Skip to contentSkip to conversation

5.15.3-rc1 review

PATCH In review Greg Kroah-Hartman 14 participants 945 messages Nov 15, 2021 – Nov 24, 2021 Page 4 of 7

Subsystems (15)
lore
Share & tools
lore
author maintainer
#392 lore raw

PATCH 381/917 media: dvb-usb: fix ununit-value in az6027_rc_query

In reply to Greg Kroah-Hartman

From: Pavel Skripkin <[email protected]>

[ Upstream commit afae4ef7d5ad913cab1316137854a36bea6268a5 ]

Syzbot reported ununit-value bug in az6027_rc_query(). The problem was in missing state pointer initialization. Since this function does nothing we can simply initialize state to REMOTE_NO_KEY_PRESSED.

Reported-and-tested-by [email protected]
Fixes 76f9a820c867 ("V4L/DVB: AZ6027: Initial import of the driver")
---
 drivers/media/usb/dvb-usb/az6027.c | 1 +
 1 file changed, 1 insertion(+)
drivers/media/usb/dvb-usb/az6027.c +1 -0
index 1c39b61cde29b..86788771175b7 100644
@@ -391,6 +391,7 @@ static struct rc_map_table rc_map_az6027_table[] = {
 /* remote control stuff (does not work with my box) */
 static int az6027_rc_query(struct dvb_usb_device *d, u32 *event, int *state)
 {
+	*state = REMOTE_NO_KEY_PRESSED;
 	return 0;
 }
 
Signature · 1 line
2.33.0
author maintainer
#393 lore raw

PATCH 382/917 media: imx258: Fix getting clock frequency

In reply to Greg Kroah-Hartman

From: Sakari Ailus <[email protected]>

[ Upstream commit d170b0ea1760989fe8ac053bef83e61f3bf87992 ]

Obtain the clock frequency by reading the clock-frequency property if there's no clock.

Fixes 9fda25332c4b ("media: i2c: imx258: get clock from device properties and enable it via runtime PM")
---
 drivers/media/i2c/imx258.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
drivers/media/i2c/imx258.c +6 -6
index 81cdf37216ca7..c249507aa2dbc 100644
@@ -1260,18 +1260,18 @@ static int imx258_probe(struct i2c_client *client)
 		return -ENOMEM;
 
 	imx258->clk = devm_clk_get_optional(&client->dev, NULL);
+	if (IS_ERR(imx258->clk))
+		return dev_err_probe(&client->dev, PTR_ERR(imx258->clk),
+				     "error getting clock\n");
 	if (!imx258->clk) {
 		dev_dbg(&client->dev,
 			"no clock provided, using clock-frequency property\n");
 
 		device_property_read_u32(&client->dev, "clock-frequency", &val);
-		if (val != IMX258_INPUT_CLOCK_FREQ)
-			return -EINVAL;
-	} else if (IS_ERR(imx258->clk)) {
-		return dev_err_probe(&client->dev, PTR_ERR(imx258->clk),
-				     "error getting clock\n");
+	} else {
+		val = clk_get_rate(imx258->clk);
 	}
-	if (clk_get_rate(imx258->clk) != IMX258_INPUT_CLOCK_FREQ) {
+	if (val != IMX258_INPUT_CLOCK_FREQ) {
 		dev_err(&client->dev, "input clock frequency not supported\n");
 		return -EINVAL;
 	}
Signature · 1 line
2.33.0
author maintainer
#394 lore raw

PATCH 383/917 media: v4l2-ioctl: S_CTRL output the right value

In reply to Greg Kroah-Hartman

From: Ricardo Ribalda <[email protected]>

[ Upstream commit c87ed93574e3cd8346c05bd934c617596c12541b ]

If the driver does not implement s_ctrl, but it does implement s_ext_ctrls, we convert the call.

When that happens we have also to convert back the response from s_ext_ctrls.

Fixes v4l2_compliance:
Control ioctls (Input 0):
fail: v4l2-test-controls.cpp(411): returned control value out of range
fail: v4l2-test-controls.cpp(507): invalid control 00980900
test VIDIOC_G/S_CTRL: FAIL

Fixes 35ea11ff8471 ("V4L/DVB (8430): videodev: move some functions from v4l2-dev.h to v4l2-common.h or v4l2-ioctl.h")
---
 drivers/media/v4l2-core/v4l2-ioctl.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
drivers/media/v4l2-core/v4l2-ioctl.c +6 -3
index da5b4002a466a..f4f67b385d00a 100644
@@ -2224,6 +2224,7 @@ static int v4l_s_ctrl(const struct v4l2_ioctl_ops *ops,
 		test_bit(V4L2_FL_USES_V4L2_FH, &vfd->flags) ? fh : NULL;
 	struct v4l2_ext_controls ctrls;
 	struct v4l2_ext_control ctrl;
+	int ret;
 
 	if (vfh && vfh->ctrl_handler)
 		return v4l2_s_ctrl(vfh, vfh->ctrl_handler, p);
@@ -2239,9 +2240,11 @@ static int v4l_s_ctrl(const struct v4l2_ioctl_ops *ops,
 	ctrls.controls = &ctrl;
 	ctrl.id = p->id;
 	ctrl.value = p->value;
-	if (check_ext_ctrls(&ctrls, VIDIOC_S_CTRL))
-		return ops->vidioc_s_ext_ctrls(file, fh, &ctrls);
-	return -EINVAL;
+	if (!check_ext_ctrls(&ctrls, VIDIOC_S_CTRL))
+		return -EINVAL;
+	ret = ops->vidioc_s_ext_ctrls(file, fh, &ctrls);
+	p->value = ctrl.value;
+	return ret;
 }
 
 static int v4l_g_ext_ctrls(const struct v4l2_ioctl_ops *ops,
Signature · 1 line
2.33.0
author maintainer
#395 lore raw

PATCH 384/917 media: mtk-vcodec: venc: fix return value when start_streaming fails

In reply to Greg Kroah-Hartman

From: Dafna Hirschfeld <[email protected]>

[ Upstream commit 065a7c66bd8b21db212fa86187ff12f0cac6ea6d ]

In case vb2ops_venc_start_streaming fails, the error value is overwritten by the ret value of pm_runtime_put which might be 0. Fix it.

Fixes 985c73693fe5a (" media: mtk-vcodec: Separating mtk encoder driver")
---
 drivers/media/platform/mtk-vcodec/mtk_vcodec_enc.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/media/platform/mtk-vcodec/mtk_vcodec_enc.c +4 -4
index 416f356af363d..d97a6765693f1 100644
@@ -793,7 +793,7 @@ static int vb2ops_venc_start_streaming(struct vb2_queue *q, unsigned int count)
 {
 	struct mtk_vcodec_ctx *ctx = vb2_get_drv_priv(q);
 	struct venc_enc_param param;
-	int ret;
+	int ret, pm_ret;
 	int i;
 
 	/* Once state turn into MTK_STATE_ABORT, we need stop_streaming
@@ -845,9 +845,9 @@ static int vb2ops_venc_start_streaming(struct vb2_queue *q, unsigned int count)
 	return 0;
 
 err_set_param:
-	ret = pm_runtime_put(&ctx->dev->plat_dev->dev);
-	if (ret < 0)
-		mtk_v4l2_err("pm_runtime_put fail %d", ret);
+	pm_ret = pm_runtime_put(&ctx->dev->plat_dev->dev);
+	if (pm_ret < 0)
+		mtk_v4l2_err("pm_runtime_put fail %d", pm_ret);
 
 err_start_stream:
 	for (i = 0; i < q->num_buffers; ++i) {
Signature · 1 line
2.33.0
author maintainer
#396 lore raw

PATCH 385/917 media: TDA1997x: handle short reads of hdmi info frame.

In reply to Greg Kroah-Hartman

From: Tom Rix <[email protected]>

[ Upstream commit 48d219f9cc667bc6fbc3e3af0b1bfd75db94fce4 ]

Static analysis reports this representative problem

tda1997x.c:1939: warning: 7th function call argument is an uninitialized value

The 7th argument is buffer[0], which is set in the earlier call to io_readn(). When io_readn() call to io_read() fails with the first read, buffer[0] is not set and 0 is returned and stored in len.

The later call to hdmi_infoframe_unpack()'s size parameter is the static size of buffer, always 40, so a short read is not caught in hdmi_infoframe_unpacks()'s checking. The variable len should be used instead.

Zero initialize buffer to 0 so it is in a known start state.

Fixes 9ac0038db9a7 ("media: i2c: Add TDA1997x HDMI receiver driver")
---
 drivers/media/i2c/tda1997x.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/media/i2c/tda1997x.c +4 -4
index 6070aaf0b32ea..4dafa9f1cf522 100644
@@ -1248,13 +1248,13 @@ tda1997x_parse_infoframe(struct tda1997x_state *state, u16 addr)
 {
 	struct v4l2_subdev *sd = &state->sd;
 	union hdmi_infoframe frame;
-	u8 buffer[40];
+	u8 buffer[40] = { 0 };
 	u8 reg;
 	int len, err;
 
 	/* read data */
 	len = io_readn(sd, addr, sizeof(buffer), buffer);
-	err = hdmi_infoframe_unpack(&frame, buffer, sizeof(buffer));
+	err = hdmi_infoframe_unpack(&frame, buffer, len);
 	if (err) {
 		v4l_err(state->client,
 			"failed parsing %d byte infoframe: 0x%04x/0x%02x\n",
@@ -1928,13 +1928,13 @@ static int tda1997x_log_infoframe(struct v4l2_subdev *sd, int addr)
 {
 	struct tda1997x_state *state = to_state(sd);
 	union hdmi_infoframe frame;
-	u8 buffer[40];
+	u8 buffer[40] = { 0 };
 	int len, err;
 
 	/* read data */
 	len = io_readn(sd, addr, sizeof(buffer), buffer);
 	v4l2_dbg(1, debug, sd, "infoframe: addr=%d len=%d\n", addr, len);
-	err = hdmi_infoframe_unpack(&frame, buffer, sizeof(buffer));
+	err = hdmi_infoframe_unpack(&frame, buffer, len);
 	if (err) {
 		v4l_err(state->client,
 			"failed parsing %d byte infoframe: 0x%04x/0x%02x\n",
Signature · 1 line
2.33.0
author maintainer
#397 lore raw

PATCH 386/917 media: mtk-vpu: Fix a resource leak in the error handling path of mtk_vpu_probe()

In reply to Greg Kroah-Hartman

From: Christophe JAILLET <[email protected]>

[ Upstream commit 2143ad413c05c7be24c3a92760e367b7f6aaac92 ]

A successful 'clk_prepare()' call should be balanced by a corresponding 'clk_unprepare()' call in the error handling path of the probe, as already done in the remove function.

Update the error handling path accordingly.

Fixes 3003a180ef6b ("[media] VPU: mediatek: support Mediatek VPU")
---
 drivers/media/platform/mtk-vpu/mtk_vpu.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/media/platform/mtk-vpu/mtk_vpu.c +4 -1
index ec290dde59cfd..7f1647da0ade0 100644
@@ -848,7 +848,8 @@ static int mtk_vpu_probe(struct platform_device *pdev)
 	vpu->wdt.wq = create_singlethread_workqueue("vpu_wdt");
 	if (!vpu->wdt.wq) {
 		dev_err(dev, "initialize wdt workqueue failed\n");
-		return -ENOMEM;
+		ret = -ENOMEM;
+		goto clk_unprepare;
 	}
 	INIT_WORK(&vpu->wdt.ws, vpu_wdt_reset_func);
 	mutex_init(&vpu->vpu_mutex);
@@ -942,6 +943,8 @@ disable_vpu_clk:
 	vpu_clock_disable(vpu);
 workqueue_destroy:
 	destroy_workqueue(vpu->wdt.wq);
+clk_unprepare:
+	clk_unprepare(vpu->clk);
 
 	return ret;
 }
Signature · 1 line
2.33.0
author maintainer
#398 lore raw

PATCH 387/917 media: imx-jpeg: Fix the error handling path of mxc_jpeg_probe()

In reply to Greg Kroah-Hartman

From: Christophe JAILLET <[email protected]>

[ Upstream commit 5c47dc6657543b3c4dffcbe741fb693b9b96796d ]

A successful 'mxc_jpeg_attach_pm_domains()' call should be balanced by a corresponding 'mxc_jpeg_detach_pm_domains()' call in the error handling path of the probe, as already done in the remove function.

Update the error handling path accordingly.

Fixes 2db16c6ed72c ("media: imx-jpeg: Add V4L2 driver for i.MX8 JPEG Encoder/Decoder")
---
 drivers/media/platform/imx-jpeg/mxc-jpeg.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/media/platform/imx-jpeg/mxc-jpeg.c +2 -0
index 33e7604271cdf..fc905ea78b175 100644
@@ -2092,6 +2092,8 @@ err_m2m:
 	v4l2_device_unregister(&jpeg->v4l2_dev);
 
 err_register:
+	mxc_jpeg_detach_pm_domains(jpeg);
+
 err_irq:
 	return ret;
 }
Signature · 1 line
2.33.0
author maintainer
#399 lore raw

PATCH 388/917 media: i2c: ths8200 needs V4L2_ASYNC

In reply to Greg Kroah-Hartman

From: Randy Dunlap <[email protected]>

[ Upstream commit e4625044d656f3c33ece0cc9da22577bc10ca5d3 ]

Fix the build errors reported by the kernel test robot by selecting V4L2_ASYNC:

mips-linux-ld: drivers/media/i2c/ths8200.o: in function `ths8200_remove':
ths8200.c:(.text+0x1ec): undefined reference to `v4l2_async_unregister_subdev'
mips-linux-ld: drivers/media/i2c/ths8200.o: in function `ths8200_probe':
ths8200.c:(.text+0x404): undefined reference to `v4l2_async_register_subdev'

---
 drivers/media/i2c/Kconfig | 1 +
 1 file changed, 1 insertion(+)
drivers/media/i2c/Kconfig +1 -0
index 08feb3e8c1bf6..6157e73eef24e 100644
@@ -597,6 +597,7 @@ config VIDEO_AK881X
 config VIDEO_THS8200
 	tristate "Texas Instruments THS8200 video encoder"
 	depends on VIDEO_V4L2 && I2C
+	select V4L2_ASYNC
 	help
 	  Support for the Texas Instruments THS8200 video encoder.
 
Signature · 1 line
2.33.0
author maintainer
#400 lore raw

PATCH 389/917 media: sun6i-csi: Allow the video device to be open multiple times

In reply to Greg Kroah-Hartman

From: Ondrej Jirman <[email protected]>

[ Upstream commit 8ed852834683ebe064157e069af8dfb41cad6403 ]

Previously it was possible, but a recent fix for uninitialized ret variable broke this behavior.

v4l2_fh_is_singular_file() check is there just to determine whether the power needs to be enabled, and it's not a failure if it returns false.

Fixes ba9139116bc0 ("media: sun6i-csi: add a missing return code")
---
 drivers/media/platform/sunxi/sun6i-csi/sun6i_video.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)
drivers/media/platform/sunxi/sun6i-csi/sun6i_video.c +2 -4
index 07b2161392d21..5ba3e29f794fd 100644
@@ -467,7 +467,7 @@ static const struct v4l2_ioctl_ops sun6i_video_ioctl_ops = {
 static int sun6i_video_open(struct file *file)
 {
 	struct sun6i_video *video = video_drvdata(file);
-	int ret;
+	int ret = 0;
 
 	if (mutex_lock_interruptible(&video->lock))
 		return -ERESTARTSYS;
@@ -481,10 +481,8 @@ static int sun6i_video_open(struct file *file)
 		goto fh_release;
 
 	/* check if already powered */
-	if (!v4l2_fh_is_singular_file(file)) {
-		ret = -EBUSY;
+	if (!v4l2_fh_is_singular_file(file))
 		goto unlock;
-	}
 
 	ret = sun6i_csi_set_power(video->csi, true);
 	if (ret < 0)
Signature · 1 line
2.33.0
author maintainer
#401 lore raw

PATCH 390/917 media: radio-wl1273: Avoid card name truncation

In reply to Greg Kroah-Hartman

From: Kees Cook <[email protected]>

[ Upstream commit dfadec236aa99f6086141949c9dc3ec50f3ff20d ]

The "card" string only holds 31 characters (and the terminating NUL). In order to avoid truncation, use a shorter card description instead of the current result, "Texas Instruments Wl1273 FM Rad".

Fixes 87d1a50ce451 ("[media] V4L2: WL1273 FM Radio: TI WL1273 FM radio driver")
---
 drivers/media/radio/radio-wl1273.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/media/radio/radio-wl1273.c +1 -1
index 1123768731676..484046471c03f 100644
@@ -1279,7 +1279,7 @@ static int wl1273_fm_vidioc_querycap(struct file *file, void *priv,
 
 	strscpy(capability->driver, WL1273_FM_DRIVER_NAME,
 		sizeof(capability->driver));
-	strscpy(capability->card, "Texas Instruments Wl1273 FM Radio",
+	strscpy(capability->card, "TI Wl1273 FM Radio",
 		sizeof(capability->card));
 	strscpy(capability->bus_info, radio->bus_type,
 		sizeof(capability->bus_info));
Signature · 1 line
2.33.0
author maintainer
#402 lore raw

PATCH 391/917 media: si470x: Avoid card name truncation

In reply to Greg Kroah-Hartman

From: Kees Cook <[email protected]>

[ Upstream commit 2908249f3878a591f7918368fdf0b7b0a6c3158c ]

The "card" string only holds 31 characters (and the terminating NUL). In order to avoid truncation, use a shorter card description instead of the current result, "Silicon Labs Si470x FM Radio Re".

Fixes 78656acdcf48 ("V4L/DVB (7038): USB radio driver for Silicon Labs Si470x FM Radio Receivers")
Fixes cc35bbddfe10 ("V4L/DVB (12416): radio-si470x: add i2c driver for si470x")
---
 drivers/media/radio/si470x/radio-si470x-i2c.c | 2 +-
 drivers/media/radio/si470x/radio-si470x-usb.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
drivers/media/radio/si470x/radio-si470x-i2c.c +1 -1
index f491420d7b538..a972c0705ac79 100644
@@ -11,7 +11,7 @@
 
 /* driver definitions */
 #define DRIVER_AUTHOR "Joonyoung Shim <[email protected]>";
-#define DRIVER_CARD "Silicon Labs Si470x FM Radio Receiver"
+#define DRIVER_CARD "Silicon Labs Si470x FM Radio"
 #define DRIVER_DESC "I2C radio driver for Si470x FM Radio Receivers"
 #define DRIVER_VERSION "1.0.2"
 
drivers/media/radio/si470x/radio-si470x-usb.c +1 -1
index fedff68d8c496..3f8634a465730 100644
@@ -16,7 +16,7 @@
 
 /* driver definitions */
 #define DRIVER_AUTHOR "Tobias Lorenz <[email protected]>"
-#define DRIVER_CARD "Silicon Labs Si470x FM Radio Receiver"
+#define DRIVER_CARD "Silicon Labs Si470x FM Radio"
 #define DRIVER_DESC "USB radio driver for Si470x FM Radio Receivers"
 #define DRIVER_VERSION "1.0.10"
 
Signature · 1 line
2.33.0
author maintainer
#403 lore raw

PATCH 392/917 media: tm6000: Avoid card name truncation

In reply to Greg Kroah-Hartman

From: Kees Cook <[email protected]>

[ Upstream commit 42bb98e420d454fef3614b70ea11cc59068395f6 ]

The "card" string only holds 31 characters (and the terminating NUL). In order to avoid truncation, use a shorter card description instead of the current result, "Trident TVMaster TM5600/6000/60".

Fixes e28f49b0b2a8 ("V4L/DVB: tm6000: fix some info messages")
---
 drivers/media/usb/tm6000/tm6000-video.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/media/usb/tm6000/tm6000-video.c +1 -2
index 3f650ede0c3dc..e293f6f3d1bc9 100644
@@ -852,8 +852,7 @@ static int vidioc_querycap(struct file *file, void  *priv,
 	struct tm6000_core *dev = ((struct tm6000_fh *)priv)->dev;
 
 	strscpy(cap->driver, "tm6000", sizeof(cap->driver));
-	strscpy(cap->card, "Trident TVMaster TM5600/6000/6010",
-		sizeof(cap->card));
+	strscpy(cap->card, "Trident TM5600/6000/6010", sizeof(cap->card));
 	usb_make_path(dev->udev, cap->bus_info, sizeof(cap->bus_info));
 	cap->capabilities = V4L2_CAP_VIDEO_CAPTURE | V4L2_CAP_READWRITE |
 			    V4L2_CAP_DEVICE_CAPS;
Signature · 1 line
2.33.0
author maintainer
#404 lore raw

PATCH 393/917 media: cx23885: Fix snd_card_free call on null card pointer

In reply to Greg Kroah-Hartman

From: Colin Ian King <[email protected]>

[ Upstream commit 7266dda2f1dfe151b12ef0c14eb4d4e622fb211c ]

Currently a call to snd_card_new that fails will set card with a NULL pointer, this causes a null pointer dereference on the error cleanup path when card it passed to snd_card_free. Fix this by adding a new error exit path that does not call snd_card_free and exiting via this new path.

Addresses-Coverity: ("Explicit null dereference")

Fixes 9e44d63246a9 ("[media] cx23885: Add ALSA support")
---
 drivers/media/pci/cx23885/cx23885-alsa.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/media/pci/cx23885/cx23885-alsa.c +2 -1
index ab14d35214aa8..25dc8d4dc5b73 100644
@@ -550,7 +550,7 @@ struct cx23885_audio_dev *cx23885_audio_register(struct cx23885_dev *dev)
 			   SNDRV_DEFAULT_IDX1, SNDRV_DEFAULT_STR1,
 			THIS_MODULE, sizeof(struct cx23885_audio_dev), &card);
 	if (err < 0)
-		goto error;
+		goto error_msg;
 
 	chip = (struct cx23885_audio_dev *) card->private_data;
 	chip->dev = dev;
@@ -576,6 +576,7 @@ struct cx23885_audio_dev *cx23885_audio_register(struct cx23885_dev *dev)
 
 error:
 	snd_card_free(card);
+error_msg:
 	pr_err("%s(): Failed to register analog audio adapter\n",
 	       __func__);
 
Signature · 1 line
2.33.0
author maintainer
#405 lore raw

PATCH 394/917 media: atmel: fix the ispck initialization

In reply to Greg Kroah-Hartman

From: Eugen Hristev <[email protected]>

[ Upstream commit d7f26849ed7cc875d0ff7480c2efebeeccea2bad ]

The runtime enabling of the ISPCK (internally clocks the pipeline inside the ISC) has to be done after the pm_runtime for the ISC dev has been started.

After the commit by Mauro:
the ISC failed to probe with the error:

atmel-sama5d2-isc f0008000.isc: failed to enable ispck: -13
atmel-sama5d2-isc: probe of f0008000.isc failed with error -13

This is because the enabling of the ispck is done too early in the probe, and the PM runtime returns invalid request.
Thus, moved this clock enabling after pm_runtime_idle is called.

The ISPCK is required only for sama5d2 type of ISC.
Thus, add a bool inside the isc struct that is platform dependent. For the sama7g5-isc, the enabling of the ISPCK is wrong and does not make sense. Removed it from the sama7g5 probe. In sama7g5-isc, there is only one clock, the MCK, which also clocks the internal pipeline of the ISC.

Adapted the clk_prepare and clk_unprepare to request the runtime PM for both clocks (MCK and ISPCK) in case of sama5d2-isc, and the single clock (MCK) in case of sama7g5-isc.

Fixes dd97908ee350 ("media: atmel: properly get pm_runtime")
---
 drivers/media/platform/atmel/atmel-isc-base.c | 25 ++++++------
 drivers/media/platform/atmel/atmel-isc.h      |  2 +
 .../media/platform/atmel/atmel-sama5d2-isc.c  | 39 ++++++++++---------
 .../media/platform/atmel/atmel-sama7g5-isc.c  | 22 ++---------
 4 files changed, 38 insertions(+), 50 deletions(-)
drivers/media/platform/atmel/atmel-isc-base.c +11 -14
index 136ab7cf36edc..ebf264b980f91 100644
@@ -123,11 +123,9 @@ static int isc_clk_prepare(struct clk_hw *hw)
 	struct isc_clk *isc_clk = to_isc_clk(hw);
 	int ret;
 
-	if (isc_clk->id == ISC_ISPCK) {
-		ret = pm_runtime_resume_and_get(isc_clk->dev);
-		if (ret < 0)
-			return ret;
-	}
+	ret = pm_runtime_resume_and_get(isc_clk->dev);
+	if (ret < 0)
+		return ret;
 
 	return isc_wait_clk_stable(hw);
 }
@@ -138,8 +136,7 @@ static void isc_clk_unprepare(struct clk_hw *hw)
 
 	isc_wait_clk_stable(hw);
 
-	if (isc_clk->id == ISC_ISPCK)
-		pm_runtime_put_sync(isc_clk->dev);
+	pm_runtime_put_sync(isc_clk->dev);
 }
 
 static int isc_clk_enable(struct clk_hw *hw)
@@ -186,16 +183,13 @@ static int isc_clk_is_enabled(struct clk_hw *hw)
 	u32 status;
 	int ret;
 
-	if (isc_clk->id == ISC_ISPCK) {
-		ret = pm_runtime_resume_and_get(isc_clk->dev);
-		if (ret < 0)
-			return 0;
-	}
+	ret = pm_runtime_resume_and_get(isc_clk->dev);
+	if (ret < 0)
+		return 0;
 
 	regmap_read(isc_clk->regmap, ISC_CLKSR, &status);
 
-	if (isc_clk->id == ISC_ISPCK)
-		pm_runtime_put_sync(isc_clk->dev);
+	pm_runtime_put_sync(isc_clk->dev);
 
 	return status & ISC_CLK(isc_clk->id) ? 1 : 0;
 }
@@ -325,6 +319,9 @@ static int isc_clk_register(struct isc_device *isc, unsigned int id)
 	const char *parent_names[3];
 	int num_parents;
 
+	if (id == ISC_ISPCK && !isc->ispck_required)
+		return 0;
+
 	num_parents = of_clk_get_parent_count(np);
 	if (num_parents < 1 || num_parents > 3)
 		return -EINVAL;
drivers/media/platform/atmel/atmel-isc.h +2 -0
index 19cc60dfcbe0f..2bfcb135ef13b 100644
@@ -178,6 +178,7 @@ struct isc_reg_offsets {
  * @hclock:		Hclock clock input (refer datasheet)
  * @ispck:		iscpck clock (refer datasheet)
  * @isc_clks:		ISC clocks
+ * @ispck_required:	ISC requires ISP Clock initialization
  * @dcfg:		DMA master configuration, architecture dependent
  *
  * @dev:		Registered device driver
@@ -252,6 +253,7 @@ struct isc_device {
 	struct clk		*hclock;
 	struct clk		*ispck;
 	struct isc_clk		isc_clks[2];
+	bool			ispck_required;
 	u32			dcfg;
 
 	struct device		*dev;
drivers/media/platform/atmel/atmel-sama5d2-isc.c +21 -18
index b66f1d174e9d7..e29a9193bac81 100644
@@ -454,6 +454,9 @@ static int atmel_isc_probe(struct platform_device *pdev)
 	/* sama5d2-isc - 8 bits per beat */
 	isc->dcfg = ISC_DCFG_YMBSIZE_BEATS8 | ISC_DCFG_CMBSIZE_BEATS8;
 
+	/* sama5d2-isc : ISPCK is required and mandatory */
+	isc->ispck_required = true;
+
 	ret = isc_pipeline_init(isc);
 	if (ret)
 		return ret;
@@ -476,22 +479,6 @@ static int atmel_isc_probe(struct platform_device *pdev)
 		dev_err(dev, "failed to init isc clock: %d\n", ret);
 		goto unprepare_hclk;
 	}
-
-	isc->ispck = isc->isc_clks[ISC_ISPCK].clk;
-
-	ret = clk_prepare_enable(isc->ispck);
-	if (ret) {
-		dev_err(dev, "failed to enable ispck: %d\n", ret);
-		goto unprepare_hclk;
-	}
-
-	/* ispck should be greater or equal to hclock */
-	ret = clk_set_rate(isc->ispck, clk_get_rate(isc->hclock));
-	if (ret) {
-		dev_err(dev, "failed to set ispck rate: %d\n", ret);
-		goto unprepare_clk;
-	}
-
 	ret = v4l2_device_register(dev, &isc->v4l2_dev);
 	if (ret) {
 		dev_err(dev, "unable to register v4l2 device.\n");
@@ -545,19 +532,35 @@ static int atmel_isc_probe(struct platform_device *pdev)
 	pm_runtime_enable(dev);
 	pm_request_idle(dev);
 
+	isc->ispck = isc->isc_clks[ISC_ISPCK].clk;
+
+	ret = clk_prepare_enable(isc->ispck);
+	if (ret) {
+		dev_err(dev, "failed to enable ispck: %d\n", ret);
+		goto cleanup_subdev;
+	}
+
+	/* ispck should be greater or equal to hclock */
+	ret = clk_set_rate(isc->ispck, clk_get_rate(isc->hclock));
+	if (ret) {
+		dev_err(dev, "failed to set ispck rate: %d\n", ret);
+		goto unprepare_clk;
+	}
+
 	regmap_read(isc->regmap, ISC_VERSION + isc->offsets.version, &ver);
 	dev_info(dev, "Microchip ISC version %x\n", ver);
 
 	return 0;
 
+unprepare_clk:
+	clk_disable_unprepare(isc->ispck);
+
 cleanup_subdev:
 	isc_subdev_cleanup(isc);
 
 unregister_v4l2_device:
 	v4l2_device_unregister(&isc->v4l2_dev);
 
-unprepare_clk:
-	clk_disable_unprepare(isc->ispck);
 unprepare_hclk:
 	clk_disable_unprepare(isc->hclock);
 
drivers/media/platform/atmel/atmel-sama7g5-isc.c +4 -18
index f2785131ff569..9c05acafd0724 100644
@@ -447,6 +447,9 @@ static int microchip_xisc_probe(struct platform_device *pdev)
 	/* sama7g5-isc RAM access port is full AXI4 - 32 bits per beat */
 	isc->dcfg = ISC_DCFG_YMBSIZE_BEATS32 | ISC_DCFG_CMBSIZE_BEATS32;
 
+	/* sama7g5-isc : ISPCK does not exist, ISC is clocked by MCK */
+	isc->ispck_required = false;
+
 	ret = isc_pipeline_init(isc);
 	if (ret)
 		return ret;
@@ -470,25 +473,10 @@ static int microchip_xisc_probe(struct platform_device *pdev)
 		goto unprepare_hclk;
 	}
 
-	isc->ispck = isc->isc_clks[ISC_ISPCK].clk;
-
-	ret = clk_prepare_enable(isc->ispck);
-	if (ret) {
-		dev_err(dev, "failed to enable ispck: %d\n", ret);
-		goto unprepare_hclk;
-	}
-
-	/* ispck should be greater or equal to hclock */
-	ret = clk_set_rate(isc->ispck, clk_get_rate(isc->hclock));
-	if (ret) {
-		dev_err(dev, "failed to set ispck rate: %d\n", ret);
-		goto unprepare_clk;
-	}
-
 	ret = v4l2_device_register(dev, &isc->v4l2_dev);
 	if (ret) {
 		dev_err(dev, "unable to register v4l2 device.\n");
-		goto unprepare_clk;
+		goto unprepare_hclk;
 	}
 
 	ret = xisc_parse_dt(dev, isc);
@@ -549,8 +537,6 @@ cleanup_subdev:
 unregister_v4l2_device:
 	v4l2_device_unregister(&isc->v4l2_dev);
 
-unprepare_clk:
-	clk_disable_unprepare(isc->ispck);
 unprepare_hclk:
 	clk_disable_unprepare(isc->hclock);
 
Signature · 1 line
2.33.0
author maintainer
#406 lore raw

PATCH 395/917 scs: Release kasan vmalloc poison in scs_free process

In reply to Greg Kroah-Hartman

From: Yee Lee <[email protected]>

[ Upstream commit 528a4ab45300fa6283556d9b48e26b45a8aa15c4 ]

Since scs allocation is moved to vmalloc region, the shadow stack is protected by kasan_posion_vmalloc. However, the vfree_atomic operation needs to access its context for scs_free process and causes kasan error as the dump info below.

This patch Adds kasan_unpoison_vmalloc() before vfree_atomic, which aligns to the prior flow as using kmem_cache. The vmalloc region will go back posioned in the following vumap() operations.

==================================================================

 BUG: KASAN: vmalloc-out-of-bounds in llist_add_batch+0x60/0xd4
 Write of size 8 at addr ffff8000100b9000 by task kthreadd/2

 CPU: 0 PID: 2 Comm: kthreadd Not tainted 5.15.0-rc2-11681-g92477dd1faa6-dirty #1
 Hardware name: linux,dummy-virt (DT)
 Call trace:
  dump_backtrace+0x0/0x43c
  show_stack+0x1c/0x2c
  dump_stack_lvl+0x68/0x84
  print_address_description+0x80/0x394
  kasan_report+0x180/0x1dc
  __asan_report_store8_noabort+0x48/0x58
  llist_add_batch+0x60/0xd4
  vfree_atomic+0x60/0xe0
  scs_free+0x1dc/0x1fc
  scs_release+0xa4/0xd4
  free_task+0x30/0xe4
  __put_task_struct+0x1ec/0x2e0
  delayed_put_task_struct+0x5c/0xa0
  rcu_do_batch+0x62c/0x8a0
  rcu_core+0x60c/0xc14
  rcu_core_si+0x14/0x24
  __do_softirq+0x19c/0x68c
  irq_exit+0x118/0x2dc
  handle_domain_irq+0xcc/0x134
  gic_handle_irq+0x7c/0x1bc
  call_on_irq_stack+0x40/0x70
  do_interrupt_handler+0x78/0x9c
  el1_interrupt+0x34/0x60
  el1h_64_irq_handler+0x1c/0x2c
  el1h_64_irq+0x78/0x7c
  _raw_spin_unlock_irqrestore+0x40/0xcc
  sched_fork+0x4f0/0xb00
  copy_process+0xacc/0x3648
  kernel_clone+0x168/0x534
  kernel_thread+0x13c/0x1b0
  kthreadd+0x2bc/0x400
  ret_from_fork+0x10/0x20

 Memory state around the buggy address:
  ffff8000100b8f00: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8
  ffff8000100b8f80: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8
 >ffff8000100b9000: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8
                    ^
  ffff8000100b9080: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8
  ffff8000100b9100: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8
 ==================================================================
---
 kernel/scs.c | 1 +
 1 file changed, 1 insertion(+)
kernel/scs.c +1 -0
index e2a71fc82fa06..579841be88646 100644
@@ -78,6 +78,7 @@ void scs_free(void *s)
 		if (this_cpu_cmpxchg(scs_cache[i], 0, s) == NULL)
 			return;
 
+	kasan_unpoison_vmalloc(s, SCS_SIZE);
 	vfree_atomic(s);
 }
 
Signature · 1 line
2.33.0
author maintainer
#407 lore raw

PATCH 396/917 kprobes: Do not use local variable when creating debugfs file

In reply to Greg Kroah-Hartman

From: Punit Agrawal <[email protected]>

[ Upstream commit 8f7262cd66699a4b02eb7549b35c81b2116aad95 ]

debugfs_create_file() takes a pointer argument that can be used during file operation callbacks (accessible via i_private in the inode structure). An obvious requirement is for the pointer to refer to valid memory when used.

When creating the debugfs file to dynamically enable / disable kprobes, a pointer to local variable is passed to debugfs_create_file(); which will go out of scope when the init function returns. The reason this hasn't triggered random memory corruption is because the pointer is not accessed during the debugfs file callbacks.

Since the enabled state is managed by the kprobes_all_disabled global variable, the local variable is not needed. Fix the incorrect (and unnecessary) usage of local variable during debugfs_file_create() by passing NULL instead.

Fixes bf8f6e5b3e51 ("Kprobes: The ON/OFF knob thru debugfs")
---
 kernel/kprobes.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
kernel/kprobes.c +1 -2
index 790a573bbe00c..1cf8bca1ea861 100644
@@ -2809,13 +2809,12 @@ static const struct file_operations fops_kp = {
 static int __init debugfs_kprobe_init(void)
 {
 	struct dentry *dir;
-	unsigned int value = 1;
 
 	dir = debugfs_create_dir("kprobes", NULL);
 
 	debugfs_create_file("list", 0400, dir, NULL, &kprobes_fops);
 
-	debugfs_create_file("enabled", 0600, dir, &value, &fops_kp);
+	debugfs_create_file("enabled", 0600, dir, NULL, &fops_kp);
 
 	debugfs_create_file("blacklist", 0400, dir, NULL,
 			    &kprobe_blacklist_fops);
Signature · 1 line
2.33.0
author maintainer
#408 lore raw

PATCH 397/917 crypto: ecc - fix CRYPTO_DEFAULT_RNG dependency

In reply to Greg Kroah-Hartman

From: Arnd Bergmann <[email protected]>

[ Upstream commit 38aa192a05f22f9778f9420e630f0322525ef12e ]

The ecc.c file started out as part of the ECDH algorithm but got moved out into a standalone module later. It does not build without CRYPTO_DEFAULT_RNG, so now that other modules are using it as well we can run into this link error:

aarch64-linux-ld: ecc.c:(.text+0xfc8): undefined reference to `crypto_default_rng'
aarch64-linux-ld: ecc.c:(.text+0xff4): undefined reference to `crypto_put_default_rng'

Move the 'select CRYPTO_DEFAULT_RNG' statement into the correct symbol.

Fixes 0d7a78643f69 ("crypto: ecrdsa - add EC-RDSA (GOST 34.10) algorithm")
Fixes 4e6602916bc6 ("crypto: ecdsa - Add support for ECDSA signature verification")
---
 crypto/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
crypto/Kconfig +1 -1
index 536df4b6b825c..285f82647d2b7 100644
@@ -233,12 +233,12 @@ config CRYPTO_DH
 
 config CRYPTO_ECC
 	tristate
+	select CRYPTO_RNG_DEFAULT
 
 config CRYPTO_ECDH
 	tristate "ECDH algorithm"
 	select CRYPTO_ECC
 	select CRYPTO_KPP
-	select CRYPTO_RNG_DEFAULT
 	help
 	  Generic implementation of the ECDH algorithm
 
Signature · 1 line
2.33.0
author maintainer
#409 lore raw

PATCH 398/917 drm: fb_helper: fix CONFIG_FB dependency

In reply to Greg Kroah-Hartman

From: Arnd Bergmann <[email protected]>

[ Upstream commit 606b102876e3741851dfb09d53f3ee57f650a52c ]

With CONFIG_FB=m and CONFIG_DRM=y, we get a link error in the fb helper:

aarch64-linux-ld: drivers/gpu/drm/drm_fb_helper.o: in function `drm_fb_helper_alloc_fbi':
(.text+0x10cc): undefined reference to `framebuffer_alloc'

Tighten the dependency so it is only allowed in the case that DRM can link against FB.

---
 drivers/gpu/drm/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/Kconfig +1 -1
index cea777ae7fb92..9199f53861cac 100644
@@ -103,7 +103,7 @@ config DRM_DEBUG_DP_MST_TOPOLOGY_REFS
 config DRM_FBDEV_EMULATION
 	bool "Enable legacy fbdev support for your modesetting driver"
 	depends on DRM
-	depends on FB
+	depends on FB=y || FB=DRM
 	select DRM_KMS_HELPER
 	select FB_CFB_FILLRECT
 	select FB_CFB_COPYAREA
Signature · 1 line
2.33.0
author maintainer
#410 lore raw

PATCH 399/917 cpuidle: Fix kobject memory leaks in error paths

In reply to Greg Kroah-Hartman

From: Anel Orazgaliyeva <[email protected]>

[ Upstream commit e5f5a66c9aa9c331da5527c2e3fd9394e7091e01 ]

Commit c343bf1ba5ef ("cpuidle: Fix three reference count leaks") fixes the cleanup of kobjects; however, it removes kfree() calls altogether, leading to memory leaks.

Fix those and also defer the initialization of dev->kobj_dev until after the error check, so that we do not end up with a dangling pointer.

Fixes c343bf1ba5ef ("cpuidle: Fix three reference count leaks")
[ rjw: Subject edits ]
---
 drivers/cpuidle/sysfs.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/cpuidle/sysfs.c +4 -1
index 53ec9585ccd44..469e18547d06c 100644
@@ -488,6 +488,7 @@ static int cpuidle_add_state_sysfs(struct cpuidle_device *device)
 					   &kdev->kobj, "state%d", i);
 		if (ret) {
 			kobject_put(&kobj->kobj);
+			kfree(kobj);
 			goto error_state;
 		}
 		cpuidle_add_s2idle_attr_group(kobj);
@@ -619,6 +620,7 @@ static int cpuidle_add_driver_sysfs(struct cpuidle_device *dev)
 				   &kdev->kobj, "driver");
 	if (ret) {
 		kobject_put(&kdrv->kobj);
+		kfree(kdrv);
 		return ret;
 	}
 
@@ -705,7 +707,6 @@ int cpuidle_add_sysfs(struct cpuidle_device *dev)
 	if (!kdev)
 		return -ENOMEM;
 	kdev->dev = dev;
-	dev->kobj_dev = kdev;
 
 	init_completion(&kdev->kobj_unregister);
 
@@ -713,9 +714,11 @@ int cpuidle_add_sysfs(struct cpuidle_device *dev)
 				   "cpuidle");
 	if (error) {
 		kobject_put(&kdev->kobj);
+		kfree(kdev);
 		return error;
 	}
 
+	dev->kobj_dev = kdev;
 	kobject_uevent(&kdev->kobj, KOBJ_ADD);
 
 	return 0;
Signature · 1 line
2.33.0
author maintainer
#411 lore raw

PATCH 400/917 media: em28xx: Dont use ops->suspend if it is NULL

In reply to Greg Kroah-Hartman

From: Colin Ian King <[email protected]>

[ Upstream commit 51fa3b70d27342baf1ea8aaab3e96e5f4f26d5b2 ]

The call to ops->suspend for the dev->dev_next case can currently trigger a call on a null function pointer if ops->suspend is null. Skip over the use of function ops->suspend if it is null.

Addresses-Coverity: ("Dereference after null check")

Fixes be7fd3c3a8c5 ("media: em28xx: Hauppauge DualHD second tuner functionality")
---
 drivers/media/usb/em28xx/em28xx-core.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/media/usb/em28xx/em28xx-core.c +3 -2
index 584fa400cd7d8..acc0bf7dbe2b1 100644
@@ -1154,8 +1154,9 @@ int em28xx_suspend_extension(struct em28xx *dev)
 	dev_info(&dev->intf->dev, "Suspending extensions\n");
 	mutex_lock(&em28xx_devlist_mutex);
 	list_for_each_entry(ops, &em28xx_extension_devlist, next) {
-		if (ops->suspend)
-			ops->suspend(dev);
+		if (!ops->suspend)
+			continue;
+		ops->suspend(dev);
 		if (dev->dev_next)
 			ops->suspend(dev->dev_next);
 	}
Signature · 1 line
2.33.0
author maintainer
#412 lore raw

PATCH 401/917 ath10k: Dont always treat modem stop events as crashes

In reply to Greg Kroah-Hartman

From: Stephen Boyd <[email protected]>

[ Upstream commit 747ff7d3d7424876111b7559b7f6704762f89796 ]

When rebooting on sc7180 Trogdor devices I see the following crash from the wifi driver.

ath10k_snoc 18800000.wifi: firmware crashed! (guid 83493570-29a2-4e98-a83e-70048c47669c)

This is because a modem stop event looks just like a firmware crash to the driver, the qmi connection is closed in both cases. Use the qcom ssr notifier block to stop treating the qmi connection close event as a firmware crash signal when the modem hasn't actually crashed. See ath10k_qmi_event_server_exit() for more details.

This silences the crash message seen during every reboot.

Fixes 3f14b73c3843 ("ath10k: Enable MSA region dump support for WCN3990")
Cc Youghandhar Chintala <[email protected]>
Cc Abhishek Kumar <[email protected]>
Cc Steev Klimaszewski <[email protected]>
Cc Matthias Kaehlcke <[email protected]>
Cc Rakesh Pillai <[email protected]>
---
 drivers/net/wireless/ath/ath10k/qmi.c  |  3 +-
 drivers/net/wireless/ath/ath10k/snoc.c | 77 ++++++++++++++++++++++++++
 drivers/net/wireless/ath/ath10k/snoc.h |  5 ++
 3 files changed, 84 insertions(+), 1 deletion(-)
drivers/net/wireless/ath/ath10k/qmi.c +2 -1
index 07e478f9a808c..80fcb917fe4e1 100644
@@ -864,7 +864,8 @@ static void ath10k_qmi_event_server_exit(struct ath10k_qmi *qmi)
 
 	ath10k_qmi_remove_msa_permission(qmi);
 	ath10k_core_free_board_files(ar);
-	if (!test_bit(ATH10K_SNOC_FLAG_UNREGISTERING, &ar_snoc->flags))
+	if (!test_bit(ATH10K_SNOC_FLAG_UNREGISTERING, &ar_snoc->flags) &&
+	    !test_bit(ATH10K_SNOC_FLAG_MODEM_STOPPED, &ar_snoc->flags))
 		ath10k_snoc_fw_crashed_dump(ar);
 
 	ath10k_snoc_fw_indication(ar, ATH10K_QMI_EVENT_FW_DOWN_IND);
drivers/net/wireless/ath/ath10k/snoc.c +77 -0

105 lines not on this page · raw message

drivers/net/wireless/ath/ath10k/snoc.h +5 -0
index 5095d1893681b..d4bce17076960 100644
@@ -6,6 +6,8 @@
 #ifndef _SNOC_H_
 #define _SNOC_H_
 
+#include <linux/notifier.h>
+
 #include "hw.h"
 #include "ce.h"
 #include "qmi.h"
@@ -45,6 +47,7 @@ struct ath10k_snoc_ce_irq {
 enum ath10k_snoc_flags {
 	ATH10K_SNOC_FLAG_REGISTERED,
 	ATH10K_SNOC_FLAG_UNREGISTERING,
+	ATH10K_SNOC_FLAG_MODEM_STOPPED,
 	ATH10K_SNOC_FLAG_RECOVERY,
 	ATH10K_SNOC_FLAG_8BIT_HOST_CAP_QUIRK,
 };
@@ -75,6 +78,8 @@ struct ath10k_snoc {
 	struct clk_bulk_data *clks;
 	size_t num_clks;
 	struct ath10k_qmi *qmi;
+	struct notifier_block nb;
+	void *notifier;
 	unsigned long flags;
 	bool xo_cal_supported;
 	u32 xo_cal_data;
Signature · 1 line
2.33.0
author maintainer
#413 lore raw

PATCH 402/917 ath9k: Fix potential interrupt storm on queue reset

In reply to Greg Kroah-Hartman

From: Linus Lüssing <[email protected]>

[ Upstream commit 4925642d541278575ad1948c5924d71ffd57ef14 ]

In tests with two Lima boards from 8devices (QCA4531 based) on OpenWrt 19.07 we could force a silent restart of a device with no serial output when we were sending a high amount of UDP traffic (iperf3 at 80 MBit/s in both directions from external hosts, saturating the wifi and causing a load of about 4.5 to 6) and were then triggering an ath9k_queue_reset().

