MAX + 1]; struct macsec_secy *secy; - bool was_active; int ret; if (!attrs[MACSEC_ATTR_IFINDEX]) @@ -1904,7 +1903,6 @@ static int macsec_add_rxsc(struct sk_buff *skb, struct genl_info *info) return PTR_ERR(rx_sc); } - was_active = rx_sc->active; if (tb_rxsc[MACSEC_RXSC_ATTR_ACTIVE]) rx_sc->active = !!nla_get_u8(tb_rxsc[MACSEC_RXSC_ATTR_ACTIVE]); @@ -1931,7 +1929,8 @@ static int macsec_add_rxsc(struct sk_buff *skb, struct genl_info *info) return 0; cleanup: - rx_sc->active = was_active; + del_rx_sc(secy, sci); + free_rx_sc(rx_sc); rtnl_unlock(); return ret; } From 73a4b31c9d11f98ae3bc5286d5382930adb0e9c7 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Wed, 2 Nov 2022 22:33:14 +0100 Subject: [PATCH 042/138] macsec: fix secy->n_rx_sc accounting secy->n_rx_sc is supposed to be the number of _active_ rxsc's within a secy. This is then used by macsec_send_sci to help decide if we should add the SCI to the header or not. This logic is currently broken when we create a new RXSC and turn it off at creation, as create_rx_sc always sets ->active to true (and immediately uses that to increment n_rx_sc), and only later macsec_add_rxsc sets rx_sc->active. Fixes: c09440f7dcb3 ("macsec: introduce IEEE 802.1AE driver") Signed-off-by: Sabrina Dubroca Reviewed-by: Antoine Tenart Reviewed-by: Leon Romanovsky Signed-off-by: David S. Miller --- drivers/net/macsec.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index 0d6fe34b91ae..1b4d856f4bd7 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -1413,7 +1413,8 @@ static struct macsec_rx_sc *del_rx_sc(struct macsec_secy *secy, sci_t sci) return NULL; } -static struct macsec_rx_sc *create_rx_sc(struct net_device *dev, sci_t sci) +static struct macsec_rx_sc *create_rx_sc(struct net_device *dev, sci_t sci, + bool active) { struct macsec_rx_sc *rx_sc; struct macsec_dev *macsec; @@ -1437,7 +1438,7 @@ static struct macsec_rx_sc *create_rx_sc(struct net_device *dev, sci_t sci) } rx_sc->sci = sci; - rx_sc->active = true; + rx_sc->active = active; refcount_set(&rx_sc->refcnt, 1); secy = &macsec_priv(dev)->secy; @@ -1876,6 +1877,7 @@ static int macsec_add_rxsc(struct sk_buff *skb, struct genl_info *info) struct macsec_rx_sc *rx_sc; struct nlattr *tb_rxsc[MACSEC_RXSC_ATTR_MAX + 1]; struct macsec_secy *secy; + bool active = true; int ret; if (!attrs[MACSEC_ATTR_IFINDEX]) @@ -1897,15 +1899,15 @@ static int macsec_add_rxsc(struct sk_buff *skb, struct genl_info *info) secy = &macsec_priv(dev)->secy; sci = nla_get_sci(tb_rxsc[MACSEC_RXSC_ATTR_SCI]); - rx_sc = create_rx_sc(dev, sci); + if (tb_rxsc[MACSEC_RXSC_ATTR_ACTIVE]) + active = nla_get_u8(tb_rxsc[MACSEC_RXSC_ATTR_ACTIVE]); + + rx_sc = create_rx_sc(dev, sci, active); if (IS_ERR(rx_sc)) { rtnl_unlock(); return PTR_ERR(rx_sc); } - if (tb_rxsc[MACSEC_RXSC_ATTR_ACTIVE]) - rx_sc->active = !!nla_get_u8(tb_rxsc[MACSEC_RXSC_ATTR_ACTIVE]); - if (macsec_is_offloaded(netdev_priv(dev))) { const struct macsec_ops *ops; struct macsec_context ctx; From 80df4706357a5a06bbbc70273bf2611df1ceee04 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Wed, 2 Nov 2022 22:33:15 +0100 Subject: [PATCH 043/138] macsec: fix detection of RXSCs when toggling offloading macsec_is_configured incorrectly uses secy->n_rx_sc to check if some RXSCs exist. secy->n_rx_sc only counts the number of active RXSCs, but there can also be inactive SCs as well, which may be stored in the driver (in case we're disabling offloading), or would have to be pushed to the device (in case we're trying to enable offloading). As long as RXSCs active on creation and never turned off, the issue is not visible. Fixes: dcb780fb2795 ("net: macsec: add nla support for changing the offloading selection") Signed-off-by: Sabrina Dubroca Reviewed-by: Antoine Tenart Reviewed-by: Leon Romanovsky Signed-off-by: David S. Miller --- drivers/net/macsec.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index 1b4d856f4bd7..700a8f96c6c2 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -2571,7 +2571,7 @@ static bool macsec_is_configured(struct macsec_dev *macsec) struct macsec_tx_sc *tx_sc = &secy->tx_sc; int i; - if (secy->n_rx_sc > 0) + if (secy->rx_sc) return true; for (i = 0; i < MACSEC_NUM_AN; i++) From aaab73f8fba4fd38f4d2617440d541a1c334e819 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Wed, 2 Nov 2022 22:33:16 +0100 Subject: [PATCH 044/138] macsec: clear encryption keys from the stack after setting up offload macsec_add_rxsa and macsec_add_txsa copy the key to an on-stack offloading context to pass it to the drivers, but leaves it there when it's done. Clear it with memzero_explicit as soon as it's not needed anymore. Fixes: 3cf3227a21d1 ("net: macsec: hardware offloading infrastructure") Signed-off-by: Sabrina Dubroca Reviewed-by: Antoine Tenart Reviewed-by: Leon Romanovsky Signed-off-by: David S. Miller --- drivers/net/macsec.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index 700a8f96c6c2..85376d2f24ca 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -1839,6 +1839,7 @@ static int macsec_add_rxsa(struct sk_buff *skb, struct genl_info *info) secy->key_len); err = macsec_offload(ops->mdo_add_rxsa, &ctx); + memzero_explicit(ctx.sa.key, secy->key_len); if (err) goto cleanup; } @@ -2081,6 +2082,7 @@ static int macsec_add_txsa(struct sk_buff *skb, struct genl_info *info) secy->key_len); err = macsec_offload(ops->mdo_add_txsa, &ctx); + memzero_explicit(ctx.sa.key, secy->key_len); if (err) goto cleanup; } From 1a0c016a4831ea29be09bbc8162d4a2a0690b4b8 Mon Sep 17 00:00:00 2001 From: Roger Quadros Date: Wed, 2 Nov 2022 12:31:44 +0200 Subject: [PATCH 045/138] net: ethernet: ti: am65-cpsw: Fix segmentation fault at module unload Move am65_cpsw_nuss_phylink_cleanup() call to after am65_cpsw_nuss_cleanup_ndev() so phylink is still valid to prevent the below Segmentation fault on module remove when first slave link is up. [ 31.652944] Unable to handle kernel paging request at virtual address 00040008000005f4 [ 31.684627] Mem abort info: [ 31.687446] ESR = 0x0000000096000004 [ 31.704614] EC = 0x25: DABT (current EL), IL = 32 bits [ 31.720663] SET = 0, FnV = 0 [ 31.723729] EA = 0, S1PTW = 0 [ 31.740617] FSC = 0x04: level 0 translation fault [ 31.756624] Data abort info: [ 31.759508] ISV = 0, ISS = 0x00000004 [ 31.776705] CM = 0, WnR = 0 [ 31.779695] [00040008000005f4] address between user and kernel address ranges [ 31.808644] Internal error: Oops: 0000000096000004 [#1] PREEMPT SMP [ 31.814928] Modules linked in: wlcore_sdio wl18xx wlcore mac80211 libarc4 cfg80211 rfkill crct10dif_ce phy_gmii_sel ti_am65_cpsw_nuss(-) sch_fq_codel ipv6 [ 31.828776] CPU: 0 PID: 1026 Comm: modprobe Not tainted 6.1.0-rc2-00012-gfabfcf7dafdb-dirty #160 [ 31.837547] Hardware name: Texas Instruments AM625 (DT) [ 31.842760] pstate: 40000005 (nZcv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 31.849709] pc : phy_stop+0x18/0xf8 [ 31.853202] lr : phylink_stop+0x38/0xf8 [ 31.857031] sp : ffff80000a0839f0 [ 31.860335] x29: ffff80000a0839f0 x28: ffff000000de1c80 x27: 0000000000000000 [ 31.867462] x26: 0000000000000000 x25: 0000000000000000 x24: ffff80000a083b98 [ 31.874589] x23: 0000000000000800 x22: 0000000000000001 x21: ffff000001bfba90 [ 31.881715] x20: ffff0000015ee000 x19: 0004000800000200 x18: 0000000000000000 [ 31.888842] x17: ffff800076c45000 x16: ffff800008004000 x15: 000058e39660b106 [ 31.895969] x14: 0000000000000144 x13: 0000000000000144 x12: 0000000000000000 [ 31.903095] x11: 000000000000275f x10: 00000000000009e0 x9 : ffff80000a0837d0 [ 31.910222] x8 : ffff000000de26c0 x7 : ffff00007fbd6540 x6 : ffff00007fbd64c0 [ 31.917349] x5 : ffff00007fbd0b10 x4 : ffff00007fbd0b10 x3 : ffff00007fbd3920 [ 31.924476] x2 : d0a07fcff8b8d500 x1 : 0000000000000000 x0 : 0004000800000200 [ 31.931603] Call trace: [ 31.934042] phy_stop+0x18/0xf8 [ 31.937177] phylink_stop+0x38/0xf8 [ 31.940657] am65_cpsw_nuss_ndo_slave_stop+0x28/0x1e0 [ti_am65_cpsw_nuss] [ 31.947452] __dev_close_many+0xa4/0x140 [ 31.951371] dev_close_many+0x84/0x128 [ 31.955115] unregister_netdevice_many+0x130/0x6d0 [ 31.959897] unregister_netdevice_queue+0x94/0xd8 [ 31.964591] unregister_netdev+0x24/0x38 [ 31.968504] am65_cpsw_nuss_cleanup_ndev.isra.0+0x48/0x70 [ti_am65_cpsw_nuss] [ 31.975637] am65_cpsw_nuss_remove+0x58/0xf8 [ti_am65_cpsw_nuss] Cc: # v5.18+ Fixes: e8609e69470f ("net: ethernet: ti: am65-cpsw: Convert to PHYLINK") Signed-off-by: Roger Quadros Signed-off-by: David S. Miller --- drivers/net/ethernet/ti/am65-cpsw-nuss.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c index 7f86068f3ff6..c50b137f92d7 100644 --- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c +++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c @@ -2823,7 +2823,6 @@ static int am65_cpsw_nuss_remove(struct platform_device *pdev) if (ret < 0) return ret; - am65_cpsw_nuss_phylink_cleanup(common); am65_cpsw_unregister_devlink(common); am65_cpsw_unregister_notifiers(common); @@ -2831,6 +2830,7 @@ static int am65_cpsw_nuss_remove(struct platform_device *pdev) * dma_deconfigure(dev) before devres_release_all(dev) */ am65_cpsw_nuss_cleanup_ndev(common); + am65_cpsw_nuss_phylink_cleanup(common); of_platform_device_destroy(common->mdio_dev, NULL); From 51afe9026d0c63263abe9840e629f118d7405b36 Mon Sep 17 00:00:00 2001 From: Ratheesh Kannoth Date: Wed, 2 Nov 2022 08:41:13 +0530 Subject: [PATCH 046/138] octeontx2-pf: NIX TX overwrites SQ_CTX_HW_S[SQ_INT] In scenarios where multiple errors have occurred for a SQ before SW starts handling error interrupt, SQ_CTX[OP_INT] may get overwritten leading to NIX_LF_SQ_OP_INT returning incorrect value. To workaround this read LMT, MNQ and SQ individual error status registers to determine the cause of error. Fixes: 4ff7d1488a84 ("octeontx2-pf: Error handling support") Signed-off-by: Ratheesh Kannoth Reviewed-by: Sunil Kovvuri Goutham Signed-off-by: David S. Miller --- .../ethernet/marvell/octeontx2/nic/otx2_pf.c | 135 ++++++++++++++---- .../marvell/octeontx2/nic/otx2_struct.h | 57 ++++++++ 2 files changed, 162 insertions(+), 30 deletions(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c index 892ca88e0cf4..303930499a4c 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c @@ -15,6 +15,7 @@ #include #include #include +#include #include "otx2_reg.h" #include "otx2_common.h" @@ -1171,6 +1172,59 @@ int otx2_set_real_num_queues(struct net_device *netdev, } EXPORT_SYMBOL(otx2_set_real_num_queues); +static char *nix_sqoperr_e_str[NIX_SQOPERR_MAX] = { + "NIX_SQOPERR_OOR", + "NIX_SQOPERR_CTX_FAULT", + "NIX_SQOPERR_CTX_POISON", + "NIX_SQOPERR_DISABLED", + "NIX_SQOPERR_SIZE_ERR", + "NIX_SQOPERR_OFLOW", + "NIX_SQOPERR_SQB_NULL", + "NIX_SQOPERR_SQB_FAULT", + "NIX_SQOPERR_SQE_SZ_ZERO", +}; + +static char *nix_mnqerr_e_str[NIX_MNQERR_MAX] = { + "NIX_MNQERR_SQ_CTX_FAULT", + "NIX_MNQERR_SQ_CTX_POISON", + "NIX_MNQERR_SQB_FAULT", + "NIX_MNQERR_SQB_POISON", + "NIX_MNQERR_TOTAL_ERR", + "NIX_MNQERR_LSO_ERR", + "NIX_MNQERR_CQ_QUERY_ERR", + "NIX_MNQERR_MAX_SQE_SIZE_ERR", + "NIX_MNQERR_MAXLEN_ERR", + "NIX_MNQERR_SQE_SIZEM1_ZERO", +}; + +static char *nix_snd_status_e_str[NIX_SND_STATUS_MAX] = { + "NIX_SND_STATUS_GOOD", + "NIX_SND_STATUS_SQ_CTX_FAULT", + "NIX_SND_STATUS_SQ_CTX_POISON", + "NIX_SND_STATUS_SQB_FAULT", + "NIX_SND_STATUS_SQB_POISON", + "NIX_SND_STATUS_HDR_ERR", + "NIX_SND_STATUS_EXT_ERR", + "NIX_SND_STATUS_JUMP_FAULT", + "NIX_SND_STATUS_JUMP_POISON", + "NIX_SND_STATUS_CRC_ERR", + "NIX_SND_STATUS_IMM_ERR", + "NIX_SND_STATUS_SG_ERR", + "NIX_SND_STATUS_MEM_ERR", + "NIX_SND_STATUS_INVALID_SUBDC", + "NIX_SND_STATUS_SUBDC_ORDER_ERR", + "NIX_SND_STATUS_DATA_FAULT", + "NIX_SND_STATUS_DATA_POISON", + "NIX_SND_STATUS_NPC_DROP_ACTION", + "NIX_SND_STATUS_LOCK_VIOL", + "NIX_SND_STATUS_NPC_UCAST_CHAN_ERR", + "NIX_SND_STATUS_NPC_MCAST_CHAN_ERR", + "NIX_SND_STATUS_NPC_MCAST_ABORT", + "NIX_SND_STATUS_NPC_VTAG_PTR_ERR", + "NIX_SND_STATUS_NPC_VTAG_SIZE_ERR", + "NIX_SND_STATUS_SEND_STATS_ERR", +}; + static irqreturn_t otx2_q_intr_handler(int irq, void *data) { struct otx2_nic *pf = data; @@ -1204,46 +1258,67 @@ static irqreturn_t otx2_q_intr_handler(int irq, void *data) /* SQ */ for (qidx = 0; qidx < pf->hw.tot_tx_queues; qidx++) { + u64 sq_op_err_dbg, mnq_err_dbg, snd_err_dbg; + u8 sq_op_err_code, mnq_err_code, snd_err_code; + + /* Below debug registers captures first errors corresponding to + * those registers. We don't have to check against SQ qid as + * these are fatal errors. + */ + ptr = otx2_get_regaddr(pf, NIX_LF_SQ_OP_INT); val = otx2_atomic64_add((qidx << 44), ptr); otx2_write64(pf, NIX_LF_SQ_OP_INT, (qidx << 44) | (val & NIX_SQINT_BITS)); - if (!(val & (NIX_SQINT_BITS | BIT_ULL(42)))) - continue; - if (val & BIT_ULL(42)) { netdev_err(pf->netdev, "SQ%lld: error reading NIX_LF_SQ_OP_INT, NIX_LF_ERR_INT 0x%llx\n", qidx, otx2_read64(pf, NIX_LF_ERR_INT)); - } else { - if (val & BIT_ULL(NIX_SQINT_LMT_ERR)) { - netdev_err(pf->netdev, "SQ%lld: LMT store error NIX_LF_SQ_OP_ERR_DBG:0x%llx", - qidx, - otx2_read64(pf, - NIX_LF_SQ_OP_ERR_DBG)); - otx2_write64(pf, NIX_LF_SQ_OP_ERR_DBG, - BIT_ULL(44)); - } - if (val & BIT_ULL(NIX_SQINT_MNQ_ERR)) { - netdev_err(pf->netdev, "SQ%lld: Meta-descriptor enqueue error NIX_LF_MNQ_ERR_DGB:0x%llx\n", - qidx, - otx2_read64(pf, NIX_LF_MNQ_ERR_DBG)); - otx2_write64(pf, NIX_LF_MNQ_ERR_DBG, - BIT_ULL(44)); - } - if (val & BIT_ULL(NIX_SQINT_SEND_ERR)) { - netdev_err(pf->netdev, "SQ%lld: Send error, NIX_LF_SEND_ERR_DBG 0x%llx", - qidx, - otx2_read64(pf, - NIX_LF_SEND_ERR_DBG)); - otx2_write64(pf, NIX_LF_SEND_ERR_DBG, - BIT_ULL(44)); - } - if (val & BIT_ULL(NIX_SQINT_SQB_ALLOC_FAIL)) - netdev_err(pf->netdev, "SQ%lld: SQB allocation failed", - qidx); + goto done; } + sq_op_err_dbg = otx2_read64(pf, NIX_LF_SQ_OP_ERR_DBG); + if (!(sq_op_err_dbg & BIT(44))) + goto chk_mnq_err_dbg; + + sq_op_err_code = FIELD_GET(GENMASK(7, 0), sq_op_err_dbg); + netdev_err(pf->netdev, "SQ%lld: NIX_LF_SQ_OP_ERR_DBG(%llx) err=%s\n", + qidx, sq_op_err_dbg, nix_sqoperr_e_str[sq_op_err_code]); + + otx2_write64(pf, NIX_LF_SQ_OP_ERR_DBG, BIT_ULL(44)); + + if (sq_op_err_code == NIX_SQOPERR_SQB_NULL) + goto chk_mnq_err_dbg; + + /* Err is not NIX_SQOPERR_SQB_NULL, call aq function to read SQ structure. + * TODO: But we are in irq context. How to call mbox functions which does sleep + */ + +chk_mnq_err_dbg: + mnq_err_dbg = otx2_read64(pf, NIX_LF_MNQ_ERR_DBG); + if (!(mnq_err_dbg & BIT(44))) + goto chk_snd_err_dbg; + + mnq_err_code = FIELD_GET(GENMASK(7, 0), mnq_err_dbg); + netdev_err(pf->netdev, "SQ%lld: NIX_LF_MNQ_ERR_DBG(%llx) err=%s\n", + qidx, mnq_err_dbg, nix_mnqerr_e_str[mnq_err_code]); + otx2_write64(pf, NIX_LF_MNQ_ERR_DBG, BIT_ULL(44)); + +chk_snd_err_dbg: + snd_err_dbg = otx2_read64(pf, NIX_LF_SEND_ERR_DBG); + if (snd_err_dbg & BIT(44)) { + snd_err_code = FIELD_GET(GENMASK(7, 0), snd_err_dbg); + netdev_err(pf->netdev, "SQ%lld: NIX_LF_SND_ERR_DBG:0x%llx err=%s\n", + qidx, snd_err_dbg, nix_snd_status_e_str[snd_err_code]); + otx2_write64(pf, NIX_LF_SEND_ERR_DBG, BIT_ULL(44)); + } + +done: + /* Print values and reset */ + if (val & BIT_ULL(NIX_SQINT_SQB_ALLOC_FAIL)) + netdev_err(pf->netdev, "SQ%lld: SQB allocation failed", + qidx); + schedule_work(&pf->reset_task); } diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h index aa205a0d158f..fa37b9f312ca 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_struct.h @@ -281,4 +281,61 @@ enum nix_sqint_e { BIT_ULL(NIX_SQINT_SEND_ERR) | \ BIT_ULL(NIX_SQINT_SQB_ALLOC_FAIL)) +enum nix_sqoperr_e { + NIX_SQOPERR_OOR = 0, + NIX_SQOPERR_CTX_FAULT = 1, + NIX_SQOPERR_CTX_POISON = 2, + NIX_SQOPERR_DISABLED = 3, + NIX_SQOPERR_SIZE_ERR = 4, + NIX_SQOPERR_OFLOW = 5, + NIX_SQOPERR_SQB_NULL = 6, + NIX_SQOPERR_SQB_FAULT = 7, + NIX_SQOPERR_SQE_SZ_ZERO = 8, + NIX_SQOPERR_MAX, +}; + +enum nix_mnqerr_e { + NIX_MNQERR_SQ_CTX_FAULT = 0, + NIX_MNQERR_SQ_CTX_POISON = 1, + NIX_MNQERR_SQB_FAULT = 2, + NIX_MNQERR_SQB_POISON = 3, + NIX_MNQERR_TOTAL_ERR = 4, + NIX_MNQERR_LSO_ERR = 5, + NIX_MNQERR_CQ_QUERY_ERR = 6, + NIX_MNQERR_MAX_SQE_SIZE_ERR = 7, + NIX_MNQERR_MAXLEN_ERR = 8, + NIX_MNQERR_SQE_SIZEM1_ZERO = 9, + NIX_MNQERR_MAX, +}; + +enum nix_snd_status_e { + NIX_SND_STATUS_GOOD = 0x0, + NIX_SND_STATUS_SQ_CTX_FAULT = 0x1, + NIX_SND_STATUS_SQ_CTX_POISON = 0x2, + NIX_SND_STATUS_SQB_FAULT = 0x3, + NIX_SND_STATUS_SQB_POISON = 0x4, + NIX_SND_STATUS_HDR_ERR = 0x5, + NIX_SND_STATUS_EXT_ERR = 0x6, + NIX_SND_STATUS_JUMP_FAULT = 0x7, + NIX_SND_STATUS_JUMP_POISON = 0x8, + NIX_SND_STATUS_CRC_ERR = 0x9, + NIX_SND_STATUS_IMM_ERR = 0x10, + NIX_SND_STATUS_SG_ERR = 0x11, + NIX_SND_STATUS_MEM_ERR = 0x12, + NIX_SND_STATUS_INVALID_SUBDC = 0x13, + NIX_SND_STATUS_SUBDC_ORDER_ERR = 0x14, + NIX_SND_STATUS_DATA_FAULT = 0x15, + NIX_SND_STATUS_DATA_POISON = 0x16, + NIX_SND_STATUS_NPC_DROP_ACTION = 0x17, + NIX_SND_STATUS_LOCK_VIOL = 0x18, + NIX_SND_STATUS_NPC_UCAST_CHAN_ERR = 0x19, + NIX_SND_STATUS_NPC_MCAST_CHAN_ERR = 0x20, + NIX_SND_STATUS_NPC_MCAST_ABORT = 0x21, + NIX_SND_STATUS_NPC_VTAG_PTR_ERR = 0x22, + NIX_SND_STATUS_NPC_VTAG_SIZE_ERR = 0x23, + NIX_SND_STATUS_SEND_MEM_FAULT = 0x24, + NIX_SND_STATUS_SEND_STATS_ERR = 0x25, + NIX_SND_STATUS_MAX, +}; + #endif /* OTX2_STRUCT_H */ From 1118b2049d77ca0b505775fc1a8d1909cf19a7ec Mon Sep 17 00:00:00 2001 From: Wang Yufen Date: Wed, 2 Nov 2022 17:41:19 +0800 Subject: [PATCH 047/138] net: tun: Fix memory leaks of napi_get_frags kmemleak reports after running test_progs: unreferenced object 0xffff8881b1672dc0 (size 232): comm "test_progs", pid 394388, jiffies 4354712116 (age 841.975s) hex dump (first 32 bytes): e0 84 d7 a8 81 88 ff ff 80 2c 67 b1 81 88 ff ff .........,g..... 00 40 c5 9b 81 88 ff ff 00 00 00 00 00 00 00 00 .@.............. backtrace: [<00000000c8f01748>] napi_skb_cache_get+0xd4/0x150 [<0000000041c7fc09>] __napi_build_skb+0x15/0x50 [<00000000431c7079>] __napi_alloc_skb+0x26e/0x540 [<000000003ecfa30e>] napi_get_frags+0x59/0x140 [<0000000099b2199e>] tun_get_user+0x183d/0x3bb0 [tun] [<000000008a5adef0>] tun_chr_write_iter+0xc0/0x1b1 [tun] [<0000000049993ff4>] do_iter_readv_writev+0x19f/0x320 [<000000008f338ea2>] do_iter_write+0x135/0x630 [<000000008a3377a4>] vfs_writev+0x12e/0x440 [<00000000a6b5639a>] do_writev+0x104/0x280 [<00000000ccf065d8>] do_syscall_64+0x3b/0x90 [<00000000d776e329>] entry_SYSCALL_64_after_hwframe+0x63/0xcd The issue occurs in the following scenarios: tun_get_user() napi_gro_frags() napi_frags_finish() case GRO_NORMAL: gro_normal_one() list_add_tail(&skb->list, &napi->rx_list); <-- While napi->rx_count < READ_ONCE(gro_normal_batch), <-- gro_normal_list() is not called, napi->rx_list is not empty <-- not ask to complete the gro work, will cause memory leaks in <-- following tun_napi_del() ... tun_napi_del() netif_napi_del() __netif_napi_del() <-- &napi->rx_list is not empty, which caused memory leaks To fix, add napi_complete() after napi_gro_frags(). Fixes: 90e33d459407 ("tun: enable napi_gro_frags() for TUN/TAP driver") Signed-off-by: Wang Yufen Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/tun.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 946628050f28..eb12f3136a54 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -1977,6 +1977,7 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile, local_bh_disable(); napi_gro_frags(&tfile->napi); + napi_complete(&tfile->napi); local_bh_enable(); mutex_unlock(&tfile->napi_mutex); } else if (tfile->napi_enabled) { From eb4940d4adf590590a9d0c47e38d2799c2ff9670 Mon Sep 17 00:00:00 2001 From: Vlastimil Babka Date: Fri, 4 Nov 2022 13:57:11 +0100 Subject: [PATCH 048/138] mm/slab: remove !CONFIG_TRACING variants of kmalloc_[node_]trace() For !CONFIG_TRACING kernels, the kmalloc() implementation tries (in cases where the allocation size is build-time constant) to save a function call, by inlining kmalloc_trace() to a kmem_cache_alloc() call. However since commit 6edf2576a6cc ("mm/slub: enable debugging memory wasting of kmalloc") this path now fails to pass the original request size to be eventually recorded (for kmalloc caches with debugging enabled). We could adjust the code to call __kmem_cache_alloc_node() as the CONFIG_TRACING variant, but that would as a result inline a call with 5 parameters, bloating the kmalloc() call sites. The cost of extra function call (to kmalloc_trace()) seems like a lesser evil. It also appears that the !CONFIG_TRACING variant is incompatible with upcoming hardening efforts [1] so it's easier if we just remove it now. Kernels with no tracing are rare these days and the benefit is dubious anyway. [1] https://lore.kernel.org/linux-mm/20221101222520.never.109-kees@kernel.org/T/#m20ecf14390e406247bde0ea9cce368f469c539ed Link: https://lore.kernel.org/all/097d8fba-bd10-a312-24a3-a4068c4f424c@suse.cz/ Suggested-by: Hyeonggon Yoo <42.hyeyoo@gmail.com> Signed-off-by: Vlastimil Babka --- include/linux/slab.h | 23 ----------------------- mm/slab_common.c | 2 -- 2 files changed, 25 deletions(-) diff --git a/include/linux/slab.h b/include/linux/slab.h index 90877fcde70b..45efc6c553b8 100644 --- a/include/linux/slab.h +++ b/include/linux/slab.h @@ -470,35 +470,12 @@ void *__kmalloc_node(size_t size, gfp_t flags, int node) __assume_kmalloc_alignm void *kmem_cache_alloc_node(struct kmem_cache *s, gfp_t flags, int node) __assume_slab_alignment __malloc; -#ifdef CONFIG_TRACING void *kmalloc_trace(struct kmem_cache *s, gfp_t flags, size_t size) __assume_kmalloc_alignment __alloc_size(3); void *kmalloc_node_trace(struct kmem_cache *s, gfp_t gfpflags, int node, size_t size) __assume_kmalloc_alignment __alloc_size(4); -#else /* CONFIG_TRACING */ -/* Save a function call when CONFIG_TRACING=n */ -static __always_inline __alloc_size(3) -void *kmalloc_trace(struct kmem_cache *s, gfp_t flags, size_t size) -{ - void *ret = kmem_cache_alloc(s, flags); - - ret = kasan_kmalloc(s, ret, size, flags); - return ret; -} - -static __always_inline __alloc_size(4) -void *kmalloc_node_trace(struct kmem_cache *s, gfp_t gfpflags, - int node, size_t size) -{ - void *ret = kmem_cache_alloc_node(s, gfpflags, node); - - ret = kasan_kmalloc(s, ret, size, gfpflags); - return ret; -} -#endif /* CONFIG_TRACING */ - void *kmalloc_large(size_t size, gfp_t flags) __assume_page_alignment __alloc_size(1); diff --git a/mm/slab_common.c b/mm/slab_common.c index 74a991fd9d31..206e59051c1d 100644 --- a/mm/slab_common.c +++ b/mm/slab_common.c @@ -1040,7 +1040,6 @@ size_t __ksize(const void *object) return slab_ksize(folio_slab(folio)->slab_cache); } -#ifdef CONFIG_TRACING void *kmalloc_trace(struct kmem_cache *s, gfp_t gfpflags, size_t size) { void *ret = __kmem_cache_alloc_node(s, gfpflags, NUMA_NO_NODE, @@ -1064,7 +1063,6 @@ void *kmalloc_node_trace(struct kmem_cache *s, gfp_t gfpflags, return ret; } EXPORT_SYMBOL(kmalloc_node_trace); -#endif /* !CONFIG_TRACING */ #endif /* !CONFIG_SLOB */ gfp_t kmalloc_fix_flags(gfp_t flags) From 6e59419fd0a244dd55e53e798797f0697dec8b1c Mon Sep 17 00:00:00 2001 From: Peng Wu Date: Sun, 11 Sep 2022 06:00:53 +0000 Subject: [PATCH 049/138] phy: sunplus: Fix an IS_ERR() vs NULL bug in sp_usb_phy_probe The devm_ioremap() function returns NULL on error, it doesn't return error pointers. Fixes: 99d9ccd973852 ("phy: usb: Add USB2.0 phy driver for Sunplus SP7021") Signed-off-by: Peng Wu Link: https://lore.kernel.org/r/20220911060053.123594-1-wupeng58@huawei.com Signed-off-by: Vinod Koul --- drivers/phy/sunplus/phy-sunplus-usb2.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/phy/sunplus/phy-sunplus-usb2.c b/drivers/phy/sunplus/phy-sunplus-usb2.c index b932087c55b2..e827b79f6d49 100644 --- a/drivers/phy/sunplus/phy-sunplus-usb2.c +++ b/drivers/phy/sunplus/phy-sunplus-usb2.c @@ -256,8 +256,8 @@ static int sp_usb_phy_probe(struct platform_device *pdev) usbphy->moon4_res_mem = platform_get_resource_byname(pdev, IORESOURCE_MEM, "moon4"); usbphy->moon4_regs = devm_ioremap(&pdev->dev, usbphy->moon4_res_mem->start, resource_size(usbphy->moon4_res_mem)); - if (IS_ERR(usbphy->moon4_regs)) - return PTR_ERR(usbphy->moon4_regs); + if (!usbphy->moon4_regs) + return -ENOMEM; usbphy->phy_clk = devm_clk_get(&pdev->dev, NULL); if (IS_ERR(usbphy->phy_clk)) From b4c66425771ddb910316c7b4cd7fa0614098ec45 Mon Sep 17 00:00:00 2001 From: Vikas Gupta Date: Thu, 3 Nov 2022 19:33:24 -0400 Subject: [PATCH 050/138] bnxt_en: refactor bnxt_cancel_reservations() Introduce bnxt_clear_reservations() to clear the reserved attributes only. This will be used in the next patch to fix PCI AER handling. Signed-off-by: Vikas Gupta Signed-off-by: Michael Chan Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 23 ++++++++++++++++------- 1 file changed, 16 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 04cf7684f1b0..3743d9755eb4 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -9983,17 +9983,12 @@ static int bnxt_try_recover_fw(struct bnxt *bp) return -ENODEV; } -int bnxt_cancel_reservations(struct bnxt *bp, bool fw_reset) +static void bnxt_clear_reservations(struct bnxt *bp, bool fw_reset) { struct bnxt_hw_resc *hw_resc = &bp->hw_resc; - int rc; if (!BNXT_NEW_RM(bp)) - return 0; /* no resource reservations required */ - - rc = bnxt_hwrm_func_resc_qcaps(bp, true); - if (rc) - netdev_err(bp->dev, "resc_qcaps failed\n"); + return; /* no resource reservations required */ hw_resc->resv_cp_rings = 0; hw_resc->resv_stat_ctxs = 0; @@ -10006,6 +10001,20 @@ int bnxt_cancel_reservations(struct bnxt *bp, bool fw_reset) bp->tx_nr_rings = 0; bp->rx_nr_rings = 0; } +} + +int bnxt_cancel_reservations(struct bnxt *bp, bool fw_reset) +{ + int rc; + + if (!BNXT_NEW_RM(bp)) + return 0; /* no resource reservations required */ + + rc = bnxt_hwrm_func_resc_qcaps(bp, true); + if (rc) + netdev_err(bp->dev, "resc_qcaps failed\n"); + + bnxt_clear_reservations(bp, fw_reset); return rc; } From 0cf736a18a1e804037839bd8df9e36f0efdb8745 Mon Sep 17 00:00:00 2001 From: Vikas Gupta Date: Thu, 3 Nov 2022 19:33:25 -0400 Subject: [PATCH 051/138] bnxt_en: fix the handling of PCIE-AER Fix the sequence required for PCIE-AER. While slot reset occurs, firmware might not be ready and the driver needs to check for its recovery. We also need to remap the health registers for some chips and clear the resource reservations. The resources will be allocated again during bnxt_io_resume(). Fixes: fb1e6e562b37 ("bnxt_en: Fix AER recovery.") Signed-off-by: Vikas Gupta Signed-off-by: Michael Chan Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 29 ++++++++++++++++++- drivers/net/ethernet/broadcom/bnxt/bnxt.h | 1 + .../net/ethernet/broadcom/bnxt/bnxt_hwrm.c | 3 +- 3 files changed, 31 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 3743d9755eb4..f44f9367b7fd 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -13922,7 +13922,9 @@ static pci_ers_result_t bnxt_io_slot_reset(struct pci_dev *pdev) pci_ers_result_t result = PCI_ERS_RESULT_DISCONNECT; struct net_device *netdev = pci_get_drvdata(pdev); struct bnxt *bp = netdev_priv(netdev); - int err = 0, off; + int retry = 0; + int err = 0; + int off; netdev_info(bp->dev, "PCI Slot Reset\n"); @@ -13950,11 +13952,36 @@ static pci_ers_result_t bnxt_io_slot_reset(struct pci_dev *pdev) pci_restore_state(pdev); pci_save_state(pdev); + bnxt_inv_fw_health_reg(bp); + bnxt_try_map_fw_health_reg(bp); + + /* In some PCIe AER scenarios, firmware may take up to + * 10 seconds to become ready in the worst case. + */ + do { + err = bnxt_try_recover_fw(bp); + if (!err) + break; + retry++; + } while (retry < BNXT_FW_SLOT_RESET_RETRY); + + if (err) { + dev_err(&pdev->dev, "Firmware not ready\n"); + goto reset_exit; + } + err = bnxt_hwrm_func_reset(bp); if (!err) result = PCI_ERS_RESULT_RECOVERED; + + bnxt_ulp_irq_stop(bp); + bnxt_clear_int_mode(bp); + err = bnxt_init_int_mode(bp); + bnxt_ulp_irq_restart(bp, err); } +reset_exit: + bnxt_clear_reservations(bp, true); rtnl_unlock(); return result; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index b1b17f911300..d5fa43cfe524 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -1621,6 +1621,7 @@ struct bnxt_fw_health { #define BNXT_FW_RETRY 5 #define BNXT_FW_IF_RETRY 10 +#define BNXT_FW_SLOT_RESET_RETRY 4 enum board_idx { BCM57301, diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_hwrm.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_hwrm.c index b01d42928a53..132442f16fe6 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_hwrm.