6.10.14-rc1 review

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ajit Pandey <[email protected]>

commit fff617979f97c773aaa9432c31cf62444b3bdbd4 upstream.

In LUCID EVO PLL CAL_L_VAL and L_VAL bitfields are part of single PLL_L_VAL register. Update for L_VAL bitfield values in PLL_L_VAL register using regmap_write() API in __alpha_pll_trion_set_rate callback will override LUCID EVO PLL initial configuration related to PLL_CAL_L_VAL bit fields in PLL_L_VAL register.

Observed random PLL lock failures during PLL enable due to such override in PLL calibration value. Use regmap_update_bits() with L_VAL bitfield mask instead of regmap_write() API to update only PLL_L_VAL bitfields in __alpha_pll_trion_set_rate callback.

---
 drivers/clk/qcom/clk-alpha-pll.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -1665,7 +1665,7 @@ static int __alpha_pll_trion_set_rate(st
 	if (ret < 0)
 		return ret;
 
-	regmap_write(pll->clkr.regmap, PLL_L_VAL(pll), l);
+	regmap_update_bits(pll->clkr.regmap, PLL_L_VAL(pll), LUCID_EVO_PLL_L_VAL_MASK,  l);
 	regmap_write(pll->clkr.regmap, PLL_ALPHA_VAL(pll), a);
 
 	/* Latch the PLL input */

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: wangrong <[email protected]>

commit a421e3fe0e6abe27395078f4f0cec5daf466caea upstream.

Due to server permission control, the client does not have access to the shared root directory, but can access subdirectories normally, so users usually mount the shared subdirectories directly. In this case, queryfs should use the actual path instead of the root directory to avoid the call returning an error (EACCES).

---
 fs/smb/client/cifsfs.c   |   13 ++++++++++++-
 fs/smb/client/cifsglob.h |    2 +-
 fs/smb/client/smb1ops.c  |    2 +-
 fs/smb/client/smb2ops.c  |   19 ++++++++++++-------
 4 files changed, 26 insertions(+), 10 deletions(-)

@@ -313,8 +313,17 @@ cifs_statfs(struct dentry *dentry, struc
 	struct TCP_Server_Info *server = tcon->ses->server;
 	unsigned int xid;
 	int rc = 0;
+	const char *full_path;
+	void *page;
 
 	xid = get_xid();
+	page = alloc_dentry_path();
+
+	full_path = build_path_from_dentry(dentry, page);
+	if (IS_ERR(full_path)) {
+		rc = PTR_ERR(full_path);
+		goto statfs_out;
+	}
 
 	if (le32_to_cpu(tcon->fsAttrInfo.MaxPathNameComponentLength) > 0)
 		buf->f_namelen =

@@ -330,8 +339,10 @@ cifs_statfs(struct dentry *dentry, struc
 	buf->f_ffree = 0;	/* unlimited */
 
 	if (server->ops->queryfs)
-		rc = server->ops->queryfs(xid, tcon, cifs_sb, buf);
+		rc = server->ops->queryfs(xid, tcon, full_path, cifs_sb, buf);
 
+statfs_out:
+	free_dentry_path(page);
 	free_xid(xid);
 	return rc;
 }

@@ -482,7 +482,7 @@ struct smb_version_operations {
 			__u16 net_fid, struct cifsInodeInfo *cifs_inode);
 	/* query remote filesystem */
 	int (*queryfs)(const unsigned int, struct cifs_tcon *,
-		       struct cifs_sb_info *, struct kstatfs *);
+		       const char *, struct cifs_sb_info *, struct kstatfs *);
 	/* send mandatory brlock to the server */
 	int (*mand_lock)(const unsigned int, struct cifsFileInfo *, __u64,
 			 __u64, __u32, int, int, bool);

@@ -909,7 +909,7 @@ cifs_oplock_response(struct cifs_tcon *t
 
 static int
 cifs_queryfs(const unsigned int xid, struct cifs_tcon *tcon,
-	     struct cifs_sb_info *cifs_sb, struct kstatfs *buf)
+	     const char *path, struct cifs_sb_info *cifs_sb, struct kstatfs *buf)
 {
 	int rc = -EOPNOTSUPP;
 

@@ -2816,7 +2816,7 @@ out_free_path:
 
 static int
 smb2_queryfs(const unsigned int xid, struct cifs_tcon *tcon,
-	     struct cifs_sb_info *cifs_sb, struct kstatfs *buf)
+	     const char *path, struct cifs_sb_info *cifs_sb, struct kstatfs *buf)
 {
 	struct smb2_query_info_rsp *rsp;
 	struct smb2_fs_full_size_info *info = NULL;

@@ -2825,7 +2825,7 @@ smb2_queryfs(const unsigned int xid, str
 	int rc;
 
 
-	rc = smb2_query_info_compound(xid, tcon, "",
+	rc = smb2_query_info_compound(xid, tcon, path,
 				      FILE_READ_ATTRIBUTES,
 				      FS_FULL_SIZE_INFORMATION,
 				      SMB2_O_INFO_FILESYSTEM,

@@ -2853,28 +2853,33 @@ qfs_exit:
 
 static int
 smb311_queryfs(const unsigned int xid, struct cifs_tcon *tcon,
-	       struct cifs_sb_info *cifs_sb, struct kstatfs *buf)
+	       const char *path, struct cifs_sb_info *cifs_sb, struct kstatfs *buf)
 {
 	int rc;
-	__le16 srch_path = 0; /* Null - open root of share */
+	__le16 *utf16_path = NULL;
 	u8 oplock = SMB2_OPLOCK_LEVEL_NONE;
 	struct cifs_open_parms oparms;
 	struct cifs_fid fid;
 
 	if (!tcon->posix_extensions)
-		return smb2_queryfs(xid, tcon, cifs_sb, buf);
+		return smb2_queryfs(xid, tcon, path, cifs_sb, buf);
 
 	oparms = (struct cifs_open_parms) {
 		.tcon = tcon,
-		.path = "",
+		.path = path,
 		.desired_access = FILE_READ_ATTRIBUTES,
 		.disposition = FILE_OPEN,
 		.create_options = cifs_create_options(cifs_sb, 0),
 		.fid = &fid,
 	};
 
-	rc = SMB2_open(xid, &oparms, &srch_path, &oplock, NULL, NULL,
+	utf16_path = cifs_convert_path_to_utf16(path, cifs_sb);
+	if (utf16_path == NULL)
+		return -ENOMEM;
+
+	rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL,
 		       NULL, NULL);
+	kfree(utf16_path);
 	if (rc)
 		return rc;
 

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Steve French <[email protected]>

commit 2f3017e7cc7515e0110a3733d8dca84de2a1d23d upstream.

Commands like "chmod 0444" mark a file readonly via the attribute flag (when mapping of mode bits into the ACL are not set, or POSIX extensions are not negotiated), but they were not reported correctly for stat of directories (they were reported ok for files and for "ls"). See example below:

    root:~# ls /mnt2 -l
    total 12
    drwxr-xr-x 2 root root         0 Sep 21 18:03 normaldir
    -rwxr-xr-x 1 root root         0 Sep 21 23:24 normalfile
    dr-xr-xr-x 2 root root         0 Sep 21 17:55 readonly-dir
    -r-xr-xr-x 1 root root 209716224 Sep 21 18:15 readonly-file
    root:~# stat -c %a /mnt2/readonly-dir
    755
    root:~# stat -c %a /mnt2/readonly-file
    555

This fixes the stat of directories when ATTR_READONLY is set (in cases where the mode can not be obtained other ways).

    root:~# stat -c %a /mnt2/readonly-dir
    555

---
 fs/smb/client/inode.c |   19 +++++++++++--------
 1 file changed, 11 insertions(+), 8 deletions(-)

@@ -800,10 +800,6 @@ static void cifs_open_info_to_fattr(stru
 		fattr->cf_mode = S_IFREG | cifs_sb->ctx->file_mode;
 		fattr->cf_dtype = DT_REG;
 
-		/* clear write bits if ATTR_READONLY is set */
-		if (fattr->cf_cifsattrs & ATTR_READONLY)
-			fattr->cf_mode &= ~(S_IWUGO);
-
 		/*
 		 * Don't accept zero nlink from non-unix servers unless
 		 * delete is pending.  Instead mark it as unknown.

@@ -816,6 +812,10 @@ static void cifs_open_info_to_fattr(stru
 		}
 	}
 
+	/* clear write bits if ATTR_READONLY is set */
+	if (fattr->cf_cifsattrs & ATTR_READONLY)
+		fattr->cf_mode &= ~(S_IWUGO);
+
 out_reparse:
 	if (S_ISLNK(fattr->cf_mode)) {
 		if (likely(data->symlink_target))

@@ -1233,11 +1233,14 @@ handle_mnt_opt:
 				 __func__, rc);
 			goto out;
 		}
-	}
-
-	/* fill in remaining high mode bits e.g. SUID, VTX */
-	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL)
+	} else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL)
+		/* fill in remaining high mode bits e.g. SUID, VTX */
 		cifs_sfu_mode(fattr, full_path, cifs_sb, xid);
+	else if (!(tcon->posix_extensions))
+		/* clear write bits if ATTR_READONLY is set */
+		if (fattr->cf_cifsattrs & ATTR_READONLY)
+			fattr->cf_mode &= ~(S_IWUGO);
+
 
 	/* check for Minshall+French symlinks */
 	if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MF_SYMLINKS) {

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Barnabás Czémán <[email protected]>

commit 129464e86c7445a858b790ac2d28d35f58256bbe upstream.

Move ST2 reading with overflow handling after measurement data reading.
ST2 register read have to be read after read measurment data, because it means end of the reading and realease the lock on the data. Remove ST2 read skip on interrupt based waiting because ST2 required to be read out at and of the axis read.

---
 drivers/iio/magnetometer/ak8975.c |   32 ++++++++++++++++----------------
 1 file changed, 16 insertions(+), 16 deletions(-)

@@ -692,22 +692,8 @@ static int ak8975_start_read_axis(struct
 	if (ret < 0)
 		return ret;
 
-	/* This will be executed only for non-interrupt based waiting case */
-	if (ret & data->def->ctrl_masks[ST1_DRDY]) {
-		ret = i2c_smbus_read_byte_data(client,
-					       data->def->ctrl_regs[ST2]);
-		if (ret < 0) {
-			dev_err(&client->dev, "Error in reading ST2\n");
-			return ret;
-		}
-		if (ret & (data->def->ctrl_masks[ST2_DERR] |
-			   data->def->ctrl_masks[ST2_HOFL])) {
-			dev_err(&client->dev, "ST2 status error 0x%x\n", ret);
-			return -EINVAL;
-		}
-	}
-
-	return 0;
+	/* Return with zero if the data is ready. */
+	return !data->def->ctrl_regs[ST1_DRDY];
 }
 
 /* Retrieve raw flux value for one of the x, y, or z axis.  */

@@ -734,6 +720,20 @@ static int ak8975_read_axis(struct iio_d
 	if (ret < 0)
 		goto exit;
 
+	/* Read out ST2 for release lock on measurment data. */
+	ret = i2c_smbus_read_byte_data(client, data->def->ctrl_regs[ST2]);
+	if (ret < 0) {
+		dev_err(&client->dev, "Error in reading ST2\n");
+		goto exit;
+	}
+
+	if (ret & (data->def->ctrl_masks[ST2_DERR] |
+		   data->def->ctrl_masks[ST2_HOFL])) {
+		dev_err(&client->dev, "ST2 status error 0x%x\n", ret);
+		ret = -EINVAL;
+		goto exit;
+	}
+
 	mutex_unlock(&data->lock);
 
 	pm_runtime_mark_last_busy(&data->client->dev);

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Willem de Bruijn <[email protected]>

commit b04c4d9eb4f25b950b33218e33b04c94e7445e51 upstream.

This reverts commit 504fc6f4f7f681d2a03aa5f68aad549d90eab853.

dev_queue_xmit_nit is expected to be called with BH disabled. __dev_queue_xmit has the following:

        /* Disable soft irqs for various locks below. Also
         * stops preemption for RCU.
         */
        rcu_read_lock_bh();

VRF must follow this invariant. The referenced commit removed this protection. Which triggered a lockdep warning:

	================================
	WARNING: inconsistent lock state
	6.11.0 #1 Tainted: G        W
	--------------------------------
	inconsistent {IN-SOFTIRQ-W} -> {SOFTIRQ-ON-W} usage.
	btserver/134819 [HC0[0]:SC0[0]:HE1:SE1] takes:
	ffff8882da30c118 (rlock-AF_PACKET){+.?.}-{2:2}, at: tpacket_rcv+0x863/0x3b30
	{IN-SOFTIRQ-W} state was registered at:
	  lock_acquire+0x19a/0x4f0
	  _raw_spin_lock+0x27/0x40
	  packet_rcv+0xa33/0x1320
	  __netif_receive_skb_core.constprop.0+0xcb0/0x3a90
	  __netif_receive_skb_list_core+0x2c9/0x890
	  netif_receive_skb_list_internal+0x610/0xcc0
          [...]

	other info that might help us debug this:
	 Possible unsafe locking scenario:

	       CPU0
	       ----
	  lock(rlock-AF_PACKET);
	  <Interrupt>
	    lock(rlock-AF_PACKET);

	 *** DEADLOCK ***

	Call Trace:
	 <TASK>
	 dump_stack_lvl+0x73/0xa0
	 mark_lock+0x102e/0x16b0
	 __lock_acquire+0x9ae/0x6170
	 lock_acquire+0x19a/0x4f0
	 _raw_spin_lock+0x27/0x40
	 tpacket_rcv+0x863/0x3b30
	 dev_queue_xmit_nit+0x709/0xa40
	 vrf_finish_direct+0x26e/0x340 [vrf]
	 vrf_l3_out+0x5f4/0xe80 [vrf]
	 __ip_local_out+0x51e/0x7a0
          [...]

---
 drivers/net/vrf.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -628,7 +628,9 @@ static void vrf_finish_direct(struct sk_
 		eth_zero_addr(eth->h_dest);
 		eth->h_proto = skb->protocol;
 
+		rcu_read_lock_bh();
 		dev_queue_xmit_nit(skb, vrf_dev);
+		rcu_read_unlock_bh();
 
 		skb_pull(skb, ETH_HLEN);
 	}

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Felix Fietkau <[email protected]>

commit 17bd3bd82f9f79f3feba15476c2b2c95a9b11ff8 upstream.

Detect tcp gso fraglist skbs with corrupted geometry (see below) and pass these to skb_segment instead of skb_segment_list, as the first can segment them correctly.

Valid SKB_GSO_FRAGLIST skbs
- consist of two or more segments
- the head_skb holds the protocol headers plus first gso_size
- one or more frag_list skbs hold exactly one segment
- all but the last must be gso_size

Optional datapath hooks such as NAT and BPF (bpf_skb_pull_data) can modify these skbs, breaking these invariants.

In extreme cases they pull all data into skb linear. For TCP, this causes a NULL ptr deref in __tcpv4_gso_segment_list_csum at tcp_hdr(seg->next).

Detect invalid geometry due to pull, by checking head_skb size. Don't just drop, as this may blackhole a destination. Convert to be able to pass to regular skb_segment.

Approach and description based on a patch by Willem de Bruijn.

---
 net/ipv4/tcp_offload.c   | 10 ++++++++--
 net/ipv6/tcpv6_offload.c | 10 ++++++++--
 2 files changed, 16 insertions(+), 4 deletions(-)

index e4ad3311e148..2308665b51c5 100644
@@ -101,8 +101,14 @@ static struct sk_buff *tcp4_gso_segment(struct sk_buff *skb,
 	if (!pskb_may_pull(skb, sizeof(struct tcphdr)))
 		return ERR_PTR(-EINVAL);
 
-	if (skb_shinfo(skb)->gso_type & SKB_GSO_FRAGLIST)
-		return __tcp4_gso_segment_list(skb, features);
+	if (skb_shinfo(skb)->gso_type & SKB_GSO_FRAGLIST) {
+		struct tcphdr *th = tcp_hdr(skb);
+
+		if (skb_pagelen(skb) - th->doff * 4 == skb_shinfo(skb)->gso_size)
+			return __tcp4_gso_segment_list(skb, features);
+
+		skb->ip_summed = CHECKSUM_NONE;
+	}
 
 	if (unlikely(skb->ip_summed != CHECKSUM_PARTIAL)) {
 		const struct iphdr *iph = ip_hdr(skb);

index 23971903e66d..a45bf17cb2a1 100644
@@ -159,8 +159,14 @@ static struct sk_buff *tcp6_gso_segment(struct sk_buff *skb,
 	if (!pskb_may_pull(skb, sizeof(*th)))
 		return ERR_PTR(-EINVAL);
 
-	if (skb_shinfo(skb)->gso_type & SKB_GSO_FRAGLIST)
-		return __tcp6_gso_segment_list(skb, features);
+	if (skb_shinfo(skb)->gso_type & SKB_GSO_FRAGLIST) {
+		struct tcphdr *th = tcp_hdr(skb);
+
+		if (skb_pagelen(skb) - th->doff * 4 == skb_shinfo(skb)->gso_size)
+			return __tcp6_gso_segment_list(skb, features);
+
+		skb->ip_summed = CHECKSUM_NONE;
+	}
 
 	if (unlikely(skb->ip_summed != CHECKSUM_PARTIAL)) {
 		const struct ipv6hdr *ipv6h = ipv6_hdr(skb);

2.46.2

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Willem de Bruijn <[email protected]>

commit a1e40ac5b5e9077fe1f7ae0eb88034db0f9ae1ab upstream.