Further debugging showed that the restart was caused by the ath79 watchdog. With disabled watchdog we could observe that the device was constantly going into ath_isr() interrupt handler and was returning early after the ATH_OP_HW_RESET flag test, without clearing any interrupts. Even though ath9k_queue_reset() calls ath9k_hw_kill_interrupts().

With JTAG we could observe the following race condition:

1) ath9k_queue_reset()
...
-> ath9k_hw_kill_interrupts()
-> set_bit(ATH_OP_HW_RESET, &common->op_flags);
...
<- returns

      2) ath9k_tasklet()
         ...
         -> ath9k_hw_resume_interrupts()
         ...
         <- returns

                 3) loops around:
                    ...
                    handle_int()
                    -> ath_isr()
                       ...
                       -> if (test_bit(ATH_OP_HW_RESET,
                                       &common->op_flags))
                            return IRQ_HANDLED;

                    x) ath_reset_internal():
                       => never reached <=

And in ath_isr() we would typically see the following interrupts / interrupt causes:

* status: 0x00111030 or 0x00110030
* async_cause: 2 (AR_INTR_MAC_IPQ)
* sync_cause: 0

So the ath9k_tasklet() reenables the ath9k interrupts through ath9k_hw_resume_interrupts() which ath9k_queue_reset() had just disabled. And ath_isr() then keeps firing because it returns IRQ_HANDLED without actually clearing the interrupt.

To fix this IRQ storm also clear/disable the interrupts again when we are in reset state.

Cc Sven Eckelmann <[email protected]>
Cc Simon Wunderlich <[email protected]>
Cc Linus Lüssing <[email protected]>
Fixes 872b5d814f99 ("ath9k: do not access hardware on IRQs during reset")
---
 drivers/net/wireless/ath/ath9k/main.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/net/wireless/ath/ath9k/main.c +3 -1
index 139831539da37..98090e40e1cf4 100644
@@ -533,8 +533,10 @@ irqreturn_t ath_isr(int irq, void *dev)
 	ath9k_debug_sync_cause(sc, sync_cause);
 	status &= ah->imask;	/* discard unasked-for bits */
 
-	if (test_bit(ATH_OP_HW_RESET, &common->op_flags))
+	if (test_bit(ATH_OP_HW_RESET, &common->op_flags)) {
+		ath9k_hw_kill_interrupts(sc->sc_ah);
 		return IRQ_HANDLED;
+	}
 
 	/*
 	 * If there are no status bits set, then this interrupt was not
Signature · 1 line
2.33.0
author maintainer
#414 lore raw

PATCH 403/917 PM: EM: Fix inefficient states detection

In reply to Greg Kroah-Hartman

From: Vincent Donnefort <[email protected]>

[ Upstream commit aa1a43262ad5df010768f69530fa179ff81651d3 ]

Currently, a debug message is printed if an inefficient state is detected in the Energy Model. Unfortunately, it won't detect if the first state is inefficient or if two successive states are. Fix this behavior.

---
 kernel/power/energy_model.c | 23 ++++++++---------------
 1 file changed, 8 insertions(+), 15 deletions(-)
kernel/power/energy_model.c +8 -15
index a332ccd829e24..97e62469a6b32 100644
@@ -107,8 +107,7 @@ static void em_debug_remove_pd(struct device *dev) {}
 static int em_create_perf_table(struct device *dev, struct em_perf_domain *pd,
 				int nr_states, struct em_data_callback *cb)
 {
-	unsigned long opp_eff, prev_opp_eff = ULONG_MAX;
-	unsigned long power, freq, prev_freq = 0;
+	unsigned long power, freq, prev_freq = 0, prev_cost = ULONG_MAX;
 	struct em_perf_state *table;
 	int i, ret;
 	u64 fmax;
@@ -153,27 +152,21 @@ static int em_create_perf_table(struct device *dev, struct em_perf_domain *pd,
 
 		table[i].power = power;
 		table[i].frequency = prev_freq = freq;
-
-		/*
-		 * The hertz/watts efficiency ratio should decrease as the
-		 * frequency grows on sane platforms. But this isn't always
-		 * true in practice so warn the user if a higher OPP is more
-		 * power efficient than a lower one.
-		 */
-		opp_eff = freq / power;
-		if (opp_eff >= prev_opp_eff)
-			dev_dbg(dev, "EM: hertz/watts ratio non-monotonically decreasing: em_perf_state %d >= em_perf_state%d\n",
-					i, i - 1);
-		prev_opp_eff = opp_eff;
 	}
 
 	/* Compute the cost of each performance state. */
 	fmax = (u64) table[nr_states - 1].frequency;
-	for (i = 0; i < nr_states; i++) {
+	for (i = nr_states - 1; i >= 0; i--) {
 		unsigned long power_res = em_scale_power(table[i].power);
 
 		table[i].cost = div64_u64(fmax * power_res,
 					  table[i].frequency);
+		if (table[i].cost >= prev_cost) {
+			dev_dbg(dev, "EM: OPP:%lu is inefficient\n",
+				table[i].frequency);
+		} else {
+			prev_cost = table[i].cost;
+		}
 	}
 
 	pd->table = table;
Signature · 1 line
2.33.0
author maintainer
#415 lore raw

PATCH 404/917 x86/insn: Use get_unaligned() instead of memcpy()

In reply to Greg Kroah-Hartman

From: Borislav Petkov <[email protected]>

[ Upstream commit f96b4675839b66168f5a07bf964dde6c2f1c4885 ]

Use get_unaligned() instead of memcpy() to access potentially unaligned memory, which, when accessed through a pointer, leads to undefined behavior. get_unaligned() describes much better what is happening there anyway even if memcpy() does the job.

In addition, since perf tool builds with -Werror, it would fire with:

util/intel-pt-decoder/../../../arch/x86/lib/insn.c: In function '__insn_get_emulate_prefix':
tools/include/../include/asm-generic/unaligned.h:10:15: error: packed attribute is unnecessary [-Werror=packed]

     10 |  const struct { type x; } __packed *__pptr = (typeof(__pptr))(ptr); \

because -Werror=packed would complain if the packed attribute would have no effect on the layout of the structure.

In this case, that is intentional so disable the warning only for that compilation unit.

That part is Reported-by: Stephen Rothwell <[email protected]>

No functional changes.

Fixes 5ba1071f7554 ("x86/insn, tools/x86: Fix undefined behavior due to potential unaligned accesses")
---
 arch/x86/lib/insn.c                    |  5 +++--
 tools/arch/x86/lib/insn.c              |  5 +++--
 tools/include/asm-generic/unaligned.h  | 23 +++++++++++++++++++++++
 tools/perf/util/intel-pt-decoder/Build |  2 ++
 4 files changed, 31 insertions(+), 4 deletions(-)
 create mode 100644 tools/include/asm-generic/unaligned.h
arch/x86/lib/insn.c +3 -2
index c565def611e24..55e371cc69fd5 100644
@@ -13,6 +13,7 @@
 #endif
 #include <asm/inat.h> /*__ignore_sync_check__ */
 #include <asm/insn.h> /* __ignore_sync_check__ */
+#include <asm/unaligned.h> /* __ignore_sync_check__ */
 
 #include <linux/errno.h>
 #include <linux/kconfig.h>
@@ -37,10 +38,10 @@
 	((insn)->next_byte + sizeof(t) + n <= (insn)->end_kaddr)
 
 #define __get_next(t, insn)	\
-	({ t r; memcpy(&r, insn->next_byte, sizeof(t)); insn->next_byte += sizeof(t); leXX_to_cpu(t, r); })
+	({ t r = get_unaligned((t *)(insn)->next_byte); (insn)->next_byte += sizeof(t); leXX_to_cpu(t, r); })
 
 #define __peek_nbyte_next(t, insn, n)	\
-	({ t r; memcpy(&r, (insn)->next_byte + n, sizeof(t)); leXX_to_cpu(t, r); })
+	({ t r = get_unaligned((t *)(insn)->next_byte + n); leXX_to_cpu(t, r); })
 
 #define get_next(t, insn)	\
 	({ if (unlikely(!validate_next(t, insn, 0))) goto err_out; __get_next(t, insn); })
tools/arch/x86/lib/insn.c +3 -2
index 797699462cd8e..8fd63a067308a 100644
@@ -13,6 +13,7 @@
 #endif
 #include "../include/asm/inat.h" /* __ignore_sync_check__ */
 #include "../include/asm/insn.h" /* __ignore_sync_check__ */
+#include "../include/asm-generic/unaligned.h" /* __ignore_sync_check__ */
 
 #include <linux/errno.h>
 #include <linux/kconfig.h>
@@ -37,10 +38,10 @@
 	((insn)->next_byte + sizeof(t) + n <= (insn)->end_kaddr)
 
 #define __get_next(t, insn)	\
-	({ t r; memcpy(&r, insn->next_byte, sizeof(t)); insn->next_byte += sizeof(t); leXX_to_cpu(t, r); })
+	({ t r = get_unaligned((t *)(insn)->next_byte); (insn)->next_byte += sizeof(t); leXX_to_cpu(t, r); })
 
 #define __peek_nbyte_next(t, insn, n)	\
-	({ t r; memcpy(&r, (insn)->next_byte + n, sizeof(t)); leXX_to_cpu(t, r); })
+	({ t r = get_unaligned((t *)(insn)->next_byte + n); leXX_to_cpu(t, r); })
 
 #define get_next(t, insn)	\
 	({ if (unlikely(!validate_next(t, insn, 0))) goto err_out; __get_next(t, insn); })
tools/include/asm-generic/unaligned.h new file +23 -0
new file mode 100644
index 0000000000000..47387c607035e
@@ -0,0 +1,23 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Copied from the kernel sources to tools/perf/:
+ */
+
+#ifndef __TOOLS_LINUX_ASM_GENERIC_UNALIGNED_H
+#define __TOOLS_LINUX_ASM_GENERIC_UNALIGNED_H
+
+#define __get_unaligned_t(type, ptr) ({						\
+	const struct { type x; } __packed *__pptr = (typeof(__pptr))(ptr);	\
+	__pptr->x;								\
+})
+
+#define __put_unaligned_t(type, val, ptr) do {					\
+	struct { type x; } __packed *__pptr = (typeof(__pptr))(ptr);		\
+	__pptr->x = (val);							\
+} while (0)
+
+#define get_unaligned(ptr)	__get_unaligned_t(typeof(*(ptr)), (ptr))
+#define put_unaligned(val, ptr) __put_unaligned_t(typeof(*(ptr)), (val), (ptr))
+
+#endif /* __TOOLS_LINUX_ASM_GENERIC_UNALIGNED_H */
+
tools/perf/util/intel-pt-decoder/Build +2 -0
index bc629359826fb..b41c2e9c6f887 100644
@@ -18,3 +18,5 @@ CFLAGS_intel-pt-insn-decoder.o += -I$(OUTPUT)util/intel-pt-decoder
 ifeq ($(CC_NO_CLANG), 1)
   CFLAGS_intel-pt-insn-decoder.o += -Wno-override-init
 endif
+
+CFLAGS_intel-pt-insn-decoder.o += -Wno-packed
Signature · 1 line
2.33.0
author maintainer
#416 lore raw

PATCH 405/917 EDAC/amd64: Handle three rank interleaving mode

In reply to Greg Kroah-Hartman

From: Yazen Ghannam <[email protected]>

[ Upstream commit 9f4873fb6af7966de8fcbd95c36b61351c1c4b1f ]

AMD Rome systems and later support interleaving between three identical ranks within a channel.

Check for this mode by counting the number of enabled chip selects and comparing their masks. If there are exactly three enabled chip selects and their masks are identical, then three rank interleaving is enabled.

The size of a rank is determined from its mask value. However, three rank interleaving doesn't follow the method of swapping an interleave bit with the most significant bit. Rather, the interleave bit is flipped and the most significant bit remains the same. There is only a single interleave bit in this case.

Account for this when determining the chip select size by keeping the most significant bit at its original value and ignoring any zero bits. This will return a full bitmask in [MSB:1].

Fixes e53a3b267fb0 ("EDAC/amd64: Find Chip Select memory size using Address Mask")
---
 drivers/edac/amd64_edac.c | 22 +++++++++++++++++++++-
 1 file changed, 21 insertions(+), 1 deletion(-)
drivers/edac/amd64_edac.c +21 -1
index 99b06a3e8fb12..4fce75013674f 100644
@@ -1065,12 +1065,14 @@ static void debug_dump_dramcfg_low(struct amd64_pvt *pvt, u32 dclr, int chan)
 #define CS_ODD_PRIMARY		BIT(1)
 #define CS_EVEN_SECONDARY	BIT(2)
 #define CS_ODD_SECONDARY	BIT(3)
+#define CS_3R_INTERLEAVE	BIT(4)
 
 #define CS_EVEN			(CS_EVEN_PRIMARY | CS_EVEN_SECONDARY)
 #define CS_ODD			(CS_ODD_PRIMARY | CS_ODD_SECONDARY)
 
 static int f17_get_cs_mode(int dimm, u8 ctrl, struct amd64_pvt *pvt)
 {
+	u8 base, count = 0;
 	int cs_mode = 0;
 
 	if (csrow_enabled(2 * dimm, ctrl, pvt))
@@ -1083,6 +1085,20 @@ static int f17_get_cs_mode(int dimm, u8 ctrl, struct amd64_pvt *pvt)
 	if (csrow_sec_enabled(2 * dimm + 1, ctrl, pvt))
 		cs_mode |= CS_ODD_SECONDARY;
 
+	/*
+	 * 3 Rank inteleaving support.
+	 * There should be only three bases enabled and their two masks should
+	 * be equal.
+	 */
+	for_each_chip_select(base, ctrl, pvt)
+		count += csrow_enabled(base, ctrl, pvt);
+
+	if (count == 3 &&
+	    pvt->csels[ctrl].csmasks[0] == pvt->csels[ctrl].csmasks[1]) {
+		edac_dbg(1, "3R interleaving in use.\n");
+		cs_mode |= CS_3R_INTERLEAVE;
+	}
+
 	return cs_mode;
 }
 
@@ -1891,10 +1907,14 @@ static int f17_addr_mask_to_cs_size(struct amd64_pvt *pvt, u8 umc,
 	 *
 	 * The MSB is the number of bits in the full mask because BIT[0] is
 	 * always 0.
+	 *
+	 * In the special 3 Rank interleaving case, a single bit is flipped
+	 * without swapping with the most significant bit. This can be handled
+	 * by keeping the MSB where it is and ignoring the single zero bit.
 	 */
 	msb = fls(addr_mask_orig) - 1;
 	weight = hweight_long(addr_mask_orig);
-	num_zero_bits = msb - weight;
+	num_zero_bits = msb - weight - !!(cs_mode & CS_3R_INTERLEAVE);
 
 	/* Take the number of zero bits off from the top of the mask. */
 	addr_mask_deinterleaved = GENMASK_ULL(msb - num_zero_bits, 1);
Signature · 1 line
2.33.0
author maintainer
#417 lore raw

PATCH 406/917 rcu: Always inline rcu_dynticks_task*_{enter,exit}()

In reply to Greg Kroah-Hartman

From: Peter Zijlstra <[email protected]>

[ Upstream commit 7663ad9a5dbcc27f3090e6bfd192c7e59222709f ]

RCU managed to grow a few noinstr violations:

vmlinux.o: warning: objtool: rcu_dynticks_eqs_enter()+0x0: call to rcu_dynticks_task_trace_enter() leaves .noinstr.text section vmlinux.o: warning: objtool: rcu_dynticks_eqs_exit()+0xe: call to rcu_dynticks_task_trace_exit() leaves .noinstr.text section

Fix them by adding __always_inline to the relevant trivial functions.

Also replace the noinstr with __always_inline for the existing rcu_dynticks_task_*() functions since noinstr would force noinline them, even when empty, which seems silly.

Fixes 7d0c9c50c5a1 ("rcu-tasks: Avoid IPIing userspace/idle tasks if kernel is so built")
---
 kernel/rcu/tree_plugin.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
kernel/rcu/tree_plugin.h +4 -4
index d070059163d70..0d21a5cdc7247 100644
@@ -1480,7 +1480,7 @@ static void rcu_bind_gp_kthread(void)
 }
 
 /* Record the current task on dyntick-idle entry. */
-static void noinstr rcu_dynticks_task_enter(void)
+static __always_inline void rcu_dynticks_task_enter(void)
 {
 #if defined(CONFIG_TASKS_RCU) && defined(CONFIG_NO_HZ_FULL)
 	WRITE_ONCE(current->rcu_tasks_idle_cpu, smp_processor_id());
@@ -1488,7 +1488,7 @@ static void noinstr rcu_dynticks_task_enter(void)
 }
 
 /* Record no current task on dyntick-idle exit. */
-static void noinstr rcu_dynticks_task_exit(void)
+static __always_inline void rcu_dynticks_task_exit(void)
 {
 #if defined(CONFIG_TASKS_RCU) && defined(CONFIG_NO_HZ_FULL)
 	WRITE_ONCE(current->rcu_tasks_idle_cpu, -1);
@@ -1496,7 +1496,7 @@ static void noinstr rcu_dynticks_task_exit(void)
 }
 
 /* Turn on heavyweight RCU tasks trace readers on idle/user entry. */
-static void rcu_dynticks_task_trace_enter(void)
+static __always_inline void rcu_dynticks_task_trace_enter(void)
 {
 #ifdef CONFIG_TASKS_TRACE_RCU
 	if (IS_ENABLED(CONFIG_TASKS_TRACE_RCU_READ_MB))
@@ -1505,7 +1505,7 @@ static void rcu_dynticks_task_trace_enter(void)
 }
 
 /* Turn off heavyweight RCU tasks trace readers on idle/user exit. */
-static void rcu_dynticks_task_trace_exit(void)
+static __always_inline void rcu_dynticks_task_trace_exit(void)
 {
 #ifdef CONFIG_TASKS_TRACE_RCU
 	if (IS_ENABLED(CONFIG_TASKS_TRACE_RCU_READ_MB))
Signature · 1 line
2.33.0
author maintainer
#418 lore raw

PATCH 407/917 rcu: Fix rcu_dynticks_curr_cpu_in_eqs() vs noinstr

In reply to Greg Kroah-Hartman

From: Peter Zijlstra <[email protected]>

[ Upstream commit 74aece72f95f399dd29363669dc32a1344c8fab4 ]

vmlinux.o: warning: objtool: rcu_nmi_enter()+0x36: call to __kasan_check_read() leaves .noinstr.text section

noinstr cannot have atomic_*() functions in because they're explicitly annotated, use arch_atomic_*().

Fixes 2be57f732889 ("rcu: Weaken ->dynticks accesses and updates")
---
 kernel/rcu/tree.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/rcu/tree.c +1 -1
index bce848e50512e..bdd1dc6de71ab 100644
@@ -327,7 +327,7 @@ static void rcu_dynticks_eqs_online(void)
  */
 static __always_inline bool rcu_dynticks_curr_cpu_in_eqs(void)
 {
-	return !(atomic_read(this_cpu_ptr(&rcu_data.dynticks)) & 0x1);
+	return !(arch_atomic_read(this_cpu_ptr(&rcu_data.dynticks)) & 0x1);
 }
 
 /*
Signature · 1 line
2.33.0
author maintainer
#419 lore raw

PATCH 408/917 netfilter: nft_dynset: relax superfluous check on set updates

In reply to Greg Kroah-Hartman

From: Pablo Neira Ayuso <[email protected]>

[ Upstream commit 7b1394892de8d95748d05e3ee41e85edb4abbfa1 ]

Relax this condition to make add and update commands idempotent for sets with no timeout. The eval function already checks if the set element timeout is available and updates it if the update command is used.

Fixes 22fe54d5fefc ("netfilter: nf_tables: add support for dynamic set updates")
---
 net/netfilter/nft_dynset.c | 11 +----------
 1 file changed, 1 insertion(+), 10 deletions(-)
net/netfilter/nft_dynset.c +1 -10
index 6ba3256fa8449..87f3af4645d9c 100644
@@ -198,17 +198,8 @@ static int nft_dynset_init(const struct nft_ctx *ctx,
 		return -EBUSY;
 
 	priv->op = ntohl(nla_get_be32(tb[NFTA_DYNSET_OP]));
-	switch (priv->op) {
-	case NFT_DYNSET_OP_ADD:
-	case NFT_DYNSET_OP_DELETE:
-		break;
-	case NFT_DYNSET_OP_UPDATE:
-		if (!(set->flags & NFT_SET_TIMEOUT))
-			return -EOPNOTSUPP;
-		break;
-	default:
+	if (priv->op > NFT_DYNSET_OP_DELETE)
 		return -EOPNOTSUPP;
-	}
 
 	timeout = 0;
 	if (tb[NFTA_DYNSET_TIMEOUT] != NULL) {
Signature · 1 line
2.33.0
author maintainer
#420 lore raw

PATCH 409/917 media: venus: fix vpp frequency calculation for decoder

In reply to Greg Kroah-Hartman

From: Mansur Alisha Shaik <[email protected]>

[ Upstream commit 1444232152ea33f5ae41fc14bade3e74d642b634 ]

In existing video driver implementation vpp frequency calculation in calculate_inst_freq() is always zero because the value of vpp_freq_per_mb is always zero for decoder.

Fixed this by correcting the calculating the vpp frequency calculation for decoder.

Fixes 3cfe5815ce0e ("media: venus: Enable low power setting for encoder")
---
 drivers/media/platform/qcom/venus/pm_helpers.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)
drivers/media/platform/qcom/venus/pm_helpers.c +6 -2
index 3e2345eb47f7c..e031fd17f4e75 100644
@@ -1085,12 +1085,16 @@ static unsigned long calculate_inst_freq(struct venus_inst *inst,
 	if (inst->state != INST_START)
 		return 0;
 
-	if (inst->session_type == VIDC_SESSION_TYPE_ENC)
+	if (inst->session_type == VIDC_SESSION_TYPE_ENC) {
 		vpp_freq_per_mb = inst->flags & VENUS_LOW_POWER ?
 			inst->clk_data.low_power_freq :
 			inst->clk_data.vpp_freq;
 
-	vpp_freq = mbs_per_sec * vpp_freq_per_mb;
+		vpp_freq = mbs_per_sec * vpp_freq_per_mb;
+	} else {
+		vpp_freq = mbs_per_sec * inst->clk_data.vpp_freq;
+	}
+
 	/* 21 / 20 is overhead factor */
 	vpp_freq += vpp_freq / 20;
 	vsp_freq = mbs_per_sec * inst->clk_data.vsp_freq;
Signature · 1 line
2.33.0
author maintainer
#421 lore raw

PATCH 410/917 media: dvb-frontends: mn88443x: Handle errors of clk_prepare_enable()

In reply to Greg Kroah-Hartman

From: Evgeny Novikov <[email protected]>

[ Upstream commit 69a10678e2fba3d182e78ea041f2d1b1a6058764 ]

mn88443x_cmn_power_on() did not handle possible errors of clk_prepare_enable() and always finished successfully so that its caller mn88443x_probe() did not care about failed preparing/enabling of clocks as well.

Add missed error handling in both mn88443x_cmn_power_on() and mn88443x_probe(). This required to change the return value of the former from "void" to "int".

Found by Linux Driver Verification project (linuxtesting.org).

Fixes 0f408ce8941f ("media: dvb-frontends: add Socionext MN88443x ISDB-S/T demodulator driver")
---
 drivers/media/dvb-frontends/mn88443x.c | 18 +++++++++++++++---
 1 file changed, 15 insertions(+), 3 deletions(-)
drivers/media/dvb-frontends/mn88443x.c +15 -3
index e4528784f8477..fff212c0bf3b5 100644
@@ -204,11 +204,18 @@ struct mn88443x_priv {
 	struct regmap *regmap_t;
 };
 
-static void mn88443x_cmn_power_on(struct mn88443x_priv *chip)
+static int mn88443x_cmn_power_on(struct mn88443x_priv *chip)
 {
+	struct device *dev = &chip->client_s->dev;
 	struct regmap *r_t = chip->regmap_t;
+	int ret;
 
-	clk_prepare_enable(chip->mclk);
+	ret = clk_prepare_enable(chip->mclk);
+	if (ret) {
+		dev_err(dev, "Failed to prepare and enable mclk: %d\n",
+			ret);
+		return ret;
+	}
 
 	gpiod_set_value_cansleep(chip->reset_gpio, 1);
 	usleep_range(100, 1000);
@@ -222,6 +229,8 @@ static void mn88443x_cmn_power_on(struct mn88443x_priv *chip)
 	} else {
 		regmap_write(r_t, HIZSET3, 0x8f);
 	}
+
+	return 0;
 }
 
 static void mn88443x_cmn_power_off(struct mn88443x_priv *chip)
@@ -738,7 +747,10 @@ static int mn88443x_probe(struct i2c_client *client,
 	chip->fe.demodulator_priv = chip;
 	i2c_set_clientdata(client, chip);
 
-	mn88443x_cmn_power_on(chip);
+	ret = mn88443x_cmn_power_on(chip);
+	if (ret)
+		goto err_i2c_t;
+
 	mn88443x_s_sleep(chip);
 	mn88443x_t_sleep(chip);
 
Signature · 1 line
2.33.0
author maintainer
#422 lore raw

PATCH 411/917 crypto: ccree - avoid out-of-range warnings from clang

In reply to Greg Kroah-Hartman

From: Arnd Bergmann <[email protected]>

[ Upstream commit cfd6fb45cfaf46fa9547421d8da387dc9c7997d4 ]

clang points out inconsistencies in the FIELD_PREP() invocation in this driver that result from the 'mask' being a 32-bit value:

drivers/crypto/ccree/cc_driver.c:117:18: error: result of comparison of constant 18446744073709551615 with expression of type 'u32' (aka 'unsigned int') is always false [-Werror,-Wtautological-constant-out-of-range-compare]

        cache_params |= FIELD_PREP(mask, val);
                        ^~~~~~~~~~~~~~~~~~~~~
include/linux/bitfield.h:94:3: note: expanded from macro 'FIELD_PREP'
                __BF_FIELD_CHECK(_mask, 0ULL, _val, "FIELD_PREP: ");    \
                ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
include/linux/bitfield.h:52:28: note: expanded from macro '__BF_FIELD_CHECK'
                BUILD_BUG_ON_MSG((_mask) > (typeof(_reg))~0ull,         \
                ~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

This does not happen in other places that just pass a constant here.

Work around the warnings by widening the type of the temporary variable.

Fixes 05c2a705917b ("crypto: ccree - rework cache parameters handling")
---
 drivers/crypto/ccree/cc_driver.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/crypto/ccree/cc_driver.c +2 -1
index e599ac6dc162a..790fa9058a36d 100644
@@ -103,7 +103,8 @@ MODULE_DEVICE_TABLE(of, arm_ccree_dev_of_match);
 static void init_cc_cache_params(struct cc_drvdata *drvdata)
 {
 	struct device *dev = drvdata_to_dev(drvdata);
-	u32 cache_params, ace_const, val, mask;
+	u32 cache_params, ace_const, val;
+	u64 mask;
 
 	/* compute CC_AXIM_CACHE_PARAMS */
 	cache_params = cc_ioread(drvdata, CC_REG(AXIM_CACHE_PARAMS));
Signature · 1 line
2.33.0
author maintainer
#423 lore raw

PATCH 412/917 crypto: qat - detect PFVF collision after ACK

In reply to Greg Kroah-Hartman

From: Giovanni Cabiddu <[email protected]>

[ Upstream commit 9b768e8a3909ac1ab39ed44a3933716da7761a6f ]

Detect a PFVF collision between the local and the remote function by checking if the message on the PFVF CSR has been overwritten. This is done after the remote function confirms that the message has been received, by clearing the interrupt bit, or the maximum number of attempts (ADF_IOV_MSG_ACK_MAX_RETRY) to check the CSR has been exceeded.

Fixes ed8ccaef52fa ("crypto: qat - Add support for SRIOV")
---
 drivers/crypto/qat/qat_common/adf_pf2vf_msg.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/crypto/qat/qat_common/adf_pf2vf_msg.c +7 -0
index 976b9ab7617cd..789a4135e28c0 100644
@@ -156,6 +156,13 @@ static int __adf_iov_putmsg(struct adf_accel_dev *accel_dev, u32 msg, u8 vf_nr)
 		val = ADF_CSR_RD(pmisc_bar_addr, pf2vf_offset);
 	} while ((val & int_bit) && (count++ < ADF_IOV_MSG_ACK_MAX_RETRY));
 
+	if (val != msg) {
+		dev_dbg(&GET_DEV(accel_dev),
+			"Collision - PFVF CSR overwritten by remote function\n");
+		ret = -EIO;
+		goto out;
+	}
+
 	if (val & int_bit) {
 		dev_dbg(&GET_DEV(accel_dev), "ACK not received from remote\n");
 		val &= ~int_bit;
Signature · 1 line
2.33.0
author maintainer
#424 lore raw

PATCH 413/917 crypto: qat - disregard spurious PFVF interrupts

In reply to Greg Kroah-Hartman

From: Giovanni Cabiddu <[email protected]>

[ Upstream commit 18fcba469ba5359c1de7e3fb16f7b9e8cd1b8e02 ]

Upon receiving a PFVF message, check if the interrupt bit is set in the message. If it is not, that means that the interrupt was probably triggered by a collision. In this case, disregard the message and re-enable the interrupts.

Fixes ed8ccaef52fa ("crypto: qat - Add support for SRIOV")
---
 drivers/crypto/qat/qat_common/adf_pf2vf_msg.c | 6 ++++++
 drivers/crypto/qat/qat_common/adf_vf_isr.c    | 6 ++++++
 2 files changed, 12 insertions(+)
drivers/crypto/qat/qat_common/adf_pf2vf_msg.c +6 -0
index 789a4135e28c0..5a41beb8f20f6 100644
@@ -211,6 +211,11 @@ void adf_vf2pf_req_hndl(struct adf_accel_vf_info *vf_info)
 
 	/* Read message from the VF */
 	msg = ADF_CSR_RD(pmisc_addr, hw_data->get_pf2vf_offset(vf_nr));
+	if (!(msg & ADF_VF2PF_INT)) {
+		dev_info(&GET_DEV(accel_dev),
+			 "Spurious VF2PF interrupt, msg %X. Ignored\n", msg);
+		goto out;
+	}
 
 	/* To ACK, clear the VF2PFINT bit */
 	msg &= ~ADF_VF2PF_INT;
@@ -294,6 +299,7 @@ void adf_vf2pf_req_hndl(struct adf_accel_vf_info *vf_info)
 	if (resp && adf_iov_putmsg(accel_dev, resp, vf_nr))
 		dev_err(&GET_DEV(accel_dev), "Failed to send response to VF\n");
 
+out:
 	/* re-enable interrupt on PF from this VF */
 	adf_enable_vf2pf_interrupts(accel_dev, (1 << vf_nr));
 
drivers/crypto/qat/qat_common/adf_vf_isr.c +6 -0
index 7828a6573f3e2..2e300c255ab94 100644
@@ -101,6 +101,11 @@ static void adf_pf2vf_bh_handler(void *data)
 
 	/* Read the message from PF */
 	msg = ADF_CSR_RD(pmisc_bar_addr, hw_data->get_pf2vf_offset(0));
+	if (!(msg & ADF_PF2VF_INT)) {
+		dev_info(&GET_DEV(accel_dev),
+			 "Spurious PF2VF interrupt, msg %X. Ignored\n", msg);
+		goto out;
+	}
 
 	if (!(msg & ADF_PF2VF_MSGORIGIN_SYSTEM))
 		/* Ignore legacy non-system (non-kernel) PF2VF messages */
@@ -149,6 +154,7 @@ static void adf_pf2vf_bh_handler(void *data)
 	msg &= ~ADF_PF2VF_INT;
 	ADF_CSR_WR(pmisc_bar_addr, hw_data->get_pf2vf_offset(0), msg);
 
+out:
 	/* Re-enable PF2VF interrupts */
 	adf_enable_pf2vf_interrupts(accel_dev);
 	return;
Signature · 1 line
2.33.0
author maintainer
#425 lore raw

PATCH 414/917 hwrng: mtk - Force runtime pm ops for sleep ops

In reply to Greg Kroah-Hartman

From: Markus Schneider-Pargmann <[email protected]>

[ Upstream commit b6f5f0c8f72d348b2d07b20d7b680ef13a7ffe98 ]

Currently mtk_rng_runtime_suspend/resume is called for both runtime pm and system sleep operations.

This is wrong as these should only be runtime ops as the name already suggests. Currently freezing the system will lead to a call to mtk_rng_runtime_suspend even if the device currently isn't active. This leads to a clock warning because it is disabled/unprepared although it isn't enabled/prepared currently.

This patch fixes this by only setting the runtime pm ops and forces to call the runtime pm ops from the system sleep ops as well if active but not otherwise.

Fixes 81d2b34508c6 ("hwrng: mtk - add runtime PM support")
---
 drivers/char/hw_random/mtk-rng.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)
drivers/char/hw_random/mtk-rng.c +7 -2
index 8ad7b515a51b8..6c00ea0085553 100644
@@ -166,8 +166,13 @@ static int mtk_rng_runtime_resume(struct device *dev)
 	return mtk_rng_init(&priv->rng);
 }
 
-static UNIVERSAL_DEV_PM_OPS(mtk_rng_pm_ops, mtk_rng_runtime_suspend,
-			    mtk_rng_runtime_resume, NULL);
+static const struct dev_pm_ops mtk_rng_pm_ops = {
+	SET_RUNTIME_PM_OPS(mtk_rng_runtime_suspend,
+			   mtk_rng_runtime_resume, NULL)
+	SET_SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend,
+				pm_runtime_force_resume)
+};
+
 #define MTK_RNG_PM_OPS (&mtk_rng_pm_ops)
 #else	/* CONFIG_PM */
 #define MTK_RNG_PM_OPS NULL
Signature · 1 line
2.33.0
author maintainer
#426 lore raw

PATCH 415/917 ima: fix deadlock when traversing "ima_default_rules".

In reply to Greg Kroah-Hartman

From: liqiong <[email protected]>

[ Upstream commit eb0782bbdfd0d7c4786216659277c3fd585afc0e ]

The current IMA ruleset is identified by the variable "ima_rules" that default to "&ima_default_rules". When loading a custom policy for the first time, the variable is updated to "&ima_policy_rules" instead. That update isn't RCU-safe, and deadlocks are possible. Indeed, some functions like ima_match_policy() may loop indefinitely when traversing "ima_default_rules" with list_for_each_entry_rcu().

When iterating over the default ruleset back to head, if the list head is "ima_default_rules", and "ima_rules" have been updated to "&ima_policy_rules", the loop condition (&entry->list != ima_rules) stays always true, traversing won't terminate, causing a soft lockup and RCU stalls.

Introduce a temporary value for "ima_rules" when iterating over the ruleset to avoid the deadlocks.

---
 security/integrity/ima/ima_policy.c | 27 ++++++++++++++++++---------
 1 file changed, 18 insertions(+), 9 deletions(-)
security/integrity/ima/ima_policy.c +18 -9
index 87b9b71cb8201..12e8adcd80a2a 100644
@@ -228,7 +228,7 @@ static struct ima_rule_entry *arch_policy_entry __ro_after_init;
 static LIST_HEAD(ima_default_rules);
 static LIST_HEAD(ima_policy_rules);
 static LIST_HEAD(ima_temp_rules);
-static struct list_head *ima_rules = &ima_default_rules;
+static struct list_head __rcu *ima_rules = (struct list_head __rcu *)(&ima_default_rules);
 
 static int ima_policy __initdata;
 
@@ -675,12 +675,14 @@ int ima_match_policy(struct user_namespace *mnt_userns, struct inode *inode,
 {
 	struct ima_rule_entry *entry;
 	int action = 0, actmask = flags | (flags << 1);
+	struct list_head *ima_rules_tmp;
 
 	if (template_desc && !*template_desc)
 		*template_desc = ima_template_desc_current();
 
 	rcu_read_lock();
-	list_for_each_entry_rcu(entry, ima_rules, list) {
+	ima_rules_tmp = rcu_dereference(ima_rules);
+	list_for_each_entry_rcu(entry, ima_rules_tmp, list) {
 
 		if (!(entry->action & actmask))
 			continue;
@@ -741,9 +743,11 @@ void ima_update_policy_flags(void)
 {
 	struct ima_rule_entry *entry;
 	int new_policy_flag = 0;
+	struct list_head *ima_rules_tmp;
 
 	rcu_read_lock();
-	list_for_each_entry(entry, ima_rules, list) {
+	ima_rules_tmp = rcu_dereference(ima_rules);
+	list_for_each_entry_rcu(entry, ima_rules_tmp, list) {
 		/*
 		 * SETXATTR_CHECK rules do not implement a full policy check
 		 * because rule checking would probably have an important
@@ -968,10 +972,10 @@ void ima_update_policy(void)
 
 	list_splice_tail_init_rcu(&ima_temp_rules, policy, synchronize_rcu);
 
-	if (ima_rules != policy) {
+	if (ima_rules != (struct list_head __rcu *)policy) {
 		ima_policy_flag = 0;
-		ima_rules = policy;
 
+		rcu_assign_pointer(ima_rules, policy);
 		/*
 		 * IMA architecture specific policy rules are specified
 		 * as strings and converted to an array of ima_entry_rules
@@ -1061,7 +1065,7 @@ static int ima_lsm_rule_init(struct ima_rule_entry *entry,
 		pr_warn("rule for LSM \'%s\' is undefined\n",
 			entry->lsm[lsm_rule].args_p);
 
-		if (ima_rules == &ima_default_rules) {
+		if (ima_rules == (struct list_head __rcu *)(&ima_default_rules)) {
 			kfree(entry->lsm[lsm_rule].args_p);
 			entry->lsm[lsm_rule].args_p = NULL;
 			result = -EINVAL;
@@ -1768,9 +1772,11 @@ void *ima_policy_start(struct seq_file *m, loff_t *pos)
 {
 	loff_t l = *pos;
 	struct ima_rule_entry *entry;
+	struct list_head *ima_rules_tmp;
 
 	rcu_read_lock();
-	list_for_each_entry_rcu(entry, ima_rules, list) {
+	ima_rules_tmp = rcu_dereference(ima_rules);
+	list_for_each_entry_rcu(entry, ima_rules_tmp, list) {
 		if (!l--) {
 			rcu_read_unlock();
 			return entry;
@@ -1789,7 +1795,8 @@ void *ima_policy_next(struct seq_file *m, void *v, loff_t *pos)
 	rcu_read_unlock();
 	(*pos)++;
 
-	return (&entry->list == ima_rules) ? NULL : entry;
+	return (&entry->list == &ima_default_rules ||
+		&entry->list == &ima_policy_rules) ? NULL : entry;
 }
 
 void ima_policy_stop(struct seq_file *m, void *v)
@@ -2014,6 +2021,7 @@ bool ima_appraise_signature(enum kernel_read_file_id id)
 	struct ima_rule_entry *entry;
 	bool found = false;
 	enum ima_hooks func;
+	struct list_head *ima_rules_tmp;
 
 	if (id >= READING_MAX_ID)
 		return false;
@@ -2021,7 +2029,8 @@ bool ima_appraise_signature(enum kernel_read_file_id id)
 	func = read_idmap[id] ?: FILE_CHECK;
 
 	rcu_read_lock();
-	list_for_each_entry_rcu(entry, ima_rules, list) {
+	ima_rules_tmp = rcu_dereference(ima_rules);
+	list_for_each_entry_rcu(entry, ima_rules_tmp, list) {
 		if (entry->action != APPRAISE)
 			continue;
 
Signature · 1 line
2.33.0
author maintainer
#427 lore raw

PATCH 416/917 b43legacy: fix a lower bounds test

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

[ Upstream commit c1c8380b0320ab757e60ed90efc8b1992a943256 ]

The problem is that "channel" is an unsigned int, when it's less 5 the value of "channel - 5" is not a negative number as one would expect but is very high positive value instead.

This means that "start" becomes a very high positive value. The result of that is that we never enter the "for (i = start; i <= end; i++) {" loop. Instead of storing the result from b43legacy_radio_aci_detect() it just uses zero.

Fixes 75388acd0cd8 ("[B43LEGACY]: add mac80211-based driver for legacy BCM43xx devices")
---
 drivers/net/wireless/broadcom/b43legacy/radio.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/broadcom/b43legacy/radio.c +1 -1
index 06891b4f837b9..fdf78c10a05c2 100644
@@ -283,7 +283,7 @@ u8 b43legacy_radio_aci_scan(struct b43legacy_wldev *dev)
 			    & 0x7FFF);
 	b43legacy_set_all_gains(dev, 3, 8, 1);
 
-	start = (channel - 5 > 0) ? channel - 5 : 1;
+	start = (channel > 5) ? channel - 5 : 1;
 	end = (channel + 5 < 14) ? channel + 5 : 13;
 
 	for (i = start; i <= end; i++) {
Signature · 1 line
2.33.0
author maintainer
#428 lore raw

PATCH 417/917 b43: fix a lower bounds test

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

[ Upstream commit 9b793db5fca44d01f72d3564a168171acf7c4076 ]

The problem is that "channel" is an unsigned int, when it's less 5 the value of "channel - 5" is not a negative number as one would expect but is very high positive value instead.

This means that "start" becomes a very high positive value. The result of that is that we never enter the "for (i = start; i <= end; i++) {" loop. Instead of storing the result from b43legacy_radio_aci_detect() it just uses zero.

---
 drivers/net/wireless/broadcom/b43/phy_g.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/broadcom/b43/phy_g.c +1 -1
index d5a1a5c582366..ac72ca39e409b 100644
@@ -2297,7 +2297,7 @@ static u8 b43_gphy_aci_scan(struct b43_wldev *dev)
 	b43_phy_mask(dev, B43_PHY_G_CRS, 0x7FFF);
 	b43_set_all_gains(dev, 3, 8, 1);
 
-	start = (channel - 5 > 0) ? channel - 5 : 1;
+	start = (channel > 5) ? channel - 5 : 1;
 	end = (channel + 5 < 14) ? channel + 5 : 13;
 
 	for (i = start; i <= end; i++) {
Signature · 1 line
2.33.0
author maintainer
#429 lore raw

PATCH 418/917 gve: Recover from queue stall due to missed IRQ

In reply to Greg Kroah-Hartman

From: John Fraker <[email protected]>

[ Upstream commit 87a7f321bb6a45e54b7d6c90d032ee5636a6ad97 ]

Don't always reset the driver on a TX timeout. Attempt to recover by kicking the queue in case an IRQ was missed.

Fixes 9e5f7d26a4c08 ("gve: Add workqueue and reset support")
---
 drivers/net/ethernet/google/gve/gve.h        |  4 +-
 drivers/net/ethernet/google/gve/gve_adminq.h |  1 +
 drivers/net/ethernet/google/gve/gve_main.c   | 48 +++++++++++++++++++-
 3 files changed, 51 insertions(+), 2 deletions(-)
drivers/net/ethernet/google/gve/gve.h +3 -1
index 2f93ed4705905..c1d4042671f9f 100644
@@ -30,7 +30,7 @@
 #define GVE_MIN_MSIX 3
 
 /* Numbers of gve tx/rx stats in stats report. */
-#define GVE_TX_STATS_REPORT_NUM	5
+#define GVE_TX_STATS_REPORT_NUM	6
 #define GVE_RX_STATS_REPORT_NUM	2
 
 /* Interval to schedule a stats report update, 20000ms. */
@@ -413,7 +413,9 @@ struct gve_tx_ring {
 	u32 q_num ____cacheline_aligned; /* queue idx */
 	u32 stop_queue; /* count of queue stops */
 	u32 wake_queue; /* count of queue wakes */
+	u32 queue_timeout; /* count of queue timeouts */
 	u32 ntfy_id; /* notification block index */
+	u32 last_kick_msec; /* Last time the queue was kicked */
 	dma_addr_t bus; /* dma address of the descr ring */
 	dma_addr_t q_resources_bus; /* dma address of the queue resources */
 	dma_addr_t complq_bus_dqo; /* dma address of the dqo.compl_ring */
drivers/net/ethernet/google/gve/gve_adminq.h +1 -0
index 47c3d8f313fcf..3953f6f7a4273 100644
@@ -270,6 +270,7 @@ enum gve_stat_names {
 	TX_LAST_COMPLETION_PROCESSED	= 5,
 	RX_NEXT_EXPECTED_SEQUENCE	= 6,
 	RX_BUFFERS_POSTED		= 7,
+	TX_TIMEOUT_CNT			= 8,
 	// stats from NIC
 	RX_QUEUE_DROP_CNT		= 65,
 	RX_NO_BUFFERS_POSTED		= 66,
drivers/net/ethernet/google/gve/gve_main.c +47 -1

68 lines not on this page · raw message

Signature · 1 line
2.33.0
author maintainer
#430 lore raw

PATCH 419/917 gve: Track RX buffer allocation failures

In reply to Greg Kroah-Hartman

From: Catherine Sullivan <[email protected]>

[ Upstream commit 1b4d1c9bab091ac6e20a3ff80c30c5cefe192bf4 ]

The rx_buf_alloc_fail counter wasn't getting updated.

Fixes 433e274b8f7b0 ("gve: Add stats for gve.")
---
 drivers/net/ethernet/google/gve/gve_rx.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/net/ethernet/google/gve/gve_rx.c +6 -1
index 94941d4e47449..16169f291ad9f 100644
@@ -514,8 +514,13 @@ static bool gve_rx_refill_buffers(struct gve_priv *priv, struct gve_rx_ring *rx)
 
 				gve_rx_free_buffer(dev, page_info, data_slot);
 				page_info->page = NULL;
-				if (gve_rx_alloc_buffer(priv, dev, page_info, data_slot))
+				if (gve_rx_alloc_buffer(priv, dev, page_info,
+							data_slot)) {
+					u64_stats_update_begin(&rx->statss);
+					rx->rx_buf_alloc_fail++;
+					u64_stats_update_end(&rx->statss);
 					break;
+				}
 			}
 		}
 		fill_cnt++;
Signature · 1 line
2.33.0
author maintainer
#431 lore raw

PATCH 420/917 mmc: sdhci-omap: Fix NULL pointer exception if regulator is not configured

In reply to Greg Kroah-Hartman

From: Tony Lindgren <[email protected]>

[ Upstream commit 8e0e7bd38b1ec7f9e5d18725ad41828be4e09859 ]

If sdhci-omap is configured for an unused device instance and the device is not set as disabled, we can get a NULL pointer dereference:

Unable to handle kernel NULL pointer dereference at virtual address

00000045
...
(regulator_set_voltage) from [<c07d7008>] (mmc_regulator_set_ocr+0x44/0xd0) (mmc_regulator_set_ocr) from [<c07e2d80>] (sdhci_set_ios+0xa4/0x490) (sdhci_set_ios) from [<c07ea690>] (sdhci_omap_set_ios+0x124/0x160) (sdhci_omap_set_ios) from [<c07c8e94>] (mmc_power_up.part.0+0x3c/0x154) (mmc_power_up.part.0) from [<c07c9d20>] (mmc_start_host+0x88/0x9c) (mmc_start_host) from [<c07cad34>] (mmc_add_host+0x58/0x7c) (mmc_add_host) from [<c07e2574>] (__sdhci_add_host+0xf0/0x22c) (__sdhci_add_host) from [<c07eaf68>] (sdhci_omap_probe+0x318/0x72c) (sdhci_omap_probe) from [<c06a39d8>] (platform_probe+0x58/0xb8)

AFAIK we are not seeing this with the devices configured in the mainline kernel but this can cause issues for folks bringing up their boards.

Fixes 7d326930d352 ("mmc: sdhci-omap: Add OMAP SDHCI driver")
---
 drivers/mmc/host/sdhci-omap.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/mmc/host/sdhci-omap.c +2 -1
index 8f4d1f003f656..3ddced779e965 100644
@@ -682,7 +682,8 @@ static void sdhci_omap_set_power(struct sdhci_host *host, unsigned char mode,
 {
 	struct mmc_host *mmc = host->mmc;
 
-	mmc_regulator_set_ocr(mmc, mmc->supply.vmmc, vdd);
+	if (!IS_ERR(mmc->supply.vmmc))
+		mmc_regulator_set_ocr(mmc, mmc->supply.vmmc, vdd);
 }
 
 static int sdhci_omap_enable_dma(struct sdhci_host *host)
Signature · 1 line
2.33.0
author maintainer
#432 lore raw

PATCH 421/917 mmc: sdhci-omap: Fix context restore

In reply to Greg Kroah-Hartman

From: Tony Lindgren <[email protected]>

[ Upstream commit d806e334d0390502cd2a820ad33d65d7f9bba618 ]

We need to restore context in a specified order with HCTL set in two phases. This is similar to what omap_hsmmc_context_restore() is doing. Otherwise SDIO can stop working on resume.

And for PM runtime and SDIO cards, we need to also save SYSCTL, IE and ISE.

This should not be a problem currently, and these patches can be applied whenever suitable.

Fixes ee0f309263a6 ("mmc: sdhci-omap: Add Support for Suspend/Resume")
---
 drivers/mmc/host/sdhci-omap.c | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)
drivers/mmc/host/sdhci-omap.c +14 -1
index 3ddced779e965..fd188b6d88f49 100644
@@ -62,6 +62,8 @@
 #define SDHCI_OMAP_IE		0x234
 #define INT_CC_EN		BIT(0)
 
+#define SDHCI_OMAP_ISE		0x238
+
 #define SDHCI_OMAP_AC12		0x23c
 #define AC12_V1V8_SIGEN		BIT(19)
 #define AC12_SCLK_SEL		BIT(23)
@@ -113,6 +115,8 @@ struct sdhci_omap_host {
 	u32			hctl;
 	u32			sysctl;
 	u32			capa;
+	u32			ie;
+	u32			ise;
 };
 
 static void sdhci_omap_start_clock(struct sdhci_omap_host *omap_host);
@@ -1245,14 +1249,23 @@ static void sdhci_omap_context_save(struct sdhci_omap_host *omap_host)
 {
 	omap_host->con = sdhci_omap_readl(omap_host, SDHCI_OMAP_CON);
 	omap_host->hctl = sdhci_omap_readl(omap_host, SDHCI_OMAP_HCTL);
+	omap_host->sysctl = sdhci_omap_readl(omap_host, SDHCI_OMAP_SYSCTL);
 	omap_host->capa = sdhci_omap_readl(omap_host, SDHCI_OMAP_CAPA);
+	omap_host->ie = sdhci_omap_readl(omap_host, SDHCI_OMAP_IE);
+	omap_host->ise = sdhci_omap_readl(omap_host, SDHCI_OMAP_ISE);
 }
 
+/* Order matters here, HCTL must be restored in two phases */
 static void sdhci_omap_context_restore(struct sdhci_omap_host *omap_host)
 {
-	sdhci_omap_writel(omap_host, SDHCI_OMAP_CON, omap_host->con);
 	sdhci_omap_writel(omap_host, SDHCI_OMAP_HCTL, omap_host->hctl);
 	sdhci_omap_writel(omap_host, SDHCI_OMAP_CAPA, omap_host->capa);
+	sdhci_omap_writel(omap_host, SDHCI_OMAP_HCTL, omap_host->hctl);
+
+	sdhci_omap_writel(omap_host, SDHCI_OMAP_SYSCTL, omap_host->sysctl);
+	sdhci_omap_writel(omap_host, SDHCI_OMAP_CON, omap_host->con);
+	sdhci_omap_writel(omap_host, SDHCI_OMAP_IE, omap_host->ie);
+	sdhci_omap_writel(omap_host, SDHCI_OMAP_ISE, omap_host->ise);
 }
 
 static int __maybe_unused sdhci_omap_suspend(struct device *dev)
Signature · 1 line
2.33.0
author maintainer
#433 lore raw

PATCH 422/917 memstick: avoid out-of-range warning

In reply to Greg Kroah-Hartman

From: Arnd Bergmann <[email protected]>

[ Upstream commit 4853396f03c3019eccf5cd113e464231e9ddf0b3 ]

clang-14 complains about a sanity check that always passes when the page size is 64KB or larger:

drivers/memstick/core/ms_block.c:1739:21: error: result of comparison of constant 65536 with expression of type 'unsigned short' is always false [-Werror,-Wtautological-constant-out-of-range-compare]

        if (msb->page_size > PAGE_SIZE) {
            ~~~~~~~~~~~~~~ ^ ~~~~~~~~~

This is fine, it will still work on all architectures, so just shut up that warning with a cast.

Fixes 0ab30494bc4f ("memstick: add support for legacy memorysticks")
---
 drivers/memstick/core/ms_block.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/memstick/core/ms_block.c +1 -1
index acf36676e388d..487e4cc2951e0 100644
@@ -1736,7 +1736,7 @@ static int msb_init_card(struct memstick_dev *card)
 	msb->pages_in_block = boot_block->attr.block_size * 2;
 	msb->block_size = msb->page_size * msb->pages_in_block;
 
-	if (msb->page_size > PAGE_SIZE) {
+	if ((size_t)msb->page_size > PAGE_SIZE) {
 		/* this isn't supported by linux at all, anyway*/
 		dbg("device page %d size isn't supported", msb->page_size);
 		return -EINVAL;
Signature · 1 line
2.33.0
author maintainer
#434 lore raw

PATCH 423/917 memstick: jmb38x_ms: use appropriate free function in jmb38x_ms_alloc_host()

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

[ Upstream commit beae4a6258e64af609ad5995cc6b6056eb0d898e ]

The "msh" pointer is device managed, meaning that memstick_alloc_host() calls device_initialize() on it. That means that it can't be free using kfree() but must instead be freed with memstick_free_host(). Otherwise it leads to a tiny memory leak of device resources.

Fixes 60fdd931d577 ("memstick: add support for JMicron jmb38x MemoryStick host controller")
---
 drivers/memstick/host/jmb38x_ms.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/memstick/host/jmb38x_ms.c +1 -1
index f9a93b0565e15..435d4c058b20e 100644
@@ -882,7 +882,7 @@ static struct memstick_host *jmb38x_ms_alloc_host(struct jmb38x_ms *jm, int cnt)
 
 	iounmap(host->addr);
 err_out_free:
-	kfree(msh);
+	memstick_free_host(msh);
 	return NULL;
 }
 
Signature · 1 line
2.33.0
author maintainer
#435 lore raw

PATCH 424/917 net, neigh: Fix NTF_EXT_LEARNED in combination with NTF_USE

In reply to Greg Kroah-Hartman

From: Daniel Borkmann <[email protected]>

[ Upstream commit e4400bbf5b15750e1b59bf4722d18d99be60c69f ]

The NTF_EXT_LEARNED neigh flag is usually propagated back to user space upon dump of the neighbor table. However, when used in combination with NTF_USE flag this is not the case despite exempting the entry from the garbage collector. This results in inconsistent state since entries are typically marked in neigh->flags with NTF_EXT_LEARNED, but here they are not. Fix it by propagating the creation flag to ___neigh_create().

Before fix:

  # ./ip/ip n replace 192.168.178.30 dev enp5s0 use extern_learn
  # ./ip/ip n
  192.168.178.30 dev enp5s0 lladdr f4:8c:50:5e:71:9a REACHABLE
  [...]

After fix:

  # ./ip/ip n replace 192.168.178.30 dev enp5s0 use extern_learn
  # ./ip/ip n
  192.168.178.30 dev enp5s0 lladdr f4:8c:50:5e:71:9a extern_learn REACHABLE
  [...]
Fixes 9ce33e46531d ("neighbour: support for NTF_EXT_LEARNED flag")
---
 net/core/neighbour.c | 26 ++++++++++++++------------
 1 file changed, 14 insertions(+), 12 deletions(-)
net/core/neighbour.c +14 -12
index 2d5bc3a75faec..8457d5f97022b 100644
@@ -379,7 +379,7 @@ EXPORT_SYMBOL(neigh_ifdown);
 
 static struct neighbour *neigh_alloc(struct neigh_table *tbl,
 				     struct net_device *dev,
-				     bool exempt_from_gc)
+				     u8 flags, bool exempt_from_gc)
 {
 	struct neighbour *n = NULL;
 	unsigned long now = jiffies;
@@ -412,6 +412,7 @@ do_alloc:
 	n->updated	  = n->used = now;
 	n->nud_state	  = NUD_NONE;
 	n->output	  = neigh_blackhole;
+	n->flags	  = flags;
 	seqlock_init(&n->hh.hh_lock);
 	n->parms	  = neigh_parms_clone(&tbl->parms);
 	timer_setup(&n->timer, neigh_timer_handler, 0);
@@ -575,19 +576,18 @@ struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, struct net *net,
 }
 EXPORT_SYMBOL(neigh_lookup_nodev);
 
-static struct neighbour *___neigh_create(struct neigh_table *tbl,
-					 const void *pkey,
-					 struct net_device *dev,
-					 bool exempt_from_gc, bool want_ref)
+static struct neighbour *
+___neigh_create(struct neigh_table *tbl, const void *pkey,
+		struct net_device *dev, u8 flags,
+		bool exempt_from_gc, bool want_ref)
 {
-	struct neighbour *n1, *rc, *n = neigh_alloc(tbl, dev, exempt_from_gc);
-	u32 hash_val;
-	unsigned int key_len = tbl->key_len;
-	int error;
+	u32 hash_val, key_len = tbl->key_len;
+	struct neighbour *n1, *rc, *n;
 	struct neigh_hash_table *nht;
+	int error;
 
+	n = neigh_alloc(tbl, dev, flags, exempt_from_gc);
 	trace_neigh_create(tbl, dev, pkey, n, exempt_from_gc);
-
 	if (!n) {
 		rc = ERR_PTR(-ENOBUFS);
 		goto out;
@@ -674,7 +674,7 @@ out_neigh_release:
 struct neighbour *__neigh_create(struct neigh_table *tbl, const void *pkey,
 				 struct net_device *dev, bool want_ref)
 {
-	return ___neigh_create(tbl, pkey, dev, false, want_ref);
+	return ___neigh_create(tbl, pkey, dev, 0, false, want_ref);
 }
 EXPORT_SYMBOL(__neigh_create);
 
@@ -1942,7 +1942,9 @@ static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh,
 
 		exempt_from_gc = ndm->ndm_state & NUD_PERMANENT ||
 				 ndm->ndm_flags & NTF_EXT_LEARNED;
-		neigh = ___neigh_create(tbl, dst, dev, exempt_from_gc, true);
+		neigh = ___neigh_create(tbl, dst, dev,
+					ndm->ndm_flags & NTF_EXT_LEARNED,
+					exempt_from_gc, true);
 		if (IS_ERR(neigh)) {
 			err = PTR_ERR(neigh);
 			goto out;
Signature · 1 line
2.33.0
author maintainer
#436 lore raw

PATCH 425/917 hwmon: Fix possible memleak in __hwmon_device_register()

In reply to Greg Kroah-Hartman

From: Yang Yingliang <[email protected]>

[ Upstream commit ada61aa0b1184a8fda1a89a340c7d6cc4e59aee5 ]

I got memory leak as follows when doing fault injection test:

unreferenced object 0xffff888102740438 (size 8):
comm "27", pid 859, jiffies 4295031351 (age 143.992s) hex dump (first 8 bytes):

    68 77 6d 6f 6e 30 00 00                          hwmon0..
  backtrace:
    [<00000000544b5996>] __kmalloc_track_caller+0x1a6/0x300
    [<00000000df0d62b9>] kvasprintf+0xad/0x140
    [<00000000d3d2a3da>] kvasprintf_const+0x62/0x190
    [<000000005f8f0f29>] kobject_set_name_vargs+0x56/0x140
    [<00000000b739e4b9>] dev_set_name+0xb0/0xe0
    [<0000000095b69c25>] __hwmon_device_register+0xf19/0x1e50 [hwmon]
    [<00000000a7e65b52>] hwmon_device_register_with_info+0xcb/0x110 [hwmon]
    [<000000006f181e86>] devm_hwmon_device_register_with_info+0x85/0x100 [hwmon]
    [<0000000081bdc567>] tmp421_probe+0x2d2/0x465 [tmp421]
    [<00000000502cc3f8>] i2c_device_probe+0x4e1/0xbb0
    [<00000000f90bda3b>] really_probe+0x285/0xc30
    [<000000007eac7b77>] __driver_probe_device+0x35f/0x4f0
    [<000000004953d43d>] driver_probe_device+0x4f/0x140
    [<000000002ada2d41>] __device_attach_driver+0x24c/0x330
    [<00000000b3977977>] bus_for_each_drv+0x15d/0x1e0
    [<000000005bf2a8e3>] __device_attach+0x267/0x410

When device_register() returns an error, the name allocated in dev_set_name() will be leaked, the put_device() should be used instead of calling hwmon_dev_release() to give up the device reference, then the name will be freed in kobject_cleanup().

Fixes bab2243ce189 ("hwmon: Introduce hwmon_device_register_with_groups")
---
 drivers/hwmon/hwmon.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/hwmon/hwmon.c +4 -2
index 8d3b1dae31df1..3501a3ead4ba6 100644
@@ -796,8 +796,10 @@ __hwmon_device_register(struct device *dev, const char *name, void *drvdata,
 	dev_set_drvdata(hdev, drvdata);
 	dev_set_name(hdev, HWMON_ID_FORMAT, id);
 	err = device_register(hdev);
-	if (err)
-		goto free_hwmon;
+	if (err) {
+		put_device(hdev);
+		goto ida_remove;
+	}
 
 	INIT_LIST_HEAD(&hwdev->tzdata);
 
Signature · 1 line
2.33.0
author maintainer
#437 lore raw

PATCH 426/917 hwmon: (pmbus/lm25066) Let compiler determine outer dimension of lm25066_coeff

In reply to Greg Kroah-Hartman

From: Zev Weiss <[email protected]>

[ Upstream commit b7931a7b0e0df4d2a25fedd895ad32c746b77bc1 ]

Maintaining this manually is error prone (there are currently only five chips supported, not six); gcc can do it for us automatically.

Fixes 666c14906b49 ("hwmon: (pmbus/lm25066) Drop support for LM25063")
---
 drivers/hwmon/pmbus/lm25066.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/hwmon/pmbus/lm25066.c +1 -1
index 1a660c4cd19f4..66d3e88b54172 100644
@@ -51,7 +51,7 @@ struct __coeff {
 #define PSC_CURRENT_IN_L	(PSC_NUM_CLASSES)
 #define PSC_POWER_L		(PSC_NUM_CLASSES + 1)
 
-static struct __coeff lm25066_coeff[6][PSC_NUM_CLASSES + 2] = {
+static struct __coeff lm25066_coeff[][PSC_NUM_CLASSES + 2] = {
 	[lm25056] = {
 		[PSC_VOLTAGE_IN] = {
 			.m = 16296,
Signature · 1 line
2.33.0
author maintainer
#438 lore raw

PATCH 427/917 ath10k: fix max antenna gain unit

In reply to Greg Kroah-Hartman

From: Sven Eckelmann <[email protected]>

[ Upstream commit 0a491167fe0cf9f26062462de2a8688b96125d48 ]

Most of the txpower for the ath10k firmware is stored as twicepower (0.5 dB steps). This isn't the case for max_antenna_gain - which is still expected by the firmware as dB.

The firmware is converting it from dB to the internal (twicepower) representation when it calculates the limits of a channel. This can be seen in tpc_stats when configuring "12" as max_antenna_gain. Instead of the expected 12 (6 dB), the tpc_stats shows 24 (12 dB).

Tested on QCA9888 and IPQ4019 with firmware 10.4-3.5.3-00057.

Fixes 02256930d9b8 ("ath10k: use proper tx power unit")
---
 drivers/net/wireless/ath/ath10k/mac.c | 6 +++---
 drivers/net/wireless/ath/ath10k/wmi.h | 3 +++
 2 files changed, 6 insertions(+), 3 deletions(-)
drivers/net/wireless/ath/ath10k/mac.c +3 -3
index 7ca68c81d9b61..5ec19d91cf372 100644
@@ -1052,7 +1052,7 @@ static int ath10k_monitor_vdev_start(struct ath10k *ar, int vdev_id)
 	arg.channel.min_power = 0;
 	arg.channel.max_power = channel->max_power * 2;
 	arg.channel.max_reg_power = channel->max_reg_power * 2;
-	arg.channel.max_antenna_gain = channel->max_antenna_gain * 2;
+	arg.channel.max_antenna_gain = channel->max_antenna_gain;
 
 	reinit_completion(&ar->vdev_setup_done);
 	reinit_completion(&ar->vdev_delete_done);
@@ -1498,7 +1498,7 @@ static int ath10k_vdev_start_restart(struct ath10k_vif *arvif,
 	arg.channel.min_power = 0;
 	arg.channel.max_power = chandef->chan->max_power * 2;
 	arg.channel.max_reg_power = chandef->chan->max_reg_power * 2;
-	arg.channel.max_antenna_gain = chandef->chan->max_antenna_gain * 2;
+	arg.channel.max_antenna_gain = chandef->chan->max_antenna_gain;
 
 	if (arvif->vdev_type == WMI_VDEV_TYPE_AP) {
 		arg.ssid = arvif->u.ap.ssid;
@@ -3426,7 +3426,7 @@ static int ath10k_update_channel_list(struct ath10k *ar)
 			ch->min_power = 0;
 			ch->max_power = channel->max_power * 2;
 			ch->max_reg_power = channel->max_reg_power * 2;
-			ch->max_antenna_gain = channel->max_antenna_gain * 2;
+			ch->max_antenna_gain = channel->max_antenna_gain;
 			ch->reg_class_id = 0; /* FIXME */
 
 			/* FIXME: why use only legacy modes, why not any
drivers/net/wireless/ath/ath10k/wmi.h +3 -0
index 41c1a3d339c25..01bfd09a9d88c 100644
@@ -2066,7 +2066,9 @@ struct wmi_channel {
 	union {
 		__le32 reginfo1;
 		struct {
+			/* note: power unit is 1 dBm */
 			u8 antenna_max;
+			/* note: power unit is 0.5 dBm */
 			u8 max_tx_power;
 		} __packed;
 	} __packed;
@@ -2086,6 +2088,7 @@ struct wmi_channel_arg {
 	u32 min_power;
 	u32 max_power;
 	u32 max_reg_power;
+	/* note: power unit is 1 dBm */
 	u32 max_antenna_gain;
 	u32 reg_class_id;
 	enum wmi_phy_mode mode;
Signature · 1 line
2.33.0
author maintainer
#439 lore raw

PATCH 428/917 kernel/sched: Fix sched_fork() access an invalid sched_task_group

In reply to Greg Kroah-Hartman

From: Zhang Qiao <[email protected]>

[ Upstream commit 4ef0c5c6b5ba1f38f0ea1cedad0cad722f00c14a ]

There is a small race between copy_process() and sched_fork() where child->sched_task_group point to an already freed pointer.

parent doing fork() | someone moving the parent
| to another cgroup
-------------------------------+-------------------------------
copy_process()
+ dup_task_struct()<1>
parent move to another cgroup, and free the old cgroup. <2>
+ sched_fork()
+ __set_task_cpu()<3>
+ task_fork_fair()
+ sched_slice()<4>

In the worst case, this bug can lead to "use-after-free" and cause panic as shown above:

(1) parent copy its sched_task_group to child at <1>;

(2) someone move the parent to another cgroup and free the old

      cgroup at <2>;

(3) the sched_task_group and cfs_rq that belong to the old cgroup will be accessed at <3> and <4>, which cause a panic:

[] BUG: unable to handle kernel NULL pointer dereference at 0000000000000000 [] PGD 8000001fa0a86067 P4D 8000001fa0a86067 PUD 2029955067 PMD 0 [] Oops: 0000 [#1] SMP PTI
[] CPU: 7 PID: 648398 Comm: ebizzy Kdump: loaded Tainted: G OE --------- - - 4.18.0.x86_64+ #1
[] RIP: 0010:sched_slice+0x84/0xc0

[] Call Trace:

  []  task_fork_fair+0x81/0x120
  []  sched_fork+0x132/0x240
  []  copy_process.part.5+0x675/0x20e0
  []  ? __handle_mm_fault+0x63f/0x690
  []  _do_fork+0xcd/0x3b0
  []  do_syscall_64+0x5d/0x1d0
  []  entry_SYSCALL_64_after_hwframe+0x65/0xca

[] RIP: 0033:0x7f04418cd7e1

Between cgroup_can_fork() and cgroup_post_fork(), the cgroup membership and thus sched_task_group can't change. So update child's sched_task_group at sched_post_fork() and move task_fork() and __set_task_cpu() (where accees the sched_task_group) from sched_fork() to sched_post_fork().

---
 include/linux/sched/task.h |  3 ++-
 kernel/fork.c              |  2 +-
 kernel/sched/core.c        | 43 +++++++++++++++++++-------------------
 3 files changed, 25 insertions(+), 23 deletions(-)
include/linux/sched/task.h +2 -1
index ef02be869cf28..ba88a69874004 100644
@@ -54,7 +54,8 @@ extern asmlinkage void schedule_tail(struct task_struct *prev);
 extern void init_idle(struct task_struct *idle, int cpu);
 
 extern int sched_fork(unsigned long clone_flags, struct task_struct *p);
-extern void sched_post_fork(struct task_struct *p);
+extern void sched_post_fork(struct task_struct *p,
+			    struct kernel_clone_args *kargs);
 extern void sched_dead(struct task_struct *p);
 
 void __noreturn do_task_dead(void);
kernel/fork.c +1 -1
index 38681ad44c76b..0e4251dc54361 100644
@@ -2405,7 +2405,7 @@ static __latent_entropy struct task_struct *copy_process(
 	write_unlock_irq(&tasklist_lock);
 
 	proc_fork_connector(p);
-	sched_post_fork(p);
+	sched_post_fork(p, args);
 	cgroup_post_fork(p, args);
 	perf_event_fork(p);
 
kernel/sched/core.c +22 -21

62 lines not on this page · raw message

Signature · 1 line
2.33.0
author maintainer
#440 lore raw

PATCH 429/917 net: fealnx: fix build for UML

In reply to Greg Kroah-Hartman

From: Randy Dunlap <[email protected]>

[ Upstream commit cd2621d07d517473611b170c69beb6524c677740 ]

On i386, when builtin (not a loadable module), the fealnx driver inspects boot_cpu_data to see what CPU family it is running on, and then acts on that data. The "family" struct member (x86) does not exist when running on UML, so prevent that test and do the default action.

Prevents this build error on UML + i386:

../drivers/net/ethernet/fealnx.c: In function ‘netdev_open’:
../drivers/net/ethernet/fealnx.c:861:19: error: ‘struct cpuinfo_um’ has no member named ‘x86’

Fixes 68f5d3f3b654 ("um: add PCI over virtio emulation driver")
Cc Jeff Dike <[email protected]>
Cc Richard Weinberger <[email protected]>
Cc Anton Ivanov <[email protected]>
---
 drivers/net/ethernet/fealnx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/fealnx.c +1 -1
index 25c91b3c5fd30..819266d463b07 100644
@@ -857,7 +857,7 @@ static int netdev_open(struct net_device *dev)
 	np->bcrvalue |= 0x04;	/* big-endian */
 #endif
 
-#if defined(__i386__) && !defined(MODULE)
+#if defined(__i386__) && !defined(MODULE) && !defined(CONFIG_UML)
 	if (boot_cpu_data.x86 <= 4)
 		np->crvalue = 0xa00;
 	else
Signature · 9 lines
2.33.0




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

PATCH 430/917 net: intel: igc_ptp: fix build for UML

In reply to Greg Kroah-Hartman

From: Randy Dunlap <[email protected]>

[ Upstream commit 523994ba3ad1b7b55abe4a72e156897b5e2db825 ]

On a UML build, the igc_ptp driver uses CONFIG_X86_TSC for timestamp conversion. The function that is used is not available on UML builds, so have the function use the default system_counterval_t timestamp instead for UML builds.

Prevents this build error on UML:

../drivers/net/ethernet/intel/igc/igc_ptp.c: In function ?igc_device_tstamp_to_system?:
../drivers/net/ethernet/intel/igc/igc_ptp.c:777:9: error: implicit declaration of function ?convert_art_ns_to_tsc? [-Werror=implicit-function-declaration]
return convert_art_ns_to_tsc(tstamp);
../drivers/net/ethernet/intel/igc/igc_ptp.c:777:9: error: incompatible types when returning type ?int? but ?struct system_counterval_t? was expected
return convert_art_ns_to_tsc(tstamp);

Fixes 68f5d3f3b654 ("um: add PCI over virtio emulation driver")
Cc linux-um at lists.infradead.org
Cc Jeff Dike <[email protected]>
Cc Richard Weinberger <[email protected]>
Cc Anton Ivanov <[email protected]>
Cc Jesse Brandeburg <[email protected]>
Cc Tony Nguyen <[email protected]>
Cc intel-wired-lan at lists.osuosl.org
---
 drivers/net/ethernet/intel/igc/igc_ptp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/intel/igc/igc_ptp.c +1 -1
index 0f021909b430a..30568e3544cda 100644
@@ -773,7 +773,7 @@ static bool igc_is_crosststamp_supported(struct igc_adapter *adapter)
 
 static struct system_counterval_t igc_device_tstamp_to_system(u64 tstamp)
 {
-#if IS_ENABLED(CONFIG_X86_TSC)
+#if IS_ENABLED(CONFIG_X86_TSC) && !defined(CONFIG_UML)
 	return convert_art_ns_to_tsc(tstamp);
 #else
 	return (struct system_counterval_t) { };
Signature · 1 line
2.33.0
author maintainer
#442 lore raw

PATCH 431/917 net: tulip: winbond-840: fix build for UML

In reply to Greg Kroah-Hartman

From: Randy Dunlap <[email protected]>

[ Upstream commit a3d708925fcca1a2f7219bc9ce93e6341f85c1e0 ]

On i386, when builtin (not a loadable module), the winbond-840 driver inspects boot_cpu_data to see what CPU family it is running on, and then acts on that data. The "family" struct member (x86) does not exist when running on UML, so prevent that test and do the default action.

Prevents this build error on UML + i386:

../drivers/net/ethernet/dec/tulip/winbond-840.c: In function ‘init_registers’:
../drivers/net/ethernet/dec/tulip/winbond-840.c:882:19: error: ‘struct cpuinfo_um’ has no member named ‘x86’
if (boot_cpu_data.x86 <= 4) {

Fixes 68f5d3f3b654 ("um: add PCI over virtio emulation driver")
Cc Jeff Dike <[email protected]>
Cc Richard Weinberger <[email protected]>
Cc Anton Ivanov <[email protected]>
---
 drivers/net/ethernet/dec/tulip/winbond-840.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/dec/tulip/winbond-840.c +1 -1
index 85b99099c6b94..5babcf05bc2f1 100644
@@ -877,7 +877,7 @@ static void init_registers(struct net_device *dev)
 		8000	16 longwords		0200 2 longwords	2000 32 longwords
 		C000	32  longwords		0400 4 longwords */
 
-#if defined (__i386__) && !defined(MODULE)
+#if defined (__i386__) && !defined(MODULE) && !defined(CONFIG_UML)
 	/* When not a module we can work around broken '486 PCI boards. */
 	if (boot_cpu_data.x86 <= 4) {
 		i |= 0x4800;
Signature · 9 lines
2.33.0




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

PATCH 432/917 x86: Fix get_wchan() to support the ORC unwinder

In reply to Greg Kroah-Hartman

From: Qi Zheng <[email protected]>

[ Upstream commit bc9bbb81730ea667c31c5b284f95ee312bab466f ]

Currently, the kernel CONFIG_UNWINDER_ORC option is enabled by default on x86, but the implementation of get_wchan() is still based on the frame pointer unwinder, so the /proc/<pid>/wchan usually returned 0 regardless of whether the task <pid> is running.

Reimplement get_wchan() by calling stack_trace_save_tsk(), which is adapted to the ORC and frame pointer unwinders.

---
 arch/x86/kernel/process.c | 51 +++------------------------------------
 1 file changed, 3 insertions(+), 48 deletions(-)
arch/x86/kernel/process.c +3 -48
index f2f733bcb2b95..cd426c3283ee1 100644
@@ -945,58 +945,13 @@ unsigned long arch_randomize_brk(struct mm_struct *mm)
  */
 unsigned long get_wchan(struct task_struct *p)
 {
-	unsigned long start, bottom, top, sp, fp, ip, ret = 0;
-	int count = 0;
+	unsigned long entry = 0;
 
 	if (p == current || task_is_running(p))
 		return 0;
 
-	if (!try_get_task_stack(p))
-		return 0;
-
-	start = (unsigned long)task_stack_page(p);
-	if (!start)
-		goto out;
-
-	/*
-	 * Layout of the stack page:
-	 *
-	 * ----------- topmax = start + THREAD_SIZE - sizeof(unsigned long)
-	 * PADDING
-	 * ----------- top = topmax - TOP_OF_KERNEL_STACK_PADDING
-	 * stack
-	 * ----------- bottom = start
-	 *
-	 * The tasks stack pointer points at the location where the
-	 * framepointer is stored. The data on the stack is:
-	 * ... IP FP ... IP FP
-	 *
-	 * We need to read FP and IP, so we need to adjust the upper
-	 * bound by another unsigned long.
-	 */
-	top = start + THREAD_SIZE - TOP_OF_KERNEL_STACK_PADDING;
-	top -= 2 * sizeof(unsigned long);
-	bottom = start;
-
-	sp = READ_ONCE(p->thread.sp);
-	if (sp < bottom || sp > top)
-		goto out;
-
-	fp = READ_ONCE_NOCHECK(((struct inactive_task_frame *)sp)->bp);
-	do {
-		if (fp < bottom || fp > top)
-			goto out;
-		ip = READ_ONCE_NOCHECK(*(unsigned long *)(fp + sizeof(unsigned long)));
-		if (!in_sched_functions(ip)) {
-			ret = ip;
-			goto out;
-		}
-		fp = READ_ONCE_NOCHECK(*(unsigned long *)fp);
-	} while (count++ < 16 && !task_is_running(p));
-
-out:
-	put_task_stack(p);
-	return ret;
+	stack_trace_save_tsk(p, &entry, 1, 0);
+	return entry;
 }
 
 long do_arch_prctl_common(struct task_struct *task, int option,
Signature · 1 line
2.33.0
author maintainer
#444 lore raw

PATCH 433/917 tcp: switch orphan_count to bare per-cpu counters

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

[ Upstream commit 19757cebf0c5016a1f36f7fe9810a9f0b33c0832 ]

Use of percpu_counter structure to track count of orphaned sockets is causing problems on modern hosts with 256 cpus or more.

Stefan Bach reported a serious spinlock contention in real workloads, that I was able to reproduce with a netfilter rule dropping incoming FIN packets.

    53.56%  server  [kernel.kallsyms]      [k] queued_spin_lock_slowpath
            |
            ---queued_spin_lock_slowpath
               |
                --53.51%--_raw_spin_lock_irqsave
                          |
                           --53.51%--__percpu_counter_sum
                                     tcp_check_oom
                                     |
                                     |--39.03%--__tcp_close
                                     |          tcp_close
                                     |          inet_release
                                     |          inet6_release
                                     |          sock_close
                                     |          __fput
                                     |          ____fput
                                     |          task_work_run
                                     |          exit_to_usermode_loop
                                     |          do_syscall_64
                                     |          entry_SYSCALL_64_after_hwframe
                                     |          __GI___libc_close
                                     |
                                      --14.48%--tcp_out_of_resources
                                                tcp_write_timeout
                                                tcp_retransmit_timer
                                                tcp_write_timer_handler
                                                tcp_write_timer
                                                call_timer_fn
                                                expire_timers
                                                __run_timers
                                                run_timer_softirq
                                                __softirqentry_text_start

As explained in commit cf86a086a180 ("net/dst: use a smaller percpu_counter batch for dst entries accounting"), default batch size is too big for the default value of tcp_max_orphans (262144).

But even if we reduce batch sizes, there would still be cases where the estimated count of orphans is beyond the limit, and where tcp_too_many_orphans() has to call the expensive percpu_counter_sum_positive().

One solution is to use plain per-cpu counters, and have a timer to periodically refresh this cache.

Updating this cache every 100ms seems about right, tcp pressure state is not radically changing over shorter periods.

percpu_counter was nice 15 years ago while hosts had less than 16 cpus, not anymore by current standards.

v2: Fix the build issue for CONFIG_CRYPTO_DEV_CHELSIO_TLS=m,
reported by kernel test robot <[email protected]> Remove unused socket argument from tcp_too_many_orphans()

Fixes dd24c00191d5 ("net: Use a percpu_counter for orphan_count")
Cc Neal Cardwell <[email protected]>
---
 .../chelsio/inline_crypto/chtls/chtls_cm.c    |  2 +-
 .../chelsio/inline_crypto/chtls/chtls_cm.h    |  2 +-
 include/net/inet_connection_sock.h            |  2 +-
 include/net/sock.h                            |  2 +-
 include/net/tcp.h                             | 17 ++-------
 net/dccp/dccp.h                               |  2 +-
 net/dccp/proto.c                              | 14 ++-----
 net/ipv4/inet_connection_sock.c               |  4 +-
 net/ipv4/inet_hashtables.c                    |  2 +-
 net/ipv4/proc.c                               |  2 +-
 net/ipv4/tcp.c                                | 38 ++++++++++++++++---
 11 files changed, 49 insertions(+), 38 deletions(-)
drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.c +1 -1
index bcad69c480740..4af5561cbfc54 100644
@@ -870,7 +870,7 @@ static void do_abort_syn_rcv(struct sock *child, struct sock *parent)
 		 * created only after 3 way handshake is done.
 		 */
 		sock_orphan(child);
-		percpu_counter_inc((child)->sk_prot->orphan_count);
+		INC_ORPHAN_COUNT(child);
 		chtls_release_resources(child);
 		chtls_conn_done(child);
 	} else {
drivers/net/ethernet/chelsio/inline_crypto/chtls/chtls_cm.h +1 -1
index b1161bdeda4dc..f61ca657601ca 100644
@@ -95,7 +95,7 @@ struct deferred_skb_cb {
 #define WSCALE_OK(tp) ((tp)->rx_opt.wscale_ok)
 #define TSTAMP_OK(tp) ((tp)->rx_opt.tstamp_ok)
 #define SACK_OK(tp) ((tp)->rx_opt.sack_ok)
-#define INC_ORPHAN_COUNT(sk) percpu_counter_inc((sk)->sk_prot->orphan_count)
+#define INC_ORPHAN_COUNT(sk) this_cpu_inc(*(sk)->sk_prot->orphan_count)
 
 /* TLS SKB */
 #define skb_ulp_tls_inline(skb)      (ULP_SKB_CB(skb)->ulp.tls.ofld)
include/net/inet_connection_sock.h +1 -1
index b06c2d02ec84e..fa6a87246a7b8 100644
@@ -289,7 +289,7 @@ static inline void inet_csk_prepare_for_destroy_sock(struct sock *sk)
 {
 	/* The below has to be done to allow calling inet_csk_destroy_sock */
 	sock_set_flag(sk, SOCK_DEAD);
-	percpu_counter_inc(sk->sk_prot->orphan_count);
+	this_cpu_inc(*sk->sk_prot->orphan_count);
 }
 
 void inet_csk_destroy_sock(struct sock *sk);
include/net/sock.h +1 -1
index 463f390d90b3e..7b0c7f5aab676 100644
@@ -1237,7 +1237,7 @@ struct proto {
 	unsigned int		useroffset;	/* Usercopy region offset */
 	unsigned int		usersize;	/* Usercopy region size */
 
-	struct percpu_counter	*orphan_count;
+	unsigned int __percpu	*orphan_count;
 
 	struct request_sock_ops	*rsk_prot;
 	struct timewait_sock_ops *twsk_prot;
include/net/tcp.h +3 -14
index 60c384569e9cd..31d384c3778a1 100644
@@ -48,7 +48,9 @@
 
 extern struct inet_hashinfo tcp_hashinfo;
 
-extern struct percpu_counter tcp_orphan_count;
+DECLARE_PER_CPU(unsigned int, tcp_orphan_count);
+int tcp_orphan_count_sum(void);
+
 void tcp_time_wait(struct sock *sk, int state, int timeo);
 
 #define MAX_TCP_HEADER	L1_CACHE_ALIGN(128 + MAX_HEADER)
@@ -290,19 +292,6 @@ static inline bool tcp_out_of_memory(struct sock *sk)
 
 void sk_forced_mem_schedule(struct sock *sk, int size);
 
-static inline bool tcp_too_many_orphans(struct sock *sk, int shift)
-{
-	struct percpu_counter *ocp = sk->sk_prot->orphan_count;
-	int orphans = percpu_counter_read_positive(ocp);
-
-	if (orphans << shift > sysctl_tcp_max_orphans) {
-		orphans = percpu_counter_sum_positive(ocp);
-		if (orphans << shift > sysctl_tcp_max_orphans)
-			return true;
-	}
-	return false;
-}
-
 bool tcp_check_oom(struct sock *sk, int shift);
 
 
net/dccp/dccp.h +1 -1
index c5c1d2b8045e8..5183e627468d8 100644
@@ -48,7 +48,7 @@ extern bool dccp_debug;
 
 extern struct inet_hashinfo dccp_hashinfo;
 
-extern struct percpu_counter dccp_orphan_count;
+DECLARE_PER_CPU(unsigned int, dccp_orphan_count);
 
 void dccp_time_wait(struct sock *sk, int state, int timeo);
 
net/dccp/proto.c +4 -10
index abb5c596a8176..fc44dadc778bb 100644
@@ -42,8 +42,8 @@ DEFINE_SNMP_STAT(struct dccp_mib, dccp_statistics) __read_mostly;
 
 EXPORT_SYMBOL_GPL(dccp_statistics);
 
-struct percpu_counter dccp_orphan_count;
-EXPORT_SYMBOL_GPL(dccp_orphan_count);
+DEFINE_PER_CPU(unsigned int, dccp_orphan_count);
+EXPORT_PER_CPU_SYMBOL_GPL(dccp_orphan_count);
 
 struct inet_hashinfo dccp_hashinfo;
 EXPORT_SYMBOL_GPL(dccp_hashinfo);
@@ -1055,7 +1055,7 @@ adjudge_to_death:
 	bh_lock_sock(sk);
 	WARN_ON(sock_owned_by_user(sk));
 
-	percpu_counter_inc(sk->sk_prot->orphan_count);
+	this_cpu_inc(dccp_orphan_count);
 
 	/* Have we already been destroyed by a softirq or backlog? */
 	if (state != DCCP_CLOSED && sk->sk_state == DCCP_CLOSED)
@@ -1115,13 +1115,10 @@ static int __init dccp_init(void)
 
 	BUILD_BUG_ON(sizeof(struct dccp_skb_cb) >
 		     sizeof_field(struct sk_buff, cb));
-	rc = percpu_counter_init(&dccp_orphan_count, 0, GFP_KERNEL);
-	if (rc)
-		goto out_fail;
 	inet_hashinfo_init(&dccp_hashinfo);
 	rc = inet_hashinfo2_init_mod(&dccp_hashinfo);
 	if (rc)
-		goto out_free_percpu;
+		goto out_fail;
 	rc = -ENOBUFS;
 	dccp_hashinfo.bind_bucket_cachep =
 		kmem_cache_create("dccp_bind_bucket",
@@ -1226,8 +1223,6 @@ out_free_bind_bucket_cachep:
 	kmem_cache_destroy(dccp_hashinfo.bind_bucket_cachep);
 out_free_hashinfo2:
 	inet_hashinfo2_free_mod(&dccp_hashinfo);
-out_free_percpu:
-	percpu_counter_destroy(&dccp_orphan_count);
 out_fail:
 	dccp_hashinfo.bhash = NULL;
 	dccp_hashinfo.ehash = NULL;
@@ -1250,7 +1245,6 @@ static void __exit dccp_fini(void)
 	dccp_ackvec_exit();
 	dccp_sysctl_exit();
 	inet_hashinfo2_free_mod(&dccp_hashinfo);
-	percpu_counter_destroy(&dccp_orphan_count);
 }
 
 module_init(dccp_init);
net/ipv4/inet_connection_sock.c +2 -2
index f25d02ad4a8af..f7fea3a7c5e64 100644
@@ -1015,7 +1015,7 @@ void inet_csk_destroy_sock(struct sock *sk)
 
 	sk_refcnt_debug_release(sk);
 
-	percpu_counter_dec(sk->sk_prot->orphan_count);
+	this_cpu_dec(*sk->sk_prot->orphan_count);
 
 	sock_put(sk);
 }
@@ -1074,7 +1074,7 @@ static void inet_child_forget(struct sock *sk, struct request_sock *req,
 
 	sock_orphan(child);
 
-	percpu_counter_inc(sk->sk_prot->orphan_count);
+	this_cpu_inc(*sk->sk_prot->orphan_count);
 
 	if (sk->sk_protocol == IPPROTO_TCP && tcp_rsk(req)->tfo_listener) {
 		BUG_ON(rcu_access_pointer(tcp_sk(child)->fastopen_rsk) != req);
net/ipv4/inet_hashtables.c +1 -1
index bfb522e513461..75737267746f8 100644
@@ -598,7 +598,7 @@ bool inet_ehash_nolisten(struct sock *sk, struct sock *osk, bool *found_dup_sk)
 	if (ok) {
 		sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 	} else {
-		percpu_counter_inc(sk->sk_prot->orphan_count);
+		this_cpu_inc(*sk->sk_prot->orphan_count);
 		inet_sk_set_state(sk, TCP_CLOSE);
 		sock_set_flag(sk, SOCK_DEAD);
 		inet_csk_destroy_sock(sk);
net/ipv4/proc.c +1 -1
index b0d3a09dc84e7..f30273afb5399 100644
@@ -53,7 +53,7 @@ static int sockstat_seq_show(struct seq_file *seq, void *v)
 	struct net *net = seq->private;
 	int orphans, sockets;
 
-	orphans = percpu_counter_sum_positive(&tcp_orphan_count);
+	orphans = tcp_orphan_count_sum();
 	sockets = proto_sockets_allocated_sum_positive(&tcp_prot);
 
 	socket_seq_show(seq);
net/ipv4/tcp.c +33 -5
index f5c336f8b0c8e..d13050b68045b 100644
@@ -287,8 +287,8 @@ enum {
 	TCP_CMSG_TS = 2
 };
 
-struct percpu_counter tcp_orphan_count;
-EXPORT_SYMBOL_GPL(tcp_orphan_count);
+DEFINE_PER_CPU(unsigned int, tcp_orphan_count);
+EXPORT_PER_CPU_SYMBOL_GPL(tcp_orphan_count);
 
 long sysctl_tcp_mem[3] __read_mostly;
 EXPORT_SYMBOL(sysctl_tcp_mem);
@@ -2687,11 +2687,36 @@ void tcp_shutdown(struct sock *sk, int how)
 }
 EXPORT_SYMBOL(tcp_shutdown);
 
+int tcp_orphan_count_sum(void)
+{
+	int i, total = 0;
+
+	for_each_possible_cpu(i)
+		total += per_cpu(tcp_orphan_count, i);
+
+	return max(total, 0);
+}
+
+static int tcp_orphan_cache;
+static struct timer_list tcp_orphan_timer;
+#define TCP_ORPHAN_TIMER_PERIOD msecs_to_jiffies(100)
+
+static void tcp_orphan_update(struct timer_list *unused)
+{
+	WRITE_ONCE(tcp_orphan_cache, tcp_orphan_count_sum());
+	mod_timer(&tcp_orphan_timer, jiffies + TCP_ORPHAN_TIMER_PERIOD);
+}
+
+static bool tcp_too_many_orphans(int shift)
+{
+	return READ_ONCE(tcp_orphan_cache) << shift > sysctl_tcp_max_orphans;
+}
+
 bool tcp_check_oom(struct sock *sk, int shift)
 {
 	bool too_many_orphans, out_of_socket_memory;
 
-	too_many_orphans = tcp_too_many_orphans(sk, shift);
+	too_many_orphans = tcp_too_many_orphans(shift);
 	out_of_socket_memory = tcp_out_of_memory(sk);
 
 	if (too_many_orphans)
@@ -2800,7 +2825,7 @@ adjudge_to_death:
 	/* remove backlog if any, without releasing ownership. */
 	__release_sock(sk);
 
-	percpu_counter_inc(sk->sk_prot->orphan_count);
+	this_cpu_inc(tcp_orphan_count);
 
 	/* Have we already been destroyed by a softirq or backlog? */
 	if (state != TCP_CLOSE && sk->sk_state == TCP_CLOSE)
@@ -4502,7 +4527,10 @@ void __init tcp_init(void)
 		     sizeof_field(struct sk_buff, cb));
 
 	percpu_counter_init(&tcp_sockets_allocated, 0, GFP_KERNEL);
-	percpu_counter_init(&tcp_orphan_count, 0, GFP_KERNEL);
+
+	timer_setup(&tcp_orphan_timer, tcp_orphan_update, TIMER_DEFERRABLE);
+	mod_timer(&tcp_orphan_timer, jiffies + TCP_ORPHAN_TIMER_PERIOD);
+
 	inet_hashinfo_init(&tcp_hashinfo);
 	inet_hashinfo2_init(&tcp_hashinfo, "tcp_listen_portaddr_hash",
 			    thash_entries, 21,  /* one slot per 2 MB*/
Signature · 1 line
2.33.0
author maintainer
#445 lore raw

PATCH 434/917 crypto: octeontx2 - set assoclen in aead_do_fallback()

In reply to Greg Kroah-Hartman

From: Ovidiu Panait <[email protected]>

[ Upstream commit 06f6e365e2ecf799c249bb464aa9d5f055e88b56 ]

Currently, in case of aead fallback, no associated data info is set in the fallback request. To fix this, call aead_request_set_ad() to pass the assoclen.

Fixes 6f03f0e8b6c8 ("crypto: octeontx2 - register with linux crypto framework")
---
 drivers/crypto/marvell/octeontx2/otx2_cptvf_algs.c | 1 +
 1 file changed, 1 insertion(+)
drivers/crypto/marvell/octeontx2/otx2_cptvf_algs.c +1 -0
index a72723455df72..877a948469bd1 100644
@@ -1274,6 +1274,7 @@ static int aead_do_fallback(struct aead_request *req, bool is_enc)
 					  req->base.complete, req->base.data);
 		aead_request_set_crypt(&rctx->fbk_req, req->src,
 				       req->dst, req->cryptlen, req->iv);
+		aead_request_set_ad(&rctx->fbk_req, req->assoclen);
 		ret = is_enc ? crypto_aead_encrypt(&rctx->fbk_req) :
 			       crypto_aead_decrypt(&rctx->fbk_req);
 	} else {
Signature · 1 line
2.33.0
author maintainer
#446 lore raw

PATCH 435/917 thermal/core: fix a UAF bug in __thermal_cooling_device_register()

In reply to Greg Kroah-Hartman

From: Ziyang Xuan <[email protected]>

[ Upstream commit 0a5c26712f963f0500161a23e0ffff8d29f742ab ]

When device_register() return failed, program will goto out_kfree_type to release 'cdev->device' by put_device(). That will call thermal_release() to free 'cdev'. But the follow-up processes access 'cdev' continually. That trggers the UAF bug.

====================================================================

BUG: KASAN: use-after-free in __thermal_cooling_device_register+0x75b/0xa90
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1.1 04/01/2014
Call Trace:
 dump_stack_lvl+0xe2/0x152
 print_address_description.constprop.0+0x21/0x140
 ? __thermal_cooling_device_register+0x75b/0xa90
 kasan_report.cold+0x7f/0x11b
 ? __thermal_cooling_device_register+0x75b/0xa90
 __thermal_cooling_device_register+0x75b/0xa90
 ? memset+0x20/0x40
 ? __sanitizer_cov_trace_pc+0x1d/0x50
 ? __devres_alloc_node+0x130/0x180
 devm_thermal_of_cooling_device_register+0x67/0xf0
 max6650_probe.cold+0x557/0x6aa

......

Freed by task 258:
 kasan_save_stack+0x1b/0x40
 kasan_set_track+0x1c/0x30
 kasan_set_free_info+0x20/0x30
 __kasan_slab_free+0x109/0x140
 kfree+0x117/0x4c0
 thermal_release+0xa0/0x110
 device_release+0xa7/0x240
 kobject_put+0x1ce/0x540
 put_device+0x20/0x30
 __thermal_cooling_device_register+0x731/0xa90
 devm_thermal_of_cooling_device_register+0x67/0xf0
 max6650_probe.cold+0x557/0x6aa [max6650]

Do not use 'cdev' again after put_device() to fix the problem like doing in thermal_zone_device_register().

[dlezcano]: as requested by Rafael, change the affectation into two statements.

Fixes 584837618100 ("thermal/drivers/core: Use a char pointer for the cooling device name")
---
 drivers/thermal/thermal_core.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/thermal/thermal_core.c +4 -2
index d094ebbde0ed7..30134f49b037a 100644
@@ -887,7 +887,7 @@ __thermal_cooling_device_register(struct device_node *np,
 {
 	struct thermal_cooling_device *cdev;
 	struct thermal_zone_device *pos = NULL;
-	int ret;
+	int id, ret;
 
 	if (!ops || !ops->get_max_state || !ops->get_cur_state ||
 	    !ops->set_cur_state)
@@ -901,6 +901,7 @@ __thermal_cooling_device_register(struct device_node *np,
 	if (ret < 0)
 		goto out_kfree_cdev;
 	cdev->id = ret;
+	id = ret;
 
 	ret = dev_set_name(&cdev->device, "cooling_device%d", cdev->id);
 	if (ret)
@@ -944,8 +945,9 @@ __thermal_cooling_device_register(struct device_node *np,
 out_kfree_type:
 	kfree(cdev->type);
 	put_device(&cdev->device);
+	cdev = NULL;
 out_ida_remove:
-	ida_simple_remove(&thermal_cdev_ida, cdev->id);
+	ida_simple_remove(&thermal_cdev_ida, id);
 out_kfree_cdev:
 	kfree(cdev);
 	return ERR_PTR(ret);
Signature · 1 line
2.33.0
author maintainer
#447 lore raw

PATCH 436/917 drm/msm/dsi: do not enable irq handler before powering up the host

In reply to Greg Kroah-Hartman

From: Dmitry Baryshkov <[email protected]>

[ Upstream commit bf94ec093d05e3ed3142d9291b876eeb9997ba5c ]

The DSI host might be left in some state by the bootloader. If this state generates an IRQ, it might hang the system by holding the interrupt line before the driver sets up the DSI host to the known state.

Move the request_irq into msm_dsi_host_init and pass IRQF_NO_AUTOEN to it. Call enable/disable_irq after msm_dsi_host_power_on/_off() functions, so that we can be sure that the interrupt is delivered when the host is in the known state.

It is not possible to defer the interrupt enablement to a later point, because drm_panel_prepare might need to communicate with the panel over the DSI link and that requires working interrupt.

---
 drivers/gpu/drm/msm/dsi/dsi.h         |  2 ++
 drivers/gpu/drm/msm/dsi/dsi_host.c    | 48 +++++++++++++++++----------
 drivers/gpu/drm/msm/dsi/dsi_manager.c | 16 +++++++++
 3 files changed, 49 insertions(+), 17 deletions(-)
drivers/gpu/drm/msm/dsi/dsi.h +2 -0
index b50db91cb8a7e..569c8ff062ba4 100644
@@ -107,6 +107,8 @@ void msm_dsi_host_cmd_xfer_commit(struct mipi_dsi_host *host,
 					u32 dma_base, u32 len);
 int msm_dsi_host_enable(struct mipi_dsi_host *host);
 int msm_dsi_host_disable(struct mipi_dsi_host *host);
+void msm_dsi_host_enable_irq(struct mipi_dsi_host *host);
+void msm_dsi_host_disable_irq(struct mipi_dsi_host *host);
 int msm_dsi_host_power_on(struct mipi_dsi_host *host,
 			struct msm_dsi_phy_shared_timings *phy_shared_timings,
 			bool is_bonded_dsi, struct msm_dsi_phy *phy);
drivers/gpu/drm/msm/dsi/dsi_host.c +31 -17

68 lines not on this page · raw message

drivers/gpu/drm/msm/dsi/dsi_manager.c +16 -0
index c41d39f5b7cf4..fb4ccffdcfe13 100644
@@ -377,6 +377,14 @@ static void dsi_mgr_bridge_pre_enable(struct drm_bridge *bridge)
 		}
 	}
 
+	/*
+	 * Enable before preparing the panel, disable after unpreparing, so
+	 * that the panel can communicate over the DSI link.
+	 */
+	msm_dsi_host_enable_irq(host);
+	if (is_bonded_dsi && msm_dsi1)
+		msm_dsi_host_enable_irq(msm_dsi1->host);
+
 	/* Always call panel functions once, because even for dual panels,
 	 * there is only one drm_panel instance.
 	 */
@@ -411,6 +419,10 @@ host_en_fail:
 	if (panel)
 		drm_panel_unprepare(panel);
 panel_prep_fail:
+	msm_dsi_host_disable_irq(host);
+	if (is_bonded_dsi && msm_dsi1)
+		msm_dsi_host_disable_irq(msm_dsi1->host);
+
 	if (is_bonded_dsi && msm_dsi1)
 		msm_dsi_host_power_off(msm_dsi1->host);
 host1_on_fail:
@@ -523,6 +535,10 @@ static void dsi_mgr_bridge_post_disable(struct drm_bridge *bridge)
 								id, ret);
 	}
 
+	msm_dsi_host_disable_irq(host);
+	if (is_bonded_dsi && msm_dsi1)
+		msm_dsi_host_disable_irq(msm_dsi1->host);
+
 	/* Save PHY status if it is a clock source */
 	msm_dsi_phy_pll_save_state(msm_dsi->phy);
 
Signature · 1 line
2.33.0
author maintainer
#448 lore raw

PATCH 437/917 drm/msm: Fix potential Oops in a6xx_gmu_rpmh_init()

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

[ Upstream commit 3d91e50ff58364f6572ad268b508175d27800e51 ]

There are two problems here:
1) The "seqptr" is used uninitalized when we free it at the end.
2) The a6xx_gmu_get_mmio() function returns error pointers. It never returns true.

Fixes 64245fc55172 ("drm/msm/a6xx: use AOP-initialized PDC for a650")
Fixes f8fc924e088e ("drm/msm/a6xx: Fix PDC register overlap")
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/gpu/drm/msm/adreno/a6xx_gmu.c +3 -3
index 8b73f70766a47..4347a104755a9 100644
@@ -516,11 +516,11 @@ static void a6xx_gmu_rpmh_init(struct a6xx_gmu *gmu)
 	struct adreno_gpu *adreno_gpu = &a6xx_gpu->base;
 	struct platform_device *pdev = to_platform_device(gmu->dev);
 	void __iomem *pdcptr = a6xx_gmu_get_mmio(pdev, "gmu_pdc");
-	void __iomem *seqptr;
+	void __iomem *seqptr = NULL;
 	uint32_t pdc_address_offset;
 	bool pdc_in_aop = false;
 
-	if (!pdcptr)
+	if (IS_ERR(pdcptr))
 		goto err;
 
 	if (adreno_is_a650(adreno_gpu) || adreno_is_a660_family(adreno_gpu))
@@ -532,7 +532,7 @@ static void a6xx_gmu_rpmh_init(struct a6xx_gmu *gmu)
 
 	if (!pdc_in_aop) {
 		seqptr = a6xx_gmu_get_mmio(pdev, "gmu_pdc_seq");
-		if (!seqptr)
+		if (IS_ERR(seqptr))
 			goto err;
 	}
 
Signature · 1 line
2.33.0
author maintainer
#449 lore raw

PATCH 438/917 drm/msm: potential error pointer dereference in init()

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

[ Upstream commit b6816441a14bbe356ba8590de79cfea2de6a085c ]

The msm_iommu_new() returns error pointers on failure so check for that to avoid an Oops.

---
 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c +4 -0
index ae48f41821cfe..ad247c06e198f 100644
@@ -908,6 +908,10 @@ static int _dpu_kms_mmu_init(struct dpu_kms *dpu_kms)
 		return 0;
 
 	mmu = msm_iommu_new(dpu_kms->dev->dev, domain);
+	if (IS_ERR(mmu)) {
+		iommu_domain_free(domain);
+		return PTR_ERR(mmu);
+	}
 	aspace = msm_gem_address_space_create(mmu, "dpu1",
 		0x1000, 0x100000000 - 0x1000);
 
Signature · 1 line
2.33.0
author maintainer
#450 lore raw

PATCH 439/917 drm/msm: unlock on error in get_sched_entity()

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

[ Upstream commit 7425e8167507fe512d8ac0825acda4aebf0a7ca0 ]

Add a missing unlock on the error path if drm_sched_entity_init() fails.

Fixes 68002469e571 ("drm/msm: One sched entity per process per priority")
---
 drivers/gpu/drm/msm/msm_submitqueue.c | 1 +
 1 file changed, 1 insertion(+)
drivers/gpu/drm/msm/msm_submitqueue.c +1 -0
index b8621c6e05546..7cb158bcbcf67 100644
@@ -101,6 +101,7 @@ get_sched_entity(struct msm_file_private *ctx, struct msm_ringbuffer *ring,
 
 		ret = drm_sched_entity_init(entity, sched_prio, &sched, 1, NULL);
 		if (ret) {
+			mutex_unlock(&entity_lock);
 			kfree(entity);
 			return ERR_PTR(ret);
 		}
Signature · 1 line
2.33.0
author maintainer
#451 lore raw

PATCH 440/917 drm/msm: fix potential NULL dereference in cleanup

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

[ Upstream commit 027d052a36e56789a2134772bacb4fd0860f03a3 ]

The "msm_obj->node" list needs to be initialized earlier so that the list_del() in msm_gem_free_object() doesn't experience a NULL pointer dereference.

---
 drivers/gpu/drm/msm/msm_gem.c | 1 +
 1 file changed, 1 insertion(+)
drivers/gpu/drm/msm/msm_gem.c +1 -0
index 22308a1b66fc3..fd398a4eaf46e 100644
@@ -1132,6 +1132,7 @@ static int msm_gem_new_impl(struct drm_device *dev,
 	msm_obj->flags = flags;
 	msm_obj->madv = MSM_MADV_WILLNEED;
 
+	INIT_LIST_HEAD(&msm_obj->node);
 	INIT_LIST_HEAD(&msm_obj->vmas);
 
 	*obj = &msm_obj->base;
Signature · 1 line
2.33.0
author maintainer
#452 lore raw

PATCH 441/917 drm/msm: uninitialized variable in msm_gem_import()

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

[ Upstream commit 2203bd0e5c12ffc53ffdd4fbd7b12d6ba27e0424 ]

The msm_gem_new_impl() function cleans up after itself so there is no need to call drm_gem_object_put(). Conceptually, it does not make sense to call a kref_put() function until after the reference counting has been initialized which happens immediately after this call in the drm_gem_(private_)object_init() functions.

In the msm_gem_import() function the "obj" pointer is uninitialized, so it will lead to a crash.

---
 drivers/gpu/drm/msm/msm_gem.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/msm/msm_gem.c +2 -2
index fd398a4eaf46e..bd6ec04f345e1 100644
@@ -1167,7 +1167,7 @@ struct drm_gem_object *msm_gem_new(struct drm_device *dev, uint32_t size, uint32
 
 	ret = msm_gem_new_impl(dev, size, flags, &obj);
 	if (ret)
-		goto fail;
+		return ERR_PTR(ret);
 
 	msm_obj = to_msm_bo(obj);
 
@@ -1251,7 +1251,7 @@ struct drm_gem_object *msm_gem_import(struct drm_device *dev,
 
 	ret = msm_gem_new_impl(dev, size, MSM_BO_WC, &obj);
 	if (ret)
-		goto fail;
+		return ERR_PTR(ret);
 
 	drm_gem_private_object_init(dev, obj, size);
 
Signature · 1 line
2.33.0
author maintainer
#453 lore raw

PATCH 442/917 net: stream: dont purge sk_error_queue in sk_stream_kill_queues()

In reply to Greg Kroah-Hartman

From: Jakub Kicinski <[email protected]>

[ Upstream commit 24bcbe1cc69fa52dc4f7b5b2456678ed464724d8 ]

sk_stream_kill_queues() can be called on close when there are still outstanding skbs to transmit. Those skbs may try to queue notifications to the error queue (e.g. timestamps). If sk_stream_kill_queues() purges the queue without taking its lock the queue may get corrupted, and skbs leaked.

This shows up as a warning about an rmem leak:

WARNING: CPU: 24 PID: 0 at net/ipv4/af_inet.c:154 inet_sock_destruct+0x...

The leak is always a multiple of 0x300 bytes (the value is in %rax on my builds, so RAX: 0000000000000300). 0x300 is truesize of an empty sk_buff. Indeed if we dump the socket state at the time of the warning the sk_error_queue is often (but not always) corrupted. The ->next pointer points back at the list head, but not the ->prev pointer. Indeed we can find the leaked skb by scanning the kernel memory for something that looks like an skb with ->sk = socket in question, and ->truesize = 0x300. The contents of ->cb[] of the skb confirms the suspicion that it is indeed a timestamp notification (as generated in __skb_complete_tx_timestamp()).

Removing purging of sk_error_queue should be okay, since inet_sock_destruct() does it again once all socket refs are gone. Eric suggests this may cause sockets that go thru disconnect() to maintain notifications from the previous incarnations of the socket, but that should be okay since the race was there anyway, and disconnect() is not exactly dependable.

Thanks to Jonathan Lemon and Omar Sandoval for help at various stages of tracing the issue.

Fixes cb9eff097831 ("net: new user space API for time stamping of incoming and outgoing packets")
---
 net/core/stream.c | 3 ---
 1 file changed, 3 deletions(-)
net/core/stream.c +0 -3
index 4f1d4aa5fb38d..a166a32b411fa 100644
@@ -195,9 +195,6 @@ void sk_stream_kill_queues(struct sock *sk)
 	/* First the read buffer. */
 	__skb_queue_purge(&sk->sk_receive_queue);
 
-	/* Next, the error queue. */
-	__skb_queue_purge(&sk->sk_error_queue);
-
 	/* Next, the write queue. */
 	WARN_ON(!skb_queue_empty(&sk->sk_write_queue));
 
Signature · 1 line
2.33.0
author maintainer
#454 lore raw

PATCH 443/917 thermal/drivers/qcom/lmh: make QCOM_LMH depends on QCOM_SCM

In reply to Greg Kroah-Hartman

From: Jackie Liu <[email protected]>

[ Upstream commit 9e5a4fb8423081d0efbf165c71c7f4abdf5f918c ]

Without QCOM_SCM, build failed, avoid like below:

aarch64-linux-gnu-ld: Unexpected GOT/PLT entries detected!
aarch64-linux-gnu-ld: Unexpected run-time procedure linkages detected!
aarch64-linux-gnu-ld: drivers/thermal/qcom/lmh.o: in function `lmh_probe':
/data/arm/workspace/kernel-build/linux/build/../drivers/thermal/qcom/lmh.c:141: undefined reference to `qcom_scm_lmh_dcvsh_available'
aarch64-linux-gnu-ld: /data/arm/workspace/kernel-build/linux/build/../drivers/thermal/qcom/lmh.c:144: undefined reference to `qcom_scm_lmh_dcvsh'
aarch64-linux-gnu-ld: /data/arm/workspace/kernel-build/linux/build/../drivers/thermal/qcom/lmh.c:149: undefined reference to `qcom_scm_lmh_dcvsh'
aarch64-linux-gnu-ld: /data/arm/workspace/kernel-build/linux/build/../drivers/thermal/qcom/lmh.c:154: undefined reference to `qcom_scm_lmh_dcvsh'
aarch64-linux-gnu-ld: /data/arm/workspace/kernel-build/linux/build/../drivers/thermal/qcom/lmh.c:159: undefined reference to `qcom_scm_lmh_dcvsh'
aarch64-linux-gnu-ld: /data/arm/workspace/kernel-build/linux/build/../drivers/thermal/qcom/lmh.c:166: undefined reference to `qcom_scm_lmh_profile_change'
aarch64-linux-gnu-ld: /data/arm/workspace/kernel-build/linux/build/../drivers/thermal/qcom/lmh.c:173: undefined reference to `qcom_scm_lmh_dcvsh'
aarch64-linux-gnu-ld: /data/arm/workspace/kernel-build/linux/build/../drivers/thermal/qcom/lmh.c:180: undefined reference to `qcom_scm_lmh_dcvsh'
aarch64-linux-gnu-ld: /data/arm/workspace/kernel-build/linux/build/../drivers/thermal/qcom/lmh.c:187: undefined reference to `qcom_scm_lmh_dcvsh' make[1]: *** [/data/arm/workspace/kernel-build/linux/Makefile:1183: vmlinux] Error 1
make[1]: Leaving directory '/data/arm/workspace/kernel-build/linux/build'
make: *** [Makefile:219: __sub-make] Error 2
make: Leaving directory '/data/arm/workspace/kernel-build/linux'

Fixes 53bca371cdf7 ("thermal/drivers/qcom: Add support for LMh driver")
---
 drivers/thermal/qcom/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/thermal/qcom/Kconfig +1 -1
index 7d942f71e5328..bfd889422dd32 100644
@@ -34,7 +34,7 @@ config QCOM_SPMI_TEMP_ALARM
 
 config QCOM_LMH
 	tristate "Qualcomm Limits Management Hardware"
-	depends on ARCH_QCOM
+	depends on ARCH_QCOM && QCOM_SCM
 	help
 	  This enables initialization of Qualcomm limits management
 	  hardware(LMh). LMh allows for hardware-enforced mitigation for cpus based on
Signature · 1 line
2.33.0
author maintainer
#455 lore raw

PATCH 444/917 mailbox: Remove WARN_ON for async_cb.cb in cmdq_exec_done

In reply to Greg Kroah-Hartman

From: jason-jh.lin <[email protected]>

[ Upstream commit ce1537fe288469bf68ee0aabdb860a790b4755ef ]

Because mtk_drm_crtc_update_config is not using cmdq_pkt_flush_async, it won't have pkt->async_cb.cb anymore.

So remove the WARN_ON check of pkt->async_cb.cb at cmdq_exec_done.

Fixes 1b6b0ce2240e ("mailbox: mtk-cmdq: Use mailbox rx_callback")
---
 drivers/mailbox/mtk-cmdq-mailbox.c | 1 -
 1 file changed, 1 deletion(-)
drivers/mailbox/mtk-cmdq-mailbox.c +0 -1
index 64175a893312e..c591dab9d5a48 100644
@@ -195,7 +195,6 @@ static void cmdq_task_exec_done(struct cmdq_task *task, int sta)
 	struct cmdq_task_cb *cb = &task->pkt->async_cb;
 	struct cmdq_cb_data data;
 
-	WARN_ON(cb->cb == (cmdq_async_flush_cb)NULL);
 	data.sta = sta;
 	data.data = cb->data;
 	data.pkt = task->pkt;
Signature · 1 line
2.33.0
author maintainer
#456 lore raw

PATCH 445/917 media: ivtv: fix build for UML

In reply to Greg Kroah-Hartman

From: Randy Dunlap <[email protected]>

[ Upstream commit 6cb67bea945bdf0ad40e633cd2d9fbeb0855675b ]

Prevent the use of page table macros and types from 2 conflicting places. This fixes multiple build errors and warnings, e.g.:

../arch/x86/include/asm/pgtable_64_types.h:21:34: error: conflicting types for ‘pte_t’

 typedef struct { pteval_t pte; } pte_t;
                                  ^~~~~
In file included from ../include/linux/mm_types_task.h:16:0,
                 from ../include/linux/mm_types.h:5,
                 from ../include/linux/buildid.h:5,
                 from ../include/linux/module.h:14,
                 from ../drivers/media/pci/ivtv/ivtv-driver.h:40,
                 from ../drivers/media/pci/ivtv/ivtvfb.c:29:

../arch/um/include/asm/page.h:57:39: note: previous declaration of ‘pte_t’ was here
typedef struct { unsigned long pte; } pte_t;

../arch/x86/include/asm/pgtable_types.h:284:43: error: expected ‘)’ before ‘prot’
static inline pgprot_t pgprot_nx(pgprot_t prot)

                                           ^
../include/linux/pgtable.h:914:26: note: in definition of macro ‘pgprot_nx’
 #define pgprot_nx(prot) (prot)
                          ^~~~
In file included from ../arch/x86/include/asm/memtype.h:6:0,
                 from ../drivers/media/pci/ivtv/ivtvfb.c:40:

../arch/x86/include/asm/pgtable_types.h:288:0: warning: "pgprot_nx" redefined
#define pgprot_nx pgprot_nx

../arch/x86/include/asm/page_types.h:11:0: warning: "PAGE_SIZE" redefined
#define PAGE_SIZE (_AC(1,UL) << PAGE_SHIFT)

In file included from ../include/linux/mm_types_task.h:16:0,

                 from ../include/linux/mm_types.h:5,
                 from ../include/linux/buildid.h:5,
                 from ../include/linux/module.h:14,
                 from ../drivers/media/pci/ivtv/ivtv-driver.h:40,
                 from ../drivers/media/pci/ivtv/ivtvfb.c:29:

../arch/um/include/asm/page.h:14:0: note: this is the location of the previous definition
#define PAGE_SIZE (_AC(1, UL) << PAGE_SHIFT)

Fixes 68f5d3f3b654 ("um: add PCI over virtio emulation driver")
Cc Johannes Berg <[email protected]>
Cc Andy Walls <[email protected]>
Cc Richard Weinberger <[email protected]>
---
 drivers/media/pci/ivtv/ivtvfb.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/media/pci/ivtv/ivtvfb.c +2 -2
index e2d56dca5be40..5ad03b2a50bdb 100644
@@ -36,7 +36,7 @@
 #include <linux/fb.h>
 #include <linux/ivtvfb.h>
 
-#ifdef CONFIG_X86_64
+#if defined(CONFIG_X86_64) && !defined(CONFIG_UML)
 #include <asm/memtype.h>
 #endif
 
@@ -1157,7 +1157,7 @@ static int ivtvfb_init_card(struct ivtv *itv)
 {
 	int rc;
 
-#ifdef CONFIG_X86_64
+#if defined(CONFIG_X86_64) && !defined(CONFIG_UML)
 	if (pat_enabled()) {
 		if (ivtvfb_force_pat) {
 			pr_info("PAT is enabled. Write-combined framebuffer caching will be disabled.\n");
Signature · 9 lines
2.33.0




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

PATCH 446/917 media: ir_toy: assignment to be16 should be of correct type

In reply to Greg Kroah-Hartman

From: Sean Young <[email protected]>

[ Upstream commit febfe985fc2ea052a363f6525ff624b8efd5273c ]

commit f0c15b360fb6 ("media: ir_toy: prevent device from hanging during transmit") removed a cpu_to_be16() cast, which causes a sparse warning.

Fixes f0c15b360fb6 ("media: ir_toy: prevent device from hanging during transmit")
---
 drivers/media/rc/ir_toy.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/media/rc/ir_toy.c +1 -1
index 48d52baec1a1c..1aa7989e756cc 100644
@@ -310,7 +310,7 @@ static int irtoy_tx(struct rc_dev *rc, uint *txbuf, uint count)
 		buf[i] = cpu_to_be16(v);
 	}
 
-	buf[count] = 0xffff;
+	buf[count] = cpu_to_be16(0xffff);
 
 	irtoy->tx_buf = buf;
 	irtoy->tx_len = size;
Signature · 1 line
2.33.0
author maintainer
#458 lore raw

PATCH 447/917 mmc: mxs-mmc: disable regulator on error and in the remove function

In reply to Greg Kroah-Hartman

From: Christophe JAILLET <[email protected]>

[ Upstream commit ce5f6c2c9b0fcb4094f8e162cfd37fb4294204f7 ]

The 'reg_vmmc' regulator is enabled in the probe. It is never disabled. Neither in the error handling path of the probe nor in the remove function.

Register a devm_action to disable it when needed.

---
 drivers/mmc/host/mxs-mmc.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
drivers/mmc/host/mxs-mmc.c +10 -0
index 947581de78601..8c3655d3be961 100644
@@ -552,6 +552,11 @@ static const struct of_device_id mxs_mmc_dt_ids[] = {
 };
 MODULE_DEVICE_TABLE(of, mxs_mmc_dt_ids);
 
+static void mxs_mmc_regulator_disable(void *regulator)
+{
+	regulator_disable(regulator);
+}
+
 static int mxs_mmc_probe(struct platform_device *pdev)
 {
 	struct device_node *np = pdev->dev.of_node;
@@ -591,6 +596,11 @@ static int mxs_mmc_probe(struct platform_device *pdev)
 				"Failed to enable vmmc regulator: %d\n", ret);
 			goto out_mmc_free;
 		}
+
+		ret = devm_add_action_or_reset(&pdev->dev, mxs_mmc_regulator_disable,
+					       reg_vmmc);
+		if (ret)
+			goto out_mmc_free;
 	}
 
 	ssp->clk = devm_clk_get(&pdev->dev, NULL);
Signature · 1 line
2.33.0
author maintainer
#459 lore raw

PATCH 448/917 io-wq: Remove duplicate code in io_workqueue_create()

In reply to Greg Kroah-Hartman

From: Bixuan Cui <[email protected]>

[ Upstream commit 71e1cef2d794338cc7b979d4c6144e1dc12718b5 ]

While task_work_add() in io_workqueue_create() is true, then duplicate code is executed:

  -> clear_bit_unlock(0, &worker->create_state);
  -> io_worker_release(worker);
  -> atomic_dec(&acct->nr_running);
  -> io_worker_ref_put(wq);
  -> return false;

  -> clear_bit_unlock(0, &worker->create_state); // back to io_workqueue_create()
  -> io_worker_release(worker);
  -> kfree(worker);

The io_worker_release() and clear_bit_unlock() are executed twice.

Fixes 3146cba99aa2 ("io-wq: make worker creation resilient against signals")
---
 fs/io-wq.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)
fs/io-wq.c +4 -5
index 203c0e2a5dfae..5d189b24a8d4b 100644
@@ -359,8 +359,10 @@ static bool io_queue_worker_create(struct io_worker *worker,
 
 	init_task_work(&worker->create_work, func);
 	worker->create_index = acct->index;
-	if (!task_work_add(wq->task, &worker->create_work, TWA_SIGNAL))
+	if (!task_work_add(wq->task, &worker->create_work, TWA_SIGNAL)) {
+		clear_bit_unlock(0, &worker->create_state);
 		return true;
+	}
 	clear_bit_unlock(0, &worker->create_state);
 fail_release:
 	io_worker_release(worker);
@@ -765,11 +767,8 @@ static void io_workqueue_create(struct work_struct *work)
 	struct io_worker *worker = container_of(work, struct io_worker, work);
 	struct io_wqe_acct *acct = io_wqe_get_acct(worker);
 
-	if (!io_queue_worker_create(worker, acct, create_worker_cont)) {
-		clear_bit_unlock(0, &worker->create_state);
-		io_worker_release(worker);
+	if (!io_queue_worker_create(worker, acct, create_worker_cont))
 		kfree(worker);
-	}
 }
 
 static bool create_io_worker(struct io_wq *wq, struct io_wqe *wqe, int index)
Signature · 1 line
2.33.0
author maintainer
#460 lore raw

PATCH 449/917 block: ataflop: fix breakage introduced at blk-mq refactoring

In reply to Greg Kroah-Hartman

From: Michael Schmitz <[email protected]>

[ Upstream commit 86d46fdaa12ae5befc16b8d73fc85a3ca0399ea6 ]

Refactoring of the Atari floppy driver when converting to blk-mq has broken the state machine in not-so-subtle ways:

finish_fdc() must be called when operations on the floppy device have completed. This is crucial in order to relase the ST-DMA lock, which protects against concurrent access to the ST-DMA controller by other drivers (some DMA related, most just related to device register access - broken beyond compare, I know).

When rewriting the driver's old do_request() function, the fact that finish_fdc() was called only when all queued requests had completed appears to have been overlooked. Instead, the new request function calls finish_fdc() immediately after the last request has been queued. finish_fdc() executes a dummy seek after most requests, and this overwrites the state machine's interrupt hander that was set up to wait for completion of the read/write request just prior. To make matters worse, finish_fdc() is called before device interrupts are re-enabled, making certain that the read/write interupt is missed.

Shifting the finish_fdc() call into the read/write request completion handler ensures the driver waits for the request to actually complete. With a queue depth of 2, we won't see long request sequences, so calling finish_fdc() unconditionally just adds a little overhead for the dummy seeks, and keeps the code simple.

While we're at it, kill ataflop_commit_rqs() which does nothing but run finish_fdc() unconditionally, again likely wiping out an in-flight request.

---
 drivers/block/ataflop.c | 18 +++---------------
 1 file changed, 3 insertions(+), 15 deletions(-)
drivers/block/ataflop.c +3 -15
index a093644ac39fb..bbb64331cf8f4 100644
@@ -653,9 +653,6 @@ static inline void copy_buffer(void *from, void *to)
 		*p2++ = *p1++;
 }
 
-  
-  
-
 /* General Interrupt Handling */
 
 static void (*FloppyIRQHandler)( int status ) = NULL;
@@ -1228,6 +1225,7 @@ static void fd_rwsec_done1(int status)
 	}
 	else {
 		/* all sectors finished */
+		finish_fdc();
 		fd_end_request_cur(BLK_STS_OK);
 	}
 	return;
@@ -1475,15 +1473,6 @@ static void setup_req_params( int drive )
 			ReqTrack, ReqSector, (unsigned long)ReqData ));
 }
 
-static void ataflop_commit_rqs(struct blk_mq_hw_ctx *hctx)
-{
-	spin_lock_irq(&ataflop_lock);
-	atari_disable_irq(IRQ_MFP_FDC);
-	finish_fdc();
-	atari_enable_irq(IRQ_MFP_FDC);
-	spin_unlock_irq(&ataflop_lock);
-}
-
 static blk_status_t ataflop_queue_rq(struct blk_mq_hw_ctx *hctx,
 				     const struct blk_mq_queue_data *bd)
 {
@@ -1491,6 +1480,8 @@ static blk_status_t ataflop_queue_rq(struct blk_mq_hw_ctx *hctx,
 	int drive = floppy - unit;
 	int type = floppy->type;
 
+	DPRINT(("Queue request: drive %d type %d last %d\n", drive, type, bd->last));
+
 	spin_lock_irq(&ataflop_lock);
 	if (fd_request) {
 		spin_unlock_irq(&ataflop_lock);
@@ -1550,8 +1541,6 @@ static blk_status_t ataflop_queue_rq(struct blk_mq_hw_ctx *hctx,
 	setup_req_params( drive );
 	do_fd_action( drive );
 
-	if (bd->last)
-		finish_fdc();
 	atari_enable_irq( IRQ_MFP_FDC );
 
 out:
@@ -1962,7 +1951,6 @@ static const struct block_device_operations floppy_fops = {
 
 static const struct blk_mq_ops ataflop_mq_ops = {
 	.queue_rq = ataflop_queue_rq,
-	.commit_rqs = ataflop_commit_rqs,
 };
 
 static int ataflop_alloc_disk(unsigned int drive, unsigned int type)
Signature · 1 line
2.33.0
author maintainer
#461 lore raw

PATCH 450/917 blk-wbt: prevent NULL pointer dereference in wb_timer_fn

In reply to Greg Kroah-Hartman

From: Andrea Righi <[email protected]>

[ Upstream commit 480d42dc001bbfe953825a92073012fcd5a99161 ]

The timer callback used to evaluate if the latency is exceeded can be executed after the corresponding disk has been released, causing the following NULL pointer dereference:

[ 119.987108] BUG: kernel NULL pointer dereference, address: 0000000000000098
[ 119.987617] #PF: supervisor read access in kernel mode
[ 119.987971] #PF: error_code(0x0000) - not-present page
[ 119.988325] PGD 7c4a4067 P4D 7c4a4067 PUD 7bf63067 PMD 0
[ 119.988697] Oops: 0000 [#1] SMP NOPTI
[ 119.988959] CPU: 1 PID: 9353 Comm: cloud-init Not tainted 5.15-rc5+arighi #rc5+arighi
[ 119.989520] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-2 04/01/2014
[ 119.990055] RIP: 0010:wb_timer_fn+0x44/0x3c0
[ 119.990376] Code: 41 8b 9c 24 98 00 00 00 41 8b 94 24 b8 00 00 00 41 8b 84 24 d8 00 00 00 4d 8b 74 24 28 01 d3 01 c3 49 8b 44 24 60 48 8b 40 78 <4c> 8b b8 98 00 00 00 4d 85 f6 0f 84 c4 00 00 00 49 83 7c 24 30 00
[ 119.991578] RSP: 0000:ffffb5f580957da8 EFLAGS: 00010246
[ 119.991937] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000004
[ 119.992412] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff88f476d7f780
[ 119.992895] RBP: ffffb5f580957dd0 R08: 0000000000000000 R09: 0000000000000000
[ 119.993371] R10: 0000000000000004 R11: 0000000000000002 R12: ffff88f476c84500
[ 119.993847] R13: ffff88f4434390c0 R14: 0000000000000000 R15: ffff88f4bdc98c00
[ 119.994323] FS: 00007fb90bcd9c00(0000) GS:ffff88f4bdc80000(0000) knlGS:0000000000000000
[ 119.994952] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 119.995380] CR2: 0000000000000098 CR3: 000000007c0d6000 CR4: 00000000000006e0
[ 119.995906] Call Trace:
[ 119.996130] ? blk_stat_free_callback_rcu+0x30/0x30
[ 119.996505] blk_stat_timer_fn+0x138/0x140
[ 119.996830] call_timer_fn+0x2b/0x100
[ 119.997136] __run_timers.part.0+0x1d1/0x240
[ 119.997470] ? kvm_clock_get_cycles+0x11/0x20
[ 119.997826] ? ktime_get+0x3e/0xa0
[ 119.998110] ? native_apic_msr_write+0x2c/0x30
[ 119.998456] ? lapic_next_event+0x20/0x30
[ 119.998779] ? clockevents_program_event+0x94/0xf0
[ 119.999150] run_timer_softirq+0x2a/0x50
[ 119.999465] __do_softirq+0xcb/0x26f
[ 119.999764] irq_exit_rcu+0x8c/0xb0
[ 120.000057] sysvec_apic_timer_interrupt+0x43/0x90
[ 120.000429] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 120.000836] asm_sysvec_apic_timer_interrupt+0x12/0x20

In this case simply return from the timer callback (no action required) to prevent the NULL pointer dereference.

---
 block/blk-wbt.c | 3 +++
 1 file changed, 3 insertions(+)
block/blk-wbt.c +3 -0
index 874c1c37bf0c6..0c119be0e8133 100644
@@ -357,6 +357,9 @@ static void wb_timer_fn(struct blk_stat_callback *cb)
 	unsigned int inflight = wbt_inflight(rwb);
 	int status;
 
+	if (!rwb->rqos.q->disk)
+		return;
+
 	status = latency_exceeded(rwb, cb->stat);
 
 	trace_wbt_timer(rwb->rqos.q->disk->bdi, status, rqd->scale_step,
Signature · 1 line
2.33.0
author maintainer
#462 lore raw

PATCH 451/917 platform/x86: thinkpad_acpi: Fix bitwise vs. logical warning

In reply to Greg Kroah-Hartman

From: Nathan Chancellor <[email protected]>

[ Upstream commit fd96e35ea7b95f1e216277805be89d66e4ae962d ]

A new warning in clang points out a use of bitwise OR with boolean expressions in this driver:

drivers/platform/x86/thinkpad_acpi.c:9061:11: error: use of bitwise '|' with boolean operands [-Werror,-Wbitwise-instead-of-logical]

        else if ((strlencmp(cmd, "level disengaged") == 0) |
                 ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
                                                           ||

drivers/platform/x86/thinkpad_acpi.c:9061:11: note: cast one or both operands to int to silence this warning 1 error generated.

This should clearly be a logical OR so change it to fix the warning.

---
 drivers/platform/x86/thinkpad_acpi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/platform/x86/thinkpad_acpi.c +1 -1
index 50ff04c84650c..27595aba214d9 100644
@@ -9145,7 +9145,7 @@ static int fan_write_cmd_level(const char *cmd, int *rc)
 
 	if (strlencmp(cmd, "level auto") == 0)
 		level = TP_EC_FAN_AUTO;
-	else if ((strlencmp(cmd, "level disengaged") == 0) |
+	else if ((strlencmp(cmd, "level disengaged") == 0) ||
 			(strlencmp(cmd, "level full-speed") == 0))
 		level = TP_EC_FAN_FULLSPEED;
 	else if (sscanf(cmd, "level %d", &level) != 1)
Signature · 1 line
2.33.0
author maintainer
#463 lore raw

PATCH 452/917 mailbox: mtk-cmdq: Validate alias_id on probe

In reply to Greg Kroah-Hartman

From: Fei Shao <[email protected]>

[ Upstream commit 5c154b6a51c2d2d7f266b3ef49b7dd1dc8cb5b65 ]

of_alias_get_id() may return -ENODEV which leads to illegal access to the cmdq->clocks array.
Adding a check over alias_id to prevent the unexpected behavior.

Fixes 85dfdbfc13ea ("mailbox: cmdq: add multi-gce clocks support for mt8195")
---
 drivers/mailbox/mtk-cmdq-mailbox.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/mailbox/mtk-cmdq-mailbox.c +1 -1
index c591dab9d5a48..9b0cc3bb5b23a 100644
@@ -572,7 +572,7 @@ static int cmdq_probe(struct platform_device *pdev)
 			char clk_id[8];
 
 			alias_id = of_alias_get_id(node, clk_name);
-			if (alias_id < cmdq->gce_num) {
+			if (alias_id >= 0 && alias_id < cmdq->gce_num) {
 				snprintf(clk_id, sizeof(clk_id), "%s%d", clk_name, alias_id);
 				cmdq->clocks[alias_id].id = clk_id;
 				cmdq->clocks[alias_id].clk = of_clk_get(node, 0);
Signature · 1 line
2.33.0
author maintainer
#464 lore raw

PATCH 453/917 mailbox: mtk-cmdq: Fix local clock ID usage

In reply to Greg Kroah-Hartman

From: Fei Shao <[email protected]>

[ Upstream commit 0a5ad4322927ee4aaba6facc0e4faf1ab6c0d48e ]

In the probe function, the clock IDs were pointed to local variables which should only be used in the same code block, and any access to them after the probing stage becomes an use-after-free case.

Since there are only limited variants of the gce clock names so far, we can just declare them as static constants to fix the issue.

Fixes 85dfdbfc13ea ("mailbox: cmdq: add multi-gce clocks support for mt8195")
---
 drivers/mailbox/mtk-cmdq-mailbox.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)
drivers/mailbox/mtk-cmdq-mailbox.c +3 -5
index 9b0cc3bb5b23a..bb4793c7b38fd 100644
@@ -531,7 +531,8 @@ static int cmdq_probe(struct platform_device *pdev)
 	struct device_node *phandle = dev->of_node;
 	struct device_node *node;
 	int alias_id = 0;
-	char clk_name[4] = "gce";
+	static const char * const clk_name = "gce";
+	static const char * const clk_names[] = { "gce0", "gce1" };
 
 	cmdq = devm_kzalloc(dev, sizeof(*cmdq), GFP_KERNEL);
 	if (!cmdq)
@@ -569,12 +570,9 @@ static int cmdq_probe(struct platform_device *pdev)
 
 	if (cmdq->gce_num > 1) {
 		for_each_child_of_node(phandle->parent, node) {
-			char clk_id[8];
-
 			alias_id = of_alias_get_id(node, clk_name);
 			if (alias_id >= 0 && alias_id < cmdq->gce_num) {
-				snprintf(clk_id, sizeof(clk_id), "%s%d", clk_name, alias_id);
-				cmdq->clocks[alias_id].id = clk_id;
+				cmdq->clocks[alias_id].id = clk_names[alias_id];
 				cmdq->clocks[alias_id].clk = of_clk_get(node, 0);
 				if (IS_ERR(cmdq->clocks[alias_id].clk)) {
 					dev_err(dev, "failed to get gce clk: %d\n", alias_id);
Signature · 1 line
2.33.0
author maintainer
#465 lore raw

PATCH 454/917 ACPI: PM: Turn off unused wakeup power resources

In reply to Greg Kroah-Hartman

From: Rafael J. Wysocki <[email protected]>

[ Upstream commit 7a63296d6f579a02b2675b4b0fe5b1cd3235e8d3 ]

If an ACPI power resource is found to be "on" during the initialization of the list of wakeup power resources of a device, it is reference counted and its wakeup_enabled flag is set, which is problematic if the deivce in question is the only user of the given power resource, it is never runtime-suspended and it is not allowed to wake up the system from sleep, because in that case the given power resource will stay "on" until the system reboots and energy will be wasted.

It is better to simply turn off wakeup power resources that are "on" during the initialization unless their reference counters are not zero, because that may be the only opportunity to prevent them from staying in the "on" state all the time.

Fixes b5d667eb392e ("ACPI / PM: Take unusual configurations of power resources into account")
---
 drivers/acpi/power.c | 19 +++++++++----------
 1 file changed, 9 insertions(+), 10 deletions(-)
drivers/acpi/power.c +9 -10
index f0ed4414edb1f..3d34cee0cc101 100644
@@ -615,20 +615,19 @@ int acpi_power_wakeup_list_init(struct list_head *list, int *system_level_p)
 
 	list_for_each_entry(entry, list, node) {
 		struct acpi_power_resource *resource = entry->resource;
-		int result;
 		u8 state;
 
 		mutex_lock(&resource->resource_lock);
 
-		result = acpi_power_get_state(resource, &state);
-		if (result) {
-			mutex_unlock(&resource->resource_lock);
-			return result;
-		}
-		if (state == ACPI_POWER_RESOURCE_STATE_ON) {
-			resource->ref_count++;
-			resource->wakeup_enabled = true;
-		}
+		/*
+		 * Make sure that the power resource state and its reference
+		 * counter value are consistent with each other.
+		 */
+		if (!resource->ref_count &&
+		    !acpi_power_get_state(resource, &state) &&
+		    state == ACPI_POWER_RESOURCE_STATE_ON)
+			__acpi_power_off(resource);
+
 		if (system_level > resource->system_level)
 			system_level = resource->system_level;
 
Signature · 1 line
2.33.0
author maintainer
#466 lore raw

PATCH 455/917 ACPI: PM: Fix sharing of wakeup power resources

In reply to Greg Kroah-Hartman

From: Rafael J. Wysocki <[email protected]>

[ Upstream commit a2d7b2e004af6b09f21ac3d10f8f4456c16a8ddf ]

If an ACPI wakeup power resource is shared between multiple devices, it may not be managed correctly.

Suppose, for example, that two devices, A and B, share a wakeup power resource P whose wakeup_enabled flag is 0 initially. Next, suppose that wakeup power is enabled for A and B, in this order, and disabled for B. When wakeup power is enabled for A, P will be turned on and its wakeup_enabled flag will be set. Next, when wakeup power is enabled for B, P will not be touched, because its wakeup_enabled flag is set. Now, when wakeup power is disabled for B, P will be turned off which is incorrect, because A will still need P in order to signal wakeup.

Moreover, if wakeup power is enabled for A and then disabled for B, the latter will cause P to be turned off incorrectly (it will be still needed by A), because acpi_disable_wakeup_device_power() is allowed to manipulate power resources when the wakeup.prepare_count counter of the given device is 0.

While the first issue could be addressed by changing the wakeup_enabled power resource flag into a counter, addressing the second one requires modifying acpi_disable_wakeup_device_power() to do nothing when the target device's wakeup.prepare_count reference counter is zero and that would cause the new counter to be redundant. Namely, if acpi_disable_wakeup_device_power() is modified as per the above, every change of the new counter following a wakeup.prepare_count change would be reflected by the analogous change of the main reference counter of the given power resource.

Accordingly, modify acpi_disable_wakeup_device_power() to do nothing when the target device's wakeup.prepare_count reference counter is zero and drop the power resource wakeup_enabled flag altogether.

While at it, ensure that all of the power resources that can be turned off will be turned off when disabling device wakeup due to a power resource manipulation error, to prevent energy from being wasted.

Fixes b5d667eb392e ("ACPI / PM: Take unusual configurations of power resources into account")
---
 drivers/acpi/power.c | 69 +++++++++++++++-----------------------------
 1 file changed, 24 insertions(+), 45 deletions(-)
drivers/acpi/power.c +24 -45
index 3d34cee0cc101..4b42debeed455 100644
@@ -52,7 +52,6 @@ struct acpi_power_resource {
 	u32 order;
 	unsigned int ref_count;
 	u8 state;
-	bool wakeup_enabled;
 	struct mutex resource_lock;
 	struct list_head dependents;
 };
@@ -710,7 +709,6 @@ int acpi_device_sleep_wake(struct acpi_device *dev,
  */
 int acpi_enable_wakeup_device_power(struct acpi_device *dev, int sleep_state)
 {
-	struct acpi_power_resource_entry *entry;
 	int err = 0;
 
 	if (!dev || !dev->wakeup.flags.valid)
@@ -721,26 +719,13 @@ int acpi_enable_wakeup_device_power(struct acpi_device *dev, int sleep_state)
 	if (dev->wakeup.prepare_count++)
 		goto out;
 
-	list_for_each_entry(entry, &dev->wakeup.resources, node) {
-		struct acpi_power_resource *resource = entry->resource;
-
-		mutex_lock(&resource->resource_lock);
-
-		if (!resource->wakeup_enabled) {
-			err = acpi_power_on_unlocked(resource);
-			if (!err)
-				resource->wakeup_enabled = true;
-		}
-
-		mutex_unlock(&resource->resource_lock);
-
-		if (err) {
-			dev_err(&dev->dev,
-				"Cannot turn wakeup power resources on\n");
-			dev->wakeup.flags.valid = 0;
-			goto out;
-		}
+	err = acpi_power_on_list(&dev->wakeup.resources);
+	if (err) {
+		dev_err(&dev->dev, "Cannot turn on wakeup power resources\n");
+		dev->wakeup.flags.valid = 0;
+		goto out;
 	}
+
 	/*
 	 * Passing 3 as the third argument below means the device may be
 	 * put into arbitrary power state afterward.
@@ -770,39 +755,33 @@ int acpi_disable_wakeup_device_power(struct acpi_device *dev)
 
 	mutex_lock(&acpi_device_lock);
 
-	if (--dev->wakeup.prepare_count > 0)
+	if (dev->wakeup.prepare_count > 1) {
+		dev->wakeup.prepare_count--;
 		goto out;
+	}
 
-	/*
-	 * Executing the code below even if prepare_count is already zero when
-	 * the function is called may be useful, for example for initialisation.
-	 */
-	if (dev->wakeup.prepare_count < 0)
-		dev->wakeup.prepare_count = 0;
+	/* Do nothing if wakeup power has not been enabled for this device. */
+	if (!dev->wakeup.prepare_count)
+		goto out;
 
 	err = acpi_device_sleep_wake(dev, 0, 0, 0);
 	if (err)
 		goto out;
 
+	/*
+	 * All of the power resources in the list need to be turned off even if
+	 * there are errors.
+	 */
 	list_for_each_entry(entry, &dev->wakeup.resources, node) {
-		struct acpi_power_resource *resource = entry->resource;
-
-		mutex_lock(&resource->resource_lock);
-
-		if (resource->wakeup_enabled) {
-			err = acpi_power_off_unlocked(resource);
-			if (!err)
-				resource->wakeup_enabled = false;
-		}
-
-		mutex_unlock(&resource->resource_lock);
+		int ret;
 
-		if (err) {
-			dev_err(&dev->dev,
-				"Cannot turn wakeup power resources off\n");
-			dev->wakeup.flags.valid = 0;
-			break;
-		}
+		ret = acpi_power_off(entry->resource);
+		if (ret && !err)
+			err = ret;
+	}
+	if (err) {
+		dev_err(&dev->dev, "Cannot turn off wakeup power resources\n");
+		dev->wakeup.flags.valid = 0;
 	}
 
  out:
Signature · 1 line
2.33.0
author maintainer
#467 lore raw

PATCH 456/917 drm/amdkfd: Fix an inappropriate error handling in allloc memory of gpu

In reply to Greg Kroah-Hartman

From: Lang Yu <[email protected]>

[ Upstream commit 5aeeac6fa38fca450faed9770f75b1470c0e2073 ]

We should unreference a gem object instead of an amdgpu bo here.

Fixes fd9a9f8801de ("drm/amdgpu: Use GEM obj reference for KFD BOs")
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +1 -1
index 054c1a224defb..cdf46bd0d8d5b 100644
@@ -1503,7 +1503,7 @@ allocate_init_user_pages_failed:
 	remove_kgd_mem_from_kfd_bo_list(*mem, avm->process_info);
 	drm_vma_node_revoke(&gobj->vma_node, drm_priv);
 err_node_allow:
-	amdgpu_bo_unref(&bo);
+	drm_gem_object_put(gobj);
 	/* Don't unreserve system mem limit twice */
 	goto err_reserve_limit;
 err_bo_create:
Signature · 1 line
2.33.0
author maintainer
#468 lore raw

PATCH 457/917 mt76: mt7921: fix endianness in mt7921_mcu_tx_done_event

In reply to Greg Kroah-Hartman

From: Lorenzo Bianconi <[email protected]>

[ Upstream commit df040215c077de0c13aab12c222bd0360a0d3988 ]

Fix endianness in mt7921_mcu_tx_done_event event reported by the firmware.

Fixes 3cce2b98e0241 ("mt76: mt7921: introduce mac tx done handling")
---
 drivers/net/wireless/mediatek/mt76/mt7921/mcu.c |  3 ++-
 drivers/net/wireless/mediatek/mt76/mt7921/mcu.h | 10 +++++-----
 2 files changed, 7 insertions(+), 6 deletions(-)
drivers/net/wireless/mediatek/mt76/mt7921/mcu.c +2 -1
index 9fbaacc67cfad..68840e55ede7a 100644
@@ -532,7 +532,8 @@ mt7921_mcu_tx_done_event(struct mt7921_dev *dev, struct sk_buff *skb)
 		peer.g8 = !!(sta->vht_cap.cap & IEEE80211_VHT_CAP_SHORT_GI_80);
 		peer.g16 = !!(sta->vht_cap.cap & IEEE80211_VHT_CAP_SHORT_GI_160);
 		mt7921_mcu_tx_rate_parse(mphy->mt76, &peer,
-					 &msta->stats.tx_rate, event->tx_rate);
+					 &msta->stats.tx_rate,
+					 le16_to_cpu(event->tx_rate));
 
 		spin_lock_bh(&dev->sta_poll_lock);
 		break;
drivers/net/wireless/mediatek/mt76/mt7921/mcu.h +5 -5
index de3c091f67368..42e7271848956 100644
@@ -296,11 +296,11 @@ struct mt7921_txpwr_event {
 struct mt7921_mcu_tx_done_event {
 	u8 pid;
 	u8 status;
-	u16 seq;
+	__le16 seq;
 
 	u8 wlan_idx;
 	u8 tx_cnt;
-	u16 tx_rate;
+	__le16 tx_rate;
 
 	u8 flag;
 	u8 tid;
@@ -312,9 +312,9 @@ struct mt7921_mcu_tx_done_event {
 	u8 reason;
 	u8 rsv0[1];
 
-	u32 delay;
-	u32 timestamp;
-	u32 applied_flag;
+	__le32 delay;
+	__le32 timestamp;
+	__le32 applied_flag;
 
 	u8 txs[28];
 
Signature · 1 line
2.33.0
author maintainer
#469 lore raw

PATCH 458/917 mt76: mt7915: fix endianness warning in mt7915_mac_add_txs_skb

In reply to Greg Kroah-Hartman

From: Lorenzo Bianconi <[email protected]>

[ Upstream commit 08b3c8da87aed4200dab00906f149d675ca90f23 ]

Fix the following sparse warning in mt7915_mac_add_txs_skb routine:

drivers/net/wireless/mediatek/mt76/mt7915/mac.c:1235:29:
warning: cast to restricted __le32
drivers/net/wireless/mediatek/mt76/mt7915/mac.c:1235:23:
warning: restricted __le32 degrades to integer

Fixes 3de4cb1756565 ("mt76: mt7915: add support for tx status reporting")
---
 drivers/net/wireless/mediatek/mt76/mt7915/mac.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/mediatek/mt76/mt7915/mac.c +1 -1
index 2462704094b0a..bbc996f86b5c3 100644
@@ -1232,7 +1232,7 @@ mt7915_mac_add_txs_skb(struct mt7915_dev *dev, struct mt76_wcid *wcid, int pid,
 		goto out;
 
 	info = IEEE80211_SKB_CB(skb);
-	if (!(txs_data[0] & le32_to_cpu(MT_TXS0_ACK_ERROR_MASK)))
+	if (!(txs_data[0] & cpu_to_le32(MT_TXS0_ACK_ERROR_MASK)))
 		info->flags |= IEEE80211_TX_STAT_ACK;
 
 	info->status.ampdu_len = 1;
Signature · 1 line
2.33.0
author maintainer
#470 lore raw

PATCH 459/917 mt76: mt7921: fix endianness warning in mt7921_update_txs

In reply to Greg Kroah-Hartman

From: Lorenzo Bianconi <[email protected]>

[ Upstream commit 7fc167bbc9296e6aeaaa4063db3639e8a3db75f6 ]

Fix the following sparse warning in mt7921_update_txs routine:
drivers/net/wireless/mediatek/mt76/mt7921/mac.c:752:31:
warning: cast to restricted __le32
drivers/net/wireless/mediatek/mt76/mt7921/mac.c:752:31:
warning: restricted __le32 degrades to integer

Fixes e5bca8c5d2cd3 ("mt76: mt7921: improve code readability for mt7921_update_txs")
---
 drivers/net/wireless/mediatek/mt76/mt7921/mac.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/wireless/mediatek/mt76/mt7921/mac.c +2 -1
index 7fe2e3a50428f..f4714b0f6e5c4 100644
@@ -735,8 +735,9 @@ mt7921_mac_write_txwi_80211(struct mt7921_dev *dev, __le32 *txwi,
 static void mt7921_update_txs(struct mt76_wcid *wcid, __le32 *txwi)
 {
 	struct mt7921_sta *msta = container_of(wcid, struct mt7921_sta, wcid);
-	u32 pid, frame_type = FIELD_GET(MT_TXD2_FRAME_TYPE, txwi[2]);
+	u32 pid, frame_type;
 
+	frame_type = FIELD_GET(MT_TXD2_FRAME_TYPE, le32_to_cpu(txwi[2]));
 	if (!(frame_type & (IEEE80211_FTYPE_DATA >> 2)))
 		return;
 
Signature · 1 line
2.33.0
author maintainer
#471 lore raw

PATCH 460/917 mt76: mt7615: fix endianness warning in mt7615_mac_write_txwi

In reply to Greg Kroah-Hartman

From: Lorenzo Bianconi <[email protected]>

[ Upstream commit d81bfb41e30c42531536c5d2baa4d275a8309715 ]

Fix the following sparse warning in mt7615_mac_write_txwi routine:
drivers/net/wireless/mediatek/mt76/mt7615/mac.c:758:17:
warning: incorrect type in assignment
expected restricted __le32 [usertype]
got unsigned long

Fixes 04b8e65922f63 ("mt76: add mac80211 driver for MT7615 PCIe-based chipsets")
Fixes d4bf77bd74930 ("mt76: mt7615: introduce mt7663u support to mt7615_write_txwi")
---
 drivers/net/wireless/mediatek/mt76/mt7615/mac.c | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)
drivers/net/wireless/mediatek/mt76/mt7615/mac.c +9 -6
index a83e48c07a71e..5455231f51881 100644
@@ -755,12 +755,15 @@ int mt7615_mac_write_txwi(struct mt7615_dev *dev, __le32 *txwi,
 	if (info->flags & IEEE80211_TX_CTL_NO_ACK)
 		txwi[3] |= cpu_to_le32(MT_TXD3_NO_ACK);
 
-	txwi[7] = FIELD_PREP(MT_TXD7_TYPE, fc_type) |
-		  FIELD_PREP(MT_TXD7_SUB_TYPE, fc_stype) |
-		  FIELD_PREP(MT_TXD7_SPE_IDX, 0x18);
-	if (!is_mmio)
-		txwi[8] = FIELD_PREP(MT_TXD8_L_TYPE, fc_type) |
-			  FIELD_PREP(MT_TXD8_L_SUB_TYPE, fc_stype);
+	val = FIELD_PREP(MT_TXD7_TYPE, fc_type) |
+	      FIELD_PREP(MT_TXD7_SUB_TYPE, fc_stype) |
+	      FIELD_PREP(MT_TXD7_SPE_IDX, 0x18);
+	txwi[7] = cpu_to_le32(val);
+	if (!is_mmio) {
+		val = FIELD_PREP(MT_TXD8_L_TYPE, fc_type) |
+		      FIELD_PREP(MT_TXD8_L_SUB_TYPE, fc_stype);
+		txwi[8] = cpu_to_le32(val);
+	}
 
 	return 0;
 }
Signature · 1 line
2.33.0
author maintainer
#472 lore raw

PATCH 461/917 mt76: mt7915: fix info leak in mt7915_mcu_set_pre_cal()

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

[ Upstream commit 3924715ffe5e064a85f56490f77b7b2084230800 ]

Zero out all the unused members of "req" so that we don't disclose stack information.

Fixes 495184ac91bb ("mt76: mt7915: add support for applying pre-calibration data")
---
 drivers/net/wireless/mediatek/mt76/mt7915/mcu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/mediatek/mt76/mt7915/mcu.c +1 -1
index 2f30047bd80f2..caf2033c5c17e 100644
@@ -3481,7 +3481,7 @@ static int mt7915_mcu_set_pre_cal(struct mt7915_dev *dev, u8 idx,
 		u8 idx;
 		u8 rsv[4];
 		__le32 len;
-	} req;
+	} req = {};
 	struct sk_buff *skb;
 
 	skb = mt76_mcu_msg_alloc(&dev->mt76, NULL, sizeof(req) + len);
Signature · 1 line
2.33.0
author maintainer
#473 lore raw

PATCH 462/917 mt76: connac: fix mt76_connac_gtk_rekey_tlv usage

In reply to Greg Kroah-Hartman

From: Leon Yen <[email protected]>

[ Upstream commit d741abeafa47a7331cd4fe526e44db4ad3da0f62 ]

The mistaken structure is introduced since we added the GTK rekey offload to mt7663. The patch fixes mt76_connac_gtk_rekey_tlv structure according to the MT7663 and MT7921 firmware we have submitted into linux-firmware.git.

Fixes b47e21e75c80 ("mt76: mt7615: add gtk rekey offload support")
---
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h +3 -3
index 1c73beb226771..4bcd728ff97c5 100644
@@ -844,14 +844,14 @@ struct mt76_connac_gtk_rekey_tlv {
 			* 2: rekey update
 			*/
 	u8 keyid;
-	u8 pad[2];
+	u8 option; /* 1: rekey data update without enabling offload */
+	u8 pad[1];
 	__le32 proto; /* WPA-RSN-WAPI-OPSN */
 	__le32 pairwise_cipher;
 	__le32 group_cipher;
 	__le32 key_mgmt; /* NONE-PSK-IEEE802.1X */
 	__le32 mgmt_group_cipher;
-	u8 option; /* 1: rekey data update without enabling offload */
-	u8 reserverd[3];
+	u8 reserverd[4];
 } __packed;
 
 #define MT76_CONNAC_WOW_MASK_MAX_LEN			16
Signature · 1 line
2.33.0
author maintainer
#474 lore raw

PATCH 463/917 mt76: fix build error implicit enumeration conversion

In reply to Greg Kroah-Hartman

From: Sean Wang <[email protected]>

[ Upstream commit adedbc643f02f5a3193b8dcc5cfca97b4c988667 ]

drivers/net/wireless/mediatek/mt76/mt7915/mcu.c:114:10: error: implicit conversion from enumeration type 'enum mt76_cipher_type' to different enumeration type 'enum mcu_cipher_type' [-Werror,-Wenum-conversion]

                return MT_CIPHER_NONE;
                ~~~~~~ ^~~~~~~~~~~~~~

drivers/net/wireless/mediatek/mt76/mt7921/mcu.c:114:10: error: implicit conversion from enumeration type 'enum mt76_cipher_type' to different enumeration type 'enum mcu_cipher_type' [-Werror,-Wenum-conversion]

                return MT_CIPHER_NONE;
                ~~~~~~ ^~~~~~~~~~~~~~
Fixes c368362c36d3 ("mt76: fix iv and CCMP header insertion")
---
 drivers/net/wireless/mediatek/mt76/mt7915/mcu.c | 2 +-
 drivers/net/wireless/mediatek/mt76/mt7921/mcu.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
drivers/net/wireless/mediatek/mt76/mt7915/mcu.c +1 -1
index caf2033c5c17e..c08c7398f9b85 100644
@@ -1201,7 +1201,7 @@ mt7915_mcu_sta_key_tlv(struct mt7915_sta *msta, struct sk_buff *skb,
 		u8 cipher;
 
 		cipher = mt7915_mcu_get_cipher(key->cipher);
-		if (cipher == MT_CIPHER_NONE)
+		if (cipher == MCU_CIPHER_NONE)
 			return -EOPNOTSUPP;
 
 		sec_key = &sec->key[0];
drivers/net/wireless/mediatek/mt76/mt7921/mcu.c +1 -1
index 68840e55ede7a..8329b705c2ca2 100644
@@ -620,7 +620,7 @@ mt7921_mcu_sta_key_tlv(struct mt7921_sta *msta, struct sk_buff *skb,
 		u8 cipher;
 
 		cipher = mt7921_mcu_get_cipher(key->cipher);
-		if (cipher == MT_CIPHER_NONE)
+		if (cipher == MCU_CIPHER_NONE)
 			return -EOPNOTSUPP;
 
 		sec_key = &sec->key[0];
Signature · 1 line
2.33.0
author maintainer
#475 lore raw

PATCH 464/917 mt76: mt7921: fix survey-dump reporting

In reply to Greg Kroah-Hartman

From: Lorenzo Bianconi <[email protected]>

[ Upstream commit 64ed76d118c656907ec1155f2cdd24de778470a2 ]

Fix MIB tx-rx MIB counters for survey-dump reporting.

Fixes 163f4d22c118d ("mt76: mt7921: add MAC support")
---
 drivers/net/wireless/mediatek/mt76/mt7921/init.c | 4 ++++
 drivers/net/wireless/mediatek/mt76/mt7921/regs.h | 8 ++++++--
 2 files changed, 10 insertions(+), 2 deletions(-)
drivers/net/wireless/mediatek/mt76/mt7921/init.c +4 -0
index a9ce10b988273..52d40385fab6c 100644
@@ -106,6 +106,10 @@ mt7921_mac_init_band(struct mt7921_dev *dev, u8 band)
 	mt76_set(dev, MT_WF_RMAC_MIB_TIME0(band), MT_WF_RMAC_MIB_RXTIME_EN);
 	mt76_set(dev, MT_WF_RMAC_MIB_AIRTIME0(band), MT_WF_RMAC_MIB_RXTIME_EN);
 
+	/* enable MIB tx-rx time reporting */
+	mt76_set(dev, MT_MIB_SCR1(band), MT_MIB_TXDUR_EN);
+	mt76_set(dev, MT_MIB_SCR1(band), MT_MIB_RXDUR_EN);
+
 	mt76_rmw_field(dev, MT_DMA_DCR0(band), MT_DMA_DCR0_MAX_RX_LEN, 1536);
 	/* disable rx rate report by default due to hw issues */
 	mt76_clear(dev, MT_DMA_DCR0(band), MT_DMA_DCR0_RXD_G5_EN);
drivers/net/wireless/mediatek/mt76/mt7921/regs.h +6 -2
index b6944c867a573..26fb118237626 100644
@@ -96,6 +96,10 @@
 #define MT_WF_MIB_BASE(_band)		((_band) ? 0xa4800 : 0x24800)
 #define MT_WF_MIB(_band, ofs)		(MT_WF_MIB_BASE(_band) + (ofs))
 
+#define MT_MIB_SCR1(_band)		MT_WF_MIB(_band, 0x004)
+#define MT_MIB_TXDUR_EN			BIT(8)
+#define MT_MIB_RXDUR_EN			BIT(9)
+
 #define MT_MIB_SDR3(_band)		MT_WF_MIB(_band, 0x698)
 #define MT_MIB_SDR3_FCS_ERR_MASK	GENMASK(31, 16)
 
@@ -108,9 +112,9 @@
 #define MT_MIB_SDR34(_band)		MT_WF_MIB(_band, 0x090)
 #define MT_MIB_MU_BF_TX_CNT		GENMASK(15, 0)
 
-#define MT_MIB_SDR36(_band)		MT_WF_MIB(_band, 0x098)
+#define MT_MIB_SDR36(_band)		MT_WF_MIB(_band, 0x054)
 #define MT_MIB_SDR36_TXTIME_MASK	GENMASK(23, 0)
-#define MT_MIB_SDR37(_band)		MT_WF_MIB(_band, 0x09c)
+#define MT_MIB_SDR37(_band)		MT_WF_MIB(_band, 0x058)
 #define MT_MIB_SDR37_RXTIME_MASK	GENMASK(23, 0)
 
 #define MT_MIB_DR8(_band)		MT_WF_MIB(_band, 0x0c0)
Signature · 1 line
2.33.0
author maintainer
#476 lore raw

PATCH 465/917 mt76: mt76x02: fix endianness warnings in mt76x02_mac.c

In reply to Greg Kroah-Hartman

From: Lorenzo Bianconi <[email protected]>

[ Upstream commit c33edef520213feccebc22c9474c685b9fb60611 ]

Fix the following sparse warning in mt76x02_mac_write_txwi and mt76x02_mac_tx_rate_val routines:
drivers/net/wireless/mediatek/mt76/mt76x02_mac.c:237:19:
warning: restricted __le16 degrades to intege
warning: cast from restricted __le16
drivers/net/wireless/mediatek/mt76/mt76x02_mac.c:383:28:
warning: incorrect type in assignment (different base types) expected restricted __le16 [usertype] rate
got unsigned long

Fixes db9f11d3433f7 ("mt76: store wcid tx rate info in one u32 reduce locking")
---
 drivers/net/wireless/mediatek/mt76/mt76x02_mac.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)
drivers/net/wireless/mediatek/mt76/mt76x02_mac.c +7 -6
index c32e6dc687739..07b21b2085823 100644
@@ -176,7 +176,7 @@ void mt76x02_mac_wcid_set_drop(struct mt76x02_dev *dev, u8 idx, bool drop)
 		mt76_wr(dev, MT_WCID_DROP(idx), (val & ~bit) | (bit * drop));
 }
 
-static __le16
+static u16
 mt76x02_mac_tx_rate_val(struct mt76x02_dev *dev,
 			const struct ieee80211_tx_rate *rate, u8 *nss_val)
 {
@@ -222,14 +222,14 @@ mt76x02_mac_tx_rate_val(struct mt76x02_dev *dev,
 		rateval |= MT_RXWI_RATE_SGI;
 
 	*nss_val = nss;
-	return cpu_to_le16(rateval);
+	return rateval;
 }
 
 void mt76x02_mac_wcid_set_rate(struct mt76x02_dev *dev, struct mt76_wcid *wcid,
 			       const struct ieee80211_tx_rate *rate)
 {
 	s8 max_txpwr_adj = mt76x02_tx_get_max_txpwr_adj(dev, rate);
-	__le16 rateval;
+	u16 rateval;
 	u32 tx_info;
 	s8 nss;
 
@@ -342,7 +342,7 @@ void mt76x02_mac_write_txwi(struct mt76x02_dev *dev, struct mt76x02_txwi *txwi,
 	struct ieee80211_key_conf *key = info->control.hw_key;
 	u32 wcid_tx_info;
 	u16 rate_ht_mask = FIELD_PREP(MT_RXWI_RATE_PHY, BIT(1) | BIT(2));
-	u16 txwi_flags = 0;
+	u16 txwi_flags = 0, rateval;
 	u8 nss;
 	s8 txpwr_adj, max_txpwr_adj;
 	u8 ccmp_pn[8], nstreams = dev->mphy.chainmask & 0xf;
@@ -380,14 +380,15 @@ void mt76x02_mac_write_txwi(struct mt76x02_dev *dev, struct mt76x02_txwi *txwi,
 
 	if (wcid && (rate->idx < 0 || !rate->count)) {
 		wcid_tx_info = wcid->tx_info;
-		txwi->rate = FIELD_GET(MT_WCID_TX_INFO_RATE, wcid_tx_info);
+		rateval = FIELD_GET(MT_WCID_TX_INFO_RATE, wcid_tx_info);
 		max_txpwr_adj = FIELD_GET(MT_WCID_TX_INFO_TXPWR_ADJ,
 					  wcid_tx_info);
 		nss = FIELD_GET(MT_WCID_TX_INFO_NSS, wcid_tx_info);
 	} else {
-		txwi->rate = mt76x02_mac_tx_rate_val(dev, rate, &nss);
+		rateval = mt76x02_mac_tx_rate_val(dev, rate, &nss);
 		max_txpwr_adj = mt76x02_tx_get_max_txpwr_adj(dev, rate);
 	}
+	txwi->rate = cpu_to_le16(rateval);
 
 	txpwr_adj = mt76x02_tx_get_txpwr_adj(dev, dev->txpower_conf,
 					     max_txpwr_adj);
Signature · 1 line
2.33.0
author maintainer
#477 lore raw

PATCH 466/917 mt76: mt7921: Fix out of order process by invalid event pkt

In reply to Greg Kroah-Hartman

From: Deren Wu <[email protected]>

[ Upstream commit cd3f387371e941e6806b455b4ba5b9f4ca4b77c6 ]

The acceptable event report should inlcude original CMD-ID. Otherwise, drop unexpected result from fw.

Fixes 1c099ab44727c ("mt76: mt7921: add MCU support")
---
 drivers/net/wireless/mediatek/mt76/mt7921/mcu.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/net/wireless/mediatek/mt76/mt7921/mcu.c +4 -0
index 8329b705c2ca2..8ced55501d373 100644
@@ -157,6 +157,7 @@ mt7921_mcu_parse_response(struct mt76_dev *mdev, int cmd,
 			  struct sk_buff *skb, int seq)
 {
 	struct mt7921_mcu_rxd *rxd;
+	int mcu_cmd = cmd & MCU_CMD_MASK;
 	int ret = 0;
 
 	if (!skb) {
@@ -194,6 +195,9 @@ mt7921_mcu_parse_response(struct mt76_dev *mdev, int cmd,
 		skb_pull(skb, sizeof(*rxd));
 		event = (struct mt7921_mcu_uni_event *)skb->data;
 		ret = le32_to_cpu(event->status);
+		/* skip invalid event */
+		if (mcu_cmd != event->cid)
+			ret = -EAGAIN;
 		break;
 	}
 	case MCU_CMD_REG_READ: {
Signature · 1 line
2.33.0
author maintainer
#478 lore raw

PATCH 467/917 mt76: mt7915: fix potential overflow of eeprom page index

In reply to Greg Kroah-Hartman

From: Shayne Chen <[email protected]>

[ Upstream commit 82a980f82a511ce74ab57eb9f692d02225eb32f4 ]

If total eeprom size is divisible by per-page size, the i in for loop will exceed max page index, which happens in our newer chipset.

Fixes 26f18380e6ca ("mt76: mt7915: add support for flash mode")
---
 drivers/net/wireless/mediatek/mt76/mt7915/mcu.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)
drivers/net/wireless/mediatek/mt76/mt7915/mcu.c +5 -5
index c08c7398f9b85..e7e396f58c92c 100644
@@ -3391,20 +3391,20 @@ int mt7915_mcu_set_chan_info(struct mt7915_phy *phy, int cmd)
 
 static int mt7915_mcu_set_eeprom_flash(struct mt7915_dev *dev)
 {
-#define TOTAL_PAGE_MASK		GENMASK(7, 5)
+#define MAX_PAGE_IDX_MASK	GENMASK(7, 5)
 #define PAGE_IDX_MASK		GENMASK(4, 2)
 #define PER_PAGE_SIZE		0x400
 	struct mt7915_mcu_eeprom req = { .buffer_mode = EE_MODE_BUFFER };
-	u8 total = MT7915_EEPROM_SIZE / PER_PAGE_SIZE;
+	u8 total = DIV_ROUND_UP(MT7915_EEPROM_SIZE, PER_PAGE_SIZE);
 	u8 *eep = (u8 *)dev->mt76.eeprom.data;
 	int eep_len;
 	int i;
 
-	for (i = 0; i <= total; i++, eep += eep_len) {
+	for (i = 0; i < total; i++, eep += eep_len) {
 		struct sk_buff *skb;
 		int ret;
 
-		if (i == total)
+		if (i == total - 1 && !!(MT7915_EEPROM_SIZE % PER_PAGE_SIZE))
 			eep_len = MT7915_EEPROM_SIZE % PER_PAGE_SIZE;
 		else
 			eep_len = PER_PAGE_SIZE;
@@ -3414,7 +3414,7 @@ static int mt7915_mcu_set_eeprom_flash(struct mt7915_dev *dev)
 		if (!skb)
 			return -ENOMEM;
 
-		req.format = FIELD_PREP(TOTAL_PAGE_MASK, total) |
+		req.format = FIELD_PREP(MAX_PAGE_IDX_MASK, total - 1) |
 			     FIELD_PREP(PAGE_IDX_MASK, i) | EE_FORMAT_WHOLE;
 		req.len = cpu_to_le16(eep_len);
 
Signature · 1 line
2.33.0
author maintainer
#479 lore raw

PATCH 468/917 mt76: mt7915: fix bit fields for HT rate idx

In reply to Greg Kroah-Hartman

From: Shayne Chen <[email protected]>

[ Upstream commit 47f1c08db7f3aaa2d13f8e56209375462ace7b8a ]

The bit fields of tx rate idx should be 6 bits, otherwise it might be incorrect in HT mode.
For VHT/HE rates, only 4 bits are actually used by rate idx, the other 2 bits are used for other functions.

Fixes c31d94af1843 ("mt76: mt7915: fix tx rate related fields in tx descriptor")
---
 drivers/net/wireless/mediatek/mt76/mt7915/mac.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/wireless/mediatek/mt76/mt7915/mac.h +2 -1
index eb1885f4bd8eb..fee7741b5d421 100644
@@ -272,7 +272,8 @@ enum tx_mcu_port_q_idx {
 #define MT_TX_RATE_MODE			GENMASK(9, 6)
 #define MT_TX_RATE_SU_EXT_TONE		BIT(5)
 #define MT_TX_RATE_DCM			BIT(4)
-#define MT_TX_RATE_IDX			GENMASK(3, 0)
+/* VHT/HE only use bits 0-3 */
+#define MT_TX_RATE_IDX			GENMASK(5, 0)
 
 #define MT_TXP_MAX_BUF_NUM		6
 
Signature · 1 line
2.33.0
author maintainer
#480 lore raw

PATCH 469/917 mt76: mt7921: fix dma hang in rmmod

In reply to Greg Kroah-Hartman

From: Deren Wu <[email protected]>

[ Upstream commit a23f80aa9c5e6ad4ec8df88037b7ffd4162b1ec4 ]

The dma would be broken after rmmod flow. There are two different cases causing this issue.
1. dma access without privilege.
2. hw access sequence borken by another context.

This patch handle both cases to avoid hw crash.

Fixes 2b9ea5a8cf1bd ("mt76: mt7921: add mt7921_dma_cleanup in mt7921_unregister_device")
---
 drivers/net/wireless/mediatek/mt76/mt7921/init.c | 9 +++++++++
 1 file changed, 9 insertions(+)
drivers/net/wireless/mediatek/mt76/mt7921/init.c +9 -0
index 52d40385fab6c..78a00028137bd 100644
@@ -251,8 +251,17 @@ int mt7921_register_device(struct mt7921_dev *dev)
 
 void mt7921_unregister_device(struct mt7921_dev *dev)
 {
+	int i;
+	struct mt76_connac_pm *pm = &dev->pm;
+
 	mt76_unregister_device(&dev->mt76);
+	mt76_for_each_q_rx(&dev->mt76, i)
+		napi_disable(&dev->mt76.napi[i]);
+	cancel_delayed_work_sync(&pm->ps_work);
+	cancel_work_sync(&pm->wake_work);
+
 	mt7921_tx_token_put(dev);
+	mt7921_mcu_drv_pmctrl(dev);
 	mt7921_dma_cleanup(dev);
 	mt7921_mcu_exit(dev);
 
Signature · 1 line
2.33.0
author maintainer
#481 lore raw

PATCH 470/917 mt76: connac: fix GTK rekey offload failure on WPA mixed mode

In reply to Greg Kroah-Hartman

From: Leon Yen <[email protected]>

[ Upstream commit 781f62960c635cfed55a8f8c0f909bdaf8268257 ]

Update the proper firmware programming sequence to fix GTK rekey offload failure on WPA mixed mode.

In the mt76_connac_mcu_key_iter,
gtk_tlv->proto should be only set up on pairwise key and gtk_tlk->group_cipher should be only set up on the group key.

Otherwise, those parameters required by firmware would be set incorrectly to cause GTK rekey offload failure on WPA mixed mode and then disconnection follows.

Fixes b47e21e75c80 ("mt76: mt7615: add gtk rekey offload support")
---
 .../net/wireless/mediatek/mt76/mt76_connac_mcu.c  | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)
drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c +9 -6
index 5c3a81e5f559d..f57f047fce99c 100644
@@ -1929,19 +1929,22 @@ mt76_connac_mcu_key_iter(struct ieee80211_hw *hw,
 	    key->cipher != WLAN_CIPHER_SUITE_TKIP)
 		return;
 
-	if (key->cipher == WLAN_CIPHER_SUITE_TKIP) {
-		gtk_tlv->proto = cpu_to_le32(NL80211_WPA_VERSION_1);
+	if (key->cipher == WLAN_CIPHER_SUITE_TKIP)
 		cipher = BIT(3);
-	} else {
-		gtk_tlv->proto = cpu_to_le32(NL80211_WPA_VERSION_2);
+	else
 		cipher = BIT(4);
-	}
 
 	/* we are assuming here to have a single pairwise key */
 	if (key->flags & IEEE80211_KEY_FLAG_PAIRWISE) {
+		if (key->cipher == WLAN_CIPHER_SUITE_TKIP)
+			gtk_tlv->proto = cpu_to_le32(NL80211_WPA_VERSION_1);
+		else
+			gtk_tlv->proto = cpu_to_le32(NL80211_WPA_VERSION_2);
+
 		gtk_tlv->pairwise_cipher = cpu_to_le32(cipher);
-		gtk_tlv->group_cipher = cpu_to_le32(cipher);
 		gtk_tlv->keyid = key->keyidx;
+	} else {
+		gtk_tlv->group_cipher = cpu_to_le32(cipher);
 	}
 }
 
Signature · 1 line
2.33.0
author maintainer
#482 lore raw

PATCH 471/917 mt76: overwrite default reg_ops if necessary

In reply to Greg Kroah-Hartman

From: Lorenzo Bianconi <[email protected]>

[ Upstream commit f6e1f59885dae5a2553f8bbd328be2cb1c80ccb2 ]

Introduce mt76_register_debugfs_fops routine in order to define per-driver regs file operations and make sure the device is up before reading or writing its registers

Fixes 1d8efc741df8 ("mt76: mt7921: introduce Runtime PM support")
Fixes de5ff3c9d1a2 ("mt76: mt7615: introduce pm_power_save delayed work")
---
 drivers/net/wireless/mediatek/mt76/debugfs.c  | 10 ++++---
 drivers/net/wireless/mediatek/mt76/mt76.h     |  8 ++++-
 .../wireless/mediatek/mt76/mt7615/debugfs.c   | 29 ++++++++++++++++++-
 .../wireless/mediatek/mt76/mt7921/debugfs.c   | 28 +++++++++++++++++-
 4 files changed, 68 insertions(+), 7 deletions(-)
drivers/net/wireless/mediatek/mt76/debugfs.c +6 -4
index fa48cc3a7a8f7..ad97308c78534 100644
@@ -116,8 +116,11 @@ static int mt76_read_rate_txpower(struct seq_file *s, void *data)
 	return 0;
 }
 
-struct dentry *mt76_register_debugfs(struct mt76_dev *dev)
+struct dentry *
+mt76_register_debugfs_fops(struct mt76_dev *dev,
+			   const struct file_operations *ops)
 {
+	const struct file_operations *fops = ops ? ops : &fops_regval;
 	struct dentry *dir;
 
 	dir = debugfs_create_dir("mt76", dev->hw->wiphy->debugfsdir);
@@ -126,8 +129,7 @@ struct dentry *mt76_register_debugfs(struct mt76_dev *dev)
 
 	debugfs_create_u8("led_pin", 0600, dir, &dev->led_pin);
 	debugfs_create_u32("regidx", 0600, dir, &dev->debugfs_reg);
-	debugfs_create_file_unsafe("regval", 0600, dir, dev,
-				   &fops_regval);
+	debugfs_create_file_unsafe("regval", 0600, dir, dev, fops);
 	debugfs_create_file_unsafe("napi_threaded", 0600, dir, dev,
 				   &fops_napi_threaded);
 	debugfs_create_blob("eeprom", 0400, dir, &dev->eeprom);
@@ -140,4 +142,4 @@ struct dentry *mt76_register_debugfs(struct mt76_dev *dev)
 
 	return dir;
 }
-EXPORT_SYMBOL_GPL(mt76_register_debugfs);
+EXPORT_SYMBOL_GPL(mt76_register_debugfs_fops);
drivers/net/wireless/mediatek/mt76/mt76.h +7 -1
index 25c5ceef52577..4d01fd85283df 100644
@@ -869,7 +869,13 @@ struct mt76_phy *mt76_alloc_phy(struct mt76_dev *dev, unsigned int size,
 int mt76_register_phy(struct mt76_phy *phy, bool vht,
 		      struct ieee80211_rate *rates, int n_rates);
 
-struct dentry *mt76_register_debugfs(struct mt76_dev *dev);
+struct dentry *mt76_register_debugfs_fops(struct mt76_dev *dev,
+					  const struct file_operations *ops);
+static inline struct dentry *mt76_register_debugfs(struct mt76_dev *dev)
+{
+	return mt76_register_debugfs_fops(dev, NULL);
+}
+
 int mt76_queues_read(struct seq_file *s, void *data);
 void mt76_seq_puts_array(struct seq_file *file, const char *str,
 			 s8 *val, int len);
drivers/net/wireless/mediatek/mt76/mt7615/debugfs.c +28 -1
index cb4659771fd97..bda22ca0bd714 100644
@@ -2,6 +2,33 @@
 
 #include "mt7615.h"
 
+static int
+mt7615_reg_set(void *data, u64 val)
+{
+	struct mt7615_dev *dev = data;
+
+	mt7615_mutex_acquire(dev);
+	mt76_wr(dev, dev->mt76.debugfs_reg, val);
+	mt7615_mutex_release(dev);
+
+	return 0;
+}
+
+static int
+mt7615_reg_get(void *data, u64 *val)
+{
+	struct mt7615_dev *dev = data;
+
+	mt7615_mutex_acquire(dev);
+	*val = mt76_rr(dev, dev->mt76.debugfs_reg);
+	mt7615_mutex_release(dev);
+
+	return 0;
+}
+
+DEFINE_DEBUGFS_ATTRIBUTE(fops_regval, mt7615_reg_get, mt7615_reg_set,
+			 "0x%08llx\n");
+
 static int
 mt7615_radar_pattern_set(void *data, u64 val)
 {
@@ -506,7 +533,7 @@ int mt7615_init_debugfs(struct mt7615_dev *dev)
 {
 	struct dentry *dir;
 
-	dir = mt76_register_debugfs(&dev->mt76);
+	dir = mt76_register_debugfs_fops(&dev->mt76, &fops_regval);
 	if (!dir)
 		return -ENOMEM;
 
drivers/net/wireless/mediatek/mt76/mt7921/debugfs.c +27 -1
index 77468bdae460b..4c89c4ac8031a 100644
@@ -4,6 +4,32 @@
 #include "mt7921.h"
 #include "eeprom.h"
 
+static int
+mt7921_reg_set(void *data, u64 val)
+{
+	struct mt7921_dev *dev = data;
+
+	mt7921_mutex_acquire(dev);
+	mt76_wr(dev, dev->mt76.debugfs_reg, val);
+	mt7921_mutex_release(dev);
+
+	return 0;
+}
+
+static int
+mt7921_reg_get(void *data, u64 *val)
+{
+	struct mt7921_dev *dev = data;
+
+	mt7921_mutex_acquire(dev);
+	*val = mt76_rr(dev, dev->mt76.debugfs_reg);
+	mt7921_mutex_release(dev);
+
+	return 0;
+}
+
+DEFINE_DEBUGFS_ATTRIBUTE(fops_regval, mt7921_reg_get, mt7921_reg_set,
+			 "0x%08llx\n");
 static int
 mt7921_fw_debug_set(void *data, u64 val)
 {
@@ -373,7 +399,7 @@ int mt7921_init_debugfs(struct mt7921_dev *dev)
 {
 	struct dentry *dir;
 
-	dir = mt76_register_debugfs(&dev->mt76);
+	dir = mt76_register_debugfs_fops(&dev->mt76, &fops_regval);
 	if (!dir)
 		return -ENOMEM;
 
Signature · 1 line
2.33.0
author maintainer
#483 lore raw

PATCH 472/917 mt76: mt7921: report HE MU radiotap

In reply to Greg Kroah-Hartman

From: Sean Wang <[email protected]>

[ Upstream commit 4fee32153ab62356aeea9d152d8f33a5fd3a0086 ]

Report HE MU/BF radiotap.

That fixed HE MU packets dropped by mac80211 because they are missing the ieee80211_radiotap_he_mu header.

---
 .../net/wireless/mediatek/mt76/mt7921/mac.c   | 65 ++++++++++++++++---
 .../net/wireless/mediatek/mt76/mt7921/mac.h   |  8 +++
 2 files changed, 65 insertions(+), 8 deletions(-)
drivers/net/wireless/mediatek/mt76/mt7921/mac.c +57 -8
index f4714b0f6e5c4..8a16f3f4d5253 100644
@@ -180,12 +180,56 @@ mt7921_mac_decode_he_radiotap_ru(struct mt76_rx_status *status,
 				      IEEE80211_RADIOTAP_HE_DATA2_RU_OFFSET);
 }
 
+static void
+mt7921_mac_decode_he_mu_radiotap(struct sk_buff *skb,
+				 struct mt76_rx_status *status,
+				 __le32 *rxv)
+{
+	static const struct ieee80211_radiotap_he_mu mu_known = {
+		.flags1 = HE_BITS(MU_FLAGS1_SIG_B_MCS_KNOWN) |
+			  HE_BITS(MU_FLAGS1_SIG_B_DCM_KNOWN) |
+			  HE_BITS(MU_FLAGS1_CH1_RU_KNOWN) |
+			  HE_BITS(MU_FLAGS1_SIG_B_SYMS_USERS_KNOWN) |
+			  HE_BITS(MU_FLAGS1_SIG_B_COMP_KNOWN),
+		.flags2 = HE_BITS(MU_FLAGS2_BW_FROM_SIG_A_BW_KNOWN) |
+			  HE_BITS(MU_FLAGS2_PUNC_FROM_SIG_A_BW_KNOWN),
+	};
+	struct ieee80211_radiotap_he_mu *he_mu = NULL;
+
+	he_mu = skb_push(skb, sizeof(mu_known));
+	memcpy(he_mu, &mu_known, sizeof(mu_known));
+
+#define MU_PREP(f, v)	le16_encode_bits(v, IEEE80211_RADIOTAP_HE_MU_##f)
+
+	he_mu->flags1 |= MU_PREP(FLAGS1_SIG_B_MCS, status->rate_idx);
+	if (status->he_dcm)
+		he_mu->flags1 |= MU_PREP(FLAGS1_SIG_B_DCM, status->he_dcm);
+
+	he_mu->flags2 |= MU_PREP(FLAGS2_BW_FROM_SIG_A_BW, status->bw) |
+			 MU_PREP(FLAGS2_SIG_B_SYMS_USERS,
+				 le32_get_bits(rxv[2], MT_CRXV_HE_NUM_USER));
+
+	he_mu->ru_ch1[0] = FIELD_GET(MT_CRXV_HE_RU0, cpu_to_le32(rxv[3]));
+
+	if (status->bw >= RATE_INFO_BW_40) {
+		he_mu->flags1 |= HE_BITS(MU_FLAGS1_CH2_RU_KNOWN);
+		he_mu->ru_ch2[0] =
+			FIELD_GET(MT_CRXV_HE_RU1, cpu_to_le32(rxv[3]));
+	}
+
+	if (status->bw >= RATE_INFO_BW_80) {
+		he_mu->ru_ch1[1] =
+			FIELD_GET(MT_CRXV_HE_RU2, cpu_to_le32(rxv[3]));
+		he_mu->ru_ch2[1] =
+			FIELD_GET(MT_CRXV_HE_RU3, cpu_to_le32(rxv[3]));
+	}
+}
+
 static void
 mt7921_mac_decode_he_radiotap(struct sk_buff *skb,
 			      struct mt76_rx_status *status,
 			      __le32 *rxv, u32 phy)
 {
-	/* TODO: struct ieee80211_radiotap_he_mu */
 	static const struct ieee80211_radiotap_he known = {
 		.data1 = HE_BITS(DATA1_DATA_MCS_KNOWN) |
 			 HE_BITS(DATA1_DATA_DCM_KNOWN) |
@@ -193,6 +237,7 @@ mt7921_mac_decode_he_radiotap(struct sk_buff *skb,
 			 HE_BITS(DATA1_CODING_KNOWN) |
 			 HE_BITS(DATA1_LDPC_XSYMSEG_KNOWN) |
 			 HE_BITS(DATA1_DOPPLER_KNOWN) |
+			 HE_BITS(DATA1_SPTL_REUSE_KNOWN) |
 			 HE_BITS(DATA1_BSS_COLOR_KNOWN),
 		.data2 = HE_BITS(DATA2_GI_KNOWN) |
 			 HE_BITS(DATA2_TXBF_KNOWN) |
@@ -207,9 +252,12 @@ mt7921_mac_decode_he_radiotap(struct sk_buff *skb,
 
 	he->data3 = HE_PREP(DATA3_BSS_COLOR, BSS_COLOR, rxv[14]) |
 		    HE_PREP(DATA3_LDPC_XSYMSEG, LDPC_EXT_SYM, rxv[2]);
+	he->data4 = HE_PREP(DATA4_SU_MU_SPTL_REUSE, SR_MASK, rxv[11]);
 	he->data5 = HE_PREP(DATA5_PE_DISAMBIG, PE_DISAMBIG, rxv[2]) |
 		    le16_encode_bits(ltf_size,
 				     IEEE80211_RADIOTAP_HE_DATA5_LTF_SIZE);
+	if (cpu_to_le32(rxv[0]) & MT_PRXV_TXBF)
+		he->data5 |= HE_BITS(DATA5_TXBF);
 	he->data6 = HE_PREP(DATA6_TXOP, TXOP_DUR, rxv[14]) |
 		    HE_PREP(DATA6_DOPPLER, DOPPLER, rxv[14]);
 
@@ -217,8 +265,7 @@ mt7921_mac_decode_he_radiotap(struct sk_buff *skb,
 	case MT_PHY_TYPE_HE_SU:
 		he->data1 |= HE_BITS(DATA1_FORMAT_SU) |
 			     HE_BITS(DATA1_UL_DL_KNOWN) |
-			     HE_BITS(DATA1_BEAM_CHANGE_KNOWN) |
-			     HE_BITS(DATA1_SPTL_REUSE_KNOWN);
+			     HE_BITS(DATA1_BEAM_CHANGE_KNOWN);
 
 		he->data3 |= HE_PREP(DATA3_BEAM_CHANGE, BEAM_CHNG, rxv[14]) |
 			     HE_PREP(DATA3_UL_DL, UPLINK, rxv[2]);
@@ -232,17 +279,15 @@ mt7921_mac_decode_he_radiotap(struct sk_buff *skb,
 		break;
 	case MT_PHY_TYPE_HE_MU:
 		he->data1 |= HE_BITS(DATA1_FORMAT_MU) |
-			     HE_BITS(DATA1_UL_DL_KNOWN) |
-			     HE_BITS(DATA1_SPTL_REUSE_KNOWN);
+			     HE_BITS(DATA1_UL_DL_KNOWN);
 
 		he->data3 |= HE_PREP(DATA3_UL_DL, UPLINK, rxv[2]);
-		he->data4 |= HE_PREP(DATA4_SU_MU_SPTL_REUSE, SR_MASK, rxv[11]);
+		he->data4 |= HE_PREP(DATA4_MU_STA_ID, MU_AID, rxv[7]);
 
 		mt7921_mac_decode_he_radiotap_ru(status, he, rxv);
 		break;
 	case MT_PHY_TYPE_HE_TB:
 		he->data1 |= HE_BITS(DATA1_FORMAT_TRIG) |
-			     HE_BITS(DATA1_SPTL_REUSE_KNOWN) |
 			     HE_BITS(DATA1_SPTL_REUSE2_KNOWN) |
 			     HE_BITS(DATA1_SPTL_REUSE3_KNOWN) |
 			     HE_BITS(DATA1_SPTL_REUSE4_KNOWN);
@@ -606,9 +651,13 @@ int mt7921_mac_fill_rx(struct mt7921_dev *dev, struct sk_buff *skb)
 
 	mt7921_mac_assoc_rssi(dev, skb);
 
-	if (rxv && status->flag & RX_FLAG_RADIOTAP_HE)
+	if (rxv && status->flag & RX_FLAG_RADIOTAP_HE) {
 		mt7921_mac_decode_he_radiotap(skb, status, rxv, mode);
 
+		if (status->flag & RX_FLAG_RADIOTAP_HE_MU)
+			mt7921_mac_decode_he_mu_radiotap(skb, status, rxv);
+	}
+
 	if (!status->wcid || !ieee80211_is_data_qos(fc))
 		return 0;
 
drivers/net/wireless/mediatek/mt76/mt7921/mac.h +8 -0
index 3af67fac213df..f0194c8780372 100644
@@ -116,6 +116,7 @@ enum rx_pkt_type {
 #define MT_PRXV_TX_DCM			BIT(4)
 #define MT_PRXV_TX_ER_SU_106T		BIT(5)
 #define MT_PRXV_NSTS			GENMASK(9, 7)
+#define MT_PRXV_TXBF			BIT(10)
 #define MT_PRXV_HT_AD_CODE		BIT(11)
 #define MT_PRXV_FRAME_MODE		GENMASK(14, 12)
 #define MT_PRXV_SGI			GENMASK(16, 15)
@@ -138,8 +139,15 @@ enum rx_pkt_type {
 #define MT_CRXV_HE_LTF_SIZE		GENMASK(18, 17)
 #define MT_CRXV_HE_LDPC_EXT_SYM		BIT(20)
 #define MT_CRXV_HE_PE_DISAMBIG		BIT(23)
+#define MT_CRXV_HE_NUM_USER		GENMASK(30, 24)
 #define MT_CRXV_HE_UPLINK		BIT(31)
 
+#define MT_CRXV_HE_RU0			GENMASK(7, 0)
+#define MT_CRXV_HE_RU1			GENMASK(15, 8)
+#define MT_CRXV_HE_RU2			GENMASK(23, 16)
+#define MT_CRXV_HE_RU3			GENMASK(31, 24)
+#define MT_CRXV_HE_MU_AID		GENMASK(30, 20)
+
 #define MT_CRXV_HE_SR_MASK		GENMASK(11, 8)
 #define MT_CRXV_HE_SR1_MASK		GENMASK(16, 12)
 #define MT_CRXV_HE_SR2_MASK             GENMASK(20, 17)
Signature · 1 line
2.33.0
author maintainer
#484 lore raw

PATCH 473/917 mt76: mt7921: fix firmware usage of RA info using legacy rates

In reply to Greg Kroah-Hartman

From: Sean Wang <[email protected]>

[ Upstream commit 99b8e195994d9d77de3bfe0cb403c44a57c2cf79 ]

According to the firmware usage, OFDM rates should fill out bit 6 - 13 while CCK rates should fill out bit 0 - 3 in legacy field of RA info to make the rate adaption runs propertly. Otherwise, a unicast frame might be picking up the unsupported rate to send out.

