[PATCH AUTOSEL 4.4 1/8] HID: cypress: Support Varmilo Keyboards' media hotkeys

[PATCH AUTOSEL 4.4 1/8] HID: cypress: Support Varmilo Keyboards' media hotkeys

[Sasha Levin]

From: Frank Yang <puilp0502@gmail.com>

[ Upstream commit 652f3d00de523a17b0cebe7b90debccf13aa8c31 ]

The Varmilo VA104M Keyboard (04b4:07b1, reported as Varmilo Z104M)
exposes media control hotkeys as a USB HID consumer control device, but
these keys do not work in the current (5.8-rc1) kernel due to the
incorrect HID report descriptor. Fix the problem by modifying the
internal HID report descriptor.

More specifically, the keyboard report descriptor specifies the
logical boundary as 572~10754 (0x023c ~ 0x2a02) while the usage
boundary is specified as 0~10754 (0x00 ~ 0x2a02). This results in an
incorrect interpretation of input reports, causing inputs to be ignored.
By setting the Logical Minimum to zero, we align the logical boundary
with the Usage ID boundary.

Some notes:

* There seem to be multiple variants of the VA104M keyboard. This
  patch specifically targets 04b4:07b1 variant.

* The device works out-of-the-box on Windows platform with the generic
  consumer control device driver (hidserv.inf). This suggests that
  Windows either ignores the Logical Minimum/Logical Maximum or
  interprets the Usage ID assignment differently from the linux
  implementation; Maybe there are other devices out there that only
  works on Windows due to this problem?

Signed-off-by: Frank Yang <puilp0502@gmail.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/hid/hid-cypress.c | 44 ++++++++++++++++++++++++++++++++++-----
 drivers/hid/hid-ids.h     |  2 ++
 2 files changed, 41 insertions(+), 5 deletions(-)

diff --git a/drivers/hid/hid-cypress.c b/drivers/hid/hid-cypress.c
index 1689568b597d4..12c5d7c96527a 100644
--- a/drivers/hid/hid-cypress.c
+++ b/drivers/hid/hid-cypress.c
@@ -26,19 +26,17 @@
 #define CP_2WHEEL_MOUSE_HACK		0x02
 #define CP_2WHEEL_MOUSE_HACK_ON		0x04
 
+#define VA_INVAL_LOGICAL_BOUNDARY	0x08
+
 /*
  * Some USB barcode readers from cypress have usage min and usage max in
  * the wrong order
  */
-static __u8 *cp_report_fixup(struct hid_device *hdev, __u8 *rdesc,
+static __u8 *cp_rdesc_fixup(struct hid_device *hdev, __u8 *rdesc,
 		unsigned int *rsize)
 {
-	unsigned long quirks = (unsigned long)hid_get_drvdata(hdev);
 	unsigned int i;
 
-	if (!(quirks & CP_RDESC_SWAPPED_MIN_MAX))
-		return rdesc;
-
 	if (*rsize < 4)
 		return rdesc;
 
@@ -51,6 +49,40 @@ static __u8 *cp_report_fixup(struct hid_device *hdev, __u8 *rdesc,
 	return rdesc;
 }
 