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_hwrm.c @@ -476,7 +476,8 @@ static int __hwrm_send(struct bnxt *bp, struct bnxt_hwrm_ctx *ctx) memset(ctx->resp, 0, PAGE_SIZE); req_type = le16_to_cpu(ctx->req->req_type); - if (BNXT_NO_FW_ACCESS(bp) && req_type != HWRM_FUNC_RESET) { + if (BNXT_NO_FW_ACCESS(bp) && + (req_type != HWRM_FUNC_RESET && req_type != HWRM_VER_GET)) { netdev_dbg(bp->dev, "hwrm req_type 0x%x skipped, FW channel down\n", req_type); goto exit; From 6d81ea3765dfa6c8a20822613c81edad1c4a16a0 Mon Sep 17 00:00:00 2001 From: Michael Chan Date: Thu, 3 Nov 2022 19:33:26 -0400 Subject: [PATCH 052/138] bnxt_en: Fix possible crash in bnxt_hwrm_set_coal() During the error recovery sequence, the rtnl_lock is not held for the entire duration and some datastructures may be freed during the sequence. Check for the BNXT_STATE_OPEN flag instead of netif_running() to ensure that the device is fully operational before proceeding to reconfigure the coalescing settings. This will fix a possible crash like this: BUG: unable to handle kernel NULL pointer dereference at 0000000000000000 PGD 0 P4D 0 Oops: 0000 [#1] SMP NOPTI CPU: 10 PID: 181276 Comm: ethtool Kdump: loaded Tainted: G IOE --------- - - 4.18.0-348.el8.x86_64 #1 Hardware name: Dell Inc. PowerEdge R740/0F9N89, BIOS 2.3.10 08/15/2019 RIP: 0010:bnxt_hwrm_set_coal+0x1fb/0x2a0 [bnxt_en] Code: c2 66 83 4e 22 08 66 89 46 1c e8 10 cb 00 00 41 83 c6 01 44 39 b3 68 01 00 00 0f 8e a3 00 00 00 48 8b 93 c8 00 00 00 49 63 c6 <48> 8b 2c c2 48 8b 85 b8 02 00 00 48 85 c0 74 2e 48 8b 74 24 08 f6 RSP: 0018:ffffb11c8dcaba50 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff8d168a8b0ac0 RCX: 00000000000000c5 RDX: 0000000000000000 RSI: ffff8d162f72c000 RDI: ffff8d168a8b0b28 RBP: 0000000000000000 R08: b6e1f68a12e9a7eb R09: 0000000000000000 R10: 0000000000000001 R11: 0000000000000037 R12: ffff8d168a8b109c R13: ffff8d168a8b10aa R14: 0000000000000000 R15: ffffffffc01ac4e0 FS: 00007f3852e4c740(0000) GS:ffff8d24c0080000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000000 CR3: 000000041b3ee003 CR4: 00000000007706e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 PKRU: 55555554 Call Trace: ethnl_set_coalesce+0x3ce/0x4c0 genl_family_rcv_msg_doit.isra.15+0x10f/0x150 genl_family_rcv_msg+0xb3/0x160 ? coalesce_fill_reply+0x480/0x480 genl_rcv_msg+0x47/0x90 ? genl_family_rcv_msg+0x160/0x160 netlink_rcv_skb+0x4c/0x120 genl_rcv+0x24/0x40 netlink_unicast+0x196/0x230 netlink_sendmsg+0x204/0x3d0 sock_sendmsg+0x4c/0x50 __sys_sendto+0xee/0x160 ? syscall_trace_enter+0x1d3/0x2c0 ? __audit_syscall_exit+0x249/0x2a0 __x64_sys_sendto+0x24/0x30 do_syscall_64+0x5b/0x1a0 entry_SYSCALL_64_after_hwframe+0x65/0xca RIP: 0033:0x7f38524163bb Fixes: 2151fe0830fd ("bnxt_en: Handle RESET_NOTIFY async event from firmware.") Reviewed-by: Somnath Kotur Signed-off-by: Michael Chan Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index f57e524c7e30..8cad15c458b3 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -162,7 +162,7 @@ static int bnxt_set_coalesce(struct net_device *dev, } reset_coalesce: - if (netif_running(dev)) { + if (test_bit(BNXT_STATE_OPEN, &bp->state)) { if (update_stats) { rc = bnxt_close_nic(bp, true, false); if (!rc) From 02597d39145bb0aa81d04bf39b6a913ce9a9d465 Mon Sep 17 00:00:00 2001 From: Alex Barba Date: Thu, 3 Nov 2022 19:33:27 -0400 Subject: [PATCH 053/138] bnxt_en: fix potentially incorrect return value for ndo_rx_flow_steer In the bnxt_en driver ndo_rx_flow_steer returns '0' whenever an entry that we are attempting to steer is already found. This is not the correct behavior. The return code should be the value/index that corresponds to the entry. Returning zero all the time causes the RFS records to be incorrect unless entry '0' is the correct one. As flows migrate to different cores this can create entries that are not correct. Fixes: c0c050c58d84 ("bnxt_en: New Broadcom ethernet driver.") Reported-by: Akshay Navgire Signed-off-by: Alex Barba Signed-off-by: Andy Gospodarek Signed-off-by: Michael Chan Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index f44f9367b7fd..c78b6e9dea2c 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -12903,8 +12903,8 @@ static int bnxt_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb, rcu_read_lock(); hlist_for_each_entry_rcu(fltr, head, hash) { if (bnxt_fltr_match(fltr, new_fltr)) { + rc = fltr->sw_id; rcu_read_unlock(); - rc = 0; goto err_free; } } From b7cbc6740bd6ad5d43345a2504f7e4beff0d709f Mon Sep 17 00:00:00 2001 From: Sean Anderson Date: Thu, 3 Nov 2022 14:28:30 -0400 Subject: [PATCH 054/138] net: fman: Unregister ethernet device on removal When the mac device gets removed, it leaves behind the ethernet device. This will result in a segfault next time the ethernet device accesses mac_dev. Remove the ethernet device when we get removed to prevent this. This is not completely reversible, since some resources aren't cleaned up properly, but that can be addressed later. Fixes: 3933961682a3 ("fsl/fman: Add FMan MAC driver") Signed-off-by: Sean Anderson Link: https://lore.kernel.org/r/20221103182831.2248833-1-sean.anderson@seco.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/fman/mac.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/net/ethernet/freescale/fman/mac.c b/drivers/net/ethernet/freescale/fman/mac.c index 65df308bad97..13e67f2864be 100644 --- a/drivers/net/ethernet/freescale/fman/mac.c +++ b/drivers/net/ethernet/freescale/fman/mac.c @@ -487,12 +487,21 @@ static int mac_probe(struct platform_device *_of_dev) return err; } +static int mac_remove(struct platform_device *pdev) +{ + struct mac_device *mac_dev = platform_get_drvdata(pdev); + + platform_device_unregister(mac_dev->priv->eth_dev); + return 0; +} + static struct platform_driver mac_driver = { .driver = { .name = KBUILD_MODNAME, .of_match_table = mac_match, }, .probe = mac_probe, + .remove = mac_remove, }; builtin_platform_driver(mac_driver); From 46653972e3ea64f79e7f8ae3aa41a4d3fdb70a13 Mon Sep 17 00:00:00 2001 From: Gaosheng Cui Date: Mon, 31 Oct 2022 19:25:36 +0800 Subject: [PATCH 055/138] capabilities: fix undefined behavior in bit shift for CAP_TO_MASK Shifting signed 32-bit value by 31 bits is undefined, so changing significant bit to unsigned. The UBSAN warning calltrace like below: UBSAN: shift-out-of-bounds in security/commoncap.c:1252:2 left shift of 1 by 31 places cannot be represented in type 'int' Call Trace: dump_stack_lvl+0x7d/0xa5 dump_stack+0x15/0x1b ubsan_epilogue+0xe/0x4e __ubsan_handle_shift_out_of_bounds+0x1e7/0x20c cap_task_prctl+0x561/0x6f0 security_task_prctl+0x5a/0xb0 __x64_sys_prctl+0x61/0x8f0 do_syscall_64+0x58/0x80 entry_SYSCALL_64_after_hwframe+0x63/0xcd Fixes: e338d263a76a ("Add 64-bit capability support to the kernel") Signed-off-by: Gaosheng Cui Acked-by: Andrew G. Morgan Reviewed-by: Serge Hallyn Signed-off-by: Paul Moore --- include/uapi/linux/capability.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/uapi/linux/capability.h b/include/uapi/linux/capability.h index 463d1ba2232a..3d61a0ae055d 100644 --- a/include/uapi/linux/capability.h +++ b/include/uapi/linux/capability.h @@ -426,7 +426,7 @@ struct vfs_ns_cap_data { */ #define CAP_TO_INDEX(x) ((x) >> 5) /* 1 << 5 == bits in __u32 */ -#define CAP_TO_MASK(x) (1 << ((x) & 31)) /* mask for indexed __u32 */ +#define CAP_TO_MASK(x) (1U << ((x) & 31)) /* mask for indexed __u32 */ #endif /* _UAPI_LINUX_CAPABILITY_H */ From 819b885cd886c193782891c4f51bbcab3de119a4 Mon Sep 17 00:00:00 2001 From: John Thomson Date: Sat, 5 Nov 2022 06:52:41 +1000 Subject: [PATCH 056/138] phy: ralink: mt7621-pci: add sentinel to quirks table With mt7621 soc_dev_attr fixed to register the soc as a device, kernel will experience an oops in soc_device_match_attr This quirk test was introduced in the staging driver in commit 9445ccb3714c ("staging: mt7621-pci-phy: add quirks for 'E2' revision using 'soc_device_attribute'"). The staging driver was removed, and later re-added in commit d87da32372a0 ("phy: ralink: Add PHY driver for MT7621 PCIe PHY") for kernel 5.11 Link: https://lore.kernel.org/lkml/26ebbed1-0fe9-4af9-8466-65f841d0b382@app.fastmail.com Fixes: d87da32372a0 ("phy: ralink: Add PHY driver for MT7621 PCIe PHY") Signed-off-by: John Thomson Acked-by: Sergio Paracuellos Link: https://lore.kernel.org/r/20221104205242.3440388-2-git@johnthomson.fastmail.com.au Signed-off-by: Vinod Koul --- drivers/phy/ralink/phy-mt7621-pci.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/phy/ralink/phy-mt7621-pci.c b/drivers/phy/ralink/phy-mt7621-pci.c index 5e6530f545b5..85888ab2d307 100644 --- a/drivers/phy/ralink/phy-mt7621-pci.c +++ b/drivers/phy/ralink/phy-mt7621-pci.c @@ -280,7 +280,8 @@ static struct phy *mt7621_pcie_phy_of_xlate(struct device *dev, } static const struct soc_device_attribute mt7621_pci_quirks_match[] = { - { .soc_id = "mt7621", .revision = "E2" } + { .soc_id = "mt7621", .revision = "E2" }, + { /* sentinel */ } }; static const struct regmap_config mt7621_pci_phy_regmap_config = { From 328687151b2a29fe93db4736bfb90da134e52fbf Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Fri, 4 Nov 2022 23:35:34 -0700 Subject: [PATCH 057/138] mm/slab_common: Restore passing "caller" for tracing The "caller" argument was accidentally being ignored in a few places that were recently refactored. Restore these "caller" arguments, instead of _RET_IP_. Fixes: 11e9734bcb6a ("mm/slab_common: unify NUMA and UMA version of tracepoints") Cc: Hyeonggon Yoo <42.hyeyoo@gmail.com> Cc: Vlastimil Babka Cc: Christoph Lameter Cc: Pekka Enberg Cc: David Rientjes Cc: Joonsoo Kim Cc: Andrew Morton Cc: Roman Gushchin Cc: linux-mm@kvack.org Signed-off-by: Kees Cook Acked-by: Hyeonggon Yoo <42.hyeyoo@gmail.com> Signed-off-by: Vlastimil Babka --- mm/slab_common.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/mm/slab_common.c b/mm/slab_common.c index 206e59051c1d..fa784563e1ed 100644 --- a/mm/slab_common.c +++ b/mm/slab_common.c @@ -941,7 +941,7 @@ void *__do_kmalloc_node(size_t size, gfp_t flags, int node, unsigned long caller if (unlikely(size > KMALLOC_MAX_CACHE_SIZE)) { ret = __kmalloc_large_node(size, flags, node); - trace_kmalloc(_RET_IP_, ret, size, + trace_kmalloc(caller, ret, size, PAGE_SIZE << get_order(size), flags, node); return ret; } @@ -953,7 +953,7 @@ void *__do_kmalloc_node(size_t size, gfp_t flags, int node, unsigned long caller ret = __kmem_cache_alloc_node(s, flags, node, size, caller); ret = kasan_kmalloc(s, ret, size, flags); - trace_kmalloc(_RET_IP_, ret, size, s->size, flags, node); + trace_kmalloc(caller, ret, size, s->size, flags, node); return ret; } From 59dc2a7e7c6a0d4a2a2c4dafea874af484fc0592 Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Mon, 31 Oct 2022 11:17:45 +0100 Subject: [PATCH 058/138] ACPI: video: Improve Chromebook checks 2 improvements for the Chromebook handling in acpi_video_get_backlight_type(): 1. Also check for the "GOOG000C" ACPI HID used on some models 2. Move the Chromebook check to above the ACPI-video check normally Chromebooks don't have ACPI video backlight support, but when flashed with upstream coreboot builds they may have ACPI video backlight support, but native should still be used/preferred then. Suggested-by: Mr. Chromebox Signed-off-by: Hans de Goede --- drivers/acpi/video_detect.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) diff --git a/drivers/acpi/video_detect.c b/drivers/acpi/video_detect.c index 9cd8797d12bb..06aaec2e378b 100644 --- a/drivers/acpi/video_detect.c +++ b/drivers/acpi/video_detect.c @@ -670,7 +670,7 @@ static const struct dmi_system_id video_detect_dmi_table[] = { static bool google_cros_ec_present(void) { - return acpi_dev_found("GOOG0004"); + return acpi_dev_found("GOOG0004") || acpi_dev_found("GOOG000C"); } /* @@ -718,6 +718,10 @@ static enum acpi_backlight_type __acpi_video_get_backlight_type(bool native) if (apple_gmux_present()) return acpi_backlight_apple_gmux; + /* Chromebooks should always prefer native backlight control. */ + if (google_cros_ec_present() && native_available) + return acpi_backlight_native; + /* On systems with ACPI video use either native or ACPI video. */ if (video_caps & ACPI_VIDEO_BACKLIGHT) { /* @@ -735,13 +739,6 @@ static enum acpi_backlight_type __acpi_video_get_backlight_type(bool native) return acpi_backlight_video; } - /* - * Chromebooks that don't have backlight handle in ACPI table - * are supposed to use native backlight if it's available. - */ - if (google_cros_ec_present() && native_available) - return acpi_backlight_native; - /* No ACPI video (old hw), use vendor specific fw methods. */ return acpi_backlight_vendor; } From 531705a765493655472c993627106e19f7e5a6d2 Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Thu, 3 Nov 2022 17:05:37 +0800 Subject: [PATCH 059/138] net: lapbether: fix issue of dev reference count leakage in lapbeth_device_event() When following tests are performed, it will cause dev reference counting leakage. a)ip link add bond2 type bond mode balance-rr b)ip link set bond2 up c)ifenslave -f bond2 rose1 d)ip link del bond2 When new bond device is created, the default type of the bond device is ether. And the bond device is up, lapbeth_device_event() receives the message and creates a new lapbeth device. In this case, the reference count value of dev is hold once. But after "ifenslave -f bond2 rose1" command is executed, the type of the bond device is changed to rose. When the bond device is unregistered, lapbeth_device_event() will not put the dev reference count. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Zhengchao Shao Signed-off-by: David S. Miller --- drivers/net/wan/lapbether.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wan/lapbether.c b/drivers/net/wan/lapbether.c index 960f1393595c..cb360dca3250 100644 --- a/drivers/net/wan/lapbether.c +++ b/drivers/net/wan/lapbether.c @@ -446,7 +446,7 @@ static int lapbeth_device_event(struct notifier_block *this, if (dev_net(dev) != &init_net) return NOTIFY_DONE; - if (!dev_is_ethdev(dev)) + if (!dev_is_ethdev(dev) && !lapbeth_get_x25_dev(dev)) return NOTIFY_DONE; switch (event) { From 85cbaf032d3cd9f595152625eda5d4ecb1d6d78d Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Thu, 3 Nov 2022 17:09:05 +0800 Subject: [PATCH 060/138] hamradio: fix issue of dev reference count leakage in bpq_device_event() When following tests are performed, it will cause dev reference counting leakage. a)ip link add bond2 type bond mode balance-rr b)ip link set bond2 up c)ifenslave -f bond2 rose1 d)ip link del bond2 When new bond device is created, the default type of the bond device is ether. And the bond device is up, bpq_device_event() receives the message and creates a new bpq device. In this case, the reference count value of dev is hold once. But after "ifenslave -f bond2 rose1" command is executed, the type of the bond device is changed to rose. When the bond device is unregistered, bpq_device_event() will not put the dev reference count. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Zhengchao Shao Signed-off-by: David S. Miller --- drivers/net/hamradio/bpqether.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/hamradio/bpqether.c b/drivers/net/hamradio/bpqether.c index 30af0081e2be..83a16d10eedb 100644 --- a/drivers/net/hamradio/bpqether.c +++ b/drivers/net/hamradio/bpqether.c @@ -533,7 +533,7 @@ static int bpq_device_event(struct notifier_block *this, if (!net_eq(dev_net(dev), &init_net)) return NOTIFY_DONE; - if (!dev_is_ethdev(dev)) + if (!dev_is_ethdev(dev) && !bpq_get_ax25_dev(dev)) return NOTIFY_DONE; switch (event) { From f25caaca424703d5a0607310f0452f978f1f78d9 Mon Sep 17 00:00:00 2001 From: HW He Date: Thu, 3 Nov 2022 18:40:00 +0800 Subject: [PATCH 061/138] net: wwan: iosm: fix memory leak in ipc_wwan_dellink IOSM driver registers network device without setting the needs_free_netdev flag, and does NOT call free_netdev() when unregisters network device, which causes a memory leak. This patch sets needs_free_netdev to true when registers network device, which makes netdev subsystem call free_netdev() automatically after unregister_netdevice(). Fixes: 2a54f2c77934 ("net: iosm: net driver") Signed-off-by: HW He Reviewed-by: Loic Poulain Signed-off-by: Zhaoping Shu Signed-off-by: David S. Miller --- drivers/net/wwan/iosm/iosm_ipc_wwan.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/wwan/iosm/iosm_ipc_wwan.c b/drivers/net/wwan/iosm/iosm_ipc_wwan.c index 2f1f8b5d5b59..0108d8d01ff2 100644 --- a/drivers/net/wwan/iosm/iosm_ipc_wwan.c +++ b/drivers/net/wwan/iosm/iosm_ipc_wwan.c @@ -168,6 +168,7 @@ static void ipc_wwan_setup(struct net_device *iosm_dev) iosm_dev->max_mtu = ETH_MAX_MTU; iosm_dev->flags = IFF_POINTOPOINT | IFF_NOARP; + iosm_dev->needs_free_netdev = true; iosm_dev->netdev_ops = &ipc_inm_ops; } From 668205b9c9f94d5ed6ab00cce9a46a654c2b5d16 Mon Sep 17 00:00:00 2001 From: HW He Date: Thu, 3 Nov 2022 18:54:19 +0800 Subject: [PATCH 062/138] net: wwan: mhi: fix memory leak in mhi_mbim_dellink MHI driver registers network device without setting the needs_free_netdev flag, and does NOT call free_netdev() when unregisters network device, which causes a memory leak. This patch sets needs_free_netdev to true when registers network device, which makes netdev subsystem call free_netdev() automatically after unregister_netdevice(). Fixes: aa730a9905b7 ("net: wwan: Add MHI MBIM network driver") Signed-off-by: HW He Signed-off-by: Zhaoping Shu Signed-off-by: David S. Miller --- drivers/net/wwan/mhi_wwan_mbim.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/wwan/mhi_wwan_mbim.c b/drivers/net/wwan/mhi_wwan_mbim.c index 6872782e8dd8..ef70bb7c88ad 100644 --- a/drivers/net/wwan/mhi_wwan_mbim.c +++ b/drivers/net/wwan/mhi_wwan_mbim.c @@ -582,6 +582,7 @@ static void mhi_mbim_setup(struct net_device *ndev) ndev->min_mtu = ETH_MIN_MTU; ndev->max_mtu = MHI_MAX_BUF_SZ - ndev->needed_headroom; ndev->tx_queue_len = 1000; + ndev->needs_free_netdev = true; } static const struct wwan_ops mhi_mbim_wwan_ops = { From f9027f88f7d1dc2b0ed1afbf28f22992d72d7efe Mon Sep 17 00:00:00 2001 From: Zhaoping Shu Date: Thu, 3 Nov 2022 19:08:49 +0800 Subject: [PATCH 063/138] net: wwan: iosm: Remove unnecessary if_mutex lock These WWAN network interface operations (create/delete/open/close) are already protected by RTNL lock, i.e., wwan_ops.newlink(), wwan_ops.dellink(), net_device_ops.ndo_open() and net_device.ndo_stop() are called with RTNL lock held. Therefore, this patch removes the unnecessary if_mutex. Signed-off-by: Zhaoping Shu Signed-off-by: David S. Miller --- drivers/net/wwan/iosm/iosm_ipc_wwan.c | 42 ++++----------------------- 1 file changed, 6 insertions(+), 36 deletions(-) diff --git a/drivers/net/wwan/iosm/iosm_ipc_wwan.c b/drivers/net/wwan/iosm/iosm_ipc_wwan.c index 0108d8d01ff2..4c9022a93e01 100644 --- a/drivers/net/wwan/iosm/iosm_ipc_wwan.c +++ b/drivers/net/wwan/iosm/iosm_ipc_wwan.c @@ -40,13 +40,11 @@ struct iosm_netdev_priv { * @ipc_imem: Pointer to imem data-struct * @sub_netlist: List of active netdevs * @dev: Pointer device structure - * @if_mutex: Mutex used for add and remove interface id */ struct iosm_wwan { struct iosm_imem *ipc_imem; struct iosm_netdev_priv __rcu *sub_netlist[IP_MUX_SESSION_END + 1]; struct device *dev; - struct mutex if_mutex; /* Mutex used for add and remove interface id */ }; /* Bring-up the wwan net link */ @@ -55,14 +53,11 @@ static int ipc_wwan_link_open(struct net_device *netdev) struct iosm_netdev_priv *priv = wwan_netdev_drvpriv(netdev); struct iosm_wwan *ipc_wwan = priv->ipc_wwan; int if_id = priv->if_id; - int ret; if (if_id < IP_MUX_SESSION_START || if_id >= ARRAY_SIZE(ipc_wwan->sub_netlist)) return -EINVAL; - mutex_lock(&ipc_wwan->if_mutex); - /* get channel id */ priv->ch_id = ipc_imem_sys_wwan_open(ipc_wwan->ipc_imem, if_id); @@ -70,8 +65,7 @@ static int ipc_wwan_link_open(struct net_device *netdev) dev_err(ipc_wwan->dev, "cannot connect wwan0 & id %d to the IPC mem layer", if_id); - ret = -ENODEV; - goto out; + return -ENODEV; } /* enable tx path, DL data may follow */ @@ -80,10 +74,7 @@ static int ipc_wwan_link_open(struct net_device *netdev) dev_dbg(ipc_wwan->dev, "Channel id %d allocated to if_id %d", priv->ch_id, priv->if_id); - ret = 0; -out: - mutex_unlock(&ipc_wwan->if_mutex); - return ret; + return 0; } /* Bring-down the wwan net link */ @@ -93,11 +84,9 @@ static int ipc_wwan_link_stop(struct net_device *netdev) netif_stop_queue(netdev); - mutex_lock(&priv->ipc_wwan->if_mutex); ipc_imem_sys_wwan_close(priv->ipc_wwan->ipc_imem, priv->if_id, priv->ch_id); priv->ch_id = -1; - mutex_unlock(&priv->ipc_wwan->if_mutex); return 0; } @@ -190,26 +179,17 @@ static int ipc_wwan_newlink(void *ctxt, struct net_device *dev, priv->netdev = dev; priv->ipc_wwan = ipc_wwan; - mutex_lock(&ipc_wwan->if_mutex); - if (rcu_access_pointer(ipc_wwan->sub_netlist[if_id])) { - err = -EBUSY; - goto out_unlock; - } + if (rcu_access_pointer(ipc_wwan->sub_netlist[if_id])) + return -EBUSY; err = register_netdevice(dev); if (err) - goto out_unlock; + return err; rcu_assign_pointer(ipc_wwan->sub_netlist[if_id], priv); - mutex_unlock(&ipc_wwan->if_mutex); - netif_device_attach(dev); return 0; - -out_unlock: - mutex_unlock(&ipc_wwan->if_mutex); - return err; } static void ipc_wwan_dellink(void *ctxt, struct net_device *dev, @@ -223,17 +203,12 @@ static void ipc_wwan_dellink(void *ctxt, struct net_device *dev, if_id >= ARRAY_SIZE(ipc_wwan->sub_netlist))) return; - mutex_lock(&ipc_wwan->if_mutex); - if (WARN_ON(rcu_access_pointer(ipc_wwan->sub_netlist[if_id]) != priv)) - goto unlock; + return; RCU_INIT_POINTER(ipc_wwan->sub_netlist[if_id], NULL); /* unregistering includes synchronize_net() */ unregister_netdevice_queue(dev, head); - -unlock: - mutex_unlock(&ipc_wwan->if_mutex); } static const struct wwan_ops iosm_wwan_ops = { @@ -324,12 +299,9 @@ struct iosm_wwan *ipc_wwan_init(struct iosm_imem *ipc_imem, struct device *dev) ipc_wwan->dev = dev; ipc_wwan->ipc_imem = ipc_imem; - mutex_init(&ipc_wwan->if_mutex); - /* WWAN core will create a netdev for the default IP MUX channel */ if (wwan_register_ops(ipc_wwan->dev, &iosm_wwan_ops, ipc_wwan, IP_MUX_SESSION_DEFAULT)) { - mutex_destroy(&ipc_wwan->if_mutex); kfree(ipc_wwan); return NULL; } @@ -342,7 +314,5 @@ void ipc_wwan_deinit(struct iosm_wwan *ipc_wwan) /* This call will remove all child netdev(s) */ wwan_unregister_ops(ipc_wwan->dev); - mutex_destroy(&ipc_wwan->if_mutex); - kfree(ipc_wwan); } From 0c175da7b0378445f5ef53904247cfbfb87e0b78 Mon Sep 17 00:00:00 2001 From: Lu Wei Date: Fri, 4 Nov 2022 10:27:23 +0800 Subject: [PATCH 064/138] tcp: prohibit TCP_REPAIR_OPTIONS if data was already sent If setsockopt with option name of TCP_REPAIR_OPTIONS and opt_code of TCPOPT_SACK_PERM is called to enable sack after data is sent and dupacks are received , it will trigger a warning in function tcp_verify_left_out() as follows: ============================================ WARNING: CPU: 8 PID: 0 at net/ipv4/tcp_input.c:2132 tcp_timeout_mark_lost+0x154/0x160 tcp_enter_loss+0x2b/0x290 tcp_retransmit_timer+0x50b/0x640 tcp_write_timer_handler+0x1c8/0x340 tcp_write_timer+0xe5/0x140 call_timer_fn+0x3a/0x1b0 __run_timers.part.0+0x1bf/0x2d0 run_timer_softirq+0x43/0xb0 __do_softirq+0xfd/0x373 __irq_exit_rcu+0xf6/0x140 The warning is caused in the following steps: 1. a socket named socketA is created 2. socketA enters repair mode without build a connection 3. socketA calls connect() and its state is changed to TCP_ESTABLISHED directly 4. socketA leaves repair mode 5. socketA calls sendmsg() to send data, packets_out and sack_outs(dup ack receives) increase 6. socketA enters repair mode again 7. socketA calls setsockopt with TCPOPT_SACK_PERM to enable sack 8. retransmit timer expires, it calls tcp_timeout_mark_lost(), lost_out increases 9. sack_outs + lost_out > packets_out triggers since lost_out and sack_outs increase repeatly In function tcp_timeout_mark_lost(), tp->sacked_out will be cleared if Step7 not happen and the warning will not be triggered. As suggested by Denis and Eric, TCP_REPAIR_OPTIONS should be prohibited if data was already sent. socket-tcp tests in CRIU has been tested as follows: $ sudo ./test/zdtm.py run -t zdtm/static/socket-tcp* --keep-going \ --ignore-taint socket-tcp* represent all socket-tcp tests in test/zdtm/static/. Fixes: b139ba4e90dc ("tcp: Repair connection-time negotiated parameters") Signed-off-by: Lu Wei Reviewed-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv4/tcp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index ef14efa1fb70..54836a6b81d6 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -3647,7 +3647,7 @@ int do_tcp_setsockopt(struct sock *sk, int level, int optname, case TCP_REPAIR_OPTIONS: if (!tp->repair) err = -EINVAL; - else if (sk->sk_state == TCP_ESTABLISHED) + else if (sk->sk_state == TCP_ESTABLISHED && !tp->bytes_sent) err = tcp_repair_options_est(sk, optval, optlen); else err = -EPERM; From 454d61a56d5e4c8cd9905f81d475d23f86f370af Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Fri, 4 Nov 2022 18:06:29 +0100 Subject: [PATCH 065/138] ACPI: video: Make acpi_video_backlight_use_native() always return true Testing has shown that there are quite a few laptop models which rely on native backlight control and which do not support ACPI video bus backlight control, causing __acpi_video_get_backlight_type() to return vendor. Known Windows laptop models affected by this are: Acer Aspire 1640 HP Compaq nc6120 IBM ThinkPad X40 System76 Starling Star1 and the following MacBook models are affected too: Apple MacBook 2.1 Apple MacBook 4.1 Apple MacBook Pro 7.1 the list of affected Windows laptop models is likely just the top of the iceberg. So for now lets undo the change to not register native backlight class devices when __acpi_video_get_backlight_type() != native. Since as part of the backlight-detect refactor the detection code now relies on the GPU drivers calling acpi_video_backlight_use_native() to learn that native backlight support is available we cannot just remove the acpi_video_backlight_use_native() calls from the GPU drivers. Instead modify acpi_video_backlight_use_native() to always return true for now. This is meant as a temporary work-around, which will be removed again when the heuristics from __acpi_video_get_backlight_type() have been improved so that they will return native on affected models. Reported-by: Matthew Garrett Reported-by: John Warriner Reported-by: Scott Ostrander Reported-by: Matthias Rampke Reported-by: Milan Hodoscek Signed-off-by: Hans de Goede --- drivers/acpi/video_detect.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/drivers/acpi/video_detect.c b/drivers/acpi/video_detect.c index 06aaec2e378b..c02e960cdbe3 100644 --- a/drivers/acpi/video_detect.c +++ b/drivers/acpi/video_detect.c @@ -751,6 +751,18 @@ EXPORT_SYMBOL(acpi_video_get_backlight_type); bool acpi_video_backlight_use_native(void) { - return __acpi_video_get_backlight_type(true) == acpi_backlight_native; + /* + * Call __acpi_video_get_backlight_type() to let it know that + * a native backlight is available. + */ + __acpi_video_get_backlight_type(true); + + /* + * For now just always return true. There is a whole bunch of laptop + * models where (video_caps & ACPI_VIDEO_BACKLIGHT) is false causing + * __acpi_video_get_backlight_type() to return vendor, while these + * models only have a native backlight control. + */ + return true; } EXPORT_SYMBOL(acpi_video_backlight_use_native); From f46acc1efd4b5846de9fa05f966e504f328f34a6 Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Mon, 31 Oct 2022 21:20:59 +0100 Subject: [PATCH 066/138] ACPI: video: Add backlight=native DMI quirk for Dell G15 5515 The Dell G15 5515 has the WMI interface (and WMI call returns) expected by the nvidia-wmi-ec-backlight interface. But the backlight class device registered by the nvidia-wmi-ec-backlight driver does not actually work. The amdgpu_bl0 native GPU backlight class device does actually work, add a backlight=native DMI quirk for this. Reported-by: Iris Reviewed-by: Daniel Dadap Signed-off-by: Hans de Goede --- Changes in v2: - Add a comment that this needs to be revisited when dynamic-mux support gets added (suggested by: Daniel Dadap) --- drivers/acpi/video_detect.