Detect gso fraglist skbs with corrupted geometry (see below) and pass these to skb_segment instead of skb_segment_list, as the first can segment them correctly.

Valid SKB_GSO_FRAGLIST skbs
- consist of two or more segments
- the head_skb holds the protocol headers plus first gso_size
- one or more frag_list skbs hold exactly one segment
- all but the last must be gso_size

Optional datapath hooks such as NAT and BPF (bpf_skb_pull_data) can modify these skbs, breaking these invariants.

In extreme cases they pull all data into skb linear. For UDP, this causes a NULL ptr deref in __udpv4_gso_segment_list_csum at udp_hdr(seg->next)->dest.

Detect invalid geometry due to pull, by checking head_skb size. Don't just drop, as this may blackhole a destination. Convert to be able to pass to regular skb_segment.

---
 net/ipv4/udp_offload.c |   22 ++++++++++++++++++++--
 1 file changed, 20 insertions(+), 2 deletions(-)

@@ -290,8 +290,26 @@ struct sk_buff *__udp_gso_segment(struct
 		return NULL;
 	}
 
-	if (skb_shinfo(gso_skb)->gso_type & SKB_GSO_FRAGLIST)
-		return __udp_gso_segment_list(gso_skb, features, is_ipv6);
+	if (skb_shinfo(gso_skb)->gso_type & SKB_GSO_FRAGLIST) {
+		 /* Detect modified geometry and pass those to skb_segment. */
+		if (skb_pagelen(gso_skb) - sizeof(*uh) == skb_shinfo(gso_skb)->gso_size)
+			return __udp_gso_segment_list(gso_skb, features, is_ipv6);
+
+		 /* Setup csum, as fraglist skips this in udp4_gro_receive. */
+		gso_skb->csum_start = skb_transport_header(gso_skb) - gso_skb->head;
+		gso_skb->csum_offset = offsetof(struct udphdr, check);
+		gso_skb->ip_summed = CHECKSUM_PARTIAL;
+
+		uh = udp_hdr(gso_skb);
+		if (is_ipv6)
+			uh->check = ~udp_v6_check(gso_skb->len,
+						  &ipv6_hdr(gso_skb)->saddr,
+						  &ipv6_hdr(gso_skb)->daddr, 0);
+		else
+			uh->check = ~udp_v4_check(gso_skb->len,
+						  ip_hdr(gso_skb)->saddr,
+						  ip_hdr(gso_skb)->daddr, 0);
+	}
 
 	skb_pull(gso_skb, sizeof(*uh));
 

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tetsuo Handa <[email protected]>

commit ada1986d07976d60bed5017aa38b7f7cf27883f7 upstream.

Alfred Agrell found that TOMOYO cannot handle execveat(AT_EMPTY_PATH) inside chroot environment where /dev and /proc are not mounted, for commit 51f39a1f0cea ("syscalls: implement execveat() system call") missed that TOMOYO tries to canonicalize argv[0] when the filename fed to the executed program as argv[0] is supplied using potentially nonexistent pathname.

Since "/dev/fd/<fd>" already lost symlink information used for obtaining that <fd>, it is too late to reconstruct symlink's pathname. Although <filename> part of "/dev/fd/<fd>/<filename>" might not be canonicalized, TOMOYO cannot use tomoyo_realpath_nofollow() when /dev or /proc is not mounted. Therefore, fallback to tomoyo_realpath_from_path() when tomoyo_realpath_nofollow() failed.

---
 security/tomoyo/domain.c |    9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

@@ -723,10 +723,13 @@ int tomoyo_find_next_domain(struct linux
 	ee->r.obj = &ee->obj;
 	ee->obj.path1 = bprm->file->f_path;
 	/* Get symlink's pathname of program. */
-	retval = -ENOENT;
 	exename.name = tomoyo_realpath_nofollow(original_name);
-	if (!exename.name)
-		goto out;
+	if (!exename.name) {
+		/* Fallback to realpath if symlink's pathname does not exist. */
+		exename.name = tomoyo_realpath_from_path(&bprm->file->f_path);
+		if (!exename.name)
+			goto out;
+	}
 	tomoyo_fill_path_info(&exename);
 retry:
 	/* Check 'aggregator' directive. */

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: KhaiWenTan <[email protected]>

commit 675faf5a14c14a2be0b870db30a70764df81e2df upstream.

The commit b8c43360f6e4 ("net: stmmac: No need to calculate speed divider when offload is disabled") allows the "port_transmit_rate_kbps" to be set to a value of 0, which is then passed to the "div_s64" function when tc-cbs is disabled. This leads to a zero-division error.

When tc-cbs is disabled, the idleslope, sendslope, and credit values the credit values are not required to be configured. Therefore, adding a return statement after setting the txQ mode to DCB when tc-cbs is disabled would prevent a zero-division error.

---
 drivers/net/ethernet/stmicro/stmmac/stmmac_tc.c |    1 +
 1 file changed, 1 insertion(+)

@@ -396,6 +396,7 @@ static int tc_setup_cbs(struct stmmac_pr
 			return ret;
 
 		priv->plat->tx_queues_cfg[queue].mode_to_use = MTL_QUEUE_DCB;
+		return 0;
 	}
 
 	/* Final adjustments for HW */

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Krzysztof Kozlowski <[email protected]>

commit 73580e2ee6adfb40276bd420da3bb1abae204e10 upstream.

Driver is leaking an OF node reference obtained from of_parse_phandle_with_fixed_args().

---
 drivers/rtc/rtc-at91sam9.c |    1 +
 1 file changed, 1 insertion(+)

@@ -368,6 +368,7 @@ static int at91_rtc_probe(struct platfor
 		return ret;
 
 	rtc->gpbr = syscon_node_to_regmap(args.np);
+	of_node_put(args.np);
 	rtc->gpbr_offset = args.args[0];
 	if (IS_ERR(rtc->gpbr)) {
 		dev_err(&pdev->dev, "failed to retrieve gpbr regmap, aborting.\n");

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nuno Sa <[email protected]>

commit fb5cc65f973661241e4a2b7390b429aa7b330c69 upstream.

We register a devm action to call adp5589_clear_config() and then pass the i2c client as argument so that we can call i2c_get_clientdata() in order to get our device object. However, i2c_set_clientdata() is only being set at the end of the probe function which means that we'll get a NULL pointer dereference in case the probe function fails early.

---
 drivers/input/keyboard/adp5589-keys.c |    9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

@@ -936,10 +936,9 @@ static int adp5589_keypad_add(struct adp
 
 static void adp5589_clear_config(void *data)
 {
-	struct i2c_client *client = data;
-	struct adp5589_kpad *kpad = i2c_get_clientdata(client);
+	struct adp5589_kpad *kpad = data;
 
-	adp5589_write(client, kpad->var->reg(ADP5589_GENERAL_CFG), 0);
+	adp5589_write(kpad->client, kpad->var->reg(ADP5589_GENERAL_CFG), 0);
 }
 
 static int adp5589_probe(struct i2c_client *client)

@@ -983,7 +982,7 @@ static int adp5589_probe(struct i2c_clie
 	}
 
 	error = devm_add_action_or_reset(&client->dev, adp5589_clear_config,
-					 client);
+					 kpad);
 	if (error)
 		return error;
 

@@ -1010,8 +1009,6 @@ static int adp5589_probe(struct i2c_clie
 	if (error)
 		return error;
 
-	i2c_set_clientdata(client, kpad);
-
 	dev_info(&client->dev, "Rev.%d keypad, irq %d\n", revid, client->irq);
 	return 0;
 }

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nuno Sa <[email protected]>

commit c684771630e64bc39bddffeb65dd8a6612a6b249 upstream.

The adp5589 seems to have the same behavior as similar devices as explained in commit 910a9f5636f5 ("Input: adp5588-keys - get value from data out when dir is out").

Basically, when the gpio is set as output we need to get the value from ADP5589_GPO_DATA_OUT_A register instead of ADP5589_GPI_STATUS_A.

---
 drivers/input/keyboard/adp5589-keys.c |   13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

@@ -391,10 +391,17 @@ static int adp5589_gpio_get_value(struct
 	struct adp5589_kpad *kpad = gpiochip_get_data(chip);
 	unsigned int bank = kpad->var->bank(kpad->gpiomap[off]);
 	unsigned int bit = kpad->var->bit(kpad->gpiomap[off]);
+	int val;
 
-	return !!(adp5589_read(kpad->client,
-			       kpad->var->reg(ADP5589_GPI_STATUS_A) + bank) &
-			       bit);
+	mutex_lock(&kpad->gpio_lock);
+	if (kpad->dir[bank] & bit)
+		val = kpad->dat_out[bank];
+	else
+		val = adp5589_read(kpad->client,
+				   kpad->var->reg(ADP5589_GPI_STATUS_A) + bank);
+	mutex_unlock(&kpad->gpio_lock);
+
+	return !!(val & bit);
 }
 
 static void adp5589_gpio_set_value(struct gpio_chip *chip,

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Baokun Li <[email protected]>

commit da6ef2dffe6056aad3435e6cf7c6471c2a62187c upstream.

A dentry leak may be caused when a lookup cookie and a cull are concurrent:

P1 | P2
-----------------------------------------------------------
cachefiles_lookup_cookie
cachefiles_look_up_object

    lookup_one_positive_unlocked
     // get dentry
                            cachefiles_cull
                              inode->i_flags |= S_KERNEL_FILE;
    cachefiles_open_file
      cachefiles_mark_inode_in_use
        __cachefiles_mark_inode_in_use
          can_use = false
          if (!(inode->i_flags & S_KERNEL_FILE))
            can_use = true
	  return false
        return false
        // Returns an error but doesn't put dentry

After that the following WARNING will be triggered when the backend folder is umounted:

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

BUG: Dentry 000000008ad87947{i=7a,n=Dx_1_1.img}  still in use (1) [unmount of ext4 sda]
WARNING: CPU: 4 PID: 359261 at fs/dcache.c:1767 umount_check+0x5d/0x70
CPU: 4 PID: 359261 Comm: umount Not tainted 6.6.0-dirty #25
RIP: 0010:umount_check+0x5d/0x70
Call Trace:
 <TASK>
 d_walk+0xda/0x2b0
 do_one_tree+0x20/0x40
 shrink_dcache_for_umount+0x2c/0x90
 generic_shutdown_super+0x20/0x160
 kill_block_super+0x1a/0x40
 ext4_kill_sb+0x22/0x40
 deactivate_locked_super+0x35/0x80
 cleanup_mnt+0x104/0x160
==================================================================

Whether cachefiles_open_file() returns true or false, the reference count obtained by lookup_positive_unlocked() in cachefiles_look_up_object() should be released.

Therefore release that reference count in cachefiles_look_up_object() to fix the above issue and simplify the code.

---
 fs/cachefiles/namei.c |    7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

@@ -595,14 +595,12 @@ static bool cachefiles_open_file(struct
 	 * write and readdir but not lookup or open).
 	 */
 	touch_atime(&file->f_path);
-	dput(dentry);
 	return true;
 
 check_failed:
 	fscache_cookie_lookup_negative(object->cookie);
 	cachefiles_unmark_inode_in_use(object, file);
 	fput(file);
-	dput(dentry);
 	if (ret == -ESTALE)
 		return cachefiles_create_file(object);
 	return false;

@@ -611,7 +609,6 @@ error_fput:
 	fput(file);
 error:
 	cachefiles_do_unmark_inode_in_use(object, d_inode(dentry));
-	dput(dentry);
 	return false;
 }
 

@@ -654,7 +651,9 @@ bool cachefiles_look_up_object(struct ca
 		goto new_file;
 	}
 
-	if (!cachefiles_open_file(object, dentry))
+	ret = cachefiles_open_file(object, dentry);
+	dput(dentry);
+	if (!ret)
 		return false;
 
 	_leave(" = t [%lu]", file_inode(object->file)->i_ino);

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hans de Goede <[email protected]>

commit ac78288fe062b64e45a479eaae74aaaafcc8ecdd upstream.

Dell All In One (AIO) models released after 2017 may use a backlight controller board connected to an UART.

In DSDT this uart port will be defined as:

Name (_HID, "DELL0501")
Name (_CID, EisaId ("PNP0501")

The Dell OptiPlex 5480 AIO has an ACPI device for one of its UARTs with the above _HID + _CID. Loading the dell-uart-backlight driver fails with the following errors:

[   18.261353] dell_uart_backlight serial0-0: Timed out waiting for response.
[   18.261356] dell_uart_backlight serial0-0: error -ETIMEDOUT: getting firmware version
[   18.261359] dell_uart_backlight serial0-0: probe with driver dell_uart_backlight failed with error -110

Indicating that there is no backlight controller board attached to the UART, while the GPU's native backlight control method does work.

Add a quirk to use the GPU's native backlight control method on this model.

---
 drivers/acpi/video_detect.c |    9 +++++++++
 1 file changed, 9 insertions(+)

@@ -845,6 +845,15 @@ static const struct dmi_system_id video_
 	 * which need native backlight control nevertheless.
 	 */
 	{
+	 /* https://github.com/zabbly/linux/issues/26 */
+	 .callback = video_detect_force_native,
+	 /* Dell OptiPlex 5480 AIO */
+	 .matches = {
+		DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+		DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex 5480 AIO"),
+		},
+	},
+	{
 	 /* https://bugzilla.redhat.com/show_bug.cgi?id=2303936 */
 	 .callback = video_detect_force_native,
 	 /* Dell OptiPlex 7760 AIO */

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hans de Goede <[email protected]>

commit 65bdebf38e5fac7c56a9e05d3479a707e6dc783c upstream.

Commit d2aaf1996504 ("ACPI: resource: Add DMI quirks for ASUS Vivobook E1504GA and E1504GAB") does exactly what the subject says, adding DMI matches for both the E1504GA and E1504GAB.

But DMI_MATCH() does a substring match, so checking for E1504GA will also match E1504GAB.

Drop the unnecessary E1504GAB entry since that is covered already by the E1504GA entry.

---
 drivers/acpi/resource.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

index 8a4726e2eb69..1ff251fd1901 100644
@@ -511,19 +511,12 @@ static const struct dmi_system_id irq1_level_low_skip_override[] = {
 		},
 	},
 	{
-		/* Asus Vivobook E1504GA */
+		/* Asus Vivobook E1504GA* */
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
 			DMI_MATCH(DMI_BOARD_NAME, "E1504GA"),
 		},
 	},
-	{
-		/* Asus Vivobook E1504GAB */
-		.matches = {
-			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
-			DMI_MATCH(DMI_BOARD_NAME, "E1504GAB"),
-		},
-	},
 	{
 		/* Asus Vivobook Pro N6506MV */
 		.matches = {

2.46.2

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hans de Goede <[email protected]>

commit 63539defee17bf0cbd8e24078cf103efee9c6633 upstream.

Like other Asus Vivobooks, the Asus Vivobook Go E1404GA has a DSDT describing IRQ 1 as ActiveLow, while the kernel overrides to Edge_High.

    $ sudo dmesg | grep DMI:.*BIOS
    [    0.000000] DMI: ASUSTeK COMPUTER INC. Vivobook Go E1404GA_E1404GA/E1404GA, BIOS E1404GA.302 08/23/2023
    $ sudo cp /sys/firmware/acpi/tables/DSDT dsdt.dat
    $ iasl -d dsdt.dat
    $ grep -A 30 PS2K dsdt.dsl | grep IRQ -A 1
                IRQ (Level, ActiveLow, Exclusive, )
                    {1}

There already is an entry in the irq1_level_low_skip_override[] DMI match table for the "E1404GAB", change this to match on "E1404GA" to cover the E1404GA model as well (DMI_MATCH() does a substring match).

