4.14.291-rc1 review

From: Miaoqian Lin <[email protected]>

commit 4b32e054335ea0ce50967f63a7bfd4db058b14b9 upstream.

of_parse_phandle() returns a node pointer with refcount incremented, we should use of_node_put() on it when not need anymore. Add missing of_node_put() to avoid refcount leak."

---
 drivers/pinctrl/nomadik/pinctrl-nomadik.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

@@ -1455,8 +1455,10 @@ static int nmk_pinctrl_dt_subnode_to_map
 
 	has_config = nmk_pinctrl_dt_get_config(np, &configs);
 	np_config = of_parse_phandle(np, "ste,config", 0);
-	if (np_config)
+	if (np_config) {
 		has_config |= nmk_pinctrl_dt_get_config(np_config, &configs);
+		of_node_put(np_config);
+	}
 	if (has_config) {
 		const char *gpio_name;
 		const char *pin;

From: Nikita Travkin <[email protected]>

commit 44339391c666e46cba522d19c65a6ad1071c68b7 upstream.

GPIO 31, 32 can be muxed to GCC_CAMSS_GP(1,2)_CLK respectively but the function was never assigned to the pingroup (even though the function exists already).

Add this mode to the related pins.

---
 drivers/pinctrl/qcom/pinctrl-msm8916.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -852,8 +852,8 @@ static const struct msm_pingroup msm8916
 	PINGROUP(28, pwr_modem_enabled_a, NA, NA, NA, NA, NA, qdss_tracedata_b, NA, atest_combodac),
 	PINGROUP(29, cci_i2c, NA, NA, NA, NA, NA, qdss_tracedata_b, NA, atest_combodac),
 	PINGROUP(30, cci_i2c, NA, NA, NA, NA, NA, NA, NA, qdss_tracedata_b),
-	PINGROUP(31, cci_timer0, NA, NA, NA, NA, NA, NA, NA, NA),
-	PINGROUP(32, cci_timer1, NA, NA, NA, NA, NA, NA, NA, NA),
+	PINGROUP(31, cci_timer0, flash_strobe, NA, NA, NA, NA, NA, NA, NA),
+	PINGROUP(32, cci_timer1, flash_strobe, NA, NA, NA, NA, NA, NA, NA),
 	PINGROUP(33, cci_async, NA, NA, NA, NA, NA, NA, NA, qdss_tracedata_b),
 	PINGROUP(34, pwr_nav_enabled_a, NA, NA, NA, NA, NA, NA, NA, qdss_tracedata_b),
 	PINGROUP(35, pwr_crypto_enabled_a, NA, NA, NA, NA, NA, NA, NA, qdss_tracedata_b),

From: Sakari Ailus <[email protected]>

commit 85140ef275f577f64e8a2c5789447222dfc14fc4 upstream.

The value acpi_add_nondev_subnodes() returns is bool so change the return type of the function to match that.

---
 drivers/acpi/property.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

@@ -132,10 +132,10 @@ static bool acpi_nondev_subnode_ok(acpi_
 	return acpi_nondev_subnode_data_ok(handle, link, list, parent);
 }
 
-static int acpi_add_nondev_subnodes(acpi_handle scope,
-				    const union acpi_object *links,
-				    struct list_head *list,
-				    struct fwnode_handle *parent)
+static bool acpi_add_nondev_subnodes(acpi_handle scope,
+				     const union acpi_object *links,
+				     struct list_head *list,
+				     struct fwnode_handle *parent)
 {
 	bool ret = false;
 	int i;

From: Matthias May <[email protected]>

commit ca2bb69514a8bc7f83914122f0d596371352416c upstream.

According to Guillaume Nault RT_TOS should never be used for IPv6.

Quote:
RT_TOS() is an old macro used to interprete IPv4 TOS as described in the obsolete RFC 1349. It's conceptually wrong to use it even in IPv4 code, although, given the current state of the code, most of the existing calls have no consequence.

But using RT_TOS() in IPv6 code is always a bug: IPv6 never had a "TOS" field to be interpreted the RFC 1349 way. There's no historical compatibility to worry about.

---
 drivers/net/geneve.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

@@ -795,8 +795,7 @@ static struct dst_entry *geneve_get_v6_d
 		use_cache = false;
 	}
 
-	fl6->flowlabel = ip6_make_flowinfo(RT_TOS(prio),
-					   info->key.label);
+	fl6->flowlabel = ip6_make_flowinfo(prio, info->key.label);
 	dst_cache = (struct dst_cache *)&info->dst_cache;
 	if (use_cache) {
 		dst = dst_cache_get_ip6(dst_cache, &fl6->saddr);

From: Peilin Ye <[email protected]>

commit 7e97cfed9929eaabc41829c395eb0d1350fccb9d upstream.

An O_NONBLOCK vsock_connect() request may try to reschedule @connect_work. Imagine the following sequence of vsock_connect() requests:

1. The 1st, non-blocking request schedules @connect_work, which will

     expire after 200 jiffies.  Socket state is now SS_CONNECTING;

2. Later, the 2nd, blocking request gets interrupted by a signal after a few jiffies while waiting for the connection to be established. Socket state is back to SS_UNCONNECTED, but @connect_work is still pending, and will expire after 100 jiffies.

3. Now, the 3rd, non-blocking request tries to schedule @connect_work again. Since @connect_work is already scheduled, schedule_delayed_work() silently returns. sock_hold() is called twice, but sock_put() will only be called once in vsock_connect_timeout(), causing a memory leak reported by syzbot:

  BUG: memory leak
  unreferenced object 0xffff88810ea56a40 (size 1232):
    comm "syz-executor756", pid 3604, jiffies 4294947681 (age 12.350s)
    hex dump (first 32 bytes):
      00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
      28 00 07 40 00 00 00 00 00 00 00 00 00 00 00 00  (..@............
    backtrace:
      [<ffffffff837c830e>] sk_prot_alloc+0x3e/0x1b0 net/core/sock.c:1930
      [<ffffffff837cbe22>] sk_alloc+0x32/0x2e0 net/core/sock.c:1989
      [<ffffffff842ccf68>] __vsock_create.constprop.0+0x38/0x320 net/vmw_vsock/af_vsock.c:734
      [<ffffffff842ce8f1>] vsock_create+0xc1/0x2d0 net/vmw_vsock/af_vsock.c:2203
      [<ffffffff837c0cbb>] __sock_create+0x1ab/0x2b0 net/socket.c:1468
      [<ffffffff837c3acf>] sock_create net/socket.c:1519 [inline]
      [<ffffffff837c3acf>] __sys_socket+0x6f/0x140 net/socket.c:1561
      [<ffffffff837c3bba>] __do_sys_socket net/socket.c:1570 [inline]
      [<ffffffff837c3bba>] __se_sys_socket net/socket.c:1568 [inline]
      [<ffffffff837c3bba>] __x64_sys_socket+0x1a/0x20 net/socket.c:1568
      [<ffffffff84512815>] do_syscall_x64 arch/x86/entry/common.c:50 [inline]
      [<ffffffff84512815>] do_syscall_64+0x35/0x80 arch/x86/entry/common.c:80
      [<ffffffff84600068>] entry_SYSCALL_64_after_hwframe+0x44/0xae
  <...>

Use mod_delayed_work() instead: if @connect_work is already scheduled, reschedule it, and undo sock_hold() to keep the reference count balanced.

---
 net/vmw_vsock/af_vsock.c |    9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

@@ -1222,7 +1222,14 @@ static int vsock_stream_connect(struct s
 			 * timeout fires.
 			 */
 			sock_hold(sk);
-			schedule_delayed_work(&vsk->connect_work, timeout);
+
+			/* If the timeout function is already scheduled,
+			 * reschedule it, then ungrab the socket refcount to
+			 * keep it balanced.
+			 */
+			if (mod_delayed_work(system_wq, &vsk->connect_work,
+					     timeout))
+				sock_put(sk);
 
 			/* Skip ahead to preserve error code set above. */
 			goto out_wait;

From: Peilin Ye <[email protected]>

commit a3e7b29e30854ed67be0d17687e744ad0c769c4b upstream.

Imagine two non-blocking vsock_connect() requests on the same socket. The first request schedules @connect_work, and after it times out, vsock_connect_timeout() sets *sock* state back to TCP_CLOSE, but keeps *socket* state as SS_CONNECTING.

Later, the second request returns -EALREADY, meaning the socket "already has a pending connection in progress", even though the first request has already timed out.

As suggested by Stefano, fix it by setting *socket* state back to SS_UNCONNECTED, so that the second request will return -ETIMEDOUT.

---
 net/vmw_vsock/af_vsock.c |    1 +
 1 file changed, 1 insertion(+)

@@ -1125,6 +1125,7 @@ static void vsock_connect_timeout(struct
 	if (sk->sk_state == TCP_SYN_SENT &&
 	    (sk->sk_shutdown != SHUTDOWN_MASK)) {
 		sk->sk_state = TCP_CLOSE;
+		sk->sk_socket->state = SS_UNCONNECTED;
 		sk->sk_err = ETIMEDOUT;
 		sk->sk_error_report(sk);
 		vsock_transport_cancel_pkt(vsk);

From: Roberto Sassu <[email protected]>

commit 5b245985a6de5ac18b5088c37068816d413fb8ed upstream.

Switch to new EVP API for detecting libcrypto, as Fedora 36 returns an error when it encounters the deprecated function MD5_Init() and the others.

The error would be interpreted as missing libcrypto, while in reality it is not.

---
 tools/build/feature/test-libcrypto.c |   15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)

@@ -1,16 +1,23 @@
 // SPDX-License-Identifier: GPL-2.0
+#include <openssl/evp.h>
 #include <openssl/sha.h>
 #include <openssl/md5.h>
 
 int main(void)
 {
-	MD5_CTX context;
+	EVP_MD_CTX *mdctx;
 	unsigned char md[MD5_DIGEST_LENGTH + SHA_DIGEST_LENGTH];
 	unsigned char dat[] = "12345";
+	unsigned int digest_len;
 
-	MD5_Init(&context);
-	MD5_Update(&context, &dat[0], sizeof(dat));
-	MD5_Final(&md[0], &context);
+	mdctx = EVP_MD_CTX_new();
+	if (!mdctx)
+		return 0;
+
+	EVP_DigestInit_ex(mdctx, EVP_md5(), NULL);
+	EVP_DigestUpdate(mdctx, &dat[0], sizeof(dat));
+	EVP_DigestFinal_ex(mdctx, &md[0], &digest_len);
+	EVP_MD_CTX_free(mdctx);
 
 	SHA1(&dat[0], sizeof(dat), &md[0]);
 

From: Dan Carpenter <[email protected]>

commit 32ad11127b95236dfc52375f3707853194a7f4b4 upstream.