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/drivers/acpi/video_detect.c b/drivers/acpi/video_detect.c index c02e960cdbe3..b2a616287638 100644 --- a/drivers/acpi/video_detect.c +++ b/drivers/acpi/video_detect.c @@ -645,6 +645,20 @@ static const struct dmi_system_id video_detect_dmi_table[] = { }, }, + /* + * Models which have nvidia-ec-wmi support, but should not use it. + * Note this indicates a likely firmware bug on these models and should + * be revisited if/when Linux gets support for dynamic mux mode. + */ + { + .callback = video_detect_force_native, + /* Dell G15 5515 */ + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "Dell G15 5515"), + }, + }, + /* * Desktops which falsely report a backlight and which our heuristics * for this do not catch. From a231224a601c1924b9df620281ad04472900d75f Mon Sep 17 00:00:00 2001 From: Manyi Li Date: Tue, 18 Oct 2022 17:53:23 +0800 Subject: [PATCH 067/138] platform/x86: ideapad-laptop: Disable touchpad_switch Ideapads for "Lenovo Yoga 3 Pro 1370" and "ZhaoYang K4e-IML" do not use EC to switch touchpad. Reading VPCCMD_R_TOUCHPAD will return zero thus touchpad may be blocked unexpectedly. Signed-off-by: Manyi Li Link: https://lore.kernel.org/r/20221018095323.14591-1-limanyi@uniontech.com Reviewed-by: Hans de Goede Signed-off-by: Hans de Goede --- drivers/platform/x86/ideapad-laptop.c | 25 ++++++++++++++++++++++++- 1 file changed, 24 insertions(+), 1 deletion(-) diff --git a/drivers/platform/x86/ideapad-laptop.c b/drivers/platform/x86/ideapad-laptop.c index abd0c81d62c4..33b3dfdd1b08 100644 --- a/drivers/platform/x86/ideapad-laptop.c +++ b/drivers/platform/x86/ideapad-laptop.c @@ -1533,6 +1533,24 @@ static const struct dmi_system_id hw_rfkill_list[] = { {} }; +static const struct dmi_system_id no_touchpad_switch_list[] = { + { + .ident = "Lenovo Yoga 3 Pro 1370", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), + DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo YOGA 3"), + }, + }, + { + .ident = "ZhaoYang K4e-IML", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), + DMI_MATCH(DMI_PRODUCT_VERSION, "ZhaoYang K4e-IML"), + }, + }, + {} +}; + static void ideapad_check_features(struct ideapad_private *priv) { acpi_handle handle = priv->adev->handle; @@ -1541,7 +1559,12 @@ static void ideapad_check_features(struct ideapad_private *priv) priv->features.hw_rfkill_switch = dmi_check_system(hw_rfkill_list); /* Most ideapads with ELAN0634 touchpad don't use EC touchpad switch */ - priv->features.touchpad_ctrl_via_ec = !acpi_dev_present("ELAN0634", NULL, -1); + if (acpi_dev_present("ELAN0634", NULL, -1)) + priv->features.touchpad_ctrl_via_ec = 0; + else if (dmi_check_system(no_touchpad_switch_list)) + priv->features.touchpad_ctrl_via_ec = 0; + else + priv->features.touchpad_ctrl_via_ec = 1; if (!read_ec_data(handle, VPCCMD_R_FAN, &val)) priv->features.fan_mode = true; From 0df044b34bf33e7e35c32b3bf6747fde6279c162 Mon Sep 17 00:00:00 2001 From: Hans de Goede Date: Tue, 25 Oct 2022 16:11:31 +0200 Subject: [PATCH 068/138] platform/x86: touchscreen_dmi: Add info for the RCA Cambio W101 v2 2-in-1 Add touchscreen info for the RCA Cambio W101 v2 2-in-1. Link: https://github.com/onitake/gsl-firmware/discussions/193 Signed-off-by: Hans de Goede Link: https://lore.kernel.org/r/20221025141131.509211-1-hdegoede@redhat.com --- drivers/platform/x86/touchscreen_dmi.c | 25 +++++++++++++++++++++++++ 1 file changed, 25 insertions(+) diff --git a/drivers/platform/x86/touchscreen_dmi.c b/drivers/platform/x86/touchscreen_dmi.c index bc97bfa8e8a6..baae3120efd0 100644 --- a/drivers/platform/x86/touchscreen_dmi.c +++ b/drivers/platform/x86/touchscreen_dmi.c @@ -770,6 +770,22 @@ static const struct ts_dmi_data predia_basic_data = { .properties = predia_basic_props, }; +static const struct property_entry rca_cambio_w101_v2_props[] = { + PROPERTY_ENTRY_U32("touchscreen-min-x", 4), + PROPERTY_ENTRY_U32("touchscreen-min-y", 20), + PROPERTY_ENTRY_U32("touchscreen-size-x", 1644), + PROPERTY_ENTRY_U32("touchscreen-size-y", 874), + PROPERTY_ENTRY_BOOL("touchscreen-swapped-x-y"), + PROPERTY_ENTRY_STRING("firmware-name", "gsl1680-rca-cambio-w101-v2.fw"), + PROPERTY_ENTRY_U32("silead,max-fingers", 10), + { } +}; + +static const struct ts_dmi_data rca_cambio_w101_v2_data = { + .acpi_name = "MSSL1680:00", + .properties = rca_cambio_w101_v2_props, +}; + static const struct property_entry rwc_nanote_p8_props[] = { PROPERTY_ENTRY_U32("touchscreen-min-y", 46), PROPERTY_ENTRY_U32("touchscreen-size-x", 1728), @@ -1409,6 +1425,15 @@ const struct dmi_system_id touchscreen_dmi_table[] = { DMI_EXACT_MATCH(DMI_BOARD_NAME, "0E57"), }, }, + { + /* RCA Cambio W101 v2 */ + /* https://github.com/onitake/gsl-firmware/discussions/193 */ + .driver_data = (void *)&rca_cambio_w101_v2_data, + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "RCA"), + DMI_MATCH(DMI_PRODUCT_NAME, "W101SA23T1"), + }, + }, { /* RWC NANOTE P8 */ .driver_data = (void *)&rwc_nanote_p8_data, From 1598bfa8e1faa932de42e1ee7628a1c4c4263f0a Mon Sep 17 00:00:00 2001 From: Jorge Lopez Date: Fri, 28 Oct 2022 10:55:27 -0500 Subject: [PATCH 069/138] platform/x86: hp_wmi: Fix rfkill causing soft blocked wifi After upgrading BIOS to U82 01.02.01 Rev.A, the console is flooded strange char "^@" which printed out every second and makes login nearly impossible. Also the below messages were shown both in console and journal/dmesg every second: usb 1-3: Device not responding to setup address. usb 1-3: device not accepting address 4, error -71 usb 1-3: device descriptor read/all, error -71 usb usb1-port3: unable to enumerate USB device Wifi is soft blocked by checking rfkill. When unblocked manually, after few seconds it would be soft blocked again. So I was suspecting something triggered rfkill to soft block wifi. At the end it was fixed by removing hp_wmi module. The root cause is the way hp-wmi driver handles command 1B on post-2009 BIOS. In pre-2009 BIOS, command 1Bh return 0x4 to indicate that BIOS no longer controls the power for the wireless devices. Signed-off-by: Jorge Lopez Link: https://bugzilla.kernel.org/show_bug.cgi?id=216468 Reviewed-by: Mario Limonciello Link: https://lore.kernel.org/r/20221028155527.7724-1-jorge.lopez2@hp.com Cc: stable@vger.kernel.org Reviewed-by: Hans de Goede Signed-off-by: Hans de Goede --- drivers/platform/x86/hp-wmi.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/drivers/platform/x86/hp-wmi.c b/drivers/platform/x86/hp-wmi.c index 627a6d0eaf83..12449038bed1 100644 --- a/drivers/platform/x86/hp-wmi.c +++ b/drivers/platform/x86/hp-wmi.c @@ -1300,8 +1300,16 @@ static int __init hp_wmi_bios_setup(struct platform_device *device) wwan_rfkill = NULL; rfkill2_count = 0; - if (hp_wmi_rfkill_setup(device)) - hp_wmi_rfkill2_setup(device); + /* + * In pre-2009 BIOS, command 1Bh return 0x4 to indicate that + * BIOS no longer controls the power for the wireless + * devices. All features supported by this command will no + * longer be supported. + */ + if (!hp_wmi_bios_2009_later()) { + if (hp_wmi_rfkill_setup(device)) + hp_wmi_rfkill2_setup(device); + } err = hp_wmi_hwmon_init(); From bcdfa1f77ea7f67368d20384932a9d1e3047ddd2 Mon Sep 17 00:00:00 2001 From: "David E. Box" Date: Fri, 4 Nov 2022 20:42:28 -0700 Subject: [PATCH 070/138] platform/x86/intel/pmt: Sapphire Rapids PMT errata fix On Sapphire Rapids, due to a hardware issue affecting the PUNIT telemetry region, reads that are not done in QWORD quantities and alignment may return incorrect data. Use a custom 64-bit copy for this region. Signed-off-by: David E. Box Link: https://lore.kernel.org/r/20221105034228.1376677-1-david.e.box@linux.intel.com Reviewed-by: Hans de Goede Signed-off-by: Hans de Goede --- drivers/platform/x86/intel/pmt/class.c | 31 +++++++++++++++++++++++++- 1 file changed, 30 insertions(+), 1 deletion(-) diff --git a/drivers/platform/x86/intel/pmt/class.c b/drivers/platform/x86/intel/pmt/class.c index 53d7fd2943b4..46598dcb634a 100644 --- a/drivers/platform/x86/intel/pmt/class.c +++ b/drivers/platform/x86/intel/pmt/class.c @@ -9,6 +9,7 @@ */ #include +#include #include #include #include @@ -19,6 +20,7 @@ #define PMT_XA_START 0 #define PMT_XA_MAX INT_MAX #define PMT_XA_LIMIT XA_LIMIT(PMT_XA_START, PMT_XA_MAX) +#define GUID_SPR_PUNIT 0x9956f43f bool intel_pmt_is_early_client_hw(struct device *dev) { @@ -33,6 +35,29 @@ bool intel_pmt_is_early_client_hw(struct device *dev) } EXPORT_SYMBOL_GPL(intel_pmt_is_early_client_hw); +static inline int +pmt_memcpy64_fromio(void *to, const u64 __iomem *from, size_t count) +{ + int i, remain; + u64 *buf = to; + + if (!IS_ALIGNED((unsigned long)from, 8)) + return -EFAULT; + + for (i = 0; i < count/8; i++) + buf[i] = readq(&from[i]); + + /* Copy any remaining bytes */ + remain = count % 8; + if (remain) { + u64 tmp = readq(&from[i]); + + memcpy(&buf[i], &tmp, remain); + } + + return count; +} + /* * sysfs */ @@ -54,7 +79,11 @@ intel_pmt_read(struct file *filp, struct kobject *kobj, if (count > entry->size - off) count = entry->size - off; - memcpy_fromio(buf, entry->base + off, count); + if (entry->guid == GUID_SPR_PUNIT) + /* PUNIT on SPR only supports aligned 64-bit read */ + count = pmt_memcpy64_fromio(buf, entry->base + off, count); + else + memcpy_fromio(buf, entry->base + off, count); return count; } From a977ece5773b6746b814aac410da4776023db239 Mon Sep 17 00:00:00 2001 From: Ivan Hu Date: Wed, 2 Nov 2022 10:05:48 +0800 Subject: [PATCH 071/138] platform/x86/intel/hid: Add some ACPI device IDs Add INTC1076 (JasonLake), INTC1077 (MeteorLake) and INTC1078 (RaptorLake) devices IDs. Signed-off-by: Ivan Hu Link: https://lore.kernel.org/r/20221102020548.5225-1-ivan.hu@canonical.com Reviewed-by: Hans de Goede Signed-off-by: Hans de Goede --- drivers/platform/x86/intel/hid.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/platform/x86/intel/hid.c b/drivers/platform/x86/intel/hid.c index 79cff1fc675c..b6313ecd190c 100644 --- a/drivers/platform/x86/intel/hid.c +++ b/drivers/platform/x86/intel/hid.c @@ -27,6 +27,9 @@ static const struct acpi_device_id intel_hid_ids[] = { {"INTC1051", 0}, {"INTC1054", 0}, {"INTC1070", 0}, + {"INTC1076", 0}, + {"INTC1077", 0}, + {"INTC1078", 0}, {"", 0}, }; MODULE_DEVICE_TABLE(acpi, intel_hid_ids); From 53eb64c88f17b14b324fbdfd417f56c5d3fa6fee Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Fri, 4 Nov 2022 17:49:16 +0200 Subject: [PATCH 072/138] platform/x86: p2sb: Don't fail if unknown CPU is found We have accessing P2SB from a very few places for quite known hardware. When a new SoC appears in intel-family.h it's not obvious that it needs to be added to p2sb.c as well. Instead, provide default BDF and refactor p2sb_get_devfn() to always succeed. If in the future we would need to exclude something, we may add a list of unsupported IDs. Without this change the iTCO on Intel Comet Lake SoCs became unavailable: i801_smbus 0000:00:1f.4: failed to create iTCO device Fixes: 5c7b9167ddf8 ("i2c: i801: convert to use common P2SB accessor") Reported-and-tested-by: Jarkko Nikula Signed-off-by: Andy Shevchenko Link: https://lore.kernel.org/r/20221104154916.35231-1-andriy.shevchenko@linux.intel.com Reviewed-by: Hans de Goede Signed-off-by: Hans de Goede --- drivers/platform/x86/p2sb.c | 15 ++++++--------- 1 file changed, 6 insertions(+), 9 deletions(-) diff --git a/drivers/platform/x86/p2sb.c b/drivers/platform/x86/p2sb.c index 384d0962ae93..1cf2471d54dd 100644 --- a/drivers/platform/x86/p2sb.c +++ b/drivers/platform/x86/p2sb.c @@ -19,26 +19,23 @@ #define P2SBC 0xe0 #define P2SBC_HIDE BIT(8) +#define P2SB_DEVFN_DEFAULT PCI_DEVFN(31, 1) + static const struct x86_cpu_id p2sb_cpu_ids[] = { X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT, PCI_DEVFN(13, 0)), - X86_MATCH_INTEL_FAM6_MODEL(ATOM_GOLDMONT_D, PCI_DEVFN(31, 1)), - X86_MATCH_INTEL_FAM6_MODEL(ATOM_SILVERMONT_D, PCI_DEVFN(31, 1)), - X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE, PCI_DEVFN(31, 1)), - X86_MATCH_INTEL_FAM6_MODEL(KABYLAKE_L, PCI_DEVFN(31, 1)), - X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE, PCI_DEVFN(31, 1)), - X86_MATCH_INTEL_FAM6_MODEL(SKYLAKE_L, PCI_DEVFN(31, 1)), {} }; static int p2sb_get_devfn(unsigned int *devfn) { + unsigned int fn = P2SB_DEVFN_DEFAULT; const struct x86_cpu_id *id; id = x86_match_cpu(p2sb_cpu_ids); - if (!id) - return -ENODEV; + if (id) + fn = (unsigned int)id->driver_data; - *devfn = (unsigned int)id->driver_data; + *devfn = fn; return 0; } From c23fb2c82267638f9d206cb96bb93e1f93ad7828 Mon Sep 17 00:00:00 2001 From: Alexander Potapenko Date: Fri, 4 Nov 2022 11:32:16 +0100 Subject: [PATCH 073/138] ipv6: addrlabel: fix infoleak when sending struct ifaddrlblmsg to network When copying a `struct ifaddrlblmsg` to the network, __ifal_reserved remained uninitialized, resulting in a 1-byte infoleak: BUG: KMSAN: kernel-network-infoleak in __netdev_start_xmit ./include/linux/netdevice.h:4841 __netdev_start_xmit ./include/linux/netdevice.h:4841 netdev_start_xmit ./include/linux/netdevice.h:4857 xmit_one net/core/dev.c:3590 dev_hard_start_xmit+0x1dc/0x800 net/core/dev.c:3606 __dev_queue_xmit+0x17e8/0x4350 net/core/dev.c:4256 dev_queue_xmit ./include/linux/netdevice.h:3009 __netlink_deliver_tap_skb net/netlink/af_netlink.c:307 __netlink_deliver_tap+0x728/0xad0 net/netlink/af_netlink.c:325 netlink_deliver_tap net/netlink/af_netlink.c:338 __netlink_sendskb net/netlink/af_netlink.c:1263 netlink_sendskb+0x1d9/0x200 net/netlink/af_netlink.c:1272 netlink_unicast+0x56d/0xf50 net/netlink/af_netlink.c:1360 nlmsg_unicast ./include/net/netlink.h:1061 rtnl_unicast+0x5a/0x80 net/core/rtnetlink.c:758 ip6addrlbl_get+0xfad/0x10f0 net/ipv6/addrlabel.c:628 rtnetlink_rcv_msg+0xb33/0x1570 net/core/rtnetlink.c:6082 ... Uninit was created at: slab_post_alloc_hook+0x118/0xb00 mm/slab.h:742 slab_alloc_node mm/slub.c:3398 __kmem_cache_alloc_node+0x4f2/0x930 mm/slub.c:3437 __do_kmalloc_node mm/slab_common.c:954 __kmalloc_node_track_caller+0x117/0x3d0 mm/slab_common.c:975 kmalloc_reserve net/core/skbuff.c:437 __alloc_skb+0x27a/0xab0 net/core/skbuff.c:509 alloc_skb ./include/linux/skbuff.h:1267 nlmsg_new ./include/net/netlink.h:964 ip6addrlbl_get+0x490/0x10f0 net/ipv6/addrlabel.c:608 rtnetlink_rcv_msg+0xb33/0x1570 net/core/rtnetlink.c:6082 netlink_rcv_skb+0x299/0x550 net/netlink/af_netlink.c:2540 rtnetlink_rcv+0x26/0x30 net/core/rtnetlink.c:6109 netlink_unicast_kernel net/netlink/af_netlink.c:1319 netlink_unicast+0x9ab/0xf50 net/netlink/af_netlink.c:1345 netlink_sendmsg+0xebc/0x10f0 net/netlink/af_netlink.c:1921 ... This patch ensures that the reserved field is always initialized. Reported-by: syzbot+3553517af6020c4f2813f1003fe76ef3cbffe98d@syzkaller.appspotmail.com Fixes: 2a8cc6c89039 ("[IPV6] ADDRCONF: Support RFC3484 configurable address selection policy table.") Signed-off-by: Alexander Potapenko Reviewed-by: David Ahern Signed-off-by: David S. Miller --- net/ipv6/addrlabel.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ipv6/addrlabel.c b/net/ipv6/addrlabel.c index 8a22486cf270..17ac45aa7194 100644 --- a/net/ipv6/addrlabel.c +++ b/net/ipv6/addrlabel.c @@ -437,6 +437,7 @@ static void ip6addrlbl_putmsg(struct nlmsghdr *nlh, { struct ifaddrlblmsg *ifal = nlmsg_data(nlh); ifal->ifal_family = AF_INET6; + ifal->__ifal_reserved = 0; ifal->ifal_prefixlen = prefixlen; ifal->ifal_flags = 0; ifal->ifal_index = ifindex; From a3335faebe1608f3e78e24f09501c9f7d5ebf87a Mon Sep 17 00:00:00 2001 From: Chen Zhongjin Date: Mon, 31 Oct 2022 11:30:53 +0800 Subject: [PATCH 074/138] can: af_can: can_exit(): add missing dev_remove_pack() of canxl_packet In can_init(), dev_add_pack(&canxl_packet) is added but not removed in can_exit(). It breaks the packet handler list and can make kernel panic when can_init() is called for the second time. | > modprobe can && rmmod can | > rmmod xxx && modprobe can | | BUG: unable to handle page fault for address: fffffbfff807d7f4 | RIP: 0010:dev_add_pack+0x133/0x1f0 | Call Trace: | | can_init+0xaa/0x1000 [can] | do_one_initcall+0xd3/0x4e0 | ... Fixes: fb08cba12b52 ("can: canxl: update CAN infrastructure for CAN XL frames") Signed-off-by: Chen Zhongjin Acked-by: Oliver Hartkopp Link: https://lore.kernel.org/all/20221031033053.37849-1-chenzhongjin@huawei.com [mkl: adjust subject and commit message] Signed-off-by: Marc Kleine-Budde --- net/can/af_can.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/can/af_can.c b/net/can/af_can.c index 9503ab10f9b8..5e9e3e1e9825 100644 --- a/net/can/af_can.c +++ b/net/can/af_can.c @@ -902,6 +902,7 @@ static __init int can_init(void) static __exit void can_exit(void) { /* protocol unregister */ + dev_remove_pack(&canxl_packet); dev_remove_pack(&canfd_packet); dev_remove_pack(&can_packet); sock_unregister(PF_CAN); From 8aa59e355949442c408408c2d836e561794c40a1 Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Fri, 28 Oct 2022 16:56:50 +0800 Subject: [PATCH 075/138] can: af_can: fix NULL pointer dereference in can_rx_register() It causes NULL pointer dereference when testing as following: (a) use syscall(__NR_socket, 0x10ul, 3ul, 0) to create netlink socket. (b) use syscall(__NR_sendmsg, ...) to create bond link device and vxcan link device, and bind vxcan device to bond device (can also use ifenslave command to bind vxcan device to bond device). (c) use syscall(__NR_socket, 0x1dul, 3ul, 1) to create CAN socket. (d) use syscall(__NR_bind, ...) to bind the bond device to CAN socket. The bond device invokes the can-raw protocol registration interface to receive CAN packets. However, ml_priv is not allocated to the dev, dev_rcv_lists is assigned to NULL in can_rx_register(). In this case, it will occur the NULL pointer dereference issue. The following is the stack information: BUG: kernel NULL pointer dereference, address: 0000000000000008 PGD 122a4067 P4D 122a4067 PUD 1223c067 PMD 0 Oops: 0000 [#1] PREEMPT SMP RIP: 0010:can_rx_register+0x12d/0x1e0 Call Trace: raw_enable_filters+0x8d/0x120 raw_enable_allfilters+0x3b/0x130 raw_bind+0x118/0x4f0 __sys_bind+0x163/0x1a0 __x64_sys_bind+0x1e/0x30 do_syscall_64+0x35/0x80 entry_SYSCALL_64_after_hwframe+0x63/0xcd Fixes: 4e096a18867a ("net: introduce CAN specific pointer in the struct net_device") Signed-off-by: Zhengchao Shao Reviewed-by: Marc Kleine-Budde Link: https://lore.kernel.org/all/20221028085650.170470-1-shaozhengchao@huawei.com Signed-off-by: Marc Kleine-Budde --- net/can/af_can.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/can/af_can.c b/net/can/af_can.c index 5e9e3e1e9825..27dcdcc0b808 100644 --- a/net/can/af_can.c +++ b/net/can/af_can.c @@ -450,7 +450,7 @@ int can_rx_register(struct net *net, struct net_device *dev, canid_t can_id, /* insert new receiver (dev,canid,mask) -> (func,data) */ - if (dev && dev->type != ARPHRD_CAN) + if (dev && (dev->type != ARPHRD_CAN || !can_get_ml_priv(dev))) return -ENODEV; if (dev && !net_eq(net, dev_net(dev))) From 866337865f3747c68a3e7bb837611e39cec1ecd6 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Fri, 4 Nov 2022 15:25:51 +0100 Subject: [PATCH 076/138] can: isotp: fix tx state handling for echo tx processing In commit 4b7fe92c0690 ("can: isotp: add local echo tx processing for consecutive frames") the data flow for consecutive frames (CF) has been reworked to improve the reliability of long data transfers. This rework did not touch the transmission and the tx state changes of single frame (SF) transfers which likely led to the WARN in the isotp_tx_timer_handler() catching a wrong tx state. This patch makes use of the improved frame processing for SF frames and sets the ISOTP_SENDING state in isotp_sendmsg() within the cmpxchg() condition handling. A review of the state machine and the timer handling additionally revealed a missing echo timeout handling in the case of the burst mode in isotp_rcv_echo() and removes a potential timer configuration uncertainty in isotp_rcv_fc() when the receiver requests consecutive frames. Fixes: 4b7fe92c0690 ("can: isotp: add local echo tx processing for consecutive frames") Link: https://lore.kernel.org/linux-can/CAO4mrfe3dG7cMP1V5FLUkw7s+50c9vichigUMQwsxX4M=45QEw@mail.gmail.com/T/#u Reported-by: Wei Chen Cc: stable@vger.kernel.org # v6.0 Signed-off-by: Oliver Hartkopp Link: https://lore.kernel.org/all/20221104142551.16924-1-socketcan@hartkopp.net Signed-off-by: Marc Kleine-Budde --- net/can/isotp.c | 71 ++++++++++++++++++++++++++----------------------- 1 file changed, 38 insertions(+), 33 deletions(-) diff --git a/net/can/isotp.c b/net/can/isotp.c index a9d1357f8489..608f8c24ae46 100644 --- a/net/can/isotp.c +++ b/net/can/isotp.c @@ -111,6 +111,9 @@ MODULE_ALIAS("can-proto-6"); #define ISOTP_FC_WT 1 /* wait */ #define ISOTP_FC_OVFLW 2 /* overflow */ +#define ISOTP_FC_TIMEOUT 1 /* 1 sec */ +#define ISOTP_ECHO_TIMEOUT 2 /* 2 secs */ + enum { ISOTP_IDLE = 0, ISOTP_WAIT_FIRST_FC, @@ -258,7 +261,8 @@ static int isotp_send_fc(struct sock *sk, int ae, u8 flowstatus) so->lastrxcf_tstamp = ktime_set(0, 0); /* start rx timeout watchdog */ - hrtimer_start(&so->rxtimer, ktime_set(1, 0), HRTIMER_MODE_REL_SOFT); + hrtimer_start(&so->rxtimer, ktime_set(ISOTP_FC_TIMEOUT, 0), + HRTIMER_MODE_REL_SOFT); return 0; } @@ -344,6 +348,8 @@ static int check_pad(struct isotp_sock *so, struct canfd_frame *cf, return 0; } +static void isotp_send_cframe(struct isotp_sock *so); + static int isotp_rcv_fc(struct isotp_sock *so, struct canfd_frame *cf, int ae) { struct sock *sk = &so->sk; @@ -398,14 +404,15 @@ static int isotp_rcv_fc(struct isotp_sock *so, struct canfd_frame *cf, int ae) case ISOTP_FC_CTS: so->tx.bs = 0; so->tx.state = ISOTP_SENDING; - /* start cyclic timer for sending CF frame */ - hrtimer_start(&so->txtimer, so->tx_gap, + /* send CF frame and enable echo timeout handling */ + hrtimer_start(&so->txtimer, ktime_set(ISOTP_ECHO_TIMEOUT, 0), HRTIMER_MODE_REL_SOFT); + isotp_send_cframe(so); break; case ISOTP_FC_WT: /* start timer to wait for next FC frame */ - hrtimer_start(&so->txtimer, ktime_set(1, 0), + hrtimer_start(&so->txtimer, ktime_set(ISOTP_FC_TIMEOUT, 0), HRTIMER_MODE_REL_SOFT); break; @@ -600,7 +607,7 @@ static int isotp_rcv_cf(struct sock *sk, struct canfd_frame *cf, int ae, /* perform blocksize handling, if enabled */ if (!so->rxfc.bs || ++so->rx.bs < so->rxfc.bs) { /* start rx timeout watchdog */ - hrtimer_start(&so->rxtimer, ktime_set(1, 0), + hrtimer_start(&so->rxtimer, ktime_set(ISOTP_FC_TIMEOUT, 0), HRTIMER_MODE_REL_SOFT); return 0; } @@ -829,7 +836,7 @@ static void isotp_rcv_echo(struct sk_buff *skb, void *data) struct isotp_sock *so = isotp_sk(sk); struct canfd_frame *cf = (struct canfd_frame *)skb->data; - /* only handle my own local echo skb's */ + /* only handle my own local echo CF/SF skb's (no FF!) */ if (skb->sk != sk || so->cfecho != *(u32 *)cf->data) return; @@ -849,13 +856,16 @@ static void isotp_rcv_echo(struct sk_buff *skb, void *data) if (so->txfc.bs && so->tx.bs >= so->txfc.bs) { /* stop and wait for FC with timeout */ so->tx.state = ISOTP_WAIT_FC; - hrtimer_start(&so->txtimer, ktime_set(1, 0), + hrtimer_start(&so->txtimer, ktime_set(ISOTP_FC_TIMEOUT, 0), HRTIMER_MODE_REL_SOFT); return; } /* no gap between data frames needed => use burst mode */ if (!so->tx_gap) { + /* enable echo timeout handling */ + hrtimer_start(&so->txtimer, ktime_set(ISOTP_ECHO_TIMEOUT, 0), + HRTIMER_MODE_REL_SOFT); isotp_send_cframe(so); return; } @@ -879,7 +889,7 @@ static enum hrtimer_restart isotp_tx_timer_handler(struct hrtimer *hrtimer) /* start timeout for unlikely lost echo skb */ hrtimer_set_expires(&so->txtimer, ktime_add(ktime_get(), - ktime_set(2, 0))); + ktime_set(ISOTP_ECHO_TIMEOUT, 0))); restart = HRTIMER_RESTART; /* push out the next consecutive frame */ @@ -907,7 +917,8 @@ static enum hrtimer_restart isotp_tx_timer_handler(struct hrtimer *hrtimer) break; default: - WARN_ON_ONCE(1); + WARN_ONCE(1, "can-isotp: tx timer state %08X cfecho %08X\n", + so->tx.state, so->cfecho); } return restart; @@ -923,7 +934,7 @@ static int isotp_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) struct canfd_frame *cf; int ae = (so->opt.flags & CAN_ISOTP_EXTEND_ADDR) ? 1 : 0; int wait_tx_done = (so->opt.flags & CAN_ISOTP_WAIT_TX_DONE) ? 1 : 0; - s64 hrtimer_sec = 0; + s64 hrtimer_sec = ISOTP_ECHO_TIMEOUT; int off; int err; @@ -942,6 +953,8 @@ static int isotp_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) err = wait_event_interruptible(so->wait, so->tx.state == ISOTP_IDLE); if (err) goto err_out; + + so->tx.state = ISOTP_SENDING; } if (!size || size > MAX_MSG_LENGTH) { @@ -986,6 +999,10 @@ static int isotp_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) cf = (struct canfd_frame *)skb->data; skb_put_zero(skb, so->ll.mtu); + /* cfecho should have been zero'ed by init / former isotp_rcv_echo() */ + if (so->cfecho) + pr_notice_once("can-isotp: uninit cfecho %08X\n", so->cfecho); + /* check for single frame transmission depending on TX_DL */ if (size <= so->tx.ll_dl - SF_PCI_SZ4 - ae - off) { /* The message size generally fits into a SingleFrame - good. @@ -1011,11 +1028,8 @@ static int isotp_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) else cf->data[ae] |= size; - so->tx.state = ISOTP_IDLE; - wake_up_interruptible(&so->wait); - - /* don't enable wait queue for a single frame transmission */ - wait_tx_done = 0; + /* set CF echo tag for isotp_rcv_echo() (SF-mode) */ + so->cfecho = *(u32 *)cf->data; } else { /* send first frame */ @@ -1031,31 +1045,23 @@ static int isotp_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) /* disable wait for FCs due to activated block size */ so->txfc.bs = 0; - /* cfecho should have been zero'ed by init */ - if (so->cfecho) - pr_notice_once("can-isotp: no fc cfecho %08X\n", - so->cfecho); - - /* set consecutive frame echo tag */ + /* set CF echo tag for isotp_rcv_echo() (CF-mode) */ so->cfecho = *(u32 *)cf->data; - - /* switch directly to ISOTP_SENDING state */ - so->tx.state = ISOTP_SENDING; - - /* start timeout for unlikely lost echo skb */ - hrtimer_sec = 2; } else { /* standard flow control check */ so->tx.state = ISOTP_WAIT_FIRST_FC; /* start timeout for FC */ - hrtimer_sec = 1; - } + hrtimer_sec = ISOTP_FC_TIMEOUT; - hrtimer_start(&so->txtimer, ktime_set(hrtimer_sec, 0), - HRTIMER_MODE_REL_SOFT); + /* no CF echo tag for isotp_rcv_echo() (FF-mode) */ + so->cfecho = 0; + } } + hrtimer_start(&so->txtimer, ktime_set(hrtimer_sec, 0), + HRTIMER_MODE_REL_SOFT); + /* send the first or only CAN frame */ cf->flags = so->ll.tx_flags; @@ -1068,8 +1074,7 @@ static int isotp_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) __func__, ERR_PTR(err)); /* no transmission -> no timeout monitoring */ - if (hrtimer_sec) - hrtimer_cancel(&so->txtimer); + hrtimer_cancel(&so->txtimer); /* reset consecutive frame echo tag */ so->cfecho = 0; From 3eb3d283e8579a22b81dd2ac3987b77465b2a22f Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Fri, 4 Nov 2022 08:50:00 +0100 Subject: [PATCH 077/138] can: j1939: j1939_send_one(): fix missing CAN header initialization The read access to struct canxl_frame::len inside of a j1939 created skbuff revealed a missing initialization of reserved and later filled elements in struct can_frame. This patch initializes the 8 byte CAN header with zero. Fixes: 9d71dd0c7009 ("can: add support of SAE J1939 protocol") Cc: Oleksij Rempel Link: https://lore.kernel.org/linux-can/20221104052235.GA6474@pengutronix.de Reported-by: syzbot+d168ec0caca4697e03b1@syzkaller.appspotmail.com Signed-off-by: Oliver Hartkopp Link: https://lore.kernel.org/all/20221104075000.105414-1-socketcan@hartkopp.net Cc: stable@vger.kernel.org Signed-off-by: Marc Kleine-Budde --- net/can/j1939/main.