Fixes 1c099ab44727 ("mt76: mt7921: add MCU support")
---
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c | 11 ++++++++++-
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h |  2 ++
 2 files changed, 12 insertions(+), 1 deletion(-)
drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c +10 -1
index f57f047fce99c..98d233e24afcc 100644
@@ -719,6 +719,7 @@ void mt76_connac_mcu_sta_tlv(struct mt76_phy *mphy, struct sk_buff *skb,
 	struct sta_rec_state *state;
 	struct sta_rec_phy *phy;
 	struct tlv *tlv;
+	u16 supp_rates;
 
 	/* starec ht */
 	if (sta->ht_cap.ht_supported) {
@@ -767,7 +768,15 @@ void mt76_connac_mcu_sta_tlv(struct mt76_phy *mphy, struct sk_buff *skb,
 
 	tlv = mt76_connac_mcu_add_tlv(skb, STA_REC_RA, sizeof(*ra_info));
 	ra_info = (struct sta_rec_ra_info *)tlv;
-	ra_info->legacy = cpu_to_le16((u16)sta->supp_rates[band]);
+
+	supp_rates = sta->supp_rates[band];
+	if (band == NL80211_BAND_2GHZ)
+		supp_rates = FIELD_PREP(RA_LEGACY_OFDM, supp_rates >> 4) |
+			     FIELD_PREP(RA_LEGACY_CCK, supp_rates & 0xf);
+	else
+		supp_rates = FIELD_PREP(RA_LEGACY_OFDM, supp_rates);
+
+	ra_info->legacy = cpu_to_le16(supp_rates);
 
 	if (sta->ht_cap.ht_supported)
 		memcpy(ra_info->rx_mcs_bitmask, sta->ht_cap.mcs.rx_mask,
drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.h +2 -0
index 4bcd728ff97c5..77d4435e4581e 100644
@@ -124,6 +124,8 @@ struct sta_rec_state {
 	u8 rsv[1];
 } __packed;
 
+#define RA_LEGACY_OFDM GENMASK(13, 6)
+#define RA_LEGACY_CCK  GENMASK(3, 0)
 #define HT_MCS_MASK_NUM 10
 struct sta_rec_ra_info {
 	__le16 tag;
Signature · 1 line
2.33.0
author maintainer
#485 lore raw

PATCH 474/917 mt76: mt7921: fix kernel warning from cfg80211_calculate_bitrate

In reply to Greg Kroah-Hartman

From: Sean Wang <[email protected]>

[ Upstream commit 8e695328a1006b7bab2d972e7d0111fa6e6faf51 ]

Fix the kernel warning from cfg80211_calculate_bitrate due to the legacy rate is not parsed well in the current driver.

Also, zeros struct rate_info before we fill it out to avoid the old value is kept such as rate->legacy.

[  790.921560] WARNING: CPU: 7 PID: 970 at net/wireless/util.c:1298 cfg80211_calculate_bitrate+0x354/0x35c [cfg80211]
[  790.987738] Hardware name: MediaTek Asurada rev1 board (DT)
[  790.993298] pstate: a0400009 (NzCv daif +PAN -UAO)
[  790.998104] pc : cfg80211_calculate_bitrate+0x354/0x35c [cfg80211]
[  791.004295] lr : cfg80211_calculate_bitrate+0x180/0x35c [cfg80211]
[  791.010462] sp : ffffffc0129c3880
[  791.013765] x29: ffffffc0129c3880 x28: ffffffd38305bea8
[  791.019065] x27: ffffffc0129c3970 x26: 0000000000000013
[  791.024364] x25: 00000000000003ca x24: 000000000000002f
[  791.029664] x23: 00000000000000d0 x22: ffffff8d108bc000
[  791.034964] x21: ffffff8d108bc0d0 x20: ffffffc0129c39a8
[  791.040264] x19: ffffffc0129c39a8 x18: 00000000ffff0a10
[  791.045563] x17: 0000000000000050 x16: 00000000000000ec
[  791.050910] x15: ffffffd3f9ebed9c x14: 0000000000000006
[  791.056211] x13: 00000000000b2eea x12: 0000000000000000
[  791.061511] x11: 00000000ffffffff x10: 0000000000000000
[  791.066811] x9 : 0000000000000000 x8 : 0000000000000000
[  791.072110] x7 : 0000000000000000 x6 : ffffffd3fafa5a7b
[  791.077409] x5 : 0000000000000000 x4 : 0000000000000000
[  791.082708] x3 : 0000000000000000 x2 : 0000000000000000
[  791.088008] x1 : ffffff8d3f79c918 x0 : 0000000000000000
[  791.093308] Call trace:
[  791.095770]  cfg80211_calculate_bitrate+0x354/0x35c [cfg80211]
[  791.101615]  nl80211_put_sta_rate+0x6c/0x2c0 [cfg80211]
[  791.106853]  nl80211_send_station+0x980/0xaa4 [cfg80211]
[  791.112178]  nl80211_get_station+0xb4/0x134 [cfg80211]
[  791.117308]  genl_rcv_msg+0x3a0/0x440
[  791.120960]  netlink_rcv_skb+0xcc/0x118
[  791.124785]  genl_rcv+0x34/0x48
[  791.127916]  netlink_unicast+0x144/0x1dc
Fixes 1c099ab44727 ("mt76: mt7921: add MCU support")
---
 drivers/net/wireless/mediatek/mt76/mt7921/mcu.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)
drivers/net/wireless/mediatek/mt76/mt7921/mcu.c +9 -2
index 8ced55501d373..3cb53c642d242 100644
@@ -320,11 +320,13 @@ mt7921_mcu_tx_rate_parse(struct mt76_phy *mphy,
 			 struct rate_info *rate, u16 r)
 {
 	struct ieee80211_supported_band *sband;
-	u16 flags = 0;
+	u16 flags = 0, rate_idx;
 	u8 txmode = FIELD_GET(MT_WTBL_RATE_TX_MODE, r);
 	u8 gi = 0;
 	u8 bw = 0;
+	bool cck = false;
 
+	memset(rate, 0, sizeof(*rate));
 	rate->mcs = FIELD_GET(MT_WTBL_RATE_MCS, r);
 	rate->nss = FIELD_GET(MT_WTBL_RATE_NSS, r) + 1;
 
@@ -349,13 +351,18 @@ mt7921_mcu_tx_rate_parse(struct mt76_phy *mphy,
 
 	switch (txmode) {
 	case MT_PHY_TYPE_CCK:
+		cck = true;
+		fallthrough;
 	case MT_PHY_TYPE_OFDM:
 		if (mphy->chandef.chan->band == NL80211_BAND_5GHZ)
 			sband = &mphy->sband_5g.sband;
 		else
 			sband = &mphy->sband_2g.sband;
 
-		rate->legacy = sband->bitrates[rate->mcs].bitrate;
+		rate_idx = FIELD_GET(MT_TX_RATE_IDX, r);
+		rate_idx = mt76_get_rate(mphy->dev, sband, rate_idx,
+					 cck);
+		rate->legacy = sband->bitrates[rate_idx].bitrate;
 		break;
 	case MT_PHY_TYPE_HT:
 	case MT_PHY_TYPE_HT_GF:
Signature · 1 line
2.33.0
author maintainer
#486 lore raw

PATCH 475/917 mt76: mt7921: always wake device if necessary in debugfs

In reply to Greg Kroah-Hartman

From: Lorenzo Bianconi <[email protected]>

[ Upstream commit 569008744178b672ea3ad9047fa3098f1b73ca55 ]

Add missing device wakeup in debugfs code if we are accessing chip registers.

Fixes 1d8efc741df8 ("mt76: mt7921: introduce Runtime PM support")
---
 drivers/net/wireless/mediatek/mt76/mt7921/debugfs.c | 8 ++++++++
 1 file changed, 8 insertions(+)
drivers/net/wireless/mediatek/mt76/mt7921/debugfs.c +8 -0
index 4c89c4ac8031a..30f3b3085c786 100644
@@ -95,6 +95,8 @@ mt7921_tx_stats_show(struct seq_file *file, void *data)
 	struct mt7921_dev *dev = file->private;
 	int stat[8], i, n;
 
+	mt7921_mutex_acquire(dev);
+
 	mt7921_ampdu_stat_read_phy(&dev->phy, file);
 
 	/* Tx amsdu info */
@@ -104,6 +106,8 @@ mt7921_tx_stats_show(struct seq_file *file, void *data)
 		n += stat[i];
 	}
 
+	mt7921_mutex_release(dev);
+
 	for (i = 0; i < ARRAY_SIZE(stat); i++) {
 		seq_printf(file, "AMSDU pack count of %d MSDU in TXD: 0x%x ",
 			   i + 1, stat[i]);
@@ -124,6 +128,8 @@ mt7921_queues_acq(struct seq_file *s, void *data)
 	struct mt7921_dev *dev = dev_get_drvdata(s->private);
 	int i;
 
+	mt7921_mutex_acquire(dev);
+
 	for (i = 0; i < 16; i++) {
 		int j, acs = i / 4, index = i % 4;
 		u32 ctrl, val, qlen = 0;
@@ -143,6 +149,8 @@ mt7921_queues_acq(struct seq_file *s, void *data)
 		seq_printf(s, "AC%d%d: queued=%d\n", acs, index, qlen);
 	}
 
+	mt7921_mutex_release(dev);
+
 	return 0;
 }
 
Signature · 1 line
2.33.0
author maintainer
#487 lore raw

PATCH 476/917 mt76: mt7915: fix hwmon temp sensor mem use-after-free

In reply to Greg Kroah-Hartman

From: Ben Greear <[email protected]>

[ Upstream commit 0ae3ff5684514d72357240f1033a7494c51f93ed ]

Without this change, garbage is seen in the hwmon name and sensors output for mt7915 is garbled. It appears that the hwmon logic does not make a copy of the incoming string, but instead just copies a char* and expects it to never go away.

Fixes 33fe9c639c13 ("mt76: mt7915: add thermal sensor device support")
---
 drivers/net/wireless/mediatek/mt76/mt7915/init.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
drivers/net/wireless/mediatek/mt76/mt7915/init.c +6 -4
index 4798d6344305d..b171027e0cfa8 100644
@@ -130,9 +130,12 @@ static int mt7915_thermal_init(struct mt7915_phy *phy)
 	struct wiphy *wiphy = phy->mt76->hw->wiphy;
 	struct thermal_cooling_device *cdev;
 	struct device *hwmon;
+	const char *name;
 
-	cdev = thermal_cooling_device_register(wiphy_name(wiphy), phy,
-					       &mt7915_thermal_ops);
+	name = devm_kasprintf(&wiphy->dev, GFP_KERNEL, "mt7915_%s",
+			      wiphy_name(wiphy));
+
+	cdev = thermal_cooling_device_register(name, phy, &mt7915_thermal_ops);
 	if (!IS_ERR(cdev)) {
 		if (sysfs_create_link(&wiphy->dev.kobj, &cdev->device.kobj,
 				      "cooling_device") < 0)
@@ -144,8 +147,7 @@ static int mt7915_thermal_init(struct mt7915_phy *phy)
 	if (!IS_REACHABLE(CONFIG_HWMON))
 		return 0;
 
-	hwmon = devm_hwmon_device_register_with_groups(&wiphy->dev,
-						       wiphy_name(wiphy), phy,
+	hwmon = devm_hwmon_device_register_with_groups(&wiphy->dev, name, phy,
 						       mt7915_hwmon_groups);
 	if (IS_ERR(hwmon))
 		return PTR_ERR(hwmon);
Signature · 1 line
2.33.0
author maintainer
#488 lore raw

PATCH 477/917 mt76: mt7615: fix hwmon temp sensor mem use-after-free

In reply to Greg Kroah-Hartman

From: Ryder Lee <[email protected]>

[ Upstream commit 0bb4e9187ea4a59dc6658a62978deda0c0dc4b28 ]

Without this change, garbage is seen in the hwmon name and sensors output for mt7615 is garbled.

Fixes 109e505ad944 ("mt76: mt7615: add thermal sensor device support")
---
 drivers/net/wireless/mediatek/mt76/mt7615/init.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/net/wireless/mediatek/mt76/mt7615/init.c +4 -2
index 2f1ac644e018e..47f23ac905a3c 100644
@@ -49,12 +49,14 @@ int mt7615_thermal_init(struct mt7615_dev *dev)
 {
 	struct wiphy *wiphy = mt76_hw(dev)->wiphy;
 	struct device *hwmon;
+	const char *name;
 
 	if (!IS_REACHABLE(CONFIG_HWMON))
 		return 0;
 
-	hwmon = devm_hwmon_device_register_with_groups(&wiphy->dev,
-						       wiphy_name(wiphy), dev,
+	name = devm_kasprintf(&wiphy->dev, GFP_KERNEL, "mt7615_%s",
+			      wiphy_name(wiphy));
+	hwmon = devm_hwmon_device_register_with_groups(&wiphy->dev, name, dev,
 						       mt7615_hwmon_groups);
 	if (IS_ERR(hwmon))
 		return PTR_ERR(hwmon);
Signature · 1 line
2.33.0
author maintainer
#489 lore raw

PATCH 478/917 mt76: mt7915: fix possible infinite loop release semaphore

In reply to Greg Kroah-Hartman

From: Lorenzo Bianconi <[email protected]>

[ Upstream commit e500c9470e26be66eb2bc6de773ae9091149118a ]

Fix possible infinite loop in mt7915_load_patch if mt7915_mcu_patch_sem_ctrl always returns an error.

Fixes e57b7901469fc ("mt76: add mac80211 driver for MT7915 PCIe-based chipsets")
---
 drivers/net/wireless/mediatek/mt76/mt7915/mcu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/mediatek/mt76/mt7915/mcu.c +1 -1
index e7e396f58c92c..85c9c08ee2a82 100644
@@ -2790,7 +2790,7 @@ out:
 	default:
 		ret = -EAGAIN;
 		dev_err(dev->mt76.dev, "Failed to release patch semaphore\n");
-		goto out;
+		break;
 	}
 	release_firmware(fw);
 
Signature · 1 line
2.33.0
author maintainer
#490 lore raw

PATCH 479/917 mt76: mt7921: fix retrying release semaphore without end

In reply to Greg Kroah-Hartman

From: Sean Wang <[email protected]>

[ Upstream commit 02d1c7d494d8052288bc175e4ff54b56d08a3c5f ]

We should pass the error code to the caller immediately to avoid the possible infinite retry to release the semaphore.

Fixes 1c099ab44727 ("mt76: mt7921: add MCU support")
---
 drivers/net/wireless/mediatek/mt76/mt7921/mcu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/mediatek/mt76/mt7921/mcu.c +1 -1
index 3cb53c642d242..506a1909ce6d5 100644
@@ -827,7 +827,7 @@ out:
 	default:
 		ret = -EAGAIN;
 		dev_err(dev->mt76.dev, "Failed to release patch semaphore\n");
-		goto out;
+		break;
 	}
 	release_firmware(fw);
 
Signature · 1 line
2.33.0
author maintainer
#491 lore raw

PATCH 480/917 mt76: mt7615: fix monitor mode tear down crash

In reply to Greg Kroah-Hartman

From: Ryder Lee <[email protected]>

[ Upstream commit a6fdbdd1ac2996a58a84672ef37efb5cbb98fadf ]

[  103.451600] CPU 3 Unable to handle kernel paging request at virtual address 00000003, epc == 8576591c, ra == 857659f0
[  103.462226] Oops[#1]:
[  103.464499] CPU: 3 PID: 9247 Comm: ifconfig Tainted: G        W         5.4.143 #0
[  103.472031] $ 0   : 00000000 00000001 83be3854 00000000
[  103.477239] $ 4   : 8102a374 8102a374 8102f0b0 00000200
[  103.482444] $ 8   : 0000002d 000001e4 64373765 5d206337
[  103.487647] $12   : 00000000 00000005 00000000 0006d1df
[  103.492853] $16   : 83be3848 853838a8 8743d600 00010000
[  103.498059] $20   : 00000000 00000000 8553dec0 0000007f
[  103.503266] $24   : 00000003 80382084
[  103.508472] $28   : 831d4000 831d5bc0 00000001 857659f0
[  103.513678] Hi    : 00000122
[  103.516543] Lo    : d1768000
[  103.519452] epc   : 8576591c mt7615_mcu_add_bss+0xd0/0x3c0 [mt7615_common]
[  103.526306] ra    : 857659f0 mt7615_mcu_add_bss+0x1a4/0x3c0 [mt7615_common]
[  103.533232] Status: 11007c03 KERNEL EXL IE
[  103.537402] Cause : 40800008 (ExcCode 02)
[  103.541389] BadVA : 00000003
[  103.544253] PrId  : 0001992f (MIPS 1004Kc)
[  103.797086] Call Trace:
[  103.799562] [<8576591c>] mt7615_mcu_add_bss+0xd0/0x3c0 [mt7615_common]
[  103.806082] [<85760a14>] mt7615_remove_interface+0x74/0x1e0 [mt7615_common]
[  103.813280] [<85603fcc>] drv_remove_interface+0x2c/0xa0 [mac80211]
[  103.819612] [<8561a8e4>] ieee80211_del_virtual_monitor.part.22+0x74/0xe8 [mac80211]
[  103.827410] [<8561b7f0>] ieee80211_do_stop+0x4a4/0x8a0 [mac80211]
[  103.833671] [<8561bc00>] ieee80211_stop+0x14/0x24 [mac80211]
[  103.839405] [<8045a328>] __dev_close_many+0x9c/0x10c
[  103.844364] [<80463de4>] __dev_change_flags+0x16c/0x1e4
[  103.849569] [<80463e84>] dev_change_flags+0x28/0x70
[  103.854440] [<80521e54>] devinet_ioctl+0x280/0x774
[  103.859222] [<80526248>] inet_ioctl+0xa4/0x1c8
[  103.863674] [<80436830>] sock_ioctl+0x2d8/0x4bc
[  103.868201] [<801adbb4>] do_vfs_ioctl+0xb8/0x7c0
[  103.872804] [<801ae30c>] ksys_ioctl+0x50/0xb4
[  103.877156] [<80014598>] syscall_common+0x34/0x58
Fixes 04b8e65922f63 ("mt76: add mac80211 driver for MT7615 PCIe-based chipsets")
---
 .../net/wireless/mediatek/mt76/mt7615/mcu.c    | 18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)
drivers/net/wireless/mediatek/mt76/mt7615/mcu.c +13 -5
index f8a09692d3e4c..4fed3afad67cc 100644
@@ -808,7 +808,8 @@ mt7615_mcu_ctrl_pm_state(struct mt7615_dev *dev, int band, int state)
 
 static int
 mt7615_mcu_bss_basic_tlv(struct sk_buff *skb, struct ieee80211_vif *vif,
-			 struct ieee80211_sta *sta, bool enable)
+			 struct ieee80211_sta *sta, struct mt7615_phy *phy,
+			 bool enable)
 {
 	struct mt7615_vif *mvif = (struct mt7615_vif *)vif->drv_priv;
 	u32 type = vif->p2p ? NETWORK_P2P : NETWORK_INFRA;
@@ -821,6 +822,7 @@ mt7615_mcu_bss_basic_tlv(struct sk_buff *skb, struct ieee80211_vif *vif,
 	switch (vif->type) {
 	case NL80211_IFTYPE_MESH_POINT:
 	case NL80211_IFTYPE_AP:
+	case NL80211_IFTYPE_MONITOR:
 		break;
 	case NL80211_IFTYPE_STATION:
 		/* TODO: enable BSS_INFO_UAPSD & BSS_INFO_PM */
@@ -840,14 +842,19 @@ mt7615_mcu_bss_basic_tlv(struct sk_buff *skb, struct ieee80211_vif *vif,
 	}
 
 	bss = (struct bss_info_basic *)tlv;
-	memcpy(bss->bssid, vif->bss_conf.bssid, ETH_ALEN);
-	bss->bcn_interval = cpu_to_le16(vif->bss_conf.beacon_int);
 	bss->network_type = cpu_to_le32(type);
-	bss->dtim_period = vif->bss_conf.dtim_period;
 	bss->bmc_tx_wlan_idx = wlan_idx;
 	bss->wmm_idx = mvif->mt76.wmm_idx;
 	bss->active = enable;
 
+	if (vif->type != NL80211_IFTYPE_MONITOR) {
+		memcpy(bss->bssid, vif->bss_conf.bssid, ETH_ALEN);
+		bss->bcn_interval = cpu_to_le16(vif->bss_conf.beacon_int);
+		bss->dtim_period = vif->bss_conf.dtim_period;
+	} else {
+		memcpy(bss->bssid, phy->mt76->macaddr, ETH_ALEN);
+	}
+
 	return 0;
 }
 
@@ -863,6 +870,7 @@ mt7615_mcu_bss_omac_tlv(struct sk_buff *skb, struct ieee80211_vif *vif)
 	tlv = mt76_connac_mcu_add_tlv(skb, BSS_INFO_OMAC, sizeof(*omac));
 
 	switch (vif->type) {
+	case NL80211_IFTYPE_MONITOR:
 	case NL80211_IFTYPE_MESH_POINT:
 	case NL80211_IFTYPE_AP:
 		if (vif->p2p)
@@ -929,7 +937,7 @@ mt7615_mcu_add_bss(struct mt7615_phy *phy, struct ieee80211_vif *vif,
 	if (enable)
 		mt7615_mcu_bss_omac_tlv(skb, vif);
 
-	mt7615_mcu_bss_basic_tlv(skb, vif, sta, enable);
+	mt7615_mcu_bss_basic_tlv(skb, vif, sta, phy, enable);
 
 	if (enable && mvif->mt76.omac_idx >= EXT_BSSID_START &&
 	    mvif->mt76.omac_idx < REPEATER_BSSID_START)
Signature · 1 line
2.33.0
author maintainer
#492 lore raw

PATCH 481/917 mt76: connac: fix possible NULL pointer dereference in mt76_connac_get_phy_mode_v2

In reply to Greg Kroah-Hartman

From: Lorenzo Bianconi <[email protected]>

[ Upstream commit b5f2ba8a4c794e8349c0e30036352b9f685164c4 ]

Fix the following NULL pointer dereference in mt76_connac_get_phy_mode_v2 routine triggered on mt7663s device when sta is NULL

[    5.490700] mt7663s mmc0:0001:1: N9 Firmware Version: 3.1.1, Build Time: 20200604161656
[    5.490815] mt7663s mmc0:0001:1: Region number: 0x4
[    5.490868] mt7663s mmc0:0001:1: Parsing tailer Region: 0
[    5.496251] mt7663s mmc0:0001:1: Region 0, override_addr = 0x00118000
[    5.496419] mt7663s mmc0:0001:1: Parsing tailer Region: 1
[    5.624027] mt7663s mmc0:0001:1: Parsing tailer Region: 2
[    5.656999] mt7663s mmc0:0001:1: Parsing tailer Region: 3
[    5.671876] mt7663s mmc0:0001:1: override_addr = 0x00118000, option = 3
[    9.358658] BUG: kernel NULL pointer dereference, address: 0000000000000000
[    9.358775] #PF: supervisor read access in kernel mode
[    9.358831] #PF: error_code(0x0000) - not-present page
[    9.358886] PGD 0 P4D 0
[    9.358917] Oops: 0000 [#1] SMP
[    9.358960] CPU: 0 PID: 235 Comm: NetworkManager Not tainted 5.15.0-rc4-kvm-02151-g39e333d657f4-dirty #769
[    9.359057] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-4.fc34 04/01/2014
[    9.359150] RIP: 0010:mt76_connac_get_phy_mode_v2+0xc9/0x11c
[    9.359473] RAX: 0000000000000013 RBX: 0000000000000000 RCX: 0000000000000027
[    9.359546] RDX: ffff8881f9c17358 RSI: 0000000000000001 RDI: ffff8881f9c17350
[    9.359624] RBP: ffff88810bac1ed4 R08: ffffffff822a4a48 R09: 0000000000000003
[    9.359697] R10: ffffffff82234a60 R11: ffffffff82234a60 R12: ffff88810bac1eec
[    9.359779] R13: 0000000000000000 R14: ffff88810bad1648 R15: ffff88810bac1eb8
[    9.359859] FS:  00007f5f1e45bbc0(0000) GS:ffff8881f9c00000(0000) knlGS:0000000000000000
[    9.359939] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[    9.360003] CR2: 0000000000000000 CR3: 0000000105d5d000 CR4: 00000000000006b0
[    9.360083] Call Trace:
[    9.360116]  mt76_connac_mcu_uni_add_bss.cold+0x21/0x250
[    9.360175]  ? schedule_preempt_disabled+0xa/0x10
[    9.360232]  ? __mutex_lock.constprop.0+0x2ab/0x460
[    9.360286]  mt7615_remove_interface+0x63/0x1d0
[    9.360342]  drv_remove_interface+0x32/0xe0
[    9.360385]  ieee80211_do_stop+0x5da/0x800
[    9.360428]  ? dev_reset_queue+0x30/0x90
[    9.360472]  ieee80211_stop+0x3b/0xb0
[    9.360516]  __dev_close_many+0x7a/0xd0
[    9.360559]  __dev_change_flags+0xd6/0x1f0
[    9.360604]  dev_change_flags+0x21/0x60
[    9.360648]  do_setlink+0x259/0xfb0
[    9.360686]  ? __nla_validate_parse+0x51/0xb80
[    9.360742]  __rtnl_newlink+0x5b3/0x960
[    9.360785]  ? inet6_fill_ifla6_attrs+0x41d/0x470
[    9.360841]  ? __kmalloc_track_caller+0x57/0x3c0
[    9.360905]  ? netlink_trim+0x8a/0xb0
[    9.360949]  ? skb_queue_tail+0x1b/0x50
Fixes 67aa27431c7f8 ("mt76: mt7921: rely on mt76_connac_mcu common library")
---
 drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/wireless/mediatek/mt76/mt76_connac_mcu.c +2 -2
index 98d233e24afcc..d25b50e769328 100644
@@ -689,7 +689,7 @@ mt76_connac_get_phy_mode_v2(struct mt76_phy *mphy, struct ieee80211_vif *vif,
 		if (ht_cap->ht_supported)
 			mode |= PHY_TYPE_BIT_HT;
 
-		if (he_cap->has_he)
+		if (he_cap && he_cap->has_he)
 			mode |= PHY_TYPE_BIT_HE;
 	} else if (band == NL80211_BAND_5GHZ) {
 		mode |= PHY_TYPE_BIT_OFDM;
@@ -700,7 +700,7 @@ mt76_connac_get_phy_mode_v2(struct mt76_phy *mphy, struct ieee80211_vif *vif,
 		if (vht_cap->vht_supported)
 			mode |= PHY_TYPE_BIT_VHT;
 
-		if (he_cap->has_he)
+		if (he_cap && he_cap->has_he)
 			mode |= PHY_TYPE_BIT_HE;
 	}
 
Signature · 1 line
2.33.0
author maintainer
#493 lore raw

PATCH 482/917 mt76: mt7915: fix sta_rec_wtbl tag len

In reply to Greg Kroah-Hartman

From: Shayne Chen <[email protected]>

[ Upstream commit afa0370f3a3a64af6d368da0bedd72ab2a026cd0 ]

Fix tag len error for sta_rec_wtbl, which causes fw parsing error for the tags placed behind it.

Fixes e57b7901469f ("mt76: add mac80211 driver for MT7915 PCIe-based chipsets")
---
 drivers/net/wireless/mediatek/mt76/mt7915/mcu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/mediatek/mt76/mt7915/mcu.c +1 -1
index 85c9c08ee2a82..6dfe3716a63a5 100644
@@ -757,7 +757,7 @@ mt7915_mcu_alloc_wtbl_req(struct mt7915_dev *dev, struct mt7915_sta *msta,
 	}
 
 	if (sta_hdr)
-		sta_hdr->len = cpu_to_le16(sizeof(hdr));
+		le16_add_cpu(&sta_hdr->len, sizeof(hdr));
 
 	return skb_put_data(nskb, &hdr, sizeof(hdr));
 }
Signature · 1 line
2.33.0
author maintainer
#494 lore raw

PATCH 483/917 mt76: mt7915: fix muar_idx in mt7915_mcu_alloc_sta_req()

In reply to Greg Kroah-Hartman

From: Shayne Chen <[email protected]>

[ Upstream commit 161cc13912d3c3e8857001988dfba39be842454a ]

For broadcast/multicast wcid, the muar_idx should be 0xe.

Fixes e57b7901469f ("mt76: add mac80211 driver for MT7915 PCIe-based chipsets")
---
 drivers/net/wireless/mediatek/mt76/mt7915/mcu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/mediatek/mt76/mt7915/mcu.c +1 -1
index 6dfe3716a63a5..ba36d3caec8e1 100644
@@ -721,7 +721,7 @@ mt7915_mcu_alloc_sta_req(struct mt7915_dev *dev, struct mt7915_vif *mvif,
 		.bss_idx = mvif->idx,
 		.wlan_idx_lo = msta ? to_wcid_lo(msta->wcid.idx) : 0,
 		.wlan_idx_hi = msta ? to_wcid_hi(msta->wcid.idx) : 0,
-		.muar_idx = msta ? mvif->omac_idx : 0,
+		.muar_idx = msta && msta->wcid.sta ? mvif->omac_idx : 0xe,
 		.is_tlv_append = 1,
 	};
 	struct sk_buff *skb;
Signature · 1 line
2.33.0
author maintainer
#495 lore raw

PATCH 484/917 rsi: stop thread firstly in rsi_91x_init() error handling

In reply to Greg Kroah-Hartman

From: Ziyang Xuan <[email protected]>

[ Upstream commit 515e7184bdf0a3ebf1757cc77fb046b4fe282189 ]

When fail to init coex module, free 'common' and 'adapter' directly, but common->tx_thread which will access 'common' and 'adapter' is running at the same time. That will trigger the UAF bug.

==================================================================

BUG: KASAN: use-after-free in rsi_tx_scheduler_thread+0x50f/0x520 [rsi_91x]
Read of size 8 at addr ffff8880076dc000 by task Tx-Thread/124777
CPU: 0 PID: 124777 Comm: Tx-Thread Not tainted 5.15.0-rc5+ #19
Call Trace:
 dump_stack_lvl+0xe2/0x152
 print_address_description.constprop.0+0x21/0x140
 ? rsi_tx_scheduler_thread+0x50f/0x520
 kasan_report.cold+0x7f/0x11b
 ? rsi_tx_scheduler_thread+0x50f/0x520
 rsi_tx_scheduler_thread+0x50f/0x520

...

Freed by task 111873:
 kasan_save_stack+0x1b/0x40
 kasan_set_track+0x1c/0x30
 kasan_set_free_info+0x20/0x30
 __kasan_slab_free+0x109/0x140
 kfree+0x117/0x4c0
 rsi_91x_init+0x741/0x8a0 [rsi_91x]
 rsi_probe+0x9f/0x1750 [rsi_usb]

Stop thread before free 'common' and 'adapter' to fix it.

---
 drivers/net/wireless/rsi/rsi_91x_main.c | 1 +
 1 file changed, 1 insertion(+)
drivers/net/wireless/rsi/rsi_91x_main.c +1 -0
index 143224a3802ba..f1bf71e6c6081 100644
@@ -369,6 +369,7 @@ struct rsi_hw *rsi_91x_init(u16 oper_mode)
 	if (common->coex_mode > 1) {
 		if (rsi_coex_attach(common)) {
 			rsi_dbg(ERR_ZONE, "Failed to init coex module\n");
+			rsi_kill_thread(&common->tx_thread);
 			goto err;
 		}
 	}
Signature · 1 line
2.33.0
author maintainer
#496 lore raw

PATCH 485/917 mwifiex: Send DELBA requests according to spec

In reply to Greg Kroah-Hartman

From: Jonas Dreßler <[email protected]>

[ Upstream commit cc8a8bc37466f79b24d972555237f3d591150602 ]

While looking at on-air packets using Wireshark, I noticed we're never setting the initiator bit when sending DELBA requests to the AP: While we set the bit on our del_ba_param_set bitmask, we forget to actually copy that bitmask over to the command struct, which means we never actually set the initiator bit.

Fix that and copy the bitmask over to the host_cmd_ds_11n_delba command struct.

Fixes 5e6e3a92b9a4 ("wireless: mwifiex: initial commit for Marvell mwifiex driver")
---
 drivers/net/wireless/marvell/mwifiex/11n.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/net/wireless/marvell/mwifiex/11n.c +3 -2
index 6696bce561786..cf08a4af84d6d 100644
@@ -657,14 +657,15 @@ int mwifiex_send_delba(struct mwifiex_private *priv, int tid, u8 *peer_mac,
 	uint16_t del_ba_param_set;
 
 	memset(&delba, 0, sizeof(delba));
-	delba.del_ba_param_set = cpu_to_le16(tid << DELBA_TID_POS);
 
-	del_ba_param_set = le16_to_cpu(delba.del_ba_param_set);
+	del_ba_param_set = tid << DELBA_TID_POS;
+
 	if (initiator)
 		del_ba_param_set |= IEEE80211_DELBA_PARAM_INITIATOR_MASK;
 	else
 		del_ba_param_set &= ~IEEE80211_DELBA_PARAM_INITIATOR_MASK;
 
+	delba.del_ba_param_set = cpu_to_le16(del_ba_param_set);
 	memcpy(&delba.peer_mac_addr, peer_mac, ETH_ALEN);
 
 	/* We don't wait for the response of this command */
Signature · 1 line
2.33.0
author maintainer
#497 lore raw

PATCH 486/917 iwlwifi: mvm: reset PM state on unsuccessful resume

In reply to Greg Kroah-Hartman

From: Johannes Berg <[email protected]>

[ Upstream commit 2f629a7772e2a7bdaff25178917a40073f79702c ]

If resume fails for some reason, we need to set the PM state back to normal so we're able to send commands during firmware reset, rather than failing all of them because we're in D3.

---
 drivers/net/wireless/intel/iwlwifi/mvm/d3.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/net/wireless/intel/iwlwifi/mvm/d3.c +4 -1
index 9f706fffb5922..d3013a51a5096 100644
@@ -2336,7 +2336,6 @@ static int __iwl_mvm_resume(struct iwl_mvm *mvm, bool test)
 		iwl_fw_dbg_collect_desc(&mvm->fwrt, &iwl_dump_desc_assert,
 					false, 0);
 		ret = 1;
-		mvm->trans->system_pm_mode = IWL_PLAT_PM_MODE_DISABLED;
 		goto err;
 	}
 
@@ -2385,6 +2384,7 @@ static int __iwl_mvm_resume(struct iwl_mvm *mvm, bool test)
 		}
 	}
 
+	/* after the successful handshake, we're out of D3 */
 	mvm->trans->system_pm_mode = IWL_PLAT_PM_MODE_DISABLED;
 
 	/*
@@ -2455,6 +2455,9 @@ out:
 	 */
 	set_bit(IWL_MVM_STATUS_HW_RESTART_REQUESTED, &mvm->status);
 
+	/* regardless of what happened, we're now out of D3 */
+	mvm->trans->system_pm_mode = IWL_PLAT_PM_MODE_DISABLED;
+
 	return 1;
 }
 
Signature · 1 line
2.33.0
author maintainer
#498 lore raw

PATCH 487/917 iwlwifi: pnvm: dont kmemdup() more than we have

In reply to Greg Kroah-Hartman

From: Johannes Berg <[email protected]>

[ Upstream commit 0f892441d8c353144e3669b7991fa5fe0bd353e9 ]

We shouldn't kmemdup() more data than we have, that might cause the code to crash. Fix that by updating the length before the kmemdup.

---
 drivers/net/wireless/intel/iwlwifi/fw/pnvm.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)
drivers/net/wireless/intel/iwlwifi/fw/pnvm.c +3 -4
index dde22bdc87039..9b0eee53488ab 100644
@@ -284,16 +284,15 @@ int iwl_pnvm_load(struct iwl_trans *trans,
 	/* First attempt to get the PNVM from BIOS */
 	package = iwl_uefi_get_pnvm(trans, &len);
 	if (!IS_ERR_OR_NULL(package)) {
+		/* we need only the data */
+		len -= sizeof(*package);
 		data = kmemdup(package->data, len, GFP_KERNEL);
 
 		/* free package regardless of whether kmemdup succeeded */
 		kfree(package);
 
-		if (data) {
-			/* we need only the data size */
-			len -= sizeof(*package);
+		if (data)
 			goto parse;
-		}
 	}
 
 	/* If it's not available, try from the filesystem */
Signature · 1 line
2.33.0
author maintainer
#499 lore raw

PATCH 488/917 iwlwifi: pnvm: read EFI data only if long enough

In reply to Greg Kroah-Hartman

From: Johannes Berg <[email protected]>

[ Upstream commit e864a77f51d0d8113b49cf7d030bc9dc911c8176 ]

If the data we get from EFI is not even long enough for the package struct we expect then ignore it entirely.

---
 drivers/net/wireless/intel/iwlwifi/fw/pnvm.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)
drivers/net/wireless/intel/iwlwifi/fw/pnvm.c +7 -3
index 9b0eee53488ab..069fcbc46d2ba 100644
@@ -284,9 +284,13 @@ int iwl_pnvm_load(struct iwl_trans *trans,
 	/* First attempt to get the PNVM from BIOS */
 	package = iwl_uefi_get_pnvm(trans, &len);
 	if (!IS_ERR_OR_NULL(package)) {
-		/* we need only the data */
-		len -= sizeof(*package);
-		data = kmemdup(package->data, len, GFP_KERNEL);
+		if (len >= sizeof(*package)) {
+			/* we need only the data */
+			len -= sizeof(*package);
+			data = kmemdup(package->data, len, GFP_KERNEL);
+		} else {
+			data = NULL;
+		}
 
 		/* free package regardless of whether kmemdup succeeded */
 		kfree(package);
Signature · 1 line
2.33.0
author maintainer
#500 lore raw

PATCH 489/917 net: enetc: unmap DMA in enetc_send_cmd()

In reply to Greg Kroah-Hartman

From: Tim Gardner <[email protected]>

[ Upstream commit cd4bc63de774eee95e9bac26a565cd80e0fca421 ]

Coverity complains of a possible dereference of a null return value.

5. returned_null: kzalloc returns NULL. [show details]
6. var_assigned: Assigning: si_data = NULL return value from kzalloc.
488 si_data = kzalloc(data_size, __GFP_DMA | GFP_KERNEL);
489 cbd.length = cpu_to_le16(data_size);
490
491 dma = dma_map_single(&priv->si->pdev->dev, si_data,
492 data_size, DMA_FROM_DEVICE);

While this kzalloc() is unlikely to fail, I did notice that the function returned without unmapping si_data.

Fix this by refactoring the error paths and checking for kzalloc() failure.

Fixes 888ae5a3952ba ("net: enetc: add tc flower psfp offload driver")
Cc Claudiu Manoil <[email protected]>
Cc "David S. Miller" <[email protected]>
Cc Jakub Kicinski <[email protected]>
Cc [email protected] (open list)
---
 .../net/ethernet/freescale/enetc/enetc_qos.c   | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)
drivers/net/ethernet/freescale/enetc/enetc_qos.c +11 -7
index 4577226d3c6ad..0536d2c76fbc4 100644
@@ -486,14 +486,16 @@ static int enetc_streamid_hw_set(struct enetc_ndev_priv *priv,
 
 	data_size = sizeof(struct streamid_data);
 	si_data = kzalloc(data_size, __GFP_DMA | GFP_KERNEL);
+	if (!si_data)
+		return -ENOMEM;
 	cbd.length = cpu_to_le16(data_size);
 
 	dma = dma_map_single(&priv->si->pdev->dev, si_data,
 			     data_size, DMA_FROM_DEVICE);
 	if (dma_mapping_error(&priv->si->pdev->dev, dma)) {
 		netdev_err(priv->si->ndev, "DMA mapping failed!\n");
-		kfree(si_data);
-		return -ENOMEM;
+		err = -ENOMEM;
+		goto out;
 	}
 
 	cbd.addr[0] = cpu_to_le32(lower_32_bits(dma));
@@ -512,12 +514,10 @@ static int enetc_streamid_hw_set(struct enetc_ndev_priv *priv,
 
 	err = enetc_send_cmd(priv->si, &cbd);
 	if (err)
-		return -EINVAL;
+		goto out;
 
-	if (!enable) {
-		kfree(si_data);
-		return 0;
-	}
+	if (!enable)
+		goto out;
 
 	/* Enable the entry overwrite again incase space flushed by hardware */
 	memset(&cbd, 0, sizeof(cbd));
@@ -560,6 +560,10 @@ static int enetc_streamid_hw_set(struct enetc_ndev_priv *priv,
 	}
 
 	err = enetc_send_cmd(priv->si, &cbd);
+out:
+	if (!dma_mapping_error(&priv->si->pdev->dev, dma))
+		dma_unmap_single(&priv->si->pdev->dev, dma, data_size, DMA_FROM_DEVICE);
+
 	kfree(si_data);
 
 	return err;
Signature · 1 line
2.33.0
author maintainer
#501 lore raw

PATCH 490/917 phy: micrel: ksz8041nl: do not use power down mode

In reply to Greg Kroah-Hartman

From: Stefan Agner <[email protected]>

[ Upstream commit 2641b62d2fab52648e34cdc6994b2eacde2d27c1 ]

Some Micrel KSZ8041NL PHY chips exhibit continuous RX errors after using the power down mode bit (0.11). If the PHY is taken out of power down mode in a certain temperature range, the PHY enters a weird state which leads to continuously reporting RX errors. In that state, the MAC is not able to receive or send any Ethernet frames and the activity LED is constantly blinking. Since Linux is using the suspend callback when the interface is taken down, ending up in that state can easily happen during a normal startup.

Micrel confirmed the issue in errata DS80000700A [*], caused by abnormal clock recovery when using power down mode. Even the latest revision (A4, Revision ID 0x1513) seems to suffer that problem, and according to the errata is not going to be fixed.

Remove the suspend/resume callback to avoid using the power down mode completely.

[*] https://ww1.microchip.com/downloads/en/DeviceDoc/80000700A.pdf

Fixes 1a5465f5d6a2 ("phy/micrel: Add suspend/resume support to Micrel PHYs")
---
 drivers/net/phy/micrel.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/net/phy/micrel.c +3 -2
index 643b1c1827a92..aec0fcefdccd6 100644
@@ -1593,8 +1593,9 @@ static struct phy_driver ksphy_driver[] = {
 	.get_sset_count = kszphy_get_sset_count,
 	.get_strings	= kszphy_get_strings,
 	.get_stats	= kszphy_get_stats,
-	.suspend	= genphy_suspend,
-	.resume		= genphy_resume,
+	/* No suspend/resume callbacks because of errata DS80000700A,
+	 * receiver error following software power down.
+	 */
 }, {
 	.phy_id		= PHY_ID_KSZ8041RNLI,
 	.phy_id_mask	= MICREL_PHY_ID_MASK,
Signature · 1 line
2.33.0
author maintainer
#502 lore raw

PATCH 491/917 nbd: Fix use-after-free in pid_show

In reply to Greg Kroah-Hartman

From: Ye Bin <[email protected]>

[ Upstream commit 0c98057be9efa32de78dbc4685fc73da9d71faa1 ]

I got issue as follows:

[  263.886511] BUG: KASAN: use-after-free in pid_show+0x11f/0x13f
[  263.888359] Read of size 4 at addr ffff8880bf0648c0 by task cat/746
[  263.890479] CPU: 0 PID: 746 Comm: cat Not tainted 4.19.90-dirty #140
[  263.893162] Call Trace:
[  263.893509]  dump_stack+0x108/0x15f
[  263.893999]  print_address_description+0xa5/0x372
[  263.894641]  kasan_report.cold+0x236/0x2a8
[  263.895696]  __asan_report_load4_noabort+0x25/0x30
[  263.896365]  pid_show+0x11f/0x13f
[  263.897422]  dev_attr_show+0x48/0x90
[  263.898361]  sysfs_kf_seq_show+0x24d/0x4b0
[  263.899479]  kernfs_seq_show+0x14e/0x1b0
[  263.900029]  seq_read+0x43f/0x1150
[  263.900499]  kernfs_fop_read+0xc7/0x5a0
[  263.903764]  vfs_read+0x113/0x350
[  263.904231]  ksys_read+0x103/0x270
[  263.905230]  __x64_sys_read+0x77/0xc0
[  263.906284]  do_syscall_64+0x106/0x360
[  263.906797]  entry_SYSCALL_64_after_hwframe+0x44/0xa9

Reproduce this issue as follows:
1. nbd-server 8000 /tmp/disk
2. nbd-client localhost 8000 /dev/nbd1
3. cat /sys/block/nbd1/pid
Then trigger use-after-free in pid_show.

Reason is after do step '2', nbd-client progress is already exit. So it's task_struct already freed.
To solve this issue, revert part of 6521d39a64b3's modify and remove useless 'recv_task' member of nbd_device.

---
 drivers/block/nbd.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)
drivers/block/nbd.c +9 -9
index 1183f7872b713..4f1b591c3a555 100644
@@ -122,10 +122,10 @@ struct nbd_device {
 	struct work_struct remove_work;
 
 	struct list_head list;
-	struct task_struct *task_recv;
 	struct task_struct *task_setup;
 
 	unsigned long flags;
+	pid_t pid; /* pid of nbd-client, if attached */
 
 	char *backend;
 };
@@ -217,7 +217,7 @@ static ssize_t pid_show(struct device *dev,
 	struct gendisk *disk = dev_to_disk(dev);
 	struct nbd_device *nbd = (struct nbd_device *)disk->private_data;
 
-	return sprintf(buf, "%d\n", task_pid_nr(nbd->task_recv));
+	return sprintf(buf, "%d\n", nbd->pid);
 }
 
 static const struct device_attribute pid_attr = {
@@ -329,7 +329,7 @@ static int nbd_set_size(struct nbd_device *nbd, loff_t bytesize,
 	nbd->config->bytesize = bytesize;
 	nbd->config->blksize_bits = __ffs(blksize);
 
-	if (!nbd->task_recv)
+	if (!nbd->pid)
 		return 0;
 
 	if (nbd->config->flags & NBD_FLAG_SEND_TRIM) {
@@ -1241,7 +1241,7 @@ static void nbd_config_put(struct nbd_device *nbd)
 		if (test_and_clear_bit(NBD_RT_HAS_PID_FILE,
 				       &config->runtime_flags))
 			device_remove_file(disk_to_dev(nbd->disk), &pid_attr);
-		nbd->task_recv = NULL;
+		nbd->pid = 0;
 		if (test_and_clear_bit(NBD_RT_HAS_BACKEND_FILE,
 				       &config->runtime_flags)) {
 			device_remove_file(disk_to_dev(nbd->disk), &backend_attr);
@@ -1282,7 +1282,7 @@ static int nbd_start_device(struct nbd_device *nbd)
 	int num_connections = config->num_connections;
 	int error = 0, i;
 
-	if (nbd->task_recv)
+	if (nbd->pid)
 		return -EBUSY;
 	if (!config->socks)
 		return -EINVAL;
@@ -1301,7 +1301,7 @@ static int nbd_start_device(struct nbd_device *nbd)
 	}
 
 	blk_mq_update_nr_hw_queues(&nbd->tag_set, config->num_connections);
-	nbd->task_recv = current;
+	nbd->pid = task_pid_nr(current);
 
 	nbd_parse_flags(nbd);
 
@@ -1557,8 +1557,8 @@ static int nbd_dbg_tasks_show(struct seq_file *s, void *unused)
 {
 	struct nbd_device *nbd = s->private;
 
-	if (nbd->task_recv)
-		seq_printf(s, "recv: %d\n", task_pid_nr(nbd->task_recv));
+	if (nbd->pid)
+		seq_printf(s, "recv: %d\n", nbd->pid);
 
 	return 0;
 }
@@ -2135,7 +2135,7 @@ static int nbd_genl_reconfigure(struct sk_buff *skb, struct genl_info *info)
 	mutex_lock(&nbd->config_lock);
 	config = nbd->config;
 	if (!test_bit(NBD_RT_BOUND, &config->runtime_flags) ||
-	    !nbd->task_recv) {
+	    !nbd->pid) {
 		dev_err(nbd_to_dev(nbd),
 			"not configured, cannot reconfigure\n");
 		ret = -EINVAL;
Signature · 1 line
2.33.0
author maintainer
#503 lore raw

PATCH 492/917 nvme-rdma: fix error code in nvme_rdma_setup_ctrl

In reply to Greg Kroah-Hartman

From: Max Gurtovoy <[email protected]>

[ Upstream commit 09748122009aed7bfaa7acc33c10c083a4758322 ]

In case that icdoff is not zero or mandatory keyed sgls are not supported by the NVMe/RDMA target, we'll go to error flow but we'll return 0 to the caller. Fix it by returning an appropriate error code.

Fixes c66e2998c8ca ("nvme-rdma: centralize controller setup sequence")
---
 drivers/nvme/host/rdma.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/nvme/host/rdma.c +2 -0
index 042c594bc57e2..0498801542eb6 100644
@@ -1095,11 +1095,13 @@ static int nvme_rdma_setup_ctrl(struct nvme_rdma_ctrl *ctrl, bool new)
 		return ret;
 
 	if (ctrl->ctrl.icdoff) {
+		ret = -EOPNOTSUPP;
 		dev_err(ctrl->ctrl.device, "icdoff is not supported!\n");
 		goto destroy_admin;
 	}
 
 	if (!(ctrl->ctrl.sgls & (1 << 2))) {
+		ret = -EOPNOTSUPP;
 		dev_err(ctrl->ctrl.device,
 			"Mandatory keyed sgls are not supported!\n");
 		goto destroy_admin;
Signature · 1 line
2.33.0
author maintainer
#504 lore raw

PATCH 493/917 PM: hibernate: fix sparse warnings

In reply to Greg Kroah-Hartman

From: Anders Roxell <[email protected]>

[ Upstream commit 01de5fcd8b1ac0ca28d2bb0921226a54fdd62684 ]

When building the kernel with sparse enabled 'C=1' the following warnings shows up:

kernel/power/swap.c:390:29: warning: incorrect type in assignment (different base types)
kernel/power/swap.c:390:29: expected int ret
kernel/power/swap.c:390:29: got restricted blk_status_t

This is due to function hib_wait_io() returns a 'blk_status_t' which is a bitwise u8. Commit 5416da01ff6e ("PM: hibernate: Remove blk_status_to_errno in hib_wait_io") seemed to have mixed up the return type. However, the 4e4cbee93d56 ("block: switch bios to blk_status_t") actually broke the behaviour by returning the wrong type.

Rework so function hib_wait_io() returns a 'int' instead of 'blk_status_t' and make sure to call function blk_status_to_errno(hb->error)' when returning from function hib_wait_io() a int gets returned.

Fixes 4e4cbee93d56 ("block: switch bios to blk_status_t")
Fixes 5416da01ff6e ("PM: hibernate: Remove blk_status_to_errno in hib_wait_io")
---
 kernel/power/swap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/power/swap.c +1 -1
index 0aabc94125d6b..f3a1086f7cdb2 100644
@@ -299,7 +299,7 @@ static int hib_submit_io(int op, int op_flags, pgoff_t page_off, void *addr,
 	return error;
 }
 
-static blk_status_t hib_wait_io(struct hib_bio_batch *hb)
+static int hib_wait_io(struct hib_bio_batch *hb)
 {
 	/*
 	 * We are relying on the behavior of blk_plug that a thread with
Signature · 1 line
2.33.0
author maintainer
#505 lore raw

PATCH 494/917 clocksource/drivers/timer-ti-dm: Select TIMER_OF

In reply to Greg Kroah-Hartman

From: Kees Cook <[email protected]>

[ Upstream commit eda9a4f7af6ee47e9e131f20e4f8a41a97379293 ]

When building OMAP_DM_TIMER without TIMER_OF, there are orphan sections due to the use of TIMER_OF_DELCARE() without CONFIG_TIMER_OF. Select CONFIG_TIMER_OF when enaling OMAP_DM_TIMER:

arm-linux-gnueabi-ld: warning: orphan section __timer_of_table' from drivers/clocksource/timer-ti-dm-systimer.o' being placed in section `__timer_of_table'

Cc Tony Lindgren <[email protected]>
Cc Daniel Lezcano <[email protected]>
Cc Keerthy <[email protected]>
Cc Sebastian Reichel <[email protected]>
Cc Ladislav Michl <[email protected]>
Cc Grygorii Strashko <[email protected]>
Fixes 52762fbd1c47 ("clocksource/drivers/timer-ti-dm: Add clockevent and clocksource support")
---
 drivers/clocksource/Kconfig | 1 +
 1 file changed, 1 insertion(+)
drivers/clocksource/Kconfig +1 -0
index 0f5e3983951a8..08f8cb944a2ac 100644
@@ -24,6 +24,7 @@ config I8253_LOCK
 
 config OMAP_DM_TIMER
 	bool
+	select TIMER_OF
 
 config CLKBLD_I8253
 	def_bool y if CLKSRC_I8253 || CLKEVT_I8253 || I8253_LOCK
Signature · 1 line
2.33.0
author maintainer
#506 lore raw

PATCH 495/917 x86/sev: Fix stack type check in vc_switch_off_ist()

In reply to Greg Kroah-Hartman

From: Joerg Roedel <[email protected]>

[ Upstream commit 5681981fb788281b09a4ea14d310d30b2bd89132 ]

The value of STACK_TYPE_EXCEPTION_LAST points to the last _valid_ exception stack. Reflect that in the check done in the vc_switch_off_ist() function.

Fixes a13644f3a53de ("x86/entry/64: Add entry code for #VC handler")
---
 arch/x86/kernel/traps.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/kernel/traps.c +1 -1
index cc6de3a01293c..5b1984d468227 100644
@@ -743,7 +743,7 @@ asmlinkage __visible noinstr struct pt_regs *vc_switch_off_ist(struct pt_regs *r
 	stack = (unsigned long *)sp;
 
 	if (!get_stack_info_noinstr(stack, current, &info) || info.type == STACK_TYPE_ENTRY ||
-	    info.type >= STACK_TYPE_EXCEPTION_LAST)
+	    info.type > STACK_TYPE_EXCEPTION_LAST)
 		sp = __this_cpu_ist_top_va(VC2);
 
 sync:
Signature · 1 line
2.33.0
author maintainer
#507 lore raw

PATCH 496/917 drm/msm: Fix potential NULL dereference in DPU SSPP

In reply to Greg Kroah-Hartman

From: Jessica Zhang <[email protected]>

[ Upstream commit 8bf71a5719b6cc5b6ba358096081e5d50ea23ab6 ]

Move initialization of sblk in _sspp_subblk_offset() after NULL check to avoid potential NULL pointer dereference.

---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)
drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c +5 -3
index 69eed79324865..f9460672176aa 100644
@@ -138,11 +138,13 @@ static int _sspp_subblk_offset(struct dpu_hw_pipe *ctx,
 		u32 *idx)
 {
 	int rc = 0;
-	const struct dpu_sspp_sub_blks *sblk = ctx->cap->sblk;
+	const struct dpu_sspp_sub_blks *sblk;
 
-	if (!ctx)
+	if (!ctx || !ctx->cap || !ctx->cap->sblk)
 		return -EINVAL;
 
+	sblk = ctx->cap->sblk;
+
 	switch (s_id) {
 	case DPU_SSPP_SRC:
 		*idx = sblk->src_blk.base;
@@ -419,7 +421,7 @@ static void _dpu_hw_sspp_setup_scaler3(struct dpu_hw_pipe *ctx,
 
 	(void)pe;
 	if (_sspp_subblk_offset(ctx, DPU_SSPP_SCALER_QSEED3, &idx) || !sspp
-		|| !scaler3_cfg || !ctx || !ctx->cap || !ctx->cap->sblk)
+		|| !scaler3_cfg)
 		return;
 
 	dpu_hw_setup_scaler3(&ctx->hw, scaler3_cfg, idx,
Signature · 1 line
2.33.0
author maintainer
#508 lore raw

PATCH 497/917 drm/msm/dsi: fix wrong type in msm_dsi_host

In reply to Greg Kroah-Hartman

From: Jessica Zhang <[email protected]>

[ Upstream commit 409af447c2a0a6e08ba190993a1153c91d3b11bd ]

Change byte_clk_rate, pixel_clk_rate, esc_clk_rate, and src_clk_rate from u32 to unsigned long, since clk_get_rate() returns an unsigned long.

Fixes a6bcddbc2ee1 ("drm/msm: dsi: Handle dual-channel for 6G as well")
---
 drivers/gpu/drm/msm/dsi/dsi_host.c | 24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)
drivers/gpu/drm/msm/dsi/dsi_host.c +12 -12
index bccd379bdba75..ea641151e77e7 100644
@@ -115,16 +115,16 @@ struct msm_dsi_host {
 	struct clk *pixel_clk_src;
 	struct clk *byte_intf_clk;
 
-	u32 byte_clk_rate;
-	u32 pixel_clk_rate;
-	u32 esc_clk_rate;
+	unsigned long byte_clk_rate;
+	unsigned long pixel_clk_rate;
+	unsigned long esc_clk_rate;
 
 	/* DSI v2 specific clocks */
 	struct clk *src_clk;
 	struct clk *esc_clk_src;
 	struct clk *dsi_clk_src;
 
-	u32 src_clk_rate;
+	unsigned long src_clk_rate;
 
 	struct gpio_desc *disp_en_gpio;
 	struct gpio_desc *te_gpio;
@@ -498,10 +498,10 @@ int msm_dsi_runtime_resume(struct device *dev)
 
 int dsi_link_clk_set_rate_6g(struct msm_dsi_host *msm_host)
 {
-	u32 byte_intf_rate;
+	unsigned long byte_intf_rate;
 	int ret;
 
-	DBG("Set clk rates: pclk=%d, byteclk=%d",
+	DBG("Set clk rates: pclk=%d, byteclk=%lu",
 		msm_host->mode->clock, msm_host->byte_clk_rate);
 
 	ret = dev_pm_opp_set_rate(&msm_host->pdev->dev,
@@ -583,7 +583,7 @@ int dsi_link_clk_set_rate_v2(struct msm_dsi_host *msm_host)
 {
 	int ret;
 
-	DBG("Set clk rates: pclk=%d, byteclk=%d, esc_clk=%d, dsi_src_clk=%d",
+	DBG("Set clk rates: pclk=%d, byteclk=%lu, esc_clk=%lu, dsi_src_clk=%lu",
 		msm_host->mode->clock, msm_host->byte_clk_rate,
 		msm_host->esc_clk_rate, msm_host->src_clk_rate);
 
@@ -673,10 +673,10 @@ void dsi_link_clk_disable_v2(struct msm_dsi_host *msm_host)
 	clk_disable_unprepare(msm_host->byte_clk);
 }
 
-static u32 dsi_get_pclk_rate(struct msm_dsi_host *msm_host, bool is_bonded_dsi)
+static unsigned long dsi_get_pclk_rate(struct msm_dsi_host *msm_host, bool is_bonded_dsi)
 {
 	struct drm_display_mode *mode = msm_host->mode;
-	u32 pclk_rate;
+	unsigned long pclk_rate;
 
 	pclk_rate = mode->clock * 1000;
 
@@ -696,7 +696,7 @@ static void dsi_calc_pclk(struct msm_dsi_host *msm_host, bool is_bonded_dsi)
 {
 	u8 lanes = msm_host->lanes;
 	u32 bpp = dsi_get_bpp(msm_host->format);
-	u32 pclk_rate = dsi_get_pclk_rate(msm_host, is_bonded_dsi);
+	unsigned long pclk_rate = dsi_get_pclk_rate(msm_host, is_bonded_dsi);
 	u64 pclk_bpp = (u64)pclk_rate * bpp;
 
 	if (lanes == 0) {
@@ -713,7 +713,7 @@ static void dsi_calc_pclk(struct msm_dsi_host *msm_host, bool is_bonded_dsi)
 	msm_host->pixel_clk_rate = pclk_rate;
 	msm_host->byte_clk_rate = pclk_bpp;
 
-	DBG("pclk=%d, bclk=%d", msm_host->pixel_clk_rate,
+	DBG("pclk=%lu, bclk=%lu", msm_host->pixel_clk_rate,
 				msm_host->byte_clk_rate);
 
 }
@@ -772,7 +772,7 @@ int dsi_calc_clk_rate_v2(struct msm_dsi_host *msm_host, bool is_bonded_dsi)
 
 	msm_host->esc_clk_rate = msm_host->byte_clk_rate / esc_div;
 
-	DBG("esc=%d, src=%d", msm_host->esc_clk_rate,
+	DBG("esc=%lu, src=%lu", msm_host->esc_clk_rate,
 		msm_host->src_clk_rate);
 
 	return 0;
Signature · 1 line
2.33.0
author maintainer
#509 lore raw

PATCH 498/917 crypto: tcrypt - fix skcipher multi-buffer tests for 1420B blocks

In reply to Greg Kroah-Hartman

From: Horia Geantă <[email protected]>

[ Upstream commit 3ae88f676aa63366ffa9eebb8ae787c7e19f0c57 ]

Commit ad6d66bcac77e ("crypto: tcrypt - include 1420 byte blocks in aead and skcipher benchmarks") mentions:

Quoted 3 lines

power-of-2 block size. So let's add 1420 bytes explicitly, and round it up to the next blocksize multiple of the algo in question if it does not support 1420 byte blocks.

but misses updating skcipher multi-buffer tests.

Fix this by using the proper (rounded) input size.

Fixes ad6d66bcac77e ("crypto: tcrypt - include 1420 byte blocks in aead and skcipher benchmarks")
---
 crypto/tcrypt.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
crypto/tcrypt.c +2 -3
index 82b0400985a51..00149657a4bc1 100644
@@ -1333,7 +1333,7 @@ static void test_mb_skcipher_speed(const char *algo, int enc, int secs,
 
 			if (bs > XBUFSIZE * PAGE_SIZE) {
 				pr_err("template (%u) too big for buffer (%lu)\n",
-				       *b_size, XBUFSIZE * PAGE_SIZE);
+				       bs, XBUFSIZE * PAGE_SIZE);
 				goto out;
 			}
 
@@ -1386,8 +1386,7 @@ static void test_mb_skcipher_speed(const char *algo, int enc, int secs,
 				memset(cur->xbuf[p], 0xff, k);
 
 				skcipher_request_set_crypt(cur->req, cur->sg,
-							   cur->sg, *b_size,
-							   iv);
+							   cur->sg, bs, iv);
 			}
 
 			if (secs) {
Signature · 1 line
2.33.0
author maintainer
#510 lore raw

PATCH 499/917 smackfs: use netlbl_cfg_cipsov4_del() for deleting cipso_v4_doi

In reply to Greg Kroah-Hartman

From: Tetsuo Handa <[email protected]>

[ Upstream commit 0934ad42bb2c5df90a1b9de690f93de735b622fe ]

syzbot is reporting UAF at cipso_v4_doi_search() [1], for smk_cipso_doi() is calling kfree() without removing from the cipso_v4_doi_list list after netlbl_cfg_cipsov4_map_add() returned an error. We need to use netlbl_cfg_cipsov4_del() in order to remove from the list and wait for RCU grace period before kfree().

Fixes 6c2e8ac0953fccdd ("netlabel: Update kernel configuration API")
---
 security/smack/smackfs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
security/smack/smackfs.c +1 -1
index 89989d28ffc55..658eab05599e6 100644
@@ -712,7 +712,7 @@ static void smk_cipso_doi(void)
 	if (rc != 0) {
 		printk(KERN_WARNING "%s:%d map add rc = %d\n",
 		       __func__, __LINE__, rc);
-		kfree(doip);
+		netlbl_cfg_cipsov4_del(doip->doi, &nai);
 		return;
 	}
 }
Signature · 1 line
2.33.0
author maintainer
#511 lore raw

PATCH 500/917 KVM: selftests: Fix nested SVM tests when built with clang

In reply to Greg Kroah-Hartman

From: Jim Mattson <[email protected]>

[ Upstream commit ed290e1c20da19fa100a3e0f421aa31b65984960 ]

Though gcc conveniently compiles a simple memset to "rep stos," clang prefers to call the libc version of memset. If a test is dynamically linked, the libc memset isn't available in L1 (nor is the PLT or the GOT, for that matter). Even if the test is statically linked, the libc memset may choose to use some CPU features, like AVX, which may not be enabled in L1. Note that __builtin_memset doesn't solve the problem, because (a) the compiler is free to call memset anyway, and (b) __builtin_memset may also choose to use features like AVX, which may not be available in L1.

To avoid a myriad of problems, use an explicit "rep stos" to clear the VMCB in generic_svm_setup(), which is called both from L0 and L1.

Fixes 20ba262f8631a ("selftests: KVM: AMD Nested test infrastructure")

Message-Id: <[email protected]>

---
 tools/testing/selftests/kvm/lib/x86_64/svm.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)
tools/testing/selftests/kvm/lib/x86_64/svm.c +13 -1
index 2ac98d70d02bd..161eba7cd1289 100644
@@ -54,6 +54,18 @@ static void vmcb_set_seg(struct vmcb_seg *seg, u16 selector,
 	seg->base = base;
 }
 
+/*
+ * Avoid using memset to clear the vmcb, since libc may not be
+ * available in L1 (and, even if it is, features that libc memset may
+ * want to use, like AVX, may not be enabled).
+ */
+static void clear_vmcb(struct vmcb *vmcb)
+{
+	int n = sizeof(*vmcb) / sizeof(u32);
+
+	asm volatile ("rep stosl" : "+c"(n), "+D"(vmcb) : "a"(0) : "memory");
+}
+
 void generic_svm_setup(struct svm_test_data *svm, void *guest_rip, void *guest_rsp)
 {
 	struct vmcb *vmcb = svm->vmcb;
@@ -70,7 +82,7 @@ void generic_svm_setup(struct svm_test_data *svm, void *guest_rip, void *guest_r
 	wrmsr(MSR_EFER, efer | EFER_SVME);
 	wrmsr(MSR_VM_HSAVE_PA, svm->save_area_gpa);
 
-	memset(vmcb, 0, sizeof(*vmcb));
+	clear_vmcb(vmcb);
 	asm volatile ("vmsave %0\n\t" : : "a" (vmcb_gpa) : "memory");
 	vmcb_set_seg(&save->es, get_es(), 0, -1U, data_seg_attr);
 	vmcb_set_seg(&save->cs, get_cs(), 0, -1U, code_seg_attr);
Signature · 1 line
2.33.0
author maintainer
#512 lore raw

PATCH 501/917 libbpf: Fix memory leak in btf__dedup()

In reply to Greg Kroah-Hartman

From: Mauricio Vásquez <[email protected]>

[ Upstream commit 1000298c76830bc291358e98e8fa5baa3baa9b3a ]

Free btf_dedup if btf_ensure_modifiable() returns error.

Fixes 919d2b1dbb07 ("libbpf: Allow modification of BTF and add btf__add_str API")
---
 tools/lib/bpf/btf.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
tools/lib/bpf/btf.c +4 -2
index 77dc24d58302d..bf8c8676d68e5 100644
@@ -2914,8 +2914,10 @@ int btf__dedup(struct btf *btf, struct btf_ext *btf_ext,
 		return libbpf_err(-EINVAL);
 	}
 
-	if (btf_ensure_modifiable(btf))
-		return libbpf_err(-ENOMEM);
+	if (btf_ensure_modifiable(btf)) {
+		err = -ENOMEM;
+		goto done;
+	}
 
 	err = btf_dedup_prep(d);
 	if (err) {
Signature · 1 line
2.33.0
author maintainer
#513 lore raw

PATCH 502/917 bpftool: Avoid leaking the JSON writer prepared for program metadata

In reply to Greg Kroah-Hartman

From: Quentin Monnet <[email protected]>

[ Upstream commit e89ef634f81c9d90e1824ab183721f3b361472e6 ]

Bpftool creates a new JSON object for writing program metadata in plain text mode, regardless of metadata being present or not. Then this writer is freed if any metadata has been found and printed, but it leaks otherwise. We cannot destroy the object unconditionally, because the destructor prints an undesirable line break. Instead, make sure the writer is created only after we have found program metadata to print.

Found with valgrind.

---
 tools/bpf/bpftool/prog.c | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)
tools/bpf/bpftool/prog.c +9 -7
index 9c3e343b7d872..fe59404e87046 100644
@@ -308,18 +308,12 @@ static void show_prog_metadata(int fd, __u32 num_maps)
 		if (printed_header)
 			jsonw_end_object(json_wtr);
 	} else {
-		json_writer_t *btf_wtr = jsonw_new(stdout);
+		json_writer_t *btf_wtr;
 		struct btf_dumper d = {
 			.btf = btf,
-			.jw = btf_wtr,
 			.is_plain_text = true,
 		};
 
-		if (!btf_wtr) {
-			p_err("jsonw alloc failed");
-			goto out_free;
-		}
-
 		for (i = 0; i < vlen; i++, vsi++) {
 			t_var = btf__type_by_id(btf, vsi->type);
 			name = btf__name_by_offset(btf, t_var->name_off);
@@ -329,6 +323,14 @@ static void show_prog_metadata(int fd, __u32 num_maps)
 
 			if (!printed_header) {
 				printf("\tmetadata:");
+
+				btf_wtr = jsonw_new(stdout);
+				if (!btf_wtr) {
+					p_err("jsonw alloc failed");
+					goto out_free;
+				}
+				d.jw = btf_wtr,
+
 				printed_header = true;
 			}
 
Signature · 1 line
2.33.0
author maintainer
#514 lore raw

PATCH 503/917 libbpf: Fix overflow in BTF sanity checks

In reply to Greg Kroah-Hartman

From: Andrii Nakryiko <[email protected]>

[ Upstream commit 5245dafe3d49efba4d3285cf27ee1cc1eeafafc6 ]

btf_header's str_off+str_len or type_off+type_len can overflow as they are u32s. This will lead to bypassing the sanity checks during BTF parsing, resulting in crashes afterwards. Fix by using 64-bit signed integers for comparison.

Fixes d8123624506c ("libbpf: Fix BTF data layout checks and allow empty BTF")
---
 tools/lib/bpf/btf.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
tools/lib/bpf/btf.c +2 -2
index bf8c8676d68e5..cfd701debcf61 100644
@@ -236,12 +236,12 @@ static int btf_parse_hdr(struct btf *btf)
 	}
 
 	meta_left = btf->raw_size - sizeof(*hdr);
-	if (meta_left < hdr->str_off + hdr->str_len) {
+	if (meta_left < (long long)hdr->str_off + hdr->str_len) {
 		pr_debug("Invalid BTF total size:%u\n", btf->raw_size);
 		return -EINVAL;
 	}
 
-	if (hdr->type_off + hdr->type_len > hdr->str_off) {
+	if ((long long)hdr->type_off + hdr->type_len > hdr->str_off) {
 		pr_debug("Invalid BTF data sections layout: type data at %u + %u, strings data at %u + %u\n",
 			 hdr->type_off, hdr->type_len, hdr->str_off, hdr->str_len);
 		return -EINVAL;
Signature · 1 line
2.33.0
author maintainer
#515 lore raw

PATCH 504/917 libbpf: Fix BTF header parsing checks

In reply to Greg Kroah-Hartman

From: Andrii Nakryiko <[email protected]>

[ Upstream commit c825f5fee19caf301d9821cd79abaa734322de26 ]

Original code assumed fixed and correct BTF header length. That's not always the case, though, so fix this bug with a proper additional check. And use actual header length instead of sizeof(struct btf_header) in sanity checks.

---
 tools/lib/bpf/btf.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)
tools/lib/bpf/btf.c +9 -3
index cfd701debcf61..1b9341ef638b0 100644
@@ -231,13 +231,19 @@ static int btf_parse_hdr(struct btf *btf)
 		}
 		btf_bswap_hdr(hdr);
 	} else if (hdr->magic != BTF_MAGIC) {
-		pr_debug("Invalid BTF magic:%x\n", hdr->magic);
+		pr_debug("Invalid BTF magic: %x\n", hdr->magic);
 		return -EINVAL;
 	}
 
-	meta_left = btf->raw_size - sizeof(*hdr);
+	if (btf->raw_size < hdr->hdr_len) {
+		pr_debug("BTF header len %u larger than data size %u\n",
+			 hdr->hdr_len, btf->raw_size);
+		return -EINVAL;
+	}
+
+	meta_left = btf->raw_size - hdr->hdr_len;
 	if (meta_left < (long long)hdr->str_off + hdr->str_len) {
-		pr_debug("Invalid BTF total size:%u\n", btf->raw_size);
+		pr_debug("Invalid BTF total size: %u\n", btf->raw_size);
 		return -EINVAL;
 	}
 
Signature · 1 line
2.33.0
author maintainer
#516 lore raw

PATCH 505/917 mt76: mt7615: mt7622: fix ibss and meshpoint

In reply to Greg Kroah-Hartman

From: Nick Hainke <[email protected]>

[ Upstream commit 753453afacc0243bd45de45e34218a8d17493e8f ]

commit 7f4b7920318b ("mt76: mt7615: add ibss support") introduced IBSS and commit f4ec7fdf7f83 ("mt76: mt7615: enable support for mesh") meshpoint support.

Both used in the "get_omac_idx"-function:

	if (~mask & BIT(HW_BSSID_0))
		return HW_BSSID_0;

With commit d8d59f66d136 ("mt76: mt7615: support 16 interfaces") the ibss and meshpoint mode should "prefer hw bssid slot 1-3". However, with that change the ibss or meshpoint mode will not send any beacon on the mt7622 wifi anymore. Devices were still able to exchange data but only if a bssid already existed. Two mt7622 devices will never be able to communicate.

This commits reverts the preferation of slot 1-3 for ibss and meshpoint. Only NL80211_IFTYPE_STATION will still prefer slot 1-3.

Tested on Banana Pi R64.

---
 drivers/net/wireless/mediatek/mt76/mt7615/main.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/wireless/mediatek/mt76/mt7615/main.c +2 -2
index dada43d6d879e..51260a669d166 100644
@@ -135,8 +135,6 @@ static int get_omac_idx(enum nl80211_iftype type, u64 mask)
 	int i;
 
 	switch (type) {
-	case NL80211_IFTYPE_MESH_POINT:
-	case NL80211_IFTYPE_ADHOC:
 	case NL80211_IFTYPE_STATION:
 		/* prefer hw bssid slot 1-3 */
 		i = get_free_idx(mask, HW_BSSID_1, HW_BSSID_3);
@@ -160,6 +158,8 @@ static int get_omac_idx(enum nl80211_iftype type, u64 mask)
 			return HW_BSSID_0;
 
 		break;
+	case NL80211_IFTYPE_ADHOC:
+	case NL80211_IFTYPE_MESH_POINT:
 	case NL80211_IFTYPE_MONITOR:
 	case NL80211_IFTYPE_AP:
 		/* ap uses hw bssid 0 and ext bssid */
Signature · 1 line
2.33.0
author maintainer
#517 lore raw

PATCH 506/917 s390/gmap: validate VMA in __gmap_zap()

In reply to Greg Kroah-Hartman

From: David Hildenbrand <[email protected]>

[ Upstream commit 2d8fb8f3914b40e3cc12f8cbb74daefd5245349d ]

We should not walk/touch page tables outside of VMA boundaries when holding only the mmap sem in read mode. Evil user space can modify the VMA layout just before this function runs and e.g., trigger races with page table removal code since commit dd2283f2605e ("mm: mmap: zap pages with read mmap_sem in munmap"). The pure prescence in our guest_to_host radix tree does not imply that there is a VMA.

Further, we should not allocate page tables (via get_locked_pte()) outside of VMA boundaries: if evil user space decides to map hugetlbfs to these ranges, bad things will happen because we suddenly have PTE or PMD page tables where we shouldn't have them.

Similarly, we have to check if we suddenly find a hugetlbfs VMA, before calling get_locked_pte().

Note that gmap_discard() is different:
zap_page_range()->unmap_single_vma() makes sure to stay within VMA boundaries.

---
 arch/s390/mm/gmap.c | 6 ++++++
 1 file changed, 6 insertions(+)
arch/s390/mm/gmap.c +6 -0
index 4d3b33ce81c62..e0735c3437759 100644
@@ -672,6 +672,7 @@ EXPORT_SYMBOL_GPL(gmap_fault);
  */
 void __gmap_zap(struct gmap *gmap, unsigned long gaddr)
 {
+	struct vm_area_struct *vma;
 	unsigned long vmaddr;
 	spinlock_t *ptl;
 	pte_t *ptep;
@@ -681,6 +682,11 @@ void __gmap_zap(struct gmap *gmap, unsigned long gaddr)
 						   gaddr >> PMD_SHIFT);
 	if (vmaddr) {
 		vmaddr |= gaddr & ~PMD_MASK;
+
+		vma = vma_lookup(gmap->mm, vmaddr);
+		if (!vma || is_vm_hugetlb_page(vma))
+			return;
+
 		/* Get pointer to the page table entry */
 		ptep = get_locked_pte(gmap->mm, vmaddr, &ptl);
 		if (likely(ptep))
Signature · 1 line
2.33.0
author maintainer
#518 lore raw

PATCH 507/917 s390/gmap: dont unconditionally call pte_unmap_unlock() in __gmap_zap()

In reply to Greg Kroah-Hartman

From: David Hildenbrand <[email protected]>

[ Upstream commit b159f94c86b43cf7e73e654bc527255b1f4eafc4 ]

... otherwise we will try unlocking a spinlock that was never locked via a garbage pointer.

At the time we reach this code path, we usually successfully looked up a PGSTE already; however, evil user space could have manipulated the VMA layout in the meantime and triggered removal of the page table.

---
 arch/s390/mm/gmap.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
arch/s390/mm/gmap.c +3 -2
index e0735c3437759..d63c0ccc5ccda 100644
@@ -689,9 +689,10 @@ void __gmap_zap(struct gmap *gmap, unsigned long gaddr)
 
 		/* Get pointer to the page table entry */
 		ptep = get_locked_pte(gmap->mm, vmaddr, &ptl);
-		if (likely(ptep))
+		if (likely(ptep)) {
 			ptep_zap_unused(gmap->mm, vmaddr, ptep, 0);
-		pte_unmap_unlock(ptep, ptl);
+			pte_unmap_unlock(ptep, ptl);
+		}
 	}
 }
 EXPORT_SYMBOL_GPL(__gmap_zap);
Signature · 1 line
2.33.0
author maintainer
#519 lore raw

PATCH 508/917 s390/mm: validate VMA in PGSTE manipulation functions

In reply to Greg Kroah-Hartman

From: David Hildenbrand <[email protected]>

[ Upstream commit fe3d10024073f06f04c74b9674bd71ccc1d787cf ]

We should not walk/touch page tables outside of VMA boundaries when holding only the mmap sem in read mode. Evil user space can modify the VMA layout just before this function runs and e.g., trigger races with page table removal code since commit dd2283f2605e ("mm: mmap: zap pages with read mmap_sem in munmap"). gfn_to_hva() will only translate using KVM memory regions, but won't validate the VMA.

Further, we should not allocate page tables outside of VMA boundaries: if evil user space decides to map hugetlbfs to these ranges, bad things will happen because we suddenly have PTE or PMD page tables where we shouldn't have them.

Similarly, we have to check if we suddenly find a hugetlbfs VMA, before calling get_locked_pte().

---
 arch/s390/mm/pgtable.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)
arch/s390/mm/pgtable.c +13 -0
index 034721a68d8fd..2717a406edeb3 100644
@@ -988,6 +988,7 @@ EXPORT_SYMBOL(get_guest_storage_key);
 int pgste_perform_essa(struct mm_struct *mm, unsigned long hva, int orc,
 			unsigned long *oldpte, unsigned long *oldpgste)
 {
+	struct vm_area_struct *vma;
 	unsigned long pgstev;
 	spinlock_t *ptl;
 	pgste_t pgste;
@@ -997,6 +998,10 @@ int pgste_perform_essa(struct mm_struct *mm, unsigned long hva, int orc,
 	WARN_ON_ONCE(orc > ESSA_MAX);
 	if (unlikely(orc > ESSA_MAX))
 		return -EINVAL;
+
+	vma = vma_lookup(mm, hva);
+	if (!vma || is_vm_hugetlb_page(vma))
+		return -EFAULT;
 	ptep = get_locked_pte(mm, hva, &ptl);
 	if (unlikely(!ptep))
 		return -EFAULT;
@@ -1089,10 +1094,14 @@ EXPORT_SYMBOL(pgste_perform_essa);
 int set_pgste_bits(struct mm_struct *mm, unsigned long hva,
 			unsigned long bits, unsigned long value)
 {
+	struct vm_area_struct *vma;
 	spinlock_t *ptl;
 	pgste_t new;
 	pte_t *ptep;
 
+	vma = vma_lookup(mm, hva);
+	if (!vma || is_vm_hugetlb_page(vma))
+		return -EFAULT;
 	ptep = get_locked_pte(mm, hva, &ptl);
 	if (unlikely(!ptep))
 		return -EFAULT;
@@ -1117,9 +1126,13 @@ EXPORT_SYMBOL(set_pgste_bits);
  */
 int get_pgste(struct mm_struct *mm, unsigned long hva, unsigned long *pgstep)
 {
+	struct vm_area_struct *vma;
 	spinlock_t *ptl;
 	pte_t *ptep;
 
+	vma = vma_lookup(mm, hva);
+	if (!vma || is_vm_hugetlb_page(vma))
+		return -EFAULT;
 	ptep = get_locked_pte(mm, hva, &ptl);
 	if (unlikely(!ptep))
 		return -EFAULT;
Signature · 1 line
2.33.0
author maintainer
#520 lore raw

PATCH 509/917 s390/mm: fix VMA and page table handling code in storage key handling functions

In reply to Greg Kroah-Hartman

From: David Hildenbrand <[email protected]>

[ Upstream commit 949f5c1244ee6c36d2e81c588d1200eaa83a3df6 ]

There are multiple things broken about our storage key handling functions:

1. We should not walk/touch page tables outside of VMA boundaries when holding only the mmap sem in read mode. Evil user space can modify the VMA layout just before this function runs and e.g., trigger races with page table removal code since commit dd2283f2605e ("mm: mmap: zap pages with read mmap_sem in munmap"). gfn_to_hva() will only translate using KVM memory regions, but won't validate the VMA.

2. We should not allocate page tables outside of VMA boundaries: if evil user space decides to map hugetlbfs to these ranges, bad things will happen because we suddenly have PTE or PMD page tables where we shouldn't have them.

3. We don't handle large PUDs that might suddenly appeared inside our page table hierarchy.

Don't manually allocate page tables, properly validate that we have VMA and bail out on pud_large().

All callers of page table handling functions, except get_guest_storage_key(), call fixup_user_fault() in case they receive an -EFAULT and retry; this will allocate the necessary page tables if required.

To keep get_guest_storage_key() working as expected and not requiring kvm_s390_get_skeys() to call fixup_user_fault() distinguish between "there is simply no page table or huge page yet and the key is assumed to be 0" and "this is a fault to be reported".

Although commit 637ff9efe5ea ("s390/mm: Add huge pmd storage key handling") introduced most of the affected code, it was actually already broken before when using get_locked_pte() without any VMA checks.

Note: Ever since commit 637ff9efe5ea ("s390/mm: Add huge pmd storage key handling") we can no longer set a guest storage key (for example from QEMU during VM live migration) without actually resolving a fault. Although we would have created most page tables, we would choke on the !pmd_present(), requiring a call to fixup_user_fault(). I would have thought that this is problematic in combination with postcopy life migration ... but nobody noticed and this patch doesn't change the situation. So maybe it's just fine.

Fixes 9fcf93b5de06 ("KVM: S390: Create helper function get_guest_storage_key")
Fixes 24d5dd0208ed ("s390/kvm: Provide function for setting the guest storage key")
Fixes a7e19ab55ffd ("KVM: s390: handle missing storage-key facility")
---
 arch/s390/mm/pgtable.c | 57 +++++++++++++++++++++++++++++-------------
 1 file changed, 39 insertions(+), 18 deletions(-)
arch/s390/mm/pgtable.c +39 -18
index 2717a406edeb3..6ad634a27d5b9 100644
@@ -429,22 +429,36 @@ static inline pmd_t pmdp_flush_lazy(struct mm_struct *mm,
 }
 
 #ifdef CONFIG_PGSTE
-static pmd_t *pmd_alloc_map(struct mm_struct *mm, unsigned long addr)
+static int pmd_lookup(struct mm_struct *mm, unsigned long addr, pmd_t **pmdp)
 {
+	struct vm_area_struct *vma;
 	pgd_t *pgd;
 	p4d_t *p4d;
 	pud_t *pud;
-	pmd_t *pmd;
+
+	/* We need a valid VMA, otherwise this is clearly a fault. */
+	vma = vma_lookup(mm, addr);
+	if (!vma)
+		return -EFAULT;
 
 	pgd = pgd_offset(mm, addr);
-	p4d = p4d_alloc(mm, pgd, addr);
-	if (!p4d)
-		return NULL;
-	pud = pud_alloc(mm, p4d, addr);
-	if (!pud)
-		return NULL;
-	pmd = pmd_alloc(mm, pud, addr);
-	return pmd;
+	if (!pgd_present(*pgd))
+		return -ENOENT;
+
+	p4d = p4d_offset(pgd, addr);
+	if (!p4d_present(*p4d))
+		return -ENOENT;
+
+	pud = pud_offset(p4d, addr);
+	if (!pud_present(*pud))
+		return -ENOENT;
+
+	/* Large PUDs are not supported yet. */
+	if (pud_large(*pud))
+		return -EFAULT;
+
+	*pmdp = pmd_offset(pud, addr);
+	return 0;
 }
 #endif
 
@@ -778,8 +792,7 @@ int set_guest_storage_key(struct mm_struct *mm, unsigned long addr,
 	pmd_t *pmdp;
 	pte_t *ptep;
 
-	pmdp = pmd_alloc_map(mm, addr);
-	if (unlikely(!pmdp))
+	if (pmd_lookup(mm, addr, &pmdp))
 		return -EFAULT;
 
 	ptl = pmd_lock(mm, pmdp);
@@ -881,8 +894,7 @@ int reset_guest_reference_bit(struct mm_struct *mm, unsigned long addr)
 	pte_t *ptep;
 	int cc = 0;
 
-	pmdp = pmd_alloc_map(mm, addr);
-	if (unlikely(!pmdp))
+	if (pmd_lookup(mm, addr, &pmdp))
 		return -EFAULT;
 
 	ptl = pmd_lock(mm, pmdp);
@@ -935,15 +947,24 @@ int get_guest_storage_key(struct mm_struct *mm, unsigned long addr,
 	pmd_t *pmdp;
 	pte_t *ptep;
 
-	pmdp = pmd_alloc_map(mm, addr);
-	if (unlikely(!pmdp))
+	/*
+	 * If we don't have a PTE table and if there is no huge page mapped,
+	 * the storage key is 0.
+	 */
+	*key = 0;
+
+	switch (pmd_lookup(mm, addr, &pmdp)) {
+	case -ENOENT:
+		return 0;
+	case 0:
+		break;
+	default:
 		return -EFAULT;
+	}
 
 	ptl = pmd_lock(mm, pmdp);
 	if (!pmd_present(*pmdp)) {
-		/* Not yet mapped memory has a zero key */
 		spin_unlock(ptl);
-		*key = 0;
 		return 0;
 	}
 
Signature · 1 line
2.33.0
author maintainer
#521 lore raw

PATCH 510/917 s390/uv: fully validate the VMA before calling follow_page()

In reply to Greg Kroah-Hartman

From: David Hildenbrand <[email protected]>

[ Upstream commit 46c22ffd2772201662350bc7b94b9ea9d3ee5ac2 ]

We should not walk/touch page tables outside of VMA boundaries when holding only the mmap sem in read mode. Evil user space can modify the VMA layout just before this function runs and e.g., trigger races with page table removal code since commit dd2283f2605e ("mm: mmap: zap pages with read mmap_sem in munmap").

find_vma() does not check if the address is >= the VMA start address; use vma_lookup() instead.

---
 arch/s390/kernel/uv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/s390/kernel/uv.c +1 -1
index 5a656c7b7a67a..f95ccbd396925 100644
@@ -212,7 +212,7 @@ again:
 	uaddr = __gmap_translate(gmap, gaddr);
 	if (IS_ERR_VALUE(uaddr))
 		goto out;
-	vma = find_vma(gmap->mm, uaddr);
+	vma = vma_lookup(gmap->mm, uaddr);
 	if (!vma)
 		goto out;
 	/*
Signature · 1 line
2.33.0
author maintainer
#522 lore raw

PATCH 511/917 KVM: s390: pv: avoid double free of sida page

In reply to Greg Kroah-Hartman

From: Claudio Imbrenda <[email protected]>

[ Upstream commit d4074324b07a94a1fca476d452dfbb3a4e7bf656 ]

If kvm_s390_pv_destroy_cpu is called more than once, we risk calling free_page on a random page, since the sidad field is aliased with the gbea, which is not guaranteed to be zero.

This can happen, for example, if userspace calls the KVM_PV_DISABLE IOCTL, and it fails, and then userspace calls the same IOCTL again. This scenario is only possible if KVM has some serious bug or if the hardware is broken.

The solution is to simply return successfully immediately if the vCPU was already non secure.

Fixes 19e1227768863a1469797c13ef8fea1af7beac2c ("KVM: S390: protvirt: Introduce instruction data area bounce buffer")

Message-Id: <[email protected]>

---
 arch/s390/kvm/pv.c | 19 +++++++++----------
 1 file changed, 9 insertions(+), 10 deletions(-)
arch/s390/kvm/pv.c +9 -10
index c8841f476e913..0a854115100b4 100644
@@ -16,18 +16,17 @@
 
 int kvm_s390_pv_destroy_cpu(struct kvm_vcpu *vcpu, u16 *rc, u16 *rrc)
 {
-	int cc = 0;
+	int cc;
 
-	if (kvm_s390_pv_cpu_get_handle(vcpu)) {
-		cc = uv_cmd_nodata(kvm_s390_pv_cpu_get_handle(vcpu),
-				   UVC_CMD_DESTROY_SEC_CPU, rc, rrc);
+	if (!kvm_s390_pv_cpu_get_handle(vcpu))
+		return 0;
+
+	cc = uv_cmd_nodata(kvm_s390_pv_cpu_get_handle(vcpu), UVC_CMD_DESTROY_SEC_CPU, rc, rrc);
+
+	KVM_UV_EVENT(vcpu->kvm, 3, "PROTVIRT DESTROY VCPU %d: rc %x rrc %x",
+		     vcpu->vcpu_id, *rc, *rrc);
+	WARN_ONCE(cc, "protvirt destroy cpu failed rc %x rrc %x", *rc, *rrc);
 
-		KVM_UV_EVENT(vcpu->kvm, 3,
-			     "PROTVIRT DESTROY VCPU %d: rc %x rrc %x",
-			     vcpu->vcpu_id, *rc, *rrc);
-		WARN_ONCE(cc, "protvirt destroy cpu failed rc %x rrc %x",
-			  *rc, *rrc);
-	}
 	/* Intended memory leak for something that should never happen. */
 	if (!cc)
 		free_pages(vcpu->arch.pv.stor_base,
Signature · 1 line
2.33.0
author maintainer
#523 lore raw

PATCH 512/917 KVM: s390: pv: avoid stalls for kvm_s390_pv_init_vm

In reply to Greg Kroah-Hartman

From: Claudio Imbrenda <[email protected]>

[ Upstream commit 1e2aa46de526a5adafe580bca4c25856bb06f09e ]

When the system is heavily overcommitted, kvm_s390_pv_init_vm might generate stall notifications.