This code tries to store -EFAULT in an unsigned int. The xenbus_file_read() function returns type ssize_t so the negative value is returned as a positive value to the user.

This change forces another change to the min() macro. Originally, the min() macro used "unsigned" type which checkpatch complains about. Also unsigned type would break if "len" were not capped at MAX_RW_COUNT. Use size_t for the min(). (No effect on runtime for the min_t() change).

---
 drivers/xen/xenbus/xenbus_dev_frontend.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -125,7 +125,7 @@ static ssize_t xenbus_file_read(struct f
 {
 	struct xenbus_file_priv *u = filp->private_data;
 	struct read_buffer *rb;
-	unsigned i;
+	ssize_t i;
 	int ret;
 
 	mutex_lock(&u->reply_mutex);

@@ -145,7 +145,7 @@ again:
 	rb = list_entry(u->read_buffers.next, struct read_buffer, list);
 	i = 0;
 	while (i < len) {
-		unsigned sz = min((unsigned)len - i, rb->len - rb->cons);
+		size_t sz = min_t(size_t, len - i, rb->len - rb->cons);
 
 		ret = copy_to_user(ubuf + i, &rb->msg[rb->cons], sz);
 

From: Duoming Zhou <[email protected]>

commit 3f4093e2bf4673f218c0bf17d8362337c400e77b upstream.

There are use-after-free bugs caused by tst_timer. The root cause is that there are no functions to stop tst_timer in idt77252_exit(). One of the possible race conditions is shown below:

    (thread 1)          |        (thread 2)
                        |  idt77252_init_one
                        |    init_card
                        |      fill_tst
                        |        mod_timer(&card->tst_timer, ...)
idt77252_exit           |  (wait a time)
                        |  tst_timer
                        |
                        |    ...
  kfree(card) // FREE   |
                        |    card->soft_tst[e] // USE

The idt77252_dev is deallocated in idt77252_exit() and used in timer handler.

This patch adds del_timer_sync() in idt77252_exit() in order that the timer handler could be stopped before the idt77252_dev is deallocated.

---
 drivers/atm/idt77252.c |    1 +
 1 file changed, 1 insertion(+)

@@ -3768,6 +3768,7 @@ static void __exit idt77252_exit(void)
 		card = idt77252_chain;
 		dev = card->atmdev;
 		idt77252_chain = card->next;
+		del_timer_sync(&card->tst_timer);
 
 		if (dev->phy->stop)
 			dev->phy->stop(dev);

From: Al Viro <[email protected]>

commit 8535c239ac674f7ead0f2652932d35c52c4123b2 upstream.

make sure that ->orig_r2 is negative for everything except the syscalls.

---
 arch/nios2/include/asm/entry.h |    3 ++-
 arch/nios2/kernel/entry.S      |    4 +---
 2 files changed, 3 insertions(+), 4 deletions(-)

@@ -50,7 +50,8 @@
 	stw	r13, PT_R13(sp)
 	stw	r14, PT_R14(sp)
 	stw	r15, PT_R15(sp)
-	stw	r2, PT_ORIG_R2(sp)
+	movi	r24, -1
+	stw	r24, PT_ORIG_R2(sp)
 	stw	r7, PT_ORIG_R7(sp)
 
 	stw	ra, PT_RA(sp)

@@ -185,6 +185,7 @@ ENTRY(handle_system_call)
 	ldw	r5, PT_R5(sp)
 
 local_restart:
+	stw	r2, PT_ORIG_R2(sp)
 	/* Check that the requested system call is within limits */
 	movui	r1, __NR_syscalls
 	bgeu	r2, r1, ret_invsyscall

@@ -336,9 +337,6 @@ external_interrupt:
 	/* skip if no interrupt is pending */
 	beq	r12, r0, ret_from_interrupt
 
-	movi	r24, -1
-	stw	r24, PT_ORIG_R2(sp)
-
 	/*
 	 * Process an external hardware interrupt.
 	 */

From: Al Viro <[email protected]>

commit 45ec746c65097c25e77d24eae8fee0def5b6cc5d upstream.

fill the gaps in there with sys_ni_syscall, as everyone does...

---
 arch/nios2/kernel/entry.S         |    1 -
 arch/nios2/kernel/syscall_table.c |    1 +
 2 files changed, 1 insertion(+), 1 deletion(-)

@@ -193,7 +193,6 @@ local_restart:
 	movhi	r11, %hiadj(sys_call_table)
 	add	r1, r1, r11
 	ldw	r1, %lo(sys_call_table)(r1)
-	beq	r1, r0, ret_invsyscall
 
 	/* Check if we are being traced */
 	GET_THREAD_INFO r11

@@ -25,5 +25,6 @@
 #define __SYSCALL(nr, call) [nr] = (call),
 
 void *sys_call_table[__NR_syscalls] = {
+	[0 ... __NR_syscalls-1] = sys_ni_syscall,
 #include <asm/unistd.h>
 };

From: Al Viro <[email protected]>

commit 25ba820ef36bdbaf9884adeac69b6e1821a7df76 upstream.

all checks done before letting the tracer modify the register state are worthless...

---
 arch/nios2/kernel/entry.S |   11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

@@ -255,9 +255,9 @@ traced_system_call:
 	ldw	r6, PT_R6(sp)
 	ldw	r7, PT_R7(sp)
 
-	/* Fetch the syscall function, we don't need to check the boundaries
-	 * since this is already done.
-	 */
+	/* Fetch the syscall function. */
+	movui	r1, __NR_syscalls
+	bgeu	r2, r1, traced_invsyscall
 	slli	r1, r2, 2
 	movhi	r11,%hiadj(sys_call_table)
 	add	r1, r1, r11

@@ -287,6 +287,11 @@ end_translate_rc_and_ret2:
 	RESTORE_SWITCH_STACK
 	br	ret_from_exception
 
+	/* If the syscall number was invalid return ENOSYS */
+traced_invsyscall:
+	movi	r2, -ENOSYS
+	br	translate_rc_and_ret2
+
 Luser_return:
 	GET_THREAD_INFO	r11			/* get thread_info pointer */
 	ldw	r10, TI_FLAGS(r11)		/* get thread_info->flags */

From: Al Viro <[email protected]>

commit 2d631bd58fe0ea3e3350212e23c9aba1fb606514 upstream.

sys_foo() returns -512 (aka -ERESTARTSYS) => do_signal() sees 512 in r2 and 1 in r1.

sys_foo() returns 512 => do_signal() sees 512 in r2 and 0 in r1.

The former is restart-worthy; the latter obviously isn't.

---
 arch/nios2/kernel/signal.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -240,7 +240,7 @@ static int do_signal(struct pt_regs *reg
 	/*
 	 * If we were from a system call, check for system call restarting...
 	 */
-	if (regs->orig_r2 >= 0) {
+	if (regs->orig_r2 >= 0 && regs->r1) {
 		continue_addr = regs->ea;
 		restart_addr = continue_addr - 4;
 		retval = regs->r2;

From: Al Viro <[email protected]>

commit 411a76b7219555c55867466c82d70ce928d6c9e1 upstream.

---
 arch/nios2/kernel/signal.c |    1 +
 1 file changed, 1 insertion(+)

@@ -261,6 +261,7 @@ static int do_signal(struct pt_regs *reg
 			regs->ea = restart_addr;
 			break;
 		}
+		regs->orig_r2 = -1;
 	}
 
 	if (get_signal(&ksig)) {

From: Al Viro <[email protected]>

commit fd0c153daad135d0ec1a53c5dbe6936a724d6ae1 upstream.

If we use the ancient SysV syscall ABI, we'd better have tell the kernel how to claim that a negative return value is a success. Use ->orig_r2 for that - it's inaccessible via ptrace, so it's a fair game for changes and it's normally[*] non-negative on return from syscall. Set to -1; syscall is not going to be restart-worthy by definition, so we won't interfere with that use either.

[*] the only exception is rt_sigreturn(), where we skip the entire messing with r1/r2 anyway.

---
 arch/nios2/include/asm/ptrace.h |    2 ++
 arch/nios2/kernel/entry.S       |    6 ++++++
 2 files changed, 8 insertions(+)

@@ -74,6 +74,8 @@ extern void show_regs(struct pt_regs *);
 	((struct pt_regs *)((unsigned long)current_thread_info() + THREAD_SIZE)\
 		- 1)
 
+#define force_successful_syscall_return() (current_pt_regs()->orig_r2 = -1)
+
 int do_syscall_trace_enter(void);
 void do_syscall_trace_exit(void);
 #endif /* __ASSEMBLY__ */

@@ -213,6 +213,9 @@ local_restart:
 translate_rc_and_ret:
 	movi	r1, 0
 	bge	r2, zero, 3f
+	ldw	r1, PT_ORIG_R2(sp)
+	addi	r1, r1, 1
+	beq	r1, zero, 3f
 	sub	r2, zero, r2
 	movi	r1, 1
 3:

@@ -276,6 +279,9 @@ traced_system_call:
 translate_rc_and_ret2:
 	movi	r1, 0
 	bge	r2, zero, 4f
+	ldw	r1, PT_ORIG_R2(sp)
+	addi	r1, r1, 1
+	beq	r1, zero, 4f
 	sub	r2, zero, r2
 	movi	r1, 1
 4:

From: Pablo Neira Ayuso <[email protected]>

commit 271c5ca826e0c3c53e0eb4032f8eaedea1ee391c upstream.

While looping to build the bitmap of used anonymous set names, check the current set in the iteration, instead of the one that is being created.

---
 net/netfilter/nf_tables_api.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -2797,7 +2797,7 @@ cont:
 		list_for_each_entry(i, &ctx->table->sets, list) {
 			int tmp;
 
-			if (!nft_is_active_next(ctx->net, set))
+			if (!nft_is_active_next(ctx->net, i))
 				continue;
 			if (!sscanf(i->name, name, &tmp))
 				continue;

From: Michael Ellerman <[email protected]>

commit 8d48562a2729742f767b0fdd994d6b2a56a49c63 upstream.

The recent change to get_phb_number() causes a DEBUG_ATOMIC_SLEEP warning on some systems:

  BUG: sleeping function called from invalid context at kernel/locking/mutex.c:580
  in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 1, name: swapper
  preempt_count: 1, expected: 0
  RCU nest depth: 0, expected: 0
  1 lock held by swapper/1:
   #0: c157efb0 (hose_spinlock){+.+.}-{2:2}, at: pcibios_alloc_controller+0x64/0x220
  Preemption disabled at:
  [<00000000>] 0x0
  CPU: 0 PID: 1 Comm: swapper Not tainted 5.19.0-yocto-standard+ #1
  Call Trace:
  [d101dc90] [c073b264] dump_stack_lvl+0x50/0x8c (unreliable)
  [d101dcb0] [c0093b70] __might_resched+0x258/0x2a8
  [d101dcd0] [c0d3e634] __mutex_lock+0x6c/0x6ec
  [d101dd50] [c0a84174] of_alias_get_id+0x50/0xf4
  [d101dd80] [c002ec78] pcibios_alloc_controller+0x1b8/0x220
  [d101ddd0] [c140c9dc] pmac_pci_init+0x198/0x784
  [d101de50] [c140852c] discover_phbs+0x30/0x4c
  [d101de60] [c0007fd4] do_one_initcall+0x94/0x344
  [d101ded0] [c1403b40] kernel_init_freeable+0x1a8/0x22c
  [d101df10] [c00086e0] kernel_init+0x34/0x160
  [d101df30] [c001b334] ret_from_kernel_thread+0x5c/0x64

This is because pcibios_alloc_controller() holds hose_spinlock but of_alias_get_id() takes of_mutex which can sleep.

The hose_spinlock protects the phb_bitmap, and also the hose_list, but it doesn't need to be held while get_phb_number() calls the OF routines, because those are only looking up information in the device tree.

So fix it by having get_phb_number() take the hose_spinlock itself, only where required, and then dropping the lock before returning. pcibios_alloc_controller() then needs to take the lock again before the list_add() but that's safe, the order of the list is not important.

---
 arch/powerpc/kernel/pci-common.c |   16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

@@ -73,10 +73,6 @@ const struct dma_map_ops *get_pci_dma_op
 }
 EXPORT_SYMBOL(get_pci_dma_ops);
 
-/*
- * This function should run under locking protection, specifically
- * hose_spinlock.
- */
 static int get_phb_number(struct device_node *dn)
 {
 	int ret, phb_id = -1;

@@ -113,15 +109,20 @@ static int get_phb_number(struct device_
 	if (!ret)
 		phb_id = (int)(prop & (MAX_PHBS - 1));
 
+	spin_lock(&hose_spinlock);
+
 	/* We need to be sure to not use the same PHB number twice. */
 	if ((phb_id >= 0) && !test_and_set_bit(phb_id, phb_bitmap))
-		return phb_id;
+		goto out_unlock;
 
 	/* If everything fails then fallback to dynamic PHB numbering. */
 	phb_id = find_first_zero_bit(phb_bitmap, MAX_PHBS);
 	BUG_ON(phb_id >= MAX_PHBS);
 	set_bit(phb_id, phb_bitmap);
 
+out_unlock:
+	spin_unlock(&hose_spinlock);
+
 	return phb_id;
 }
 

@@ -132,10 +133,13 @@ struct pci_controller *pcibios_alloc_con
 	phb = zalloc_maybe_bootmem(sizeof(struct pci_controller), GFP_KERNEL);
 	if (phb == NULL)
 		return NULL;
-	spin_lock(&hose_spinlock);
+
 	phb->global_number = get_phb_number(dev);
+
+	spin_lock(&hose_spinlock);
 	list_add_tail(&phb->list_node, &hose_list);
 	spin_unlock(&hose_spinlock);
+
 	phb->dn = dev;
 	phb->is_dynamic = slab_is_available();
 #ifdef CONFIG_PPC64

From: Alan Brady <[email protected]>

commit 57c942bc3bef0970f0b21f8e0998e76a900ea80d upstream.

When a tx_timeout fires, the PF attempts to recover by incrementally resetting. First we try a PFR, then CORER and finally a GLOBR. If the GLOBR fails, then we keep hitting the tx_timeout and incrementing the recovery level and issuing dmesgs, which is both annoying to the user and accomplishes nothing.

If the GLOBR fails, then we're pretty much totally hosed, and there's not much else we can do to recover, so this makes it such that we just kill the VSI and stop hitting the tx_timeout in such a case.

---
 drivers/net/ethernet/intel/i40e/i40e_main.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

@@ -405,7 +405,9 @@ static void i40e_tx_timeout(struct net_d
 		set_bit(__I40E_GLOBAL_RESET_REQUESTED, pf->state);
 		break;
 	default:
-		netdev_err(netdev, "tx_timeout recovery unsuccessful\n");
+		netdev_err(netdev, "tx_timeout recovery unsuccessful, device is in non-recoverable state.\n");
+		set_bit(__I40E_DOWN_REQUESTED, pf->state);
+		set_bit(__I40E_VSI_DOWN_REQUESTED, vsi->state);
 		break;
 	}
 

From: Csókás Bence <[email protected]>

commit 61d5e2a251fb20c2c5e998c3f1d52ed6d5360319 upstream.

Code reimplements functionality already in fec_ptp_read(), but misses check for FEC_QUIRK_BUG_CAPTURE. Replace with function call.

---
 drivers/net/ethernet/freescale/fec_ptp.c |    6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

@@ -155,11 +155,7 @@ static int fec_ptp_enable_pps(struct fec
 		 * NSEC_PER_SEC - ts.tv_nsec. Add the remaining nanoseconds
 		 * to current timer would be next second.
 		 */
-		tempval = readl(fep->hwp + FEC_ATIME_CTRL);
-		tempval |= FEC_T_CTRL_CAPTURE;
-		writel(tempval, fep->hwp + FEC_ATIME_CTRL);
-
-		tempval = readl(fep->hwp + FEC_ATIME);
+		tempval = fep->cc.read(&fep->cc);
 		/* Convert the ptp local counter to 1588 timestamp */
 		ns = timecounter_cyc2time(&fep->tc, tempval);
 		ts = ns_to_timespec64(ns);

From: Lin Ma <[email protected]>

commit 6faee3d4ee8be0f0367d0c3d826afb3571b7a5e0 upstream.

The commit c23d92b80e0b ("igb: Teardown SR-IOV before unregister_netdev()") places the unregister_netdev() call after the igb_disable_sriov() call to avoid functionality issue.

However, it introduces several race conditions when detaching a device. For example, when .remove() is called, the below interleaving leads to use-after-free.

 (FREE from device detaching)      |   (USE from netdev core)
igb_remove                         |  igb_ndo_get_vf_config
 igb_disable_sriov                 |  vf >= adapter->vfs_allocated_count?
  kfree(adapter->vf_data)          |
  adapter->vfs_allocated_count = 0 |
                                   |    memcpy(... adapter->vf_data[vf]

Moreover, the igb_disable_sriov() also suffers from data race with the requests from VF driver.

 (FREE from device detaching)      |   (USE from requests)
igb_remove                         |  igb_msix_other
 igb_disable_sriov                 |   igb_msg_task
  kfree(adapter->vf_data)          |    vf < adapter->vfs_allocated_count
  adapter->vfs_allocated_count = 0 |

To this end, this commit first eliminates the data races from netdev core by using rtnl_lock (similar to commit 719479230893 ("dpaa2-eth: add MAC/PHY support through phylink")). And then adds a spinlock to eliminate races from driver requests. (similar to commit 1e53834ce541 ("ixgbe: Add locking to prevent panic when setting sriov_numvfs to zero")

---
 drivers/net/ethernet/intel/igb/igb.h      |    2 ++
 drivers/net/ethernet/intel/igb/igb_main.c |   12 +++++++++++-
 2 files changed, 13 insertions(+), 1 deletion(-)

@@ -599,6 +599,8 @@ struct igb_adapter {
 	struct igb_mac_addr *mac_table;
 	struct vf_mac_filter vf_macs;
 	struct vf_mac_filter *vf_mac_list;
+	/* lock for VF resources */
+	spinlock_t vfs_lock;
 };
 
 /* flags controlling PTP/1588 function */

@@ -2805,6 +2805,7 @@ static int igb_disable_sriov(struct pci_
 	struct net_device *netdev = pci_get_drvdata(pdev);
 	struct igb_adapter *adapter = netdev_priv(netdev);
 	struct e1000_hw *hw = &adapter->hw;
+	unsigned long flags;
 
 	/* reclaim resources allocated to VFs */
 	if (adapter->vf_data) {

@@ -2817,12 +2818,13 @@ static int igb_disable_sriov(struct pci_
 			pci_disable_sriov(pdev);
 			msleep(500);
 		}
-
+		spin_lock_irqsave(&adapter->vfs_lock, flags);
 		kfree(adapter->vf_mac_list);
 		adapter->vf_mac_list = NULL;
 		kfree(adapter->vf_data);
 		adapter->vf_data = NULL;
 		adapter->vfs_allocated_count = 0;
+		spin_unlock_irqrestore(&adapter->vfs_lock, flags);
 		wr32(E1000_IOVCTL, E1000_IOVCTL_REUSE_VFQ);
 		wrfl();
 		msleep(100);

@@ -2984,7 +2986,9 @@ static void igb_remove(struct pci_dev *p
 	igb_release_hw_control(adapter);
 
 #ifdef CONFIG_PCI_IOV
+	rtnl_lock();
 	igb_disable_sriov(pdev);
+	rtnl_unlock();
 #endif
 
 	unregister_netdev(netdev);

@@ -3137,6 +3141,9 @@ static int igb_sw_init(struct igb_adapte
 
 	spin_lock_init(&adapter->nfc_lock);
 	spin_lock_init(&adapter->stats64_lock);
+
+	/* init spinlock to avoid concurrency of VF resources */
+	spin_lock_init(&adapter->vfs_lock);
 #ifdef CONFIG_PCI_IOV
 	switch (hw->mac.type) {
 	case e1000_82576:

@@ -6776,8 +6783,10 @@ unlock:
 static void igb_msg_task(struct igb_adapter *adapter)
 {
 	struct e1000_hw *hw = &adapter->hw;
+	unsigned long flags;
 	u32 vf;
 
+	spin_lock_irqsave(&adapter->vfs_lock, flags);
 	for (vf = 0; vf < adapter->vfs_allocated_count; vf++) {
 		/* process any reset requests */
 		if (!igb_check_for_rst(hw, vf))

@@ -6791,6 +6800,7 @@ static void igb_msg_task(struct igb_adap
 		if (!igb_check_for_ack(hw, vf))
 			igb_rcv_ack_from_vf(adapter, vf);
 	}
+	spin_unlock_irqrestore(&adapter->vfs_lock, flags);
 }
 
 /**

From: Masahiro Yamada <[email protected]>

commit ce99d0bf312daf0178e640da9e3c93b773a67e7d upstream.

Currently LDFLAGS is not cleared, so same flags are accumulated in LDFLAGS when the top Makefile is recursively invoked.

I found unneeded rebuild for ARCH=arm64 when CONFIG_TRIM_UNUSED_KSYMS is enabled. If include/generated/autoksyms.h is updated, the top Makefile is recursively invoked, then arch/arm64/Makefile adds one more '-maarch64linux'. Due to the command line change, modules are rebuilt needlessly.

---
 Makefile |    1 +
 1 file changed, 1 insertion(+)

@@ -426,6 +426,7 @@ KBUILD_CFLAGS_KERNEL :=
 KBUILD_AFLAGS_MODULE  := -DMODULE
 KBUILD_CFLAGS_MODULE  := -DMODULE
 KBUILD_LDFLAGS_MODULE := -T $(srctree)/scripts/module-common.lds
+LDFLAGS :=
 GCC_PLUGINS_CFLAGS :=
 CLANG_FLAGS :=
 

From: Qu Wenruo <[email protected]>

commit bd8f7e627703ca5707833d623efcd43f104c7b3f upstream.

If we have only 8K partial write at the beginning of a full RAID56 stripe, we will write the following contents:

0 8K 32K 64K
Disk 1 (data): |XX| | |

Disk 2  (data):     |               |               |
Disk 3  (parity):   |XXXXXXXXXXXXXXX|XXXXXXXXXXXXXXX|

|X| means the sector will be written back to disk.

Note that, although we won't write any sectors from disk 2, but we will write the full 64KiB of parity to disk.

This behavior is fine for now, but not for the future (especially for RAID56J, as we waste quite some space to journal the unused parity stripes).

So here we will also utilize the btrfs_raid_bio::dbitmap, anytime we queue a higher level bio into an rbio, we will update rbio::dbitmap to indicate which vertical stripes we need to writeback.

And at finish_rmw(), we also check dbitmap to see if we need to write any sector in the vertical stripe.

So after the patch, above example will only lead to the following writeback pattern:

0 8K 32K 64K
Disk 1 (data): |XX| | |

Disk 2  (data):     |               |               |
Disk 3  (parity):   |XX|            |               |

---
 fs/btrfs/raid56.c |   55 ++++++++++++++++++++++++++++++++++++++++++++++++++----
 1 file changed, 51 insertions(+), 4 deletions(-)

@@ -334,6 +334,9 @@ static void merge_rbio(struct btrfs_raid
 {
 	bio_list_merge(&dest->bio_list, &victim->bio_list);
 	dest->bio_list_bytes += victim->bio_list_bytes;
+	/* Also inherit the bitmaps from @victim. */
+	bitmap_or(dest->dbitmap, victim->dbitmap, dest->dbitmap,
+		  dest->stripe_npages);
 	dest->generic_bio_cnt += victim->generic_bio_cnt;
 	bio_list_init(&victim->bio_list);
 }

@@ -882,6 +885,12 @@ static void rbio_orig_end_io(struct btrf
 
 	if (rbio->generic_bio_cnt)
 		btrfs_bio_counter_sub(rbio->fs_info, rbio->generic_bio_cnt);
+	/*
+	 * Clear the data bitmap, as the rbio may be cached for later usage.
+	 * do this before before unlock_stripe() so there will be no new bio
+	 * for this bio.
+	 */
+	bitmap_clear(rbio->dbitmap, 0, rbio->stripe_npages);
 
 	/*
 	 * At this moment, rbio->bio_list is empty, however since rbio does not

@@ -1204,6 +1213,9 @@ static noinline void finish_rmw(struct b
 	else
 		BUG();
 
+	/* We should have at least one data sector. */
+	ASSERT(bitmap_weight(rbio->dbitmap, rbio->stripe_npages));
+
 	/* at this point we either have a full stripe,
 	 * or we've read the full stripe from the drive.
 	 * recalculate the parity and write the new results.

@@ -1277,6 +1289,11 @@ static noinline void finish_rmw(struct b
 	for (stripe = 0; stripe < rbio->real_stripes; stripe++) {
 		for (pagenr = 0; pagenr < rbio->stripe_npages; pagenr++) {
 			struct page *page;
+
+			/* This vertical stripe has no data, skip it. */
+			if (!test_bit(pagenr, rbio->dbitmap))
+				continue;
+
 			if (stripe < rbio->nr_data) {
 				page = page_in_rbio(rbio, stripe, pagenr, 1);
 				if (!page)

@@ -1301,6 +1318,11 @@ static noinline void finish_rmw(struct b
 
 		for (pagenr = 0; pagenr < rbio->stripe_npages; pagenr++) {
 			struct page *page;
+
+			/* This vertical stripe has no data, skip it. */
+			if (!test_bit(pagenr, rbio->dbitmap))
+				continue;
+
 			if (stripe < rbio->nr_data) {
 				page = page_in_rbio(rbio, stripe, pagenr, 1);
 				if (!page)

@@ -1745,6 +1767,33 @@ static void btrfs_raid_unplug(struct blk
 	run_plug(plug);
 }
 
+/* Add the original bio into rbio->bio_list, and update rbio::dbitmap. */
+static void rbio_add_bio(struct btrfs_raid_bio *rbio, struct bio *orig_bio)
+{
+	const struct btrfs_fs_info *fs_info = rbio->fs_info;
+	const u64 orig_logical = orig_bio->bi_iter.bi_sector << SECTOR_SHIFT;
+	const u64 full_stripe_start = rbio->bbio->raid_map[0];
+	const u32 orig_len = orig_bio->bi_iter.bi_size;
+	const u32 sectorsize = fs_info->sectorsize;
+	u64 cur_logical;
+
+	ASSERT(orig_logical >= full_stripe_start &&
+	       orig_logical + orig_len <= full_stripe_start +
+	       rbio->nr_data * rbio->stripe_len);
+
+	bio_list_add(&rbio->bio_list, orig_bio);
+	rbio->bio_list_bytes += orig_bio->bi_iter.bi_size;
+
+	/* Update the dbitmap. */
+	for (cur_logical = orig_logical; cur_logical < orig_logical + orig_len;
+	     cur_logical += sectorsize) {
+		int bit = ((u32)(cur_logical - full_stripe_start) >>
+			   PAGE_SHIFT) % rbio->stripe_npages;
+
+		set_bit(bit, rbio->dbitmap);
+	}
+}
+
 /*
  * our main entry point for writes from the rest of the FS.
  */

@@ -1761,9 +1810,8 @@ int raid56_parity_write(struct btrfs_fs_
 		btrfs_put_bbio(bbio);
 		return PTR_ERR(rbio);
 	}
-	bio_list_add(&rbio->bio_list, bio);
-	rbio->bio_list_bytes = bio->bi_iter.bi_size;
 	rbio->operation = BTRFS_RBIO_WRITE;
+	rbio_add_bio(rbio, bio);
 
 	btrfs_bio_counter_inc_noblocked(fs_info);
 	rbio->generic_bio_cnt = 1;

@@ -2146,8 +2194,7 @@ int raid56_parity_recover(struct btrfs_f
 	}
 
 	rbio->operation = BTRFS_RBIO_READ_REBUILD;
-	bio_list_add(&rbio->bio_list, bio);
-	rbio->bio_list_bytes = bio->bi_iter.bi_size;
+	rbio_add_bio(rbio, bio);
 
 	rbio->faila = find_logical_bio_stripe(rbio, bio);
 	if (rbio->faila == -1) {

From: Qu Wenruo <[email protected]>

commit f6065f8edeb25f4a9dfe0b446030ad995a84a088 upstream.

[BUG]
There is a small workload which will always fail with recent kernel:
(A simplified version from btrfs/125 test case)

  mkfs.btrfs -f -m raid5 -d raid5 -b 1G $dev1 $dev2 $dev3
  mount $dev1 $mnt
  xfs_io -f -c "pwrite -S 0xee 0 1M" $mnt/file1
  sync
  umount $mnt
  btrfs dev scan -u $dev3
  mount -o degraded $dev1 $mnt
  xfs_io -f -c "pwrite -S 0xff 0 128M" $mnt/file2
  umount $mnt
  btrfs dev scan
  mount $dev1 $mnt
  btrfs balance start --full-balance $mnt
  umount $mnt

The failure is always failed to read some tree blocks:

BTRFS info (device dm-4): relocating block group 217710592 flags data|raid5 BTRFS error (device dm-4): parent transid verify failed on 38993920 wanted 9 found 7 BTRFS error (device dm-4): parent transid verify failed on 38993920 wanted 9 found 7
...

[CAUSE]
With the recently added debug output, we can see all RAID56 operations related to full stripe 38928384:

56.1183: raid56_read_partial: full_stripe=38928384 devid=2 type=DATA1 offset=0 opf=0x0 physical=9502720 len=65536 56.1185: raid56_read_partial: full_stripe=38928384 devid=3 type=DATA2 offset=16384 opf=0x0 physical=9519104 len=16384 56.1185: raid56_read_partial: full_stripe=38928384 devid=3 type=DATA2 offset=49152 opf=0x0 physical=9551872 len=16384 56.1187: raid56_write_stripe: full_stripe=38928384 devid=3 type=DATA2 offset=0 opf=0x1 physical=9502720 len=16384 56.1188: raid56_write_stripe: full_stripe=38928384 devid=3 type=DATA2 offset=32768 opf=0x1 physical=9535488 len=16384 56.1188: raid56_write_stripe: full_stripe=38928384 devid=1 type=PQ1 offset=0 opf=0x1 physical=30474240 len=16384 56.1189: raid56_write_stripe: full_stripe=38928384 devid=1 type=PQ1 offset=32768 opf=0x1 physical=30507008 len=16384 56.1218: raid56_write_stripe: full_stripe=38928384 devid=3 type=DATA2 offset=49152 opf=0x1 physical=9551872 len=16384 56.1219: raid56_write_stripe: full_stripe=38928384 devid=1 type=PQ1 offset=49152 opf=0x1 physical=30523392 len=16384 56.2721: raid56_parity_recover: full stripe=38928384 eb=39010304 mirror=2
56.2723: raid56_parity_recover: full stripe=38928384 eb=39010304 mirror=2
56.2724: raid56_parity_recover: full stripe=38928384 eb=39010304 mirror=2

Before we enter raid56_parity_recover(), we have triggered some metadata write for the full stripe 38928384, this leads to us to read all the sectors from disk.

Furthermore, btrfs raid56 write will cache its calculated P/Q sectors to avoid unnecessary read.

This means, for that full stripe, after any partial write, we will have stale data, along with P/Q calculated using that stale data.

Thankfully due to patch "btrfs: only write the sectors in the vertical stripe which has data stripes" we haven't submitted all the corrupted P/Q to disk.

When we really need to recover certain range, aka in raid56_parity_recover(), we will use the cached rbio, along with its cached sectors (the full stripe is all cached).

This explains why we have no event raid56_scrub_read_recover() triggered.

Since we have the cached P/Q which is calculated using the stale data, the recovered one will just be stale.

In our particular test case, it will always return the same incorrect metadata, thus causing the same error message "parent transid verify failed on 39010304 wanted 9 found 7" again and again.

[BTRFS DESTRUCTIVE RMW PROBLEM]

Test case btrfs/125 (and above workload) always has its trouble with the destructive read-modify-write (RMW) cycle:

        0       32K     64K
Data1:  | Good  | Good  |
Data2:  | Bad   | Bad   |
Parity: | Good  | Good  |

In above case, if we trigger any write into Data1, we will use the bad data in Data2 to re-generate parity, killing the only chance to recovery Data2, thus Data2 is lost forever.

This destructive RMW cycle is not specific to btrfs RAID56, but there are some btrfs specific behaviors making the case even worse:

- Btrfs will cache sectors for unrelated vertical stripes.

In above example, if we're only writing into 0~32K range, btrfs will still read data range (32K ~ 64K) of Data1, and (64K~128K) of Data2. This behavior is to cache sectors for later update.

Incidentally commit d4e28d9b5f04 ("btrfs: raid56: make steal_rbio() subpage compatible") has a bug which makes RAID56 to never trust the cached sectors, thus slightly improve the situation for recovery.

Unfortunately, follow up fix "btrfs: update stripe_sectors::uptodate in steal_rbio" will revert the behavior back to the old one.

- Btrfs raid56 partial write will update all P/Q sectors and cache them

This means, even if data at (64K ~ 96K) of Data2 is free space, and only (96K ~ 128K) of Data2 is really stale data.
And we write into that (96K ~ 128K), we will update all the parity sectors for the full stripe.

This unnecessary behavior will completely kill the chance of recovery.

Thankfully, an unrelated optimization "btrfs: only write the sectors in the vertical stripe which has data stripes" will prevent submitting the write bio for untouched vertical sectors.

That optimization will keep the on-disk P/Q untouched for a chance for later recovery.

[FIX]
Although we have no good way to completely fix the destructive RMW (unless we go full scrub for each partial write), we can still limit the damage.

With patch "btrfs: only write the sectors in the vertical stripe which has data stripes" now we won't really submit the P/Q of unrelated vertical stripes, so the on-disk P/Q should still be fine.

Now we really need to do is just drop all the cached sectors when doing recovery.

By this, we have a chance to read the original P/Q from disk, and have a chance to recover the stale data, while still keep the cache to speed up regular write path.

In fact, just dropping all the cache for recovery path is good enough to allow the test case btrfs/125 along with the small script to pass reliably.

The lack of metadata write after the degraded mount, and forced metadata COW is saving us this time.

So this patch will fix the behavior by not trust any cache in __raid56_parity_recover(), to solve the problem while still keep the cache useful.

But please note that this test pass DOES NOT mean we have solved the destructive RMW problem, we just do better damage control a little better.

Related patches:

- btrfs: only write the sectors in the vertical stripe
- d4e28d9b5f04 ("btrfs: raid56: make steal_rbio() subpage compatible")
- btrfs: update stripe_sectors::uptodate in steal_rbio

---
 fs/btrfs/raid56.c |   19 ++++++-------------
 1 file changed, 6 insertions(+), 13 deletions(-)

@@ -2096,9 +2096,12 @@ static int __raid56_parity_recover(struc
 	atomic_set(&rbio->error, 0);
 
 	/*
-	 * read everything that hasn't failed.  Thanks to the
-	 * stripe cache, it is possible that some or all of these
-	 * pages are going to be uptodate.
+	 * Read everything that hasn't failed. However this time we will
+	 * not trust any cached sector.
+	 * As we may read out some stale data but higher layer is not reading
+	 * that stale part.
+	 *
+	 * So here we always re-read everything in recovery path.
 	 */
 	for (stripe = 0; stripe < rbio->real_stripes; stripe++) {
 		if (rbio->faila == stripe || rbio->failb == stripe) {

@@ -2107,16 +2110,6 @@ static int __raid56_parity_recover(struc
 		}
 
 		for (pagenr = 0; pagenr < rbio->stripe_npages; pagenr++) {
-			struct page *p;
-
-			/*
-			 * the rmw code may have already read this
-			 * page in
-			 */
-			p = rbio_stripe_page(rbio, stripe, pagenr);
-			if (PageUptodate(p))
-				continue;
-
 			ret = rbio_add_io_page(rbio, &bio_list,
 				       rbio_stripe_page(rbio, stripe, pagenr),
 				       stripe, pagenr, rbio->stripe_len);

From: Liang He <[email protected]>

[ Upstream commit 91b3c8dbe898df158fd2a84675f3a284ff6666f7 ]

In this function, there are two refcount leak bugs:
(1) when breaking out of for_each_endpoint_of_node(), we need call the of_node_put() for the 'ep';
(2) we should call of_node_put() for the reference returned by of_graph_get_remote_port() when it is not used anymore.

---
 drivers/gpu/drm/meson/meson_drv.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

index 2d5f2ed3b0b2..0da33f7af654 100644
@@ -136,8 +136,11 @@ static bool meson_vpu_has_available_connectors(struct device *dev)
 	for_each_endpoint_of_node(dev->of_node, ep) {
 		/* If the endpoint node exists, consider it enabled */
 		remote = of_graph_get_remote_port(ep);
-		if (remote)
+		if (remote) {
+			of_node_put(remote);
+			of_node_put(ep);
 			return true;
+		}
 	}
 
 	return false;

2.35.1

From: Pavan Chebbi <[email protected]>

[ Upstream commit afd306a65cedb9589564bdb23a0c368abc4215fd ]

The Broadcom BCM5750x NICs may be multi-function devices. They do not advertise ACS capability. Peer-to-peer transactions are not possible between the individual functions, so it is safe to treat them as fully isolated.

Add an ACS quirk for these devices so the functions can be in independent IOMMU groups and attached individually to userspace applications using VFIO.

---
 drivers/pci/quirks.c | 3 +++
 1 file changed, 3 insertions(+)

index 98327966e5b4..0a116359b5c7 100644
@@ -4760,6 +4760,9 @@ static const struct pci_dev_acs_enabled {
 	{ PCI_VENDOR_ID_AMPERE, 0xE00C, pci_quirk_xgene_acs },
 	/* Broadcom multi-function device */
 	{ PCI_VENDOR_ID_BROADCOM, 0x16D7, pci_quirk_mf_endpoint_acs },
+	{ PCI_VENDOR_ID_BROADCOM, 0x1750, pci_quirk_mf_endpoint_acs },
+	{ PCI_VENDOR_ID_BROADCOM, 0x1751, pci_quirk_mf_endpoint_acs },
+	{ PCI_VENDOR_ID_BROADCOM, 0x1752, pci_quirk_mf_endpoint_acs },
 	{ PCI_VENDOR_ID_BROADCOM, 0xD714, pci_quirk_brcm_acs },
 	{ 0 }
 };

2.35.1

From: Sai Prakash Ranjan <[email protected]>

[ Upstream commit 443685992bda9bb4f8b17fc02c9f6c60e62b1461 ]

Fix -Woverflow warnings for tegra irqchip driver which is a result of moving arm64 custom MMIO accessor macros to asm-generic function implementations giving a bonus type-checking now and uncovering these overflow warnings.

drivers/irqchip/irq-tegra.c: In function ‘tegra_ictlr_suspend’:
drivers/irqchip/irq-tegra.c:151:18: warning: large integer implicitly truncated to unsigned type [-Woverflow]

   writel_relaxed(~0ul, ictlr + ICTLR_COP_IER_CLR);
                  ^

---
 drivers/irqchip/irq-tegra.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

index 0abc0cd1c32e..1b3048ecb600 100644
@@ -157,10 +157,10 @@ static int tegra_ictlr_suspend(void)
 		lic->cop_iep[i] = readl_relaxed(ictlr + ICTLR_COP_IEP_CLASS);
 
 		/* Disable COP interrupts */
-		writel_relaxed(~0ul, ictlr + ICTLR_COP_IER_CLR);
+		writel_relaxed(GENMASK(31, 0), ictlr + ICTLR_COP_IER_CLR);
 
 		/* Disable CPU interrupts */
-		writel_relaxed(~0ul, ictlr + ICTLR_CPU_IER_CLR);
+		writel_relaxed(GENMASK(31, 0), ictlr + ICTLR_CPU_IER_CLR);
 
 		/* Enable the wakeup sources of ictlr */
 		writel_relaxed(lic->ictlr_wake_mask[i], ictlr + ICTLR_CPU_IER_SET);

@@ -181,12 +181,12 @@ static void tegra_ictlr_resume(void)
 
 		writel_relaxed(lic->cpu_iep[i],
 			       ictlr + ICTLR_CPU_IEP_CLASS);
-		writel_relaxed(~0ul, ictlr + ICTLR_CPU_IER_CLR);
+		writel_relaxed(GENMASK(31, 0), ictlr + ICTLR_CPU_IER_CLR);
 		writel_relaxed(lic->cpu_ier[i],
 			       ictlr + ICTLR_CPU_IER_SET);
 		writel_relaxed(lic->cop_iep[i],
 			       ictlr + ICTLR_COP_IEP_CLASS);
-		writel_relaxed(~0ul, ictlr + ICTLR_COP_IER_CLR);
+		writel_relaxed(GENMASK(31, 0), ictlr + ICTLR_COP_IER_CLR);
 		writel_relaxed(lic->cop_ier[i],
 			       ictlr + ICTLR_COP_IER_SET);
 	}

@@ -321,7 +321,7 @@ static int __init tegra_ictlr_init(struct device_node *node,
 		lic->base[i] = base;
 
 		/* Disable all interrupts */
-		writel_relaxed(~0UL, base + ICTLR_CPU_IER_CLR);
+		writel_relaxed(GENMASK(31, 0), base + ICTLR_CPU_IER_CLR);
 		/* All interrupts target IRQ */
 		writel_relaxed(0, base + ICTLR_CPU_IEP_CLASS);
 

2.35.1

From: Liang He <[email protected]>

[ Upstream commit 40a959d7042bb7711e404ad2318b30e9f92c6b9b ]

In ohci_hcd_ppc_of_probe(), of_find_compatible_node() will return a node pointer with refcount incremented. We should use of_node_put() when it is not used anymore.

---
 drivers/usb/host/ohci-ppc-of.c | 1 +
 1 file changed, 1 insertion(+)

index 4f87a5c61b08..d22a70363fbf 100644
@@ -168,6 +168,7 @@ static int ohci_hcd_ppc_of_probe(struct platform_device *op)
 				release_mem_region(res.start, 0x4);
 		} else
 			pr_debug("%s: cannot get ehci offset from fdt\n", __FILE__);
+		of_node_put(np);
 	}
 
 	irq_dispose_mapping(irq);

2.35.1

From: Robert Marko <[email protected]>

[ Upstream commit 1bf7305e79aab095196131bdc87a97796e0e3fac ]

Once the usb sleep clocks are disabled, clock framework is trying to disable the sleep clock source also.

However, it seems that it cannot be disabled and trying to do so produces:

[  245.436390] ------------[ cut here ]------------
[  245.441233] gcc_sleep_clk_src status stuck at 'on'
[  245.441254] WARNING: CPU: 2 PID: 223 at clk_branch_wait+0x130/0x140
[  245.450435] Modules linked in: xhci_plat_hcd xhci_hcd dwc3 dwc3_qcom leds_gpio
[  245.456601] CPU: 2 PID: 223 Comm: sh Not tainted 5.18.0-rc4 #215
[  245.463889] Hardware name: Xiaomi AX9000 (DT)
[  245.470050] pstate: 204000c5 (nzCv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[  245.474307] pc : clk_branch_wait+0x130/0x140
[  245.481073] lr : clk_branch_wait+0x130/0x140
[  245.485588] sp : ffffffc009f2bad0
[  245.489838] x29: ffffffc009f2bad0 x28: ffffff8003e6c800 x27: 0000000000000000
[  245.493057] x26: 0000000000000000 x25: 0000000000000000 x24: ffffff800226ef20
[  245.500175] x23: ffffffc0089ff550 x22: 0000000000000000 x21: ffffffc008476ad0
[  245.507294] x20: 0000000000000000 x19: ffffffc00965ac70 x18: fffffffffffc51a7
[  245.514413] x17: 68702e3030303837 x16: 3a6d726f6674616c x15: ffffffc089f2b777
[  245.521531] x14: ffffffc0095c9d18 x13: 0000000000000129 x12: 0000000000000129
[  245.528649] x11: 00000000ffffffea x10: ffffffc009621d18 x9 : 0000000000000001
[  245.535767] x8 : 0000000000000001 x7 : 0000000000017fe8 x6 : 0000000000000001
[  245.542885] x5 : ffffff803fdca6d8 x4 : 0000000000000000 x3 : 0000000000000027
[  245.550002] x2 : 0000000000000027 x1 : 0000000000000023 x0 : 0000000000000026
[  245.557122] Call trace:
[  245.564229]  clk_branch_wait+0x130/0x140
[  245.566490]  clk_branch2_disable+0x2c/0x40
[  245.570656]  clk_core_disable+0x60/0xb0
[  245.574561]  clk_core_disable+0x68/0xb0
[  245.578293]  clk_disable+0x30/0x50
[  245.582113]  dwc3_qcom_remove+0x60/0xc0 [dwc3_qcom]
[  245.585588]  platform_remove+0x28/0x60
[  245.590361]  device_remove+0x4c/0x80
[  245.594179]  device_release_driver_internal+0x1dc/0x230
[  245.597914]  device_driver_detach+0x18/0x30
[  245.602861]  unbind_store+0xec/0x110
[  245.607027]  drv_attr_store+0x24/0x40
[  245.610847]  sysfs_kf_write+0x44/0x60
[  245.614405]  kernfs_fop_write_iter+0x128/0x1c0
[  245.618052]  new_sync_write+0xc0/0x130
[  245.622391]  vfs_write+0x1d4/0x2a0
[  245.626123]  ksys_write+0x58/0xe0
[  245.629508]  __arm64_sys_write+0x1c/0x30
[  245.632895]  invoke_syscall.constprop.0+0x5c/0x110
[  245.636890]  do_el0_svc+0xa0/0x150
[  245.641488]  el0_svc+0x18/0x60
[  245.644872]  el0t_64_sync_handler+0xa4/0x130
[  245.647914]  el0t_64_sync+0x174/0x178
[  245.652340] ---[ end trace 0000000000000000 ]---

So, add CLK_IS_CRITICAL flag to the clock so that the kernel won't try to disable the sleep clock.

---
 drivers/clk/qcom/gcc-ipq8074.c | 1 +
 1 file changed, 1 insertion(+)

index 0f735d37690f..97a951990103 100644
@@ -139,6 +139,7 @@ static struct clk_branch gcc_sleep_clk_src = {
 			},
 			.num_parents = 1,
 			.ops = &clk_branch2_ops,
+			.flags = CLK_IS_CRITICAL,
 		},
 	},
 };

2.35.1

From: Jozef Martiniak <[email protected]>

[ Upstream commit 04cb742d4d8f30dc2e83b46ac317eec09191c68e ]

after usb_ep_queue() if wait_for_completion_interruptible() is interrupted we need to wait until IRQ gets finished.

Otherwise complete() from epio_complete() can corrupt stack.

---
 drivers/usb/gadget/legacy/inode.c | 1 +
 1 file changed, 1 insertion(+)

index 53d17e1bb65a..f0f41cc8b2f1 100644
@@ -365,6 +365,7 @@ ep_io (struct ep_data *epdata, void *buf, unsigned len)
 				spin_unlock_irq (&epdata->dev->lock);
 
 				DBG (epdata->dev, "endpoint gone\n");
+				wait_for_completion(&done);
 				epdata->status = -ENODEV;
 			}
 		}

2.35.1

From: Christophe JAILLET <[email protected]>

[ Upstream commit 3a15b45b5454da862376b5d69a4967f5c6fa1368 ]

A bitmap_zalloc() must be balanced by a corresponding bitmap_free() in the error handling path of afu_allocate_irqs().

---
 drivers/misc/cxl/irq.c | 1 +
 1 file changed, 1 insertion(+)

index ce08a9f22308..0dbe78383f8f 100644
@@ -353,6 +353,7 @@ int afu_allocate_irqs(struct cxl_context *ctx, u32 count)
 
 out:
 	cxl_ops->release_irq_ranges(&ctx->irqs, ctx->afu->adapter);
+	bitmap_free(ctx->irq_bitmap);
 	afu_irq_name_free(ctx);
 	return -ENOMEM;
 }

2.35.1

From: Wentao_Liang <[email protected]>

[ Upstream commit 104212471b1c1817b311771d817fb692af983173 ]

In line 2884, "raid5_release_stripe(sh);" drops the reference to sh and may cause sh to be released. However, sh is subsequently used in lines 2886 "if (sh->batch_head && sh != sh->batch_head)". This may result in an use-after-free bug.

It can be fixed by moving "raid5_release_stripe(sh);" to the bottom of the function.

---
 drivers/md/raid5.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index 1e52443f3aca..866ba1743f9f 100644
@@ -2668,10 +2668,10 @@ static void raid5_end_write_request(struct bio *bi)
 	if (!test_and_clear_bit(R5_DOUBLE_LOCKED, &sh->dev[i].flags))
 		clear_bit(R5_LOCKED, &sh->dev[i].flags);
 	set_bit(STRIPE_HANDLE, &sh->state);
-	raid5_release_stripe(sh);
 
 	if (sh->batch_head && sh != sh->batch_head)
 		raid5_release_stripe(sh->batch_head);
+	raid5_release_stripe(sh);
 }
 
 static void raid5_error(struct mddev *mddev, struct md_rdev *rdev)

2.35.1

From: Ye Bin <[email protected]>

[ Upstream commit b24e77ef1c6d4dbf42749ad4903c97539cc9755a ]

Now if check directoy entry is corrupted, ext4_empty_dir may return true then directory will be removed when file system mounted with "errors=continue". In order not to make things worse just return false when directory is corrupted.

---
 fs/ext4/namei.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

index 4b78fa4dbcc1..e1cade9dbd30 100644
@@ -2820,11 +2820,8 @@ bool ext4_empty_dir(struct inode *inode)
 		de = (struct ext4_dir_entry_2 *) (bh->b_data +
 					(offset & (sb->s_blocksize - 1)));
 		if (ext4_check_dir_entry(inode, NULL, de, bh,
-					 bh->b_data, bh->b_size, offset)) {
-			offset = (offset | (sb->s_blocksize - 1)) + 1;
-			continue;
-		}
-		if (le32_to_cpu(de->inode)) {
+					 bh->b_data, bh->b_size, offset) ||
+		    le32_to_cpu(de->inode)) {
 			brelse(bh);
 			return false;
 		}

2.35.1

From: Kiselev, Oleg <[email protected]>

[ Upstream commit 69cb8e9d8cd97cdf5e293b26d70a9dee3e35e6bd ]

This patch avoids an attempt to resize the filesystem to an unaligned cluster boundary. An online resize to a size that is not integral to cluster size results in the last iteration attempting to grow the fs by a negative amount, which trips a BUG_ON and leaves the fs with a corrupted in-memory superblock.

---
 fs/ext4/resize.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

index a50eabffa411..30b2798244fa 100644
@@ -1951,6 +1951,16 @@ int ext4_resize_fs(struct super_block *sb, ext4_fsblk_t n_blocks_count)
 	}
 	brelse(bh);
 
+	/*
+	 * For bigalloc, trim the requested size to the nearest cluster
+	 * boundary to avoid creating an unusable filesystem. We do this
+	 * silently, instead of returning an error, to avoid breaking
+	 * callers that blindly resize the filesystem to the full size of
+	 * the underlying block device.
+	 */
+	if (ext4_has_feature_bigalloc(sb))
+		n_blocks_count &= ~((1 << EXT4_CLUSTER_BITS(sb)) - 1);
+
 retry:
 	o_blocks_count = ext4_blocks_count(es);
 

2.35.1

From: Liang He <[email protected]>

[ Upstream commit d24d7bb2cd947676f9b71fb944d045e09b8b282f ]

In soc_info(), of_find_node_by_type() will return a node pointer with refcount incremented. We should use of_node_put() when it is not used anymore.

---
 drivers/tty/serial/ucc_uart.c | 2 ++
 1 file changed, 2 insertions(+)

index 55b702775786..40b8e414f48f 100644
@@ -1143,6 +1143,8 @@ static unsigned int soc_info(unsigned int *rev_h, unsigned int *rev_l)
 		/* No compatible property, so try the name. */
 		soc_string = np->name;
 
+	of_node_put(np);
+
 	/* Extract the SOC number from the "PowerPC," string */
 	if ((sscanf(soc_string, "PowerPC,%u", &soc) != 1) || !soc)
 		return 0;

2.35.1

From: Schspa Shi <[email protected]>

[ Upstream commit 6641085e8d7b3f061911517f79a2a15a0a21b97b ]

On buffer resize failure, vfio_info_cap_add() will free the buffer, report zero for the size, and return -ENOMEM. As additional hardening, also clear the buffer pointer to prevent any chance of a double free.

---
 drivers/vfio/vfio.c | 1 +
 1 file changed, 1 insertion(+)

index 0d73d913c18b..747eb5c70238 100644
@@ -1813,6 +1813,7 @@ struct vfio_info_cap_header *vfio_info_cap_add(struct vfio_info_cap *caps,
 	buf = krealloc(caps->buf, caps->size + size, GFP_KERNEL);
 	if (!buf) {
 		kfree(caps->buf);
+		caps->buf = NULL;
 		caps->size = 0;
 		return ERR_PTR(-ENOMEM);
 	}

2.35.1

From: Liang He <[email protected]>

[ Upstream commit 7a9f743ceead60ed454c46fbc3085ee9a79cbebb ]

We should call of_node_put() for the reference 'uctl_node' returned by of_get_parent() which will increase the refcount. Otherwise, there will be a refcount leak bug.

---
 arch/mips/cavium-octeon/octeon-platform.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

index ed42fc27033c..f2a1c316608a 100644
@@ -86,11 +86,12 @@ static void octeon2_usb_clocks_start(struct device *dev)
 					 "refclk-frequency", &clock_rate);
 		if (i) {
 			dev_err(dev, "No UCTL \"refclk-frequency\"\n");
+			of_node_put(uctl_node);
 			goto exit;
 		}
 		i = of_property_read_string(uctl_node,
 					    "refclk-type", &clock_type);
-
+		of_node_put(uctl_node);
 		if (!i && strcmp("crystal", clock_type) == 0)
 			is_crystal_clock = true;
 	}

2.35.1

From: Takashi Iwai <[email protected]>

[ Upstream commit ef34a0ae7a2654bc9e58675e36898217fb2799d8 ]

Currently the call of kill_fasync() from an interrupt handler might lead to potential spin deadlocks, as spotted by syzkaller. Unfortunately, it's not so trivial to fix this lock chain as it's involved with the tasklist_lock that is touched in allover places.

As a temporary workaround, this patch provides the way to defer the async signal notification in a work. The new helper functions, snd_fasync_helper() and snd_kill_faync() are replacements for fasync_helper() and kill_fasync(), respectively. In addition, snd_fasync_free() needs to be called at the destructor of the relevant file object.

---
 include/sound/core.h |  8 ++++
 sound/core/misc.c    | 94 ++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 102 insertions(+)

index 4104a9d1001f..9d04e700b855 100644
@@ -442,4 +442,12 @@ snd_pci_quirk_lookup_id(u16 vendor, u16 device,
 }
 #endif
 
+/* async signal helpers */
+struct snd_fasync;
+
+int snd_fasync_helper(int fd, struct file *file, int on,
+		      struct snd_fasync **fasyncp);
+void snd_kill_fasync(struct snd_fasync *fasync, int signal, int poll);
+void snd_fasync_free(struct snd_fasync *fasync);
+
 #endif /* __SOUND_CORE_H */

2.35.1

From: Takashi Iwai <[email protected]>

[ Upstream commit 95cc637c1afd83fb7dd3d7c8a53710488f4caf9c ]

For avoiding the potential deadlock via kill_fasync() call, use the new fasync helpers to defer the invocation from PCI API. Note that it's merely a workaround.

---
 sound/core/timer.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

index c333ceb80d5f..720d63574bfb 100644
@@ -75,7 +75,7 @@ struct snd_timer_user {
 	unsigned int filter;
 	struct timespec tstamp;		/* trigger tstamp */
 	wait_queue_head_t qchange_sleep;
-	struct fasync_struct *fasync;
+	struct snd_fasync *fasync;
 	struct mutex ioctl_lock;
 };
 

@@ -1295,7 +1295,7 @@ static void snd_timer_user_interrupt(struct snd_timer_instance *timeri,
 	}
       __wake:
 	spin_unlock(&tu->qlock);
-	kill_fasync(&tu->fasync, SIGIO, POLL_IN);
+	snd_kill_fasync(tu->fasync, SIGIO, POLL_IN);
 	wake_up(&tu->qchange_sleep);
 }
 

@@ -1332,7 +1332,7 @@ static void snd_timer_user_ccallback(struct snd_timer_instance *timeri,
 	spin_lock_irqsave(&tu->qlock, flags);
 	snd_timer_user_append_to_tqueue(tu, &r1);
 	spin_unlock_irqrestore(&tu->qlock, flags);
-	kill_fasync(&tu->fasync, SIGIO, POLL_IN);
+	snd_kill_fasync(tu->fasync, SIGIO, POLL_IN);
 	wake_up(&tu->qchange_sleep);
 }
 

@@ -1399,7 +1399,7 @@ static void snd_timer_user_tinterrupt(struct snd_timer_instance *timeri,
 	spin_unlock(&tu->qlock);
 	if (append == 0)
 		return;
-	kill_fasync(&tu->fasync, SIGIO, POLL_IN);
+	snd_kill_fasync(tu->fasync, SIGIO, POLL_IN);
 	wake_up(&tu->qchange_sleep);
 }
 

@@ -1465,6 +1465,7 @@ static int snd_timer_user_release(struct inode *inode, struct file *file)
 		if (tu->timeri)
 			snd_timer_close(tu->timeri);
 		mutex_unlock(&tu->ioctl_lock);
+		snd_fasync_free(tu->fasync);
 		kfree(tu->queue);
 		kfree(tu->tqueue);
 		kfree(tu);

@@ -2017,7 +2018,7 @@ static int snd_timer_user_fasync(int fd, struct file * file, int on)
 	struct snd_timer_user *tu;
 
 	tu = file->private_data;
-	return fasync_helper(fd, file, on, &tu->fasync);
+	return snd_fasync_helper(fd, file, on, &tu->fasync);
 }
 
 static ssize_t snd_timer_user_read(struct file *file, char __user *buffer,

2.35.1

From: Steve French <[email protected]>

[ Upstream commit 5fa2cffba0b82336a2244d941322eb1627ff787b ]

Coverity complains about assigning a pointer based on value length before checking that value length goes beyond the end of the SMB. Although this is even more unlikely as value length is a single byte, and the pointer is not dereferenced until laterm, it is clearer to check the lengths first.

Addresses-Coverity: 1467704 ("Speculative execution data leak")

---
 fs/cifs/smb2ops.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

index 3280a801b1d7..069eb2533e7f 100644
@@ -463,9 +463,7 @@ move_smb2_ea_to_cifs(char *dst, size_t dst_size,
 	size_t name_len, value_len, user_name_len;
 
 	while (src_size > 0) {
-		name = &src->ea_data[0];
 		name_len = (size_t)src->ea_name_length;
-		value = &src->ea_data[src->ea_name_length + 1];
 		value_len = (size_t)le16_to_cpu(src->ea_value_length);
 
 		if (name_len == 0) {

@@ -478,6 +476,9 @@ move_smb2_ea_to_cifs(char *dst, size_t dst_size,
 			goto out;
 		}
 
+		name = &src->ea_data[0];
+		value = &src->ea_data[src->ea_name_length + 1];
+
 		if (ea_name) {
 			if (ea_name_len == name_len &&
 			    memcmp(ea_name, name, name_len) == 0) {

2.35.1

From: Zhouyi Zhou <[email protected]>

[ Upstream commit ca829e05d3d4f728810cc5e4b468d9ebc7745eb3 ]

On 64-bit, calling jump_label_init() in setup_feature_keys() is too late because static keys may be used in subroutines of parse_early_param() which is again subroutine of early_init_devtree().