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/can/j1939/main.c b/net/can/j1939/main.c index 144c86b0e3ff..821d4ff303b3 100644 --- a/net/can/j1939/main.c +++ b/net/can/j1939/main.c @@ -336,6 +336,9 @@ int j1939_send_one(struct j1939_priv *priv, struct sk_buff *skb) /* re-claim the CAN_HDR from the SKB */ cf = skb_push(skb, J1939_CAN_HDR); + /* initialize header structure */ + memset(cf, 0, J1939_CAN_HDR); + /* make it a full can frame again */ skb_put(skb, J1939_CAN_FTR + (8 - dlc)); From ae64438be1923e3c1102d90fd41db7afcfaf54cc Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Wed, 2 Nov 2022 10:54:31 +0100 Subject: [PATCH 078/138] can: dev: fix skb drop check In commit a6d190f8c767 ("can: skb: drop tx skb if in listen only mode") the priv->ctrlmode element is read even on virtual CAN interfaces that do not create the struct can_priv at startup. This out-of-bounds read may lead to CAN frame drops for virtual CAN interfaces like vcan and vxcan. This patch mainly reverts the original commit and adds a new helper for CAN interface drivers that provide the required information in struct can_priv. Fixes: a6d190f8c767 ("can: skb: drop tx skb if in listen only mode") Reported-by: Dariusz Stojaczyk Cc: Vincent Mailhol Cc: Max Staudt Signed-off-by: Oliver Hartkopp Acked-by: Vincent Mailhol Link: https://lore.kernel.org/all/20221102095431.36831-1-socketcan@hartkopp.net Cc: stable@vger.kernel.org # 6.0.x [mkl: patch pch_can, too] Signed-off-by: Marc Kleine-Budde --- drivers/net/can/at91_can.c | 2 +- drivers/net/can/c_can/c_can_main.c | 2 +- drivers/net/can/can327.c | 2 +- drivers/net/can/cc770/cc770.c | 2 +- drivers/net/can/ctucanfd/ctucanfd_base.c | 2 +- drivers/net/can/dev/skb.c | 10 +--------- drivers/net/can/flexcan/flexcan-core.c | 2 +- drivers/net/can/grcan.c | 2 +- drivers/net/can/ifi_canfd/ifi_canfd.c | 2 +- drivers/net/can/janz-ican3.c | 2 +- drivers/net/can/kvaser_pciefd.c | 2 +- drivers/net/can/m_can/m_can.c | 2 +- drivers/net/can/mscan/mscan.c | 2 +- drivers/net/can/pch_can.c | 2 +- drivers/net/can/peak_canfd/peak_canfd.c | 2 +- drivers/net/can/rcar/rcar_can.c | 2 +- drivers/net/can/rcar/rcar_canfd.c | 2 +- drivers/net/can/sja1000/sja1000.c | 2 +- drivers/net/can/slcan/slcan-core.c | 2 +- drivers/net/can/softing/softing_main.c | 2 +- drivers/net/can/spi/hi311x.c | 2 +- drivers/net/can/spi/mcp251x.c | 2 +- drivers/net/can/spi/mcp251xfd/mcp251xfd-tx.c | 2 +- drivers/net/can/sun4i_can.c | 2 +- drivers/net/can/ti_hecc.c | 2 +- drivers/net/can/usb/ems_usb.c | 2 +- drivers/net/can/usb/esd_usb.c | 2 +- drivers/net/can/usb/etas_es58x/es58x_core.c | 2 +- drivers/net/can/usb/gs_usb.c | 2 +- drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c | 2 +- drivers/net/can/usb/mcba_usb.c | 2 +- drivers/net/can/usb/peak_usb/pcan_usb_core.c | 2 +- drivers/net/can/usb/ucan.c | 2 +- drivers/net/can/usb/usb_8dev.c | 2 +- drivers/net/can/xilinx_can.c | 2 +- include/linux/can/dev.h | 16 ++++++++++++++++ 36 files changed, 51 insertions(+), 43 deletions(-) diff --git a/drivers/net/can/at91_can.c b/drivers/net/can/at91_can.c index 3a2d109a3792..199cb200f2bd 100644 --- a/drivers/net/can/at91_can.c +++ b/drivers/net/can/at91_can.c @@ -452,7 +452,7 @@ static netdev_tx_t at91_start_xmit(struct sk_buff *skb, struct net_device *dev) unsigned int mb, prio; u32 reg_mid, reg_mcr; - if (can_dropped_invalid_skb(dev, skb)) + if (can_dev_dropped_skb(dev, skb)) return NETDEV_TX_OK; mb = get_tx_next_mb(priv); diff --git a/drivers/net/can/c_can/c_can_main.c b/drivers/net/can/c_can/c_can_main.c index d6605dbb7737..c63f7fc1e691 100644 --- a/drivers/net/can/c_can/c_can_main.c +++ b/drivers/net/can/c_can/c_can_main.c @@ -457,7 +457,7 @@ static netdev_tx_t c_can_start_xmit(struct sk_buff *skb, struct c_can_tx_ring *tx_ring = &priv->tx; u32 idx, obj, cmd = IF_COMM_TX; - if (can_dropped_invalid_skb(dev, skb)) + if (can_dev_dropped_skb(dev, skb)) return NETDEV_TX_OK; if (c_can_tx_busy(priv, tx_ring)) diff --git a/drivers/net/can/can327.c b/drivers/net/can/can327.c index 0aa1af31d0fe..094197780776 100644 --- a/drivers/net/can/can327.c +++ b/drivers/net/can/can327.c @@ -813,7 +813,7 @@ static netdev_tx_t can327_netdev_start_xmit(struct sk_buff *skb, struct can327 *elm = netdev_priv(dev); struct can_frame *frame = (struct can_frame *)skb->data; - if (can_dropped_invalid_skb(dev, skb)) + if (can_dev_dropped_skb(dev, skb)) return NETDEV_TX_OK; /* We shouldn't get here after a hardware fault: diff --git a/drivers/net/can/cc770/cc770.c b/drivers/net/can/cc770/cc770.c index 0b9dfc76e769..30909f3aab57 100644 --- a/drivers/net/can/cc770/cc770.c +++ b/drivers/net/can/cc770/cc770.c @@ -429,7 +429,7 @@ static netdev_tx_t cc770_start_xmit(struct sk_buff *skb, struct net_device *dev) struct cc770_priv *priv = netdev_priv(dev); unsigned int mo = obj2msgobj(CC770_OBJ_TX); - if (can_dropped_invalid_skb(dev, skb)) + if (can_dev_dropped_skb(dev, skb)) return NETDEV_TX_OK; netif_stop_queue(dev); diff --git a/drivers/net/can/ctucanfd/ctucanfd_base.c b/drivers/net/can/ctucanfd/ctucanfd_base.c index b8da15ea6ad9..64c349fd4600 100644 --- a/drivers/net/can/ctucanfd/ctucanfd_base.c +++ b/drivers/net/can/ctucanfd/ctucanfd_base.c @@ -600,7 +600,7 @@ static netdev_tx_t ctucan_start_xmit(struct sk_buff *skb, struct net_device *nde bool ok; unsigned long flags; - if (can_dropped_invalid_skb(ndev, skb)) + if (can_dev_dropped_skb(ndev, skb)) return NETDEV_TX_OK; if (unlikely(!CTU_CAN_FD_TXTNF(priv))) { diff --git a/drivers/net/can/dev/skb.c b/drivers/net/can/dev/skb.c index 791a51e2f5d6..241ec636e91f 100644 --- a/drivers/net/can/dev/skb.c +++ b/drivers/net/can/dev/skb.c @@ -5,7 +5,6 @@ */ #include -#include #include #define MOD_DESC "CAN device driver interface" @@ -337,8 +336,6 @@ static bool can_skb_headroom_valid(struct net_device *dev, struct sk_buff *skb) /* Drop a given socketbuffer if it does not contain a valid CAN frame. */ bool can_dropped_invalid_skb(struct net_device *dev, struct sk_buff *skb) { - struct can_priv *priv = netdev_priv(dev); - switch (ntohs(skb->protocol)) { case ETH_P_CAN: if (!can_is_can_skb(skb)) @@ -359,13 +356,8 @@ bool can_dropped_invalid_skb(struct net_device *dev, struct sk_buff *skb) goto inval_skb; } - if (!can_skb_headroom_valid(dev, skb)) { + if (!can_skb_headroom_valid(dev, skb)) goto inval_skb; - } else if (priv->ctrlmode & CAN_CTRLMODE_LISTENONLY) { - netdev_info_once(dev, - "interface in listen only mode, dropping skb\n"); - goto inval_skb; - } return false; diff --git a/drivers/net/can/flexcan/flexcan-core.c b/drivers/net/can/flexcan/flexcan-core.c index 5ee38e586fd8..9bdadd716f4e 100644 --- a/drivers/net/can/flexcan/flexcan-core.c +++ b/drivers/net/can/flexcan/flexcan-core.c @@ -742,7 +742,7 @@ static netdev_tx_t flexcan_start_xmit(struct sk_buff *skb, struct net_device *de u32 ctrl = FLEXCAN_MB_CODE_TX_DATA | ((can_fd_len2dlc(cfd->len)) << 16); int i; - if (can_dropped_invalid_skb(dev, skb)) + if (can_dev_dropped_skb(dev, skb)) return NETDEV_TX_OK; netif_stop_queue(dev); diff --git a/drivers/net/can/grcan.c b/drivers/net/can/grcan.c index 6c37aab93eb3..4bedcc3eea0d 100644 --- a/drivers/net/can/grcan.c +++ b/drivers/net/can/grcan.c @@ -1345,7 +1345,7 @@ static netdev_tx_t grcan_start_xmit(struct sk_buff *skb, unsigned long flags; u32 oneshotmode = priv->can.ctrlmode & CAN_CTRLMODE_ONE_SHOT; - if (can_dropped_invalid_skb(dev, skb)) + if (can_dev_dropped_skb(dev, skb)) return NETDEV_TX_OK; /* Trying to transmit in silent mode will generate error interrupts, but diff --git a/drivers/net/can/ifi_canfd/ifi_canfd.c b/drivers/net/can/ifi_canfd/ifi_canfd.c index 8d42b7e6661f..07eaf724a572 100644 --- a/drivers/net/can/ifi_canfd/ifi_canfd.c +++ b/drivers/net/can/ifi_canfd/ifi_canfd.c @@ -860,7 +860,7 @@ static netdev_tx_t ifi_canfd_start_xmit(struct sk_buff *skb, u32 txst, txid, txdlc; int i; - if (can_dropped_invalid_skb(ndev, skb)) + if (can_dev_dropped_skb(ndev, skb)) return NETDEV_TX_OK; /* Check if the TX buffer is full */ diff --git a/drivers/net/can/janz-ican3.c b/drivers/net/can/janz-ican3.c index 71a2caae0757..0732a5092141 100644 --- a/drivers/net/can/janz-ican3.c +++ b/drivers/net/can/janz-ican3.c @@ -1693,7 +1693,7 @@ static netdev_tx_t ican3_xmit(struct sk_buff *skb, struct net_device *ndev) void __iomem *desc_addr; unsigned long flags; - if (can_dropped_invalid_skb(ndev, skb)) + if (can_dev_dropped_skb(ndev, skb)) return NETDEV_TX_OK; spin_lock_irqsave(&mod->lock, flags); diff --git a/drivers/net/can/kvaser_pciefd.c b/drivers/net/can/kvaser_pciefd.c index 4e9680c8eb34..bcad11709bc9 100644 --- a/drivers/net/can/kvaser_pciefd.c +++ b/drivers/net/can/kvaser_pciefd.c @@ -772,7 +772,7 @@ static netdev_tx_t kvaser_pciefd_start_xmit(struct sk_buff *skb, int nwords; u8 count; - if (can_dropped_invalid_skb(netdev, skb)) + if (can_dev_dropped_skb(netdev, skb)) return NETDEV_TX_OK; nwords = kvaser_pciefd_prepare_tx_packet(&packet, can, skb); diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c index dcb582563d5e..00d11e95fd98 100644 --- a/drivers/net/can/m_can/m_can.c +++ b/drivers/net/can/m_can/m_can.c @@ -1721,7 +1721,7 @@ static netdev_tx_t m_can_start_xmit(struct sk_buff *skb, { struct m_can_classdev *cdev = netdev_priv(dev); - if (can_dropped_invalid_skb(dev, skb)) + if (can_dev_dropped_skb(dev, skb)) return NETDEV_TX_OK; if (cdev->is_peripheral) { diff --git a/drivers/net/can/mscan/mscan.c b/drivers/net/can/mscan/mscan.c index 2119fbb287ef..a6829cdc0e81 100644 --- a/drivers/net/can/mscan/mscan.c +++ b/drivers/net/can/mscan/mscan.c @@ -191,7 +191,7 @@ static netdev_tx_t mscan_start_xmit(struct sk_buff *skb, struct net_device *dev) int i, rtr, buf_id; u32 can_id; - if (can_dropped_invalid_skb(dev, skb)) + if (can_dev_dropped_skb(dev, skb)) return NETDEV_TX_OK; out_8(®s->cantier, 0); diff --git a/drivers/net/can/pch_can.c b/drivers/net/can/pch_can.c index 0558ff67ec6a..2a44b2803e55 100644 --- a/drivers/net/can/pch_can.c +++ b/drivers/net/can/pch_can.c @@ -882,7 +882,7 @@ static netdev_tx_t pch_xmit(struct sk_buff *skb, struct net_device *ndev) int i; u32 id2; - if (can_dropped_invalid_skb(ndev, skb)) + if (can_dev_dropped_skb(ndev, skb)) return NETDEV_TX_OK; tx_obj_no = priv->tx_obj; diff --git a/drivers/net/can/peak_canfd/peak_canfd.c b/drivers/net/can/peak_canfd/peak_canfd.c index f8420cc1d907..31c9c127e24b 100644 --- a/drivers/net/can/peak_canfd/peak_canfd.c +++ b/drivers/net/can/peak_canfd/peak_canfd.c @@ -651,7 +651,7 @@ static netdev_tx_t peak_canfd_start_xmit(struct sk_buff *skb, int room_left; u8 len; - if (can_dropped_invalid_skb(ndev, skb)) + if (can_dev_dropped_skb(ndev, skb)) return NETDEV_TX_OK; msg_size = ALIGN(sizeof(*msg) + cf->len, 4); diff --git a/drivers/net/can/rcar/rcar_can.c b/drivers/net/can/rcar/rcar_can.c index 6ee968c59ac9..cc43c9c5e38c 100644 --- a/drivers/net/can/rcar/rcar_can.c +++ b/drivers/net/can/rcar/rcar_can.c @@ -590,7 +590,7 @@ static netdev_tx_t rcar_can_start_xmit(struct sk_buff *skb, struct can_frame *cf = (struct can_frame *)skb->data; u32 data, i; - if (can_dropped_invalid_skb(ndev, skb)) + if (can_dev_dropped_skb(ndev, skb)) return NETDEV_TX_OK; if (cf->can_id & CAN_EFF_FLAG) /* Extended frame format */ diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c index 198da643ee6d..d530e986f7fa 100644 --- a/drivers/net/can/rcar/rcar_canfd.c +++ b/drivers/net/can/rcar/rcar_canfd.c @@ -1481,7 +1481,7 @@ static netdev_tx_t rcar_canfd_start_xmit(struct sk_buff *skb, unsigned long flags; u32 ch = priv->channel; - if (can_dropped_invalid_skb(ndev, skb)) + if (can_dev_dropped_skb(ndev, skb)) return NETDEV_TX_OK; if (cf->can_id & CAN_EFF_FLAG) { diff --git a/drivers/net/can/sja1000/sja1000.c b/drivers/net/can/sja1000/sja1000.c index 1bb1129b0450..aac5956e4a53 100644 --- a/drivers/net/can/sja1000/sja1000.c +++ b/drivers/net/can/sja1000/sja1000.c @@ -291,7 +291,7 @@ static netdev_tx_t sja1000_start_xmit(struct sk_buff *skb, u8 cmd_reg_val = 0x00; int i; - if (can_dropped_invalid_skb(dev, skb)) + if (can_dev_dropped_skb(dev, skb)) return NETDEV_TX_OK; netif_stop_queue(dev); diff --git a/drivers/net/can/slcan/slcan-core.c b/drivers/net/can/slcan/slcan-core.c index 8d13fdf8c28a..fbb34139daa1 100644 --- a/drivers/net/can/slcan/slcan-core.c +++ b/drivers/net/can/slcan/slcan-core.c @@ -594,7 +594,7 @@ static netdev_tx_t slcan_netdev_xmit(struct sk_buff *skb, { struct slcan *sl = netdev_priv(dev); - if (can_dropped_invalid_skb(dev, skb)) + if (can_dev_dropped_skb(dev, skb)) return NETDEV_TX_OK; spin_lock(&sl->lock); diff --git a/drivers/net/can/softing/softing_main.c b/drivers/net/can/softing/softing_main.c index a5ef57f415f7..c72f505d29fe 100644 --- a/drivers/net/can/softing/softing_main.c +++ b/drivers/net/can/softing/softing_main.c @@ -60,7 +60,7 @@ static netdev_tx_t softing_netdev_start_xmit(struct sk_buff *skb, struct can_frame *cf = (struct can_frame *)skb->data; uint8_t buf[DPRAM_TX_SIZE]; - if (can_dropped_invalid_skb(dev, skb)) + if (can_dev_dropped_skb(dev, skb)) return NETDEV_TX_OK; spin_lock(&card->spin); diff --git a/drivers/net/can/spi/hi311x.c b/drivers/net/can/spi/hi311x.c index b87dc420428d..e1b8533a602e 100644 --- a/drivers/net/can/spi/hi311x.c +++ b/drivers/net/can/spi/hi311x.c @@ -373,7 +373,7 @@ static netdev_tx_t hi3110_hard_start_xmit(struct sk_buff *skb, return NETDEV_TX_BUSY; } - if (can_dropped_invalid_skb(net, skb)) + if (can_dev_dropped_skb(net, skb)) return NETDEV_TX_OK; netif_stop_queue(net); diff --git a/drivers/net/can/spi/mcp251x.c b/drivers/net/can/spi/mcp251x.c index 24883a65ca66..79c4bab5f724 100644 --- a/drivers/net/can/spi/mcp251x.c +++ b/drivers/net/can/spi/mcp251x.c @@ -789,7 +789,7 @@ static netdev_tx_t mcp251x_hard_start_xmit(struct sk_buff *skb, return NETDEV_TX_BUSY; } - if (can_dropped_invalid_skb(net, skb)) + if (can_dev_dropped_skb(net, skb)) return NETDEV_TX_OK; netif_stop_queue(net); diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-tx.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-tx.c index ffb6c36b7d9b..160528d3cc26 100644 --- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-tx.c +++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-tx.c @@ -172,7 +172,7 @@ netdev_tx_t mcp251xfd_start_xmit(struct sk_buff *skb, u8 tx_head; int err; - if (can_dropped_invalid_skb(ndev, skb)) + if (can_dev_dropped_skb(ndev, skb)) return NETDEV_TX_OK; if (mcp251xfd_tx_busy(priv, tx_ring)) diff --git a/drivers/net/can/sun4i_can.c b/drivers/net/can/sun4i_can.c index 525309da1320..2b78f9197681 100644 --- a/drivers/net/can/sun4i_can.c +++ b/drivers/net/can/sun4i_can.c @@ -429,7 +429,7 @@ static netdev_tx_t sun4ican_start_xmit(struct sk_buff *skb, struct net_device *d canid_t id; int i; - if (can_dropped_invalid_skb(dev, skb)) + if (can_dev_dropped_skb(dev, skb)) return NETDEV_TX_OK; netif_stop_queue(dev); diff --git a/drivers/net/can/ti_hecc.c b/drivers/net/can/ti_hecc.c index b218fb3c6b76..27700f72eac2 100644 --- a/drivers/net/can/ti_hecc.c +++ b/drivers/net/can/ti_hecc.c @@ -470,7 +470,7 @@ static netdev_tx_t ti_hecc_xmit(struct sk_buff *skb, struct net_device *ndev) u32 mbxno, mbx_mask, data; unsigned long flags; - if (can_dropped_invalid_skb(ndev, skb)) + if (can_dev_dropped_skb(ndev, skb)) return NETDEV_TX_OK; mbxno = get_tx_head_mb(priv); diff --git a/drivers/net/can/usb/ems_usb.c b/drivers/net/can/usb/ems_usb.c index d31191686a54..050c0b49938a 100644 --- a/drivers/net/can/usb/ems_usb.c +++ b/drivers/net/can/usb/ems_usb.c @@ -747,7 +747,7 @@ static netdev_tx_t ems_usb_start_xmit(struct sk_buff *skb, struct net_device *ne size_t size = CPC_HEADER_SIZE + CPC_MSG_HEADER_LEN + sizeof(struct cpc_can_msg); - if (can_dropped_invalid_skb(netdev, skb)) + if (can_dev_dropped_skb(netdev, skb)) return NETDEV_TX_OK; /* create a URB, and a buffer for it, and copy the data to the URB */ diff --git a/drivers/net/can/usb/esd_usb.c b/drivers/net/can/usb/esd_usb.c index 1bcfad11b1e4..81b88e9e5bdc 100644 --- a/drivers/net/can/usb/esd_usb.c +++ b/drivers/net/can/usb/esd_usb.c @@ -725,7 +725,7 @@ static netdev_tx_t esd_usb_start_xmit(struct sk_buff *skb, int ret = NETDEV_TX_OK; size_t size = sizeof(struct esd_usb_msg); - if (can_dropped_invalid_skb(netdev, skb)) + if (can_dev_dropped_skb(netdev, skb)) return NETDEV_TX_OK; /* create a URB, and a buffer for it, and copy the data to the URB */ diff --git a/drivers/net/can/usb/etas_es58x/es58x_core.c b/drivers/net/can/usb/etas_es58x/es58x_core.c index 51294b717040..25f863b4f5f0 100644 --- a/drivers/net/can/usb/etas_es58x/es58x_core.c +++ b/drivers/net/can/usb/etas_es58x/es58x_core.c @@ -1913,7 +1913,7 @@ static netdev_tx_t es58x_start_xmit(struct sk_buff *skb, unsigned int frame_len; int ret; - if (can_dropped_invalid_skb(netdev, skb)) { + if (can_dev_dropped_skb(netdev, skb)) { if (priv->tx_urb) goto xmit_commit; return NETDEV_TX_OK; diff --git a/drivers/net/can/usb/gs_usb.c b/drivers/net/can/usb/gs_usb.c index f0065d40eb24..9c2c25fde3d1 100644 --- a/drivers/net/can/usb/gs_usb.c +++ b/drivers/net/can/usb/gs_usb.c @@ -723,7 +723,7 @@ static netdev_tx_t gs_can_start_xmit(struct sk_buff *skb, unsigned int idx; struct gs_tx_context *txc; - if (can_dropped_invalid_skb(netdev, skb)) + if (can_dev_dropped_skb(netdev, skb)) return NETDEV_TX_OK; /* find an empty context to keep track of transmission */ diff --git a/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c b/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c index e91648ed7386..802e27c0eced 100644 --- a/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c +++ b/drivers/net/can/usb/kvaser_usb/kvaser_usb_core.c @@ -570,7 +570,7 @@ static netdev_tx_t kvaser_usb_start_xmit(struct sk_buff *skb, unsigned int i; unsigned long flags; - if (can_dropped_invalid_skb(netdev, skb)) + if (can_dev_dropped_skb(netdev, skb)) return NETDEV_TX_OK; urb = usb_alloc_urb(0, GFP_ATOMIC); diff --git a/drivers/net/can/usb/mcba_usb.c b/drivers/net/can/usb/mcba_usb.c index 69346c63021f..218b098b261d 100644 --- a/drivers/net/can/usb/mcba_usb.c +++ b/drivers/net/can/usb/mcba_usb.c @@ -311,7 +311,7 @@ static netdev_tx_t mcba_usb_start_xmit(struct sk_buff *skb, .cmd_id = MBCA_CMD_TRANSMIT_MESSAGE_EV }; - if (can_dropped_invalid_skb(netdev, skb)) + if (can_dev_dropped_skb(netdev, skb)) return NETDEV_TX_OK; ctx = mcba_usb_get_free_ctx(priv, cf); diff --git a/drivers/net/can/usb/peak_usb/pcan_usb_core.c b/drivers/net/can/usb/peak_usb/pcan_usb_core.c index 225697d70a9a..1d996d3320fe 100644 --- a/drivers/net/can/usb/peak_usb/pcan_usb_core.c +++ b/drivers/net/can/usb/peak_usb/pcan_usb_core.c @@ -351,7 +351,7 @@ static netdev_tx_t peak_usb_ndo_start_xmit(struct sk_buff *skb, int i, err; size_t size = dev->adapter->tx_buffer_size; - if (can_dropped_invalid_skb(netdev, skb)) + if (can_dev_dropped_skb(netdev, skb)) return NETDEV_TX_OK; for (i = 0; i < PCAN_USB_MAX_TX_URBS; i++) diff --git a/drivers/net/can/usb/ucan.c b/drivers/net/can/usb/ucan.c index 7c35f50fda4e..67c2ff407d06 100644 --- a/drivers/net/can/usb/ucan.c +++ b/drivers/net/can/usb/ucan.c @@ -1120,7 +1120,7 @@ static netdev_tx_t ucan_start_xmit(struct sk_buff *skb, struct can_frame *cf = (struct can_frame *)skb->data; /* check skb */ - if (can_dropped_invalid_skb(netdev, skb)) + if (can_dev_dropped_skb(netdev, skb)) return NETDEV_TX_OK; /* allocate a context and slow down tx path, if fifo state is low */ diff --git a/drivers/net/can/usb/usb_8dev.c b/drivers/net/can/usb/usb_8dev.c index 64c00abe91cf..8a5596ce4e46 100644 --- a/drivers/net/can/usb/usb_8dev.c +++ b/drivers/net/can/usb/usb_8dev.c @@ -602,7 +602,7 @@ static netdev_tx_t usb_8dev_start_xmit(struct sk_buff *skb, int i, err; size_t size = sizeof(struct usb_8dev_tx_msg); - if (can_dropped_invalid_skb(netdev, skb)) + if (can_dev_dropped_skb(netdev, skb)) return NETDEV_TX_OK; /* create a URB, and a buffer for it, and copy the data to the URB */ diff --git a/drivers/net/can/xilinx_can.c b/drivers/net/can/xilinx_can.c index 5d3172795ad0..43c812ea1de0 100644 --- a/drivers/net/can/xilinx_can.c +++ b/drivers/net/can/xilinx_can.c @@ -743,7 +743,7 @@ static netdev_tx_t xcan_start_xmit(struct sk_buff *skb, struct net_device *ndev) struct xcan_priv *priv = netdev_priv(ndev); int ret; - if (can_dropped_invalid_skb(ndev, skb)) + if (can_dev_dropped_skb(ndev, skb)) return NETDEV_TX_OK; if (priv->devtype.flags & XCAN_FLAG_TX_MAILBOXES) diff --git a/include/linux/can/dev.h b/include/linux/can/dev.h index 58f5431a5559..982ba245eb41 100644 --- a/include/linux/can/dev.h +++ b/include/linux/can/dev.h @@ -152,6 +152,22 @@ static inline bool can_is_canxl_dev_mtu(unsigned int mtu) return (mtu >= CANXL_MIN_MTU && mtu <= CANXL_MAX_MTU); } +/* drop skb if it does not contain a valid CAN frame for sending */ +static inline bool can_dev_dropped_skb(struct net_device *dev, struct sk_buff *skb) +{ + struct can_priv *priv = netdev_priv(dev); + + if (priv->ctrlmode & CAN_CTRLMODE_LISTENONLY) { + netdev_info_once(dev, + "interface in listen only mode, dropping skb\n"); + kfree_skb(skb); + dev->stats.tx_dropped++; + return true; + } + + return can_dropped_invalid_skb(dev, skb); +} + void can_setup(struct net_device *dev); struct net_device *alloc_candev_mqs(int sizeof_priv, unsigned int echo_skb_max, From 8b043dfb3dc7c32f9c2c0c93e3c2de346ee5e358 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Fri, 28 Oct 2022 12:06:45 +0200 Subject: [PATCH 079/138] can: rcar_canfd: Add missing ECC error checks for channels 2-7 When introducing support for R-Car V3U, which has 8 instead of 2 channels, the ECC error bitmask was extended to take into account the extra channels, but rcar_canfd_global_error() was not updated to act upon the extra bits. Replace the RCANFD_GERFL_EEF[01] macros by a new macro that takes the channel number, fixing R-Car V3U while simplifying the code. Fixes: 45721c406dcf50d4 ("can: rcar_canfd: Add support for r8a779a0 SoC") Signed-off-by: Geert Uytterhoeven Reviewed-by: Biju Das Link: https://lore.kernel.org/all/4edb2ea46cc64d0532a08a924179827481e14b4f.1666951503.git.geert+renesas@glider.be Cc: stable@vger.kernel.org Signed-off-by: Marc Kleine-Budde --- drivers/net/can/rcar/rcar_canfd.c | 13 ++++--------- 1 file changed, 4 insertions(+), 9 deletions(-) diff --git a/drivers/net/can/rcar/rcar_canfd.c b/drivers/net/can/rcar/rcar_canfd.c index d530e986f7fa..b306cf554634 100644 --- a/drivers/net/can/rcar/rcar_canfd.c +++ b/drivers/net/can/rcar/rcar_canfd.c @@ -81,8 +81,7 @@ enum rcanfd_chip_id { /* RSCFDnCFDGERFL / RSCFDnGERFL */ #define RCANFD_GERFL_EEF0_7 GENMASK(23, 16) -#define RCANFD_GERFL_EEF1 BIT(17) -#define RCANFD_GERFL_EEF0 BIT(16) +#define RCANFD_GERFL_EEF(ch) BIT(16 + (ch)) #define RCANFD_GERFL_CMPOF BIT(3) /* CAN FD only */ #define RCANFD_GERFL_THLES BIT(2) #define RCANFD_GERFL_MES BIT(1) @@ -90,7 +89,7 @@ enum rcanfd_chip_id { #define RCANFD_GERFL_ERR(gpriv, x) \ ((x) & (reg_v3u(gpriv, RCANFD_GERFL_EEF0_7, \ - RCANFD_GERFL_EEF0 | RCANFD_GERFL_EEF1) | \ + RCANFD_GERFL_EEF(0) | RCANFD_GERFL_EEF(1)) | \ RCANFD_GERFL_MES | \ ((gpriv)->fdmode ? RCANFD_GERFL_CMPOF : 0))) @@ -936,12 +935,8 @@ static void rcar_canfd_global_error(struct net_device *ndev) u32 ridx = ch + RCANFD_RFFIFO_IDX; gerfl = rcar_canfd_read(priv->base, RCANFD_GERFL); - if ((gerfl & RCANFD_GERFL_EEF0) && (ch == 0)) { - netdev_dbg(ndev, "Ch0: ECC Error flag\n"); - stats->tx_dropped++; - } - if ((gerfl & RCANFD_GERFL_EEF1) && (ch == 1)) { - netdev_dbg(ndev, "Ch1: ECC Error flag\n"); + if (gerfl & RCANFD_GERFL_EEF(ch)) { + netdev_dbg(ndev, "Ch%u: ECC Error flag\n", ch); stats->tx_dropped++; } if (gerfl & RCANFD_GERFL_MES) { From 0fca385d6ebc3cabb20f67bcf8a71f1448bdc001 Mon Sep 17 00:00:00 2001 From: Liu Shixin Date: Thu, 3 Nov 2022 16:33:01 +0800 Subject: [PATCH 080/138] btrfs: fix match incorrectly in dev_args_match_device syzkaller found a failed assertion: assertion failed: (args->devid != (u64)-1) || args->missing, in fs/btrfs/volumes.c:6921 This can be triggered when we set devid to (u64)-1 by ioctl. In this case, the match of devid will be skipped and the match of device may succeed incorrectly. Patch 562d7b1512f7 introduced this function which is used to match device. This function contains two matching scenarios, we can distinguish them by checking the value of args->missing rather than check whether args->devid and args->uuid is default value. Reported-by: syzbot+031687116258450f9853@syzkaller.appspotmail.com Fixes: 562d7b1512f7 ("btrfs: handle device lookup with btrfs_dev_lookup_args") CC: stable@vger.kernel.org # 5.16+ Reviewed-by: Nikolay Borisov Signed-off-by: Liu Shixin Signed-off-by: David Sterba --- fs/btrfs/volumes.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index a8d4bc6a1937..f09d09c259f5 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -6918,18 +6918,18 @@ static bool dev_args_match_fs_devices(const struct btrfs_dev_lookup_args *args, static bool dev_args_match_device(const struct btrfs_dev_lookup_args *args, const struct btrfs_device *device) { - ASSERT((args->devid != (u64)-1) || args->missing); + if (args->missing) { + if (test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state) && + !device->bdev) + return true; + return false; + } - if ((args->devid != (u64)-1) && device->devid != args->devid) + if (device->devid != args->devid) return false; if (args->uuid && memcmp(device->uuid, args->uuid, BTRFS_UUID_SIZE) != 0) return false; - if (!args->missing) - return true; - if (test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state) && - !device->bdev) - return true; - return false; + return true; } /* From 9b2f20344d450137d015b380ff0c2e2a6a170135 Mon Sep 17 00:00:00 2001 From: Zhang Xiaoxu Date: Tue, 1 Nov 2022 10:53:54 +0800 Subject: [PATCH 081/138] btrfs: selftests: fix wrong error check in btrfs_free_dummy_root() The btrfs_alloc_dummy_root() uses ERR_PTR as the error return value rather than NULL, if error happened, there will be a NULL pointer dereference: BUG: KASAN: null-ptr-deref in btrfs_free_dummy_root+0x21/0x50 [btrfs] Read of size 8 at addr 000000000000002c by task insmod/258926 CPU: 2 PID: 258926 Comm: insmod Tainted: G W 6.1.0-rc2+ #5 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.14.0-1.fc33 04/01/2014 Call Trace: dump_stack_lvl+0x34/0x44 kasan_report+0xb7/0x140 kasan_check_range+0x145/0x1a0 btrfs_free_dummy_root+0x21/0x50 [btrfs] btrfs_test_free_space_cache+0x1a8c/0x1add [btrfs] btrfs_run_sanity_tests+0x65/0x80 [btrfs] init_btrfs_fs+0xec/0x154 [btrfs] do_one_initcall+0x87/0x2a0 do_init_module+0xdf/0x320 load_module+0x3006/0x3390 __do_sys_finit_module+0x113/0x1b0 do_syscall_64+0x35/0x80 entry_SYSCALL_64_after_hwframe+0x46/0xb0 Fixes: aaedb55bc08f ("Btrfs: add tests for btrfs_get_extent") CC: stable@vger.kernel.org # 4.9+ Reviewed-by: Anand Jain Signed-off-by: Zhang Xiaoxu Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/tests/btrfs-tests.