---
 drivers/acpi/resource.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -504,10 +504,10 @@ static const struct dmi_system_id irq1_l
 		},
 	},
 	{
-		/* Asus Vivobook Go E1404GAB */
+		/* Asus Vivobook Go E1404GA* */
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
-			DMI_MATCH(DMI_BOARD_NAME, "E1404GAB"),
+			DMI_MATCH(DMI_BOARD_NAME, "E1404GA"),
 		},
 	},
 	{

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hans de Goede <[email protected]>

commit 2f80ce0b78c340e332f04a5801dee5e4ac8cfaeb upstream.

Like other Asus Vivobook models the X1704VAP has its keybopard IRQ (1) described as ActiveLow in the DSDT, which the kernel overrides to EdgeHigh which breaks the keyboard.

Add the X1704VAP to the irq1_level_low_skip_override[] quirk table to fix this.

---
 drivers/acpi/resource.c |    7 +++++++
 1 file changed, 7 insertions(+)

@@ -441,6 +441,13 @@ static const struct dmi_system_id irq1_l
 		},
 	},
 	{
+		/* Asus Vivobook X1704VAP */
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
+			DMI_MATCH(DMI_BOARD_NAME, "X1704VAP"),
+		},
+	},
+	{
 		/* Asus ExpertBook B1402CBA */
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hans de Goede <[email protected]>

commit 056301e7c7c886f96d799edd36f3406cc30e1822 upstream.

Like other Asus ExpertBook models the B2502CVA has its keybopard IRQ (1) described as ActiveLow in the DSDT, which the kernel overrides to EdgeHigh which breaks the keyboard.

Add the B2502CVA to the irq1_level_low_skip_override[] quirk table to fix this.

---
 drivers/acpi/resource.c |    7 +++++++
 1 file changed, 7 insertions(+)

@@ -511,6 +511,13 @@ static const struct dmi_system_id irq1_l
 		},
 	},
 	{
+		/* Asus ExpertBook B2502CVA */
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
+			DMI_MATCH(DMI_BOARD_NAME, "B2502CVA"),
+		},
+	},
+	{
 		/* Asus Vivobook Go E1404GA* */
 		.matches = {
 			DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Qu Wenruo <[email protected]>

commit c3b47f49e83197e8dffd023ec568403bcdbb774b upstream.

[BUG]
Syzbot reported a NULL pointer dereference with the following crash:

FAULT_INJECTION: forcing a failure.

   start_transaction+0x830/0x1670 fs/btrfs/transaction.c:676
   prepare_to_relocate+0x31f/0x4c0 fs/btrfs/relocation.c:3642
   relocate_block_group+0x169/0xd20 fs/btrfs/relocation.c:3678
  ...
  BTRFS info (device loop0): balance: ended with status: -12
  Oops: general protection fault, probably for non-canonical address 0xdffffc00000000cc: 0000 [#1] PREEMPT SMP KASAN NOPTI
  KASAN: null-ptr-deref in range [0x0000000000000660-0x0000000000000667]
  RIP: 0010:btrfs_update_reloc_root+0x362/0xa80 fs/btrfs/relocation.c:926
  Call Trace:
   <TASK>
   commit_fs_roots+0x2ee/0x720 fs/btrfs/transaction.c:1496
   btrfs_commit_transaction+0xfaf/0x3740 fs/btrfs/transaction.c:2430
   del_balance_item fs/btrfs/volumes.c:3678 [inline]
   reset_balance_state+0x25e/0x3c0 fs/btrfs/volumes.c:3742
   btrfs_balance+0xead/0x10c0 fs/btrfs/volumes.c:4574
   btrfs_ioctl_balance+0x493/0x7c0 fs/btrfs/ioctl.c:3673
   vfs_ioctl fs/ioctl.c:51 [inline]
   __do_sys_ioctl fs/ioctl.c:907 [inline]
   __se_sys_ioctl+0xf9/0x170 fs/ioctl.c:893
   do_syscall_x64 arch/x86/entry/common.c:52 [inline]
   do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83
   entry_SYSCALL_64_after_hwframe+0x77/0x7f

[CAUSE]
The allocation failure happens at the start_transaction() inside prepare_to_relocate(), and during the error handling we call unset_reloc_control(), which makes fs_info->balance_ctl to be NULL.

Then we continue the error path cleanup in btrfs_balance() by calling reset_balance_state() which will call del_balance_item() to fully delete the balance item in the root tree.

However during the small window between set_reloc_contrl() and unset_reloc_control(), we can have a subvolume tree update and created a reloc_root for that subvolume.

Then we go into the final btrfs_commit_transaction() of del_balance_item(), and into btrfs_update_reloc_root() inside commit_fs_roots().

That function checks if fs_info->reloc_ctl is in the merge_reloc_tree stage, but since fs_info->reloc_ctl is NULL, it results a NULL pointer dereference.

[FIX]
Just add extra check on fs_info->reloc_ctl inside
btrfs_update_reloc_root(), before checking
fs_info->reloc_ctl->merge_reloc_tree.

That DEAD_RELOC_TREE handling is to prevent further modification to the reloc tree during merge stage, but since there is no reloc_ctl at all, we do not need to bother that.

---
 fs/btrfs/relocation.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -922,7 +922,7 @@ int btrfs_update_reloc_root(struct btrfs
 	btrfs_grab_root(reloc_root);
 
 	/* root->reloc_root will stay until current relocation finished */
-	if (fs_info->reloc_ctl->merge_reloc_tree &&
+	if (fs_info->reloc_ctl && fs_info->reloc_ctl->merge_reloc_tree &&
 	    btrfs_root_refs(root_item) == 0) {
 		set_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state);
 		/*

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Josef Bacik <[email protected]>

commit db7e68b522c01eb666cfe1f31637775f18997811 upstream.

Since the inception of relocation we have maintained the backref cache across transaction commits, updating the backref cache with the new bytenr whenever we COWed blocks that were in the cache, and then updating their bytenr once we detected a transaction id change.

This works as long as we're only ever modifying blocks, not changing the structure of the tree.

However relocation does in fact change the structure of the tree. For example, if we are relocating a data extent, we will look up all the leaves that point to this data extent. We will then call do_relocation() on each of these leaves, which will COW down to the leaf and then update the file extent location.

But, a key feature of do_relocation() is the pending list. This is all the pending nodes that we modified when we updated the file extent item. We will then process all of these blocks via finish_pending_nodes, which calls do_relocation() on all of the nodes that led up to that leaf.

The purpose of this is to make sure we don't break sharing unless we absolutely have to. Consider the case that we have 3 snapshots that all point to this leaf through the same nodes, the initial COW would have created a whole new path. If we did this for all 3 snapshots we would end up with 3x the number of nodes we had originally. To avoid this we will cycle through each of the snapshots that point to each of these nodes and update their pointers to point at the new nodes.

Once we update the pointer to the new node we will drop the node we removed the link for and all of its children via btrfs_drop_subtree(). This is essentially just btrfs_drop_snapshot(), but for an arbitrary point in the snapshot.

The problem with this is that we will never reflect this in the backref cache. If we do this btrfs_drop_snapshot() for a node that is in the backref tree, we will leave the node in the backref tree. This becomes a problem when we change the transid, as now the backref cache has entire subtrees that no longer exist, but exist as if they still are pointed to by the same roots.

In the best case scenario you end up with "adding refs to an existing tree ref" errors from insert_inline_extent_backref(), where we attempt to link in nodes on roots that are no longer valid.

Worst case you will double free some random block and re-use it when there's still references to the block.

This is extremely subtle, and the consequences are quite bad. There isn't a way to make sure our backref cache is consistent between transid's.

In order to fix this we need to simply evict the entire backref cache anytime we cross transid's. This reduces performance in that we have to rebuild this backref cache every time we change transid's, but fixes the bug.

This has existed since relocation was added, and is a pretty critical bug. There's a lot more cleanup that can be done now that this functionality is going away, but this patch is as small as possible in order to fix the problem and make it easy for us to backport it to all the kernels it needs to be backported to.

Followup series will dismantle more of this code and simplify relocation drastically to remove this functionality.

We have a reproducer that reproduced the corruption within a few minutes of running. With this patch it survives several iterations/hours of running the reproducer.

---
 fs/btrfs/backref.c    |   12 +++++---
 fs/btrfs/relocation.c |   75 ++------------------------------------------------
 2 files changed, 11 insertions(+), 76 deletions(-)

@@ -3179,10 +3179,14 @@ void btrfs_backref_release_cache(struct
 		btrfs_backref_cleanup_node(cache, node);
 	}
 
-	cache->last_trans = 0;
-
-	for (i = 0; i < BTRFS_MAX_LEVEL; i++)
-		ASSERT(list_empty(&cache->pending[i]));
+	for (i = 0; i < BTRFS_MAX_LEVEL; i++) {
+		while (!list_empty(&cache->pending[i])) {
+			node = list_first_entry(&cache->pending[i],
+						struct btrfs_backref_node,
+						list);
+			btrfs_backref_cleanup_node(cache, node);
+		}
+	}
 	ASSERT(list_empty(&cache->pending_edge));
 	ASSERT(list_empty(&cache->useless_node));
 	ASSERT(list_empty(&cache->changed));

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Filipe Manana <[email protected]>

commit fa630df665aa9ddce3a96ce7b54e10a38e4d2a2b upstream.

During an incremental send we may end up sending an invalid clone operation, for the last extent of a file which ends at an unaligned offset that matches the final i_size of the file in the send snapshot, in case the file had its initial size (the size in the parent snapshot) decreased in the send snapshot. In this case the destination will fail to apply the clone operation because its end offset is not sector size aligned and it ends before the current size of the file.

Sending the truncate operation always happens when we finish processing an inode, after we process all its extents (and xattrs, names, etc). So fix this by ensuring the file has a valid size before we send a clone operation for an unaligned extent that ends at the final i_size of the file. The size we truncate to matches the start offset of the clone range but it could be any value between that start offset and the final size of the file since the clone operation will expand the i_size if the current size is smaller than the end offset. The start offset of the range was chosen because it's always sector size aligned and avoids a truncation into the middle of a page, which results in dirtying the page due to filling part of it with zeroes and then making the clone operation at the receiver trigger IO.

The following test reproduces the issue:

  $ cat test.sh
  #!/bin/bash

DEV=/dev/sdi
MNT=/mnt/sdi

mkfs.btrfs -f $DEV
mount $DEV $MNT

  # Create a file with a size of 256K + 5 bytes, having two extents, one
  # with a size of 128K and another one with a size of 128K + 5 bytes.
  last_ext_size=$((128 * 1024 + 5))
  xfs_io -f -d -c "pwrite -S 0xab -b 128K 0 128K" \
         -c "pwrite -S 0xcd -b $last_ext_size 128K $last_ext_size" \
         $MNT/foo

# Another file which we will later clone foo into, but initially with

  # a larger size than foo.
  xfs_io -f -c "pwrite -S 0xef 0 1M" $MNT/bar

btrfs subvolume snapshot -r $MNT/ $MNT/snap1

  # Now resize bar and clone foo into it.
  xfs_io -c "truncate 0" \
         -c "reflink $MNT/foo" $MNT/bar

btrfs subvolume snapshot -r $MNT/ $MNT/snap2

rm -f /tmp/send-full /tmp/send-inc
btrfs send -f /tmp/send-full $MNT/snap1 btrfs send -p $MNT/snap1 -f /tmp/send-inc $MNT/snap2

umount $MNT
mkfs.btrfs -f $DEV
mount $DEV $MNT

btrfs receive -f /tmp/send-full $MNT
btrfs receive -f /tmp/send-inc $MNT

umount $MNT

Running it before this patch:

  $ ./test.sh
  (...)
  At subvol snap1
  At snapshot snap2
  ERROR: failed to clone extents to bar: Invalid argument

A test case for fstests will be sent soon.

---
 fs/btrfs/send.c |   23 ++++++++++++++++++++++-
 1 file changed, 22 insertions(+), 1 deletion(-)

@@ -6188,8 +6188,29 @@ static int send_write_or_clone(struct se
 	if (ret < 0)
 		return ret;
 
-	if (clone_root->offset + num_bytes == info.size)
+	if (clone_root->offset + num_bytes == info.size) {
+		/*
+		 * The final size of our file matches the end offset, but it may
+		 * be that its current size is larger, so we have to truncate it
+		 * to any value between the start offset of the range and the
+		 * final i_size, otherwise the clone operation is invalid
+		 * because it's unaligned and it ends before the current EOF.
+		 * We do this truncate to the final i_size when we finish
+		 * processing the inode, but it's too late by then. And here we
+		 * truncate to the start offset of the range because it's always
+		 * sector size aligned while if it were the final i_size it
+		 * would result in dirtying part of a page, filling part of a
+		 * page with zeroes and then having the clone operation at the
+		 * receiver trigger IO and wait for it due to the dirty page.
+		 */
+		if (sctx->parent_root != NULL) {
+			ret = send_truncate(sctx, sctx->cur_ino,
+					    sctx->cur_inode_gen, offset);
+			if (ret < 0)
+				return ret;
+		}
 		goto clone_data;
+	}
 
 write_data:
 	ret = send_extent_data(sctx, path, offset, num_bytes);

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Filipe Manana <[email protected]>

commit 41fd1e94066a815a7ab0a7025359e9b40e4b3576 upstream.

During unmount, at close_ctree(), we have the following steps in this order:

1) Park the cleaner kthread - this doesn't destroy the kthread, it basically halts its execution (wake ups against it work but do nothing);

2) We stop the cleaner kthread - this results in freeing the respective struct task_struct;

3) We call btrfs_stop_all_workers() which waits for any jobs running in all the work queues and then free the work queues.

Syzbot reported a case where a fixup worker resulted in a crash when doing a delayed iput on its inode while attempting to wake up the cleaner at btrfs_add_delayed_iput(), because the task_struct of the cleaner kthread was already freed. This can happen during unmount because we don't wait for any fixup workers still running before we call kthread_stop() against the cleaner kthread, which stops and free all its resources.

Fix this by waiting for any fixup workers at close_ctree() before we call kthread_stop() against the cleaner and run pending delayed iputs.

The stack traces reported by syzbot were the following:

  BUG: KASAN: slab-use-after-free in __lock_acquire+0x77/0x2050 kernel/locking/lockdep.c:5065
  Read of size 8 at addr ffff8880272a8a18 by task kworker/u8:3/52

  CPU: 1 UID: 0 PID: 52 Comm: kworker/u8:3 Not tainted 6.12.0-rc1-syzkaller #0
  Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 09/13/2024
  Workqueue: btrfs-fixup btrfs_work_helper
  Call Trace:
   <TASK>
   __dump_stack lib/dump_stack.c:94 [inline]
   dump_stack_lvl+0x241/0x360 lib/dump_stack.c:120
   print_address_description mm/kasan/report.c:377 [inline]
   print_report+0x169/0x550 mm/kasan/report.c:488
   kasan_report+0x143/0x180 mm/kasan/report.c:601
   __lock_acquire+0x77/0x2050 kernel/locking/lockdep.c:5065
   lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5825
   __raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
   _raw_spin_lock_irqsave+0xd5/0x120 kernel/locking/spinlock.c:162
   class_raw_spinlock_irqsave_constructor include/linux/spinlock.h:551 [inline]
   try_to_wake_up+0xb0/0x1480 kernel/sched/core.c:4154
   btrfs_writepage_fixup_worker+0xc16/0xdf0 fs/btrfs/inode.c:2842
   btrfs_work_helper+0x390/0xc50 fs/btrfs/async-thread.c:314
   process_one_work kernel/workqueue.c:3229 [inline]
   process_scheduled_works+0xa63/0x1850 kernel/workqueue.c:3310
   worker_thread+0x870/0xd30 kernel/workqueue.c:3391
   kthread+0x2f0/0x390 kernel/kthread.c:389
   ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147
   ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244
   </TASK>

  Allocated by task 2:
   kasan_save_stack mm/kasan/common.c:47 [inline]
   kasan_save_track+0x3f/0x80 mm/kasan/common.c:68
   unpoison_slab_object mm/kasan/common.c:319 [inline]
   __kasan_slab_alloc+0x66/0x80 mm/kasan/common.c:345
   kasan_slab_alloc include/linux/kasan.h:247 [inline]
   slab_post_alloc_hook mm/slub.c:4086 [inline]
   slab_alloc_node mm/slub.c:4135 [inline]
   kmem_cache_alloc_node_noprof+0x16b/0x320 mm/slub.c:4187
   alloc_task_struct_node kernel/fork.c:180 [inline]
   dup_task_struct+0x57/0x8c0 kernel/fork.c:1107
   copy_process+0x5d1/0x3d50 kernel/fork.c:2206
   kernel_clone+0x223/0x880 kernel/fork.c:2787
   kernel_thread+0x1bc/0x240 kernel/fork.c:2849
   create_kthread kernel/kthread.c:412 [inline]
   kthreadd+0x60d/0x810 kernel/kthread.c:765
   ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147
   ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244