For example booting with "threadirqs":

static_key_enable_cpuslocked(): static key '0xc000000002953260' used before call to jump_label_init()

  WARNING: CPU: 0 PID: 0 at kernel/jump_label.c:166 static_key_enable_cpuslocked+0xfc/0x120
  ...
  NIP static_key_enable_cpuslocked+0xfc/0x120
  LR  static_key_enable_cpuslocked+0xf8/0x120
  Call Trace:
    static_key_enable_cpuslocked+0xf8/0x120 (unreliable)
    static_key_enable+0x30/0x50
    setup_forced_irqthreads+0x28/0x40
    do_early_param+0xa0/0x108
    parse_args+0x290/0x4e0
    parse_early_options+0x48/0x5c
    parse_early_param+0x58/0x84
    early_init_devtree+0xd4/0x518
    early_setup+0xb4/0x214

So call jump_label_init() just before parse_early_param() in early_init_devtree().

---
 arch/powerpc/kernel/prom.c | 7 +++++++
 1 file changed, 7 insertions(+)

index 99409d0420d2..ffc571edac28 100644
@@ -710,6 +710,13 @@ void __init early_init_devtree(void *params)
 	of_scan_flat_dt(early_init_dt_scan_root, NULL);
 	of_scan_flat_dt(early_init_dt_scan_memory_ppc, NULL);
 
+	/*
+	 * As generic code authors expect to be able to use static keys
+	 * in early_param() handlers, we initialize the static keys just
+	 * before parsing early params (it's fine to call jump_label_init()
+	 * more than once).
+	 */
+	jump_label_init();
 	parse_early_param();
 
 	/* make sure we've parsed cmdline for mem= before this */

2.35.1

From: Zheyu Ma <[email protected]>

[ Upstream commit 40bf722f8064f50200b8c4f8946cd625b441dda9 ]

Since the user can control the arguments of the ioctl() from the user space, under special arguments that may result in a divide-by-zero bug.

If the user provides an improper 'pixclock' value that makes the argumet of i740_calc_vclk() less than 'I740_RFREQ_FIX', it will cause a divide-by-zero bug in:

    drivers/video/fbdev/i740fb.c:353 p_best = min(15, ilog2(I740_MAX_VCO_FREQ / (freq / I740_RFREQ_FIX)));

The following log can reveal it:

divide error: 0000 [#1] PREEMPT SMP KASAN PTI

RIP: 0010:i740_calc_vclk drivers/video/fbdev/i740fb.c:353 [inline]
RIP: 0010:i740fb_decode_var drivers/video/fbdev/i740fb.c:646 [inline]
RIP: 0010:i740fb_set_par+0x163f/0x3b70 drivers/video/fbdev/i740fb.c:742
Call Trace:
 fb_set_var+0x604/0xeb0 drivers/video/fbdev/core/fbmem.c:1034
 do_fb_ioctl+0x234/0x670 drivers/video/fbdev/core/fbmem.c:1110
 fb_ioctl+0xdd/0x130 drivers/video/fbdev/core/fbmem.c:1189

Fix this by checking the argument of i740_calc_vclk() first.

---
 drivers/video/fbdev/i740fb.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

index 7bc5f6056c77..4147a9534179 100644
@@ -399,7 +399,7 @@ static int i740fb_decode_var(const struct fb_var_screeninfo *var,
 	u32 xres, right, hslen, left, xtotal;
 	u32 yres, lower, vslen, upper, ytotal;
 	u32 vxres, xoffset, vyres, yoffset;
-	u32 bpp, base, dacspeed24, mem;
+	u32 bpp, base, dacspeed24, mem, freq;
 	u8 r7;
 	int i;
 

@@ -641,7 +641,12 @@ static int i740fb_decode_var(const struct fb_var_screeninfo *var,
 	par->atc[VGA_ATC_OVERSCAN] = 0;
 
 	/* Calculate VCLK that most closely matches the requested dot clock */
-	i740_calc_vclk((((u32)1e9) / var->pixclock) * (u32)(1e3), par);
+	freq = (((u32)1e9) / var->pixclock) * (u32)(1e3);
+	if (freq < I740_RFREQ_FIX) {
+		fb_dbg(info, "invalid pixclock\n");
+		freq = I740_RFREQ_FIX;
+	}
+	i740_calc_vclk(freq, par);
 
 	/* Since we program the clocks ourselves, always use VCLK2. */
 	par->misc |= 0x0C;

2.35.1

From: Nathan Chancellor <[email protected]>

[ Upstream commit 74de14fe05dd6b151d73cb0c73c8ec874cbdcde6 ]

When CONFIG_XPA is enabled, Clang warns:

  arch/mips/mm/tlbex.c:629:24: error: converting the result of '<<' to a boolean; did you mean '(1 << _PAGE_NO_EXEC_SHIFT) != 0'? [-Werror,-Wint-in-bool-context]
          if (cpu_has_rixi && !!_PAGE_NO_EXEC) {
                              ^
  arch/mips/include/asm/pgtable-bits.h:174:28: note: expanded from macro '_PAGE_NO_EXEC'
  # define _PAGE_NO_EXEC          (1 << _PAGE_NO_EXEC_SHIFT)
                                     ^
  arch/mips/mm/tlbex.c:2568:24: error: converting the result of '<<' to a boolean; did you mean '(1 << _PAGE_NO_EXEC_SHIFT) != 0'? [-Werror,-Wint-in-bool-context]
          if (!cpu_has_rixi || !_PAGE_NO_EXEC) {
                                ^
  arch/mips/include/asm/pgtable-bits.h:174:28: note: expanded from macro '_PAGE_NO_EXEC'
  # define _PAGE_NO_EXEC          (1 << _PAGE_NO_EXEC_SHIFT)
                                     ^
  2 errors generated.

_PAGE_NO_EXEC can be '0' or '1 << _PAGE_NO_EXEC_SHIFT' depending on the build and runtime configuration, which is what the negation operators are trying to convey. To silence the warning, explicitly compare against 0 so the result of the '<<' operator is not implicitly converted to a boolean.

According to its documentation, GCC enables -Wint-in-bool-context with -Wall but this warning is not visible when building the same configuration with GCC. It appears GCC only warns when compiling C++, not C, although the documentation makes no note of this:
https://godbolt.org/z/x39q3brxf

---
 arch/mips/mm/tlbex.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

index b55c74a7f7a4..82cd14e7b20d 100644
@@ -634,7 +634,7 @@ static __maybe_unused void build_convert_pte_to_entrylo(u32 **p,
 		return;
 	}
 
-	if (cpu_has_rixi && !!_PAGE_NO_EXEC) {
+	if (cpu_has_rixi && _PAGE_NO_EXEC != 0) {
 		if (fill_includes_sw_bits) {
 			UASM_i_ROTR(p, reg, reg, ilog2(_PAGE_GLOBAL));
 		} else {

@@ -2577,7 +2577,7 @@ static void check_pabits(void)
 	unsigned long entry;
 	unsigned pabits, fillbits;
 
-	if (!cpu_has_rixi || !_PAGE_NO_EXEC) {
+	if (!cpu_has_rixi || _PAGE_NO_EXEC == 0) {
 		/*
 		 * We'll only be making use of the fact that we can rotate bits
 		 * into the fill if the CPU supports RIXI, so don't bother

2.35.1

Build results:

	total: 170 pass: 170 fail: 0
Qemu test results:
	total: 421 pass: 421 fail: 0

Guenter

Results from Linaro's test farm.
No regressions on arm64, arm, x86_64, and i386.

## Build
* kernel: 4.14.291-rc1
* git: https://gitlab.com/Linaro/lkft/mirrors/stable/linux-stable-rc
* git branch: linux-4.14.y
* git commit: 4d54ef55c38e2d37b9c16b20b9d1b3febf73a2e2
* git describe: v4.14.290-230-g4d54ef55c38e
* test details:
https://qa-reports.linaro.org/lkft/linux-stable-rc-linux-4.14.y/build/v4.14.290-230-g4d54ef55c38e

## No test Regressions (compared to v4.14.290)

## No metric Regressions (compared to v4.14.290)

## No test Fixes (compared to v4.14.290)

## No metric Fixes (compared to v4.14.290)

## Test result summary
total: 85673, pass: 77996, fail: 641, skip: 6721, xfail: 315

## Build Summary
* arc: 20 total, 20 passed, 0 failed
* arm: 561 total, 551 passed, 10 failed
* arm64: 100 total, 94 passed, 6 failed
* i386: 52 total, 50 passed, 2 failed
* mips: 60 total, 60 passed, 0 failed
* parisc: 24 total, 24 passed, 0 failed
* powerpc: 32 total, 32 passed, 0 failed
* s390: 24 total, 18 passed, 6 failed
* sh: 48 total, 48 passed, 0 failed
* sparc: 24 total, 24 passed, 0 failed
* x86_64: 96 total, 94 passed, 2 failed

## Test suites summary
* fwts
* kunit
* libhugetlbfs
* log-parser-boot
* log-parser-test
* ltp-cap_bounds
* ltp-commands
* ltp-containers
* ltp-controllers
* ltp-cpuhotplug
* ltp-crypto
* ltp-cve
* ltp-dio
* ltp-fcntl-locktests
* ltp-filecaps
* ltp-fs
* ltp-fs_bind
* ltp-fs_perms_simple
* ltp-fsx
* ltp-hugetlb
* ltp-io
* ltp-ipc
* ltp-math
* ltp-mm
* ltp-nptl
* ltp-open-posix-tests
* ltp-pty
* ltp-sched
* ltp-securebits
* ltp-smoke
* ltp-syscalls
* ltp-tracing
* network-basic-tests
* packetdrill
* rcutorture
* v4l2-compliance
* vdso

--
Linaro LKFT
https://lkft.linaro.org