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/btrfs/tests/btrfs-tests.c b/fs/btrfs/tests/btrfs-tests.c index 9c478fa256f6..d43cb5242fec 100644 --- a/fs/btrfs/tests/btrfs-tests.c +++ b/fs/btrfs/tests/btrfs-tests.c @@ -200,7 +200,7 @@ void btrfs_free_dummy_fs_info(struct btrfs_fs_info *fs_info) void btrfs_free_dummy_root(struct btrfs_root *root) { - if (!root) + if (IS_ERR_OR_NULL(root)) return; /* Will be freed by btrfs_free_fs_roots */ if (WARN_ON(test_bit(BTRFS_ROOT_IN_RADIX, &root->state))) From 8bb808c6ad91ec3d332f072ce8f8aa4b16e307e0 Mon Sep 17 00:00:00 2001 From: David Sterba Date: Thu, 3 Nov 2022 14:39:01 +0100 Subject: [PATCH 082/138] btrfs: don't print stack trace when transaction is aborted due to ENOMEM Add ENOMEM among the error codes that don't print stack trace on transaction abort. We've got several reports from syzbot that detects stacks as errors but caused by limiting memory. As this is an artificial condition we don't need to know where exactly the error happens, the abort and error cleanup will continue like e.g. for EIO. As the transaction aborts code needs to be inline in a lot of code, the implementation cases about minimal bloat. The error codes are in a separate function and the WARN uses the condition directly. This increases the code size by 571 bytes on release build. Alternatives considered: add -ENOMEM among the errors, this increases size by 2340 bytes, various attempts to combine the WARN and helper calls, increase by 700 or more bytes. Example syzbot reports (error -12): - https://syzkaller.appspot.com/bug?extid=5244d35be7f589cf093e - https://syzkaller.appspot.com/bug?extid=9c37714c07194d816417 Signed-off-by: David Sterba --- fs/btrfs/ctree.c | 16 ++++++++++++++++ fs/btrfs/ctree.h | 11 +++++++---- 2 files changed, 23 insertions(+), 4 deletions(-) diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index b39b339fbf96..a9543f01184c 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -113,6 +113,22 @@ noinline void btrfs_release_path(struct btrfs_path *p) } } +/* + * We want the transaction abort to print stack trace only for errors where the + * cause could be a bug, eg. due to ENOSPC, and not for common errors that are + * caused by external factors. + */ +bool __cold abort_should_print_stack(int errno) +{ + switch (errno) { + case -EIO: + case -EROFS: + case -ENOMEM: + return false; + } + return true; +} + /* * safely gets a reference on the root node of a tree. A lock * is not taken, so a concurrent writer may put a different node diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index f677b49df8ae..9e6d48ff4597 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -3796,9 +3796,11 @@ void __btrfs_abort_transaction(struct btrfs_trans_handle *trans, const char *function, unsigned int line, int errno, bool first_hit); +bool __cold abort_should_print_stack(int errno); + /* * Call btrfs_abort_transaction as early as possible when an error condition is - * detected, that way the exact line number is reported. + * detected, that way the exact stack trace is reported for some errors. */ #define btrfs_abort_transaction(trans, errno) \ do { \ @@ -3807,10 +3809,11 @@ do { \ if (!test_and_set_bit(BTRFS_FS_STATE_TRANS_ABORTED, \ &((trans)->fs_info->fs_state))) { \ first = true; \ - if ((errno) != -EIO && (errno) != -EROFS) { \ - WARN(1, KERN_DEBUG \ + if (WARN(abort_should_print_stack(errno), \ + KERN_DEBUG \ "BTRFS: Transaction aborted (error %d)\n", \ - (errno)); \ + (errno))) { \ + /* Stack trace printed. */ \ } else { \ btrfs_debug((trans)->fs_info, \ "Transaction aborted (error %d)", \ From b75b51f886e9dd8cdfca1392ad43f4e542611c00 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Mon, 7 Nov 2022 07:23:26 +0800 Subject: [PATCH 083/138] Revert "btrfs: scrub: use larger block size for data extent scrub" This reverts commit 786672e9e1a39a231806313e3c445c236588ceef. [BUG] Since commit 786672e9e1a3 ("btrfs: scrub: use larger block size for data extent scrub"), btrfs scrub no longer reports errors if the corruption is not in the first sector of a STRIPE_LEN. The following script can expose the problem: mkfs.btrfs -f $dev mount $dev $mnt xfs_io -f -c "pwrite -S 0xff 0 8k" $mnt/foobar umount $mnt # 13631488 is the logical bytenr of above 8K extent btrfs-map-logical -l 13631488 -b 4096 $dev mirror 1 logical 13631488 physical 13631488 device /dev/test/scratch1 # Corrupt the 2nd sector of that extent xfs_io -f -c "pwrite -S 0x00 13635584 4k" $dev mount $dev $mnt btrfs scrub start -B $mnt scrub done for 54e63f9f-0c30-4c84-a33b-5c56014629b7 Scrub started: Mon Nov 7 07:18:27 2022 Status: finished Duration: 0:00:00 Total to scrub: 536.00MiB Rate: 0.00B/s Error summary: no errors found <<< [CAUSE] That offending commit enlarges the data extent scrub size from sector size to BTRFS_STRIPE_LEN, to avoid extra scrub_block to be allocated. But unfortunately the data extent scrub is still heavily relying on the fact that there is only one scrub_sector per scrub_block. Thus it will only check the first sector, and ignoring the remaining sectors. Furthermore the error reporting is not able to handle multiple sectors either. [FIX] For now just revert the offending commit. The consequence is just extra memory usage during scrub. We will need a proper change to make the remaining data scrub path to handle multiple sectors before we enlarging the data scrub size. Reported-by: Li Zhang Signed-off-by: Qu Wenruo Signed-off-by: David Sterba --- fs/btrfs/scrub.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index f260c53829e5..b659c67af1e0 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -2672,17 +2672,11 @@ static int scrub_extent(struct scrub_ctx *sctx, struct map_lookup *map, u8 csum[BTRFS_CSUM_SIZE]; u32 blocksize; - /* - * Block size determines how many scrub_block will be allocated. Here - * we use BTRFS_STRIPE_LEN (64KiB) as default limit, so we won't - * allocate too many scrub_block, while still won't cause too large - * bios for large extents. - */ if (flags & BTRFS_EXTENT_FLAG_DATA) { if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) blocksize = map->stripe_len; else - blocksize = BTRFS_STRIPE_LEN; + blocksize = sctx->fs_info->sectorsize; spin_lock(&sctx->stat_lock); sctx->stat.data_extents_scrubbed++; sctx->stat.data_bytes_scrubbed += len; From 21e61ec6d0bb786818490e926aa9aeb4de95ad0d Mon Sep 17 00:00:00 2001 From: Johannes Thumshirn Date: Fri, 4 Nov 2022 07:12:33 -0700 Subject: [PATCH 084/138] btrfs: zoned: clone zoned device info when cloning a device When cloning a btrfs_device, we're not cloning the associated btrfs_zoned_device_info structure of the device in case of a zoned filesystem. Later on this leads to a NULL pointer dereference when accessing the device's zone_info for instance when setting a zone as active. This was uncovered by fstests' testcase btrfs/161. CC: stable@vger.kernel.org # 5.15+ Signed-off-by: Johannes Thumshirn Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/volumes.c | 12 ++++++++++++ fs/btrfs/zoned.c | 40 ++++++++++++++++++++++++++++++++++++++++ fs/btrfs/zoned.h | 11 +++++++++++ 3 files changed, 63 insertions(+) diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index f09d09c259f5..3cb968ede675 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -1011,6 +1011,18 @@ static struct btrfs_fs_devices *clone_fs_devices(struct btrfs_fs_devices *orig) rcu_assign_pointer(device->name, name); } + if (orig_dev->zone_info) { + struct btrfs_zoned_device_info *zone_info; + + zone_info = btrfs_clone_dev_zone_info(orig_dev); + if (!zone_info) { + btrfs_free_device(device); + ret = -ENOMEM; + goto error; + } + device->zone_info = zone_info; + } + list_add(&device->dev_list, &fs_devices->devices); device->fs_devices = fs_devices; fs_devices->num_devices++; diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c index e2d073b08a7d..1912abf6d020 100644 --- a/fs/btrfs/zoned.c +++ b/fs/btrfs/zoned.c @@ -639,6 +639,46 @@ void btrfs_destroy_dev_zone_info(struct btrfs_device *device) device->zone_info = NULL; } +struct btrfs_zoned_device_info *btrfs_clone_dev_zone_info(struct btrfs_device *orig_dev) +{ + struct btrfs_zoned_device_info *zone_info; + + zone_info = kmemdup(orig_dev->zone_info, sizeof(*zone_info), GFP_KERNEL); + if (!zone_info) + return NULL; + + zone_info->seq_zones = bitmap_zalloc(zone_info->nr_zones, GFP_KERNEL); + if (!zone_info->seq_zones) + goto out; + + bitmap_copy(zone_info->seq_zones, orig_dev->zone_info->seq_zones, + zone_info->nr_zones); + + zone_info->empty_zones = bitmap_zalloc(zone_info->nr_zones, GFP_KERNEL); + if (!zone_info->empty_zones) + goto out; + + bitmap_copy(zone_info->empty_zones, orig_dev->zone_info->empty_zones, + zone_info->nr_zones); + + zone_info->active_zones = bitmap_zalloc(zone_info->nr_zones, GFP_KERNEL); + if (!zone_info->active_zones) + goto out; + + bitmap_copy(zone_info->active_zones, orig_dev->zone_info->active_zones, + zone_info->nr_zones); + zone_info->zone_cache = NULL; + + return zone_info; + +out: + bitmap_free(zone_info->seq_zones); + bitmap_free(zone_info->empty_zones); + bitmap_free(zone_info->active_zones); + kfree(zone_info); + return NULL; +} + int btrfs_get_dev_zone(struct btrfs_device *device, u64 pos, struct blk_zone *zone) { diff --git a/fs/btrfs/zoned.h b/fs/btrfs/zoned.h index e17462db3a84..8bd16d40b7c6 100644 --- a/fs/btrfs/zoned.h +++ b/fs/btrfs/zoned.h @@ -36,6 +36,7 @@ int btrfs_get_dev_zone(struct btrfs_device *device, u64 pos, int btrfs_get_dev_zone_info_all_devices(struct btrfs_fs_info *fs_info); int btrfs_get_dev_zone_info(struct btrfs_device *device, bool populate_cache); void btrfs_destroy_dev_zone_info(struct btrfs_device *device); +struct btrfs_zoned_device_info *btrfs_clone_dev_zone_info(struct btrfs_device *orig_dev); int btrfs_check_zoned_mode(struct btrfs_fs_info *fs_info); int btrfs_check_mountopts_zoned(struct btrfs_fs_info *info); int btrfs_sb_log_location_bdev(struct block_device *bdev, int mirror, int rw, @@ -103,6 +104,16 @@ static inline int btrfs_get_dev_zone_info(struct btrfs_device *device, static inline void btrfs_destroy_dev_zone_info(struct btrfs_device *device) { } +/* + * In case the kernel is compiled without CONFIG_BLK_DEV_ZONED we'll never call + * into btrfs_clone_dev_zone_info() so it's safe to return NULL here. + */ +static inline struct btrfs_zoned_device_info *btrfs_clone_dev_zone_info( + struct btrfs_device *orig_dev) +{ + return NULL; +} + static inline int btrfs_check_zoned_mode(const struct btrfs_fs_info *fs_info) { if (!btrfs_is_zoned(fs_info)) From a8d1b1647bf8244a5f270538e9e636e2657fffa3 Mon Sep 17 00:00:00 2001 From: Johannes Thumshirn Date: Fri, 4 Nov 2022 07:12:34 -0700 Subject: [PATCH 085/138] btrfs: zoned: initialize device's zone info for seeding When performing seeding on a zoned filesystem it is necessary to initialize each zoned device's btrfs_zoned_device_info structure, otherwise mounting the filesystem will cause a NULL pointer dereference. This was uncovered by fstests' testcase btrfs/163. CC: stable@vger.kernel.org # 5.15+ Signed-off-by: Johannes Thumshirn Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 4 +++- fs/btrfs/volumes.c | 11 +++++++++-- fs/btrfs/volumes.h | 2 +- 3 files changed, 13 insertions(+), 4 deletions(-) diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 4b28263c3d32..d99bf7c64611 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -2551,7 +2551,9 @@ static int btrfs_read_roots(struct btrfs_fs_info *fs_info) fs_info->dev_root = root; } /* Initialize fs_info for all devices in any case */ - btrfs_init_devices_late(fs_info); + ret = btrfs_init_devices_late(fs_info); + if (ret) + goto out; /* * This tree can share blocks with some other fs tree during relocation diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 3cb968ede675..635f45f1a2ef 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -7756,10 +7756,11 @@ int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info) return ret; } -void btrfs_init_devices_late(struct btrfs_fs_info *fs_info) +int btrfs_init_devices_late(struct btrfs_fs_info *fs_info) { struct btrfs_fs_devices *fs_devices = fs_info->fs_devices, *seed_devs; struct btrfs_device *device; + int ret = 0; fs_devices->fs_info = fs_info; @@ -7768,12 +7769,18 @@ void btrfs_init_devices_late(struct btrfs_fs_info *fs_info) device->fs_info = fs_info; list_for_each_entry(seed_devs, &fs_devices->seed_list, seed_list) { - list_for_each_entry(device, &seed_devs->devices, dev_list) + list_for_each_entry(device, &seed_devs->devices, dev_list) { device->fs_info = fs_info; + ret = btrfs_get_dev_zone_info(device, false); + if (ret) + break; + } seed_devs->fs_info = fs_info; } mutex_unlock(&fs_devices->device_list_mutex); + + return ret; } static u64 btrfs_dev_stats_value(const struct extent_buffer *eb, diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index f8b668dc8bf8..099def5613b8 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -671,7 +671,7 @@ int find_free_dev_extent(struct btrfs_device *device, u64 num_bytes, void btrfs_dev_stat_inc_and_print(struct btrfs_device *dev, int index); int btrfs_get_dev_stats(struct btrfs_fs_info *fs_info, struct btrfs_ioctl_get_dev_stats *stats); -void btrfs_init_devices_late(struct btrfs_fs_info *fs_info); +int btrfs_init_devices_late(struct btrfs_fs_info *fs_info); int btrfs_init_dev_stats(struct btrfs_fs_info *fs_info); int btrfs_run_dev_stats(struct btrfs_trans_handle *trans); void btrfs_rm_dev_replace_remove_srcdev(struct btrfs_device *srcdev); From c62f6bec53e63b11112e1ebce6bbaa39ce6f6706 Mon Sep 17 00:00:00 2001 From: Johannes Thumshirn Date: Fri, 4 Nov 2022 07:12:35 -0700 Subject: [PATCH 086/138] btrfs: zoned: fix locking imbalance on scrub If we're doing device replace on a zoned filesystem and discover in scrub_enumerate_chunks() that we don't have to copy the block group it is unlocked before it gets skipped. But as the block group hasn't yet been locked before it leads to a locking imbalance. To fix this simply remove the unlock. This was uncovered by fstests' testcase btrfs/163. Fixes: 9283b9e09a6d ("btrfs: remove lock protection for BLOCK_GROUP_FLAG_TO_COPY") Signed-off-by: Johannes Thumshirn Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/scrub.c | 1 - 1 file changed, 1 deletion(-) diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index b659c67af1e0..196c4c6ed1ed 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -3911,7 +3911,6 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, if (sctx->is_dev_replace && btrfs_is_zoned(fs_info)) { if (!test_bit(BLOCK_GROUP_FLAG_TO_COPY, &cache->runtime_flags)) { - spin_unlock(&cache->lock); btrfs_put_block_group(cache); goto skip; } From c18c20f16219516b12a4f2fd29c25e06be97e064 Mon Sep 17 00:00:00 2001 From: Vlastimil Babka Date: Mon, 7 Nov 2022 17:11:27 +0100 Subject: [PATCH 087/138] mm, slab: remove duplicate kernel-doc comment for ksize() Akira reports: > "make htmldocs" reports duplicate C declaration of ksize() as follows: > /linux/Documentation/core-api/mm-api:43: ./mm/slab_common.c:1428: WARNING: Duplicate C declaration, also defined at core-api/mm-api:212. > Declaration is '.. c:function:: size_t ksize (const void *objp)'. > This is due to the kernel-doc comment for ksize() declaration added in > include/linux/slab.h by commit 05a940656e1e ("slab: Introduce > kmalloc_size_roundup()"). There is an older kernel-doc comment for ksize() definition in mm/slab_common.c, which is not only duplicated, but also contradicts the new one - the additional storage discovered by ksize() should not be used by callers anymore. Delete the old kernel-doc. Reported-by: Akira Yokosawa Link: https://lore.kernel.org/all/d33440f6-40cf-9747-3340-e54ffaf7afb8@gmail.com/ Fixes: 05a940656e1e ("slab: Introduce kmalloc_size_roundup()") Cc: Kees Cook Signed-off-by: Vlastimil Babka --- mm/slab_common.c | 14 -------------- 1 file changed, 14 deletions(-) diff --git a/mm/slab_common.c b/mm/slab_common.c index fa784563e1ed..0042fb2730d1 100644 --- a/mm/slab_common.c +++ b/mm/slab_common.c @@ -1409,20 +1409,6 @@ void kfree_sensitive(const void *p) } EXPORT_SYMBOL(kfree_sensitive); -/** - * ksize - get the actual amount of memory allocated for a given object - * @objp: Pointer to the object - * - * kmalloc may internally round up allocations and return more memory - * than requested. ksize() can be used to determine the actual amount of - * memory allocated. The caller may use this additional memory, even though - * a smaller amount of memory was initially specified with the kmalloc call. - * The caller must guarantee that objp points to a valid object previously - * allocated with either kmalloc() or kmem_cache_alloc(). The object - * must not be freed during the duration of the call. - * - * Return: size of the actual memory used by @objp in bytes - */ size_t ksize(const void *objp) { size_t size; From 59f2f4b8a757412fce372f6d0767bdb55da127a8 Mon Sep 17 00:00:00 2001 From: Liam Howlett Date: Mon, 7 Nov 2022 20:11:42 +0000 Subject: [PATCH 088/138] fs/userfaultfd: Fix maple tree iterator in userfaultfd_unregister() When iterating the VMAs, the maple state needs to be invalidated if the tree is modified by a split or merge to ensure the maple tree node contained in the maple state is still valid. These invalidations were missed, so add them to the paths which alter the tree. Reported-by: syzbot+0d2014e4da2ccced5b41@syzkaller.appspotmail.com Fixes: 69dbe6daf104 (userfaultfd: use maple tree iterator to iterate VMAs) Signed-off-by: Liam R. Howlett Signed-off-by: Linus Torvalds --- fs/userfaultfd.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c index 07c81ab3fd4d..98ac37e34e3d 100644 --- a/fs/userfaultfd.c +++ b/fs/userfaultfd.c @@ -1630,17 +1630,20 @@ static int userfaultfd_unregister(struct userfaultfd_ctx *ctx, NULL_VM_UFFD_CTX, anon_vma_name(vma)); if (prev) { vma = prev; + mas_pause(&mas); goto next; } if (vma->vm_start < start) { ret = split_vma(mm, vma, start, 1); if (ret) break; + mas_pause(&mas); } if (vma->vm_end > end) { ret = split_vma(mm, vma, end, 0); if (ret) break; + mas_pause(&mas); } next: /* From ed4314f7729714d788698ade4f9905ee5378ebc0 Mon Sep 17 00:00:00 2001 From: Rasmus Villemoes Date: Fri, 4 Nov 2022 09:30:04 +0100 Subject: [PATCH 089/138] net: stmmac: dwmac-meson8b: fix meson8b_devm_clk_prepare_enable() There are two problems with meson8b_devm_clk_prepare_enable(), introduced in commit a54dc4a49045 ("net: stmmac: dwmac-meson8b: Make the clock enabling code re-usable"): - It doesn't pass the clk argument, but instead always the rgmii_tx_clk of the device. - It silently ignores the return value of devm_add_action_or_reset(). The former didn't become an actual bug until another user showed up in the next commit 9308c47640d5 ("net: stmmac: dwmac-meson8b: add support for the RX delay configuration"). The latter means the callers could end up with the clock not actually prepared/enabled. Fixes: a54dc4a49045 ("net: stmmac: dwmac-meson8b: Make the clock enabling code re-usable") Signed-off-by: Rasmus Villemoes Reviewed-by: Martin Blumenstingl Link: https://lore.kernel.org/r/20221104083004.2212520-1-linux@rasmusvillemoes.dk Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c index c7a6588d9398..e8b507f88fbc 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c @@ -272,11 +272,9 @@ static int meson8b_devm_clk_prepare_enable(struct meson8b_dwmac *dwmac, if (ret) return ret; - devm_add_action_or_reset(dwmac->dev, - (void(*)(void *))clk_disable_unprepare, - dwmac->rgmii_tx_clk); - - return 0; + return devm_add_action_or_reset(dwmac->dev, + (void(*)(void *))clk_disable_unprepare, + clk); } static int meson8b_init_rgmii_delays(struct meson8b_dwmac *dwmac) From ec683f02a150b9c4428f08accd387c8c216ea0e5 Mon Sep 17 00:00:00 2001 From: Miquel Raynal Date: Fri, 4 Nov 2022 17:21:47 +0100 Subject: [PATCH 090/138] dt-bindings: net: tsnep: Fix typo on generic nvmem property While working on the nvmem description I figured out this file had the "nvmem-cell-names" property name misspelled. Fix the typo, as "nvmem-cells-names" has never existed. Fixes: 603094b2cdb7 ("dt-bindings: net: Add tsnep Ethernet controller") Signed-off-by: Miquel Raynal Reviewed-by: Gerhard Engleder Acked-by: Rob Herring Link: https://lore.kernel.org/r/20221104162147.1288230-1-miquel.raynal@bootlin.com Signed-off-by: Jakub Kicinski --- Documentation/devicetree/bindings/net/engleder,tsnep.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/net/engleder,tsnep.yaml b/Documentation/devicetree/bindings/net/engleder,tsnep.yaml index 5bd964a46a9d..a6921e805e37 100644 --- a/Documentation/devicetree/bindings/net/engleder,tsnep.yaml +++ b/Documentation/devicetree/bindings/net/engleder,tsnep.yaml @@ -47,7 +47,7 @@ properties: nvmem-cells: true - nvmem-cells-names: true + nvmem-cell-names: true phy-connection-type: enum: From 02f5999e652952d69c341a03d4313310703fd7f1 Mon Sep 17 00:00:00 2001 From: Yang Yingliang Date: Sat, 5 Nov 2022 14:34:42 +0800 Subject: [PATCH 091/138] octeontx2-pf: fix build error when CONFIG_OCTEONTX2_PF=y If CONFIG_MACSEC=m and CONFIG_OCTEONTX2_PF=y, it leads a build error: ld: drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.o: in function `otx2_pfaf_mbox_up_handler': otx2_pf.c:(.text+0x181c): undefined reference to `cn10k_handle_mcs_event' ld: drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.o: in function `otx2_probe': otx2_pf.c:(.text+0x437e): undefined reference to `cn10k_mcs_init' ld: drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.o: in function `otx2_remove': otx2_pf.c:(.text+0x5031): undefined reference to `cn10k_mcs_free' ld: drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.o: in function `otx2_mbox_up_handler_mcs_intr_notify': otx2_pf.c:(.text+0x5f11): undefined reference to `cn10k_handle_mcs_event' Make CONFIG_OCTEONTX2_PF depends on CONFIG_MACSEC to fix it. Because it has empty stub functions of cn10k, CONFIG_OCTEONTX2_PF can be enabled if CONFIG_MACSEC is disabled Fixes: c54ffc73601c ("octeontx2-pf: mcs: Introduce MACSEC hardware offloading") Reported-by: kernel test robot Signed-off-by: Yang Yingliang Link: https://lore.kernel.org/r/20221105063442.2013981-1-yangyingliang@huawei.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/marvell/octeontx2/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/marvell/octeontx2/Kconfig b/drivers/net/ethernet/marvell/octeontx2/Kconfig index 993ac180a5db..6b4f640163f7 100644 --- a/drivers/net/ethernet/marvell/octeontx2/Kconfig +++ b/drivers/net/ethernet/marvell/octeontx2/Kconfig @@ -32,6 +32,7 @@ config OCTEONTX2_PF tristate "Marvell OcteonTX2 NIC Physical Function driver" select OCTEONTX2_MBOX select NET_DEVLINK + depends on MACSEC || !MACSEC depends on (64BIT && COMPILE_TEST) || ARM64 select DIMLIB depends on PCI From 8d820bc9d12b8beebca836cceaf2bbe68216c2f8 Mon Sep 17 00:00:00 2001 From: YueHaibing Date: Sat, 5 Nov 2022 17:02:45 +0800 Subject: [PATCH 092/138] net: broadcom: Fix BCMGENET Kconfig While BCMGENET select BROADCOM_PHY as y, but PTP_1588_CLOCK_OPTIONAL is m, kconfig warning and build errors: WARNING: unmet direct dependencies detected for BROADCOM_PHY Depends on [m]: NETDEVICES [=y] && PHYLIB [=y] && PTP_1588_CLOCK_OPTIONAL [=m] Selected by [y]: - BCMGENET [=y] && NETDEVICES [=y] && ETHERNET [=y] && NET_VENDOR_BROADCOM [=y] && HAS_IOMEM [=y] && ARCH_BCM2835 [=y] drivers/net/phy/broadcom.o: In function `bcm54xx_suspend': broadcom.c:(.text+0x6ac): undefined reference to `bcm_ptp_stop' drivers/net/phy/broadcom.o: In function `bcm54xx_phy_probe': broadcom.c:(.text+0x784): undefined reference to `bcm_ptp_probe' drivers/net/phy/broadcom.o: In function `bcm54xx_config_init': broadcom.c:(.text+0xd4c): undefined reference to `bcm_ptp_config_init' Fixes: 99addbe31f55 ("net: broadcom: Select BROADCOM_PHY for BCMGENET") Signed-off-by: YueHaibing Acked-by: Florian Fainelli Link: https://lore.kernel.org/r/20221105090245.8508-1-yuehaibing@huawei.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/Kconfig b/drivers/net/ethernet/broadcom/Kconfig index f4e1ca68d831..55dfdb34e37b 100644 --- a/drivers/net/ethernet/broadcom/Kconfig +++ b/drivers/net/ethernet/broadcom/Kconfig @@ -77,7 +77,7 @@ config BCMGENET select BCM7XXX_PHY select MDIO_BCM_UNIMAC select DIMLIB - select BROADCOM_PHY if ARCH_BCM2835 + select BROADCOM_PHY if (ARCH_BCM2835 && PTP_1588_CLOCK_OPTIONAL) help This driver supports the built-in Ethernet MACs found in the Broadcom BCM7xxx Set Top Box family chipset. From 1c075b192fe41030457cd4a5f7dea730412bca40 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Fri, 4 Nov 2022 16:48:53 -0400 Subject: [PATCH 093/138] tipc: fix the msg->req tlv len check in tipc_nl_compat_name_table_dump_header This is a follow-up for commit 974cb0e3e7c9 ("tipc: fix uninit-value in tipc_nl_compat_name_table_dump") where it should have type casted sizeof(..) to int to work when TLV_GET_DATA_LEN() returns a negative value. syzbot reported a call trace because of it: BUG: KMSAN: uninit-value in ... tipc_nl_compat_name_table_dump+0x841/0xea0 net/tipc/netlink_compat.c:934 __tipc_nl_compat_dumpit+0xab2/0x1320 net/tipc/netlink_compat.c:238 tipc_nl_compat_dumpit+0x991/0xb50 net/tipc/netlink_compat.c:321 tipc_nl_compat_recv+0xb6e/0x1640 net/tipc/netlink_compat.c:1324 genl_family_rcv_msg_doit net/netlink/genetlink.c:731 [inline] genl_family_rcv_msg net/netlink/genetlink.c:775 [inline] genl_rcv_msg+0x103f/0x1260 net/netlink/genetlink.c:792 netlink_rcv_skb+0x3a5/0x6c0 net/netlink/af_netlink.c:2501 genl_rcv+0x3c/0x50 net/netlink/genetlink.c:803 netlink_unicast_kernel net/netlink/af_netlink.c:1319 [inline] netlink_unicast+0xf3b/0x1270 net/netlink/af_netlink.c:1345 netlink_sendmsg+0x1288/0x1440 net/netlink/af_netlink.c:1921 sock_sendmsg_nosec net/socket.c:714 [inline] sock_sendmsg net/socket.c:734 [inline] Reported-by: syzbot+e5dbaaa238680ce206ea@syzkaller.appspotmail.com Fixes: 974cb0e3e7c9 ("tipc: fix uninit-value in tipc_nl_compat_name_table_dump") Signed-off-by: Xin Long Link: https://lore.kernel.org/r/ccd6a7ea801b15aec092c3b532a883b4c5708695.1667594933.git.lucien.xin@gmail.com Signed-off-by: Jakub Kicinski --- net/tipc/netlink_compat.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/tipc/netlink_compat.c b/net/tipc/netlink_compat.c index fc68733673ba..dfea27a906f2 100644 --- a/net/tipc/netlink_compat.c +++ b/net/tipc/netlink_compat.c @@ -880,7 +880,7 @@ static int tipc_nl_compat_name_table_dump_header(struct tipc_nl_compat_msg *msg) }; ntq = (struct tipc_name_table_query *)TLV_DATA(msg->req); - if (TLV_GET_DATA_LEN(msg->req) < sizeof(struct tipc_name_table_query)) + if (TLV_GET_DATA_LEN(msg->req) < (int)sizeof(struct tipc_name_table_query)) return -EINVAL; depth = ntohl(ntq->depth); From 9f0b773210c27a8f5d98ddb2fc4ba60a42a3285f Mon Sep 17 00:00:00 2001 From: Xin Long Date: Fri, 4 Nov 2022 17:45:15 -0400 Subject: [PATCH 094/138] sctp: remove the unnecessary sinfo_stream check in sctp_prsctp_prune_unsent Since commit 5bbbbe32a431 ("sctp: introduce stream scheduler foundations"), sctp_stream_outq_migrate() has been called in sctp_stream_init/update to removes those chunks to streams higher than the new max. There is no longer need to do such check in sctp_prsctp_prune_unsent(). Signed-off-by: Xin Long Signed-off-by: Jakub Kicinski --- net/sctp/outqueue.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c index e213aaf45d67..c99fe3dc19bc 100644 --- a/net/sctp/outqueue.c +++ b/net/sctp/outqueue.c @@ -384,6 +384,7 @@ static int sctp_prsctp_prune_unsent(struct sctp_association *asoc, { struct sctp_outq *q = &asoc->outqueue; struct sctp_chunk *chk, *temp; + struct sctp_stream_out *sout; q->sched->unsched_all(&asoc->stream); @@ -398,12 +399,9 @@ static int sctp_prsctp_prune_unsent(struct sctp_association *asoc, sctp_sched_dequeue_common(q, chk); asoc->sent_cnt_removable--; asoc->abandoned_unsent[SCTP_PR_INDEX(PRIO)]++; - if (chk->sinfo.sinfo_stream < asoc->stream.outcnt) { - struct sctp_stream_out *streamout = - SCTP_SO(&asoc->stream, chk->sinfo.sinfo_stream); - streamout->ext->abandoned_unsent[SCTP_PR_INDEX(PRIO)]++; - } + sout = SCTP_SO(&asoc->stream, chk->sinfo.sinfo_stream); + sout->ext->abandoned_unsent[SCTP_PR_INDEX(PRIO)]++; msg_len -= chk->skb->truesize + sizeof(struct sctp_chunk); sctp_chunk_free(chk); From 2f201ae14ae0f91dbf1cffea7bb1e29e81d4d108 Mon Sep 17 00:00:00 2001 From: Xin Long Date: Fri, 4 Nov 2022 17:45:16 -0400 Subject: [PATCH 095/138] sctp: clear out_curr if all frag chunks of current msg are pruned A crash was reported by Zhen Chen: list_del corruption, ffffa035ddf01c18->next is NULL WARNING: CPU: 1 PID: 250682 at lib/list_debug.c:49 __list_del_entry_valid+0x59/0xe0 RIP: 0010:__list_del_entry_valid+0x59/0xe0 Call Trace: sctp_sched_dequeue_common+0x17/0x70 [sctp] sctp_sched_fcfs_dequeue+0x37/0x50 [sctp] sctp_outq_flush_data+0x85/0x360 [sctp] sctp_outq_uncork+0x77/0xa0 [sctp] sctp_cmd_interpreter.constprop.0+0x164/0x1450 [sctp] sctp_side_effects+0x37/0xe0 [sctp] sctp_do_sm+0xd0/0x230 [sctp] sctp_primitive_SEND+0x2f/0x40 [sctp] sctp_sendmsg_to_asoc+0x3fa/0x5c0 [sctp] sctp_sendmsg+0x3d5/0x440 [sctp] sock_sendmsg+0x5b/0x70 and in sctp_sched_fcfs_dequeue() it dequeued a chunk from stream out_curr outq while this outq was empty. Normally stream->out_curr must be set to NULL once all frag chunks of current msg are dequeued, as we can see in sctp_sched_dequeue_done(). However, in sctp_prsctp_prune_unsent() as it is not a proper dequeue, sctp_sched_dequeue_done() is not called to do this. This patch is to fix it by simply setting out_curr to NULL when the last frag chunk of current msg is dequeued from out_curr stream in sctp_prsctp_prune_unsent(). Fixes: 5bbbbe32a431 ("sctp: introduce stream scheduler foundations") Reported-by: Zhen Chen Tested-by: Caowangbao Signed-off-by: Xin Long Signed-off-by: Jakub Kicinski --- net/sctp/outqueue.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c index c99fe3dc19bc..20831079fb09 100644 --- a/net/sctp/outqueue.c +++ b/net/sctp/outqueue.c @@ -403,6 +403,11 @@ static int sctp_prsctp_prune_unsent(struct sctp_association *asoc, sout = SCTP_SO(&asoc->stream, chk->sinfo.sinfo_stream); sout->ext->abandoned_unsent[SCTP_PR_INDEX(PRIO)]++; + /* clear out_curr if all frag chunks are pruned */ + if (asoc->stream.out_curr == sout && + list_is_last(&chk->frag_list, &chk->msg->chunks)) + asoc->stream.out_curr = NULL; + msg_len -= chk->skb->truesize + sizeof(struct sctp_chunk); sctp_chunk_free(chk); if (msg_len <= 0) From 3faf7e14ec0c3462c2d747fa6793b8645d1391df Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Mon, 7 Nov 2022 09:14:45 +0800 Subject: [PATCH 096/138] net: lapbether: fix issue of invalid opcode in lapbeth_open() If lapb_register() failed when lapb device goes to up for the first time, the NAPI is not disabled. As a result, the invalid opcode issue is reported when the lapb device goes to up for the second time. The stack info is as follows: [ 1958.311422][T11356] kernel BUG at net/core/dev.c:6442! [ 1958.312206][T11356] invalid opcode: 0000 [#1] PREEMPT SMP KASAN [ 1958.315979][T11356] RIP: 0010:napi_enable+0x16a/0x1f0 [ 1958.332310][T11356] Call Trace: [ 1958.332817][T11356] [ 1958.336135][T11356] lapbeth_open+0x18/0x90 [ 1958.337446][T11356] __dev_open+0x258/0x490 [ 1958.341672][T11356] __dev_change_flags+0x4d4/0x6a0 [ 1958.345325][T11356] dev_change_flags+0x93/0x160 [ 1958.346027][T11356] devinet_ioctl+0x1276/0x1bf0 [ 1958.346738][T11356] inet_ioctl+0x1c8/0x2d0 [ 1958.349638][T11356] sock_ioctl+0x5d1/0x750 [ 1958.356059][T11356] __x64_sys_ioctl+0x3ec/0x1790 [ 1958.365594][T11356] do_syscall_64+0x35/0x80 [ 1958.366239][T11356] entry_SYSCALL_64_after_hwframe+0x46/0xb0 [ 1958.377381][T11356] Fixes: 514e1150da9c ("net: x25: Queue received packets in the drivers instead of per-CPU queues") Signed-off-by: Zhengchao Shao Link: https://lore.kernel.org/r/20221107011445.207372-1-shaozhengchao@huawei.com Signed-off-by: Paolo Abeni --- drivers/net/wan/lapbether.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/wan/lapbether.c b/drivers/net/wan/lapbether.c index cb360dca3250..d62a904d2e42 100644 --- a/drivers/net/wan/lapbether.c +++ b/drivers/net/wan/lapbether.c @@ -325,6 +325,7 @@ static int lapbeth_open(struct net_device *dev) err = lapb_register(dev, &lapbeth_callbacks); if (err != LAPB_OK) { + napi_disable(&lapbeth->napi); pr_err("lapb_register error: %d\n", err); return -ENODEV; } From b0c09c7f08c2467b2089bdf4adb2fbbc2464f4a8 Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Mon, 7 Nov 2022 09:21:59 +0800 Subject: [PATCH 097/138] net: ethernet: mtk-star-emac: disable napi when connect and start PHY failed in mtk_star_enable() When failed to connect to and start PHY in mtk_star_enable() for opening device, napi isn't disabled. When open mtk star device next time, it will reports a invalid opcode issue. Fix it. Only be compiled, not be tested. Fixes: 8c7bd5a454ff ("net: ethernet: mtk-star-emac: new driver") Signed-off-by: Zhengchao Shao Reviewed-by: Leon Romanovsky Link: https://lore.kernel.org/r/20221107012159.211387-1-shaozhengchao@huawei.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/mediatek/mtk_star_emac.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/mediatek/mtk_star_emac.c b/drivers/net/ethernet/mediatek/mtk_star_emac.c index 7e890f81148e..7050351250b7 100644 --- a/drivers/net/ethernet/mediatek/mtk_star_emac.c +++ b/drivers/net/ethernet/mediatek/mtk_star_emac.c @@ -1026,6 +1026,8 @@ static int mtk_star_enable(struct net_device *ndev) return 0; err_free_irq: + napi_disable(&priv->rx_napi); + napi_disable(&priv->tx_napi); free_irq(ndev->irq, ndev); err_free_skbs: mtk_star_free_rx_skbs(priv); From f0dfc4c88ef39be0ba736aa0ce6119263fc19aeb Mon Sep 17 00:00:00 2001 From: Ratheesh Kannoth Date: Mon, 7 Nov 2022 09:05:05 +0530 Subject: [PATCH 098/138] octeontx2-pf: Fix SQE threshold checking Current way of checking available SQE count which is based on HW updated SQB count could result in driver submitting an SQE even before CQE for the previously transmitted SQE at the same index is processed in NAPI resulting losing SKB pointers, hence a leak. Fix this by checking a consumer index which is updated once CQE is processed. Fixes: 3ca6c4c882a7 ("octeontx2-pf: Add packet transmission support") Signed-off-by: Ratheesh Kannoth Reviewed-by: Sunil Kovvuri Goutham Link: https://lore.kernel.org/r/20221107033505.2491464-1-rkannoth@marvell.com Signed-off-by: Paolo Abeni --- .../marvell/octeontx2/nic/otx2_common.c | 1 + .../marvell/octeontx2/nic/otx2_txrx.c | 32 +++++++++++-------- .../marvell/octeontx2/nic/otx2_txrx.h | 1 + 3 files changed, 21 insertions(+), 13 deletions(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c index 9ac9e6615ae7..9e10e7471b88 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c @@ -898,6 +898,7 @@ static int otx2_sq_init(struct otx2_nic *pfvf, u16 qidx, u16 sqb_aura) } sq->head = 0; + sq->cons_head = 0; sq->sqe_per_sqb = (pfvf->hw.sqb_size / sq->sqe_size) - 1; sq->num_sqbs = (qset->sqe_cnt + sq->sqe_per_sqb) / sq->sqe_per_sqb; /* Set SQE threshold to 10% of total SQEs */ diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c index 5ec11d71bf60..ef10aef3cda0 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c @@ -441,6 +441,7 @@ static int otx2_tx_napi_handler(struct otx2_nic *pfvf, struct otx2_cq_queue *cq, int budget) { int tx_pkts = 0, tx_bytes = 0, qidx; + struct otx2_snd_queue *sq; struct nix_cqe_tx_s *cqe; int processed_cqe = 0; @@ -451,6 +452,9 @@ static int otx2_tx_napi_handler(struct otx2_nic *pfvf, return 0; process_cqe: + qidx = cq->cq_idx - pfvf->hw.rx_queues; + sq = &pfvf->qset.sq[qidx]; + while (likely(processed_cqe < budget) && cq->pend_cqe) { cqe = (struct nix_cqe_tx_s *)otx2_get_next_cqe(cq); if (unlikely(!cqe)) { @@ -458,18 +462,20 @@ static int otx2_tx_napi_handler(struct otx2_nic *pfvf, return 0; break; } + if (cq->cq_type == CQ_XDP) { - qidx = cq->cq_idx - pfvf->hw.rx_queues; - otx2_xdp_snd_pkt_handler(pfvf, &pfvf->qset.sq[qidx], - cqe); + otx2_xdp_snd_pkt_handler(pfvf, sq, cqe); } else { - otx2_snd_pkt_handler(pfvf, cq, - &pfvf->qset.sq[cq->cint_idx], - cqe, budget, &tx_pkts, &tx_bytes); + otx2_snd_pkt_handler(pfvf, cq, sq, cqe, budget, + &tx_pkts, &tx_bytes); } + cqe->hdr.cqe_type = NIX_XQE_TYPE_INVALID; processed_cqe++; cq->pend_cqe--; + + sq->cons_head++; + sq->cons_head &= (sq->sqe_cnt - 1); } /* Free CQEs to HW */ @@ -1072,17 +1078,17 @@ bool otx2_sq_append_skb(struct net_device *netdev, struct otx2_snd_queue *sq, { struct netdev_queue *txq = netdev_get_tx_queue(netdev, qidx); struct otx2_nic *pfvf = netdev_priv(netdev); - int offset, num_segs, free_sqe; + int offset, num_segs, free_desc; struct nix_sqe_hdr_s *sqe_hdr; - /* Check if there is room for new SQE. - * 'Num of SQBs freed to SQ's pool - SQ's Aura count' - * will give free SQE count. + /* Check if there is enough room between producer + * and consumer index. */ - free_sqe = (sq->num_sqbs - *sq->aura_fc_addr) * sq->sqe_per_sqb; + free_desc = (sq->cons_head - sq->head - 1 + sq->sqe_cnt) & (sq->sqe_cnt - 1); + if (free_desc < sq->sqe_thresh) + return false; - if (free_sqe < sq->sqe_thresh || - free_sqe < otx2_get_sqe_count(pfvf, skb)) + if (free_desc < otx2_get_sqe_count(pfvf, skb)) return false; num_segs = skb_shinfo(skb)->nr_frags + 1; diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h index fbe62bbfb789..93cac2c2664c 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h @@ -79,6 +79,7 @@ struct sg_list { struct otx2_snd_queue { u8 aura_id; u16 head; + u16 cons_head; u16 sqe_size; u32 sqe_cnt; u16 num_sqbs; From ce9e57feeed81d17d5e80ed86f516ff0d39c3867 Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Mon, 7 Nov 2022 12:30:32 +0800 Subject: [PATCH 099/138] drivers: net: xgene: disable napi when register irq failed in xgene_enet_open() When failed to register irq in xgene_enet_open() for opening device, napi isn't disabled. When open xgene device next time, it will reports a invalid opcode issue. Fix it. Only be compiled, not be tested. Fixes: aeb20b6b3f4e ("drivers: net: xgene: fix: ifconfig up/down crash") Signed-off-by: Zhengchao Shao Link: https://lore.kernel.org/r/20221107043032.357673-1-shaozhengchao@huawei.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/apm/xgene/xgene_enet_main.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c index d6cfea65a714..390671640388 100644 --- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c @@ -1004,8 +1004,10 @@ static int xgene_enet_open(struct net_device *ndev) xgene_enet_napi_enable(pdata); ret = xgene_enet_register_irq(ndev); - if (ret) + if (ret) { + xgene_enet_napi_disable(pdata); return ret; + } if (ndev->phydev) { phy_start(ndev->phydev); From 03832a32bf8ff0a8305d94ddd3979835a807248f Mon Sep 17 00:00:00 2001 From: Ziyang Xuan Date: Thu, 3 Nov 2022 09:12:02 +0800 Subject: [PATCH 100/138] netfilter: nfnetlink: fix potential dead lock in nfnetlink_rcv_msg() When type is NFNL_CB_MUTEX and -EAGAIN error occur in nfnetlink_rcv_msg(), it does not execute nfnl_unlock(). That would trigger potential dead lock. Fixes: 50f2db9e368f ("netfilter: nfnetlink: consolidate callback types") Signed-off-by: Ziyang Xuan Signed-off-by: Florian Westphal --- net/netfilter/nfnetlink.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c index 9c44518cb70f..6d18fb346868 100644 --- a/net/netfilter/nfnetlink.c +++ b/net/netfilter/nfnetlink.c @@ -294,6 +294,7 @@ static int nfnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, nfnl_lock(subsys_id); if (nfnl_dereference_protected(subsys_id) != ss || nfnetlink_find_client(type, ss) != nc) { + nfnl_unlock(subsys_id); err = -EAGAIN; break; } From 03c1f1ef1584c981935fab2fa0c45d3e43e2c235 Mon Sep 17 00:00:00 2001 From: Shigeru Yoshida Date: Thu, 3 Nov 2022 22:08:49 +0900 Subject: [PATCH 101/138] netfilter: Cleanup nft_net->module_list from nf_tables_exit_net() syzbot reported a warning like below [1]: WARNING: CPU: 3 PID: 9 at net/netfilter/nf_tables_api.c:10096 nf_tables_exit_net+0x71c/0x840 Modules linked in: CPU: 2 PID: 9 Comm: kworker/u8:0 Tainted: G W 6.1.0-rc3-00072-g8e5423e991e8 #47 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.0-1.fc36 04/01/2014 Workqueue: netns cleanup_net RIP: 0010:nf_tables_exit_net+0x71c/0x840 ... Call Trace: ? __nft_release_table+0xfc0/0xfc0 ops_exit_list+0xb5/0x180 cleanup_net+0x506/0xb10 ? unregister_pernet_device+0x80/0x80 process_one_work+0xa38/0x1730 ? pwq_dec_nr_in_flight+0x2b0/0x2b0 ? rwlock_bug.part.0+0x90/0x90 ? _raw_spin_lock_irq+0x46/0x50 worker_thread+0x67e/0x10e0 ? process_one_work+0x1730/0x1730 kthread+0x2e5/0x3a0 ? kthread_complete_and_exit+0x40/0x40 ret_from_fork+0x1f/0x30 In nf_tables_exit_net(), there is a case where nft_net->commit_list is empty but nft_net->module_list is not empty. Such a case occurs with the following scenario: 1. nfnetlink_rcv_batch() is called 2. nf_tables_newset() returns -EAGAIN and NFNL_BATCH_FAILURE bit is set to status 3. nf_tables_abort() is called with NFNL_ABORT_AUTOLOAD (nft_net->commit_list is released, but nft_net->module_list is not because of NFNL_ABORT_AUTOLOAD flag) 4. Jump to replay label 5. netlink_skb_clone() fails and returns from the function (this is caused by fault injection in the reproducer of syzbot) This patch fixes this issue by calling __nf_tables_abort() when nft_net->module_list is not empty in nf_tables_exit_net(). Fixes: eb014de4fd41 ("netfilter: nf_tables: autoload modules from the abort path") Link: https://syzkaller.appspot.com/bug?id=802aba2422de4218ad0c01b46c9525cc9d4e4aa3 [1] Reported-by: syzbot+178efee9e2d7f87f5103@syzkaller.appspotmail.com Signed-off-by: Shigeru Yoshida Signed-off-by: Florian Westphal --- net/netfilter/nf_tables_api.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/netfilter/nf_tables_api.c b/net/netfilter/nf_tables_api.c index 76bd4d03dbda..e7152d599d73 100644 --- a/net/netfilter/nf_tables_api.c +++ b/net/netfilter/nf_tables_api.c @@ -10090,7 +10090,8 @@ static void __net_exit nf_tables_exit_net(struct net *net) struct nftables_pernet *nft_net = nft_pernet(net); mutex_lock(&nft_net->commit_mutex); - if (!list_empty(&nft_net->commit_list)) + if (!list_empty(&nft_net->commit_list) || + !list_empty(&nft_net->module_list)) __nf_tables_abort(net, NFNL_ABORT_NONE); __nft_release_tables(net); mutex_unlock(&nft_net->commit_mutex); From 519b58bbfa825f042fcf80261cc18e1e35f85ffd Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Tue, 8 Nov 2022 10:56:07 +0800 Subject: [PATCH 102/138] net: marvell: prestera: fix memory leak in prestera_rxtx_switch_init() When prestera_sdma_switch_init() failed, the memory pointed to by sw->rxtx isn't released. Fix it. Only be compiled, not be tested. Fixes: 501ef3066c89 ("net: marvell: prestera: Add driver for Prestera family ASIC devices") Signed-off-by: Zhengchao Shao Reviewed-by: Vadym Kochan Link: https://lore.kernel.org/r/20221108025607.338450-1-shaozhengchao@huawei.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/marvell/prestera/prestera_rxtx.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/marvell/prestera/prestera_rxtx.c b/drivers/net/ethernet/marvell/prestera/prestera_rxtx.c index 42ee963e9f75..9277a8fd1339 100644 --- a/drivers/net/ethernet/marvell/prestera/prestera_rxtx.c +++ b/drivers/net/ethernet/marvell/prestera/prestera_rxtx.c @@ -776,6 +776,7 @@ static netdev_tx_t prestera_sdma_xmit(struct prestera_sdma *sdma, int prestera_rxtx_switch_init(struct prestera_switch *sw) { struct prestera_rxtx *rxtx; + int err; rxtx = kzalloc(sizeof(*rxtx), GFP_KERNEL); if (!rxtx) @@ -783,7 +784,11 @@ int prestera_rxtx_switch_init(struct prestera_switch *sw) sw->rxtx = rxtx; - return prestera_sdma_switch_init(sw); + err = prestera_sdma_switch_init(sw); + if (err) + kfree(rxtx); + + return err; } void prestera_rxtx_switch_fini(struct prestera_switch *sw) From 07d120aa33cc9d9115753d159f64d20c94458781 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 7 Nov 2022 18:00:11 +0000 Subject: [PATCH 103/138] net: tun: call napi_schedule_prep() to ensure we own a napi A recent patch exposed another issue in napi_get_frags() caught by syzbot [1] Before feeding packets to GRO, and calling napi_complete() we must first grab NAPI_STATE_SCHED. [1] WARNING: CPU: 0 PID: 3612 at net/core/dev.c:6076 napi_complete_done+0x45b/0x880 net/core/dev.c:6076 Modules linked in: CPU: 0 PID: 3612 Comm: syz-executor408 Not tainted 6.1.0-rc3-syzkaller-00175-g1118b2049d77 #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/26/2022 RIP: 0010:napi_complete_done+0x45b/0x880 net/core/dev.c:6076 Code: c1 ea 03 0f b6 14 02 4c 89 f0 83 e0 07 83 c0 03 38 d0 7c 08 84 d2 0f 85 24 04 00 00 41 89 5d 1c e9 73 fc ff ff e8 b5 53 22 fa <0f> 0b e9 82 fe ff ff e8 a9 53 22 fa 48 8b 5c 24 08 31 ff 48 89 de RSP: 0018:ffffc90003c4f920 EFLAGS: 00010293 RAX: 0000000000000000 RBX: 0000000000000030 RCX: 0000000000000000 RDX: ffff8880251c0000 RSI: ffffffff875a58db RDI: 0000000000000007 RBP: 0000000000000001 R08: 0000000000000007 R09: 0000000000000000 R10: 0000000000000001 R11: 0000000000000001 R12: ffff888072d02628 R13: ffff888072d02618 R14: ffff888072d02634 R15: 0000000000000000 FS: 0000555555f13300(0000) GS:ffff8880b9a00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 000055c44d3892b8 CR3: 00000000172d2000 CR4: 00000000003506f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: napi_complete include/linux/netdevice.h:510 [inline] tun_get_user+0x206d/0x3a60 drivers/net/tun.c:1980 tun_chr_write_iter+0xdb/0x200 drivers/net/tun.c:2027 call_write_iter include/linux/fs.h:2191 [inline] do_iter_readv_writev+0x20b/0x3b0 fs/read_write.c:735 do_iter_write+0x182/0x700 fs/read_write.c:861 vfs_writev+0x1aa/0x630 fs/read_write.c:934 do_writev+0x133/0x2f0 fs/read_write.c:977 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd RIP: 0033:0x7f37021a3c19 Fixes: 1118b2049d77 ("net: tun: Fix memory leaks of napi_get_frags") Reported-by: syzbot Signed-off-by: Eric Dumazet Cc: Wang Yufen Link: https://lore.kernel.org/r/20221107180011.188437-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- drivers/net/tun.c | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-) diff --git a/drivers/net/tun.c b/drivers/net/tun.c index eb12f3136a54..7a3ab3427369 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -1967,18 +1967,25 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile, skb_headlen(skb)); if (unlikely(headlen > skb_headlen(skb))) { + WARN_ON_ONCE(1); + err = -ENOMEM; dev_core_stats_rx_dropped_inc(tun->dev); +napi_busy: napi_free_frags(&tfile->napi); rcu_read_unlock(); mutex_unlock(&tfile->napi_mutex); - WARN_ON(1); - return -ENOMEM; + return err; } - local_bh_disable(); - napi_gro_frags(&tfile->napi); - napi_complete(&tfile->napi); - local_bh_enable(); + if (likely(napi_schedule_prep(&tfile->napi))) { + local_bh_disable(); + napi_gro_frags(&tfile->napi); + napi_complete(&tfile->napi); + local_bh_enable(); + } else { + err = -EBUSY; + goto napi_busy; + } mutex_unlock(&tfile->napi_mutex); } else if (tfile->napi_enabled) { struct sk_buff_head *queue = &tfile->sk.sk_write_queue; From b06334919c7a068d54ba5b219c05e919d89943f7 Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Mon, 7 Nov 2022 18:14:43 +0800 Subject: [PATCH 104/138] net: nixge: disable napi when enable interrupts failed in nixge_open() When failed to enable interrupts in nixge_open() for opening device, napi isn't disabled. When open nixge device next time, it will reports a invalid opcode issue. Fix it. Only be compiled, not be tested. Fixes: 492caffa8a1a ("net: ethernet: nixge: Add support for National Instruments XGE netdev") Signed-off-by: Zhengchao Shao Link: https://lore.kernel.org/r/20221107101443.120205-1-shaozhengchao@huawei.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ni/nixge.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/ni/nixge.c b/drivers/net/ethernet/ni/nixge.c index 3db4a2431741..19d043b593cc 100644 --- a/drivers/net/ethernet/ni/nixge.c +++ b/drivers/net/ethernet/ni/nixge.c @@ -900,6 +900,7 @@ static int nixge_open(struct net_device *ndev) err_rx_irq: free_irq(priv->tx_irq, ndev); err_tx_irq: + napi_disable(&priv->napi); phy_stop(phy); phy_disconnect(phy); tasklet_kill(&priv->dma_err_tasklet); From 742c60e1285ca40642e988f7e3db92232171b27d Mon Sep 17 00:00:00 2001 From: Nick Child Date: Mon, 7 Nov 2022 14:32:15 -0600 Subject: [PATCH 105/138] ibmveth: Reduce default tx queues to 8 Previously, the default number of transmit queues was 16. Due to resource concerns, set to 8 queues instead. Still allow the user to set more queues (max 16) if they like. Since the driver is virtualized away from the physical NIC, the purpose of multiple queues is purely to allow for parallel calls to the hypervisor. Therefore, there is no noticeable effect on performance by reducing queue count to 8. Fixes: d926793c1de9 ("ibmveth: Implement multi queue on xmit") Reported-by: Dave Taht Signed-off-by: Nick Child Link: https://lore.kernel.org/r/20221107203215.58206-1-nnac123@linux.ibm.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ibm/ibmveth.c | 3 ++- drivers/net/ethernet/ibm/ibmveth.h | 1 + 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/ibm/ibmveth.c b/drivers/net/ethernet/ibm/ibmveth.c index 3b14dc93f59d..5b96cd94dcd2 100644 --- a/drivers/net/ethernet/ibm/ibmveth.c +++ b/drivers/net/ethernet/ibm/ibmveth.c @@ -1757,7 +1757,8 @@ static int ibmveth_probe(struct vio_dev *dev, const struct vio_device_id *id) kobject_uevent(kobj, KOBJ_ADD); } - rc = netif_set_real_num_tx_queues(netdev, ibmveth_real_max_tx_queues()); + rc = netif_set_real_num_tx_queues(netdev, min(num_online_cpus(), + IBMVETH_DEFAULT_QUEUES)); if (rc) { netdev_dbg(netdev, "failed to set number of tx queues rc=%d\n", rc); diff --git a/drivers/net/ethernet/ibm/ibmveth.h b/drivers/net/ethernet/ibm/ibmveth.h index daf6f615c03f..115d4c45aa77 100644 --- a/drivers/net/ethernet/ibm/ibmveth.h +++ b/drivers/net/ethernet/ibm/ibmveth.h @@ -100,6 +100,7 @@ static inline long h_illan_attributes(unsigned long unit_address, #define IBMVETH_MAX_BUF_SIZE (1024 * 128) #define IBMVETH_MAX_TX_BUF_SIZE (1024 * 64) #define IBMVETH_MAX_QUEUES 16U +#define IBMVETH_DEFAULT_QUEUES 8U static int pool_size[] = { 512, 1024 * 2, 1024 * 16, 1024 * 32, 1024 * 64 }; static int pool_count[] = { 256, 512, 256, 256, 256 }; From 58bb78ce02269c0cf5b1f2bd2e4a605500b44c6b Mon Sep 17 00:00:00 2001 From: Phil Sutter Date: Wed, 9 Nov 2022 10:06:04 +0100 Subject: [PATCH 106/138] selftests: netfilter: Fix and review rpath.sh Address a few problems with the initial test script version: * On systems with ip6tables but no ip6tables-legacy, testing for ip6tables was disabled by accident. * Firewall setup phase did not respect possibly unavailable tools. * Consistently call nft via '$nft'. Fixes: 6e31ce831c63b ("selftests: netfilter: Test reverse path filtering") Signed-off-by: Phil Sutter Signed-off-by: Pablo Neira Ayuso --- tools/testing/selftests/netfilter/rpath.sh | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/tools/testing/selftests/netfilter/rpath.sh b/tools/testing/selftests/netfilter/rpath.sh index 2d8da7bd8ab7..f7311e66d219 100755 --- a/tools/testing/selftests/netfilter/rpath.sh +++ b/tools/testing/selftests/netfilter/rpath.sh @@ -15,7 +15,7 @@ fi if ip6tables-legacy --version >/dev/null 2>&1; then ip6tables='ip6tables-legacy' -elif ! ip6tables --version >/dev/null 2>&1; then +elif ip6tables --version >/dev/null 2>&1; then ip6tables='ip6tables' else ip6tables='' @@ -62,9 +62,11 @@ ip -net "$ns1" a a fec0:42::2/64 dev v0 nodad ip -net "$ns2" a a fec0:42::1/64 dev d0 nodad # firewall matches to test -ip netns exec "$ns2" "$iptables" -t raw -A PREROUTING -s 192.168.0.0/16 -m rpfilter -ip netns exec "$ns2" "$ip6tables" -t raw -A PREROUTING -s fec0::/16 -m rpfilter -ip netns exec "$ns2" nft -f - < Date: Mon, 7 Nov 2022 13:04:49 +0530 Subject: [PATCH 107/138] net: wwan: iosm: fix memory leak in ipc_pcie_read_bios_cfg ipc_pcie_read_bios_cfg() is using the acpi_evaluate_dsm() to obtain the wwan power state configuration from BIOS but is not freeing the acpi_object. The acpi_evaluate_dsm() returned acpi_object to be freed. Free the acpi_object after use. Fixes: 7e98d785ae61 ("net: iosm: entry point") Signed-off-by: M Chetan Kumar Signed-off-by: David S. Miller --- drivers/net/wwan/iosm/iosm_ipc_pcie.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/net/wwan/iosm/iosm_ipc_pcie.c b/drivers/net/wwan/iosm/iosm_ipc_pcie.c index 31f57b986df2..97cb6846c6ae 100644 --- a/drivers/net/wwan/iosm/iosm_ipc_pcie.c +++ b/drivers/net/wwan/iosm/iosm_ipc_pcie.c @@ -232,6 +232,7 @@ static void ipc_pcie_config_init(struct iosm_pcie *ipc_pcie) */ static enum ipc_pcie_sleep_state ipc_pcie_read_bios_cfg(struct device *dev) { + enum ipc_pcie_sleep_state sleep_state = IPC_PCIE_D0L12; union acpi_object *object; acpi_handle handle_acpi; @@ -242,12 +243,16 @@ static enum ipc_pcie_sleep_state ipc_pcie_read_bios_cfg(struct device *dev) } object = acpi_evaluate_dsm(handle_acpi, &wwan_acpi_guid, 0, 3, NULL); + if (!object) + goto default_ret; - if (object && object->integer.value == 3) - return IPC_PCIE_D3L2; + if (object->integer.value == 3) + sleep_state = IPC_PCIE_D3L2; + + kfree(object); default_ret: - return IPC_PCIE_D0L12; + return sleep_state; } static int ipc_pcie_probe(struct pci_dev *pci, From 035e3befc191347331dd2530c3686e05a8acfbb2 Mon Sep 17 00:00:00 2001 From: M Chetan Kumar Date: Mon, 7 Nov 2022 13:05:02 +0530 Subject: [PATCH 108/138] net: wwan: iosm: fix driver not working with INTEL_IOMMU disabled With INTEL_IOMMU disable config or by forcing intel_iommu=off from grub some of the features of IOSM driver like browsing, flashing & coredump collection is not working. When driver calls DMA API - dma_map_single() for tx transfers. It is resulting in dma mapping error. Set the device DMA addressing capabilities using dma_set_mask() and remove the INTEL_IOMMU dependency in kconfig so that driver follows the platform config either INTEL_IOMMU enable or disable. Fixes: f7af616c632e ("net: iosm: infrastructure") Signed-off-by: M Chetan Kumar Signed-off-by: David S. Miller --- drivers/net/wwan/Kconfig | 2 +- drivers/net/wwan/iosm/iosm_ipc_pcie.c | 7 +++++++ 2 files changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/net/wwan/Kconfig b/drivers/net/wwan/Kconfig index 3486ffe94ac4..ac4d73b5626f 100644 --- a/drivers/net/wwan/Kconfig +++ b/drivers/net/wwan/Kconfig @@ -94,7 +94,7 @@ config RPMSG_WWAN_CTRL config IOSM tristate "IOSM Driver for Intel M.2 WWAN Device" - depends on INTEL_IOMMU + depends on PCI select NET_DEVLINK select RELAY if WWAN_DEBUGFS help diff --git a/drivers/net/wwan/iosm/iosm_ipc_pcie.c b/drivers/net/wwan/iosm/iosm_ipc_pcie.c index 97cb6846c6ae..d3d34d1c4704 100644 --- a/drivers/net/wwan/iosm/iosm_ipc_pcie.c +++ b/drivers/net/wwan/iosm/iosm_ipc_pcie.c @@ -259,6 +259,7 @@ static int ipc_pcie_probe(struct pci_dev *pci, const struct pci_device_id *pci_id) { struct iosm_pcie *ipc_pcie = kzalloc(sizeof(*ipc_pcie), GFP_KERNEL); + int ret; pr_debug("Probing device 0x%X from the vendor 0x%X", pci_id->device, pci_id->vendor); @@ -291,6 +292,12 @@ static int ipc_pcie_probe(struct pci_dev *pci, goto pci_enable_fail; } + ret = dma_set_mask(ipc_pcie->dev, DMA_BIT_MASK(64)); + if (ret) { + dev_err(ipc_pcie->dev, "Could not set PCI DMA mask: %d", ret); + return ret; + } + ipc_pcie_config_aspm(ipc_pcie); dev_dbg(ipc_pcie->dev, "PCIe device enabled."); From 02d2d2ea4a3bc2391f6ac31f6854da83e8a63829 Mon Sep 17 00:00:00 2001 From: M Chetan Kumar Date: Mon, 7 Nov 2022 13:05:13 +0530 Subject: [PATCH 109/138] net: wwan: iosm: fix invalid mux header type Data stall seen during peak DL throughput test & packets are dropped by mux layer due to invalid header type in datagram. During initlization Mux aggregration protocol is set to default UL/DL size and TD count of Mux lite protocol. This configuration mismatch between device and driver is resulting in data stall/packet drops. Override the UL/DL size and TD count for Mux aggregation protocol. Fixes: 1f52d7b62285 ("net: wwan: iosm: Enable M.2 7360 WWAN card support") Signed-off-by: M Chetan Kumar Signed-off-by: David S. Miller --- drivers/net/wwan/iosm/iosm_ipc_imem_ops.c | 8 ++++++++ drivers/net/wwan/iosm/iosm_ipc_mux.h | 1 + 2 files changed, 9 insertions(+) diff --git a/drivers/net/wwan/iosm/iosm_ipc_imem_ops.c b/drivers/net/wwan/iosm/iosm_ipc_imem_ops.c index b7f9237dedf7..66b90cc4c346 100644 --- a/drivers/net/wwan/iosm/iosm_ipc_imem_ops.c +++ b/drivers/net/wwan/iosm/iosm_ipc_imem_ops.c @@ -91,6 +91,14 @@ void ipc_imem_wwan_channel_init(struct iosm_imem *ipc_imem, } ipc_chnl_cfg_get(&chnl_cfg, ipc_imem->nr_of_channels); + + if (ipc_imem->mmio->mux_protocol == MUX_AGGREGATION && + ipc_imem->nr_of_channels == IPC_MEM_IP_CHL_ID_0) { + chnl_cfg.ul_nr_of_entries = IPC_MEM_MAX_TDS_MUX_AGGR_UL; + chnl_cfg.dl_nr_of_entries = IPC_MEM_MAX_TDS_MUX_AGGR_DL; + chnl_cfg.dl_buf_size = IPC_MEM_MAX_ADB_BUF_SIZE; + } + ipc_imem_channel_init(ipc_imem, IPC_CTYPE_WWAN, chnl_cfg, IRQ_MOD_OFF); diff --git a/drivers/net/wwan/iosm/iosm_ipc_mux.h b/drivers/net/wwan/iosm/iosm_ipc_mux.h index cd9d74cc097f..9968bb885c1f 100644 --- a/drivers/net/wwan/iosm/iosm_ipc_mux.h +++ b/drivers/net/wwan/iosm/iosm_ipc_mux.h @@ -10,6 +10,7 @@ #define IPC_MEM_MAX_UL_DG_ENTRIES 100 #define IPC_MEM_MAX_TDS_MUX_AGGR_UL 60 +#define IPC_MEM_MAX_TDS_MUX_AGGR_DL 60 #define IPC_MEM_MAX_ADB_BUF_SIZE (16 * 1024) #define IPC_MEM_MAX_UL_ADB_BUF_SIZE IPC_MEM_MAX_ADB_BUF_SIZE From 980ec04a88c9f0046c1da65833fb77b2ffa34b04 Mon Sep 17 00:00:00 2001 From: M Chetan Kumar Date: Mon, 7 Nov 2022 13:05:24 +0530 Subject: [PATCH 110/138] net: wwan: iosm: fix kernel test robot reported errors Include linux/vmalloc.h in iosm_ipc_coredump.c & iosm_ipc_devlink.c to resolve kernel test robot errors. Reported-by: kernel test robot Signed-off-by: M Chetan Kumar Signed-off-by: David S. Miller --- drivers/net/wwan/iosm/iosm_ipc_coredump.c | 1 + drivers/net/wwan/iosm/iosm_ipc_devlink.c | 1 + 2 files changed, 2 insertions(+) diff --git a/drivers/net/wwan/iosm/iosm_ipc_coredump.c b/drivers/net/wwan/iosm/iosm_ipc_coredump.c index 9acd87724c9d..26ca30476f40 100644 --- a/drivers/net/wwan/iosm/iosm_ipc_coredump.c +++ b/drivers/net/wwan/iosm/iosm_ipc_coredump.c @@ -2,6 +2,7 @@ /* * Copyright (C) 2020-2021 Intel Corporation. */ +#include #include "iosm_ipc_coredump.h" diff --git a/drivers/net/wwan/iosm/iosm_ipc_devlink.c b/drivers/net/wwan/iosm/iosm_ipc_devlink.c index 17da85a8f337..2fe724d623c0 100644 --- a/drivers/net/wwan/iosm/iosm_ipc_devlink.c +++ b/drivers/net/wwan/iosm/iosm_ipc_devlink.c @@ -2,6 +2,7 @@ /* * Copyright (C) 2020-2021 Intel Corporation. */ +#include #include "iosm_ipc_chnl_cfg.h" #include "iosm_ipc_coredump.h" From 15f8f168952f54d3c86d734dc764f20844e423ac Mon Sep 17 00:00:00 2001 From: Vlad Buslov Date: Wed, 2 Nov 2022 23:55:37 -0700 Subject: [PATCH 111/138] net/mlx5: Bridge, verify LAG state when adding bond to bridge Mlx5 LAG is initialized asynchronously on a workqueue which means that for a brief moment after setting mlx5 UL representors as lower devices of a bond netdevice the LAG itself is not fully initialized in the driver. When adding such bond device to a bridge mlx5 bridge code will not consider it as offload-capable, skip creating necessary bookkeeping and fail any further bridge offload-related commands with it (setting VLANs, offloading FDBs, etc.). In order to make the error explicit during bridge initialization stage