- Linux-stable-mirror - lists.linaro.org

patch "iio: imu: fxos8700: fix IMU data bits returned to user space" added to char-misc-linus

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled iio: imu: fxos8700: fix IMU data bits returned to user space to my char-misc git tree which can be found at git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc.git in the char-misc-linus branch. The patch will show up in the next release of the linux-next tree (usually sometime within the next 24 hours during the week.) The patch will hopefully also be merged in Linus's tree for the next -rc kernel release. If you have any questions about this process, please let me know. From a53f945879c0cb9de3a4c05a665f5157884b5208 Mon Sep 17 00:00:00 2001 From: Carlos Song <carlos.song(a)nxp.com> Date: Thu, 8 Dec 2022 15:19:08 +0800 Subject: iio: imu: fxos8700: fix IMU data bits returned to user space ACCEL output data registers contain the X-axis, Y-axis, and Z-axis 14-bit left-justified sample data and MAGN output data registers contain the X-axis, Y-axis, and Z-axis 16-bit sample data. The ACCEL raw register output data should be divided by 4 before sent to userspace. Apply a 2 bits signed right shift to the raw data from ACCEL output data register but keep that from MAGN sensor as the origin. Fixes: 84e5ddd5c46e ("iio: imu: Add support for the FXOS8700 IMU") Signed-off-by: Carlos Song <carlos.song(a)nxp.com> Link: https://lore.kernel.org/r/20221208071911.2405922-5-carlos.song@nxp.com Cc: <Stable(a)vger.kernel.org> Signed-off-by: Jonathan Cameron <Jonathan.Cameron(a)huawei.com> --- drivers/iio/imu/fxos8700_core.c | 28 +++++++++++++++++++++++++++- 1 file changed, 27 insertions(+), 1 deletion(-) diff --git a/drivers/iio/imu/fxos8700_core.c b/drivers/iio/imu/fxos8700_core.c index b62bc92bbacc..06948a8cc315 100644 --- a/drivers/iio/imu/fxos8700_core.c +++ b/drivers/iio/imu/fxos8700_core.c @@ -394,6 +394,7 @@ static int fxos8700_get_data(struct fxos8700_data *data, int chan_type, int axis, int *val) { u8 base, reg; + s16 tmp; int ret; /* @@ -421,8 +422,33 @@ static int fxos8700_get_data(struct fxos8700_data *data, int chan_type, /* Convert axis to buffer index */ reg = axis - IIO_MOD_X; + /* + * Convert to native endianness. The accel data and magn data + * are signed, so a forced type conversion is needed. + */ + tmp = be16_to_cpu(data->buf[reg]); + + /* + * ACCEL output data registers contain the X-axis, Y-axis, and Z-axis + * 14-bit left-justified sample data and MAGN output data registers + * contain the X-axis, Y-axis, and Z-axis 16-bit sample data. Apply + * a signed 2 bits right shift to the readback raw data from ACCEL + * output data register and keep that from MAGN sensor as the origin. + * Value should be extended to 32 bit. + */ + switch (chan_type) { + case IIO_ACCEL: + tmp = tmp >> 2; + break; + case IIO_MAGN: + /* Nothing to do */ + break; + default: + return -EINVAL; + } + /* Convert to native endianness */ - *val = sign_extend32(be16_to_cpu(data->buf[reg]), 15); + *val = sign_extend32(tmp, 15); return 0; } -- 2.39.1

2 years, 9 months

1
0
0 0

patch "iio: imu: fxos8700: fix incomplete ACCEL and MAGN channels readback" added to char-misc-linus