+static __u8 *va_logical_boundary_fixup(struct hid_device *hdev, __u8 *rdesc,
+		unsigned int *rsize)
+{
+	/*
+	 * Varmilo VA104M (with VID Cypress and device ID 07B1) incorrectly
+	 * reports Logical Minimum of its Consumer Control device as 572
+	 * (0x02 0x3c). Fix this by setting its Logical Minimum to zero.
+	 */
+	if (*rsize == 25 &&
+			rdesc[0] == 0x05 && rdesc[1] == 0x0c &&
+			rdesc[2] == 0x09 && rdesc[3] == 0x01 &&
+			rdesc[6] == 0x19 && rdesc[7] == 0x00 &&
+			rdesc[11] == 0x16 && rdesc[12] == 0x3c && rdesc[13] == 0x02) {
+		hid_info(hdev,
+			 "fixing up varmilo VA104M consumer control report descriptor\n");
+		rdesc[12] = 0x00;
+		rdesc[13] = 0x00;
+	}
+	return rdesc;
+}
+
+static __u8 *cp_report_fixup(struct hid_device *hdev, __u8 *rdesc,
+		unsigned int *rsize)
+{
+	unsigned long quirks = (unsigned long)hid_get_drvdata(hdev);
+
+	if (quirks & CP_RDESC_SWAPPED_MIN_MAX)
+		rdesc = cp_rdesc_fixup(hdev, rdesc, rsize);
+	if (quirks & VA_INVAL_LOGICAL_BOUNDARY)
+		rdesc = va_logical_boundary_fixup(hdev, rdesc, rsize);
+
+	return rdesc;
+}
+
 static int cp_input_mapped(struct hid_device *hdev, struct hid_input *hi,
 		struct hid_field *field, struct hid_usage *usage,
 		unsigned long **bit, int *max)
@@ -131,6 +163,8 @@ static const struct hid_device_id cp_devices[] = {
 		.driver_data = CP_RDESC_SWAPPED_MIN_MAX },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_CYPRESS, USB_DEVICE_ID_CYPRESS_MOUSE),
 		.driver_data = CP_2WHEEL_MOUSE_HACK },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_CYPRESS, USB_DEVICE_ID_CYPRESS_VARMILO_VA104M_07B1),
+		.driver_data = VA_INVAL_LOGICAL_BOUNDARY },
 	{ }
 };
 MODULE_DEVICE_TABLE(hid, cp_devices);
diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h
index 33d2b5948d7fc..773452c6edfab 100644
--- a/drivers/hid/hid-ids.h
+++ b/drivers/hid/hid-ids.h
@@ -279,6 +279,8 @@
 #define USB_DEVICE_ID_CYPRESS_BARCODE_4	0xed81
 #define USB_DEVICE_ID_CYPRESS_TRUETOUCH	0xc001
 
+#define USB_DEVICE_ID_CYPRESS_VARMILO_VA104M_07B1   0X07b1
+
 #define USB_VENDOR_ID_DATA_MODUL	0x7374
 #define USB_VENDOR_ID_DATA_MODUL_EASYMAXTOUCH	0x1201
 
-- 
2.27.0

[PATCH AUTOSEL 4.4 2/8] Input: i8042 - allow insmod to succeed on devices without an i8042 controller

[Sasha Levin]

From: Hans de Goede <hdegoede@redhat.com>

[ Upstream commit b1884583fcd17d6a1b1bba94bbb5826e6b5c6e17 ]

The i8042 module exports several symbols which may be used by other
modules.

Before this commit it would refuse to load (when built as a module itself)
on systems without an i8042 controller.

This is a problem specifically for the asus-nb-wmi module. Many Asus
laptops support the Asus WMI interface. Some of them have an i8042
controller and need to use i8042_install_filter() to filter some kbd
events. Other models do not have an i8042 controller (e.g. they use an
USB attached kbd).

Before this commit the asus-nb-wmi driver could not be loaded on Asus
models without an i8042 controller, when the i8042 code was built as
a module (as Arch Linux does) because the module_init function of the
i8042 module would fail with -ENODEV and thus the i8042_install_filter
symbol could not be loaded.

This commit fixes this by exiting from module_init with a return code
of 0 if no controller is found.  It also adds a i8042_present bool to
make the module_exit function a no-op in this case and also adds a
check for i8042_present to the exported i8042_command function.

The latter i8042_present check should not really be necessary because
when builtin that function can already be used on systems without
an i8042 controller, but better safe then sorry.

Reported-and-tested-by: Marius Iacob <themariusus@gmail.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20201008112628.3979-2-hdegoede@redhat.com
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/input/serio/i8042.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/drivers/input/serio/i8042.c b/drivers/input/serio/i8042.c
index c84c685056b99..6b648339733fa 100644
--- a/drivers/input/serio/i8042.c
+++ b/drivers/input/serio/i8042.c
@@ -125,6 +125,7 @@ module_param_named(unmask_kbd_data, i8042_unmask_kbd_data, bool, 0600);
 MODULE_PARM_DESC(unmask_kbd_data, "Unconditional enable (may reveal sensitive data) of normally sanitize-filtered kbd data traffic debug log [pre-condition: i8042.debug=1 enabled]");
 #endif
 
+static bool i8042_present;
 static bool i8042_bypass_aux_irq_test;
 static char i8042_kbd_firmware_id[128];
 static char i8042_aux_firmware_id[128];
@@ -343,6 +344,9 @@ int i8042_command(unsigned char *param, int command)
 	unsigned long flags;
 	int retval;
 
+	if (!i8042_present)
+		return -1;
+
 	spin_lock_irqsave(&i8042_lock, flags);
 	retval = __i8042_command(param, command);
 	spin_unlock_irqrestore(&i8042_lock, flags);
@@ -1597,12 +1601,15 @@ static int __init i8042_init(void)
 
 	err = i8042_platform_init();
 	if (err)
-		return err;
+		return (err == -ENODEV) ? 0 : err;
 
 	err = i8042_controller_check();
 	if (err)
 		goto err_platform_exit;
 
+	/* Set this before creating the dev to allow i8042_command to work right away */
+	i8042_present = true;
+
 	pdev = platform_create_bundle(&i8042_driver, i8042_probe, NULL, 0, NULL, 0);
 	if (IS_ERR(pdev)) {
 		err = PTR_ERR(pdev);
@@ -1621,6 +1628,9 @@ static int __init i8042_init(void)
 
 static void __exit i8042_exit(void)
 {
+	if (!i8042_present)
+		return;
+
 	platform_device_unregister(i8042_platform_device);
 	platform_driver_unregister(&i8042_driver);
 	i8042_platform_exit();
-- 
2.27.0

[PATCH AUTOSEL 4.4 3/8] HID: hid-sensor-hub: Fix issue with devices with no report ID

[Sasha Levin]

From: Pablo Ceballos <pceballos@google.com>

[ Upstream commit 34a9fa2025d9d3177c99351c7aaf256c5f50691f ]

Some HID devices don't use a report ID because they only have a single
report. In those cases, the report ID in struct hid_report will be zero
and the data for the report will start at the first byte, so don't skip
over the first byte.

Signed-off-by: Pablo Ceballos <pceballos@google.com>
Acked-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/hid/hid-sensor-hub.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/hid/hid-sensor-hub.c b/drivers/hid/hid-sensor-hub.c
index 8efaa88329aa3..83e45d5801a98 100644
--- a/drivers/hid/hid-sensor-hub.c
+++ b/drivers/hid/hid-sensor-hub.c
@@ -473,7 +473,8 @@ static int sensor_hub_raw_event(struct hid_device *hdev,
 		return 1;
 
 	ptr = raw_data;
-	ptr++; /* Skip report id */
+	if (report->id)
+		ptr++; /* Skip report id */
 
 	spin_lock_irqsave(&pdata->lock, flags);
 
-- 
2.27.0

[PATCH AUTOSEL 4.4 4/8] x86/xen: don't unbind uninitialized lock_kicker_irq

[Sasha Levin]

From: Brian Masney <bmasney@redhat.com>

[ Upstream commit 65cae18882f943215d0505ddc7e70495877308e6 ]

When booting a hyperthreaded system with the kernel parameter
'mitigations=auto,nosmt', the following warning occurs:

    WARNING: CPU: 0 PID: 1 at drivers/xen/events/events_base.c:1112 unbind_from_irqhandler+0x4e/0x60
    ...
    Hardware name: Xen HVM domU, BIOS 4.2.amazon 08/24/2006
    ...
    Call Trace:
     xen_uninit_lock_cpu+0x28/0x62
     xen_hvm_cpu_die+0x21/0x30
     takedown_cpu+0x9c/0xe0
     ? trace_suspend_resume+0x60/0x60
     cpuhp_invoke_callback+0x9a/0x530
     _cpu_up+0x11a/0x130
     cpu_up+0x7e/0xc0
     bringup_nonboot_cpus+0x48/0x50
     smp_init+0x26/0x79
     kernel_init_freeable+0xea/0x229
     ? rest_init+0xaa/0xaa
     kernel_init+0xa/0x106
     ret_from_fork+0x35/0x40

The secondary CPUs are not activated with the nosmt mitigations and only
the primary thread on each CPU core is used. In this situation,
xen_hvm_smp_prepare_cpus(), and more importantly xen_init_lock_cpu(), is
not called, so the lock_kicker_irq is not initialized for the secondary
CPUs. Let's fix this by exiting early in xen_uninit_lock_cpu() if the
irq is not set to avoid the warning from above for each secondary CPU.

Signed-off-by: Brian Masney <bmasney@redhat.com>
Link: https://lore.kernel.org/r/20201107011119.631442-1-bmasney@redhat.com
Reviewed-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 arch/x86/xen/spinlock.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c
index 85872a08994a1..e9fc0f7df0da8 100644
--- a/arch/x86/xen/spinlock.c
+++ b/arch/x86/xen/spinlock.c
@@ -301,10 +301,20 @@ void xen_init_lock_cpu(int cpu)
 
 void xen_uninit_lock_cpu(int cpu)
 {
+	int irq;
+
 	if (!xen_pvspin)
 		return;
 
-	unbind_from_irqhandler(per_cpu(lock_kicker_irq, cpu), NULL);
+	/*
+	 * When booting the kernel with 'mitigations=auto,nosmt', the secondary
+	 * CPUs are not activated, and lock_kicker_irq is not initialized.
+	 */
+	irq = per_cpu(lock_kicker_irq, cpu);
+	if (irq == -1)
+		return;
+
+	unbind_from_irqhandler(irq, NULL);
 	per_cpu(lock_kicker_irq, cpu) = -1;
 	kfree(per_cpu(irq_name, cpu));
 	per_cpu(irq_name, cpu) = NULL;
-- 
2.27.0

[PATCH AUTOSEL 4.4 5/8] proc: don't allow async path resolution of /proc/self components

[Sasha Levin]

From: Jens Axboe <axboe@kernel.dk>

[ Upstream commit 8d4c3e76e3be11a64df95ddee52e99092d42fc19 ]

If this is attempted by a kthread, then return -EOPNOTSUPP as we don't
currently support that. Once we can get task_pid_ptr() doing the right
thing, then this can go away again.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/proc/self.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/fs/proc/self.c b/fs/proc/self.c
index 2dcc2558b3aa7..dffbe533d53fc 100644
--- a/fs/proc/self.c
+++ b/fs/proc/self.c
@@ -24,6 +24,13 @@ static const char *proc_self_follow_link(struct dentry *dentry, void **cookie)
 	pid_t tgid = task_tgid_nr_ns(current, ns);
 	char *name;
 
+	/*
+	 * Not currently supported. Once we can inherit all of struct pid,
+	 * we can allow this.
+	 */
+	if (current->flags & PF_KTHREAD)
+		return ERR_PTR(-EOPNOTSUPP);
+
 	if (!tgid)
 		return ERR_PTR(-ENOENT);
 	/* 11 for max length of signed int in decimal + NULL term */
-- 
2.27.0

[PATCH AUTOSEL 4.4 6/8] dmaengine: pl330: _prep_dma_memcpy: Fix wrong burst size

[Sasha Levin]

From: Sugar Zhang <sugar.zhang@rock-chips.com>

[ Upstream commit e773ca7da8beeca7f17fe4c9d1284a2b66839cc1 ]

Actually, burst size is equal to '1 << desc->rqcfg.brst_size'.
we should use burst size, not desc->rqcfg.brst_size.

dma memcpy performance on Rockchip RV1126
@ 1512MHz A7, 1056MHz LPDDR3, 200MHz DMA:

dmatest:

/# echo dma0chan0 > /sys/module/dmatest/parameters/channel
/# echo 4194304 > /sys/module/dmatest/parameters/test_buf_size
/# echo 8 > /sys/module/dmatest/parameters/iterations
/# echo y > /sys/module/dmatest/parameters/norandom
/# echo y > /sys/module/dmatest/parameters/verbose
/# echo 1 > /sys/module/dmatest/parameters/run

dmatest: dma0chan0-copy0: result #1: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000
dmatest: dma0chan0-copy0: result #2: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000
dmatest: dma0chan0-copy0: result #3: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000
dmatest: dma0chan0-copy0: result #4: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000
dmatest: dma0chan0-copy0: result #5: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000
dmatest: dma0chan0-copy0: result #6: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000
dmatest: dma0chan0-copy0: result #7: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000
dmatest: dma0chan0-copy0: result #8: 'test passed' with src_off=0x0 dst_off=0x0 len=0x400000

Before:

  dmatest: dma0chan0-copy0: summary 8 tests, 0 failures 48 iops 200338 KB/s (0)

After this patch:

  dmatest: dma0chan0-copy0: summary 8 tests, 0 failures 179 iops 734873 KB/s (0)

After this patch and increase dma clk to 400MHz:

  dmatest: dma0chan0-copy0: summary 8 tests, 0 failures 259 iops 1062929 KB/s (0)

Signed-off-by: Sugar Zhang <sugar.zhang@rock-chips.com>
Link: https://lore.kernel.org/r/1605326106-55681-1-git-send-email-sugar.zhang@rock-chips.com
Signed-off-by: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/dma/pl330.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c
index 9aa57b37381a9..7f66ae1945b24 100644
--- a/drivers/dma/pl330.c
+++ b/drivers/dma/pl330.c
@@ -2634,7 +2634,7 @@ pl330_prep_dma_memcpy(struct dma_chan *chan, dma_addr_t dst,
 	 * If burst size is smaller than bus width then make sure we only
 	 * transfer one at a time to avoid a burst stradling an MFIFO entry.
 	 */
-	if (desc->rqcfg.brst_size * 8 < pl330->pcfg.data_bus_width)
+	if (burst * 8 < pl330->pcfg.data_bus_width)
 		desc->rqcfg.brst_len = 1;
 
 	desc->bytes_requested = len;
-- 
2.27.0

[PATCH AUTOSEL 4.4 7/8] scsi: libiscsi: Fix NOP race condition

[Sasha Levin]

From: Lee Duncan <lduncan@suse.com>

[ Upstream commit fe0a8a95e7134d0b44cd407bc0085b9ba8d8fe31 ]

iSCSI NOPs are sometimes "lost", mistakenly sent to the user-land iscsid
daemon instead of handled in the kernel, as they should be, resulting in a
message from the daemon like:

  iscsid: Got nop in, but kernel supports nop handling.

This can occur because of the new forward- and back-locks, and the fact
that an iSCSI NOP response can occur before processing of the NOP send is
complete. This can result in "conn->ping_task" being NULL in
iscsi_nop_out_rsp(), when the pointer is actually in the process of being
set.

To work around this, we add a new state to the "ping_task" pointer. In
addition to NULL (not assigned) and a pointer (assigned), we add the state
"being set", which is signaled with an INVALID pointer (using "-1").

Link: https://lore.kernel.org/r/20201106193317.16993-1-leeman.duncan@gmail.com
Reviewed-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Lee Duncan <lduncan@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/scsi/libiscsi.c | 23 +++++++++++++++--------
 include/scsi/libiscsi.h |  3 +++
 2 files changed, 18 insertions(+), 8 deletions(-)

diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index b4fbcf4cade8f..36e415487fe53 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -570,8 +570,8 @@ static void iscsi_complete_task(struct iscsi_task *task, int state)
 	if (conn->task == task)
 		conn->task = NULL;
 
-	if (conn->ping_task == task)
-		conn->ping_task = NULL;
+	if (READ_ONCE(conn->ping_task) == task)
+		WRITE_ONCE(conn->ping_task, NULL);
 
 	/* release get from queueing */
 	__iscsi_put_task(task);
@@ -780,6 +780,9 @@ __iscsi_conn_send_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr,
 						   task->conn->session->age);
 	}
 
+	if (unlikely(READ_ONCE(conn->ping_task) == INVALID_SCSI_TASK))
+		WRITE_ONCE(conn->ping_task, task);
+
 	if (!ihost->workq) {
 		if (iscsi_prep_mgmt_task(conn, task))
 			goto free_task;
@@ -987,8 +990,11 @@ static int iscsi_send_nopout(struct iscsi_conn *conn, struct iscsi_nopin *rhdr)
         struct iscsi_nopout hdr;
 	struct iscsi_task *task;
 
-	if (!rhdr && conn->ping_task)
-		return -EINVAL;
+	if (!rhdr) {
+		if (READ_ONCE(conn->ping_task))
+			return -EINVAL;
+		WRITE_ONCE(conn->ping_task, INVALID_SCSI_TASK);
+	}
 
 	memset(&hdr, 0, sizeof(struct iscsi_nopout));
 	hdr.opcode = ISCSI_OP_NOOP_OUT | ISCSI_OP_IMMEDIATE;
@@ -1003,11 +1009,12 @@ static int iscsi_send_nopout(struct iscsi_conn *conn, struct iscsi_nopin *rhdr)
 
 	task = __iscsi_conn_send_pdu(conn, (struct iscsi_hdr *)&hdr, NULL, 0);
 	if (!task) {
+		if (!rhdr)
+			WRITE_ONCE(conn->ping_task, NULL);
 		iscsi_conn_printk(KERN_ERR, conn, "Could not send nopout\n");
 		return -EIO;
 	} else if (!rhdr) {
 		/* only track our nops */
-		conn->ping_task = task;
 		conn->last_ping = jiffies;
 	}
 
@@ -1020,7 +1027,7 @@ static int iscsi_nop_out_rsp(struct iscsi_task *task,
 	struct iscsi_conn *conn = task->conn;
 	int rc = 0;
 
-	if (conn->ping_task != task) {
+	if (READ_ONCE(conn->ping_task) != task) {
 		/*
 		 * If this is not in response to one of our
 		 * nops then it must be from userspace.
@@ -1960,7 +1967,7 @@ static void iscsi_start_tx(struct iscsi_conn *conn)
  */
 static int iscsi_has_ping_timed_out(struct iscsi_conn *conn)
 {
-	if (conn->ping_task &&
+	if (READ_ONCE(conn->ping_task) &&
 	    time_before_eq(conn->last_recv + (conn->recv_timeout * HZ) +
 			   (conn->ping_timeout * HZ), jiffies))
 		return 1;
@@ -2095,7 +2102,7 @@ static enum blk_eh_timer_return iscsi_eh_cmd_timed_out(struct scsi_cmnd *sc)
 	 * Checking the transport already or nop from a cmd timeout still
 	 * running
 	 */
-	if (conn->ping_task) {
+	if (READ_ONCE(conn->ping_task)) {
 		task->have_checked_conn = true;
 		rc = BLK_EH_RESET_TIMER;
 		goto done;
diff --git a/include/scsi/libiscsi.h b/include/scsi/libiscsi.h
index c7b1dc713cdd7..9c7f4aad6db66 100644
--- a/include/scsi/libiscsi.h
+++ b/include/scsi/libiscsi.h
@@ -144,6 +144,9 @@ struct iscsi_task {
 	void			*dd_data;	/* driver/transport data */
 };
 