implement the code that detects such condition during NETDEV_PRECHANGEUPPER event and returns an error. Fixes: ff9b7521468b ("net/mlx5: Bridge, support LAG") Signed-off-by: Vlad Buslov Reviewed-by: Roi Dayan Reviewed-by: Mark Bloch Signed-off-by: Saeed Mahameed --- .../mellanox/mlx5/core/en/rep/bridge.c | 31 +++++++++++++++++++ 1 file changed, 31 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c index 39ef2a2561a3..8099a21e674c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bridge.c @@ -164,6 +164,36 @@ static int mlx5_esw_bridge_port_changeupper(struct notifier_block *nb, void *ptr return err; } +static int +mlx5_esw_bridge_changeupper_validate_netdev(void *ptr) +{ + struct net_device *dev = netdev_notifier_info_to_dev(ptr); + struct netdev_notifier_changeupper_info *info = ptr; + struct net_device *upper = info->upper_dev; + struct net_device *lower; + struct list_head *iter; + + if (!netif_is_bridge_master(upper) || !netif_is_lag_master(dev)) + return 0; + + netdev_for_each_lower_dev(dev, lower, iter) { + struct mlx5_core_dev *mdev; + struct mlx5e_priv *priv; + + if (!mlx5e_eswitch_rep(lower)) + continue; + + priv = netdev_priv(lower); + mdev = priv->mdev; + if (!mlx5_lag_is_active(mdev)) + return -EAGAIN; + if (!mlx5_lag_is_shared_fdb(mdev)) + return -EOPNOTSUPP; + } + + return 0; +} + static int mlx5_esw_bridge_switchdev_port_event(struct notifier_block *nb, unsigned long event, void *ptr) { @@ -171,6 +201,7 @@ static int mlx5_esw_bridge_switchdev_port_event(struct notifier_block *nb, switch (event) { case NETDEV_PRECHANGEUPPER: + err = mlx5_esw_bridge_changeupper_validate_netdev(ptr); break; case NETDEV_CHANGEUPPER: From 2808b37b59288ad8f1897e3546c2296df3384b65 Mon Sep 17 00:00:00 2001 From: Roy Novich Date: Wed, 2 Nov 2022 23:55:38 -0700 Subject: [PATCH 112/138] net/mlx5: Allow async trigger completion execution on single CPU systems MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit For a single CPU system, the kernel thread executing mlx5_cmd_flush() never releases the CPU but calls down_trylock(&cmd→sem) in a busy loop. On a single processor system, this leads to a deadlock as the kernel thread which executes mlx5_cmd_invoke() never gets scheduled. Fix this, by adding the cond_resched() call to the loop, allow the command completion kernel thread to execute. Fixes: 8e715cd613a1 ("net/mlx5: Set command entry semaphore up once got index free") Signed-off-by: Alexander Schmidt Signed-off-by: Roy Novich Reviewed-by: Moshe Shemesh Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index 46ba4c2faad2..2e0d59ca62b5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -1770,12 +1770,17 @@ void mlx5_cmd_flush(struct mlx5_core_dev *dev) struct mlx5_cmd *cmd = &dev->cmd; int i; - for (i = 0; i < cmd->max_reg_cmds; i++) - while (down_trylock(&cmd->sem)) + for (i = 0; i < cmd->max_reg_cmds; i++) { + while (down_trylock(&cmd->sem)) { mlx5_cmd_trigger_completions(dev); + cond_resched(); + } + } - while (down_trylock(&cmd->pages_sem)) + while (down_trylock(&cmd->pages_sem)) { mlx5_cmd_trigger_completions(dev); + cond_resched(); + } /* Unlock cmdif */ up(&cmd->pages_sem); From e12de39c07a7872c1ac7250311bb60b74ff29f25 Mon Sep 17 00:00:00 2001 From: Chris Mi Date: Wed, 2 Nov 2022 23:55:39 -0700 Subject: [PATCH 113/138] net/mlx5: E-switch, Set to legacy mode if failed to change switchdev mode No need to rollback to the other mode because probably will fail again. Just set to legacy mode and clear fdb table created flag. So that fdb table will not be cleared again. Fixes: f019679ea5f2 ("net/mlx5: E-switch, Remove dependency between sriov and eswitch mode") Signed-off-by: Chris Mi Reviewed-by: Roi Dayan Signed-off-by: Saeed Mahameed --- .../net/ethernet/mellanox/mlx5/core/eswitch.c | 14 ++++++++------ .../mellanox/mlx5/core/eswitch_offloads.c | 18 +++--------------- 2 files changed, 11 insertions(+), 21 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index c59107fa9e6d..2169486c4bfb 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -1387,12 +1387,14 @@ void mlx5_eswitch_disable_locked(struct mlx5_eswitch *esw) esw->mode == MLX5_ESWITCH_LEGACY ? "LEGACY" : "OFFLOADS", esw->esw_funcs.num_vfs, esw->enabled_vports); - esw->fdb_table.flags &= ~MLX5_ESW_FDB_CREATED; - if (esw->mode == MLX5_ESWITCH_OFFLOADS) - esw_offloads_disable(esw); - else if (esw->mode == MLX5_ESWITCH_LEGACY) - esw_legacy_disable(esw); - mlx5_esw_acls_ns_cleanup(esw); + if (esw->fdb_table.flags & MLX5_ESW_FDB_CREATED) { + esw->fdb_table.flags &= ~MLX5_ESW_FDB_CREATED; + if (esw->mode == MLX5_ESWITCH_OFFLOADS) + esw_offloads_disable(esw); + else if (esw->mode == MLX5_ESWITCH_LEGACY) + esw_legacy_disable(esw); + mlx5_esw_acls_ns_cleanup(esw); + } if (esw->mode == MLX5_ESWITCH_OFFLOADS) devl_rate_nodes_destroy(devlink); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 4e50df3139c6..728ca9f2bb9d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -2310,7 +2310,7 @@ static int esw_create_restore_table(struct mlx5_eswitch *esw) static int esw_offloads_start(struct mlx5_eswitch *esw, struct netlink_ext_ack *extack) { - int err, err1; + int err; esw->mode = MLX5_ESWITCH_OFFLOADS; err = mlx5_eswitch_enable_locked(esw, esw->dev->priv.sriov.num_vfs); @@ -2318,11 +2318,6 @@ static int esw_offloads_start(struct mlx5_eswitch *esw, NL_SET_ERR_MSG_MOD(extack, "Failed setting eswitch to offloads"); esw->mode = MLX5_ESWITCH_LEGACY; - err1 = mlx5_eswitch_enable_locked(esw, MLX5_ESWITCH_IGNORE_NUM_VFS); - if (err1) { - NL_SET_ERR_MSG_MOD(extack, - "Failed setting eswitch back to legacy"); - } mlx5_rescan_drivers(esw->dev); } if (esw->offloads.inline_mode == MLX5_INLINE_MODE_NONE) { @@ -3389,19 +3384,12 @@ int esw_offloads_enable(struct mlx5_eswitch *esw) static int esw_offloads_stop(struct mlx5_eswitch *esw, struct netlink_ext_ack *extack) { - int err, err1; + int err; esw->mode = MLX5_ESWITCH_LEGACY; err = mlx5_eswitch_enable_locked(esw, MLX5_ESWITCH_IGNORE_NUM_VFS); - if (err) { + if (err) NL_SET_ERR_MSG_MOD(extack, "Failed setting eswitch to legacy"); - esw->mode = MLX5_ESWITCH_OFFLOADS; - err1 = mlx5_eswitch_enable_locked(esw, MLX5_ESWITCH_IGNORE_NUM_VFS); - if (err1) { - NL_SET_ERR_MSG_MOD(extack, - "Failed setting eswitch back to offloads"); - } - } return err; } From 7d167b4a4c810919ba1affd02fc6b7f40e7e6eeb Mon Sep 17 00:00:00 2001 From: Shay Drory Date: Wed, 2 Nov 2022 23:55:40 -0700 Subject: [PATCH 114/138] net/mlx5: fw_reset: Don't try to load device in case PCI isn't working In case PCI reads fail after unload, there is no use in trying to load the device. Fixes: 5ec697446f46 ("net/mlx5: Add support for devlink reload action fw activate") Signed-off-by: Shay Drory Reviewed-by: Moshe Shemesh Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c index 07c583996c29..9d908a0ccfef 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c @@ -152,7 +152,8 @@ static void mlx5_fw_reset_complete_reload(struct mlx5_core_dev *dev) mlx5_unload_one(dev); if (mlx5_health_wait_pci_up(dev)) mlx5_core_err(dev, "reset reload flow aborted, PCI reads still not working\n"); - mlx5_load_one(dev, false); + else + mlx5_load_one(dev, false); devlink_remote_reload_actions_performed(priv_to_devlink(dev), 0, BIT(DEVLINK_RELOAD_ACTION_DRIVER_REINIT) | BIT(DEVLINK_RELOAD_ACTION_FW_ACTIVATE)); From f9c955b4fe5c8626d11b8a9b93ccc0ba77358fec Mon Sep 17 00:00:00 2001 From: Maxim Mikityanskiy Date: Wed, 2 Nov 2022 23:55:42 -0700 Subject: [PATCH 115/138] net/mlx5e: Add missing sanity checks for max TX WQE size The commit cited below started using the firmware capability for the maximum TX WQE size. This commit adds an important check to verify that the driver doesn't attempt to exceed this capability, and also restores another check mistakenly removed in the cited commit (a WQE must not exceed the page size). Fixes: c27bd1718c06 ("net/mlx5e: Read max WQEBBs on the SQ from firmware") Signed-off-by: Maxim Mikityanskiy Reviewed-by: Tariq Toukan Signed-off-by: Saeed Mahameed --- .../net/ethernet/mellanox/mlx5/core/en/txrx.h | 24 ++++++++++++++++++- .../net/ethernet/mellanox/mlx5/core/en_main.c | 7 ++++++ .../net/ethernet/mellanox/mlx5/core/en_tx.c | 5 ++++ 3 files changed, 35 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h index cb164b62f543..853f312cd757 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h @@ -11,6 +11,27 @@ #define INL_HDR_START_SZ (sizeof(((struct mlx5_wqe_eth_seg *)NULL)->inline_hdr.start)) +/* IPSEC inline data includes: + * 1. ESP trailer: up to 255 bytes of padding, 1 byte for pad length, 1 byte for + * next header. + * 2. ESP authentication data: 16 bytes for ICV. + */ +#define MLX5E_MAX_TX_IPSEC_DS DIV_ROUND_UP(sizeof(struct mlx5_wqe_inline_seg) + \ + 255 + 1 + 1 + 16, MLX5_SEND_WQE_DS) + +/* 366 should be big enough to cover all L2, L3 and L4 headers with possible + * encapsulations. + */ +#define MLX5E_MAX_TX_INLINE_DS DIV_ROUND_UP(366 - INL_HDR_START_SZ + VLAN_HLEN, \ + MLX5_SEND_WQE_DS) + +/* Sync the calculation with mlx5e_sq_calc_wqe_attr. */ +#define MLX5E_MAX_TX_WQEBBS DIV_ROUND_UP(MLX5E_TX_WQE_EMPTY_DS_COUNT + \ + MLX5E_MAX_TX_INLINE_DS + \ + MLX5E_MAX_TX_IPSEC_DS + \ + MAX_SKB_FRAGS + 1, \ + MLX5_SEND_WQEBB_NUM_DS) + #define MLX5E_RX_ERR_CQE(cqe) (get_cqe_opcode(cqe) != MLX5_CQE_RESP_SEND) static inline @@ -424,6 +445,8 @@ mlx5e_set_eseg_swp(struct sk_buff *skb, struct mlx5_wqe_eth_seg *eseg, static inline u16 mlx5e_stop_room_for_wqe(struct mlx5_core_dev *mdev, u16 wqe_size) { + WARN_ON_ONCE(PAGE_SIZE / MLX5_SEND_WQE_BB < mlx5e_get_max_sq_wqebbs(mdev)); + /* A WQE must not cross the page boundary, hence two conditions: * 1. Its size must not exceed the page size. * 2. If the WQE size is X, and the space remaining in a page is less @@ -436,7 +459,6 @@ static inline u16 mlx5e_stop_room_for_wqe(struct mlx5_core_dev *mdev, u16 wqe_si "wqe_size %u is greater than max SQ WQEBBs %u", wqe_size, mlx5e_get_max_sq_wqebbs(mdev)); - return MLX5E_STOP_ROOM(wqe_size); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 364f04309149..e3a4f01bcceb 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -5694,6 +5694,13 @@ int mlx5e_attach_netdev(struct mlx5e_priv *priv) mlx5e_fs_set_state_destroy(priv->fs, !test_bit(MLX5E_STATE_DESTROYING, &priv->state)); + /* Validate the max_wqe_size_sq capability. */ + if (WARN_ON_ONCE(mlx5e_get_max_sq_wqebbs(priv->mdev) < MLX5E_MAX_TX_WQEBBS)) { + mlx5_core_warn(priv->mdev, "MLX5E: Max SQ WQEBBs firmware capability: %u, needed %lu\n", + mlx5e_get_max_sq_wqebbs(priv->mdev), MLX5E_MAX_TX_WQEBBS); + return -EIO; + } + /* max number of channels may have changed */ max_nch = mlx5e_calc_max_nch(priv->mdev, priv->netdev, profile); if (priv->channels.params.num_channels > max_nch) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index 6adca01fbdc9..f7897ddb29c5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -305,6 +305,8 @@ static void mlx5e_sq_calc_wqe_attr(struct sk_buff *skb, const struct mlx5e_tx_at u16 ds_cnt_inl = 0; u16 ds_cnt_ids = 0; + /* Sync the calculation with MLX5E_MAX_TX_WQEBBS. */ + if (attr->insz) ds_cnt_ids = DIV_ROUND_UP(sizeof(struct mlx5_wqe_inline_seg) + attr->insz, MLX5_SEND_WQE_DS); @@ -317,6 +319,9 @@ static void mlx5e_sq_calc_wqe_attr(struct sk_buff *skb, const struct mlx5e_tx_at inl += VLAN_HLEN; ds_cnt_inl = DIV_ROUND_UP(inl, MLX5_SEND_WQE_DS); + if (WARN_ON_ONCE(ds_cnt_inl > MLX5E_MAX_TX_INLINE_DS)) + netdev_warn(skb->dev, "ds_cnt_inl = %u > max %u\n", ds_cnt_inl, + (u16)MLX5E_MAX_TX_INLINE_DS); ds_cnt += ds_cnt_inl; } From 8d4b475e9d0f100c3920d8bf7d392a6dac88583e Mon Sep 17 00:00:00 2001 From: Maxim Mikityanskiy Date: Wed, 2 Nov 2022 23:55:43 -0700 Subject: [PATCH 116/138] net/mlx5e: Fix usage of DMA sync API DMA sync functions should use the same direction that was used by DMA mapping. Use DMA_BIDIRECTIONAL for XDP_TX from regular RQ, which reuses the same mapping that was used for RX, and DMA_TO_DEVICE for XDP_TX from XSK RQ and XDP_REDIRECT, which establish a new mapping in this direction. On the RX side, use the same direction that was used when setting up the mapping (DMA_BIDIRECTIONAL for XDP, DMA_FROM_DEVICE otherwise). Also don't skip sync for device when establishing a DMA_FROM_DEVICE mapping for RX, as some architectures (ARM) may require invalidating caches before the device can use the mapping. It doesn't break the bugfix made in commit 0b7cfa4082fb ("net/mlx5e: Fix page DMA map/unmap attributes"), since the bug happened on unmap. Fixes: 0b7cfa4082fb ("net/mlx5e: Fix page DMA map/unmap attributes") Fixes: b5503b994ed5 ("net/mlx5e: XDP TX forwarding support") Signed-off-by: Maxim Mikityanskiy Reviewed-by: Gal Pressman Reviewed-by: Tariq Toukan Signed-off-by: Saeed Mahameed --- .../net/ethernet/mellanox/mlx5/core/en/xdp.c | 4 +-- .../net/ethernet/mellanox/mlx5/core/en_rx.c | 27 ++++++++++--------- 2 files changed, 16 insertions(+), 15 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c index 4685c652c97e..20507ef2f956 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c @@ -117,7 +117,7 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq, xdpi.page.rq = rq; dma_addr = page_pool_get_dma_addr(page) + (xdpf->data - (void *)xdpf); - dma_sync_single_for_device(sq->pdev, dma_addr, xdptxd.len, DMA_TO_DEVICE); + dma_sync_single_for_device(sq->pdev, dma_addr, xdptxd.len, DMA_BIDIRECTIONAL); if (unlikely(xdp_frame_has_frags(xdpf))) { sinfo = xdp_get_shared_info_from_frame(xdpf); @@ -131,7 +131,7 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq, skb_frag_off(frag); len = skb_frag_size(frag); dma_sync_single_for_device(sq->pdev, addr, len, - DMA_TO_DEVICE); + DMA_BIDIRECTIONAL); } } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index 58084650151f..a61a43fc8d5c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -266,7 +266,7 @@ static inline bool mlx5e_rx_cache_get(struct mlx5e_rq *rq, union mlx5e_alloc_uni addr = page_pool_get_dma_addr(au->page); /* Non-XSK always uses PAGE_SIZE. */ - dma_sync_single_for_device(rq->pdev, addr, PAGE_SIZE, DMA_FROM_DEVICE); + dma_sync_single_for_device(rq->pdev, addr, PAGE_SIZE, rq->buff.map_dir); return true; } @@ -282,8 +282,7 @@ static inline int mlx5e_page_alloc_pool(struct mlx5e_rq *rq, union mlx5e_alloc_u return -ENOMEM; /* Non-XSK always uses PAGE_SIZE. */ - addr = dma_map_page_attrs(rq->pdev, au->page, 0, PAGE_SIZE, - rq->buff.map_dir, DMA_ATTR_SKIP_CPU_SYNC); + addr = dma_map_page(rq->pdev, au->page, 0, PAGE_SIZE, rq->buff.map_dir); if (unlikely(dma_mapping_error(rq->pdev, addr))) { page_pool_recycle_direct(rq->page_pool, au->page); au->page = NULL; @@ -427,14 +426,15 @@ mlx5e_add_skb_frag(struct mlx5e_rq *rq, struct sk_buff *skb, { dma_addr_t addr = page_pool_get_dma_addr(au->page); - dma_sync_single_for_cpu(rq->pdev, addr + frag_offset, len, DMA_FROM_DEVICE); + dma_sync_single_for_cpu(rq->pdev, addr + frag_offset, len, + rq->buff.map_dir); page_ref_inc(au->page); skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, au->page, frag_offset, len, truesize); } static inline void -mlx5e_copy_skb_header(struct device *pdev, struct sk_buff *skb, +mlx5e_copy_skb_header(struct mlx5e_rq *rq, struct sk_buff *skb, struct page *page, dma_addr_t addr, int offset_from, int dma_offset, u32 headlen) { @@ -442,7 +442,8 @@ mlx5e_copy_skb_header(struct device *pdev, struct sk_buff *skb, /* Aligning len to sizeof(long) optimizes memcpy performance */ unsigned int len = ALIGN(headlen, sizeof(long)); - dma_sync_single_for_cpu(pdev, addr + dma_offset, len, DMA_FROM_DEVICE); + dma_sync_single_for_cpu(rq->pdev, addr + dma_offset, len, + rq->buff.map_dir); skb_copy_to_linear_data(skb, from, len); } @@ -1538,7 +1539,7 @@ mlx5e_skb_from_cqe_linear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi, addr = page_pool_get_dma_addr(au->page); dma_sync_single_range_for_cpu(rq->pdev, addr, wi->offset, - frag_size, DMA_FROM_DEVICE); + frag_size, rq->buff.map_dir); net_prefetch(data); prog = rcu_dereference(rq->xdp_prog); @@ -1587,7 +1588,7 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi addr = page_pool_get_dma_addr(au->page); dma_sync_single_range_for_cpu(rq->pdev, addr, wi->offset, - rq->buff.frame0_sz, DMA_FROM_DEVICE); + rq->buff.frame0_sz, rq->buff.map_dir); net_prefetchw(va); /* xdp_frame data area */ net_prefetch(va + rx_headroom); @@ -1608,7 +1609,7 @@ mlx5e_skb_from_cqe_nonlinear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi addr = page_pool_get_dma_addr(au->page); dma_sync_single_for_cpu(rq->pdev, addr + wi->offset, - frag_consumed_bytes, DMA_FROM_DEVICE); + frag_consumed_bytes, rq->buff.map_dir); if (!xdp_buff_has_frags(&xdp)) { /* Init on the first fragment to avoid cold cache access @@ -1905,7 +1906,7 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w mlx5e_fill_skb_data(skb, rq, au, byte_cnt, frag_offset); /* copy header */ addr = page_pool_get_dma_addr(head_au->page); - mlx5e_copy_skb_header(rq->pdev, skb, head_au->page, addr, + mlx5e_copy_skb_header(rq, skb, head_au->page, addr, head_offset, head_offset, headlen); /* skb linear part was allocated with headlen and aligned to long */ skb->tail += headlen; @@ -1939,7 +1940,7 @@ mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, addr = page_pool_get_dma_addr(au->page); dma_sync_single_range_for_cpu(rq->pdev, addr, head_offset, - frag_size, DMA_FROM_DEVICE); + frag_size, rq->buff.map_dir); net_prefetch(data); prog = rcu_dereference(rq->xdp_prog); @@ -1987,7 +1988,7 @@ mlx5e_skb_from_cqe_shampo(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, if (likely(frag_size <= BIT(MLX5E_SHAMPO_LOG_MAX_HEADER_ENTRY_SIZE))) { /* build SKB around header */ - dma_sync_single_range_for_cpu(rq->pdev, head->addr, 0, frag_size, DMA_FROM_DEVICE); + dma_sync_single_range_for_cpu(rq->pdev, head->addr, 0, frag_size, rq->buff.map_dir); prefetchw(hdr); prefetch(data); skb = mlx5e_build_linear_skb(rq, hdr, frag_size, rx_headroom, head_size, 0); @@ -2009,7 +2010,7 @@ mlx5e_skb_from_cqe_shampo(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, } prefetchw(skb->data); - mlx5e_copy_skb_header(rq->pdev, skb, head->page, head->addr, + mlx5e_copy_skb_header(rq, skb, head->page, head->addr, head_offset + rx_headroom, rx_headroom, head_size); /* skb linear part was allocated with headlen and aligned to long */ From 08912ea799cd2a43b8999457e316967fe4e2f327 Mon Sep 17 00:00:00 2001 From: Roi Dayan Date: Wed, 2 Nov 2022 23:55:44 -0700 Subject: [PATCH 117/138] net/mlx5e: Fix tc acts array not to be dependent on enum order The tc acts array should not be dependent on kernel internal flow action id enum. Fix the array initialization. Fixes: fad547906980 ("net/mlx5e: Add tc action infrastructure") Signed-off-by: Roi Dayan Reviewed-by: Maor Dickman Reviewed-by: Tariq Toukan Signed-off-by: Saeed Mahameed --- .../mellanox/mlx5/core/en/tc/act/act.c | 92 +++++++------------ 1 file changed, 32 insertions(+), 60 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c index 305fde62a78d..3337241cfd84 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc/act/act.c @@ -6,70 +6,42 @@ #include "en/tc_priv.h" #include "mlx5_core.h" -/* Must be aligned with enum flow_action_id. */ static struct mlx5e_tc_act *tc_acts_fdb[NUM_FLOW_ACTIONS] = { - &mlx5e_tc_act_accept, - &mlx5e_tc_act_drop, - &mlx5e_tc_act_trap, - &mlx5e_tc_act_goto, - &mlx5e_tc_act_mirred, - &mlx5e_tc_act_mirred, - &mlx5e_tc_act_redirect_ingress, - NULL, /* FLOW_ACTION_MIRRED_INGRESS, */ - &mlx5e_tc_act_vlan, - &mlx5e_tc_act_vlan, - &mlx5e_tc_act_vlan_mangle, - &mlx5e_tc_act_tun_encap, - &mlx5e_tc_act_tun_decap, - &mlx5e_tc_act_pedit, - &mlx5e_tc_act_pedit, - &mlx5e_tc_act_csum, - NULL, /* FLOW_ACTION_MARK, */ - &mlx5e_tc_act_ptype, - NULL, /* FLOW_ACTION_PRIORITY, */ - NULL, /* FLOW_ACTION_WAKE, */ - NULL, /* FLOW_ACTION_QUEUE, */ - &mlx5e_tc_act_sample, - &mlx5e_tc_act_police, - &mlx5e_tc_act_ct, - NULL, /* FLOW_ACTION_CT_METADATA, */ - &mlx5e_tc_act_mpls_push, - &mlx5e_tc_act_mpls_pop, - NULL, /* FLOW_ACTION_MPLS_MANGLE, */ - NULL, /* FLOW_ACTION_GATE, */ - NULL, /* FLOW_ACTION_PPPOE_PUSH, */ - NULL, /* FLOW_ACTION_JUMP, */ - NULL, /* FLOW_ACTION_PIPE, */ - &mlx5e_tc_act_vlan, - &mlx5e_tc_act_vlan, + [FLOW_ACTION_ACCEPT] = &mlx5e_tc_act_accept, + [FLOW_ACTION_DROP] = &mlx5e_tc_act_drop, + [FLOW_ACTION_TRAP] = &mlx5e_tc_act_trap, + [FLOW_ACTION_GOTO] = &mlx5e_tc_act_goto, + [FLOW_ACTION_REDIRECT] = &mlx5e_tc_act_mirred, + [FLOW_ACTION_MIRRED] = &mlx5e_tc_act_mirred, + [FLOW_ACTION_REDIRECT_INGRESS] = &mlx5e_tc_act_redirect_ingress, + [FLOW_ACTION_VLAN_PUSH] = &mlx5e_tc_act_vlan, + [FLOW_ACTION_VLAN_POP] = &mlx5e_tc_act_vlan, + [FLOW_ACTION_VLAN_MANGLE] = &mlx5e_tc_act_vlan_mangle, + [FLOW_ACTION_TUNNEL_ENCAP] = &mlx5e_tc_act_tun_encap, + [FLOW_ACTION_TUNNEL_DECAP] = &mlx5e_tc_act_tun_decap, + [FLOW_ACTION_MANGLE] = &mlx5e_tc_act_pedit, + [FLOW_ACTION_ADD] = &mlx5e_tc_act_pedit, + [FLOW_ACTION_CSUM] = &mlx5e_tc_act_csum, + [FLOW_ACTION_PTYPE] = &mlx5e_tc_act_ptype, + [FLOW_ACTION_SAMPLE] = &mlx5e_tc_act_sample, + [FLOW_ACTION_POLICE] = &mlx5e_tc_act_police, + [FLOW_ACTION_CT] = &mlx5e_tc_act_ct, + [FLOW_ACTION_MPLS_PUSH] = &mlx5e_tc_act_mpls_push, + [FLOW_ACTION_MPLS_POP] = &mlx5e_tc_act_mpls_pop, + [FLOW_ACTION_VLAN_PUSH_ETH] = &mlx5e_tc_act_vlan, + [FLOW_ACTION_VLAN_POP_ETH] = &mlx5e_tc_act_vlan, }; -/* Must be aligned with enum flow_action_id. */ static struct mlx5e_tc_act *tc_acts_nic[NUM_FLOW_ACTIONS] = { - &mlx5e_tc_act_accept, - &mlx5e_tc_act_drop, - NULL, /* FLOW_ACTION_TRAP, */ - &mlx5e_tc_act_goto, - &mlx5e_tc_act_mirred_nic, - NULL, /* FLOW_ACTION_MIRRED, */ - NULL, /* FLOW_ACTION_REDIRECT_INGRESS, */ - NULL, /* FLOW_ACTION_MIRRED_INGRESS, */ - NULL, /* FLOW_ACTION_VLAN_PUSH, */ - NULL, /* FLOW_ACTION_VLAN_POP, */ - NULL, /* FLOW_ACTION_VLAN_MANGLE, */ - NULL, /* FLOW_ACTION_TUNNEL_ENCAP, */ - NULL, /* FLOW_ACTION_TUNNEL_DECAP, */ - &mlx5e_tc_act_pedit, - &mlx5e_tc_act_pedit, - &mlx5e_tc_act_csum, - &mlx5e_tc_act_mark, - NULL, /* FLOW_ACTION_PTYPE, */ - NULL, /* FLOW_ACTION_PRIORITY, */ - NULL, /* FLOW_ACTION_WAKE, */ - NULL, /* FLOW_ACTION_QUEUE, */ - NULL, /* FLOW_ACTION_SAMPLE, */ - NULL, /* FLOW_ACTION_POLICE, */ - &mlx5e_tc_act_ct, + [FLOW_ACTION_ACCEPT] = &mlx5e_tc_act_accept, + [FLOW_ACTION_DROP] = &mlx5e_tc_act_drop, + [FLOW_ACTION_GOTO] = &mlx5e_tc_act_goto, + [FLOW_ACTION_REDIRECT] = &mlx5e_tc_act_mirred_nic, + [FLOW_ACTION_MANGLE] = &mlx5e_tc_act_pedit, + [FLOW_ACTION_ADD] = &mlx5e_tc_act_pedit, + [FLOW_ACTION_CSUM] = &mlx5e_tc_act_csum, + [FLOW_ACTION_MARK] = &mlx5e_tc_act_mark, + [FLOW_ACTION_CT] = &mlx5e_tc_act_ct, }; /** From 9e06430841363a1d2932d546fdce1cc5edb3c2a0 Mon Sep 17 00:00:00 2001 From: Jianbo Liu Date: Wed, 2 Nov 2022 23:55:45 -0700 Subject: [PATCH 118/138] net/mlx5e: TC, Fix wrong rejection of packet-per-second policing In the bellow commit, we added support for PPS policing without removing the check which block offload of such cases. Fix it by removing this check. Fixes: a8d52b024d6d ("net/mlx5e: TC, Support offloading police action") Signed-off-by: Jianbo Liu Reviewed-by: Maor Dickman Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index dd6fea9e9a5b..372dfb89e396 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -4758,12 +4758,6 @@ int mlx5e_policer_validate(const struct flow_action *action, return -EOPNOTSUPP; } - if (act->police.rate_pkt_ps) { - NL_SET_ERR_MSG_MOD(extack, - "QoS offload not support packets per second"); - return -EOPNOTSUPP; - } - return 0; } From f4f4096b410e8d31c3f07f39de3b17d144edd53d Mon Sep 17 00:00:00 2001 From: Roi Dayan Date: Wed, 2 Nov 2022 23:55:46 -0700 Subject: [PATCH 119/138] net/mlx5e: E-Switch, Fix comparing termination table instance The pkt_reformat pointer being saved under flow_act and not dest attribute in the termination table instance. Fix the comparison pointers. Also fix returning success if one pkt_reformat pointer is null and the other is not. Fixes: 249ccc3c95bd ("net/mlx5e: Add support for offloading traffic from uplink to uplink") Signed-off-by: Roi Dayan Reviewed-by: Chris Mi Signed-off-by: Saeed Mahameed --- .../mellanox/mlx5/core/eswitch_offloads_termtbl.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c index ee568bf34ae2..108a3503f413 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads_termtbl.c @@ -30,9 +30,9 @@ mlx5_eswitch_termtbl_hash(struct mlx5_flow_act *flow_act, sizeof(dest->vport.num), hash); hash = jhash((const void *)&dest->vport.vhca_id, sizeof(dest->vport.num), hash); - if (dest->vport.pkt_reformat) - hash = jhash(dest->vport.pkt_reformat, - sizeof(*dest->vport.pkt_reformat), + if (flow_act->pkt_reformat) + hash = jhash(flow_act->pkt_reformat, + sizeof(*flow_act->pkt_reformat), hash); return hash; } @@ -53,9 +53,11 @@ mlx5_eswitch_termtbl_cmp(struct mlx5_flow_act *flow_act1, if (ret) return ret; - return dest1->vport.pkt_reformat && dest2->vport.pkt_reformat ? - memcmp(dest1->vport.pkt_reformat, dest2->vport.pkt_reformat, - sizeof(*dest1->vport.pkt_reformat)) : 0; + if (flow_act1->pkt_reformat && flow_act2->pkt_reformat) + return memcmp(flow_act1->pkt_reformat, flow_act2->pkt_reformat, + sizeof(*flow_act1->pkt_reformat)); + + return !(flow_act1->pkt_reformat == flow_act2->pkt_reformat); } static int From 7f1a6d4b9e820b08479a07f6e14c36ccfa641751 Mon Sep 17 00:00:00 2001 From: Roi Dayan Date: Wed, 2 Nov 2022 23:55:47 -0700 Subject: [PATCH 120/138] net/mlx5e: TC, Fix slab-out-of-bounds in parse_tc_actions esw_attr is only allocated if namespace is fdb. BUG: KASAN: slab-out-of-bounds in parse_tc_actions+0xdc6/0x10e0 [mlx5_core] Write of size 4 at addr ffff88815f185b04 by task tc/2135 CPU: 5 PID: 2135 Comm: tc Not tainted 6.1.0-rc2+ #2 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 Call Trace: dump_stack_lvl+0x57/0x7d print_report+0x170/0x471 ? parse_tc_actions+0xdc6/0x10e0 [mlx5_core] kasan_report+0xbc/0xf0 ? parse_tc_actions+0xdc6/0x10e0 [mlx5_core] parse_tc_actions+0xdc6/0x10e0 [mlx5_core] Fixes: 94d651739e17 ("net/mlx5e: TC, Fix cloned flow attr instance dests are not zeroed") Signed-off-by: Roi Dayan Reviewed-by: Maor Dickman Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 372dfb89e396..5a6aa61ec82a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -3633,10 +3633,14 @@ mlx5e_clone_flow_attr_for_post_act(struct mlx5_flow_attr *attr, attr2->action = 0; attr2->flags = 0; attr2->parse_attr = parse_attr; - attr2->esw_attr->out_count = 0; - attr2->esw_attr->split_count = 0; attr2->dest_chain = 0; attr2->dest_ft = NULL; + + if (ns_type == MLX5_FLOW_NAMESPACE_FDB) { + attr2->esw_attr->out_count = 0; + attr2->esw_attr->split_count = 0; + } + return attr2; } From f23df5220d2bf8d5e639f074b76f206a736d09e1 Mon Sep 17 00:00:00 2001 From: Norbert Zulinski Date: Mon, 10 Oct 2022 10:22:22 -0400 Subject: [PATCH 121/138] ice: Fix spurious interrupt during removal of trusted VF Previously, during removal of trusted VF when VF is down there was number of spurious interrupt equal to number of queues on VF. Add check if VF already has inactive queues. If VF is disabled and has inactive rx queues then do not disable rx queues. Add check in ice_vsi_stop_tx_ring if it's VF's vsi and if VF is disabled. Fixes: efe41860008e ("ice: Fix memory corruption in VF driver") Signed-off-by: Norbert Zulinski Signed-off-by: Mateusz Palczewski Tested-by: Konrad Jankowski Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_base.c | 2 +- drivers/net/ethernet/intel/ice/ice_lib.c | 25 +++++++++++++++++++++ drivers/net/ethernet/intel/ice/ice_lib.h | 1 + drivers/net/ethernet/intel/ice/ice_vf_lib.c | 5 ++++- 4 files changed, 31 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c index 9e36f01dfa4f..e864634d66bc 100644 --- a/drivers/net/ethernet/intel/ice/ice_base.c +++ b/drivers/net/ethernet/intel/ice/ice_base.c @@ -958,7 +958,7 @@ ice_vsi_stop_tx_ring(struct ice_vsi *vsi, enum ice_disq_rst_src rst_src, * associated to the queue to schedule NAPI handler */ q_vector = ring->q_vector; - if (q_vector) + if (q_vector && !