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled iio: imu: fxos8700: fix incomplete ACCEL and MAGN channels readback to my char-misc git tree which can be found at git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc.git in the char-misc-linus branch. The patch will show up in the next release of the linux-next tree (usually sometime within the next 24 hours during the week.) The patch will hopefully also be merged in Linus's tree for the next -rc kernel release. If you have any questions about this process, please let me know. From 37a94d86d7050665d6d01378b2c916c28e454f10 Mon Sep 17 00:00:00 2001 From: Carlos Song <carlos.song(a)nxp.com> Date: Thu, 8 Dec 2022 15:19:07 +0800 Subject: iio: imu: fxos8700: fix incomplete ACCEL and MAGN channels readback The length of ACCEL and MAGN 3-axis channels output data is 6 byte individually. However block only read 3 bytes data into buffer from ACCEL or MAGN output data registers every time. It causes an incomplete ACCEL and MAGN channels readback. Set correct value count for regmap_bulk_read to get 6 bytes ACCEL and MAGN channels readback. Fixes: 84e5ddd5c46e ("iio: imu: Add support for the FXOS8700 IMU") Signed-off-by: Carlos Song <carlos.song(a)nxp.com> Link: https://lore.kernel.org/r/20221208071911.2405922-4-carlos.song@nxp.com Cc: <Stable(a)vger.kernel.org> Signed-off-by: Jonathan Cameron <Jonathan.Cameron(a)huawei.com> --- drivers/iio/imu/fxos8700_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/imu/fxos8700_core.c b/drivers/iio/imu/fxos8700_core.c index 977eb7dc7dbd..b62bc92bbacc 100644 --- a/drivers/iio/imu/fxos8700_core.c +++ b/drivers/iio/imu/fxos8700_core.c @@ -414,7 +414,7 @@ static int fxos8700_get_data(struct fxos8700_data *data, int chan_type, /* Block read 6 bytes of device output registers to avoid data loss */ ret = regmap_bulk_read(data->regmap, base, data->buf, - FXOS8700_DATA_BUF_SIZE); + sizeof(data->buf)); if (ret) return ret; -- 2.39.1

2 years, 9 months

1
0
0 0

patch "iio: imu: fxos8700: fix swapped ACCEL and MAGN channels readback" added to char-misc-linus

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled iio: imu: fxos8700: fix swapped ACCEL and MAGN channels readback to my char-misc git tree which can be found at git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc.git in the char-misc-linus branch. The patch will show up in the next release of the linux-next tree (usually sometime within the next 24 hours during the week.) The patch will hopefully also be merged in Linus's tree for the next -rc kernel release. If you have any questions about this process, please let me know. From c68b44bc7d9b1469774a1c985ee71d2cbc5ebef5 Mon Sep 17 00:00:00 2001 From: Carlos Song <carlos.song(a)nxp.com> Date: Thu, 8 Dec 2022 15:19:06 +0800 Subject: iio: imu: fxos8700: fix swapped ACCEL and MAGN channels readback Because ACCEL and MAGN channels data register base address is swapped the accelerometer and magnetometer channels readback is swapped. Fixes: 84e5ddd5c46e ("iio: imu: Add support for the FXOS8700 IMU") Signed-off-by: Carlos Song <carlos.song(a)nxp.com> Link: https://lore.kernel.org/r/20221208071911.2405922-3-carlos.song@nxp.com Cc: <Stable(a)vger.kernel.org> Signed-off-by: Jonathan Cameron <Jonathan.Cameron(a)huawei.com> --- drivers/iio/imu/fxos8700_core.c | 17 +++++++++++++++-- 1 file changed, 15 insertions(+), 2 deletions(-) diff --git a/drivers/iio/imu/fxos8700_core.c b/drivers/iio/imu/fxos8700_core.c index 235b02b2f4e5..977eb7dc7dbd 100644 --- a/drivers/iio/imu/fxos8700_core.c +++ b/drivers/iio/imu/fxos8700_core.c @@ -395,9 +395,22 @@ static int fxos8700_get_data(struct fxos8700_data *data, int chan_type, { u8 base, reg; int ret; - enum fxos8700_sensor type = fxos8700_to_sensor(chan_type); - base = type ? FXOS8700_OUT_X_MSB : FXOS8700_M_OUT_X_MSB; + /* + * Different register base addresses varies with channel types. + * This bug hasn't been noticed before because using an enum is + * really hard to read. Use an a switch statement to take over that. + */ + switch (chan_type) { + case IIO_ACCEL: + base = FXOS8700_OUT_X_MSB; + break; + case IIO_MAGN: + base = FXOS8700_M_OUT_X_MSB; + break; + default: + return -EINVAL; + } /* Block read 6 bytes of device output registers to avoid data loss */ ret = regmap_bulk_read(data->regmap, base, data->buf, -- 2.39.1

2 years, 9 months

1
0
0 0

patch "iio: imu: fxos8700: fix map label of channel type to MAGN sensor" added to char-misc-linus