+/* invalid scsi_task pointer */
+#define	INVALID_SCSI_TASK	(struct iscsi_task *)-1l
+
 static inline int iscsi_task_has_unsol_data(struct iscsi_task *task)
 {
 	return task->unsol_r2t.data_length > task->unsol_r2t.sent;
-- 
2.27.0

[PATCH AUTOSEL 4.4 8/8] scsi: target: iscsi: Fix cmd abort fabric stop race

[Sasha Levin]

From: Mike Christie <michael.christie@oracle.com>

[ Upstream commit f36199355c64a39fe82cfddc7623d827c7e050da ]

Maurizio found a race where the abort and cmd stop paths can race as
follows:

 1. thread1 runs iscsit_release_commands_from_conn and sets
    CMD_T_FABRIC_STOP.

 2. thread2 runs iscsit_aborted_task and then does __iscsit_free_cmd. It
    then returns from the aborted_task callout and we finish
    target_handle_abort and do:

    target_handle_abort -> transport_cmd_check_stop_to_fabric ->
	lio_check_stop_free -> target_put_sess_cmd

    The cmd is now freed.

 3. thread1 now finishes iscsit_release_commands_from_conn and runs
    iscsit_free_cmd while accessing a command we just released.

In __target_check_io_state we check for CMD_T_FABRIC_STOP and set the
CMD_T_ABORTED if the driver is not cleaning up the cmd because of a session
shutdown. However, iscsit_release_commands_from_conn only sets the
CMD_T_FABRIC_STOP and does not check to see if the abort path has claimed
completion ownership of the command.

This adds a check in iscsit_release_commands_from_conn so only the abort or
fabric stop path cleanup the command.

Link: https://lore.kernel.org/r/1605318378-9269-1-git-send-email-michael.christie@oracle.com
Reported-by: Maurizio Lombardi <mlombard@redhat.com>
Reviewed-by: Maurizio Lombardi <mlombard@redhat.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/target/iscsi/iscsi_target.c | 17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c
index cbb4414edd71b..c48aca1360c89 100644
--- a/drivers/target/iscsi/iscsi_target.c
+++ b/drivers/target/iscsi/iscsi_target.c
@@ -493,8 +493,7 @@ static void iscsit_aborted_task(struct iscsi_conn *conn, struct iscsi_cmd *cmd)
 	bool scsi_cmd = (cmd->iscsi_opcode == ISCSI_OP_SCSI_CMD);
 
 	spin_lock_bh(&conn->cmd_lock);
-	if (!list_empty(&cmd->i_conn_node) &&
-	    !(cmd->se_cmd.transport_state & CMD_T_FABRIC_STOP))
+	if (!list_empty(&cmd->i_conn_node))
 		list_del_init(&cmd->i_conn_node);
 	spin_unlock_bh(&conn->cmd_lock);
 
@@ -4228,12 +4227,22 @@ static void iscsit_release_commands_from_conn(struct iscsi_conn *conn)
 	spin_lock_bh(&conn->cmd_lock);
 	list_splice_init(&conn->conn_cmd_list, &tmp_list);
 
-	list_for_each_entry(cmd, &tmp_list, i_conn_node) {
+	list_for_each_entry_safe(cmd, cmd_tmp, &tmp_list, i_conn_node) {
 		struct se_cmd *se_cmd = &cmd->se_cmd;
 
 		if (se_cmd->se_tfo != NULL) {
 			spin_lock_irq(&se_cmd->t_state_lock);
-			se_cmd->transport_state |= CMD_T_FABRIC_STOP;
+			if (se_cmd->transport_state & CMD_T_ABORTED) {
+				/*
+				 * LIO's abort path owns the cleanup for this,
+				 * so put it back on the list and let
+				 * aborted_task handle it.
+				 */
+				list_move_tail(&cmd->i_conn_node,
+					       &conn->conn_cmd_list);
+			} else {
+				se_cmd->transport_state |= CMD_T_FABRIC_STOP;
+			}
 			spin_unlock_irq(&se_cmd->t_state_lock);
 		}
 	}
-- 
2.27.0