(vsi->vf && ice_is_vf_disabled(vsi->vf))) ice_trigger_sw_intr(hw, q_vector); status = ice_dis_vsi_txq(vsi->port_info, txq_meta->vsi_idx, diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c index 938ba8c215cb..7276badfa19e 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_lib.c @@ -2239,6 +2239,31 @@ int ice_vsi_stop_xdp_tx_rings(struct ice_vsi *vsi) return ice_vsi_stop_tx_rings(vsi, ICE_NO_RESET, 0, vsi->xdp_rings, vsi->num_xdp_txq); } +/** + * ice_vsi_is_rx_queue_active + * @vsi: the VSI being configured + * + * Return true if at least one queue is active. + */ +bool ice_vsi_is_rx_queue_active(struct ice_vsi *vsi) +{ + struct ice_pf *pf = vsi->back; + struct ice_hw *hw = &pf->hw; + int i; + + ice_for_each_rxq(vsi, i) { + u32 rx_reg; + int pf_q; + + pf_q = vsi->rxq_map[i]; + rx_reg = rd32(hw, QRX_CTRL(pf_q)); + if (rx_reg & QRX_CTRL_QENA_STAT_M) + return true; + } + + return false; +} + /** * ice_vsi_is_vlan_pruning_ena - check if VLAN pruning is enabled or not * @vsi: VSI to check whether or not VLAN pruning is enabled. diff --git a/drivers/net/ethernet/intel/ice/ice_lib.h b/drivers/net/ethernet/intel/ice/ice_lib.h index ec4bf0c89857..dcdf69a693e9 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.h +++ b/drivers/net/ethernet/intel/ice/ice_lib.h @@ -129,4 +129,5 @@ u16 ice_vsi_num_non_zero_vlans(struct ice_vsi *vsi); bool ice_is_feature_supported(struct ice_pf *pf, enum ice_feature f); void ice_clear_feature_support(struct ice_pf *pf, enum ice_feature f); void ice_init_feature_support(struct ice_pf *pf); +bool ice_vsi_is_rx_queue_active(struct ice_vsi *vsi); #endif /* !_ICE_LIB_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_vf_lib.c b/drivers/net/ethernet/intel/ice/ice_vf_lib.c index 0abeed092de1..1c51778db951 100644 --- a/drivers/net/ethernet/intel/ice/ice_vf_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_vf_lib.c @@ -576,7 +576,10 @@ int ice_reset_vf(struct ice_vf *vf, u32 flags) return -EINVAL; } ice_vsi_stop_lan_tx_rings(vsi, ICE_NO_RESET, vf->vf_id); - ice_vsi_stop_all_rx_rings(vsi); + + if (ice_vsi_is_rx_queue_active(vsi)) + ice_vsi_stop_all_rx_rings(vsi); + dev_dbg(dev, "VF is already disabled, there is no need for resetting it, telling VM, all is fine %d\n", vf->vf_id); return 0; From 0e710a3ffd0caaf23b8791b041e8792f252f8e4f Mon Sep 17 00:00:00 2001 From: Michal Jaron Date: Fri, 14 Oct 2022 10:45:37 +0200 Subject: [PATCH 122/138] iavf: Fix VF driver counting VLAN 0 filters VF driver mistakenly counts VLAN 0 filters, when no PF driver counts them. Do not count VLAN 0 filters, when VLAN_V2 is engaged. Counting those filters in, will affect filters size by -1, when sending batched VLAN addition message. Fixes: 968996c070ef ("iavf: Fix VLAN_V2 addition/rejection") Signed-off-by: Przemyslaw Patynowski Signed-off-by: Michal Jaron Signed-off-by: Kamil Maziarz Tested-by: Konrad Jankowski Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/iavf/iavf_virtchnl.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c b/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c index 5a9e6563923e..24a701fd140e 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c +++ b/drivers/net/ethernet/intel/iavf/iavf_virtchnl.c @@ -2438,6 +2438,8 @@ void iavf_virtchnl_completion(struct iavf_adapter *adapter, list_for_each_entry(f, &adapter->vlan_filter_list, list) { if (f->is_new_vlan) { f->is_new_vlan = false; + if (!f->vlan.vid) + continue; if (f->vlan.tpid == ETH_P_8021Q) set_bit(f->vlan.vid, adapter->vsi.active_cvlans); From 6d47b53fb3f363a74538a1dbd09954af3d8d4131 Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Wed, 9 Nov 2022 09:15:37 +0800 Subject: [PATCH 123/138] net: cpsw: disable napi in cpsw_ndo_open() When failed to create xdp rxqs or fill rx channels in cpsw_ndo_open() for opening device, napi isn't disabled. When open cpsw device next time, it will report a invalid opcode issue. Compiled tested only. Fixes: d354eb85d618 ("drivers: net: cpsw: dual_emac: simplify napi usage") Signed-off-by: Zhengchao Shao Link: https://lore.kernel.org/r/20221109011537.96975-1-shaozhengchao@huawei.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/ti/cpsw.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index 709ca6dd6ecb..13c9c2d6b79b 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -854,6 +854,8 @@ static int cpsw_ndo_open(struct net_device *ndev) err_cleanup: if (!cpsw->usage_count) { + napi_disable(&cpsw->napi_rx); + napi_disable(&cpsw->napi_tx); cpdma_ctlr_stop(cpsw->dma); cpsw_destroy_xdp_rxqs(cpsw); } From d75aed1428da787cbe42bc073d76f1354f364d92 Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Wed, 9 Nov 2022 10:14:51 +0800 Subject: [PATCH 124/138] net: cxgb3_main: disable napi when bind qsets failed in cxgb_up() When failed to bind qsets in cxgb_up() for opening device, napi isn't disabled. When open cxgb3 device next time, it will trigger a BUG_ON() in napi_enable(). Compile tested only. Fixes: 48c4b6dbb7e2 ("cxgb3 - fix port up/down error path") Signed-off-by: Zhengchao Shao Link: https://lore.kernel.org/r/20221109021451.121490-1-shaozhengchao@huawei.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c index a52e6b6e2876..9b84c8d8d309 100644 --- a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c +++ b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c @@ -1301,6 +1301,7 @@ static int cxgb_up(struct adapter *adap) if (ret < 0) { CH_ERR(adap, "failed to bind qsets, err %d\n", ret); t3_intr_disable(adap); + quiesce_rx(adap); free_irq_resources(adap); err = ret; goto out; From dcea1a8107c04b9521dee1dd37971757a22db162 Mon Sep 17 00:00:00 2001 From: "Tan, Tee Min" Date: Mon, 7 Nov 2022 21:08:11 -0500 Subject: [PATCH 125/138] stmmac: intel: Update PCH PTP clock rate from 200MHz to 204.8MHz Current Intel platform has an output of ~976ms interval when probed on 1 Pulse-per-Second(PPS) hardware pin. The correct PTP clock frequency for PCH GbE should be 204.8MHz instead of 200MHz. PSE GbE PTP clock rate remains at 200MHz. Fixes: 58da0cfa6cf1 ("net: stmmac: create dwmac-intel.c to contain all Intel platform") Signed-off-by: Ling Pei Lee Signed-off-by: Tan, Tee Min Signed-off-by: Voon Weifeng Signed-off-by: Gan Yi Fang Link: https://lore.kernel.org/r/20221108020811.12919-1-yi.fang.gan@intel.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c index 0a2afc1a3124..7deb1f817dac 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c @@ -629,7 +629,6 @@ static int ehl_common_data(struct pci_dev *pdev, { plat->rx_queues_to_use = 8; plat->tx_queues_to_use = 8; - plat->clk_ptp_rate = 200000000; plat->use_phy_wol = 1; plat->safety_feat_cfg->tsoee = 1; @@ -654,6 +653,8 @@ static int ehl_sgmii_data(struct pci_dev *pdev, plat->serdes_powerup = intel_serdes_powerup; plat->serdes_powerdown = intel_serdes_powerdown; + plat->clk_ptp_rate = 204800000; + return ehl_common_data(pdev, plat); } @@ -667,6 +668,8 @@ static int ehl_rgmii_data(struct pci_dev *pdev, plat->bus_id = 1; plat->phy_interface = PHY_INTERFACE_MODE_RGMII; + plat->clk_ptp_rate = 204800000; + return ehl_common_data(pdev, plat); } @@ -683,6 +686,8 @@ static int ehl_pse0_common_data(struct pci_dev *pdev, plat->bus_id = 2; plat->addr64 = 32; + plat->clk_ptp_rate = 200000000; + intel_mgbe_pse_crossts_adj(intel_priv, EHL_PSE_ART_MHZ); return ehl_common_data(pdev, plat); @@ -722,6 +727,8 @@ static int ehl_pse1_common_data(struct pci_dev *pdev, plat->bus_id = 3; plat->addr64 = 32; + plat->clk_ptp_rate = 200000000; + intel_mgbe_pse_crossts_adj(intel_priv, EHL_PSE_ART_MHZ); return ehl_common_data(pdev, plat); @@ -757,7 +764,7 @@ static int tgl_common_data(struct pci_dev *pdev, { plat->rx_queues_to_use = 6; plat->tx_queues_to_use = 4; - plat->clk_ptp_rate = 200000000; + plat->clk_ptp_rate = 204800000; plat->speed_mode_2500 = intel_speed_mode_2500; plat->safety_feat_cfg->tsoee = 1; From d4072058af4fd8fb4658e7452289042a406a9398 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 8 Nov 2022 09:55:17 +0000 Subject: [PATCH 126/138] mctp: Fix an error handling path in mctp_init() If mctp_neigh_init() return error, the routes resources should be released in the error handling path. Otherwise some resources leak. Fixes: 4d8b9319282a ("mctp: Add neighbour implementation") Signed-off-by: Wei Yongjun Acked-by: Matt Johnston Link: https://lore.kernel.org/r/20221108095517.620115-1-weiyongjun@huaweicloud.com Signed-off-by: Jakub Kicinski --- net/mctp/af_mctp.c | 4 +++- net/mctp/route.c | 2 +- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/net/mctp/af_mctp.c b/net/mctp/af_mctp.c index b6b5e496fa40..fc9e728b6333 100644 --- a/net/mctp/af_mctp.c +++ b/net/mctp/af_mctp.c @@ -665,12 +665,14 @@ static __init int mctp_init(void) rc = mctp_neigh_init(); if (rc) - goto err_unreg_proto; + goto err_unreg_routes; mctp_device_init(); return 0; +err_unreg_routes: + mctp_routes_exit(); err_unreg_proto: proto_unregister(&mctp_proto); err_unreg_sock: diff --git a/net/mctp/route.c b/net/mctp/route.c index 2155f15a074c..f9a80b82dc51 100644 --- a/net/mctp/route.c +++ b/net/mctp/route.c @@ -1400,7 +1400,7 @@ int __init mctp_routes_init(void) return register_pernet_subsys(&mctp_net_ops); } -void __exit mctp_routes_exit(void) +void mctp_routes_exit(void) { unregister_pernet_subsys(&mctp_net_ops); rtnl_unregister(PF_MCTP, RTM_DELROUTE); From c6092ea1e6d7bd12acd881f6aa2b5054cd70e096 Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Wed, 9 Nov 2022 09:21:00 +0800 Subject: [PATCH 127/138] cxgb4vf: shut down the adapter when t4vf_update_port_info() failed in cxgb4vf_open() When t4vf_update_port_info() failed in cxgb4vf_open(), resources applied during adapter goes up are not cleared. Fix it. Only be compiled, not be tested. Fixes: 18d79f721e0a ("cxgb4vf: Update port information in cxgb4vf_open()") Signed-off-by: Zhengchao Shao Link: https://lore.kernel.org/r/20221109012100.99132-1-shaozhengchao@huawei.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c index 54db79f4dcfe..63b2bd084130 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c @@ -858,7 +858,7 @@ static int cxgb4vf_open(struct net_device *dev) */ err = t4vf_update_port_info(pi); if (err < 0) - return err; + goto err_unwind; /* * Note that this interface is up and start everything up ... From f2d45fdf9a0ed2c94c01c422a0d0add8ffd42099 Mon Sep 17 00:00:00 2001 From: Yang Yingliang Date: Tue, 8 Nov 2022 19:46:45 +0800 Subject: [PATCH 128/138] stmmac: dwmac-loongson: fix missing pci_disable_msi() while module exiting pci_enable_msi() has been called in loongson_dwmac_probe(), so pci_disable_msi() needs be called in remove path and error path of probe(). Fixes: 30bba69d7db4 ("stmmac: pci: Add dwmac support for Loongson") Signed-off-by: Yang Yingliang Signed-off-by: Paolo Abeni --- .../net/ethernet/stmicro/stmmac/dwmac-loongson.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c index 79fa7870563b..16915b4d9505 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c @@ -125,6 +125,7 @@ static int loongson_dwmac_probe(struct pci_dev *pdev, const struct pci_device_id if (res.irq < 0) { dev_err(&pdev->dev, "IRQ macirq not found\n"); ret = -ENODEV; + goto err_disable_msi; } res.wol_irq = of_irq_get_byname(np, "eth_wake_irq"); @@ -137,9 +138,18 @@ static int loongson_dwmac_probe(struct pci_dev *pdev, const struct pci_device_id if (res.lpi_irq < 0) { dev_err(&pdev->dev, "IRQ eth_lpi not found\n"); ret = -ENODEV; + goto err_disable_msi; } - return stmmac_dvr_probe(&pdev->dev, plat, &res); + ret = stmmac_dvr_probe(&pdev->dev, plat, &res); + if (ret) + goto err_disable_msi; + + return ret; + +err_disable_msi: + pci_disable_msi(pdev); + return ret; } static void loongson_dwmac_remove(struct pci_dev *pdev) @@ -155,6 +165,7 @@ static void loongson_dwmac_remove(struct pci_dev *pdev) break; } + pci_disable_msi(pdev); pci_disable_device(pdev); } From fe5b3ce8b4377e543960220f539b989a927afd8a Mon Sep 17 00:00:00 2001 From: Yang Yingliang Date: Tue, 8 Nov 2022 19:46:46 +0800 Subject: [PATCH 129/138] stmmac: dwmac-loongson: fix missing pci_disable_device() in loongson_dwmac_probe() Add missing pci_disable_device() in the error path in loongson_dwmac_probe(). Fixes: 30bba69d7db4 ("stmmac: pci: Add dwmac support for Loongson") Signed-off-by: Yang Yingliang Signed-off-by: Paolo Abeni --- drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c index 16915b4d9505..2d480bc49c51 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c @@ -97,7 +97,7 @@ static int loongson_dwmac_probe(struct pci_dev *pdev, const struct pci_device_id continue; ret = pcim_iomap_regions(pdev, BIT(0), pci_name(pdev)); if (ret) - return ret; + goto err_disable_device; break; } @@ -108,7 +108,8 @@ static int loongson_dwmac_probe(struct pci_dev *pdev, const struct pci_device_id phy_mode = device_get_phy_mode(&pdev->dev); if (phy_mode < 0) { dev_err(&pdev->dev, "phy_mode not found\n"); - return phy_mode; + ret = phy_mode; + goto err_disable_device; } plat->phy_interface = phy_mode; @@ -149,6 +150,8 @@ static int loongson_dwmac_probe(struct pci_dev *pdev, const struct pci_device_id err_disable_msi: pci_disable_msi(pdev); +err_disable_device: + pci_disable_device(pdev); return ret; } From 7f94d0498f9c763f37172c08059ae91804c3075a Mon Sep 17 00:00:00 2001 From: Yang Yingliang Date: Tue, 8 Nov 2022 19:46:47 +0800 Subject: [PATCH 130/138] stmmac: dwmac-loongson: fix missing of_node_put() while module exiting The node returned by of_get_child_by_name() with refcount decremented, of_node_put() needs be called when finish using it. So add it in the error path in loongson_dwmac_probe() and in loongson_dwmac_remove(). Fixes: 2ae34111fe4e ("stmmac: dwmac-loongson: fix invalid mdio_node") Signed-off-by: Yang Yingliang Signed-off-by: Paolo Abeni --- .../ethernet/stmicro/stmmac/dwmac-loongson.c | 19 ++++++++++++++----- 1 file changed, 14 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c index 2d480bc49c51..a25c187d3185 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-loongson.c @@ -75,20 +75,24 @@ static int loongson_dwmac_probe(struct pci_dev *pdev, const struct pci_device_id plat->mdio_bus_data = devm_kzalloc(&pdev->dev, sizeof(*plat->mdio_bus_data), GFP_KERNEL); - if (!plat->mdio_bus_data) - return -ENOMEM; + if (!plat->mdio_bus_data) { + ret = -ENOMEM; + goto err_put_node; + } plat->mdio_bus_data->needs_reset = true; } plat->dma_cfg = devm_kzalloc(&pdev->dev, sizeof(*plat->dma_cfg), GFP_KERNEL); - if (!plat->dma_cfg) - return -ENOMEM; + if (!plat->dma_cfg) { + ret = -ENOMEM; + goto err_put_node; + } /* Enable pci device */ ret = pci_enable_device(pdev); if (ret) { dev_err(&pdev->dev, "%s: ERROR: failed to enable device\n", __func__); - return ret; + goto err_put_node; } /* Get the base address of device */ @@ -152,13 +156,18 @@ static int loongson_dwmac_probe(struct pci_dev *pdev, const struct pci_device_id pci_disable_msi(pdev); err_disable_device: pci_disable_device(pdev); +err_put_node: + of_node_put(plat->mdio_node); return ret; } static void loongson_dwmac_remove(struct pci_dev *pdev) { + struct net_device *ndev = dev_get_drvdata(&pdev->dev); + struct stmmac_priv *priv = netdev_priv(ndev); int i; + of_node_put(priv->plat->mdio_node); stmmac_dvr_remove(&pdev->dev); for (i = 0; i < PCI_STD_NUM_BARS; i++) { From 1b16b3fdf675cca15a537572bac50cc5354368fc Mon Sep 17 00:00:00 2001 From: Antoine Tenart Date: Tue, 8 Nov 2022 16:34:58 +0100 Subject: [PATCH 131/138] net: phy: mscc: macsec: clear encryption keys when freeing a flow Commit aaab73f8fba4 ("macsec: clear encryption keys from the stack after setting up offload") made sure to clean encryption keys from the stack after setting up offloading, but the MSCC PHY driver made a copy, kept it in the flow data and did not clear it when freeing a flow. Fix this. Fixes: 28c5107aa904 ("net: phy: mscc: macsec support") Signed-off-by: Antoine Tenart Signed-off-by: Paolo Abeni --- drivers/net/phy/mscc/mscc_macsec.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/phy/mscc/mscc_macsec.c b/drivers/net/phy/mscc/mscc_macsec.c index ee5b17edca39..f81b077618f4 100644 --- a/drivers/net/phy/mscc/mscc_macsec.c +++ b/drivers/net/phy/mscc/mscc_macsec.c @@ -632,6 +632,7 @@ static void vsc8584_macsec_free_flow(struct vsc8531_private *priv, list_del(&flow->list); clear_bit(flow->index, bitmap); + memzero_explicit(flow->key, sizeof(flow->key)); kfree(flow); } From 879785def0f5e71d54399de0f8a5cb399db14171 Mon Sep 17 00:00:00 2001 From: Antoine Tenart Date: Tue, 8 Nov 2022 16:34:59 +0100 Subject: [PATCH 132/138] net: atlantic: macsec: clear encryption keys from the stack Commit aaab73f8fba4 ("macsec: clear encryption keys from the stack after setting up offload") made sure to clean encryption keys from the stack after setting up offloading, but the atlantic driver made a copy and did not clear it. Fix this. [4 Fixes tags below, all part of the same series, no need to split this] Fixes: 9ff40a751a6f ("net: atlantic: MACSec ingress offload implementation") Fixes: b8f8a0b7b5cb ("net: atlantic: MACSec ingress offload HW bindings") Fixes: 27736563ce32 ("net: atlantic: MACSec egress offload implementation") Fixes: 9d106c6dd81b ("net: atlantic: MACSec egress offload HW bindings") Signed-off-by: Antoine Tenart Signed-off-by: Paolo Abeni --- .../net/ethernet/aquantia/atlantic/aq_macsec.c | 2 ++ .../aquantia/atlantic/macsec/macsec_api.c | 18 +++++++++++------- 2 files changed, 13 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_macsec.c b/drivers/net/ethernet/aquantia/atlantic/aq_macsec.c index a0180811305d..7eb5851eb95d 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_macsec.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_macsec.c @@ -570,6 +570,7 @@ static int aq_update_txsa(struct aq_nic_s *nic, const unsigned int sc_idx, ret = aq_mss_set_egress_sakey_record(hw, &key_rec, sa_idx); + memzero_explicit(&key_rec, sizeof(key_rec)); return ret; } @@ -899,6 +900,7 @@ static int aq_update_rxsa(struct aq_nic_s *nic, const unsigned int sc_idx, ret = aq_mss_set_ingress_sakey_record(hw, &sa_key_record, sa_idx); + memzero_explicit(&sa_key_record, sizeof(sa_key_record)); return ret; } diff --git a/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.c b/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.c index 36c7cf05630a..431924959520 100644 --- a/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.c +++ b/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.c @@ -757,6 +757,7 @@ set_ingress_sakey_record(struct aq_hw_s *hw, u16 table_index) { u16 packed_record[18]; + int ret; if (table_index >= NUMROWS_INGRESSSAKEYRECORD) return -EINVAL; @@ -789,9 +790,12 @@ set_ingress_sakey_record(struct aq_hw_s *hw, packed_record[16] = rec->key_len & 0x3; - return set_raw_ingress_record(hw, packed_record, 18, 2, - ROWOFFSET_INGRESSSAKEYRECORD + - table_index); + ret = set_raw_ingress_record(hw, packed_record, 18, 2, + ROWOFFSET_INGRESSSAKEYRECORD + + table_index); + + memzero_explicit(packed_record, sizeof(packed_record)); + return ret; } int aq_mss_set_ingress_sakey_record(struct aq_hw_s *hw, @@ -1739,14 +1743,14 @@ static int set_egress_sakey_record(struct aq_hw_s *hw, ret = set_raw_egress_record(hw, packed_record, 8, 2, ROWOFFSET_EGRESSSAKEYRECORD + table_index); if (unlikely(ret)) - return ret; + goto clear_key; ret = set_raw_egress_record(hw, packed_record + 8, 8, 2, ROWOFFSET_EGRESSSAKEYRECORD + table_index - 32); - if (unlikely(ret)) - return ret; - return 0; +clear_key: + memzero_explicit(packed_record, sizeof(packed_record)); + return ret; } int aq_mss_set_egress_sakey_record(struct aq_hw_s *hw, From 0348c1ab980c1d43fb37b758d4b760990c066cb5 Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Wed, 9 Nov 2022 10:37:41 +0800 Subject: [PATCH 133/138] ethernet: s2io: disable napi when start nic failed in s2io_card_up() When failed to start nic or add interrupt service routine in s2io_card_up() for opening device, napi isn't disabled. When open s2io device next time, it will trigger a BUG_ON()in napi_enable(). Compile tested only. Fixes: 5f490c968056 ("S2io: Fixed synchronization between scheduling of napi with card reset and close") Signed-off-by: Zhengchao Shao Link: https://lore.kernel.org/r/20221109023741.131552-1-shaozhengchao@huawei.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/neterion/s2io.c | 29 +++++++++++++++++++--------- 1 file changed, 20 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/neterion/s2io.c b/drivers/net/ethernet/neterion/s2io.c index dcf8212119f9..1d3c4474b7cb 100644 --- a/drivers/net/ethernet/neterion/s2io.c +++ b/drivers/net/ethernet/neterion/s2io.c @@ -7128,9 +7128,8 @@ static int s2io_card_up(struct s2io_nic *sp) if (ret) { DBG_PRINT(ERR_DBG, "%s: Out of memory in Open\n", dev->name); - s2io_reset(sp); - free_rx_buffers(sp); - return -ENOMEM; + ret = -ENOMEM; + goto err_fill_buff; } DBG_PRINT(INFO_DBG, "Buf in ring:%d is %d:\n", i, ring->rx_bufs_left); @@ -7168,18 +7167,16 @@ static int s2io_card_up(struct s2io_nic *sp) /* Enable Rx Traffic and interrupts on the NIC */ if (start_nic(sp)) { DBG_PRINT(ERR_DBG, "%s: Starting NIC failed\n", dev->name); - s2io_reset(sp); - free_rx_buffers(sp); - return -ENODEV; + ret = -ENODEV; + goto err_out; } /* Add interrupt service routine */ if (s2io_add_isr(sp) != 0) { if (sp->config.intr_type == MSI_X) s2io_rem_isr(sp); - s2io_reset(sp); - free_rx_buffers(sp); - return -ENODEV; + ret = -ENODEV; + goto err_out; } timer_setup(&sp->alarm_timer, s2io_alarm_handle, 0); @@ -7199,6 +7196,20 @@ static int s2io_card_up(struct s2io_nic *sp) } return 0; + +err_out: + if (config->napi) { + if (config->intr_type == MSI_X) { + for (i = 0; i < sp->config.rx_ring_num; i++) + napi_disable(&sp->mac_control.rings[i].napi); + } else { + napi_disable(&sp->napi); + } + } +err_fill_buff: + s2io_reset(sp); + free_rx_buffers(sp); + return ret; } /** From f111606b63ff2282428ffbac0447c871eb957b6c Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Wed, 9 Nov 2022 10:54:32 +0800 Subject: [PATCH 134/138] net: mv643xx_eth: disable napi when init rxq or txq failed in mv643xx_eth_open() When failed to init rxq or txq in mv643xx_eth_open() for opening device, napi isn't disabled. When open mv643xx_eth device next time, it will trigger a BUG_ON() in napi_enable(). Compile tested only. Fixes: 2257e05c1705 ("mv643xx_eth: get rid of receive-side locking") Signed-off-by: Zhengchao Shao Link: https://lore.kernel.org/r/20221109025432.80900-1-shaozhengchao@huawei.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/marvell/mv643xx_eth.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/marvell/mv643xx_eth.c b/drivers/net/ethernet/marvell/mv643xx_eth.c index 707993b445d1..8941f69d93e9 100644 --- a/drivers/net/ethernet/marvell/mv643xx_eth.c +++ b/drivers/net/ethernet/marvell/mv643xx_eth.c @@ -2481,6 +2481,7 @@ static int mv643xx_eth_open(struct net_device *dev) for (i = 0; i < mp->rxq_count; i++) rxq_deinit(mp->rxq + i); out: + napi_disable(&mp->napi); free_irq(dev->irq, dev); return err; From acce40037041f97baad18142bb253064491ebde3 Mon Sep 17 00:00:00 2001 From: Zhengchao Shao Date: Wed, 9 Nov 2022 12:40:16 +0800 Subject: [PATCH 135/138] ethernet: tundra: free irq when alloc ring failed in tsi108_open() When alloc tx/rx ring failed in tsi108_open(), it doesn't free irq. Fix it. Fixes: 5e123b844a1c ("[PATCH] Add tsi108/9 On Chip Ethernet device driver support") Signed-off-by: Zhengchao Shao Link: https://lore.kernel.org/r/20221109044016.126866-1-shaozhengchao@huawei.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/tundra/tsi108_eth.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/tundra/tsi108_eth.c b/drivers/net/ethernet/tundra/tsi108_eth.c index 2cd2afc3fff0..d09d352e1c0a 100644 --- a/drivers/net/ethernet/tundra/tsi108_eth.c +++ b/drivers/net/ethernet/tundra/tsi108_eth.c @@ -1290,12 +1290,15 @@ static int tsi108_open(struct net_device *dev) data->rxring = dma_alloc_coherent(&data->pdev->dev, rxring_size, &data->rxdma, GFP_KERNEL); - if (!data->rxring) + if (!data->rxring) { + free_irq(data->irq_num, dev); return -ENOMEM; + } data->txring = dma_alloc_coherent(&data->pdev->dev, txring_size, &data->txdma, GFP_KERNEL); if (!data->txring) { + free_irq(data->irq_num, dev); dma_free_coherent(&data->pdev->dev, rxring_size, data->rxring, data->rxdma); return -ENOMEM; From 23569b5652ee8e8e55a12f7835f59af6f3cefc30 Mon Sep 17 00:00:00 2001 From: Chuang Wang Date: Wed, 9 Nov 2022 17:07:34 +0800 Subject: [PATCH 136/138] net: macvlan: fix memory leaks of macvlan_common_newlink kmemleak reports memory leaks in macvlan_common_newlink, as follows: ip link add link eth0 name .. type macvlan mode source macaddr add kmemleak reports: unreferenced object 0xffff8880109bb140 (size 64): comm "ip", pid 284, jiffies 4294986150 (age 430.108s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 b8 aa 5a 12 80 88 ff ff ..........Z..... 80 1b fa 0d 80 88 ff ff 1e ff ac af c7 c1 6b 6b ..............kk backtrace: [] kmem_cache_alloc_trace+0x1c7/0x300 [] macvlan_hash_add_source+0x45/0xc0 [] macvlan_changelink_sources+0xd7/0x170 [] macvlan_common_newlink+0x38c/0x5a0 [] macvlan_newlink+0xe/0x20 [] __rtnl_newlink+0x7af/0xa50 [] rtnl_newlink+0x48/0x70 ... In the scenario where the macvlan mode is configured as 'source', macvlan_changelink_sources() will be execured to reconfigure list of remote source mac addresses, at the same time, if register_netdevice() return an error, the resource generated by macvlan_changelink_sources() is not cleaned up. Using this patch, in the case of an error, it will execute macvlan_flush_sources() to ensure that the resource is cleaned up. Fixes: aa5fd0fb7748 ("driver: macvlan: Destroy new macvlan port if macvlan_common_newlink failed.") Signed-off-by: Chuang Wang Link: https://lore.kernel.org/r/20221109090735.690500-1-nashuiliang@gmail.com Signed-off-by: Paolo Abeni --- drivers/net/macvlan.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index c5cfe8555199..578897aaada0 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -1533,8 +1533,10 @@ int macvlan_common_newlink(struct net *src_net, struct net_device *dev, /* the macvlan port may be freed by macvlan_uninit when fail to register. * so we destroy the macvlan port only when it's valid. */ - if (create && macvlan_port_get_rtnl(lowerdev)) + if (create && macvlan_port_get_rtnl(lowerdev)) { + macvlan_flush_sources(port, vlan); macvlan_port_destroy(port->dev); + } return err; } EXPORT_SYMBOL_GPL(macvlan_common_newlink); From 6ce3df596be25d95c2cddc9e1b673394948e3732 Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Wed, 9 Nov 2022 15:19:07 -0800 Subject: [PATCH 137/138] MAINTAINERS: Move Vivien to CREDITS Last patch from Vivien was nearly 3 years ago and he has not reviewed or responded to DSA patches since then, move to CREDITS. Signed-off-by: Florian Fainelli Reviewed-by: Andrew Lunn Link: https://lore.kernel.org/r/20221109231907.621678-1-f.fainelli@gmail.com Signed-off-by: Jakub Kicinski --- CREDITS | 5 +++++ MAINTAINERS | 2 -- 2 files changed, 5 insertions(+), 2 deletions(-) diff --git a/CREDITS b/CREDITS index 1841184c834d..54672cbcd719 100644 --- a/CREDITS +++ b/CREDITS @@ -918,6 +918,11 @@ S: Ottawa, Ontario S: K1N 6Z9 S: CANADA +N: Vivien Didelot +E: vivien.didelot@gmail.com +D: DSA framework and MV88E6XXX driver +S: Montreal, Quebec, Canada + N: Jeff Dike E: jdike@karaya.com W: http://user-mode-linux.sourceforge.net diff --git a/MAINTAINERS b/MAINTAINERS index e1bc31a6624b..239407627be2 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -12226,7 +12226,6 @@ F: arch/mips/boot/dts/img/pistachio* MARVELL 88E6XXX ETHERNET SWITCH FABRIC DRIVER M: Andrew Lunn -M: Vivien Didelot L: netdev@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/net/dsa/marvell.txt @@ -14324,7 +14323,6 @@ F: drivers/net/wireless/ NETWORKING [DSA] M: Andrew Lunn -M: Vivien Didelot M: Florian Fainelli M: Vladimir Oltean S: Maintained From de91b3197d15172407608b2c357aab7ac1451e2b Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Wed, 9 Nov 2022 15:01:16 +0000 Subject: [PATCH 138/138] eth: sp7021: drop free_netdev() from spl2sw_init_netdev() It's not necessary to free netdev allocated with devm_alloc_etherdev() and using free_netdev() leads to double free. Fixes: fd3040b9394c ("net: ethernet: Add driver for Sunplus SP7021") Signed-off-by: Wei Yongjun Link: https://lore.kernel.org/r/20221109150116.2988194-1-weiyongjun@huaweicloud.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/sunplus/spl2sw_driver.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/net/ethernet/sunplus/spl2sw_driver.c b/drivers/net/ethernet/sunplus/spl2sw_driver.c index 9be585237277..c499a14314f1 100644 --- a/drivers/net/ethernet/sunplus/spl2sw_driver.c +++ b/drivers/net/ethernet/sunplus/spl2sw_driver.c @@ -287,7 +287,6 @@ static u32 spl2sw_init_netdev(struct platform_device *pdev, u8 *mac_addr, if (ret) { dev_err(&pdev->dev, "Failed to register net device \"%s\"!\n", ndev->name); - free_netdev(ndev); *r_ndev = NULL; return ret; }