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled iio: imu: fxos8700: fix map label of channel type to MAGN sensor to my char-misc git tree which can be found at git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc.git in the char-misc-linus branch. The patch will show up in the next release of the linux-next tree (usually sometime within the next 24 hours during the week.) The patch will hopefully also be merged in Linus's tree for the next -rc kernel release. If you have any questions about this process, please let me know. From 429e1e8ec696e0e7a0742904e3dc2f83b7b23dfb Mon Sep 17 00:00:00 2001 From: Carlos Song <carlos.song(a)nxp.com> Date: Thu, 8 Dec 2022 15:19:05 +0800 Subject: iio: imu: fxos8700: fix map label of channel type to MAGN sensor FXOS8700 is an IMU sensor with ACCEL sensor and MAGN sensor. Sensor type is indexed by corresponding channel type in a switch. IIO_ANGL_VEL channel type mapped to MAGN sensor has caused confusion. Fix the mapping label of "IIO_MAGN" channel type instead of "IIO_ANGL_VEL" channel type to MAGN sensor. Fixes: 84e5ddd5c46e ("iio: imu: Add support for the FXOS8700 IMU") Signed-off-by: Carlos Song <carlos.song(a)nxp.com> Link: https://lore.kernel.org/r/20221208071911.2405922-2-carlos.song@nxp.com Cc: <Stable(a)vger.kernel.org> Signed-off-by: Jonathan Cameron <Jonathan.Cameron(a)huawei.com> --- drivers/iio/imu/fxos8700_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/imu/fxos8700_core.c b/drivers/iio/imu/fxos8700_core.c index 423cfe526f2a..235b02b2f4e5 100644 --- a/drivers/iio/imu/fxos8700_core.c +++ b/drivers/iio/imu/fxos8700_core.c @@ -320,7 +320,7 @@ static enum fxos8700_sensor fxos8700_to_sensor(enum iio_chan_type iio_type) switch (iio_type) { case IIO_ACCEL: return FXOS8700_ACCEL; - case IIO_ANGL_VEL: + case IIO_MAGN: return FXOS8700_MAGN; default: return -EINVAL; -- 2.39.1

2 years, 9 months

1
0
0 0

patch "iio:adc:twl6030: Enable measurements of VUSB, VBAT and others" added to char-misc-linus

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled iio:adc:twl6030: Enable measurements of VUSB, VBAT and others to my char-misc git tree which can be found at git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc.git in the char-misc-linus branch. The patch will show up in the next release of the linux-next tree (usually sometime within the next 24 hours during the week.) The patch will hopefully also be merged in Linus's tree for the next -rc kernel release. If you have any questions about this process, please let me know. From f804bd0dc28683a93a60f271aaefb2fc5b0853dd Mon Sep 17 00:00:00 2001 From: Andreas Kemnade <andreas(a)kemnade.info> Date: Thu, 1 Dec 2022 19:16:35 +0100 Subject: iio:adc:twl6030: Enable measurements of VUSB, VBAT and others Some inputs need to be wired up to produce proper measurements, without this change only near zero values are reported. Signed-off-by: Andreas Kemnade <andreas(a)kemnade.info> Fixes: 1696f36482e70 ("iio: twl6030-gpadc: TWL6030, TWL6032 GPADC driver") Link: https://lore.kernel.org/r/20221201181635.3522962-1-andreas@kemnade.info Cc: <Stable(a)vger.kernel.org> Signed-off-by: Jonathan Cameron <Jonathan.Cameron(a)huawei.com> --- drivers/iio/adc/twl6030-gpadc.c | 32 ++++++++++++++++++++++++++++++++ 1 file changed, 32 insertions(+) diff --git a/drivers/iio/adc/twl6030-gpadc.c b/drivers/iio/adc/twl6030-gpadc.c index f53e8558b560..40438e5b4970 100644 --- a/drivers/iio/adc/twl6030-gpadc.c +++ b/drivers/iio/adc/twl6030-gpadc.c @@ -57,6 +57,18 @@ #define TWL6030_GPADCS BIT(1) #define TWL6030_GPADCR BIT(0) +#define USB_VBUS_CTRL_SET 0x04 +#define USB_ID_CTRL_SET 0x06 + +#define TWL6030_MISC1 0xE4 +#define VBUS_MEAS 0x01 +#define ID_MEAS 0x01 + +#define VAC_MEAS 0x04 +#define VBAT_MEAS 0x02 +#define BB_MEAS 0x01 + + /** * struct twl6030_chnl_calib - channel calibration * @gain: slope coefficient for ideal curve @@ -927,6 +939,26 @@ static int twl6030_gpadc_probe(struct platform_device *pdev) return ret; } + ret = twl_i2c_write_u8(TWL_MODULE_USB, VBUS_MEAS, USB_VBUS_CTRL_SET); + if (ret < 0) { + dev_err(dev, "failed to wire up inputs\n"); + return ret; + } + + ret = twl_i2c_write_u8(TWL_MODULE_USB, ID_MEAS, USB_ID_CTRL_SET); + if (ret < 0) { + dev_err(dev, "failed to wire up inputs\n"); + return ret; + } + + ret = twl_i2c_write_u8(TWL6030_MODULE_ID0, + VBAT_MEAS | BB_MEAS | BB_MEAS, + TWL6030_MISC1); + if (ret < 0) { + dev_err(dev, "failed to wire up inputs\n"); + return ret; + } + indio_dev->name = DRIVER_NAME; indio_dev->info = &twl6030_gpadc_iio_info; indio_dev->modes = INDIO_DIRECT_MODE; -- 2.39.1