  Freed by task 61:
   kasan_save_stack mm/kasan/common.c:47 [inline]
   kasan_save_track+0x3f/0x80 mm/kasan/common.c:68
   kasan_save_free_info+0x40/0x50 mm/kasan/generic.c:579
   poison_slab_object mm/kasan/common.c:247 [inline]
   __kasan_slab_free+0x59/0x70 mm/kasan/common.c:264
   kasan_slab_free include/linux/kasan.h:230 [inline]
   slab_free_hook mm/slub.c:2343 [inline]
   slab_free mm/slub.c:4580 [inline]
   kmem_cache_free+0x1a2/0x420 mm/slub.c:4682
   put_task_struct include/linux/sched/task.h:144 [inline]
   delayed_put_task_struct+0x125/0x300 kernel/exit.c:228
   rcu_do_batch kernel/rcu/tree.c:2567 [inline]
   rcu_core+0xaaa/0x17a0 kernel/rcu/tree.c:2823
   handle_softirqs+0x2c5/0x980 kernel/softirq.c:554
   __do_softirq kernel/softirq.c:588 [inline]
   invoke_softirq kernel/softirq.c:428 [inline]
   __irq_exit_rcu+0xf4/0x1c0 kernel/softirq.c:637
   irq_exit_rcu+0x9/0x30 kernel/softirq.c:649
   instr_sysvec_apic_timer_interrupt arch/x86/kernel/apic/apic.c:1037 [inline]
   sysvec_apic_timer_interrupt+0xa6/0xc0 arch/x86/kernel/apic/apic.c:1037
   asm_sysvec_apic_timer_interrupt+0x1a/0x20 arch/x86/include/asm/idtentry.h:702

  Last potentially related work creation:
   kasan_save_stack+0x3f/0x60 mm/kasan/common.c:47
   __kasan_record_aux_stack+0xac/0xc0 mm/kasan/generic.c:541
   __call_rcu_common kernel/rcu/tree.c:3086 [inline]
   call_rcu+0x167/0xa70 kernel/rcu/tree.c:3190
   context_switch kernel/sched/core.c:5318 [inline]
   __schedule+0x184b/0x4ae0 kernel/sched/core.c:6675
   schedule_idle+0x56/0x90 kernel/sched/core.c:6793
   do_idle+0x56a/0x5d0 kernel/sched/idle.c:354
   cpu_startup_entry+0x42/0x60 kernel/sched/idle.c:424
   start_secondary+0x102/0x110 arch/x86/kernel/smpboot.c:314
   common_startup_64+0x13e/0x147

  The buggy address belongs to the object at ffff8880272a8000
   which belongs to the cache task_struct of size 7424
  The buggy address is located 2584 bytes inside of
   freed 7424-byte region [ffff8880272a8000, ffff8880272a9d00)

  The buggy address belongs to the physical page:
  page: refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x272a8
  head: order:3 mapcount:0 entire_mapcount:0 nr_pages_mapped:0 pincount:0
  flags: 0xfff00000000040(head|node=0|zone=1|lastcpupid=0x7ff)
  page_type: f5(slab)
  raw: 00fff00000000040 ffff88801bafa500 dead000000000122 0000000000000000
  raw: 0000000000000000 0000000080040004 00000001f5000000 0000000000000000
  head: 00fff00000000040 ffff88801bafa500 dead000000000122 0000000000000000
  head: 0000000000000000 0000000080040004 00000001f5000000 0000000000000000
  head: 00fff00000000003 ffffea00009caa01 ffffffffffffffff 0000000000000000
  head: 0000000000000008 0000000000000000 00000000ffffffff 0000000000000000
  page dumped because: kasan: bad access detected
  page_owner tracks the page as allocated
  page last allocated via order 3, migratetype Unmovable, gfp_mask 0xd20c0(__GFP_IO|__GFP_FS|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC), pid 2, tgid 2 (kthreadd), ts 71247381401, free_ts 71214998153
   set_page_owner include/linux/page_owner.h:32 [inline]
   post_alloc_hook+0x1f3/0x230 mm/page_alloc.c:1537
   prep_new_page mm/page_alloc.c:1545 [inline]
   get_page_from_freelist+0x3039/0x3180 mm/page_alloc.c:3457
   __alloc_pages_noprof+0x256/0x6c0 mm/page_alloc.c:4733
   alloc_pages_mpol_noprof+0x3e8/0x680 mm/mempolicy.c:2265
   alloc_slab_page+0x6a/0x120 mm/slub.c:2413
   allocate_slab+0x5a/0x2f0 mm/slub.c:2579
   new_slab mm/slub.c:2632 [inline]
   ___slab_alloc+0xcd1/0x14b0 mm/slub.c:3819
   __slab_alloc+0x58/0xa0 mm/slub.c:3909
   __slab_alloc_node mm/slub.c:3962 [inline]
   slab_alloc_node mm/slub.c:4123 [inline]
   kmem_cache_alloc_node_noprof+0x1fe/0x320 mm/slub.c:4187
   alloc_task_struct_node kernel/fork.c:180 [inline]
   dup_task_struct+0x57/0x8c0 kernel/fork.c:1107
   copy_process+0x5d1/0x3d50 kernel/fork.c:2206
   kernel_clone+0x223/0x880 kernel/fork.c:2787
   kernel_thread+0x1bc/0x240 kernel/fork.c:2849
   create_kthread kernel/kthread.c:412 [inline]
   kthreadd+0x60d/0x810 kernel/kthread.c:765
   ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147
   ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244
  page last free pid 5230 tgid 5230 stack trace:
   reset_page_owner include/linux/page_owner.h:25 [inline]
   free_pages_prepare mm/page_alloc.c:1108 [inline]
   free_unref_page+0xcd0/0xf00 mm/page_alloc.c:2638
   discard_slab mm/slub.c:2678 [inline]
   __put_partials+0xeb/0x130 mm/slub.c:3146
   put_cpu_partial+0x17c/0x250 mm/slub.c:3221
   __slab_free+0x2ea/0x3d0 mm/slub.c:4450
   qlink_free mm/kasan/quarantine.c:163 [inline]
   qlist_free_all+0x9a/0x140 mm/kasan/quarantine.c:179
   kasan_quarantine_reduce+0x14f/0x170 mm/kasan/quarantine.c:286
   __kasan_slab_alloc+0x23/0x80 mm/kasan/common.c:329
   kasan_slab_alloc include/linux/kasan.h:247 [inline]
   slab_post_alloc_hook mm/slub.c:4086 [inline]
   slab_alloc_node mm/slub.c:4135 [inline]
   kmem_cache_alloc_noprof+0x135/0x2a0 mm/slub.c:4142
   getname_flags+0xb7/0x540 fs/namei.c:139
   do_sys_openat2+0xd2/0x1d0 fs/open.c:1409
   do_sys_open fs/open.c:1430 [inline]
   __do_sys_openat fs/open.c:1446 [inline]
   __se_sys_openat fs/open.c:1441 [inline]
   __x64_sys_openat+0x247/0x2a0 fs/open.c:1441
   do_syscall_x64 arch/x86/entry/common.c:52 [inline]
   do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83
   entry_SYSCALL_64_after_hwframe+0x77/0x7f

  Memory state around the buggy address:
   ffff8880272a8900: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
   ffff8880272a8980: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
  >ffff8880272a8a00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
                              ^
   ffff8880272a8a80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
   ffff8880272a8b00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
  ==================================================================

---
 fs/btrfs/disk-io.c |   11 +++++++++++
 1 file changed, 11 insertions(+)

@@ -4272,6 +4272,17 @@ void __cold close_ctree(struct btrfs_fs_
 	btrfs_cleanup_defrag_inodes(fs_info);
 
 	/*
+	 * Wait for any fixup workers to complete.
+	 * If we don't wait for them here and they are still running by the time
+	 * we call kthread_stop() against the cleaner kthread further below, we
+	 * get an use-after-free on the cleaner because the fixup worker adds an
+	 * inode to the list of delayed iputs and then attempts to wakeup the
+	 * cleaner kthread, which was already stopped and destroyed. We parked
+	 * already the cleaner, but below we run all pending delayed iputs.
+	 */
+	btrfs_flush_workqueue(fs_info->fixup_workers);
+
+	/*
 	 * After we parked the cleaner kthread, ordered extents may have
 	 * completed and created new delayed iputs. If one of the async reclaim
 	 * tasks is running and in the RUN_DELAYED_IPUTS flush state, then we

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Miquel Sabaté Solà <[email protected]>

commit c0f02536fffbbec71aced36d52a765f8c4493dc2 upstream.

In the parse_perf_domain function, if the call to of_parse_phandle_with_args returns an error, then the reference to the CPU device node that was acquired at the start of the function would not be properly decremented.

Address this by declaring the variable with the __free(device_node) cleanup attribute.

---
 include/linux/cpufreq.h |    6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

@@ -1113,10 +1113,9 @@ static inline int parse_perf_domain(int
 				    const char *cell_name,
 				    struct of_phandle_args *args)
 {
-	struct device_node *cpu_np;
 	int ret;
 
-	cpu_np = of_cpu_device_node_get(cpu);
+	struct device_node *cpu_np __free(device_node) = of_cpu_device_node_get(cpu);
 	if (!cpu_np)
 		return -ENODEV;
 

@@ -1124,9 +1123,6 @@ static inline int parse_perf_domain(int
 					 args);
 	if (ret < 0)
 		return ret;
-
-	of_node_put(cpu_np);
-
 	return 0;
 }
 

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Emanuele Ghidoli <[email protected]>

commit 3360d41f4ac490282fddc3ccc0b58679aa5c065d upstream.

On a few platforms such as TI's AM69 device, disable_irq() fails to keep track of the interrupts that happen between disable_irq() and enable_irq() and those interrupts are missed. Use the ->irq_unmask() and ->irq_mask() methods instead of ->irq_enable() and ->irq_disable() to correctly keep track of edges when disable_irq is called.

This solves the issue of disable_irq() not working as expected on such platforms.

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

@@ -289,7 +289,7 @@ static int davinci_gpio_probe(struct pla
  * serve as EDMA event triggers.
  */
 
-static void gpio_irq_disable(struct irq_data *d)
+static void gpio_irq_mask(struct irq_data *d)
 {
 	struct davinci_gpio_regs __iomem *g = irq2regs(d);
 	uintptr_t mask = (uintptr_t)irq_data_get_irq_handler_data(d);

@@ -298,7 +298,7 @@ static void gpio_irq_disable(struct irq_
 	writel_relaxed(mask, &g->clr_rising);
 }
 
-static void gpio_irq_enable(struct irq_data *d)
+static void gpio_irq_unmask(struct irq_data *d)
 {
 	struct davinci_gpio_regs __iomem *g = irq2regs(d);
 	uintptr_t mask = (uintptr_t)irq_data_get_irq_handler_data(d);

@@ -324,8 +324,8 @@ static int gpio_irq_type(struct irq_data
 
 static struct irq_chip gpio_irqchip = {
 	.name		= "GPIO",
-	.irq_enable	= gpio_irq_enable,
-	.irq_disable	= gpio_irq_disable,
+	.irq_unmask	= gpio_irq_unmask,
+	.irq_mask	= gpio_irq_mask,
 	.irq_set_type	= gpio_irq_type,
 	.flags		= IRQCHIP_SET_TYPE_MASKED | IRQCHIP_SKIP_SET_WAKE,
 };

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jiawen Wu <[email protected]>

commit 93ef6ee5c20e9330477930ec6347672c9e0cf5a6 upstream.

The value is read from the register TXGBE_RX_GEN_CTL3, and it should be written back to TXGBE_RX_GEN_CTL3 when it changes some fields.

---
 drivers/net/pcs/pcs-xpcs-wx.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -109,7 +109,7 @@ static void txgbe_pma_config_1g(struct d
 	txgbe_write_pma(xpcs, TXGBE_DFE_TAP_CTL0, 0);
 	val = txgbe_read_pma(xpcs, TXGBE_RX_GEN_CTL3);
 	val = u16_replace_bits(val, 0x4, TXGBE_RX_GEN_CTL3_LOS_TRSHLD0);
-	txgbe_write_pma(xpcs, TXGBE_RX_EQ_ATTN_CTL, val);
+	txgbe_write_pma(xpcs, TXGBE_RX_GEN_CTL3, val);
 
 	txgbe_write_pma(xpcs, TXGBE_MPLLA_CTL0, 0x20);
 	txgbe_write_pma(xpcs, TXGBE_MPLLA_CTL3, 0x46);

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Luiz Augusto von Dentz <[email protected]>

commit b25e11f978b63cb7857890edb3a698599cddb10e upstream.

This aligned BR/EDR JUST_WORKS method with LE which since 92516cd97fd4 ("Bluetooth: Always request for user confirmation for Just Works") always request user confirmation with confirm_hint set since the likes of bluetoothd have dedicated policy around JUST_WORKS method (e.g. main.conf:JustWorksRepairing).

CVE: CVE-2024-8805

---
 net/bluetooth/hci_event.c |   13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

@@ -5325,19 +5325,16 @@ static void hci_user_confirm_request_evt
 		goto unlock;
 	}
 
-	/* If no side requires MITM protection; auto-accept */
+	/* If no side requires MITM protection; use JUST_CFM method */
 	if ((!loc_mitm || conn->remote_cap == HCI_IO_NO_INPUT_OUTPUT) &&
 	    (!rem_mitm || conn->io_capability == HCI_IO_NO_INPUT_OUTPUT)) {
 
-		/* If we're not the initiators request authorization to
-		 * proceed from user space (mgmt_user_confirm with
-		 * confirm_hint set to 1). The exception is if neither
-		 * side had MITM or if the local IO capability is
-		 * NoInputNoOutput, in which case we do auto-accept
+		/* If we're not the initiator of request authorization and the
+		 * local IO capability is not NoInputNoOutput, use JUST_WORKS
+		 * method (mgmt_user_confirm with confirm_hint set to 1).
 		 */
 		if (!test_bit(HCI_CONN_AUTH_PEND, &conn->flags) &&
-		    conn->io_capability != HCI_IO_NO_INPUT_OUTPUT &&
-		    (loc_mitm || rem_mitm)) {
+		    conn->io_capability != HCI_IO_NO_INPUT_OUTPUT) {
 			bt_dev_dbg(hdev, "Confirming auto-accept as acceptor");
 			confirm_hint = 1;
 			goto confirm;

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jiawei Ye <[email protected]>

commit bff1709b3980bd7f80be6786f64cc9a9ee9e56da upstream.

In the mac802154_scan_worker function, the scan_req->type field was accessed after the RCU read-side critical section was unlocked. According to RCU usage rules, this is illegal and can lead to unpredictable behavior, such as accessing memory that has been updated or causing use-after-free issues.

This possible bug was identified using a static analysis tool developed by myself, specifically designed to detect RCU-related issues.

To address this, the scan_req->type value is now stored in a local variable scan_req_type while still within the RCU read-side critical section. The scan_req_type is then used after the RCU lock is released, ensuring that the type value is safely accessed without violating RCU rules.

---
 net/mac802154/scan.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

@@ -176,6 +176,7 @@ void mac802154_scan_worker(struct work_s
 	struct ieee802154_local *local =
 		container_of(work, struct ieee802154_local, scan_work.work);
 	struct cfg802154_scan_request *scan_req;
+	enum nl802154_scan_types scan_req_type;
 	struct ieee802154_sub_if_data *sdata;
 	unsigned int scan_duration = 0;
 	struct wpan_phy *wpan_phy;

@@ -209,6 +210,7 @@ void mac802154_scan_worker(struct work_s
 	}
 
 	wpan_phy = scan_req->wpan_phy;
+	scan_req_type = scan_req->type;
 	scan_req_duration = scan_req->duration;
 
 	/* Look for the next valid chan */

@@ -246,7 +248,7 @@ void mac802154_scan_worker(struct work_s
 		goto end_scan;
 	}
 
-	if (scan_req->type == NL802154_SCAN_ACTIVE) {
+	if (scan_req_type == NL802154_SCAN_ACTIVE) {
 		ret = mac802154_transmit_beacon_req(local, sdata);
 		if (ret)
 			dev_err(&sdata->dev->dev,

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jens Axboe <[email protected]>

commit c314094cb4cfa6fc5a17f4881ead2dfebfa717a7 upstream.

If the recv returns zero, or an error, then it doesn't matter if more data has already been received for this buffer. A condition like that should terminate the multishot receive. Rather than pass in the collected return value, pass in whether to terminate or keep the recv going separately.

Note that this isn't a bug right now, as the only way to get there is via setting MSG_WAITALL with multishot receive. And if an application does that, then -EINVAL is returned anyway. But it seems like an easy bug to introduce, so let's make it a bit more explicit.

---
 io_uring/net.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

@@ -1116,6 +1116,7 @@ int io_recv(struct io_kiocb *req, unsign
 	int ret, min_ret = 0;
 	bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK;
 	size_t len = sr->len;
+	bool mshot_finished;
 
 	if (!(req->flags & REQ_F_POLLED) &&
 	    (sr->flags & IORING_RECVSEND_POLL_FIRST))

@@ -1170,6 +1171,7 @@ out_free:
 		req_set_fail(req);
 	}
 
+	mshot_finished = ret <= 0;
 	if (ret > 0)
 		ret += sr->done_io;
 	else if (sr->done_io)

@@ -1177,7 +1179,7 @@ out_free:
 	else
 		io_kbuf_recycle(req, issue_flags);
 
-	if (!io_recv_finish(req, &ret, kmsg, ret <= 0, issue_flags))
+	if (!io_recv_finish(req, &ret, kmsg, mshot_finished, issue_flags))
 		goto retry_multishot;
 
 	return ret;

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Patrick Donnelly <[email protected]>

commit ccda9910d8490f4fb067131598e4b2e986faa5a0 upstream.

Log recovered from a user's cluster:

    <7>[ 5413.970692] ceph:  get_cap_refs 00000000958c114b ret 1 got Fr
    <7>[ 5413.970695] ceph:  start_read 00000000958c114b, no cache cap
    ...
    <7>[ 5473.934609] ceph:   my wanted = Fr, used = Fr, dirty -
    <7>[ 5473.934616] ceph:  revocation: pAsLsXsFr -> pAsLsXs (revoking Fr)
    <7>[ 5473.934632] ceph:  __ceph_caps_issued 00000000958c114b cap 00000000f7784259 issued pAsLsXs
    <7>[ 5473.934638] ceph:  check_caps 10000000e68.fffffffffffffffe file_want - used Fr dirty - flushing - issued pAsLsXs revoking Fr retain pAsLsXsFsr  AUTHONLY NOINVAL FLUSH_FORCE

The MDS subsequently complains that the kernel client is late releasing caps.

Approximately, a series of changes to this code by commits 49870056005c ("ceph: convert ceph_readpages to ceph_readahead"), 2de160417315 ("netfs: Change ->init_request() to return an error code") and a5c9dc445139 ("ceph: Make ceph_init_request() check caps on readahead") resulted in subtle resource cleanup to be missed. The main culprit is the change in error handling in 2de160417315 which meant that a failure in init_request() would no longer cause cleanup to be called. That would prevent the ceph_put_cap_refs() call which would cleanup the leaked cap ref.

---
 fs/ceph/addr.c |    5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

@@ -468,8 +468,11 @@ static int ceph_init_request(struct netf
 	rreq->netfs_priv = priv;
 
 out:
-	if (ret < 0)
+	if (ret < 0) {
+		if (got)
+			ceph_put_cap_refs(ceph_inode(inode), got);
 		kfree(priv);
+	}
 
 	return ret;
 }

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wei Li <[email protected]>

commit 2a13ca2e8abb12ee43ada8a107dadca83f140937 upstream.

The cpuhp online/offline processing race also exists in percpu-mode hwlat tracer in theory, apply the fix too. That is:

    T1                       | T2
    [CPUHP_ONLINE]           | cpu_device_down()
     hwlat_hotplug_workfn()  |
                             |     cpus_write_lock()
                             |     takedown_cpu(1)
                             |     cpus_write_unlock()
    [CPUHP_OFFLINE]          |
        cpus_read_lock()     |
        start_kthread(1)     |
        cpus_read_unlock()   |

---
 kernel/trace/trace_hwlat.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -520,6 +520,8 @@ static void hwlat_hotplug_workfn(struct
 	if (!hwlat_busy || hwlat_data.thread_mode != MODE_PER_CPU)
 		goto out_unlock;
 
+	if (!cpu_online(cpu))
+		goto out_unlock;
 	if (!cpumask_test_cpu(cpu, tr->tracing_cpumask))
 		goto out_unlock;
 

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wei Li <[email protected]>

commit b484a02c9cedf8703eff8f0756f94618004bd165 upstream.

stop_kthread() is the offline callback for "trace/osnoise:online", since commit 5bfbcd1ee57b ("tracing/timerlat: Add interface_lock around clearing of kthread in stop_kthread()"), the following ABBA deadlock scenario is introduced:

T1                            | T2 [BP]               | T3 [AP]
osnoise_hotplug_workfn()      | work_for_cpu_fn()     | cpuhp_thread_fun()
                              |   _cpu_down()         |   osnoise_cpu_die()
  mutex_lock(&interface_lock) |                       |     stop_kthread()
                              |     cpus_write_lock() |       mutex_lock(&interface_lock)
  cpus_read_lock()            |     cpuhp_kick_ap()   |

As the interface_lock here in just for protecting the "kthread" field of the osn_var, use xchg() instead to fix this issue. Also use for_each_online_cpu() back in stop_per_cpu_kthreads() as it can take cpu_read_lock() again.

---
 kernel/trace/trace_osnoise.c |   13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)

@@ -1953,12 +1953,8 @@ static void stop_kthread(unsigned int cp
 {
 	struct task_struct *kthread;
 
-	mutex_lock(&interface_lock);
-	kthread = per_cpu(per_cpu_osnoise_var, cpu).kthread;
+	kthread = xchg_relaxed(&(per_cpu(per_cpu_osnoise_var, cpu).kthread), NULL);
 	if (kthread) {
-		per_cpu(per_cpu_osnoise_var, cpu).kthread = NULL;
-		mutex_unlock(&interface_lock);
-
 		if (cpumask_test_and_clear_cpu(cpu, &kthread_cpumask) &&
 		    !WARN_ON(!test_bit(OSN_WORKLOAD, &osnoise_options))) {
 			kthread_stop(kthread);

@@ -1972,7 +1968,6 @@ static void stop_kthread(unsigned int cp
 			put_task_struct(kthread);
 		}
 	} else {
-		mutex_unlock(&interface_lock);
 		/* if no workload, just return */
 		if (!test_bit(OSN_WORKLOAD, &osnoise_options)) {
 			/*

@@ -1994,8 +1989,12 @@ static void stop_per_cpu_kthreads(void)
 {
 	int cpu;
 
-	for_each_possible_cpu(cpu)
+	cpus_read_lock();
+
+	for_each_online_cpu(cpu)
 		stop_kthread(cpu);
+
+	cpus_read_unlock();
 }
 
 /*

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wei Li <[email protected]>

commit 829e0c9f0855f26b3ae830d17b24aec103f7e915 upstream.

There is another found exception that the "timerlat/1" thread was scheduled on CPU0, and lead to timer corruption finally:

```
ODEBUG: init active (active state 0) object: ffff888237c2e108 object type: hrtimer hint: timerlat_irq+0x0/0x220
WARNING: CPU: 0 PID: 426 at lib/debugobjects.c:518 debug_print_object+0x7d/0xb0
Modules linked in:
CPU: 0 UID: 0 PID: 426 Comm: timerlat/1 Not tainted 6.11.0-rc7+ #45
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1.1 04/01/2014
RIP: 0010:debug_print_object+0x7d/0xb0
...
Call Trace:
 <TASK>
 ? __warn+0x7c/0x110
 ? debug_print_object+0x7d/0xb0
 ? report_bug+0xf1/0x1d0
 ? prb_read_valid+0x17/0x20
 ? handle_bug+0x3f/0x70
 ? exc_invalid_op+0x13/0x60
 ? asm_exc_invalid_op+0x16/0x20
 ? debug_print_object+0x7d/0xb0
 ? debug_print_object+0x7d/0xb0
 ? __pfx_timerlat_irq+0x10/0x10
 __debug_object_init+0x110/0x150
 hrtimer_init+0x1d/0x60
 timerlat_main+0xab/0x2d0
 ? __pfx_timerlat_main+0x10/0x10
 kthread+0xb7/0xe0
 ? __pfx_kthread+0x10/0x10
 ret_from_fork+0x2d/0x40
 ? __pfx_kthread+0x10/0x10
 ret_from_fork_asm+0x1a/0x30
 </TASK>
```

After tracing the scheduling event, it was discovered that the migration of the "timerlat/1" thread was performed during thread creation. Further analysis confirmed that it is because the CPU online processing for osnoise is implemented through workers, which is asynchronous with the offline processing. When the worker was scheduled to create a thread, the CPU may has already been removed from the cpu_online_mask during the offline process, resulting in the inability to select the right CPU:

T1                       | T2
[CPUHP_ONLINE]           | cpu_device_down()
osnoise_hotplug_workfn() |
                         |     cpus_write_lock()
                         |     takedown_cpu(1)
                         |     cpus_write_unlock()
[CPUHP_OFFLINE]          |
    cpus_read_lock()     |
    start_kthread(1)     |
    cpus_read_unlock()   |

To fix this, skip online processing if the CPU is already offline.

---
 kernel/trace/trace_osnoise.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -2094,6 +2094,8 @@ static void osnoise_hotplug_workfn(struc
 	mutex_lock(&interface_lock);
 	cpus_read_lock();
 
+	if (!cpu_online(cpu))
+		goto out_unlock;
 	if (!cpumask_test_cpu(cpu, &osnoise_cpumask))
 		goto out_unlock;
 

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wei Li <[email protected]>

commit 0bb0a5c12ecf36ad561542bbb95f96355e036a02 upstream.

osnoise_hotplug_workfn() is the asynchronous online callback for "trace/osnoise:online". It may be congested when a CPU goes online and offline repeatedly and is invoked for multiple times after a certain online.

This will lead to kthread leak and timer corruption. Add a check in start_kthread() to prevent this situation.

---
 kernel/trace/trace_osnoise.c |    7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

@@ -2006,6 +2006,10 @@ static int start_kthread(unsigned int cp
 	void *main = osnoise_main;
 	char comm[24];
 
+	/* Do not start a new thread if it is already running */
+	if (per_cpu(per_cpu_osnoise_var, cpu).kthread)
+		return 0;
+
 	if (timerlat_enabled()) {
 		snprintf(comm, 24, "timerlat/%d", cpu);
 		main = timerlat_main;

@@ -2060,11 +2064,10 @@ static int start_per_cpu_kthreads(void)
 		if (cpumask_test_and_clear_cpu(cpu, &kthread_cpumask)) {
 			struct task_struct *kthread;
 
-			kthread = per_cpu(per_cpu_osnoise_var, cpu).kthread;
+			kthread = xchg_relaxed(&(per_cpu(per_cpu_osnoise_var, cpu).kthread), NULL);
 			if (!WARN_ON(!kthread))
 				kthread_stop(kthread);
 		}
-		per_cpu(per_cpu_osnoise_var, cpu).kthread = NULL;
 	}
 
 	for_each_cpu(cpu, current_mask) {

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eder Zulian <[email protected]>

commit 3d7b8ea7a8a20a45d019382c4dc6ed79e8bb95cf upstream.

The help text in osnoise top and timerlat top had some minor errors and omissions. The -d option was missing the 's' (second) abbreviation and the error message for '-d' used '-D'.

---
 tools/tracing/rtla/src/osnoise_top.c  |    2 +-
 tools/tracing/rtla/src/timerlat_top.c |    4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

@@ -434,7 +434,7 @@ struct osnoise_top_params *osnoise_top_p
 		case 'd':
 			params->duration = parse_seconds_duration(optarg);
 			if (!params->duration)
-				osnoise_top_usage(params, "Invalid -D duration\n");
+				osnoise_top_usage(params, "Invalid -d duration\n");
 			break;
 		case 'e':
 			tevent = trace_event_alloc(optarg);

@@ -459,7 +459,7 @@ static void timerlat_top_usage(char *usa
 		"	  -c/--cpus cpus: run the tracer only on the given cpus",
 		"	  -H/--house-keeping cpus: run rtla control threads only on the given cpus",
 		"	  -C/--cgroup[=cgroup_name]: set cgroup, if no cgroup_name is passed, the rtla's cgroup will be inherited",
-		"	  -d/--duration time[m|h|d]: duration of the session in seconds",
+		"	  -d/--duration time[s|m|h|d]: duration of the session",
 		"	  -D/--debug: print debug info",
 		"	     --dump-tasks: prints the task running on all CPUs if stop conditions are met (depends on !--no-aa)",
 		"	  -t/--trace[file]: save the stopped trace to [file|timerlat_trace.txt]",

@@ -613,7 +613,7 @@ static struct timerlat_top_params
 		case 'd':
 			params->duration = parse_seconds_duration(optarg);
 			if (!params->duration)
-				timerlat_top_usage("Invalid -D duration\n");
+				timerlat_top_usage("Invalid -d duration\n");
 			break;
 		case 'e':
 			tevent = trace_event_alloc(optarg);

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Thomas Zimmermann <[email protected]>

commit ad604f0a4c040dcb8faf44dc72db25e457c28076 upstream.

The sysfb framebuffer handling only operates on graphics devices that provide the system's firmware framebuffer. If that device is not known, assume that any graphics device has been initialized by firmware.

Fixes a problem on i915 where sysfb does not release the firmware framebuffer after the native graphics driver loaded.

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

index 02a07d3d0d40..a3df782fa687 100644
@@ -67,9 +67,11 @@ static bool sysfb_unregister(void)
 void sysfb_disable(struct device *dev)
 {
 	struct screen_info *si = &screen_info;
+	struct device *parent;
 
 	mutex_lock(&disable_lock);
-	if (!dev || dev == sysfb_parent_dev(si)) {
+	parent = sysfb_parent_dev(si);
+	if (!dev || !parent || dev == parent) {
 		sysfb_unregister();
 		disabled = true;
 	}

2.46.2

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Al Viro <[email protected]>

commit 678379e1d4f7443b170939525d3312cfc37bf86b upstream.

Cloning a descriptor table picks the size that would cover all currently opened files. That's fine for clone() and unshare(), but for close_range() there's an additional twist - we clone before we close, and it would be a shame to have
close_range(3, ~0U, CLOSE_RANGE_UNSHARE)
leave us with a huge descriptor table when we are not going to keep anything past stderr, just because some large file descriptor used to be open before our call has taken it out.

Unfortunately, it had been dealt with in an inherently racy way - sane_fdtable_size() gets a "don't copy anything past that" argument (passed via unshare_fd() and dup_fd()), close_range() decides how much should be trimmed and passes that to unshare_fd().

The problem is, a range that used to extend to the end of descriptor table back when close_range() had looked at it might very well have stuff grown after it by the time dup_fd() has allocated a new files_struct and started to figure out the capacity of fdtable to be attached to that.

That leads to interesting pathological cases; at the very least it's a QoI issue, since unshare(CLONE_FILES) is atomic in a sense that it takes a snapshot of descriptor table one might have observed at some point. Since CLOSE_RANGE_UNSHARE close_range() is supposed to be a combination of unshare(CLONE_FILES) with plain close_range(), ending up with a weird state that would never occur with unshare(2) is confusing, to put it mildly.

It's not hard to get rid of - all it takes is passing both ends of the range down to sane_fdtable_size(). There we are under ->files_lock, so the race is trivially avoided.

So we do the following:
* switch close_files() from calling unshare_fd() to calling
dup_fd().
* undo the calling convention change done to unshare_fd() in
60997c3d45d9 "close_range: add CLOSE_RANGE_UNSHARE"
* introduce struct fd_range, pass a pointer to that to dup_fd()
and sane_fdtable_size() instead of "trim everything past that point" they are currently getting. NULL means "we are not going to be punching any holes"; NR_OPEN_MAX is gone.
* make sane_fdtable_size() use find_last_bit() instead of
open-coding it; it's easier to follow that way.
* while we are at it, have dup_fd() report errors by returning
ERR_PTR(), no need to use a separate int *errorp argument.

---
 fs/file.c               |   93 +++++++++++++++++-------------------------------
 include/linux/fdtable.h |    8 ++--
 kernel/fork.c           |   32 +++++++---------
 3 files changed, 51 insertions(+), 82 deletions(-)

@@ -272,59 +272,45 @@ static inline bool fd_is_open(unsigned i
 	return test_bit(fd, fdt->open_fds);
 }
 
-static unsigned int count_open_files(struct fdtable *fdt)
-{
-	unsigned int size = fdt->max_fds;
-	unsigned int i;
-
-	/* Find the last open fd */
-	for (i = size / BITS_PER_LONG; i > 0; ) {
-		if (fdt->open_fds[--i])
-			break;
-	}
-	i = (i + 1) * BITS_PER_LONG;
-	return i;
-}
-
 /*
  * Note that a sane fdtable size always has to be a multiple of
  * BITS_PER_LONG, since we have bitmaps that are sized by this.
  *
- * 'max_fds' will normally already be properly aligned, but it
- * turns out that in the close_range() -> __close_range() ->
- * unshare_fd() -> dup_fd() -> sane_fdtable_size() we can end
- * up having a 'max_fds' value that isn't already aligned.
- *
- * Rather than make close_range() have to worry about this,
- * just make that BITS_PER_LONG alignment be part of a sane
- * fdtable size. Becuase that's really what it is.
+ * punch_hole is optional - when close_range() is asked to unshare
+ * and close, we don't need to copy descriptors in that range, so
+ * a smaller cloned descriptor table might suffice if the last
+ * currently opened descriptor falls into that range.
  */
-static unsigned int sane_fdtable_size(struct fdtable *fdt, unsigned int max_fds)
+static unsigned int sane_fdtable_size(struct fdtable *fdt, struct fd_range *punch_hole)
 {
-	unsigned int count;
+	unsigned int last = find_last_bit(fdt->open_fds, fdt->max_fds);
 
-	count = count_open_files(fdt);
-	if (max_fds < NR_OPEN_DEFAULT)
-		max_fds = NR_OPEN_DEFAULT;
-	return ALIGN(min(count, max_fds), BITS_PER_LONG);
+	if (last == fdt->max_fds)
+		return NR_OPEN_DEFAULT;
+	if (punch_hole && punch_hole->to >= last && punch_hole->from <= last) {
+		last = find_last_bit(fdt->open_fds, punch_hole->from);
+		if (last == punch_hole->from)
+			return NR_OPEN_DEFAULT;
+	}
+	return ALIGN(last + 1, BITS_PER_LONG);
 }
 
 /*
- * Allocate a new files structure and copy contents from the
- * passed in files structure.
- * errorp will be valid only when the returned files_struct is NULL.
+ * Allocate a new descriptor table and copy contents from the passed in
+ * instance.  Returns a pointer to cloned table on success, ERR_PTR()
+ * on failure.  For 'punch_hole' see sane_fdtable_size().
  */
-struct files_struct *dup_fd(struct files_struct *oldf, unsigned int max_fds, int *errorp)
+struct files_struct *dup_fd(struct files_struct *oldf, struct fd_range *punch_hole)
 {
 	struct files_struct *newf;
 	struct file **old_fds, **new_fds;
 	unsigned int open_files, i;
 	struct fdtable *old_fdt, *new_fdt;
+	int error;
 
-	*errorp = -ENOMEM;
 	newf = kmem_cache_alloc(files_cachep, GFP_KERNEL);
 	if (!newf)
-		goto out;
+		return ERR_PTR(-ENOMEM);
 
 	atomic_set(&newf->count, 1);
 

@@ -341,7 +327,7 @@ struct files_struct *dup_fd(struct files
 
 	spin_lock(&oldf->file_lock);
 	old_fdt = files_fdtable(oldf);
-	open_files = sane_fdtable_size(old_fdt, max_fds);
+	open_files = sane_fdtable_size(old_fdt, punch_hole);
 
 	/*
 	 * Check whether we need to allocate a larger fd array and fd set.

@@ -354,14 +340,14 @@ struct files_struct *dup_fd(struct files
 
 		new_fdt = alloc_fdtable(open_files - 1);
 		if (!new_fdt) {
-			*errorp = -ENOMEM;
+			error = -ENOMEM;
 			goto out_release;
 		}
 
 		/* beyond sysctl_nr_open; nothing to do */
 		if (unlikely(new_fdt->max_fds < open_files)) {
 			__free_fdtable(new_fdt);
-			*errorp = -EMFILE;
+			error = -EMFILE;
 			goto out_release;
 		}
 

@@ -372,7 +358,7 @@ struct files_struct *dup_fd(struct files
 		 */
 		spin_lock(&oldf->file_lock);
 		old_fdt = files_fdtable(oldf);
-		open_files = sane_fdtable_size(old_fdt, max_fds);
+		open_files = sane_fdtable_size(old_fdt, punch_hole);
 	}
 
 	copy_fd_bitmaps(new_fdt, old_fdt, open_files / BITS_PER_LONG);

@@ -406,8 +392,7 @@ struct files_struct *dup_fd(struct files
 
 out_release:
 	kmem_cache_free(files_cachep, newf);
-out:
-	return NULL;
+	return ERR_PTR(error);
 }
 
 static struct fdtable *close_files(struct files_struct * files)

@@ -748,37 +733,25 @@ int __close_range(unsigned fd, unsigned
 	if (fd > max_fd)
 		return -EINVAL;
 
-	if (flags & CLOSE_RANGE_UNSHARE) {
-		int ret;
-		unsigned int max_unshare_fds = NR_OPEN_MAX;
+	if ((flags & CLOSE_RANGE_UNSHARE) && atomic_read(&cur_fds->count) > 1) {
+		struct fd_range range = {fd, max_fd}, *punch_hole = ⦥
 
 		/*
 		 * If the caller requested all fds to be made cloexec we always
 		 * copy all of the file descriptors since they still want to
 		 * use them.
 		 */
-		if (!(flags & CLOSE_RANGE_CLOEXEC)) {
-			/*
-			 * If the requested range is greater than the current
-			 * maximum, we're closing everything so only copy all
-			 * file descriptors beneath the lowest file descriptor.
-			 */
-			rcu_read_lock();
-			if (max_fd >= last_fd(files_fdtable(cur_fds)))
-				max_unshare_fds = fd;
-			rcu_read_unlock();
-		}
-
-		ret = unshare_fd(CLONE_FILES, max_unshare_fds, &fds);
-		if (ret)
-			return ret;
+		if (flags & CLOSE_RANGE_CLOEXEC)
+			punch_hole = NULL;
 
+		fds = dup_fd(cur_fds, punch_hole);
+		if (IS_ERR(fds))
+			return PTR_ERR(fds);
 		/*
 		 * We used to share our file descriptor table, and have now
 		 * created a private one, make sure we're using it below.
 		 */
-		if (fds)
-			swap(cur_fds, fds);
+		swap(cur_fds, fds);
 	}
 
 	if (flags & CLOSE_RANGE_CLOEXEC)

@@ -22,7 +22,6 @@
  * as this is the granularity returned by copy_fdset().
  */
 #define NR_OPEN_DEFAULT BITS_PER_LONG
-#define NR_OPEN_MAX ~0U
 
 struct fdtable {
 	unsigned int max_fds;

@@ -106,7 +105,10 @@ struct task_struct;
 
 void put_files_struct(struct files_struct *fs);
 int unshare_files(void);
-struct files_struct *dup_fd(struct files_struct *, unsigned, int *) __latent_entropy;
+struct fd_range {
+	unsigned int from, to;
+};
+struct files_struct *dup_fd(struct files_struct *, struct fd_range *) __latent_entropy;
 void do_close_on_exec(struct files_struct *);
 int iterate_fd(struct files_struct *, unsigned,
 		int (*)(const void *, struct file *, unsigned),

@@ -115,8 +117,6 @@ int iterate_fd(struct files_struct *, un
 extern int close_fd(unsigned int fd);
 extern int __close_range(unsigned int fd, unsigned int max_fd, unsigned int flags);
 extern struct file *file_close_fd(unsigned int fd);
-extern int unshare_fd(unsigned long unshare_flags, unsigned int max_fds,
-		      struct files_struct **new_fdp);
 
 extern struct kmem_cache *files_cachep;
 

@@ -1770,33 +1770,30 @@ static int copy_files(unsigned long clon
 		      int no_files)
 {
 	struct files_struct *oldf, *newf;
-	int error = 0;
 
 	/*
 	 * A background process may not have any files ...
 	 */
 	oldf = current->files;
 	if (!oldf)
-		goto out;
+		return 0;
 
 	if (no_files) {
 		tsk->files = NULL;
-		goto out;
+		return 0;
 	}
 
 	if (clone_flags & CLONE_FILES) {
 		atomic_inc(&oldf->count);
-		goto out;
+		return 0;
 	}
 
-	newf = dup_fd(oldf, NR_OPEN_MAX, &error);
-	if (!newf)
-		goto out;
+	newf = dup_fd(oldf, NULL);
+	if (IS_ERR(newf))
+		return PTR_ERR(newf);
 
 	tsk->files = newf;
-	error = 0;
-out:
-	return error;
+	return 0;
 }
 
 static int copy_sighand(unsigned long clone_flags, struct task_struct *tsk)

@@ -3246,17 +3243,16 @@ static int unshare_fs(unsigned long unsh
 /*
  * Unshare file descriptor table if it is being shared
  */
-int unshare_fd(unsigned long unshare_flags, unsigned int max_fds,
-	       struct files_struct **new_fdp)
+static int unshare_fd(unsigned long unshare_flags, struct files_struct **new_fdp)
 {
 	struct files_struct *fd = current->files;
-	int error = 0;
 
 	if ((unshare_flags & CLONE_FILES) &&
 	    (fd && atomic_read(&fd->count) > 1)) {
-		*new_fdp = dup_fd(fd, max_fds, &error);
-		if (!*new_fdp)
-			return error;
+		fd = dup_fd(fd, NULL);
+		if (IS_ERR(fd))
+			return PTR_ERR(fd);
+		*new_fdp = fd;
 	}
 
 	return 0;

@@ -3314,7 +3310,7 @@ int ksys_unshare(unsigned long unshare_f
 	err = unshare_fs(unshare_flags, &new_fs);
 	if (err)
 		goto bad_unshare_out;
-	err = unshare_fd(unshare_flags, NR_OPEN_MAX, &new_fd);
+	err = unshare_fd(unshare_flags, &new_fd);
 	if (err)
 		goto bad_unshare_cleanup_fs;
 	err = unshare_userns(unshare_flags, &new_cred);

@@ -3406,7 +3402,7 @@ int unshare_files(void)
 	struct files_struct *old, *copy = NULL;
 	int error;
 
-	error = unshare_fd(CLONE_FILES, NR_OPEN_MAX, &copy);
+	error = unshare_fd(CLONE_FILES, &copy);
 	if (error || !copy)
 		return error;
 

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jani Nikula <[email protected]>

commit 394b52462020b6cceff1f7f47fdebd03589574f3 upstream.

CONFIG_DRM_I915_USERFAULT_AUTOSUSPEND is an int, defaulting to 250. When the wakeref is non-zero, it's either -1 or a dynamically allocated pointer, depending on CONFIG_DRM_I915_DEBUG_RUNTIME_PM. It's likely that the code works by coincidence with the bitwise AND, but with CONFIG_DRM_I915_DEBUG_RUNTIME_PM=y, there's the off chance that the condition evaluates to false, and intel_wakeref_auto() doesn't get called. Switch to the intended logical AND.

v2: Use != to avoid clang -Wconstant-logical-operand (Nathan)

---
 drivers/gpu/drm/i915/gem/i915_gem_ttm.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -1131,7 +1131,7 @@ static vm_fault_t vm_fault_ttm(struct vm
 		GEM_WARN_ON(!i915_ttm_cpu_maps_iomem(bo->resource));
 	}
 
-	if (wakeref & CONFIG_DRM_I915_USERFAULT_AUTOSUSPEND)
+	if (wakeref && CONFIG_DRM_I915_USERFAULT_AUTOSUSPEND != 0)
 		intel_wakeref_auto(&to_i915(obj->base.dev)->runtime_pm.userfault_wakeref,
 				   msecs_to_jiffies_timeout(CONFIG_DRM_I915_USERFAULT_AUTOSUSPEND));
 

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Boris Brezillon <[email protected]>

commit f9e7ac6e2e9986c2ee63224992cb5c8276e46b2a upstream.

The only user (the mesa gallium driver) is already assuming explicit synchronization and doing the export/import dance on shared BOs. The only reason we were registering ourselves as writers on external BOs is because Xe, which was the reference back when we developed Panthor, was doing so. Turns out Xe was wrong, and we really want bookkeep on all registered fences, so userspace can explicitly upgrade those to read/write when needed.

---
 drivers/gpu/drm/panthor/panthor_sched.c |    7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

@@ -3436,13 +3436,8 @@ void panthor_job_update_resvs(struct drm
 {
 	struct panthor_job *job = container_of(sched_job, struct panthor_job, base);
 
-	/* Still not sure why we want USAGE_WRITE for external objects, since I
-	 * was assuming this would be handled through explicit syncs being imported
-	 * to external BOs with DMA_BUF_IOCTL_IMPORT_SYNC_FILE, but other drivers
-	 * seem to pass DMA_RESV_USAGE_WRITE, so there must be a good reason.
-	 */
 	panthor_vm_update_resvs(job->group->vm, exec, &sched_job->s_fence->finished,
-				DMA_RESV_USAGE_BOOKKEEP, DMA_RESV_USAGE_WRITE);
+				DMA_RESV_USAGE_BOOKKEEP, DMA_RESV_USAGE_BOOKKEEP);
 }
 
 void panthor_sched_unplug(struct panthor_device *ptdev)

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Boris Brezillon <[email protected]>

commit 282864cc5d3f144af0cdea1868ee2dc2c5110f0d upstream.

The group variable can't be used to retrieve ptdev in our second loop, because it points to the previously iterated list_head, not a valid group. Get the ptdev object from the scheduler instead.

---
 drivers/gpu/drm/panthor/panthor_sched.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

index a8a939a9fb51..145d983bb129 100644
@@ -2046,6 +2046,7 @@ static void
 tick_ctx_cleanup(struct panthor_scheduler *sched,
 		 struct panthor_sched_tick_ctx *ctx)
 {
+	struct panthor_device *ptdev = sched->ptdev;
 	struct panthor_group *group, *tmp;
 	u32 i;
 

@@ -2054,7 +2055,7 @@ tick_ctx_cleanup(struct panthor_scheduler *sched,
 			/* If everything went fine, we should only have groups
 			 * to be terminated in the old_groups lists.
 			 */
-			drm_WARN_ON(&group->ptdev->base, !ctx->csg_upd_failed_mask &&
+			drm_WARN_ON(&ptdev->base, !ctx->csg_upd_failed_mask &&
 				    group_can_run(group));
 
 			if (!group_can_run(group)) {

@@ -2077,7 +2078,7 @@ tick_ctx_cleanup(struct panthor_scheduler *sched,
 		/* If everything went fine, the groups to schedule lists should
 		 * be empty.
 		 */
-		drm_WARN_ON(&group->ptdev->base,
+		drm_WARN_ON(&ptdev->base,
 			    !ctx->csg_upd_failed_mask && !list_empty(&ctx->groups[i]));
 
 		list_for_each_entry_safe(group, tmp, &ctx->groups[i], run_node) {

2.46.2

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Boris Brezillon <[email protected]>

commit 7a1f30afe97294281a2ba05977688385744f9844 upstream.

If deferred operations are pending, we want to wait for those to land before declaring the queue blocked on a SYNC_WAIT. We need this to deal with the case where the sync object is signalled through a deferred SYNC_{ADD,SET} from the same queue. If we don't do that and the group gets scheduled out before the deferred SYNC_{SET,ADD} is executed, we'll end up with a timeout, because no external SYNC_{SET,ADD} will make the scheduler reconsider the group for execution.

---
 drivers/gpu/drm/panthor/panthor_sched.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

index 145d983bb129..2aff02ba6949 100644
@@ -1103,7 +1103,13 @@ cs_slot_sync_queue_state_locked(struct panthor_device *ptdev, u32 csg_id, u32 cs
 			list_move_tail(&group->wait_node,
 				       &group->ptdev->scheduler->groups.waiting);
 		}
-		group->blocked_queues |= BIT(cs_id);
+
+		/* The queue is only blocked if there's no deferred operation
+		 * pending, which can be checked through the scoreboard status.
+		 */
+		if (!cs_iface->output->status_scoreboards)
+			group->blocked_queues |= BIT(cs_id);
+
 		queue->syncwait.gpu_va = cs_iface->output->status_wait_sync_ptr;
 		queue->syncwait.ref = cs_iface->output->status_wait_sync_value;
 		status_wait_cond = cs_iface->output->status_wait & CS_STATUS_WAIT_SYNC_COND_MASK;

2.46.2

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rob Clark <[email protected]>

commit 440d52b370b03b366fd26ace36bab20552116145 upstream.

Fixes a race condition reported here: https://github.com/AsahiLinux/linux/issues/309#issuecomment-2238968609

The whole premise of lockless access to a single-producer-single- consumer queue is that there is just a single producer and single consumer. That means we can't call drm_sched_can_queue() (which is about queueing more work to the hw, not to the spsc queue) from anywhere other than the consumer (wq).

This call in the producer is just an optimization to avoid scheduling the consuming worker if it cannot yet queue more work to the hw. It is safe to drop this optimization to avoid the race condition.

---
 drivers/gpu/drm/scheduler/sched_entity.c | 4 ++--
 drivers/gpu/drm/scheduler/sched_main.c   | 7 ++-----
 include/drm/gpu_scheduler.h              | 2 +-
 3 files changed, 5 insertions(+), 8 deletions(-)

index 58c8161289fe..567e5ace6d0c 100644
@@ -380,7 +380,7 @@ static void drm_sched_entity_wakeup(struct dma_fence *f,
 		container_of(cb, struct drm_sched_entity, cb);
 
 	drm_sched_entity_clear_dep(f, cb);
-	drm_sched_wakeup(entity->rq->sched, entity);
+	drm_sched_wakeup(entity->rq->sched);
 }
 
 /**

@@ -612,7 +612,7 @@ void drm_sched_entity_push_job(struct drm_sched_job *sched_job)
 		if (drm_sched_policy == DRM_SCHED_POLICY_FIFO)
 			drm_sched_rq_update_fifo(entity, submit_ts);
 
-		drm_sched_wakeup(entity->rq->sched, entity);
+		drm_sched_wakeup(entity->rq->sched);
 	}
 }
 EXPORT_SYMBOL(drm_sched_entity_push_job);

index 7e90c9f95611..a124d5e77b5e 100644
@@ -1022,15 +1022,12 @@ EXPORT_SYMBOL(drm_sched_job_cleanup);
 /**
  * drm_sched_wakeup - Wake up the scheduler if it is ready to queue
  * @sched: scheduler instance
- * @entity: the scheduler entity
  *
  * Wake up the scheduler if we can queue jobs.
  */
-void drm_sched_wakeup(struct drm_gpu_scheduler *sched,
-		      struct drm_sched_entity *entity)
+void drm_sched_wakeup(struct drm_gpu_scheduler *sched)
 {
-	if (drm_sched_can_queue(sched, entity))
-		drm_sched_run_job_queue(sched);
+	drm_sched_run_job_queue(sched);
 }
 
 /**

index 5acc64954a88..e28bc649b5c9 100644
@@ -574,7 +574,7 @@ void drm_sched_entity_modify_sched(struct drm_sched_entity *entity,
 
 void drm_sched_tdr_queue_imm(struct drm_gpu_scheduler *sched);
 void drm_sched_job_cleanup(struct drm_sched_job *job);
-void drm_sched_wakeup(struct drm_gpu_scheduler *sched, struct drm_sched_entity *entity);
+void drm_sched_wakeup(struct drm_gpu_scheduler *sched);
 bool drm_sched_wqueue_ready(struct drm_gpu_scheduler *sched);
 void drm_sched_wqueue_stop(struct drm_gpu_scheduler *sched);
 void drm_sched_wqueue_start(struct drm_gpu_scheduler *sched);

2.46.2

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tvrtko Ursulin <[email protected]>

commit 4286cc2c953983d44d248c9de1c81d3a9643345c upstream.

Without the locking amdgpu currently can race between amdgpu_ctx_set_entity_priority() (via drm_sched_entity_modify_sched()) and drm_sched_job_arm(), leading to the latter accesing potentially inconsitent entity->sched_list and entity->num_sched_list pair.

v2:
* Improve commit message. (Philipp)

---
 drivers/gpu/drm/scheduler/sched_entity.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -133,8 +133,10 @@ void drm_sched_entity_modify_sched(struc
 {
 	WARN_ON(!num_sched_list || !sched_list);
 
+	spin_lock(&entity->rq_lock);
 	entity->sched_list = sched_list;
 	entity->num_sched_list = num_sched_list;
+	spin_unlock(&entity->rq_lock);
 }
 EXPORT_SYMBOL(drm_sched_entity_modify_sched);
 

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tvrtko Ursulin <[email protected]>

commit cbc8764e29c2318229261a679b2aafd0f9072885 upstream.

Since drm_sched_entity_modify_sched() can modify the entities run queue, lets make sure to only dereference the pointer once so both adding and waking up are guaranteed to be consistent.

Alternative of moving the spin_unlock to after the wake up would for now be more problematic since the same lock is taken inside drm_sched_rq_update_fifo().

v2:
* Improve commit message. (Philipp)
* Cache the scheduler pointer directly. (Christian)

---
 drivers/gpu/drm/scheduler/sched_entity.c |   10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

@@ -599,6 +599,9 @@ void drm_sched_entity_push_job(struct dr
 
 	/* first job wakes up scheduler */
 	if (first) {
+		struct drm_gpu_scheduler *sched;
+		struct drm_sched_rq *rq;
+
 		/* Add the entity to the run queue */
 		spin_lock(&entity->rq_lock);
 		if (entity->stopped) {

@@ -608,13 +611,16 @@ void drm_sched_entity_push_job(struct dr
 			return;
 		}
 
-		drm_sched_rq_add_entity(entity->rq, entity);
+		rq = entity->rq;
+		sched = rq->sched;
+
+		drm_sched_rq_add_entity(rq, entity);
 		spin_unlock(&entity->rq_lock);
 
 		if (drm_sched_policy == DRM_SCHED_POLICY_FIFO)
 			drm_sched_rq_update_fifo(entity, submit_ts);
 
-		drm_sched_wakeup(entity->rq->sched);
+		drm_sched_wakeup(sched);
 	}
 }
 EXPORT_SYMBOL(drm_sched_entity_push_job);

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tvrtko Ursulin <[email protected]>

commit 087913e0ba2b3b9d7ccbafb2acf5dab9e35ae1d5 upstream.

Entities run queue can change during drm_sched_entity_push_job() so make sure to update the score consistently.

---
 drivers/gpu/drm/scheduler/sched_entity.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -586,7 +586,6 @@ void drm_sched_entity_push_job(struct dr
 	ktime_t submit_ts;
 
 	trace_drm_sched_job(sched_job, entity);
-	atomic_inc(entity->rq->sched->score);
 	WRITE_ONCE(entity->last_user, current->group_leader);
 
 	/*

@@ -614,6 +613,7 @@ void drm_sched_entity_push_job(struct dr
 		rq = entity->rq;
 		sched = rq->sched;
 
+		atomic_inc(sched->score);
 		drm_sched_rq_add_entity(rq, entity);
 		spin_unlock(&entity->rq_lock);
 

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alex Hung <[email protected]>

commit 05af800704ee7187d9edd461ec90f3679b1c4aba upstream.

[WHY & HOW]
Some eDP panels suffer from flicking when HDR is enabled in KDE. This quirk works around it by skipping VSC that is incompatible with eDP panels.

---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c         |   11 ++++++++++-
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c |    4 ++++
 drivers/gpu/drm/amd/display/dc/dc_types.h                 |    1 +
 3 files changed, 15 insertions(+), 1 deletion(-)

@@ -6389,12 +6389,21 @@ create_stream_for_sink(struct drm_connec
 	if (stream->signal == SIGNAL_TYPE_DISPLAY_PORT ||
 	    stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST ||
 	    stream->signal == SIGNAL_TYPE_EDP) {
+		const struct dc_edid_caps *edid_caps;
+		unsigned int disable_colorimetry = 0;
+
+		if (aconnector->dc_sink) {
+			edid_caps = &aconnector->dc_sink->edid_caps;
+			disable_colorimetry = edid_caps->panel_patch.disable_colorimetry;
+		}
+
 		//
 		// should decide stream support vsc sdp colorimetry capability
 		// before building vsc info packet
 		//
 		stream->use_vsc_sdp_for_colorimetry = stream->link->dpcd_caps.dpcd_rev.raw >= 0x14 &&
-						      stream->link->dpcd_caps.dprx_feature.bits.VSC_SDP_COLORIMETRY_SUPPORTED;
+						      stream->link->dpcd_caps.dprx_feature.bits.VSC_SDP_COLORIMETRY_SUPPORTED &&
+						      !disable_colorimetry;
 
 		if (stream->out_transfer_func.tf == TRANSFER_FUNCTION_GAMMA22)
 			tf = TRANSFER_FUNC_GAMMA_22;

@@ -72,6 +72,10 @@ static void apply_edid_quirks(struct edi
 		DRM_DEBUG_DRIVER("Clearing DPCD 0x317 on monitor with panel id %X\n", panel_id);
 		edid_caps->panel_patch.remove_sink_ext_caps = true;
 		break;
+	case drm_edid_encode_panel_id('S', 'D', 'C', 0x4154):
+		DRM_DEBUG_DRIVER("Disabling VSC on monitor with panel id %X\n", panel_id);
+		edid_caps->panel_patch.disable_colorimetry = true;
+		break;
 	default:
 		return;
 	}

@@ -178,6 +178,7 @@ struct dc_panel_patch {
 	unsigned int skip_avmute;
 	unsigned int mst_start_top_delay;
 	unsigned int remove_sink_ext_caps;
+	unsigned int disable_colorimetry;
 };
 
 struct dc_edid_caps {

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yihan Zhu <[email protected]>

commit 0d5e5e8a0aa49ea2163abf128da3b509a6c58286 upstream.

[WHY & HOW]
Mismatch in DCN35 DML2 cause bw validation failed to acquire unexpected DPP pipe to cause grey screen and system hang. Remove EnhancedPrefetchScheduleAccelerationFinal value override to match HW spec.

---
 drivers/gpu/drm/amd/display/dc/dml2/dml2_policy.c |    1 -
 1 file changed, 1 deletion(-)

@@ -303,7 +303,6 @@ void build_unoptimized_policy_settings(e
 	if (project == dml_project_dcn35 ||
 		project == dml_project_dcn351) {
 		policy->DCCProgrammingAssumesScanDirectionUnknownFinal = false;
-		policy->EnhancedPrefetchScheduleAccelerationFinal = 0;
 		policy->AllowForPStateChangeOrStutterInVBlankFinal = dml_prefetch_support_uclk_fclk_and_stutter_if_possible; /*new*/
 		policy->UseOnlyMaxPrefetchModes = 1;
 	}

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tom Chung <[email protected]>

commit 52d4e3fb3d340447dcdac0e14ff21a764f326907 upstream.

[Why]
Connected with a Thunderbolt monitor and do the suspend and the system may hang while resume.

The TBT monitor HPD will be triggered during the resume procedure and call the drm_client_modeset_probe() while
struct drm_connector connector->dev->master is NULL.

It will mess up the pipe topology after resume.

[How]
Skip the TBT monitor HPD during the resume procedure because we currently will probe the connectors after resume by default.

---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c |    6 ++++++
 1 file changed, 6 insertions(+)

@@ -758,6 +758,12 @@ static void dmub_hpd_callback(struct amd
 		return;
 	}
 
+	/* Skip DMUB HPD IRQ in suspend/resume. We will probe them later. */
+	if (notify->type == DMUB_NOTIFICATION_HPD && adev->in_suspend) {
+		DRM_INFO("Skip DMUB HPD IRQ callback in suspend/resume\n");
+		return;
+	}
+
 	link_index = notify->link_index;
 	link = adev->dm.dc->links[link_index];
 	dev = adev->dm.ddev;

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Uwe Kleine-König <[email protected]>

commit 8b4865cd904650cbed7f2407e653934c621b8127 upstream.

notify_hwp_interrupt() is called via sysvec_thermal() -> smp_thermal_vector() -> intel_thermal_interrupt() in hard irq context. For this reason it must not use a simple spin_lock that sleeps with PREEMPT_RT enabled. So convert it to a raw spinlock.

---
 drivers/cpufreq/intel_pstate.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

index c31914a9876fa..b694e474acece 100644
@@ -1622,7 +1622,7 @@ static void intel_pstate_notify_work(struct work_struct *work)
 	wrmsrl_on_cpu(cpudata->cpu, MSR_HWP_STATUS, 0);
 }
 
-static DEFINE_SPINLOCK(hwp_notify_lock);
+static DEFINE_RAW_SPINLOCK(hwp_notify_lock);
 static cpumask_t hwp_intr_enable_mask;
 
 void notify_hwp_interrupt(void)

@@ -1638,7 +1638,7 @@ void notify_hwp_interrupt(void)
 	if (!(value & 0x01))
 		return;
 
-	spin_lock_irqsave(&hwp_notify_lock, flags);
+	raw_spin_lock_irqsave(&hwp_notify_lock, flags);
 
 	if (!cpumask_test_cpu(this_cpu, &hwp_intr_enable_mask))
 		goto ack_intr;

@@ -1646,13 +1646,13 @@ void notify_hwp_interrupt(void)
 	schedule_delayed_work(&all_cpu_data[this_cpu]->hwp_notify_work,
 			      msecs_to_jiffies(10));
 
-	spin_unlock_irqrestore(&hwp_notify_lock, flags);
+	raw_spin_unlock_irqrestore(&hwp_notify_lock, flags);
 
 	return;
 
 ack_intr:
 	wrmsrl_safe(MSR_HWP_STATUS, 0);
-	spin_unlock_irqrestore(&hwp_notify_lock, flags);
+	raw_spin_unlock_irqrestore(&hwp_notify_lock, flags);
 }
 
 static void intel_pstate_disable_hwp_interrupt(struct cpudata *cpudata)

@@ -1665,9 +1665,9 @@ static void intel_pstate_disable_hwp_interrupt(struct cpudata *cpudata)
 	/* wrmsrl_on_cpu has to be outside spinlock as this can result in IPC */
 	wrmsrl_on_cpu(cpudata->cpu, MSR_HWP_INTERRUPT, 0x00);
 
-	spin_lock_irq(&hwp_notify_lock);
+	raw_spin_lock_irq(&hwp_notify_lock);
 	cancel_work = cpumask_test_and_clear_cpu(cpudata->cpu, &hwp_intr_enable_mask);
-	spin_unlock_irq(&hwp_notify_lock);
+	raw_spin_unlock_irq(&hwp_notify_lock);
 
 	if (cancel_work)
 		cancel_delayed_work_sync(&cpudata->hwp_notify_work);

@@ -1677,10 +1677,10 @@ static void intel_pstate_enable_hwp_interrupt(struct cpudata *cpudata)
 {
 	/* Enable HWP notification interrupt for guaranteed performance change */
 	if (boot_cpu_has(X86_FEATURE_HWP_NOTIFY)) {
-		spin_lock_irq(&hwp_notify_lock);
+		raw_spin_lock_irq(&hwp_notify_lock);
 		INIT_DELAYED_WORK(&cpudata->hwp_notify_work, intel_pstate_notify_work);
 		cpumask_set_cpu(cpudata->cpu, &hwp_intr_enable_mask);
-		spin_unlock_irq(&hwp_notify_lock);
+		raw_spin_unlock_irq(&hwp_notify_lock);
 
 		/* wrmsrl_on_cpu has to be outside spinlock as this can result in IPC */
 		wrmsrl_on_cpu(cpudata->cpu, MSR_HWP_INTERRUPT, 0x01);

2.43.0

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Masahiro Yamada <[email protected]>

[ Upstream commit 984ed20ece1c6c20789ece040cbff3eb1a388fa9 ]

If you enable "Option -> Show Debug Info" and click a link, the program terminates with the following error:

*** buffer overflow detected ***: terminated

The buffer overflow is caused by the following line:

    strcat(data, "$");

The buffer needs one more byte to accommodate the additional character.

---
 scripts/kconfig/qconf.cc | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index c6c42c0f4e5d5..b7fc5aeb78cc0 100644
@@ -1174,7 +1174,7 @@ void ConfigInfoView::clicked(const QUrl &url)
 {
 	QByteArray str = url.toEncoded();
 	const std::size_t count = str.size();
-	char *data = new char[count + 1];
+	char *data = new char[count + 2];  // '$' + '\0'
 	struct symbol **result;
 	struct menu *m = NULL;
 

2.43.0

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mark Rutland <[email protected]>

[ Upstream commit 924725707d80bc2588cefafef76ff3f164d299bc ]

Add cputype definitions for Neoverse-N3. These will be used for errata detection in subsequent patches.

These values can be found in Table A-261 ("MIDR_EL1 bit descriptions") in issue 02 of the Neoverse-N3 TRM, which can be found at:

https://developer.arm.com/documentation/107997/0000/?lang=en

---
 arch/arm64/include/asm/cputype.h | 2 ++
 1 file changed, 2 insertions(+)

index 5a7dfeb8e8eb5..488f8e7513495 100644
@@ -94,6 +94,7 @@
 #define ARM_CPU_PART_NEOVERSE_V3	0xD84
 #define ARM_CPU_PART_CORTEX_X925	0xD85
 #define ARM_CPU_PART_CORTEX_A725	0xD87
+#define ARM_CPU_PART_NEOVERSE_N3	0xD8E
 
 #define APM_CPU_PART_XGENE		0x000
 #define APM_CPU_VAR_POTENZA		0x00

@@ -176,6 +177,7 @@
 #define MIDR_NEOVERSE_V3 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_V3)
 #define MIDR_CORTEX_X925 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X925)
 #define MIDR_CORTEX_A725 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A725)
+#define MIDR_NEOVERSE_N3 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_N3)
 #define MIDR_THUNDERX	MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX)
 #define MIDR_THUNDERX_81XX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX_81XX)
 #define MIDR_THUNDERX_83XX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX_83XX)

2.43.0

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mark Rutland <[email protected]>

[ Upstream commit 081eb7932c2b244f63317a982c5e3990e2c7fbdd ]

A number of Arm Ltd CPUs suffer from errata whereby an MSR to the SSBS special-purpose register does not affect subsequent speculative instructions, permitting speculative store bypassing for a window of time.

We worked around this for a number of CPUs in commits:

* 7187bb7d0b5c7dfa ("arm64: errata: Add workaround for Arm errata 3194386 and 3312417")
* 75b3c43eab594bfb ("arm64: errata: Expand speculative SSBS workaround")
* 145502cac7ea70b5 ("arm64: errata: Expand speculative SSBS workaround (again)")

Since then, a (hopefully final) batch of updates have been published, with two more affected CPUs. For the affected CPUs the existing mitigation is sufficient, as described in their respective Software Developer Errata Notice (SDEN) documents:

* Cortex-A715 (MP148) SDEN v15.0, erratum 3456084
https://developer.arm.com/documentation/SDEN-2148827/1500/

* Neoverse-N3 (MP195) SDEN v5.0, erratum 3456111
https://developer.arm.com/documentation/SDEN-3050973/0500/

Enable the existing mitigation by adding the relevant MIDRs to erratum_spec_ssbs_list, and update silicon-errata.rst and the Kconfig text accordingly.

---
 Documentation/arch/arm64/silicon-errata.rst | 4 ++++
 arch/arm64/Kconfig                          | 2 ++
 arch/arm64/kernel/cpu_errata.c              | 2 ++
 3 files changed, 8 insertions(+)

index 3bc51669ead7d..8cd4f365044b6 100644
@@ -146,6 +146,8 @@ stable kernels.
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Cortex-A715     | #2645198        | ARM64_ERRATUM_2645198       |
 +----------------+-----------------+-----------------+-----------------------------+
+| ARM            | Cortex-A715     | #3456084        | ARM64_ERRATUM_3194386       |
++----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Cortex-A720     | #3456091        | ARM64_ERRATUM_3194386       |
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Cortex-A725     | #3456106        | ARM64_ERRATUM_3194386       |

@@ -186,6 +188,8 @@ stable kernels.
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Neoverse-N2     | #3324339        | ARM64_ERRATUM_3194386       |
 +----------------+-----------------+-----------------+-----------------------------+
+| ARM            | Neoverse-N3     | #3456111        | ARM64_ERRATUM_3194386       |
++----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Neoverse-V1     | #1619801        | N/A                         |
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Neoverse-V1     | #3324341        | ARM64_ERRATUM_3194386       |

index 03faaecfa2444..43d79f87fa180 100644
@@ -1078,6 +1078,7 @@ config ARM64_ERRATUM_3194386
 	  * ARM Cortex-A78C erratum 3324346
 	  * ARM Cortex-A78C erratum 3324347
 	  * ARM Cortex-A710 erratam 3324338
+	  * ARM Cortex-A715 errartum 3456084
 	  * ARM Cortex-A720 erratum 3456091
 	  * ARM Cortex-A725 erratum 3456106
 	  * ARM Cortex-X1 erratum 3324344

@@ -1088,6 +1089,7 @@ config ARM64_ERRATUM_3194386
 	  * ARM Cortex-X925 erratum 3324334
 	  * ARM Neoverse-N1 erratum 3324349
 	  * ARM Neoverse N2 erratum 3324339
+	  * ARM Neoverse-N3 erratum 3456111
 	  * ARM Neoverse-V1 erratum 3324341
 	  * ARM Neoverse V2 erratum 3324336
 	  * ARM Neoverse-V3 erratum 3312417

index aec2867daadc2..a78f247029aec 100644
@@ -439,6 +439,7 @@ static const struct midr_range erratum_spec_ssbs_list[] = {
 	MIDR_ALL_VERSIONS(MIDR_CORTEX_A78),
 	MIDR_ALL_VERSIONS(MIDR_CORTEX_A78C),
 	MIDR_ALL_VERSIONS(MIDR_CORTEX_A710),
+	MIDR_ALL_VERSIONS(MIDR_CORTEX_A715),
 	MIDR_ALL_VERSIONS(MIDR_CORTEX_A720),
 	MIDR_ALL_VERSIONS(MIDR_CORTEX_A725),
 	MIDR_ALL_VERSIONS(MIDR_CORTEX_X1),

@@ -450,6 +451,7 @@ static const struct midr_range erratum_spec_ssbs_list[] = {
 	MIDR_ALL_VERSIONS(MIDR_MICROSOFT_AZURE_COBALT_100),
 	MIDR_ALL_VERSIONS(MIDR_NEOVERSE_N1),
 	MIDR_ALL_VERSIONS(MIDR_NEOVERSE_N2),
+	MIDR_ALL_VERSIONS(MIDR_NEOVERSE_N3),
 	MIDR_ALL_VERSIONS(MIDR_NEOVERSE_V1),
 	MIDR_ALL_VERSIONS(MIDR_NEOVERSE_V2),
 	MIDR_ALL_VERSIONS(MIDR_NEOVERSE_V3),

2.43.0

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Oleg Nesterov <[email protected]>

commit 34820304cc2cd1804ee1f8f3504ec77813d29c8e upstream.

xol_add_vma() maps the uninitialized page allocated by __create_xol_area() into userspace. On some architectures (x86) this memory is readable even without VM_READ, VM_EXEC results in the same pgprot_t as VM_EXEC|VM_READ, although this doesn't really matter, debugger can read this memory anyway.

---
 kernel/events/uprobes.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index 28c678c8daef3..3dd1f14643648 100644
@@ -1491,7 +1491,7 @@ static struct xol_area *__create_xol_area(unsigned long vaddr)
 
 	area->xol_mapping.name = "[uprobes]";
 	area->xol_mapping.pages = area->pages;
-	area->pages[0] = alloc_page(GFP_HIGHUSER);
+	area->pages[0] = alloc_page(GFP_HIGHUSER | __GFP_ZERO);
 	if (!area->pages[0])
 		goto free_bitmap;
 	area->pages[1] = NULL;

2.43.0

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yosry Ahmed <[email protected]>

The z3fold compressed pages allocator is rarely used, most users use zsmalloc. The only disadvantage of zsmalloc in comparison is the dependency on MMU, and zbud is a more common option for !MMU as it was the default zswap allocator for a long time.

Historically, zsmalloc had worse latency than zbud and z3fold but offered better memory savings. This is no longer the case as shown by a simple recent analysis [1]. That analysis showed that z3fold does not have any advantage over zsmalloc or zbud considering both performance and memory usage. In a kernel build test on tmpfs in a limited cgroup, z3fold took 3% more time and used 1.8% more memory. The latency of zswap_load() was 7% higher, and that of zswap_store() was 10% higher. Zsmalloc is better in all metrics.

Moreover, z3fold apparently has latent bugs, which was made noticeable by a recent soft lockup bug report with z3fold [2]. Switching to zsmalloc not only fixed the problem, but also reduced the swap usage from 6~8G to 1~2G. Other users have also reported being bitten by mistakenly enabling z3fold.

Other than hurting users, z3fold is repeatedly causing wasted engineering effort. Apart from investigating the above bug, it came up in multiple development discussions (e.g. [3]) as something we need to handle, when there aren't any legit users (at least not intentionally).

The natural course of action is to deprecate z3fold, and remove in a few cycles if no objections are raised from active users. Next on the list should be zbud, as it offers marginal latency gains at the cost of huge memory waste when compared to zsmalloc. That one will need to wait until zsmalloc does not depend on MMU.

Rename the user-visible config option from CONFIG_Z3FOLD to CONFIG_Z3FOLD_DEPRECATED so that users with CONFIG_Z3FOLD=y get a new prompt with explanation during make oldconfig. Also, remove CONFIG_Z3FOLD=y from defconfigs.

[1]https://lore.kernel.org/lkml/CAJD7tkbRF6od-2x_L8-A1QL3=2Ww13sCj4S3i4bNndqF+3+_Vg@mail.gmail.com/ [2]https://lore.kernel.org/lkml/[email protected]/ [3]https://lore.kernel.org/lkml/CAJD7tkbnmeVugfunffSovJf9FAgy9rhBVt_tx=nxUveLUfqVsA@mail.gmail.com/

---
 arch/loongarch/configs/loongson3_defconfig |  1 -
 arch/powerpc/configs/ppc64_defconfig       |  1 -
 mm/Kconfig                                 | 25 ++++++++++++++++------
 3 files changed, 19 insertions(+), 8 deletions(-)

index b4252c357c8e2..75b366407a60a 100644
@@ -96,7 +96,6 @@ CONFIG_ZPOOL=y
 CONFIG_ZSWAP=y
 CONFIG_ZSWAP_COMPRESSOR_DEFAULT_ZSTD=y
 CONFIG_ZBUD=y
-CONFIG_Z3FOLD=y
 CONFIG_ZSMALLOC=m
 # CONFIG_COMPAT_BRK is not set
 CONFIG_MEMORY_HOTPLUG=y

index 544a65fda77bc..d39284489aa26 100644
@@ -81,7 +81,6 @@ CONFIG_MODULE_SIG_SHA512=y
 CONFIG_PARTITION_ADVANCED=y
 CONFIG_BINFMT_MISC=m
 CONFIG_ZSWAP=y
-CONFIG_Z3FOLD=y
 CONFIG_ZSMALLOC=y
 # CONFIG_SLAB_MERGE_DEFAULT is not set
 CONFIG_SLAB_FREELIST_RANDOM=y

index b4cb45255a541..baf7ce6a888c0 100644
@@ -146,12 +146,15 @@ config ZSWAP_ZPOOL_DEFAULT_ZBUD
 	help
 	  Use the zbud allocator as the default allocator.
 
-config ZSWAP_ZPOOL_DEFAULT_Z3FOLD
-	bool "z3fold"
-	select Z3FOLD
+config ZSWAP_ZPOOL_DEFAULT_Z3FOLD_DEPRECATED
+	bool "z3foldi (DEPRECATED)"
+	select Z3FOLD_DEPRECATED
 	help
 	  Use the z3fold allocator as the default allocator.
 
+	  Deprecated and scheduled for removal in a few cycles,
+	  see CONFIG_Z3FOLD_DEPRECATED.
+
 config ZSWAP_ZPOOL_DEFAULT_ZSMALLOC
 	bool "zsmalloc"
 	select ZSMALLOC

@@ -163,7 +166,7 @@ config ZSWAP_ZPOOL_DEFAULT
        string
        depends on ZSWAP
        default "zbud" if ZSWAP_ZPOOL_DEFAULT_ZBUD
-       default "z3fold" if ZSWAP_ZPOOL_DEFAULT_Z3FOLD
+       default "z3fold" if ZSWAP_ZPOOL_DEFAULT_Z3FOLD_DEPRECATED
        default "zsmalloc" if ZSWAP_ZPOOL_DEFAULT_ZSMALLOC
        default ""
 

@@ -177,15 +180,25 @@ config ZBUD
 	  deterministic reclaim properties that make it preferable to a higher
 	  density approach when reclaim will be used.
 
-config Z3FOLD
-	tristate "3:1 compression allocator (z3fold)"
+config Z3FOLD_DEPRECATED
+	tristate "3:1 compression allocator (z3fold) (DEPRECATED)"
 	depends on ZSWAP
 	help
+	  Deprecated and scheduled for removal in a few cycles. If you have
+	  a good reason for using Z3FOLD over ZSMALLOC, please contact
+	  [email protected] and the zswap maintainers.
+
 	  A special purpose allocator for storing compressed pages.
 	  It is designed to store up to three compressed pages per physical
 	  page. It is a ZBUD derivative so the simplicity and determinism are
 	  still there.
 
+config Z3FOLD
+	tristate
+	default y if Z3FOLD_DEPRECATED=y
+	default m if Z3FOLD_DEPRECATED=m
+	depends on Z3FOLD_DEPRECATED
+
 config ZSMALLOC
 	tristate
 	prompt "N:1 compression allocator (zsmalloc)" if ZSWAP

2.43.0

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mario Limonciello <[email protected]>

[ Upstream commit 87d749a6aab73d8069d0345afaa98297816cb220 ]

The issue with panel power savings compatibility below AMDGPU_DM_DEFAULT_MIN_BACKLIGHT happens at AMDGPU_DM_DEFAULT_MIN_BACKLIGHT as well.

That issue will be fixed separately, so don't prevent the backlight brightness from going that low.

---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index d2554b4a14fe9..83f4ff9e848d7 100644
@@ -4176,7 +4176,7 @@ static void amdgpu_dm_update_backlight_caps(struct amdgpu_display_manager *dm,
 		int spread = caps.max_input_signal - caps.min_input_signal;
 
 		if (caps.max_input_signal > AMDGPU_DM_DEFAULT_MAX_BACKLIGHT ||
-		    caps.min_input_signal < AMDGPU_DM_DEFAULT_MIN_BACKLIGHT ||
+		    caps.min_input_signal < 0 ||
 		    spread > AMDGPU_DM_DEFAULT_MAX_BACKLIGHT ||
 		    spread < AMDGPU_DM_MIN_SPREAD) {
 			DRM_DEBUG_KMS("DM: Invalid backlight caps: min=%d, max=%d\n",

2.43.0

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alexey Dobriyan <[email protected]>

[ Upstream commit 961a2851324561caed579764ffbee3db82b32829 ]

Neither ELF spec not ELF loader require program header to be placed right after ELF header, but build-id code very much assumes such placement:

See

	find_get_page(vma->vm_file->f_mapping, 0);

line and checks against PAGE_SIZE.

Returns errors for now until someone rewrites build-id parser to be more inline with load_elf_binary().

Stable-dep-of: 905415ff3ffb ("lib/buildid: harden build ID parsing logic")

---
 lib/buildid.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

index 7954dd92e36c0..e02b5507418b4 100644
@@ -73,6 +73,13 @@ static int get_build_id_32(const void *page_addr, unsigned char *build_id,
 	Elf32_Phdr *phdr;
 	int i;
 
+	/*
+	 * FIXME
+	 * Neither ELF spec nor ELF loader require that program headers
+	 * start immediately after ELF header.
+	 */
+	if (ehdr->e_phoff != sizeof(Elf32_Ehdr))
+		return -EINVAL;
 	/* only supports phdr that fits in one page */
 	if (ehdr->e_phnum >
 	    (PAGE_SIZE - sizeof(Elf32_Ehdr)) / sizeof(Elf32_Phdr))

@@ -98,6 +105,13 @@ static int get_build_id_64(const void *page_addr, unsigned char *build_id,
 	Elf64_Phdr *phdr;
 	int i;
 
+	/*
+	 * FIXME
+	 * Neither ELF spec nor ELF loader require that program headers
+	 * start immediately after ELF header.
+	 */
+	if (ehdr->e_phoff != sizeof(Elf64_Ehdr))
+		return -EINVAL;
 	/* only supports phdr that fits in one page */
 	if (ehdr->e_phnum >
 	    (PAGE_SIZE - sizeof(Elf64_Ehdr)) / sizeof(Elf64_Phdr))

2.43.0

6.10-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andrii Nakryiko <[email protected]>

[ Upstream commit 905415ff3ffb1d7e5afa62bacabd79776bd24606 ]

Harden build ID parsing logic, adding explicit READ_ONCE() where it's important to have a consistent value read and validated just once.

Also, as pointed out by Andi Kleen, we need to make sure that entire ELF note is within a page bounds, so move the overflow check up and add an extra note_size boundaries validation.

Fixes tag below points to the code that moved this code into lib/buildid.c, and then subsequently was used in perf subsystem, making this code exposed to perf_event_open() users in v5.12+.