Fix this by using uv_call_sched instead of just uv_call. This is ok because we are not holding spinlocks.

Fixes 214d9bbcd3a672 ("s390/mm: provide memory management functions for protected KVM guests")

Message-Id: <[email protected]>

---
 arch/s390/kvm/pv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/s390/kvm/pv.c +1 -1
index 0a854115100b4..00d272d134c24 100644
@@ -195,7 +195,7 @@ int kvm_s390_pv_init_vm(struct kvm *kvm, u16 *rc, u16 *rrc)
 	uvcb.conf_base_stor_origin = (u64)kvm->arch.pv.stor_base;
 	uvcb.conf_virt_stor_origin = (u64)kvm->arch.pv.stor_var;
 
-	cc = uv_call(0, (u64)&uvcb);
+	cc = uv_call_sched(0, (u64)&uvcb);
 	*rc = uvcb.header.rc;
 	*rrc = uvcb.header.rrc;
 	KVM_UV_EVENT(kvm, 3, "PROTVIRT CREATE VM: handle %llx len %llx rc %x rrc %x",
Signature · 1 line
2.33.0
author maintainer
#524 lore raw

PATCH 513/917 irq: mips: avoid nested irq_enter()

In reply to Greg Kroah-Hartman

From: Mark Rutland <[email protected]>

[ Upstream commit c65b52d02f6c1a06ddb20cba175ad49eccd6410d ]

As bcm6345_l1_irq_handle() is a chained irqchip handler, it will be invoked within the context of the root irqchip handler, which must have entered IRQ context already.

When bcm6345_l1_irq_handle() calls arch/mips's do_IRQ() , this will nest another call to irq_enter(), and the resulting nested increment to rcu_data.dynticks_nmi_nesting will cause rcu_is_cpu_rrupt_from_idle() to fail to identify wakeups from idle, resulting in failure to preempt, and RCU stalls.

Chained irqchip handlers must invoke IRQ handlers by way of thee core irqchip code, i.e. generic_handle_irq() or generic_handle_domain_irq() and should not call do_IRQ(), which is intended only for root irqchip handlers.

Fix bcm6345_l1_irq_handle() by calling generic_handle_irq() directly.

Fixes c7c42ec2baa1de7a ("irqchips/bmips: Add bcm6345-l1 interrupt controller")
Cc Thomas Gleixner <[email protected]>
---
 drivers/irqchip/irq-bcm6345-l1.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/irqchip/irq-bcm6345-l1.c +1 -1
index e3483789f4df3..1bd0621c4ce2a 100644
@@ -140,7 +140,7 @@ static void bcm6345_l1_irq_handle(struct irq_desc *desc)
 		for_each_set_bit(hwirq, &pending, IRQS_PER_WORD) {
 			irq = irq_linear_revmap(intc->domain, base + hwirq);
 			if (irq)
-				do_IRQ(irq);
+				generic_handle_irq(irq);
 			else
 				spurious_interrupt();
 		}
Signature · 1 line
2.33.0
author maintainer
#525 lore raw

PATCH 514/917 net: dsa: avoid refcount warnings when ->port_{fdb,mdb}_del returns error

In reply to Greg Kroah-Hartman

From: Vladimir Oltean <[email protected]>

[ Upstream commit 232deb3f9567ce37d99b8616a6c07c1fc0436abf ]

At present, when either of ds->ops->port_fdb_del() or ds->ops->port_mdb_del() return a non-zero error code, we attempt to save the day and keep the data structure associated with that switchdev object, as the deletion procedure did not complete.

However, the way in which we do this is suspicious to the checker in lib/refcount.c, who thinks it is buggy to increment a refcount that became zero, and that this is indicative of a use-after-free.

Fixes 161ca59d39e9 ("net: dsa: reference count the MDB entries at the cross-chip notifier level")
---
 net/dsa/switch.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
net/dsa/switch.c +2 -2
index 6466d0539af9f..44558fbdc65b3 100644
@@ -264,7 +264,7 @@ static int dsa_switch_do_mdb_del(struct dsa_switch *ds, int port,
 
 	err = ds->ops->port_mdb_del(ds, port, mdb);
 	if (err) {
-		refcount_inc(&a->refcount);
+		refcount_set(&a->refcount, 1);
 		return err;
 	}
 
@@ -329,7 +329,7 @@ static int dsa_switch_do_fdb_del(struct dsa_switch *ds, int port,
 
 	err = ds->ops->port_fdb_del(ds, port, addr, vid);
 	if (err) {
-		refcount_inc(&a->refcount);
+		refcount_set(&a->refcount, 1);
 		return err;
 	}
 
Signature · 1 line
2.33.0
author maintainer
#526 lore raw

PATCH 515/917 ARM: 9142/1: kasan: work around LPAE build warning

In reply to Greg Kroah-Hartman

From: Arnd Bergmann <[email protected]>

[ Upstream commit c2e6df3eaaf120cde5e7c3a70590dd82e427458a ]

pgd_page_vaddr() returns an 'unsigned long' address, causing a warning with the memcpy() call in kasan_init():

arch/arm/mm/kasan_init.c: In function 'kasan_init':
include/asm-generic/pgtable-nop4d.h:44:50: error: passing argument 2 of '__memcpy' makes pointer from integer without a cast [-Werror=int-conversion]

   44 | #define pgd_page_vaddr(pgd)                     ((unsigned long)(p4d_pgtable((p4d_t){ pgd })))
      |                                                 ~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
      |                                                  |
      |                                                  long unsigned int

arch/arm/include/asm/string.h:58:45: note: in definition of macro 'memcpy'
58 | #define memcpy(dst, src, len) __memcpy(dst, src, len)
| ^~~
arch/arm/mm/kasan_init.c:229:16: note: in expansion of macro 'pgd_page_vaddr'

  229 |                pgd_page_vaddr(*pgd_offset_k(KASAN_SHADOW_START)),
      |                ^~~~~~~~~~~~~~
arch/arm/include/asm/string.h:21:47: note: expected 'const void *' but argument is of type 'long unsigned int'
   21 | extern void *__memcpy(void *dest, const void *src, __kernel_size_t n);
      |                                   ~~~~~~~~~~~~^~~

Avoid this by adding an explicit typecast.

Fixes 5615f69bc209 ("ARM: 9016/2: Initialize the mapping of KASan shadow memory")
---
 arch/arm/mm/kasan_init.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/mm/kasan_init.c +1 -1
index 9c348042a7244..4b1619584b23c 100644
@@ -226,7 +226,7 @@ void __init kasan_init(void)
 	BUILD_BUG_ON(pgd_index(KASAN_SHADOW_START) !=
 		     pgd_index(KASAN_SHADOW_END));
 	memcpy(tmp_pmd_table,
-	       pgd_page_vaddr(*pgd_offset_k(KASAN_SHADOW_START)),
+	       (void*)pgd_page_vaddr(*pgd_offset_k(KASAN_SHADOW_START)),
 	       sizeof(tmp_pmd_table));
 	set_pgd(&tmp_pgd_table[pgd_index(KASAN_SHADOW_START)],
 		__pgd(__pa(tmp_pmd_table) | PMD_TYPE_TABLE | L_PGD_SWAPPER));
Signature · 1 line
2.33.0
author maintainer
#527 lore raw

PATCH 516/917 ath10k: fix module load regression with iram-recovery feature

In reply to Greg Kroah-Hartman

From: Abinaya Kalaiselvan <[email protected]>

[ Upstream commit 6f8c8bf4c7c9be1c42088689fd4370e06b46608a ]

Commit 9af7c32ceca8 ("ath10k: add target IRAM recovery feature support") introduced a new firmware feature flag ATH10K_FW_FEATURE_IRAM_RECOVERY. But this caused ath10k_pci module load to fail if ATH10K_FW_CRASH_DUMP_RAM_DATA bit was not enabled in the ath10k coredump_mask module parameter:

[ 2209.328190] ath10k_pci 0000:02:00.0: qca9984/qca9994 hw1.0 target 0x01000000 chip_id 0x00000000 sub 168c:cafe
[ 2209.434414] ath10k_pci 0000:02:00.0: kconfig debug 1 debugfs 1 tracing 1 dfs 1 testmode 1
[ 2209.547191] ath10k_pci 0000:02:00.0: firmware ver 10.4-3.9.0.2-00099 api 5 features no-p2p,mfp,peer-flow-ctrl,btcoex-param,allows-mesh-bcast,no-ps,peer-fixed-rate,iram-recovery crc32 cbade90a
[ 2210.896485] ath10k_pci 0000:02:00.0: board_file api 1 bmi_id 0:1 crc32 a040efc2
[ 2213.603339] ath10k_pci 0000:02:00.0: failed to copy target iram contents: -12
[ 2213.839027] ath10k_pci 0000:02:00.0: could not init core (-12)
[ 2213.933910] ath10k_pci 0000:02:00.0: could not probe fw (-12)

And by default coredump_mask does not have ATH10K_FW_CRASH_DUMP_RAM_DATA enabled so anyone using a firmware with iram-recovery feature would fail. To my knowledge only QCA9984 firmwares starting from release 10.4-3.9.0.2-00099 enabled the feature.

The reason for regression was that ath10k_core_copy_target_iram() used ath10k_coredump_get_mem_layout() to get the memory layout, but when ATH10K_FW_CRASH_DUMP_RAM_DATA was disabled it would get just NULL and bail out with an error.

While looking at all this I noticed another bug: if CONFIG_DEV_COREDUMP is disabled but the firmware has iram-recovery enabled the module load fails with similar error messages. I fixed that by returning 0 from ath10k_core_copy_target_iram() when _ath10k_coredump_get_mem_layout() returns NULL.

Tested-on: QCA9984 hw2.0 PCI 10.4-3.9.0.2-00139

Fixes 9af7c32ceca8 ("ath10k: add target IRAM recovery feature support")
---
 drivers/net/wireless/ath/ath10k/core.c     | 11 +++++++++--
 drivers/net/wireless/ath/ath10k/coredump.c | 11 ++++++++---
 drivers/net/wireless/ath/ath10k/coredump.h |  7 +++++++
 3 files changed, 24 insertions(+), 5 deletions(-)
drivers/net/wireless/ath/ath10k/core.c +9 -2
index 2f9be182fbfbb..64c7145b51a2e 100644
@@ -2690,9 +2690,16 @@ static int ath10k_core_copy_target_iram(struct ath10k *ar)
 	int i, ret;
 	u32 len, remaining_len;
 
-	hw_mem = ath10k_coredump_get_mem_layout(ar);
+	/* copy target iram feature must work also when
+	 * ATH10K_FW_CRASH_DUMP_RAM_DATA is disabled, so
+	 * _ath10k_coredump_get_mem_layout() to accomplist that
+	 */
+	hw_mem = _ath10k_coredump_get_mem_layout(ar);
 	if (!hw_mem)
-		return -ENOMEM;
+		/* if CONFIG_DEV_COREDUMP is disabled we get NULL, then
+		 * just silently disable the feature by doing nothing
+		 */
+		return 0;
 
 	for (i = 0; i < hw_mem->region_table.size; i++) {
 		tmp = &hw_mem->region_table.regions[i];
drivers/net/wireless/ath/ath10k/coredump.c +8 -3
index 7eb72290a925c..55e7e11d06d94 100644
@@ -1447,11 +1447,17 @@ static u32 ath10k_coredump_get_ramdump_size(struct ath10k *ar)
 
 const struct ath10k_hw_mem_layout *ath10k_coredump_get_mem_layout(struct ath10k *ar)
 {
-	int i;
-
 	if (!test_bit(ATH10K_FW_CRASH_DUMP_RAM_DATA, &ath10k_coredump_mask))
 		return NULL;
 
+	return _ath10k_coredump_get_mem_layout(ar);
+}
+EXPORT_SYMBOL(ath10k_coredump_get_mem_layout);
+
+const struct ath10k_hw_mem_layout *_ath10k_coredump_get_mem_layout(struct ath10k *ar)
+{
+	int i;
+
 	if (WARN_ON(ar->target_version == 0))
 		return NULL;
 
@@ -1464,7 +1470,6 @@ const struct ath10k_hw_mem_layout *ath10k_coredump_get_mem_layout(struct ath10k
 
 	return NULL;
 }
-EXPORT_SYMBOL(ath10k_coredump_get_mem_layout);
 
 struct ath10k_fw_crash_data *ath10k_coredump_new(struct ath10k *ar)
 {
drivers/net/wireless/ath/ath10k/coredump.h +7 -0
index 42404e246e0e9..240d705150888 100644
@@ -176,6 +176,7 @@ int ath10k_coredump_register(struct ath10k *ar);
 void ath10k_coredump_unregister(struct ath10k *ar);
 void ath10k_coredump_destroy(struct ath10k *ar);
 
+const struct ath10k_hw_mem_layout *_ath10k_coredump_get_mem_layout(struct ath10k *ar);
 const struct ath10k_hw_mem_layout *ath10k_coredump_get_mem_layout(struct ath10k *ar);
 
 #else /* CONFIG_DEV_COREDUMP */
@@ -214,6 +215,12 @@ ath10k_coredump_get_mem_layout(struct ath10k *ar)
 	return NULL;
 }
 
+static inline const struct ath10k_hw_mem_layout *
+_ath10k_coredump_get_mem_layout(struct ath10k *ar)
+{
+	return NULL;
+}
+
 #endif /* CONFIG_DEV_COREDUMP */
 
 #endif /* _COREDUMP_H_ */
Signature · 1 line
2.33.0
author maintainer
#528 lore raw

PATCH 517/917 block: ataflop: more blk-mq refactoring fixes

In reply to Greg Kroah-Hartman

From: Michael Schmitz <[email protected]>

[ Upstream commit d28e4dff085c5a87025c9a0a85fb798bd8e9ca17 ]

As it turns out, my earlier patch in commit 86d46fdaa12a (block:
ataflop: fix breakage introduced at blk-mq refactoring) was incomplete. This patch fixes any remaining issues found during more testing and code review.

Requests exceeding 4 k are handled in 4k segments but __blk_mq_end_request() is never called on these (still sectors outstanding on the request). With redo_fd_request() removed, there is no provision to kick off processing of the next segment, causing requests exceeding 4k to hang. (By setting /sys/block/fd0/queue/max_sectors_k <= 4 as workaround, this behaviour can be avoided).

Instead of reintroducing redo_fd_request(), requeue the remainder of the request by calling blk_mq_requeue_request() on incomplete requests (i.e. when blk_update_request() still returns true), and rely on the block layer to queue the residual as new request.

Both error handling and formatting needs to release the ST-DMA lock, so call finish_fdc() on these (this was previously handled by redo_fd_request()). finish_fdc() may be called legitimately without the ST-DMA lock held - make sure we only release the lock if we actually held it. In a similar way, early exit due to errors in ataflop_queue_rq() must release the lock.

After minor errors, fd_error sets up to recalibrate the drive but never re-runs the current operation (another task handled by redo_fd_request() before). Call do_fd_action() to get the next steps (seek, retry read/write) underway.

---
 drivers/block/ataflop.c | 45 +++++++++++++++++++++++++++++++++++------
 1 file changed, 39 insertions(+), 6 deletions(-)
drivers/block/ataflop.c +39 -6
index bbb64331cf8f4..4947e41f89b7d 100644
@@ -456,10 +456,20 @@ static DEFINE_TIMER(fd_timer, check_change);
 	
 static void fd_end_request_cur(blk_status_t err)
 {
+	DPRINT(("fd_end_request_cur(), bytes %d of %d\n",
+		blk_rq_cur_bytes(fd_request),
+		blk_rq_bytes(fd_request)));
+
 	if (!blk_update_request(fd_request, err,
 				blk_rq_cur_bytes(fd_request))) {
+		DPRINT(("calling __blk_mq_end_request()\n"));
 		__blk_mq_end_request(fd_request, err);
 		fd_request = NULL;
+	} else {
+		/* requeue rest of request */
+		DPRINT(("calling blk_mq_requeue_request()\n"));
+		blk_mq_requeue_request(fd_request, true);
+		fd_request = NULL;
 	}
 }
 
@@ -697,12 +707,21 @@ static void fd_error( void )
 	if (fd_request->error_count >= MAX_ERRORS) {
 		printk(KERN_ERR "fd%d: too many errors.\n", SelectedDrive );
 		fd_end_request_cur(BLK_STS_IOERR);
+		finish_fdc();
+		return;
 	}
 	else if (fd_request->error_count == RECALIBRATE_ERRORS) {
 		printk(KERN_WARNING "fd%d: recalibrating\n", SelectedDrive );
 		if (SelectedDrive != -1)
 			SUD.track = -1;
 	}
+	/* need to re-run request to recalibrate */
+	atari_disable_irq( IRQ_MFP_FDC );
+
+	setup_req_params( SelectedDrive );
+	do_fd_action( SelectedDrive );
+
+	atari_enable_irq( IRQ_MFP_FDC );
 }
 
 
@@ -729,8 +748,10 @@ static int do_format(int drive, int type, struct atari_format_descr *desc)
 	if (type) {
 		type--;
 		if (type >= NUM_DISK_MINORS ||
-		    minor2disktype[type].drive_types > DriveType)
+		    minor2disktype[type].drive_types > DriveType) {
+			finish_fdc();
 			return -EINVAL;
+		}
 	}
 
 	q = unit[drive].disk[type]->queue;
@@ -748,6 +769,7 @@ static int do_format(int drive, int type, struct atari_format_descr *desc)
 	}
 
 	if (!UDT || desc->track >= UDT->blocks/UDT->spt/2 || desc->head >= 2) {
+		finish_fdc();
 		ret = -EINVAL;
 		goto out;
 	}
@@ -788,6 +810,7 @@ static int do_format(int drive, int type, struct atari_format_descr *desc)
 
 	wait_for_completion(&format_wait);
 
+	finish_fdc();
 	ret = FormatError ? -EIO : 0;
 out:
 	blk_mq_unquiesce_queue(q);
@@ -822,6 +845,7 @@ static void do_fd_action( int drive )
 		    else {
 			/* all sectors finished */
 			fd_end_request_cur(BLK_STS_OK);
+			finish_fdc();
 			return;
 		    }
 		}
@@ -1225,8 +1249,8 @@ static void fd_rwsec_done1(int status)
 	}
 	else {
 		/* all sectors finished */
-		finish_fdc();
 		fd_end_request_cur(BLK_STS_OK);
+		finish_fdc();
 	}
 	return;
   
@@ -1348,7 +1372,7 @@ static void fd_times_out(struct timer_list *unused)
 
 static void finish_fdc( void )
 {
-	if (!NeedSeek) {
+	if (!NeedSeek || !stdma_is_locked_by(floppy_irq)) {
 		finish_fdc_done( 0 );
 	}
 	else {
@@ -1383,7 +1407,8 @@ static void finish_fdc_done( int dummy )
 	start_motor_off_timer();
 
 	local_irq_save(flags);
-	stdma_release();
+	if (stdma_is_locked_by(floppy_irq))
+		stdma_release();
 	local_irq_restore(flags);
 
 	DPRINT(("finish_fdc() finished\n"));
@@ -1480,7 +1505,9 @@ static blk_status_t ataflop_queue_rq(struct blk_mq_hw_ctx *hctx,
 	int drive = floppy - unit;
 	int type = floppy->type;
 
-	DPRINT(("Queue request: drive %d type %d last %d\n", drive, type, bd->last));
+	DPRINT(("Queue request: drive %d type %d sectors %d of %d last %d\n",
+		drive, type, blk_rq_cur_sectors(bd->rq),
+		blk_rq_sectors(bd->rq), bd->last));
 
 	spin_lock_irq(&ataflop_lock);
 	if (fd_request) {
@@ -1502,6 +1529,7 @@ static blk_status_t ataflop_queue_rq(struct blk_mq_hw_ctx *hctx,
 		/* drive not connected */
 		printk(KERN_ERR "Unknown Device: fd%d\n", drive );
 		fd_end_request_cur(BLK_STS_IOERR);
+		stdma_release();
 		goto out;
 	}
 		
@@ -1518,11 +1546,13 @@ static blk_status_t ataflop_queue_rq(struct blk_mq_hw_ctx *hctx,
 		if (--type >= NUM_DISK_MINORS) {
 			printk(KERN_WARNING "fd%d: invalid disk format", drive );
 			fd_end_request_cur(BLK_STS_IOERR);
+			stdma_release();
 			goto out;
 		}
 		if (minor2disktype[type].drive_types > DriveType)  {
 			printk(KERN_WARNING "fd%d: unsupported disk format", drive );
 			fd_end_request_cur(BLK_STS_IOERR);
+			stdma_release();
 			goto out;
 		}
 		type = minor2disktype[type].index;
@@ -1623,6 +1653,7 @@ static int fd_locked_ioctl(struct block_device *bdev, fmode_t mode,
 		/* what if type > 0 here? Overwrite specified entry ? */
 		if (type) {
 		        /* refuse to re-set a predefined type for now */
+			finish_fdc();
 			return -EINVAL;
 		}
 
@@ -1690,8 +1721,10 @@ static int fd_locked_ioctl(struct block_device *bdev, fmode_t mode,
 
 		/* sanity check */
 		if (setprm.track != dtp->blocks/dtp->spt/2 ||
-		    setprm.head != 2)
+		    setprm.head != 2) {
+			finish_fdc();
 			return -EINVAL;
+		}
 
 		UDT = dtp;
 		set_capacity(disk, UDT->blocks);
Signature · 1 line
2.33.0
author maintainer
#529 lore raw

PATCH 518/917 blk-cgroup: synchronize blkg creation against policy deactivation

In reply to Greg Kroah-Hartman

From: Yu Kuai <[email protected]>

[ Upstream commit 0c9d338c8443b06da8e8d3bfce824c5ea6d3488f ]

Our test reports a null pointer dereference:

[  168.534653] ==================================================================
[  168.535614] Disabling lock debugging due to kernel taint
[  168.536346] BUG: kernel NULL pointer dereference, address: 0000000000000008
[  168.537274] #PF: supervisor read access in kernel mode
[  168.537964] #PF: error_code(0x0000) - not-present page
[  168.538667] PGD 0 P4D 0
[  168.539025] Oops: 0000 [#1] PREEMPT SMP KASAN
[  168.539656] CPU: 13 PID: 759 Comm: bash Tainted: G    B             5.15.0-rc2-next-202100
[  168.540954] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS ?-20190727_0738364
[  168.542736] RIP: 0010:bfq_pd_init+0x88/0x1e0
[  168.543318] Code: 98 00 00 00 e8 c9 e4 5b ff 4c 8b 65 00 49 8d 7c 24 08 e8 bb e4 5b ff 4d0
[  168.545803] RSP: 0018:ffff88817095f9c0 EFLAGS: 00010002
[  168.546497] RAX: 0000000000000001 RBX: ffff888101a1c000 RCX: 0000000000000000
[  168.547438] RDX: 0000000000000003 RSI: 0000000000000002 RDI: ffff888106553428
[  168.548402] RBP: ffff888106553400 R08: ffffffff961bcaf4 R09: 0000000000000001
[  168.549365] R10: ffffffffa2e16c27 R11: fffffbfff45c2d84 R12: 0000000000000000
[  168.550291] R13: ffff888101a1c098 R14: ffff88810c7a08c8 R15: ffffffffa55541a0
[  168.551221] FS:  00007fac75227700(0000) GS:ffff88839ba80000(0000) knlGS:0000000000000000
[  168.552278] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  168.553040] CR2: 0000000000000008 CR3: 0000000165ce7000 CR4: 00000000000006e0
[  168.554000] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[  168.554929] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[  168.555888] Call Trace:
[  168.556221]  <TASK>
[  168.556510]  blkg_create+0x1c0/0x8c0
[  168.556989]  blkg_conf_prep+0x574/0x650
[  168.557502]  ? stack_trace_save+0x99/0xd0
[  168.558033]  ? blkcg_conf_open_bdev+0x1b0/0x1b0
[  168.558629]  tg_set_conf.constprop.0+0xb9/0x280
[  168.559231]  ? kasan_set_track+0x29/0x40
[  168.559758]  ? kasan_set_free_info+0x30/0x60
[  168.560344]  ? tg_set_limit+0xae0/0xae0
[  168.560853]  ? do_sys_openat2+0x33b/0x640
[  168.561383]  ? do_sys_open+0xa2/0x100
[  168.561877]  ? __x64_sys_open+0x4e/0x60
[  168.562383]  ? __kasan_check_write+0x20/0x30
[  168.562951]  ? copyin+0x48/0x70
[  168.563390]  ? _copy_from_iter+0x234/0x9e0
[  168.563948]  tg_set_conf_u64+0x17/0x20
[  168.564467]  cgroup_file_write+0x1ad/0x380
[  168.565014]  ? cgroup_file_poll+0x80/0x80
[  168.565568]  ? __mutex_lock_slowpath+0x30/0x30
[  168.566165]  ? pgd_free+0x100/0x160
[  168.566649]  kernfs_fop_write_iter+0x21d/0x340
[  168.567246]  ? cgroup_file_poll+0x80/0x80
[  168.567796]  new_sync_write+0x29f/0x3c0
[  168.568314]  ? new_sync_read+0x410/0x410
[  168.568840]  ? __handle_mm_fault+0x1c97/0x2d80
[  168.569425]  ? copy_page_range+0x2b10/0x2b10
[  168.570007]  ? _raw_read_lock_bh+0xa0/0xa0
[  168.570622]  vfs_write+0x46e/0x630
[  168.571091]  ksys_write+0xcd/0x1e0
[  168.571563]  ? __x64_sys_read+0x60/0x60
[  168.572081]  ? __kasan_check_write+0x20/0x30
[  168.572659]  ? do_user_addr_fault+0x446/0xff0
[  168.573264]  __x64_sys_write+0x46/0x60
[  168.573774]  do_syscall_64+0x35/0x80
[  168.574264]  entry_SYSCALL_64_after_hwframe+0x44/0xae
[  168.574960] RIP: 0033:0x7fac74915130
[  168.575456] Code: 73 01 c3 48 8b 0d 58 ed 2c 00 f7 d8 64 89 01 48 83 c8 ff c3 66 0f 1f 444
[  168.577969] RSP: 002b:00007ffc3080e288 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
[  168.578986] RAX: ffffffffffffffda RBX: 0000000000000009 RCX: 00007fac74915130
[  168.579937] RDX: 0000000000000009 RSI: 000056007669f080 RDI: 0000000000000001
[  168.580884] RBP: 000056007669f080 R08: 000000000000000a R09: 00007fac75227700
[  168.581841] R10: 000056007655c8f0 R11: 0000000000000246 R12: 0000000000000009
[  168.582796] R13: 0000000000000001 R14: 00007fac74be55e0 R15: 00007fac74be08c0
[  168.583757]  </TASK>
[  168.584063] Modules linked in:
[  168.584494] CR2: 0000000000000008
[  168.584964] ---[ end trace 2475611ad0f77a1a ]---

This is because blkg_alloc() is called from blkg_conf_prep() without holding 'q->queue_lock', and elevator is exited before blkg_create():

thread 1 thread 2
blkg_conf_prep
spin_lock_irq(&q->queue_lock);
blkg_lookup_check -> return NULL
spin_unlock_irq(&q->queue_lock);

blkg_alloc
blkcg_policy_enabled -> true
pd = ->pd_alloc_fn
blkg->pd[i] = pd

                                   blk_mq_exit_sched
                                    bfq_exit_queue
                                     blkcg_deactivate_policy
                                      spin_lock_irq(&q->queue_lock);
                                      __clear_bit(pol->plid, q->blkcg_pols);
                                      spin_unlock_irq(&q->queue_lock);
                                    q->elevator = NULL;
  spin_lock_irq(&q->queue_lock);
   blkg_create
    if (blkg->pd[i])
     ->pd_init_fn -> q->elevator is NULL

spin_unlock_irq(&q->queue_lock);

Because blkcg_deactivate_policy() requires queue to be frozen, we can grab q_usage_counter to synchoronize blkg_conf_prep() against blkcg_deactivate_policy().

Fixes e21b7a0b9887 ("block, bfq: add full hierarchical scheduling and cgroups support")
---
 block/blk-cgroup.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
block/blk-cgroup.c +10 -0
index 9a1c5839dd469..ebff1af402e5b 100644
@@ -633,6 +633,14 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol,
 
 	q = bdev->bd_disk->queue;
 
+	/*
+	 * blkcg_deactivate_policy() requires queue to be frozen, we can grab
+	 * q_usage_counter to prevent concurrent with blkcg_deactivate_policy().
+	 */
+	ret = blk_queue_enter(q, 0);
+	if (ret)
+		return ret;
+
 	rcu_read_lock();
 	spin_lock_irq(&q->queue_lock);
 
@@ -702,6 +710,7 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol,
 			goto success;
 	}
 success:
+	blk_queue_exit(q);
 	ctx->bdev = bdev;
 	ctx->blkg = blkg;
 	ctx->body = input;
@@ -714,6 +723,7 @@ fail_unlock:
 	rcu_read_unlock();
 fail:
 	blkdev_put_no_open(bdev);
+	blk_queue_exit(q);
 	/*
 	 * If queue was bypassing, we should retry.  Do so after a
 	 * short msleep().  It isn't strictly necessary but queue
Signature · 1 line
2.33.0
author maintainer
#530 lore raw

PATCH 519/917 libbpf: Fix off-by-one bug in bpf_core_apply_relo()

In reply to Greg Kroah-Hartman

From: Andrii Nakryiko <[email protected]>

[ Upstream commit de5d0dcef602de39070c31c7e56c58249c56ba37 ]

Fix instruction index validity check which has off-by-one error.

Fixes 3ee4f5335511 ("libbpf: Split bpf_core_apply_relo() into bpf_program independent helper.")
---
 tools/lib/bpf/libbpf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/lib/bpf/libbpf.c +1 -1
index 51180f300d2e1..7145463a4a562 100644
@@ -5138,7 +5138,7 @@ static int bpf_core_apply_relo(struct bpf_program *prog,
 	 * relocated, so it's enough to just subtract in-section offset
 	 */
 	insn_idx = insn_idx - prog->sec_insn_off;
-	if (insn_idx > prog->insns_cnt)
+	if (insn_idx >= prog->insns_cnt)
 		return -EINVAL;
 	insn = &prog->insns[insn_idx];
 
Signature · 1 line
2.33.0
author maintainer
#531 lore raw

PATCH 520/917 tpm: fix Atmel TPM crash caused by too frequent queries

In reply to Greg Kroah-Hartman

From: Hao Wu <[email protected]>

[ Upstream commit 79ca6f74dae067681a779fd573c2eb59649989bc ]

The Atmel TPM 1.2 chips crash with error tpm_try_transmit: send(): error -62 since kernel 4.14. It is observed from the kernel log after running tpm_sealdata -z. The error thrown from the command is as follows

```
$ tpm_sealdata -z
Tspi_Key_LoadKey failed: 0x00001087 - layer=tddl,
code=0087 (135), I/O error
```

The issue was reproduced with the following Atmel TPM chip:

```
$ tpm_version
T0  TPM 1.2 Version Info:
  Chip Version:        1.2.66.1
  Spec Level:          2
  Errata Revision:     3
  TPM Vendor ID:       ATML
  TPM Version:         01010000
  Manufacturer Info:   41544d4c
```

The root cause of the issue is due to the TPM calls to msleep() were replaced with usleep_range() [1], which reduces the actual timeout. Via experiments, it is observed that the original msleep(5) actually sleeps for 15ms. Because of a known timeout issue in Atmel TPM 1.2 chip, the shorter timeout than 15ms can cause the error described above.

A few further changes in kernel 4.16 [2] and 4.18 [3, 4] further reduced the timeout to less than 1ms. With experiments, the problematic timeout in the latest kernel is the one for wait_for_tpm_stat.

To fix it, the patch reverts the timeout of wait_for_tpm_stat to 15ms for all Atmel TPM 1.2 chips, but leave it untouched for Ateml TPM 2.0 chip, and chips from other vendors.
As explained above, the chosen 15ms timeout is
the actual timeout before this issue introduced,
thus the old value is used here.
Particularly, TPM_ATML_TIMEOUT_WAIT_STAT_MIN is set to 14700us, TPM_ATML_TIMEOUT_WAIT_STAT_MIN is set to 15000us according to the existing TPM_TIMEOUT_RANGE_US (300us).
The fixed has been tested in the system with the affected Atmel chip with no issues observed after boot up.

[1] 9f3fc7bcddcb tpm: replace msleep() with usleep_range() in TPM 1.2/2.0 generic drivers
[2] cf151a9a44d5 tpm: reduce tpm polling delay in tpm_tis_core
[3] 59f5a6b07f64 tpm: reduce poll sleep time in tpm_transmit()
[4] 424eaf910c32 tpm: reduce polling time to usecs for even finer granularity

Fixes 9f3fc7bcddcb ("tpm: replace msleep() with usleep_range() in TPM 1.2/2.0 generic drivers")
---
 drivers/char/tpm/tpm_tis_core.c | 26 ++++++++++++++++++--------
 drivers/char/tpm/tpm_tis_core.h |  4 ++++
 include/linux/tpm.h             |  1 +
 3 files changed, 23 insertions(+), 8 deletions(-)
drivers/char/tpm/tpm_tis_core.c +18 -8
index 69579efb247b3..b2659a4c40168 100644
@@ -48,6 +48,7 @@ static int wait_for_tpm_stat(struct tpm_chip *chip, u8 mask,
 		unsigned long timeout, wait_queue_head_t *queue,
 		bool check_cancel)
 {
+	struct tpm_tis_data *priv = dev_get_drvdata(&chip->dev);
 	unsigned long stop;
 	long rc;
 	u8 status;
@@ -80,8 +81,8 @@ again:
 		}
 	} else {
 		do {
-			usleep_range(TPM_TIMEOUT_USECS_MIN,
-				     TPM_TIMEOUT_USECS_MAX);
+			usleep_range(priv->timeout_min,
+				     priv->timeout_max);
 			status = chip->ops->status(chip);
 			if ((status & mask) == mask)
 				return 0;
@@ -945,7 +946,22 @@ int tpm_tis_core_init(struct device *dev, struct tpm_tis_data *priv, int irq,
 	chip->timeout_b = msecs_to_jiffies(TIS_TIMEOUT_B_MAX);
 	chip->timeout_c = msecs_to_jiffies(TIS_TIMEOUT_C_MAX);
 	chip->timeout_d = msecs_to_jiffies(TIS_TIMEOUT_D_MAX);
+	priv->timeout_min = TPM_TIMEOUT_USECS_MIN;
+	priv->timeout_max = TPM_TIMEOUT_USECS_MAX;
 	priv->phy_ops = phy_ops;
+
+	rc = tpm_tis_read32(priv, TPM_DID_VID(0), &vendor);
+	if (rc < 0)
+		goto out_err;
+
+	priv->manufacturer_id = vendor;
+
+	if (priv->manufacturer_id == TPM_VID_ATML &&
+		!(chip->flags & TPM_CHIP_FLAG_TPM2)) {
+		priv->timeout_min = TIS_TIMEOUT_MIN_ATML;
+		priv->timeout_max = TIS_TIMEOUT_MAX_ATML;
+	}
+
 	dev_set_drvdata(&chip->dev, priv);
 
 	if (is_bsw()) {
@@ -988,12 +1004,6 @@ int tpm_tis_core_init(struct device *dev, struct tpm_tis_data *priv, int irq,
 	if (rc)
 		goto out_err;
 
-	rc = tpm_tis_read32(priv, TPM_DID_VID(0), &vendor);
-	if (rc < 0)
-		goto out_err;
-
-	priv->manufacturer_id = vendor;
-
 	rc = tpm_tis_read8(priv, TPM_RID(0), &rid);
 	if (rc < 0)
 		goto out_err;
drivers/char/tpm/tpm_tis_core.h +4 -0
index b2a3c6c72882d..3be24f221e32a 100644
@@ -54,6 +54,8 @@ enum tis_defaults {
 	TIS_MEM_LEN = 0x5000,
 	TIS_SHORT_TIMEOUT = 750,	/* ms */
 	TIS_LONG_TIMEOUT = 2000,	/* 2 sec */
+	TIS_TIMEOUT_MIN_ATML = 14700,	/* usecs */
+	TIS_TIMEOUT_MAX_ATML = 15000,	/* usecs */
 };
 
 /* Some timeout values are needed before it is known whether the chip is
@@ -98,6 +100,8 @@ struct tpm_tis_data {
 	wait_queue_head_t read_queue;
 	const struct tpm_tis_phy_ops *phy_ops;
 	unsigned short rng_quality;
+	unsigned int timeout_min; /* usecs */
+	unsigned int timeout_max; /* usecs */
 };
 
 struct tpm_tis_phy_ops {
include/linux/tpm.h +1 -0
index aa11fe323c56b..12d827734686d 100644
@@ -269,6 +269,7 @@ enum tpm2_cc_attrs {
 #define TPM_VID_INTEL    0x8086
 #define TPM_VID_WINBOND  0x1050
 #define TPM_VID_STM      0x104A
+#define TPM_VID_ATML     0x1114
 
 enum tpm_chip_flags {
 	TPM_CHIP_FLAG_TPM2		= BIT(1),
Signature · 1 line
2.33.0
author maintainer
#532 lore raw

PATCH 521/917 tpm_tis_spi: Add missing SPI ID

In reply to Greg Kroah-Hartman

From: Mark Brown <[email protected]>

[ Upstream commit 7eba41fe8c7bb01ff3d4b757bd622375792bc720 ]

In commit c46ed2281bbe ("tpm_tis_spi: add missing SPI device ID entries") we added SPI IDs for all the DT aliases to handle the fact that we always use SPI modaliases to load modules even when probed via DT however the mentioned commit missed that the SPI and OF device ID entries did not match and were different and so DT nodes with compatible "tcg,tpm_tis-spi" will not match. Add an extra ID for tpm_tis-spi rather than just fix the existing one since what's currently there is going to be better for anyone actually using SPI IDs to instantiate.

Fixes c46ed2281bbe ("tpm_tis_spi: add missing SPI device ID entries")
Fixes 96c8395e2166 ("spi: Revert modalias changes")
---
 drivers/char/tpm/tpm_tis_spi_main.c | 1 +
 1 file changed, 1 insertion(+)
drivers/char/tpm/tpm_tis_spi_main.c +1 -0
index 54584b4b00d19..aaa59a00eeaef 100644
@@ -267,6 +267,7 @@ static const struct spi_device_id tpm_tis_spi_id[] = {
 	{ "st33htpm-spi", (unsigned long)tpm_tis_spi_probe },
 	{ "slb9670", (unsigned long)tpm_tis_spi_probe },
 	{ "tpm_tis_spi", (unsigned long)tpm_tis_spi_probe },
+	{ "tpm_tis-spi", (unsigned long)tpm_tis_spi_probe },
 	{ "cr50", (unsigned long)cr50_spi_probe },
 	{}
 };
Signature · 1 line
2.33.0
author maintainer
#533 lore raw

PATCH 522/917 libbpf: Fix endianness detection in BPF_CORE_READ_BITFIELD_PROBED()

In reply to Greg Kroah-Hartman

From: Ilya Leoshkevich <[email protected]>

[ Upstream commit 45f2bebc8079788f62f22d9e8b2819afb1789d7b ]

__BYTE_ORDER is supposed to be defined by a libc, and __BYTE_ORDER__ - by a compiler. bpf_core_read.h checks __BYTE_ORDER == __LITTLE_ENDIAN, which is true if neither are defined, leading to incorrect behavior on big-endian hosts if libc headers are not included, which is often the case.

Fixes ee26dade0e3b ("libbpf: Add support for relocatable bitfields")
---
 tools/lib/bpf/bpf_core_read.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/lib/bpf/bpf_core_read.h +1 -1
index 09ebe3db5f2f8..e4aa9996a5501 100644
@@ -40,7 +40,7 @@ enum bpf_enum_value_kind {
 #define __CORE_RELO(src, field, info)					      \
 	__builtin_preserve_field_info((src)->field, BPF_FIELD_##info)
 
-#if __BYTE_ORDER == __LITTLE_ENDIAN
+#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
 #define __CORE_BITFIELD_PROBE_READ(dst, src, fld)			      \
 	bpf_probe_read_kernel(						      \
 			(void *)dst,				      \
Signature · 1 line
2.33.0
author maintainer
#534 lore raw

PATCH 523/917 tcp: dont free a FIN sk_buff in tcp_remove_empty_skb()

In reply to Greg Kroah-Hartman

From: Jon Maxwell <[email protected]>

[ Upstream commit cf12e6f9124629b18a6182deefc0315f0a73a199 ]

v1: Implement a more general statement as recommended by Eric Dumazet. The sequence number will be advanced, so this check will fix the FIN case and other cases.

A customer reported sockets stuck in the CLOSING state. A Vmcore revealed that the write_queue was not empty as determined by tcp_write_queue_empty() but the sk_buff containing the FIN flag had been freed and the socket was zombied in that state. Corresponding pcaps show no FIN from the Linux kernel on the wire.

Some instrumentation was added to the kernel and it was found that there is a timing window where tcp_sendmsg() can run after tcp_send_fin().

tcp_sendmsg() will hit an error, for example:

1269 ▹ if (sk->sk_err || (sk->sk_shutdown & SEND_SHUTDOWN))↩
1270 ▹ ▹ goto do_error;↩

tcp_remove_empty_skb() will then free the FIN sk_buff as "skb->len == 0". The TCP socket is now wedged in the FIN-WAIT-1 state because the FIN is never sent.

If the other side sends a FIN packet the socket will transition to CLOSING and remain that way until the system is rebooted.

Fix this by checking for the FIN flag in the sk_buff and don't free it if that is the case. Testing confirmed that fixed the issue.

Fixes fdfc5c8594c2 ("tcp: remove empty skb from write queue in error cases")
---
 net/ipv4/tcp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/ipv4/tcp.c +1 -1
index d13050b68045b..8affba5909bdf 100644
@@ -952,7 +952,7 @@ int tcp_send_mss(struct sock *sk, int *size_goal, int flags)
  */
 void tcp_remove_empty_skb(struct sock *sk, struct sk_buff *skb)
 {
-	if (skb && !skb->len) {
+	if (skb && TCP_SKB_CB(skb)->seq == TCP_SKB_CB(skb)->end_seq) {
 		tcp_unlink_write_queue(skb, sk);
 		if (tcp_write_queue_empty(sk))
 			tcp_chrono_stop(sk, TCP_CHRONO_BUSY);
Signature · 1 line
2.33.0
author maintainer
#535 lore raw

PATCH 524/917 tracing: Fix missing trace_boot_init_histograms kstrdup NULL checks

In reply to Greg Kroah-Hartman

From: Mathieu Desnoyers <[email protected]>

[ Upstream commit 3c20bd3af535d64771b193bb4dd41ed662c464ce ]

trace_boot_init_histograms misses NULL pointer checks for kstrdup failure.

Fixes 64dc7f6958ef5 ("tracing/boot: Show correct histogram error command")
---
 kernel/trace/trace_boot.c | 4 ++++
 1 file changed, 4 insertions(+)
kernel/trace/trace_boot.c +4 -0
index 8d252f63cd784..0580287d7a0d1 100644
@@ -430,6 +430,8 @@ trace_boot_init_histograms(struct trace_event_file *file,
 		/* All digit started node should be instances. */
 		if (trace_boot_compose_hist_cmd(node, buf, size) == 0) {
 			tmp = kstrdup(buf, GFP_KERNEL);
+			if (!tmp)
+				return;
 			if (trigger_process_regex(file, buf) < 0)
 				pr_err("Failed to apply hist trigger: %s\n", tmp);
 			kfree(tmp);
@@ -439,6 +441,8 @@ trace_boot_init_histograms(struct trace_event_file *file,
 	if (xbc_node_find_subkey(hnode, "keys")) {
 		if (trace_boot_compose_hist_cmd(hnode, buf, size) == 0) {
 			tmp = kstrdup(buf, GFP_KERNEL);
+			if (!tmp)
+				return;
 			if (trigger_process_regex(file, buf) < 0)
 				pr_err("Failed to apply hist trigger: %s\n", tmp);
 			kfree(tmp);
Signature · 1 line
2.33.0
author maintainer
#536 lore raw

PATCH 525/917 cpufreq: intel_pstate: Fix cpu->pstate.turbo_freq initialization

In reply to Greg Kroah-Hartman

From: Zhang Rui <[email protected]>

[ Upstream commit c72bcf0ab87a92634e58af62e89af0f40dfd0b88 ]

Fix a problem in active mode that cpu->pstate.turbo_freq is initialized only if HWP-to-frequency scaling factor is refined.

In passive mode, this problem is not exposed, because cpu->pstate.turbo_freq is set again, later in intel_cpufreq_cpu_init()->intel_pstate_get_hwp_cap().

Fixes eb3693f0521e ("cpufreq: intel_pstate: hybrid: CPU-specific scaling factor")
---
 drivers/cpufreq/intel_pstate.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/cpufreq/intel_pstate.c +2 -1
index 8c176b7dae415..fc7a429f22d33 100644
@@ -537,7 +537,8 @@ static void intel_pstate_hybrid_hwp_adjust(struct cpudata *cpu)
 	 * scaling factor is too high, so recompute it to make the HWP_CAP
 	 * highest performance correspond to the maximum turbo frequency.
 	 */
-	if (turbo_freq < cpu->pstate.turbo_pstate * scaling) {
+	cpu->pstate.turbo_freq = cpu->pstate.turbo_pstate * scaling;
+	if (turbo_freq < cpu->pstate.turbo_freq) {
 		cpu->pstate.turbo_freq = turbo_freq;
 		scaling = DIV_ROUND_UP(turbo_freq, cpu->pstate.turbo_pstate);
 		cpu->pstate.scaling = scaling;
Signature · 1 line
2.33.0
author maintainer
#537 lore raw

PATCH 526/917 spi: spi-rpc-if: Check return value of rpcif_sw_init()

In reply to Greg Kroah-Hartman

From: Lad Prabhakar <[email protected]>

[ Upstream commit 0b0a281ed7001d4c4f4c47bdc84680c4997761ca ]

rpcif_sw_init() can fail so make sure we check the return value of it and on error exit rpcif_spi_probe() callback with error code.

---
 drivers/spi/spi-rpc-if.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/spi/spi-rpc-if.c +3 -1
index c53138ce00309..83796a4ead34a 100644
@@ -139,7 +139,9 @@ static int rpcif_spi_probe(struct platform_device *pdev)
 		return -ENOMEM;
 
 	rpc = spi_controller_get_devdata(ctlr);
-	rpcif_sw_init(rpc, parent);
+	error = rpcif_sw_init(rpc, parent);
+	if (error)
+		return error;
 
 	platform_set_drvdata(pdev, ctlr);
 
Signature · 1 line
2.33.0