2 years, 9 months

1
0
0 0

patch "iio: imx8qxp-adc: fix irq flood when call imx8qxp_adc_read_raw()" added to char-misc-linus

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled iio: imx8qxp-adc: fix irq flood when call imx8qxp_adc_read_raw() to my char-misc git tree which can be found at git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc.git in the char-misc-linus branch. The patch will show up in the next release of the linux-next tree (usually sometime within the next 24 hours during the week.) The patch will hopefully also be merged in Linus's tree for the next -rc kernel release. If you have any questions about this process, please let me know. From 0fc3562a993c3dc41d1177b3983d9300d0db1d4d Mon Sep 17 00:00:00 2001 From: Frank Li <Frank.Li(a)nxp.com> Date: Thu, 1 Dec 2022 09:01:10 -0500 Subject: iio: imx8qxp-adc: fix irq flood when call imx8qxp_adc_read_raw() irq flood happen when run cat /sys/bus/iio/devices/iio:device0/in_voltage1_raw imx8qxp_adc_read_raw() { ... enable irq /* adc start */ writel(1, adc->regs + IMX8QXP_ADR_ADC_SWTRIG); ^^^^ trigger irq flood. wait_for_completion_interruptible_timeout(); readl(adc->regs + IMX8QXP_ADR_ADC_RESFIFO); ^^^^ clear irq here. ... } There is only FIFO watermark interrupt at this ADC controller. IRQ line will be assert until software read data from FIFO. So IRQ flood happen during wait_for_completion_interruptible_timeout(). Move FIFO read into irq handle to avoid irq flood. Fixes: 1e23dcaa1a9f ("iio: imx8qxp-adc: Add driver support for NXP IMX8QXP ADC") Cc: stable(a)vger.kernel.org Signed-off-by: Frank Li <Frank.Li(a)nxp.com> Reviewed-by: Cai Huoqing <cai.huoqing(a)linux.dev> Reviewed-by: Haibo Chen <haibo.chen(a)nxp.com> Link: https://lore.kernel.org/r/20221201140110.2653501-1-Frank.Li@nxp.com Cc: <Stable(a)vger.kernel.org> Signed-off-by: Jonathan Cameron <Jonathan.Cameron(a)huawei.com> --- drivers/iio/adc/imx8qxp-adc.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/iio/adc/imx8qxp-adc.c b/drivers/iio/adc/imx8qxp-adc.c index 36777b827165..f5a0fc9e64c5 100644 --- a/drivers/iio/adc/imx8qxp-adc.c +++ b/drivers/iio/adc/imx8qxp-adc.c @@ -86,6 +86,8 @@ #define IMX8QXP_ADC_TIMEOUT msecs_to_jiffies(100) +#define IMX8QXP_ADC_MAX_FIFO_SIZE 16 + struct imx8qxp_adc { struct device *dev; void __iomem *regs; @@ -95,6 +97,7 @@ struct imx8qxp_adc { /* Serialise ADC channel reads */ struct mutex lock; struct completion completion; + u32 fifo[IMX8QXP_ADC_MAX_FIFO_SIZE]; }; #define IMX8QXP_ADC_CHAN(_idx) { \ @@ -238,8 +241,7 @@ static int imx8qxp_adc_read_raw(struct iio_dev *indio_dev, return ret; } - *val = FIELD_GET(IMX8QXP_ADC_RESFIFO_VAL_MASK, - readl(adc->regs + IMX8QXP_ADR_ADC_RESFIFO)); + *val = adc->fifo[0]; mutex_unlock(&adc->lock); return IIO_VAL_INT; @@ -265,10 +267,15 @@ static irqreturn_t imx8qxp_adc_isr(int irq, void *dev_id) { struct imx8qxp_adc *adc = dev_id; u32 fifo_count; + int i; fifo_count = FIELD_GET(IMX8QXP_ADC_FCTRL_FCOUNT_MASK, readl(adc->regs + IMX8QXP_ADR_ADC_FCTRL)); + for (i = 0; i < fifo_count; i++) + adc->fifo[i] = FIELD_GET(IMX8QXP_ADC_RESFIFO_VAL_MASK, + readl_relaxed(adc->regs + IMX8QXP_ADR_ADC_RESFIFO)); + if (fifo_count) complete(&adc->completion); -- 2.39.1

2 years, 9 months

1
0
0 0

patch "iio: adc: xilinx-ams: fix devm_krealloc() return value check" added to char-misc-linus

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled iio: adc: xilinx-ams: fix devm_krealloc() return value check to my char-misc git tree which can be found at git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc.git in the char-misc-linus branch. The patch will show up in the next release of the linux-next tree (usually sometime within the next 24 hours during the week.) The patch will hopefully also be merged in Linus's tree for the next -rc kernel release. If you have any questions about this process, please let me know. From 6794ed0cfcc6ce737240eccc48b3e8190df36703 Mon Sep 17 00:00:00 2001 From: Marco Pagani <marpagan(a)redhat.com> Date: Fri, 25 Nov 2022 12:31:12 +0100 Subject: iio: adc: xilinx-ams: fix devm_krealloc() return value check The clang-analyzer reported a warning: "Value stored to 'ret' is never read". Fix the return value check if devm_krealloc() fails to resize ams_channels. Fixes: d5c70627a794 ("iio: adc: Add Xilinx AMS driver") Signed-off-by: Marco Pagani <marpagan(a)redhat.com> Acked-by: Michal Simek <michal.simek(a)amd.com> Link: https://lore.kernel.org/r/20221125113112.219290-1-marpagan@redhat.com Cc: <stable(a)vger.kernel.org> Signed-off-by: Jonathan Cameron <Jonathan.Cameron(a)huawei.com> --- drivers/iio/adc/xilinx-ams.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/adc/xilinx-ams.c b/drivers/iio/adc/xilinx-ams.c index 5b4bdf3a26bb..a507d2e17079 100644 --- a/drivers/iio/adc/xilinx-ams.c +++ b/drivers/iio/adc/xilinx-ams.c @@ -1329,7 +1329,7 @@ static int ams_parse_firmware(struct iio_dev *indio_dev) dev_channels = devm_krealloc(dev, ams_channels, dev_size, GFP_KERNEL); if (!dev_channels) - ret = -ENOMEM; + return -ENOMEM; indio_dev->channels = dev_channels; indio_dev->num_channels = num_channels; -- 2.39.1

2 years, 9 months

1
0
0 0

patch "iio: adc: berlin2-adc: Add missing of_node_put() in error path" added to char-misc-linus

by gregkh＠linuxfoundation.org

This is a note to let you know that I've just added the patch titled iio: adc: berlin2-adc: Add missing of_node_put() in error path to my char-misc git tree which can be found at git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc.git in the char-misc-linus branch. The patch will show up in the next release of the linux-next tree (usually sometime within the next 24 hours during the week.) The patch will hopefully also be merged in Linus's tree for the next -rc kernel release. If you have any questions about this process, please let me know. From cbd3a0153cd18a2cbef6bf3cf31bb406c3fc9f55 Mon Sep 17 00:00:00 2001 From: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> Date: Tue, 29 Nov 2022 10:03:16 +0800 Subject: iio: adc: berlin2-adc: Add missing of_node_put() in error path of_get_parent() will return a device_node pointer with refcount incremented. We need to use of_node_put() on it when done. Add the missing of_node_put() in the error path of berlin2_adc_probe(); Fixes: 70f1937911ca ("iio: adc: add support for Berlin") Signed-off-by: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> Link: https://lore.kernel.org/r/20221129020316.191731-1-wangxiongfeng2@huawei.com Cc: <Stable(a)vger.kernel.org> Signed-off-by: Jonathan Cameron <Jonathan.Cameron(a)huawei.com> --- drivers/iio/adc/berlin2-adc.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/iio/adc/berlin2-adc.c b/drivers/iio/adc/berlin2-adc.c index 3d2e8b4db61a..a4e7c7eff5ac 100644 --- a/drivers/iio/adc/berlin2-adc.c +++ b/drivers/iio/adc/berlin2-adc.c @@ -298,8 +298,10 @@ static int berlin2_adc_probe(struct platform_device *pdev) int ret; indio_dev = devm_iio_device_alloc(&pdev->dev, sizeof(*priv)); - if (!indio_dev) + if (!indio_dev) { + of_node_put(parent_np); return -ENOMEM; + } priv = iio_priv(indio_dev); -- 2.39.1

2 years, 9 months

1
0
0 0

[PATCH v2] x86/debug: Fix stack recursion caused by DR7 accesses

by Joerg Roedel

From: Joerg Roedel <jroedel(a)suse.de> In kernels compiled with CONFIG_PARAVIRT=n the compiler re-orders the DR7 read in exc_nmi() to happen before the call to sev_es_ist_enter(). This is problematic when running as an SEV-ES guest because in this environment the DR7 read might cause a #VC exception, and taking #VC exceptions is not safe in exc_nmi() before sev_es_ist_enter() has run. The result is stack recursion if the NMI was caused on the #VC IST stack, because a subsequent #VC exception in the NMI handler will overwrite the stack frame of the interrupted #VC handler. As there are no compiler barriers affecting the ordering of DR7 reads/writes, make the accesses to this register volatile, forbidding the compiler to re-order them. [ bp: Massage text. ] Cc: stable(a)vger.kernel.org Fixes: 315562c9af3d ("x86/sev-es: Adjust #VC IST Stack on entering NMI handler") Signed-off-by: Joerg Roedel <jroedel(a)suse.de> --- arch/x86/include/asm/debugreg.h | 27 ++++++++++++++++++++++++--- 1 file changed, 24 insertions(+), 3 deletions(-) diff --git a/arch/x86/include/asm/debugreg.h b/arch/x86/include/asm/debugreg.h index b049d950612f..d54d5dbfb4b9 100644 --- a/arch/x86/include/asm/debugreg.h +++ b/arch/x86/include/asm/debugreg.h @@ -39,7 +39,20 @@ static __always_inline unsigned long native_get_debugreg(int regno) asm("mov %%db6, %0" :"=r" (val)); break; case 7: - asm("mov %%db7, %0" :"=r" (val)); + /* + * Apply __FORCE_ORDER to DR7 reads to forbid re-ordering them + * with other code. + * + * This is needed because a DR7 access can cause a #VC exception + * when running under SEV-ES. Taking a #VC exception is not a + * safe thing to do just anywhere in the entry code and + * re-ordering might place the access into an unsafe location. + * + * This happened in the NMI handler, where the DR7 read was + * re-ordered to happen before the call to sev_es_ist_enter(), + * causing stack recursion. + */ + asm("mov %%db7, %0" : "=r" (val) : __FORCE_ORDER); break; default: BUG(); @@ -66,8 +79,16 @@ static __always_inline void native_set_debugreg(int regno, unsigned long value) asm("mov %0, %%db6" ::"r" (value)); break; case 7: - asm("mov %0, %%db7" ::"r" (value)); - break; + /* + * Apply __FORCE_ORDER to DR7 writes to forbid re-ordering them + * with other code. + * + * While is didn't happen with a DR7 write (see the DR7 read + * comment above which explains where it happened), add the + * __FORCE_ORDER here too to avoid similar problems in the + * future. + */ + asm("mov %0, %%db7" ::"r" (value), __FORCE_ORDER); break; default: BUG(); } -- 2.39.0

2 years, 9 months

1
0
0 0

Re: [PATCH 6.1 000/306] 6.1.9-rc3 review

by Ronald Warsow

Hi Greg 6.1.9-rc3 compiles, boots and runs here on x86_64 (Intel i5-11400, Fedora 37) Thanks Tested-by: Ronald Warsow <rwarsow(a)gmx.de>

2 years, 9 months

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror