Skip to contentSkip to conversation

6.1.142-rc1 review

PATCH In review Greg Kroah-Hartman 5 participants 510 messages Jun 23, 2025 – Jun 24, 2025 Page 4 of 4

Subsystems (4)
lore
Share & tools
lore
author maintainer
#453 lore raw

PATCH 454/508 ksmbd: fix null pointer dereference in destroy_previous_session

In reply to Greg Kroah-Hartman

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

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

From: Namjae Jeon <[email protected]>

commit 7ac5b66acafcc9292fb935d7e03790f2b8b2dc0e upstream.

If client set ->PreviousSessionId on kerberos session setup stage, NULL pointer dereference error will happen. Since sess->user is not set yet, It can pass the user argument as NULL to destroy_previous_session. sess->user will be set in ksmbd_krb5_authenticate(). So this patch move calling destroy_previous_session() after ksmbd_krb5_authenticate().

---
 fs/smb/server/smb2pdu.c |   11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)
fs/smb/server/smb2pdu.c +6 -5
@@ -1610,17 +1610,18 @@ static int krb5_authenticate(struct ksmb
 	out_len = work->response_sz -
 		(le16_to_cpu(rsp->SecurityBufferOffset) + 4);
 
-	/* Check previous session */
-	prev_sess_id = le64_to_cpu(req->PreviousSessionId);
-	if (prev_sess_id && prev_sess_id != sess->id)
-		destroy_previous_session(conn, sess->user, prev_sess_id);
-
 	retval = ksmbd_krb5_authenticate(sess, in_blob, in_len,
 					 out_blob, &out_len);
 	if (retval) {
 		ksmbd_debug(SMB, "krb5 authentication failed\n");
 		return -EINVAL;
 	}
+
+	/* Check previous session */
+	prev_sess_id = le64_to_cpu(req->PreviousSessionId);
+	if (prev_sess_id && prev_sess_id != sess->id)
+		destroy_previous_session(conn, sess->user, prev_sess_id);
+
 	rsp->SecurityBufferLength = cpu_to_le16(out_len);
 
 	if ((conn->sign || server_conf.enforced_signing) ||
author maintainer
#454 lore raw

PATCH 455/508 selinux: fix selinux_xfrm_alloc_user() to set correct ctx_len

In reply to Greg Kroah-Hartman

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

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

From: Stephen Smalley <[email protected]>

commit 86c8db86af43f52f682e53a0f2f0828683be1e52 upstream.

We should count the terminating NUL byte as part of the ctx_len. Otherwise, UBSAN logs a warning:

  UBSAN: array-index-out-of-bounds in security/selinux/xfrm.c:99:14
  index 60 is out of range for type 'char [*]'

The allocation itself is correct so there is no actual out of bounds indexing, just a warning.

---
 security/selinux/xfrm.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
security/selinux/xfrm.c +1 -1
@@ -95,7 +95,7 @@ static int selinux_xfrm_alloc_user(struc
 
 	ctx->ctx_doi = XFRM_SC_DOI_LSM;
 	ctx->ctx_alg = XFRM_SC_ALG_SELINUX;
-	ctx->ctx_len = str_len;
+	ctx->ctx_len = str_len + 1;
 	memcpy(ctx->ctx_str, &uctx[1], str_len);
 	ctx->ctx_str[str_len] = '\0';
 	rc = security_context_to_sid(&selinux_state, ctx->ctx_str, str_len,
author maintainer
#455 lore raw

PATCH 456/508 atm: Revert atm_account_tx() if copy_from_iter_full() fails.

In reply to Greg Kroah-Hartman

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

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

From: Kuniyuki Iwashima <[email protected]>

commit 7851263998d4269125fd6cb3fdbfc7c6db853859 upstream.

In vcc_sendmsg(), we account skb->truesize to sk->sk_wmem_alloc by atm_account_tx().

It is expected to be reverted by atm_pop_raw() later called by vcc->dev->ops->send(vcc, skb).

However, vcc_sendmsg() misses the same revert when copy_from_iter_full() fails, and then we will leak a socket.

Let's factorise the revert part as atm_return_tx() and call it in the failure path.

Note that the corresponding sk_wmem_alloc operation can be found in alloc_tx() as of the blamed commit.

  $ git blame -L:alloc_tx net/atm/common.c c55fa3cccbc2c~
---
 include/linux/atmdev.h |    6 ++++++
 net/atm/common.c       |    1 +
 net/atm/raw.c          |    2 +-
 3 files changed, 8 insertions(+), 1 deletion(-)
include/linux/atmdev.h +6 -0
@@ -249,6 +249,12 @@ static inline void atm_account_tx(struct
 	ATM_SKB(skb)->atm_options = vcc->atm_options;
 }
 
+static inline void atm_return_tx(struct atm_vcc *vcc, struct sk_buff *skb)
+{
+	WARN_ON_ONCE(refcount_sub_and_test(ATM_SKB(skb)->acct_truesize,
+					   &sk_atm(vcc)->sk_wmem_alloc));
+}
+
 static inline void atm_force_charge(struct atm_vcc *vcc,int truesize)
 {
 	atomic_add(truesize, &sk_atm(vcc)->sk_rmem_alloc);
net/atm/common.c +1 -0
@@ -635,6 +635,7 @@ int vcc_sendmsg(struct socket *sock, str
 
 	skb->dev = NULL; /* for paths shared with net_device interfaces */
 	if (!copy_from_iter_full(skb_put(skb, size), size, &m->msg_iter)) {
+		atm_return_tx(vcc, skb);
 		kfree_skb(skb);
 		error = -EFAULT;
 		goto out;
net/atm/raw.c +1 -1
@@ -36,7 +36,7 @@ static void atm_pop_raw(struct atm_vcc *
 
 	pr_debug("(%d) %d -= %d\n",
 		 vcc->vci, sk_wmem_alloc_get(sk), ATM_SKB(skb)->acct_truesize);
-	WARN_ON(refcount_sub_and_test(ATM_SKB(skb)->acct_truesize, &sk->sk_wmem_alloc));
+	atm_return_tx(vcc, skb);
 	dev_kfree_skb_any(skb);
 	sk->sk_write_space(sk);
 }
author maintainer
#456 lore raw

PATCH 457/508 Input: sparcspkr - avoid unannotated fall-through

In reply to Greg Kroah-Hartman

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

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

From: WangYuli <[email protected]>

commit 8b1d858cbd4e1800e9336404ba7892b5a721230d upstream.

Fix follow warnings with clang-21i (and reformat for clarity):
drivers/input/misc/sparcspkr.c:78:3: warning: unannotated fall-through between switch labels [-Wimplicit-fallthrough]

     78 |                 case SND_TONE: break;
        |                 ^
  drivers/input/misc/sparcspkr.c:78:3: note: insert 'break;' to avoid fall-through
     78 |                 case SND_TONE: break;
        |                 ^
        |                 break;
  drivers/input/misc/sparcspkr.c:113:3: warning: unannotated fall-through between switch labels [-Wimplicit-fallthrough]
    113 |                 case SND_TONE: break;
        |                 ^
  drivers/input/misc/sparcspkr.c:113:3: note: insert 'break;' to avoid fall-through
    113 |                 case SND_TONE: break;
        |                 ^
        |                 break;

2 warnings generated.

---
 drivers/input/misc/sparcspkr.c |   22 ++++++++++++++++------
 1 file changed, 16 insertions(+), 6 deletions(-)
drivers/input/misc/sparcspkr.c +16 -6
@@ -74,9 +74,14 @@ static int bbc_spkr_event(struct input_d
 		return -1;
 
 	switch (code) {
-		case SND_BELL: if (value) value = 1000;
-		case SND_TONE: break;
-		default: return -1;
+	case SND_BELL:
+		if (value)
+			value = 1000;
+		break;
+	case SND_TONE:
+		break;
+	default:
+		return -1;
 	}
 
 	if (value > 20 && value < 32767)
@@ -112,9 +117,14 @@ static int grover_spkr_event(struct inpu
 		return -1;
 
 	switch (code) {
-		case SND_BELL: if (value) value = 1000;
-		case SND_TONE: break;
-		default: return -1;
+	case SND_BELL:
+		if (value)
+			value = 1000;
+		break;
+	case SND_TONE:
+		break;
+	default:
+		return -1;
 	}
 
 	if (value > 20 && value < 32767)
author maintainer
#457 lore raw

PATCH 458/508 wifi: cfg80211: init wiphy_work before allocating rfkill fails

In reply to Greg Kroah-Hartman

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

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

From: Edward Adam Davis <[email protected]>

commit fc88dee89d7b63eeb17699393eb659aadf9d9b7c upstream.

syzbort reported a uninitialize wiphy_work_lock in cfg80211_dev_free. [1]

After rfkill allocation fails, the wiphy release process will be performed, which will cause cfg80211_dev_free to access the uninitialized wiphy_work related data.

Move the initialization of wiphy_work to before rfkill initialization to avoid this issue.

[1]
INFO: trying to register non-static key.
The code is fine but needs lockdep annotation, or maybe you didn't initialize this object before use? turning off the locking correctness validator.

CPU: 0 UID: 0 PID: 5935 Comm: syz-executor550 Not tainted 6.14.0-rc6-syzkaller-00103-g4003c9e78778 #0
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014
Call Trace:
 <TASK>
 __dump_stack lib/dump_stack.c:94 [inline]
 dump_stack_lvl+0x116/0x1f0 lib/dump_stack.c:120
 assign_lock_key kernel/locking/lockdep.c:983 [inline]
 register_lock_class+0xc39/0x1240 kernel/locking/lockdep.c:1297
 __lock_acquire+0x135/0x3c40 kernel/locking/lockdep.c:5103
 lock_acquire.part.0+0x11b/0x380 kernel/locking/lockdep.c:5851
 __raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
 _raw_spin_lock_irqsave+0x3a/0x60 kernel/locking/spinlock.c:162
 cfg80211_dev_free+0x30/0x3d0 net/wireless/core.c:1196
 device_release+0xa1/0x240 drivers/base/core.c:2568
 kobject_cleanup lib/kobject.c:689 [inline]
 kobject_release lib/kobject.c:720 [inline]
 kref_put include/linux/kref.h:65 [inline]
 kobject_put+0x1e4/0x5a0 lib/kobject.c:737
 put_device+0x1f/0x30 drivers/base/core.c:3774
 wiphy_free net/wireless/core.c:1224 [inline]
 wiphy_new_nm+0x1c1f/0x2160 net/wireless/core.c:562
 ieee80211_alloc_hw_nm+0x1b7a/0x2260 net/mac80211/main.c:835
 mac80211_hwsim_new_radio+0x1d6/0x54e0 drivers/net/wireless/virtual/mac80211_hwsim.c:5185
 hwsim_new_radio_nl+0xb42/0x12b0 drivers/net/wireless/virtual/mac80211_hwsim.c:6242
 genl_family_rcv_msg_doit+0x202/0x2f0 net/netlink/genetlink.c:1115
 genl_family_rcv_msg net/netlink/genetlink.c:1195 [inline]
 genl_rcv_msg+0x565/0x800 net/netlink/genetlink.c:1210
 netlink_rcv_skb+0x16b/0x440 net/netlink/af_netlink.c:2533
 genl_rcv+0x28/0x40 net/netlink/genetlink.c:1219
 netlink_unicast_kernel net/netlink/af_netlink.c:1312 [inline]
 netlink_unicast+0x53c/0x7f0 net/netlink/af_netlink.c:1338
 netlink_sendmsg+0x8b8/0xd70 net/netlink/af_netlink.c:1882
 sock_sendmsg_nosec net/socket.c:718 [inline]
 __sock_sendmsg net/socket.c:733 [inline]
 ____sys_sendmsg+0xaaf/0xc90 net/socket.c:2573
 ___sys_sendmsg+0x135/0x1e0 net/socket.c:2627
 __sys_sendmsg+0x16e/0x220 net/socket.c:2659
 do_syscall_x64 arch/x86/entry/common.c:52 [inline]
 do_syscall_64+0xcd/0x250 arch/x86/entry/common.c:83
Fixes 72d520476a2f ("wifi: cfg80211: cancel wiphy_work before freeing wiphy")
Reported-by [email protected]

Close: https://syzkaller.appspot.com/bug?extid=aaf0488c83d1d5f4f029

---
 net/wireless/core.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
net/wireless/core.c +3 -3
@@ -546,6 +546,9 @@ use_default_name:
 	INIT_WORK(&rdev->mgmt_registrations_update_wk,
 		  cfg80211_mgmt_registrations_update_wk);
 	spin_lock_init(&rdev->mgmt_registrations_lock);
+	INIT_WORK(&rdev->wiphy_work, cfg80211_wiphy_work);
+	INIT_LIST_HEAD(&rdev->wiphy_work_list);
+	spin_lock_init(&rdev->wiphy_work_lock);
 
 #ifdef CONFIG_CFG80211_DEFAULT_PS
 	rdev->wiphy.flags |= WIPHY_FLAG_PS_ON_BY_DEFAULT;
@@ -563,9 +566,6 @@ use_default_name:
 		return NULL;
 	}
 
-	INIT_WORK(&rdev->wiphy_work, cfg80211_wiphy_work);
-	INIT_LIST_HEAD(&rdev->wiphy_work_list);
-	spin_lock_init(&rdev->wiphy_work_lock);
 	INIT_WORK(&rdev->rfkill_block, cfg80211_rfkill_block_work);
 	INIT_WORK(&rdev->conn_work, cfg80211_conn_work);
 	INIT_WORK(&rdev->event_work, cfg80211_event_work);
author maintainer
#458 lore raw

PATCH 459/508 ALSA: usb-audio: Rename ALSA kcontrol PCM and PCM1 for the KTMicro sound card

In reply to Greg Kroah-Hartman

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

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

From: wangdicheng <[email protected]>

commit 93adf20ff4d6e865e0b974110d3cf2f07c057177 upstream.

PCM1 not in Pulseaudio's control list; standardize control to "Speaker" and "Headphone".

---
 sound/usb/mixer_maps.c |   12 ++++++++++++
 1 file changed, 12 insertions(+)
sound/usb/mixer_maps.c +12 -0
@@ -383,6 +383,13 @@ static const struct usbmix_name_map ms_u
 	{ 0 }   /* terminator */
 };
 
+/* KTMicro USB */
+static struct usbmix_name_map s31b2_0022_map[] = {
+	{ 23, "Speaker Playback" },
+	{ 18, "Headphone Playback" },
+	{ 0 }
+};
+
 /* ASUS ROG Zenith II with Realtek ALC1220-VB */
 static const struct usbmix_name_map asus_zenith_ii_map[] = {
 	{ 19, NULL, 12 }, /* FU, Input Gain Pad - broken response, disabled */
@@ -692,6 +699,11 @@ static const struct usbmix_ctl_map usbmi
 		.id = USB_ID(0x045e, 0x083c),
 		.map = ms_usb_link_map,
 	},
+	{
+		/* KTMicro USB */
+		.id = USB_ID(0X31b2, 0x0022),
+		.map = s31b2_0022_map,
+	},
 	{ 0 } /* terminator */
 };
 
author maintainer
#459 lore raw

PATCH 460/508 ALSA: hda/intel: Add Thinkpad E15 to PM deny list

In reply to Greg Kroah-Hartman

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

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

From: Takashi Iwai <[email protected]>

commit c987a390f1b3b8bdac11031d7004e3410fe259bd upstream.

Lenovo Thinkpad E15 with Conexant CX8070 codec seems causing ugly noises after runtime-PM suspend. Disable the codec runtime PM as a workaround.

---
 sound/pci/hda/hda_intel.c |    2 ++
 1 file changed, 2 insertions(+)
sound/pci/hda/hda_intel.c +2 -0
@@ -2261,6 +2261,8 @@ static const struct snd_pci_quirk power_
 	SND_PCI_QUIRK(0x1734, 0x1232, "KONTRON SinglePC", 0),
 	/* Dell ALC3271 */
 	SND_PCI_QUIRK(0x1028, 0x0962, "Dell ALC3271", 0),
+	/* https://bugzilla.kernel.org/show_bug.cgi?id=220210 */
+	SND_PCI_QUIRK(0x17aa, 0x5079, "Lenovo Thinkpad E15", 0),
 	{}
 };
 #endif /* CONFIG_PM */
author maintainer
#460 lore raw

PATCH 461/508 ALSA: hda/realtek: enable headset mic on Latitude 5420 Rugged

In reply to Greg Kroah-Hartman

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

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

From: Jonathan Lane <[email protected]>

commit efa6bdf1bc75e26cafaa5f1d775e8bb7c5b0c431 upstream.

Like many Dell laptops, the 3.5mm port by default can not detect a combined headphones+mic headset or even a pure microphone. This change enables the port's functionality.

---
 sound/pci/hda/patch_realtek.c |    1 +
 1 file changed, 1 insertion(+)
sound/pci/hda/patch_realtek.c +1 -0
@@ -9809,6 +9809,7 @@ static const struct snd_pci_quirk alc269
 	SND_PCI_QUIRK(0x1028, 0x0871, "Dell Precision 3630", ALC255_FIXUP_DELL_HEADSET_MIC),
 	SND_PCI_QUIRK(0x1028, 0x0872, "Dell Precision 3630", ALC255_FIXUP_DELL_HEADSET_MIC),
 	SND_PCI_QUIRK(0x1028, 0x0873, "Dell Precision 3930", ALC255_FIXUP_DUMMY_LINEOUT_VERB),
+	SND_PCI_QUIRK(0x1028, 0x0879, "Dell Latitude 5420 Rugged", ALC269_FIXUP_DELL4_MIC_NO_PRESENCE),
 	SND_PCI_QUIRK(0x1028, 0x08ad, "Dell WYSE AIO", ALC225_FIXUP_DELL_WYSE_AIO_MIC_NO_PRESENCE),
 	SND_PCI_QUIRK(0x1028, 0x08ae, "Dell WYSE NB", ALC225_FIXUP_DELL1_MIC_NO_PRESENCE),
 	SND_PCI_QUIRK(0x1028, 0x0935, "Dell", ALC274_FIXUP_DELL_AIO_LINEOUT_VERB),
author maintainer
#461 lore raw

PATCH 462/508 iio: accel: fxls8962af: Fix temperature calculation

In reply to Greg Kroah-Hartman

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

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

From: Sean Nyekjaer <[email protected]>

commit 16038474e3a0263572f36326ef85057aaf341814 upstream.

According to spec temperature should be returned in milli degrees Celsius. Add in_temp_scale to calculate from Celsius to milli Celsius.

Fixes a3e0b51884ee ("iio: accel: add support for FXLS8962AF/FXLS8964AF accelerometers")
---
 drivers/iio/accel/fxls8962af-core.c |   14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)
drivers/iio/accel/fxls8962af-core.c +12 -2
@@ -20,6 +20,7 @@
 #include <linux/pm_runtime.h>
 #include <linux/regulator/consumer.h>
 #include <linux/regmap.h>
+#include <linux/units.h>
 
 #include <linux/iio/buffer.h>
 #include <linux/iio/events.h>
@@ -435,8 +436,16 @@ static int fxls8962af_read_raw(struct ii
 		*val = FXLS8962AF_TEMP_CENTER_VAL;
 		return IIO_VAL_INT;
 	case IIO_CHAN_INFO_SCALE:
-		*val = 0;
-		return fxls8962af_read_full_scale(data, val2);
+		switch (chan->type) {
+		case IIO_TEMP:
+			*val = MILLIDEGREE_PER_DEGREE;
+			return IIO_VAL_INT;
+		case IIO_ACCEL:
+			*val = 0;
+			return fxls8962af_read_full_scale(data, val2);
+		default:
+			return -EINVAL;
+		}
 	case IIO_CHAN_INFO_SAMP_FREQ:
 		return fxls8962af_read_samp_freq(data, val, val2);
 	default:
@@ -735,6 +744,7 @@ static const struct iio_event_spec fxls8
 	.type = IIO_TEMP, \
 	.address = FXLS8962AF_TEMP_OUT, \
 	.info_mask_separate = BIT(IIO_CHAN_INFO_RAW) | \
+			      BIT(IIO_CHAN_INFO_SCALE) | \
 			      BIT(IIO_CHAN_INFO_OFFSET),\
 	.scan_index = -1, \
 	.scan_type = { \
author maintainer
#462 lore raw

PATCH 463/508 mm/hugetlb: unshare page tables during VMA split, not before

In reply to Greg Kroah-Hartman

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

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

From: Jann Horn <[email protected]>

commit 081056dc00a27bccb55ccc3c6f230a3d5fd3f7e0 upstream.

Currently, __split_vma() triggers hugetlb page table unsharing through vm_ops->may_split(). This happens before the VMA lock and rmap locks are taken - which is too early, it allows racing VMA-locked page faults in our process and racing rmap walks from other processes to cause page tables to be shared again before we actually perform the split.

Fix it by explicitly calling into the hugetlb unshare logic from __split_vma() in the same place where THP splitting also happens. At that point, both the VMA and the rmap(s) are write-locked.

An annoying detail is that we can now call into the helper hugetlb_unshare_pmds() from two different locking contexts:

1. from hugetlb_split(), holding:
- mmap lock (exclusively)
- VMA lock
- file rmap lock (exclusively)
2. hugetlb_unshare_all_pmds(), which I think is designed to be able to call us with only the mmap lock held (in shared mode), but currently only runs while holding mmap lock (exclusively) and VMA lock

Backporting note:
This commit fixes a racy protection that was introduced in commit b30c14cd6102 ("hugetlb: unshare some PMDs when splitting VMAs"); that commit claimed to fix an issue introduced in 5.13, but it should actually also go all the way back.

Fixes 39dde65c9940 ("[PATCH] shared page table for hugetlb page")
Cc Liam Howlett <[email protected]>
Cc Lorenzo Stoakes <[email protected]>
Cc Vlastimil Babka <[email protected]>
Cc <[email protected]> [b30c14cd6102: hugetlb: unshare some PMDs when splitting VMAs]
[stable backport: code got moved around, VMA splitting is in __vma_adjust]
---
 include/linux/hugetlb.h |    3 ++
 mm/hugetlb.c            |   60 +++++++++++++++++++++++++++++++++++-------------
 mm/mmap.c               |    8 ++++++
 3 files changed, 55 insertions(+), 16 deletions(-)
include/linux/hugetlb.h +3 -0
@@ -239,6 +239,7 @@ unsigned long hugetlb_change_protection(
 
 bool is_hugetlb_entry_migration(pte_t pte);
 void hugetlb_unshare_all_pmds(struct vm_area_struct *vma);
+void hugetlb_split(struct vm_area_struct *vma, unsigned long addr);
 
 #else /* !CONFIG_HUGETLB_PAGE */
 
@@ -472,6 +473,8 @@ static inline vm_fault_t hugetlb_fault(s
 
 static inline void hugetlb_unshare_all_pmds(struct vm_area_struct *vma) { }
 
+static inline void hugetlb_split(struct vm_area_struct *vma, unsigned long addr) {}
+
 #endif /* !CONFIG_HUGETLB_PAGE */
 /*
  * hugepages at page global directory. If arch support
mm/hugetlb.c +44 -16

109 lines not on this page · raw message

mm/mmap.c +8 -0
@@ -727,7 +727,15 @@ int __vma_adjust(struct vm_area_struct *
 		return -ENOMEM;
 	}
 
+	/*
+	 * Get rid of huge pages and shared page tables straddling the split
+	 * boundary.
+	 */
 	vma_adjust_trans_huge(orig_vma, start, end, adjust_next);
+	if (is_vm_hugetlb_page(orig_vma)) {
+		hugetlb_split(orig_vma, start);
+		hugetlb_split(orig_vma, end);
+	}
 	if (file) {
 		mapping = file->f_mapping;
 		root = &mapping->i_mmap;
author maintainer
#463 lore raw

PATCH 464/508 mm: hugetlb: independent PMD page table shared count

In reply to Greg Kroah-Hartman

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

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

From: Liu Shixin <[email protected]>

commit 59d9094df3d79443937add8700b2ef1a866b1081 upstream.

The folio refcount may be increased unexpectly through try_get_folio() by caller such as split_huge_pages. In huge_pmd_unshare(), we use refcount to check whether a pmd page table is shared. The check is incorrect if the refcount is increased by the above caller, and this can cause the page table leaked:

 BUG: Bad page state in process sh  pfn:109324
 page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x66 pfn:0x109324
 flags: 0x17ffff800000000(node=0|zone=2|lastcpupid=0xfffff)
 page_type: f2(table)
 raw: 017ffff800000000 0000000000000000 0000000000000000 0000000000000000
 raw: 0000000000000066 0000000000000000 00000000f2000000 0000000000000000
 page dumped because: nonzero mapcount
 ...
 CPU: 31 UID: 0 PID: 7515 Comm: sh Kdump: loaded Tainted: G    B              6.13.0-rc2master+ #7
 Tainted: [B]=BAD_PAGE
 Hardware name: QEMU KVM Virtual Machine, BIOS 0.0.0 02/06/2015
 Call trace:
  show_stack+0x20/0x38 (C)
  dump_stack_lvl+0x80/0xf8
  dump_stack+0x18/0x28
  bad_page+0x8c/0x130
  free_page_is_bad_report+0xa4/0xb0
  free_unref_page+0x3cc/0x620
  __folio_put+0xf4/0x158
  split_huge_pages_all+0x1e0/0x3e8
  split_huge_pages_write+0x25c/0x2d8
  full_proxy_write+0x64/0xd8
  vfs_write+0xcc/0x280
  ksys_write+0x70/0x110
  __arm64_sys_write+0x24/0x38
  invoke_syscall+0x50/0x120
  el0_svc_common.constprop.0+0xc8/0xf0
  do_el0_svc+0x24/0x38
  el0_svc+0x34/0x128
  el0t_64_sync_handler+0xc8/0xd0
  el0t_64_sync+0x190/0x198

The issue may be triggered by damon, offline_page, page_idle, etc, which will increase the refcount of page table.

1. The page table itself will be discarded after reporting the "nonzero mapcount".

2. The HugeTLB page mapped by the page table miss freeing since we treat the page table as shared and a shared page table will not be unmapped.

Fix it by introducing independent PMD page table shared count. As described by comment, pt_index/pt_mm/pt_frag_refcount are used for s390 gmap, x86 pgds and powerpc, pt_share_count is used for x86/arm64/riscv pmds, so we can reuse the field as pt_share_count.

Fixes 39dde65c9940 ("[PATCH] shared page table for hugetlb page")
Cc Kefeng Wang <[email protected]>
Cc Ken Chen <[email protected]>
Cc Muchun Song <[email protected]>
Cc Nanyong Sun <[email protected]>
Cc Jane Chu <[email protected]>

[backport note: struct ptdesc did not exist yet, stuff it equivalently into struct page instead]

---
 include/linux/mm.h       |    3 +++
 include/linux/mm_types.h |    3 +++
 mm/hugetlb.c             |   16 +++++++---------
 3 files changed, 13 insertions(+), 9 deletions(-)
include/linux/mm.h +3 -0
@@ -2537,6 +2537,9 @@ static inline bool pgtable_pmd_page_ctor
 	if (!pmd_ptlock_init(page))
 		return false;
 	__SetPageTable(page);
+#ifdef CONFIG_ARCH_WANT_HUGE_PMD_SHARE
+	atomic_set(&page->pt_share_count, 0);
+#endif
 	inc_lruvec_page_state(page, NR_PAGETABLE);
 	return true;
 }
include/linux/mm_types.h +3 -0
@@ -160,6 +160,9 @@ struct page {
 			union {
 				struct mm_struct *pt_mm; /* x86 pgds only */
 				atomic_t pt_frag_refcount; /* powerpc */
+#ifdef CONFIG_ARCH_WANT_HUGE_PMD_SHARE
+				atomic_t pt_share_count;
+#endif
 			};
 #if ALLOC_SPLIT_PTLOCKS
 			spinlock_t *ptl;
mm/hugetlb.c +7 -9
@@ -7114,7 +7114,7 @@ pte_t *huge_pmd_share(struct mm_struct *
 			spte = huge_pte_offset(svma->vm_mm, saddr,
 					       vma_mmu_pagesize(svma));
 			if (spte) {
-				get_page(virt_to_page(spte));
+				atomic_inc(&virt_to_page(spte)->pt_share_count);
 				break;
 			}
 		}
@@ -7129,7 +7129,7 @@ pte_t *huge_pmd_share(struct mm_struct *
 				(pmd_t *)((unsigned long)spte & PAGE_MASK));
 		mm_inc_nr_pmds(mm);
 	} else {
-		put_page(virt_to_page(spte));
+		atomic_dec(&virt_to_page(spte)->pt_share_count);
 	}
 	spin_unlock(ptl);
 out:
@@ -7141,10 +7141,6 @@ out:
 /*
  * unmap huge page backed by shared pte.
  *
- * Hugetlb pte page is ref counted at the time of mapping.  If pte is shared
- * indicated by page_count > 1, unmap is achieved by clearing pud and
- * decrementing the ref count. If count == 1, the pte page is not shared.
- *
  * Called with page table lock held.
  *
  * returns: 1 successfully unmapped a shared pte page
@@ -7153,18 +7149,20 @@ out:
 int huge_pmd_unshare(struct mm_struct *mm, struct vm_area_struct *vma,
 					unsigned long addr, pte_t *ptep)
 {
+	unsigned long sz = huge_page_size(hstate_vma(vma));
 	pgd_t *pgd = pgd_offset(mm, addr);
 	p4d_t *p4d = p4d_offset(pgd, addr);
 	pud_t *pud = pud_offset(p4d, addr);
 
 	i_mmap_assert_write_locked(vma->vm_file->f_mapping);
 	hugetlb_vma_assert_locked(vma);
-	BUG_ON(page_count(virt_to_page(ptep)) == 0);
-	if (page_count(virt_to_page(ptep)) == 1)
+	if (sz != PMD_SIZE)
+		return 0;
+	if (!atomic_read(&virt_to_page(ptep)->pt_share_count))
 		return 0;
 
 	pud_clear(pud);
-	put_page(virt_to_page(ptep));
+	atomic_dec(&virt_to_page(ptep)->pt_share_count);
 	mm_dec_nr_pmds(mm);
 	return 1;
 }
author maintainer
#464 lore raw

PATCH 465/508 mm/hugetlb: fix huge_pmd_unshare() vs GUP-fast race

In reply to Greg Kroah-Hartman

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

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

From: Jann Horn <[email protected]>

commit 1013af4f585fccc4d3e5c5824d174de2257f7d6d upstream.

huge_pmd_unshare() drops a reference on a page table that may have previously been shared across processes, potentially turning it into a normal page table used in another process in which unrelated VMAs can afterwards be installed.

If this happens in the middle of a concurrent gup_fast(), gup_fast() could end up walking the page tables of another process. While I don't see any way in which that immediately leads to kernel memory corruption, it is really weird and unexpected.

Fix it with an explicit broadcast IPI through tlb_remove_table_sync_one(), just like we do in khugepaged when removing page tables for a THP collapse.

---
 mm/hugetlb.c |    7 +++++++
 1 file changed, 7 insertions(+)
mm/hugetlb.c +7 -0
@@ -7162,6 +7162,13 @@ int huge_pmd_unshare(struct mm_struct *m
 		return 0;
 
 	pud_clear(pud);
+	/*
+	 * Once our caller drops the rmap lock, some other process might be
+	 * using this page table as a normal, non-hugetlb page table.
+	 * Wait for pending gup_fast() in other threads to finish before letting
+	 * that happen.
+	 */
+	tlb_remove_table_sync_one();
 	atomic_dec(&virt_to_page(ptep)->pt_share_count);
 	mm_dec_nr_pmds(mm);
 	return 1;
author maintainer
#465 lore raw

PATCH 466/508 mm/huge_memory: fix dereferencing invalid pmd migration entry

In reply to Greg Kroah-Hartman

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

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

From: Gavin Guo <[email protected]>

commit be6e843fc51a584672dfd9c4a6a24c8cb81d5fb7 upstream.

When migrating a THP, concurrent access to the PMD migration entry during a deferred split scan can lead to an invalid address access, as illustrated below. To prevent this invalid access, it is necessary to check the PMD migration entry and return early. In this context, there is no need to use pmd_to_swp_entry and pfn_swap_entry_to_page to verify the equality of the target folio. Since the PMD migration entry is locked, it cannot be served as the target.

Mailing list discussion and explanation from Hugh Dickins: "An anon_vma lookup points to a location which may contain the folio of interest, but might instead contain another folio: and weeding out those other folios is precisely what the "folio != pmd_folio((*pmd)" check (and the "risk of replacing the wrong folio" comment a few lines above it) is for."

BUG: unable to handle page fault for address: ffffea60001db008
CPU: 0 UID: 0 PID: 2199114 Comm: tee Not tainted 6.14.0+ #4 NONE
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04/01/2014
RIP: 0010:split_huge_pmd_locked+0x3b5/0x2b60
Call Trace:
<TASK>
try_to_migrate_one+0x28c/0x3730
rmap_walk_anon+0x4f6/0x770
unmap_folio+0x196/0x1f0
split_huge_page_to_list_to_order+0x9f6/0x1560
deferred_split_scan+0xac5/0x12a0
shrinker_debugfs_scan_write+0x376/0x470
full_proxy_write+0x15c/0x220
vfs_write+0x2fc/0xcb0
ksys_write+0x146/0x250
do_syscall_64+0x6a/0x120
entry_SYSCALL_64_after_hwframe+0x76/0x7e

The bug is found by syzkaller on an internal kernel, then confirmed on upstream.

---
 mm/huge_memory.c |   11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)
mm/huge_memory.c +6 -5
@@ -2282,12 +2282,14 @@ void __split_huge_pmd(struct vm_area_str
 {
 	spinlock_t *ptl;
 	struct mmu_notifier_range range;
+	bool pmd_migration;
 
 	mmu_notifier_range_init(&range, MMU_NOTIFY_CLEAR, 0, vma, vma->vm_mm,
 				address & HPAGE_PMD_MASK,
 				(address & HPAGE_PMD_MASK) + HPAGE_PMD_SIZE);
 	mmu_notifier_invalidate_range_start(&range);
 	ptl = pmd_lock(vma->vm_mm, pmd);
+	pmd_migration = is_pmd_migration_entry(*pmd);
 
 	/*
 	 * If caller asks to setup a migration entry, we need a folio to check
@@ -2296,13 +2298,12 @@ void __split_huge_pmd(struct vm_area_str
 	VM_BUG_ON(freeze && !folio);
 	VM_WARN_ON_ONCE(folio && !folio_test_locked(folio));
 
-	if (pmd_trans_huge(*pmd) || pmd_devmap(*pmd) ||
-	    is_pmd_migration_entry(*pmd)) {
+	if (pmd_trans_huge(*pmd) || pmd_devmap(*pmd) || pmd_migration) {
 		/*
-		 * It's safe to call pmd_page when folio is set because it's
-		 * guaranteed that pmd is present.
+		 * Do not apply pmd_folio() to a migration entry; and folio lock
+		 * guarantees that it must be of the wrong folio anyway.
 		 */
-		if (folio && folio != page_folio(pmd_page(*pmd)))
+		if (folio && (pmd_migration || folio != page_folio(pmd_page(*pmd))))
 			goto out;
 		__split_huge_pmd_locked(vma, pmd, range.start, freeze);
 	}
author maintainer
#466 lore raw

PATCH 467/508 net: Fix checksum update for ILA adj-transport

In reply to Greg Kroah-Hartman

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

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

From: Paul Chaignon <[email protected]>

commit 6043b794c7668c19dabc4a93c75b924a19474d59 upstream.

During ILA address translations, the L4 checksums can be handled in different ways. One of them, adj-transport, consist in parsing the transport layer and updating any found checksum. This logic relies on inet_proto_csum_replace_by_diff and produces an incorrect skb->csum when in state CHECKSUM_COMPLETE.

This bug can be reproduced with a simple ILA to SIR mapping, assuming packets are received with CHECKSUM_COMPLETE:

  $ ip a show dev eth0
  14: eth0@if15: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
      link/ether 62:ae:35:9e:0f:8d brd ff:ff:ff:ff:ff:ff link-netnsid 0
      inet6 3333:0:0:1::c078/64 scope global
         valid_lft forever preferred_lft forever
      inet6 fd00:10:244:1::c078/128 scope global nodad
         valid_lft forever preferred_lft forever
      inet6 fe80::60ae:35ff:fe9e:f8d/64 scope link proto kernel_ll
         valid_lft forever preferred_lft forever
  $ ip ila add loc_match fd00:10:244:1 loc 3333:0:0:1 \
      csum-mode adj-transport ident-type luid dev eth0

Then I hit [fd00:10:244:1::c078]:8000 with a server listening only on [3333:0:0:1::c078]:8000. With the bug, the SYN packet is dropped with SKB_DROP_REASON_TCP_CSUM after inet_proto_csum_replace_by_diff changed skb->csum. The translation and drop are visible on pwru [1] traces:

  IFACE   TUPLE                                                        FUNC
  eth0:9  [fd00:10:244:3::3d8]:51420->[fd00:10:244:1::c078]:8000(tcp)  ipv6_rcv
  eth0:9  [fd00:10:244:3::3d8]:51420->[fd00:10:244:1::c078]:8000(tcp)  ip6_rcv_core
  eth0:9  [fd00:10:244:3::3d8]:51420->[fd00:10:244:1::c078]:8000(tcp)  nf_hook_slow
  eth0:9  [fd00:10:244:3::3d8]:51420->[fd00:10:244:1::c078]:8000(tcp)  inet_proto_csum_replace_by_diff
  eth0:9  [fd00:10:244:3::3d8]:51420->[3333:0:0:1::c078]:8000(tcp)     tcp_v6_early_demux
  eth0:9  [fd00:10:244:3::3d8]:51420->[3333:0:0:1::c078]:8000(tcp)     ip6_route_input
  eth0:9  [fd00:10:244:3::3d8]:51420->[3333:0:0:1::c078]:8000(tcp)     ip6_input
  eth0:9  [fd00:10:244:3::3d8]:51420->[3333:0:0:1::c078]:8000(tcp)     ip6_input_finish
  eth0:9  [fd00:10:244:3::3d8]:51420->[3333:0:0:1::c078]:8000(tcp)     ip6_protocol_deliver_rcu
  eth0:9  [fd00:10:244:3::3d8]:51420->[3333:0:0:1::c078]:8000(tcp)     raw6_local_deliver
  eth0:9  [fd00:10:244:3::3d8]:51420->[3333:0:0:1::c078]:8000(tcp)     ipv6_raw_deliver
  eth0:9  [fd00:10:244:3::3d8]:51420->[3333:0:0:1::c078]:8000(tcp)     tcp_v6_rcv
  eth0:9  [fd00:10:244:3::3d8]:51420->[3333:0:0:1::c078]:8000(tcp)     __skb_checksum_complete
  eth0:9  [fd00:10:244:3::3d8]:51420->[3333:0:0:1::c078]:8000(tcp)     kfree_skb_reason(SKB_DROP_REASON_TCP_CSUM)
  eth0:9  [fd00:10:244:3::3d8]:51420->[3333:0:0:1::c078]:8000(tcp)     skb_release_head_state
  eth0:9  [fd00:10:244:3::3d8]:51420->[3333:0:0:1::c078]:8000(tcp)     skb_release_data
  eth0:9  [fd00:10:244:3::3d8]:51420->[3333:0:0:1::c078]:8000(tcp)     skb_free_head
  eth0:9  [fd00:10:244:3::3d8]:51420->[3333:0:0:1::c078]:8000(tcp)     kfree_skbmem

This is happening because inet_proto_csum_replace_by_diff is updating skb->csum when it shouldn't. The L4 checksum is updated such that it "cancels" the IPv6 address change in terms of checksum computation, so the impact on skb->csum is null.

Note this would be different for an IPv4 packet since three fields would be updated: the IPv4 address, the IP checksum, and the L4 checksum. Two would cancel each other and skb->csum would still need to be updated to take the L4 checksum change into account.

This patch fixes it by passing an ipv6 flag to inet_proto_csum_replace_by_diff, to skip the skb->csum update if we're in the IPv6 case. Note the behavior of the only other user of inet_proto_csum_replace_by_diff, the BPF subsystem, is left as is in this patch and fixed in the subsequent patch.

With the fix, using the reproduction from above, I can confirm skb->csum is not touched by inet_proto_csum_replace_by_diff and the TCP SYN proceeds to the application after the ILA translation.

---
 include/net/checksum.h    |    2 +-
 net/core/filter.c         |    2 +-
 net/core/utils.c          |    4 ++--
 net/ipv6/ila/ila_common.c |    6 +++---
 4 files changed, 7 insertions(+), 7 deletions(-)
include/net/checksum.h +1 -1
@@ -156,7 +156,7 @@ void inet_proto_csum_replace16(__sum16 *
 			       const __be32 *from, const __be32 *to,
 			       bool pseudohdr);
 void inet_proto_csum_replace_by_diff(__sum16 *sum, struct sk_buff *skb,
-				     __wsum diff, bool pseudohdr);
+				     __wsum diff, bool pseudohdr, bool ipv6);
 
 static __always_inline
 void inet_proto_csum_replace2(__sum16 *sum, struct sk_buff *skb,
net/core/filter.c +1 -1
@@ -1977,7 +1977,7 @@ BPF_CALL_5(bpf_l4_csum_replace, struct s
 		if (unlikely(from != 0))
 			return -EINVAL;
 
-		inet_proto_csum_replace_by_diff(ptr, skb, to, is_pseudo);
+		inet_proto_csum_replace_by_diff(ptr, skb, to, is_pseudo, false);
 		break;
 	case 2:
 		inet_proto_csum_replace2(ptr, skb, from, to, is_pseudo);
net/core/utils.c +2 -2
@@ -473,11 +473,11 @@ void inet_proto_csum_replace16(__sum16 *
 EXPORT_SYMBOL(inet_proto_csum_replace16);
 
 void inet_proto_csum_replace_by_diff(__sum16 *sum, struct sk_buff *skb,
-				     __wsum diff, bool pseudohdr)
+				     __wsum diff, bool pseudohdr, bool ipv6)
 {
 	if (skb->ip_summed != CHECKSUM_PARTIAL) {
 		csum_replace_by_diff(sum, diff);
-		if (skb->ip_summed == CHECKSUM_COMPLETE && pseudohdr)
+		if (skb->ip_summed == CHECKSUM_COMPLETE && pseudohdr && !ipv6)
 			skb->csum = ~csum_sub(diff, skb->csum);
 	} else if (pseudohdr) {
 		*sum = ~csum_fold(csum_add(diff, csum_unfold(*sum)));
net/ipv6/ila/ila_common.c +3 -3
@@ -86,7 +86,7 @@ static void ila_csum_adjust_transport(st
 
 			diff = get_csum_diff(ip6h, p);
 			inet_proto_csum_replace_by_diff(&th->check, skb,
-							diff, true);
+							diff, true, true);
 		}
 		break;
 	case NEXTHDR_UDP:
@@ -97,7 +97,7 @@ static void ila_csum_adjust_transport(st
 			if (uh->check || skb->ip_summed == CHECKSUM_PARTIAL) {
 				diff = get_csum_diff(ip6h, p);
 				inet_proto_csum_replace_by_diff(&uh->check, skb,
-								diff, true);
+								diff, true, true);
 				if (!uh->check)
 					uh->check = CSUM_MANGLED_0;
 			}
@@ -111,7 +111,7 @@ static void ila_csum_adjust_transport(st
 
 			diff = get_csum_diff(ip6h, p);
 			inet_proto_csum_replace_by_diff(&ih->icmp6_cksum, skb,
-							diff, true);
+							diff, true, true);
 		}
 		break;
 	}
author maintainer
#467 lore raw

PATCH 468/508 bpf: Fix L4 csum update on IPv6 in CHECKSUM_COMPLETE

In reply to Greg Kroah-Hartman

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

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

From: Paul Chaignon <[email protected]>

commit ead7f9b8de65632ef8060b84b0c55049a33cfea1 upstream.

In Cilium, we use bpf_csum_diff + bpf_l4_csum_replace to, among other things, update the L4 checksum after reverse SNATing IPv6 packets. That use case is however not currently supported and leads to invalid skb->csum values in some cases. This patch adds support for IPv6 address changes in bpf_l4_csum_update via a new flag.

When calling bpf_l4_csum_replace in Cilium, it ends up calling inet_proto_csum_replace_by_diff:

    1:  void inet_proto_csum_replace_by_diff(__sum16 *sum, struct sk_buff *skb,
    2:                                       __wsum diff, bool pseudohdr)
    3:  {
    4:      if (skb->ip_summed != CHECKSUM_PARTIAL) {
    5:          csum_replace_by_diff(sum, diff);
    6:          if (skb->ip_summed == CHECKSUM_COMPLETE && pseudohdr)
    7:              skb->csum = ~csum_sub(diff, skb->csum);
    8:      } else if (pseudohdr) {
    9:          *sum = ~csum_fold(csum_add(diff, csum_unfold(*sum)));
    10:     }
    11: }

The bug happens when we're in the CHECKSUM_COMPLETE state. We've just updated one of the IPv6 addresses. The helper now updates the L4 header checksum on line 5. Next, it updates skb->csum on line 7. It shouldn't.

For an IPv6 packet, the updates of the IPv6 address and of the L4 checksum will cancel each other. The checksums are set such that computing a checksum over the packet including its checksum will result in a sum of 0. So the same is true here when we update the L4 checksum on line 5. We'll update it as to cancel the previous IPv6 address update. Hence skb->csum should remain untouched in this case.

The same bug doesn't affect IPv4 packets because, in that case, three fields are updated: the IPv4 address, the IP checksum, and the L4 checksum. The change to the IPv4 address and one of the checksums still cancel each other in skb->csum, but we're left with one checksum update and should therefore update skb->csum accordingly. That's exactly what inet_proto_csum_replace_by_diff does.

This special case for IPv6 L4 checksums is also described atop inet_proto_csum_replace16, the function we should be using in this case.

This patch introduces a new bpf_l4_csum_replace flag, BPF_F_IPV6, to indicate that we're updating the L4 checksum of an IPv6 packet. When the flag is set, inet_proto_csum_replace_by_diff will skip the skb->csum update.

---
 include/uapi/linux/bpf.h       |    2 ++
 net/core/filter.c              |    5 +++--
 tools/include/uapi/linux/bpf.h |    2 ++
 3 files changed, 7 insertions(+), 2 deletions(-)
include/uapi/linux/bpf.h +2 -0
@@ -1787,6 +1787,7 @@ union bpf_attr {
  * 		for updates resulting in a null checksum the value is set to
  * 		**CSUM_MANGLED_0** instead. Flag **BPF_F_PSEUDO_HDR** indicates
  * 		the checksum is to be computed against a pseudo-header.
+ * 		Flag **BPF_F_IPV6** should be set for IPv6 packets.
  *
  * 		This helper works in combination with **bpf_csum_diff**\ (),
  * 		which does not update the checksum in-place, but offers more
@@ -5715,6 +5716,7 @@ enum {
 	BPF_F_PSEUDO_HDR		= (1ULL << 4),
 	BPF_F_MARK_MANGLED_0		= (1ULL << 5),
 	BPF_F_MARK_ENFORCE		= (1ULL << 6),
+	BPF_F_IPV6			= (1ULL << 7),
 };
 
 /* BPF_FUNC_skb_set_tunnel_key and BPF_FUNC_skb_get_tunnel_key flags. */
net/core/filter.c +3 -2
@@ -1958,10 +1958,11 @@ BPF_CALL_5(bpf_l4_csum_replace, struct s
 	bool is_pseudo = flags & BPF_F_PSEUDO_HDR;
 	bool is_mmzero = flags & BPF_F_MARK_MANGLED_0;
 	bool do_mforce = flags & BPF_F_MARK_ENFORCE;
+	bool is_ipv6   = flags & BPF_F_IPV6;
 	__sum16 *ptr;
 
 	if (unlikely(flags & ~(BPF_F_MARK_MANGLED_0 | BPF_F_MARK_ENFORCE |
-			       BPF_F_PSEUDO_HDR | BPF_F_HDR_FIELD_MASK)))
+			       BPF_F_PSEUDO_HDR | BPF_F_HDR_FIELD_MASK | BPF_F_IPV6)))
 		return -EINVAL;
 	if (unlikely(offset > 0xffff || offset & 1))
 		return -EFAULT;
@@ -1977,7 +1978,7 @@ BPF_CALL_5(bpf_l4_csum_replace, struct s
 		if (unlikely(from != 0))
 			return -EINVAL;
 
-		inet_proto_csum_replace_by_diff(ptr, skb, to, is_pseudo, false);
+		inet_proto_csum_replace_by_diff(ptr, skb, to, is_pseudo, is_ipv6);
 		break;
 	case 2:
 		inet_proto_csum_replace2(ptr, skb, from, to, is_pseudo);
tools/include/uapi/linux/bpf.h +2 -0
@@ -1787,6 +1787,7 @@ union bpf_attr {
  * 		for updates resulting in a null checksum the value is set to
  * 		**CSUM_MANGLED_0** instead. Flag **BPF_F_PSEUDO_HDR** indicates
  * 		the checksum is to be computed against a pseudo-header.
+ * 		Flag **BPF_F_IPV6** should be set for IPv6 packets.
  *
  * 		This helper works in combination with **bpf_csum_diff**\ (),
  * 		which does not update the checksum in-place, but offers more
@@ -5715,6 +5716,7 @@ enum {
 	BPF_F_PSEUDO_HDR		= (1ULL << 4),
 	BPF_F_MARK_MANGLED_0		= (1ULL << 5),
 	BPF_F_MARK_ENFORCE		= (1ULL << 6),
+	BPF_F_IPV6			= (1ULL << 7),
 };
 
 /* BPF_FUNC_clone_redirect and BPF_FUNC_redirect flags. */
author maintainer
#468 lore raw

PATCH 469/508 erofs: remove unused trace event erofs_destroy_inode

In reply to Greg Kroah-Hartman

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

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

From: Gao Xiang <[email protected]>

commit 30b58444807c93bffeaba7d776110f2a909d2f9a upstream.

The trace event erofs_destroy_inode was added but remains unused. This unused event contributes approximately 5KB to the kernel module size.

---
 include/trace/events/erofs.h |   18 ------------------
 1 file changed, 18 deletions(-)
include/trace/events/erofs.h +0 -18
@@ -232,24 +232,6 @@ DEFINE_EVENT(erofs__map_blocks_exit, z_e
 	TP_ARGS(inode, map, flags, ret)
 );
 
-TRACE_EVENT(erofs_destroy_inode,
-	TP_PROTO(struct inode *inode),
-
-	TP_ARGS(inode),
-
-	TP_STRUCT__entry(
-		__field(	dev_t,		dev		)
-		__field(	erofs_nid_t,	nid		)
-	),
-
-	TP_fast_assign(
-		__entry->dev	= inode->i_sb->s_dev;
-		__entry->nid	= EROFS_I(inode)->nid;
-	),
-
-	TP_printk("dev = (%d,%d), nid = %llu", show_dev_nid(__entry))
-);
-
 #endif /* _TRACE_EROFS_H */
 
  /* This part must be outside protection */
author maintainer
#469 lore raw

PATCH 470/508 drm/msm/disp: Correct porch timing for SDM845

In reply to Greg Kroah-Hartman

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

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

From: James A. MacInnes <[email protected]>

[ Upstream commit 146e87f3e11de0dfa091ff87e34b4bc6eec761a4 ]

Type-C DisplayPort inoperable due to incorrect porch settings.
- Re-used wide_bus_en as flag to prevent porch shifting

Fixes c943b4948b58 ("drm/msm/dp: add displayPort driver support")

Patchwork: https://patchwork.freedesktop.org/patch/636945/

---
 .../gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c   | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)
drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c +9 -5
index aba2488c32fa1..4b7952c4e639f 100644
@@ -91,17 +91,21 @@ static void drm_mode_to_intf_timing_params(
 		timing->vsync_polarity = 0;
 	}
 
-	/* for DP/EDP, Shift timings to align it to bottom right */
-	if (phys_enc->hw_intf->cap->type == INTF_DP) {
+	timing->wide_bus_en = dpu_encoder_is_widebus_enabled(phys_enc->parent);
+	timing->compression_en = dpu_encoder_is_dsc_enabled(phys_enc->parent);
+
+	/*
+	 *  For DP/EDP, Shift timings to align it to bottom right.
+	 *  wide_bus_en is set for everything excluding SDM845 &
+	 *  porch changes cause DisplayPort failure and HDMI tearing.
+	 */
+	if (phys_enc->hw_intf->cap->type == INTF_DP && timing->wide_bus_en) {
 		timing->h_back_porch += timing->h_front_porch;
 		timing->h_front_porch = 0;
 		timing->v_back_porch += timing->v_front_porch;
 		timing->v_front_porch = 0;
 	}
 
-	timing->wide_bus_en = dpu_encoder_is_widebus_enabled(phys_enc->parent);
-	timing->compression_en = dpu_encoder_is_dsc_enabled(phys_enc->parent);
-
 	/*
 	 * for DP, divide the horizonal parameters by 2 when
 	 * widebus is enabled
Signature · 1 line
2.39.5
author maintainer
#470 lore raw

PATCH 471/508 drm/msm/dsi/dsi_phy_10nm: Fix missing initial VCO rate

In reply to Greg Kroah-Hartman

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

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

From: Krzysztof Kozlowski <[email protected]>

[ Upstream commit 8a48e35becb214743214f5504e726c3ec131cd6d ]

Driver unconditionally saves current state on first init in dsi_pll_10nm_init(), but does not save the VCO rate, only some of the divider registers. The state is then restored during probe/enable via msm_dsi_phy_enable() -> msm_dsi_phy_pll_restore_state() -> dsi_10nm_pll_restore_state().

Restoring calls dsi_pll_10nm_vco_set_rate() with pll_10nm->vco_current_rate=0, which basically overwrites existing rate of VCO and messes with clock hierarchy, by setting frequency to 0 to clock tree. This makes anyway little sense - VCO rate was not saved, so should not be restored.

If PLL was not configured configure it to minimum rate to avoid glitches and configuring entire in clock hierarchy to 0 Hz.

Patchwork: https://patchwork.freedesktop.org/patch/654796/

---
 drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c +7 -0
index 27b592c776a30..1c111969342a7 100644
@@ -716,6 +716,13 @@ static int dsi_pll_10nm_init(struct msm_dsi_phy *phy)
 	/* TODO: Remove this when we have proper display handover support */
 	msm_dsi_phy_pll_save_state(phy);
 
+	/*
+	 * Store also proper vco_current_rate, because its value will be used in
+	 * dsi_10nm_pll_restore_state().
+	 */
+	if (!dsi_pll_10nm_vco_recalc_rate(&pll_10nm->clk_hw, VCO_REF_CLK_RATE))
+		pll_10nm->vco_current_rate = pll_10nm->phy->cfg->min_pll_rate;
+
 	return 0;
 }
 
Signature · 1 line
2.39.5
author maintainer
#471 lore raw

PATCH 472/508 ionic: Prevent driver/fw getting out of sync on devcmd(s)

In reply to Greg Kroah-Hartman

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

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

From: Brett Creeley <[email protected]>

[ Upstream commit 5466491c9e3309ed5c7adbb8fad6e93fcc9a8fe9 ]

Some stress/negative firmware testing around devcmd(s) returning EAGAIN found that the done bit could get out of sync in the firmware when it wasn't cleared in a retry case.

While here, change the type of the local done variable to a bool to match the return type from ionic_dev_cmd_done().

---
 drivers/net/ethernet/pensando/ionic/ionic_main.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/ethernet/pensando/ionic/ionic_main.c +2 -1
index d2038ff316ca5..16f96ed73909c 100644
@@ -461,9 +461,9 @@ static int __ionic_dev_cmd_wait(struct ionic *ionic, unsigned long max_seconds,
 	unsigned long start_time;
 	unsigned long max_wait;
 	unsigned long duration;
-	int done = 0;
 	bool fw_up;
 	int opcode;
+	bool done;
 	int err;
 
 	/* Wait for dev cmd to complete, retrying if we get EAGAIN,
@@ -471,6 +471,7 @@ static int __ionic_dev_cmd_wait(struct ionic *ionic, unsigned long max_seconds,
 	 */
 	max_wait = jiffies + (max_seconds * HZ);
 try_again:
+	done = false;
 	opcode = idev->opcode;
 	start_time = jiffies;
 	for (fw_up = ionic_is_fw_running(idev);
Signature · 1 line
2.39.5
author maintainer
#472 lore raw

PATCH 473/508 drm/nouveau/bl: increase buffer size to avoid truncate warning

In reply to Greg Kroah-Hartman

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

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

From: Jacob Keller <[email protected]>

[ Upstream commit 61b2b3737499f1fb361a54a16828db24a8345e85 ]

The nouveau_get_backlight_name() function generates a unique name for the backlight interface, appending an id from 1 to 99 for all backlight devices after the first.

GCC 15 (and likely other compilers) produce the following -Wformat-truncation warning:

nouveau_backlight.c: In function ‘nouveau_backlight_init’:
nouveau_backlight.c:56:69: error: ‘%d’ directive output may be truncated writing between 1 and 10 bytes into a region of size 3 [-Werror=format-truncation=]
56 | snprintf(backlight_name, BL_NAME_SIZE, "nv_backlight%d", nb);
| ^~
In function ‘nouveau_get_backlight_name’,
inlined from ‘nouveau_backlight_init’ at nouveau_backlight.c:351:7:
nouveau_backlight.c:56:56: note: directive argument in the range [1, 2147483647]
56 | snprintf(backlight_name, BL_NAME_SIZE, "nv_backlight%d", nb);
| ^~~~~~~~~~~~~~~~
nouveau_backlight.c:56:17: note: ‘snprintf’ output between 14 and 23 bytes into a destination of size 15

   56 |                 snprintf(backlight_name, BL_NAME_SIZE, "nv_backlight%d", nb);
      |                 ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

The warning started appearing after commit ab244be47a8f ("drm/nouveau:
Fix a potential theorical leak in nouveau_get_backlight_name()") This fix for the ida usage removed the explicit value check for ids larger than 99. The compiler is unable to intuit that the ida_alloc_max() limits the returned value range between 0 and 99.

Because the compiler can no longer infer that the number ranges from 0 to 99, it thinks that it could use as many as 11 digits (10 + the potential - sign for negative numbers).

The warning has gone unfixed for some time, with at least one kernel test robot report. The code breaks W=1 builds, which is especially frustrating with the introduction of CONFIG_WERROR.

The string is stored temporarily on the stack and then copied into the device name. Its not a big deal to use 11 more bytes of stack rounding out to an even 24 bytes. Increase BL_NAME_SIZE to 24 to avoid the truncation warning. This fixes the W=1 builds that include this driver.

Compile tested only.

---
 drivers/gpu/drm/nouveau/nouveau_backlight.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/nouveau/nouveau_backlight.c +1 -1
index a614582779cae..6469820e46151 100644
@@ -41,7 +41,7 @@
 #include "nouveau_acpi.h"
 
 static struct ida bl_ida;
-#define BL_NAME_SIZE 15 // 12 for name + 2 for digits + 1 for '\0'
+#define BL_NAME_SIZE 24 // 12 for name + 11 for digits + 1 for '\0'
 
 static bool
 nouveau_get_backlight_name(char backlight_name[BL_NAME_SIZE],
Signature · 1 line
2.39.5
author maintainer
#473 lore raw

PATCH 474/508 hwmon: (occ) Rework attribute registration for stack usage

In reply to Greg Kroah-Hartman

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

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

From: Arnd Bergmann <[email protected]>

[ Upstream commit 744c2fe950e936c4d62430de899d6253424200ed ]

clang produces an output with excessive stack usage when building the occ_setup_sensor_attrs() function, apparently the result of having a lot of struct literals and building with the -fno-strict-overflow option that leads clang to skip some optimization in case the 'attr' pointer overruns:

drivers/hwmon/occ/common.c:775:12: error: stack frame size (1392) exceeds limit (1280) in 'occ_setup_sensor_attrs' [-Werror,-Wframe-larger-than]

Replace the custom macros for initializing the attributes with a simpler function call that does not run into this corner case.

Fixes 54076cb3b5ff ("hwmon (occ): Add sensor attributes and register hwmon device")
---
 drivers/hwmon/occ/common.c | 212 +++++++++++++++----------------------
 1 file changed, 85 insertions(+), 127 deletions(-)
drivers/hwmon/occ/common.c +85 -127
index dd690f700d499..256cda99fdc95 100644
@@ -747,29 +747,30 @@ static ssize_t occ_show_extended(struct device *dev,
 }
 
 /*
- * Some helper macros to make it easier to define an occ_attribute. Since these
- * are dynamically allocated, we shouldn't use the existing kernel macros which
+ * A helper to make it easier to define an occ_attribute. Since these
+ * are dynamically allocated, we cannot use the existing kernel macros which
  * stringify the name argument.
  */
-#define ATTR_OCC(_name, _mode, _show, _store) {				\
-	.attr	= {							\
-		.name = _name,						\
-		.mode = VERIFY_OCTAL_PERMISSIONS(_mode),		\
-	},								\
-	.show	= _show,						\
-	.store	= _store,						\
-}
-
-#define SENSOR_ATTR_OCC(_name, _mode, _show, _store, _nr, _index) {	\
-	.dev_attr	= ATTR_OCC(_name, _mode, _show, _store),	\
-	.index		= _index,					\
-	.nr		= _nr,						\
+static void occ_init_attribute(struct occ_attribute *attr, int mode,
+	ssize_t (*show)(struct device *dev, struct device_attribute *attr, char *buf),
+	ssize_t (*store)(struct device *dev, struct device_attribute *attr,
+				   const char *buf, size_t count),
+	int nr, int index, const char *fmt, ...)
+{
+	va_list args;
+
+	va_start(args, fmt);
+	vsnprintf(attr->name, sizeof(attr->name), fmt, args);
+	va_end(args);
+
+	attr->sensor.dev_attr.attr.name = attr->name;
+	attr->sensor.dev_attr.attr.mode = mode;
+	attr->sensor.dev_attr.show = show;
+	attr->sensor.dev_attr.store = store;
+	attr->sensor.index = index;
+	attr->sensor.nr = nr;
 }
 
-#define OCC_INIT_ATTR(_name, _mode, _show, _store, _nr, _index)		\
-	((struct sensor_device_attribute_2)				\
-		SENSOR_ATTR_OCC(_name, _mode, _show, _store, _nr, _index))
-
 /*
  * Allocate and instatiate sensor_device_attribute_2s. It's most efficient to
  * use our own instead of the built-in hwmon attribute types.
@@ -855,14 +856,15 @@ static int occ_setup_sensor_attrs(struct occ *occ)
 		sensors->extended.num_sensors = 0;
 	}
 
-	occ->attrs = devm_kzalloc(dev, sizeof(*occ->attrs) * num_attrs,
+	occ->attrs = devm_kcalloc(dev, num_attrs, sizeof(*occ->attrs),
 				  GFP_KERNEL);
 	if (!occ->attrs)
 		return -ENOMEM;
 
 	/* null-terminated list */
-	occ->group.attrs = devm_kzalloc(dev, sizeof(*occ->group.attrs) *
-					num_attrs + 1, GFP_KERNEL);
+	occ->group.attrs = devm_kcalloc(dev, num_attrs + 1,
+					sizeof(*occ->group.attrs),
+					GFP_KERNEL);
 	if (!occ->group.attrs)
 		return -ENOMEM;
 
@@ -872,43 +874,33 @@ static int occ_setup_sensor_attrs(struct occ *occ)
 		s = i + 1;
 		temp = ((struct temp_sensor_2 *)sensors->temp.data) + i;
 
-		snprintf(attr->name, sizeof(attr->name), "temp%d_label", s);
-		attr->sensor = OCC_INIT_ATTR(attr->name, 0444, show_temp, NULL,
-					     0, i);
+		occ_init_attribute(attr, 0444, show_temp, NULL,
+				   0, i, "temp%d_label", s);
 		attr++;
 
 		if (sensors->temp.version == 2 &&
 		    temp->fru_type == OCC_FRU_TYPE_VRM) {
-			snprintf(attr->name, sizeof(attr->name),
-				 "temp%d_alarm", s);
+			occ_init_attribute(attr, 0444, show_temp, NULL,
+					   1, i, "temp%d_alarm", s);
 		} else {
-			snprintf(attr->name, sizeof(attr->name),
-				 "temp%d_input", s);
+			occ_init_attribute(attr, 0444, show_temp, NULL,
+					   1, i, "temp%d_input", s);
 		}
 
-		attr->sensor = OCC_INIT_ATTR(attr->name, 0444, show_temp, NULL,
-					     1, i);
 		attr++;
 
 		if (sensors->temp.version > 1) {
-			snprintf(attr->name, sizeof(attr->name),
-				 "temp%d_fru_type", s);
-			attr->sensor = OCC_INIT_ATTR(attr->name, 0444,
-						     show_temp, NULL, 2, i);
+			occ_init_attribute(attr, 0444, show_temp, NULL,
+					   2, i, "temp%d_fru_type", s);
 			attr++;
 
-			snprintf(attr->name, sizeof(attr->name),
-				 "temp%d_fault", s);
-			attr->sensor = OCC_INIT_ATTR(attr->name, 0444,
-						     show_temp, NULL, 3, i);
+			occ_init_attribute(attr, 0444, show_temp, NULL,
+					   3, i, "temp%d_fault", s);
 			attr++;
 
 			if (sensors->temp.version == 0x10) {
-				snprintf(attr->name, sizeof(attr->name),
-					 "temp%d_max", s);
-				attr->sensor = OCC_INIT_ATTR(attr->name, 0444,
-							     show_temp, NULL,
-							     4, i);
+				occ_init_attribute(attr, 0444, show_temp, NULL,
+						   4, i, "temp%d_max", s);
 				attr++;
 			}
 		}
@@ -917,14 +909,12 @@ static int occ_setup_sensor_attrs(struct occ *occ)
 	for (i = 0; i < sensors->freq.num_sensors; ++i) {
 		s = i + 1;
 
-		snprintf(attr->name, sizeof(attr->name), "freq%d_label", s);
-		attr->sensor = OCC_INIT_ATTR(attr->name, 0444, show_freq, NULL,
-					     0, i);
+		occ_init_attribute(attr, 0444, show_freq, NULL,
+				   0, i, "freq%d_label", s);
 		attr++;
 
-		snprintf(attr->name, sizeof(attr->name), "freq%d_input", s);
-		attr->sensor = OCC_INIT_ATTR(attr->name, 0444, show_freq, NULL,
-					     1, i);
+		occ_init_attribute(attr, 0444, show_freq, NULL,
+				   1, i, "freq%d_input", s);
 		attr++;
 	}
 
@@ -940,32 +930,24 @@ static int occ_setup_sensor_attrs(struct occ *occ)
 			s = (i * 4) + 1;
 
 			for (j = 0; j < 4; ++j) {
-				snprintf(attr->name, sizeof(attr->name),
-					 "power%d_label", s);
-				attr->sensor = OCC_INIT_ATTR(attr->name, 0444,
-							     show_power, NULL,
-							     nr++, i);
+				occ_init_attribute(attr, 0444, show_power,
+						   NULL, nr++, i,
+						   "power%d_label", s);
 				attr++;
 
-				snprintf(attr->name, sizeof(attr->name),
-					 "power%d_average", s);
-				attr->sensor = OCC_INIT_ATTR(attr->name, 0444,
-							     show_power, NULL,
-							     nr++, i);
+				occ_init_attribute(attr, 0444, show_power,
+						   NULL, nr++, i,
+						   "power%d_average", s);
 				attr++;
 
-				snprintf(attr->name, sizeof(attr->name),
-					 "power%d_average_interval", s);
-				attr->sensor = OCC_INIT_ATTR(attr->name, 0444,
-							     show_power, NULL,
-							     nr++, i);
+				occ_init_attribute(attr, 0444, show_power,
+						   NULL, nr++, i,
+						   "power%d_average_interval", s);
 				attr++;
 
-				snprintf(attr->name, sizeof(attr->name),
-					 "power%d_input", s);
-				attr->sensor = OCC_INIT_ATTR(attr->name, 0444,
-							     show_power, NULL,
-							     nr++, i);
+				occ_init_attribute(attr, 0444, show_power,
+						   NULL, nr++, i,
+						   "power%d_input", s);
 				attr++;
 
 				s++;
@@ -977,28 +959,20 @@ static int occ_setup_sensor_attrs(struct occ *occ)
 		for (i = 0; i < sensors->power.num_sensors; ++i) {
 			s = i + 1;
 
-			snprintf(attr->name, sizeof(attr->name),
-				 "power%d_label", s);
-			attr->sensor = OCC_INIT_ATTR(attr->name, 0444,
-						     show_power, NULL, 0, i);
+			occ_init_attribute(attr, 0444, show_power, NULL,
+					   0, i, "power%d_label", s);
 			attr++;
 
-			snprintf(attr->name, sizeof(attr->name),
-				 "power%d_average", s);
-			attr->sensor = OCC_INIT_ATTR(attr->name, 0444,
-						     show_power, NULL, 1, i);
+			occ_init_attribute(attr, 0444, show_power, NULL,
+					   1, i, "power%d_average", s);
 			attr++;
 
-			snprintf(attr->name, sizeof(attr->name),
-				 "power%d_average_interval", s);
-			attr->sensor = OCC_INIT_ATTR(attr->name, 0444,
-						     show_power, NULL, 2, i);
+			occ_init_attribute(attr, 0444, show_power, NULL,
+					   2, i, "power%d_average_interval", s);
 			attr++;
 
-			snprintf(attr->name, sizeof(attr->name),
-				 "power%d_input", s);
-			attr->sensor = OCC_INIT_ATTR(attr->name, 0444,
-						     show_power, NULL, 3, i);
+			occ_init_attribute(attr, 0444, show_power, NULL,
+					   3, i, "power%d_input", s);
 			attr++;
 		}
 
@@ -1006,56 +980,43 @@ static int occ_setup_sensor_attrs(struct occ *occ)
 	}
 
 	if (sensors->caps.num_sensors >= 1) {
-		snprintf(attr->name, sizeof(attr->name), "power%d_label", s);
-		attr->sensor = OCC_INIT_ATTR(attr->name, 0444, show_caps, NULL,
-					     0, 0);
+		occ_init_attribute(attr, 0444, show_caps, NULL,
+				   0, 0, "power%d_label", s);
 		attr++;
 
-		snprintf(attr->name, sizeof(attr->name), "power%d_cap", s);
-		attr->sensor = OCC_INIT_ATTR(attr->name, 0444, show_caps, NULL,
-					     1, 0);
+		occ_init_attribute(attr, 0444, show_caps, NULL,
+				   1, 0, "power%d_cap", s);
 		attr++;
 
-		snprintf(attr->name, sizeof(attr->name), "power%d_input", s);
-		attr->sensor = OCC_INIT_ATTR(attr->name, 0444, show_caps, NULL,
-					     2, 0);
+		occ_init_attribute(attr, 0444, show_caps, NULL,
+				   2, 0, "power%d_input", s);
 		attr++;
 
-		snprintf(attr->name, sizeof(attr->name),
-			 "power%d_cap_not_redundant", s);
-		attr->sensor = OCC_INIT_ATTR(attr->name, 0444, show_caps, NULL,
-					     3, 0);
+		occ_init_attribute(attr, 0444, show_caps, NULL,
+				   3, 0, "power%d_cap_not_redundant", s);
 		attr++;
 
-		snprintf(attr->name, sizeof(attr->name), "power%d_cap_max", s);
-		attr->sensor = OCC_INIT_ATTR(attr->name, 0444, show_caps, NULL,
-					     4, 0);
+		occ_init_attribute(attr, 0444, show_caps, NULL,
+				   4, 0, "power%d_cap_max", s);
 		attr++;
 
-		snprintf(attr->name, sizeof(attr->name), "power%d_cap_min", s);
-		attr->sensor = OCC_INIT_ATTR(attr->name, 0444, show_caps, NULL,
-					     5, 0);
+		occ_init_attribute(attr, 0444, show_caps, NULL,
+				   5, 0, "power%d_cap_min", s);
 		attr++;
 
-		snprintf(attr->name, sizeof(attr->name), "power%d_cap_user",
-			 s);
-		attr->sensor = OCC_INIT_ATTR(attr->name, 0644, show_caps,
-					     occ_store_caps_user, 6, 0);
+		occ_init_attribute(attr, 0644, show_caps, occ_store_caps_user,
+				   6, 0, "power%d_cap_user", s);
 		attr++;
 
 		if (sensors->caps.version > 1) {
-			snprintf(attr->name, sizeof(attr->name),
-				 "power%d_cap_user_source", s);
-			attr->sensor = OCC_INIT_ATTR(attr->name, 0444,
-						     show_caps, NULL, 7, 0);
+			occ_init_attribute(attr, 0444, show_caps, NULL,
+					   7, 0, "power%d_cap_user_source", s);
 			attr++;
 
 			if (sensors->caps.version > 2) {
-				snprintf(attr->name, sizeof(attr->name),
-					 "power%d_cap_min_soft", s);
-				attr->sensor = OCC_INIT_ATTR(attr->name, 0444,
-							     show_caps, NULL,
-							     8, 0);
+				occ_init_attribute(attr, 0444, show_caps, NULL,
+						   8, 0,
+						   "power%d_cap_min_soft", s);
 				attr++;
 			}
 		}
@@ -1064,19 +1025,16 @@ static int occ_setup_sensor_attrs(struct occ *occ)
 	for (i = 0; i < sensors->extended.num_sensors; ++i) {
 		s = i + 1;
 
-		snprintf(attr->name, sizeof(attr->name), "extn%d_label", s);
-		attr->sensor = OCC_INIT_ATTR(attr->name, 0444,
-					     occ_show_extended, NULL, 0, i);
+		occ_init_attribute(attr, 0444, occ_show_extended, NULL,
+				   0, i, "extn%d_label", s);
 		attr++;
 
-		snprintf(attr->name, sizeof(attr->name), "extn%d_flags", s);
-		attr->sensor = OCC_INIT_ATTR(attr->name, 0444,
-					     occ_show_extended, NULL, 1, i);
+		occ_init_attribute(attr, 0444, occ_show_extended, NULL,
+				   1, i, "extn%d_flags", s);
 		attr++;
 
-		snprintf(attr->name, sizeof(attr->name), "extn%d_input", s);
-		attr->sensor = OCC_INIT_ATTR(attr->name, 0444,
-					     occ_show_extended, NULL, 2, i);
+		occ_init_attribute(attr, 0444, occ_show_extended, NULL,
+				   2, i, "extn%d_input", s);
 		attr++;
 	}
 
Signature · 1 line
2.39.5
author maintainer
#474 lore raw

PATCH 475/508 hwmon: (occ) fix unaligned accesses

In reply to Greg Kroah-Hartman

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

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

From: Arnd Bergmann <[email protected]>

[ Upstream commit 2c021b45c154958566aad0cae9f74ab26a2d5732 ]

Passing a pointer to an unaligned integer as a function argument is undefined behavior:

drivers/hwmon/occ/common.c:492:27: warning: taking address of packed member 'accumulator' of class or structure 'power_sensor_2' may result in an unaligned pointer value [-Waddress-of-packed-member]
492 | val = occ_get_powr_avg(&power->accumulator,

      |                           ^~~~~~~~~~~~~~~~~~
drivers/hwmon/occ/common.c:493:13: warning: taking address of packed member 'update_tag' of class or structure 'power_sensor_2' may result in an unaligned pointer value [-Waddress-of-packed-member]
  493 |            &power->update_tag);
      |             ^~~~~~~~~~~~~~~~~

Move the get_unaligned() calls out of the function and pass these through argument registers instead.

Fixes c10e753d43eb ("hwmon (occ): Add sensor types and versions")
---
 drivers/hwmon/occ/common.c | 28 +++++++++++++---------------
 1 file changed, 13 insertions(+), 15 deletions(-)
drivers/hwmon/occ/common.c +13 -15
index 256cda99fdc95..483f79b394298 100644
@@ -459,12 +459,10 @@ static ssize_t occ_show_power_1(struct device *dev,
 	return sysfs_emit(buf, "%llu\n", val);
 }
 
-static u64 occ_get_powr_avg(u64 *accum, u32 *samples)
+static u64 occ_get_powr_avg(u64 accum, u32 samples)
 {
-	u64 divisor = get_unaligned_be32(samples);
-
-	return (divisor == 0) ? 0 :
-		div64_u64(get_unaligned_be64(accum) * 1000000ULL, divisor);
+	return (samples == 0) ? 0 :
+		mul_u64_u32_div(accum, 1000000UL, samples);
 }
 
 static ssize_t occ_show_power_2(struct device *dev,
@@ -489,8 +487,8 @@ static ssize_t occ_show_power_2(struct device *dev,
 				  get_unaligned_be32(&power->sensor_id),
 				  power->function_id, power->apss_channel);
 	case 1:
-		val = occ_get_powr_avg(&power->accumulator,
-				       &power->update_tag);
+		val = occ_get_powr_avg(get_unaligned_be64(&power->accumulator),
+				       get_unaligned_be32(&power->update_tag));
 		break;
 	case 2:
 		val = (u64)get_unaligned_be32(&power->update_tag) *
@@ -527,8 +525,8 @@ static ssize_t occ_show_power_a0(struct device *dev,
 		return sysfs_emit(buf, "%u_system\n",
 				  get_unaligned_be32(&power->sensor_id));
 	case 1:
-		val = occ_get_powr_avg(&power->system.accumulator,
-				       &power->system.update_tag);
+		val = occ_get_powr_avg(get_unaligned_be64(&power->system.accumulator),
+				       get_unaligned_be32(&power->system.update_tag));
 		break;
 	case 2:
 		val = (u64)get_unaligned_be32(&power->system.update_tag) *
@@ -541,8 +539,8 @@ static ssize_t occ_show_power_a0(struct device *dev,
 		return sysfs_emit(buf, "%u_proc\n",
 				  get_unaligned_be32(&power->sensor_id));
 	case 5:
-		val = occ_get_powr_avg(&power->proc.accumulator,
-				       &power->proc.update_tag);
+		val = occ_get_powr_avg(get_unaligned_be64(&power->proc.accumulator),
+				       get_unaligned_be32(&power->proc.update_tag));
 		break;
 	case 6:
 		val = (u64)get_unaligned_be32(&power->proc.update_tag) *
@@ -555,8 +553,8 @@ static ssize_t occ_show_power_a0(struct device *dev,
 		return sysfs_emit(buf, "%u_vdd\n",
 				  get_unaligned_be32(&power->sensor_id));
 	case 9:
-		val = occ_get_powr_avg(&power->vdd.accumulator,
-				       &power->vdd.update_tag);
+		val = occ_get_powr_avg(get_unaligned_be64(&power->vdd.accumulator),
+				       get_unaligned_be32(&power->vdd.update_tag));
 		break;
 	case 10:
 		val = (u64)get_unaligned_be32(&power->vdd.update_tag) *
@@ -569,8 +567,8 @@ static ssize_t occ_show_power_a0(struct device *dev,
 		return sysfs_emit(buf, "%u_vdn\n",
 				  get_unaligned_be32(&power->sensor_id));
 	case 13:
-		val = occ_get_powr_avg(&power->vdn.accumulator,
-				       &power->vdn.update_tag);
+		val = occ_get_powr_avg(get_unaligned_be64(&power->vdn.accumulator),
+				       get_unaligned_be32(&power->vdn.update_tag));
 		break;
 	case 14:
 		val = (u64)get_unaligned_be32(&power->vdn.update_tag) *
Signature · 1 line
2.39.5
author maintainer
#475 lore raw

PATCH 476/508 pldmfw: Select CRC32 when PLDMFW is selected

In reply to Greg Kroah-Hartman

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

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

From: Simon Horman <[email protected]>

[ Upstream commit 1224b218a4b9203656ecc932152f4c81a97b4fcc ]

pldmfw calls crc32 code and depends on it being enabled, else there is a link error as follows. So PLDMFW should select CRC32.

lib/pldmfw/pldmfw.o: In function `pldmfw_flash_image':
pldmfw.c:(.text+0x70f): undefined reference to `crc32_le_base'

This problem was introduced by commit b8265621f488 ("Add pldmfw library for PLDM firmware update").

It manifests as of commit d69ea414c9b4 ("ice: implement device flash update via devlink").

And is more likely to occur as of commit 9ad19171b6d6 ("lib/crc: remove unnecessary prompt for CONFIG_CRC32 and drop 'default y'").

Found by chance while exercising builds based on tinyconfig.

Fixes b8265621f488 ("Add pldmfw library for PLDM firmware update")
---
 lib/Kconfig | 1 +
 1 file changed, 1 insertion(+)
lib/Kconfig +1 -0
index 9bbf8a4b2108e..0610fad65a366 100644
@@ -744,6 +744,7 @@ config GENERIC_LIB_DEVMEM_IS_ALLOWED
 
 config PLDMFW
 	bool
+	select CRC32
 	default n
 
 config ASN1_ENCODER
Signature · 1 line
2.39.5
author maintainer
#476 lore raw

PATCH 477/508 aoe: clean device rq_list in aoedev_downdev()

In reply to Greg Kroah-Hartman

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

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

From: Justin Sanders <[email protected]>

[ Upstream commit 7f90d45e57cb2ef1f0adcaf925ddffdfc5e680ca ]

An aoe device's rq_list contains accepted block requests that are waiting to be transmitted to the aoe target. This queue was added as part of the conversion to blk_mq. However, the queue was not cleaned out when an aoe device is downed which caused blk_mq_freeze_queue() to sleep indefinitely waiting for those requests to complete, causing a hang. This fix cleans out the queue before calling blk_mq_freeze_queue().

---
 drivers/block/aoe/aoedev.c | 8 ++++++++
 1 file changed, 8 insertions(+)
drivers/block/aoe/aoedev.c +8 -0
index 3523dd82d7a00..280679bde3a50 100644
@@ -198,6 +198,7 @@ aoedev_downdev(struct aoedev *d)
 {
 	struct aoetgt *t, **tt, **te;
 	struct list_head *head, *pos, *nx;
+	struct request *rq, *rqnext;
 	int i;
 
 	d->flags &= ~DEVFL_UP;
@@ -223,6 +224,13 @@ aoedev_downdev(struct aoedev *d)
 	/* clean out the in-process request (if any) */
 	aoe_failip(d);
 
+	/* clean out any queued block requests */
+	list_for_each_entry_safe(rq, rqnext, &d->rq_list, queuelist) {
+		list_del_init(&rq->queuelist);
+		blk_mq_start_request(rq);
+		blk_mq_end_request(rq, BLK_STS_IOERR);
+	}
+
 	/* fast fail all pending I/O */
 	if (d->blkq) {
 		/* UP is cleared, freeze+quiesce to insure all are errored */
Signature · 1 line
2.39.5
author maintainer
#477 lore raw

PATCH 478/508 net: ice: Perform accurate aRFS flow match

In reply to Greg Kroah-Hartman

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

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

From: Krishna Kumar <[email protected]>

[ Upstream commit 5d3bc9e5e725aa36cca9b794e340057feb6880b4 ]

This patch fixes an issue seen in a large-scale deployment under heavy incoming pkts where the aRFS flow wrongly matches a flow and reprograms the NIC with wrong settings. That mis-steering causes RX-path latency spikes and noisy neighbor effects when many connections collide on the same hash (some of our production servers have 20-30K connections).

set_rps_cpu() calls ndo_rx_flow_steer() with flow_id that is calculated by hashing the skb sized by the per rx-queue table size. This results in multiple connections (even across different rx-queues) getting the same hash value. The driver steer function modifies the wrong flow to use this rx-queue, e.g.: Flow#1 is first added:

    Flow#1:  <ip1, port1, ip2, port2>, Hash 'h', q#10

Later when a new flow needs to be added:

	    Flow#2:  <ip3, port3, ip4, port4>, Hash 'h', q#20

The driver finds the hash 'h' from Flow#1 and updates it to use q#20. This results in both flows getting un-optimized - packets for Flow#1 goes to q#20, and then reprogrammed back to q#10 later and so on; and Flow #2 programming is never done as Flow#1 is matched first for all misses. Many flows may wrongly share the same hash and reprogram rules of the original flow each with their own q#.

Tested on two 144-core servers with 16K netperf sessions for 180s. Netperf clients are pinned to cores 0-71 sequentially (so that wrong packets on q#s 72-143 can be measured). IRQs are set 1:1 for queues -> CPUs, enable XPS, enable aRFS (global value is 144 * rps_flow_cnt).

Test notes about results from ice_rx_flow_steer():
---------------------------------------------------
1. "Skip:" counter increments here:

    if (fltr_info->q_index == rxq_idx ||
	arfs_entry->fltr_state != ICE_ARFS_ACTIVE)
	    goto out;
2. "Add:" counter increments here:
    ret = arfs_entry->fltr_info.fltr_id;
    INIT_HLIST_NODE(&arfs_entry->list_entry);
3. "Update:" counter increments here:
    /* update the queue to forward to on an already existing flow */

Runtime comparison: original code vs with the patch for different rps_flow_cnt values.

+-------------------------------+--------------+--------------+
| rps_flow_cnt                  |      512     |    2048      |
+-------------------------------+--------------+--------------+
| Ratio of Pkts on Good:Bad q's | 214 vs 822K  | 1.1M vs 980K |
| Avoid wrong aRFS programming  | 0 vs 310K    | 0 vs 30K     |
| CPU User                      | 216 vs 183   | 216 vs 206   |
| CPU System                    | 1441 vs 1171 | 1447 vs 1320 |
| CPU Softirq                   | 1245 vs 920  | 1238 vs 961  |
| CPU Total                     | 29 vs 22.7   | 29 vs 24.9   |
| aRFS Update                   | 533K vs 59   | 521K vs 32   |
| aRFS Skip                     | 82M vs 77M   | 7.2M vs 4.5M |
+-------------------------------+--------------+--------------+

A separate TCP_STREAM and TCP_RR with 1,4,8,16,64,128,256,512 connections showed no performance degradation.

Some points on the patch/aRFS behavior:
1. Enabling full tuple matching ensures flows are always correctly matched, even with smaller hash sizes.
2. 5-6% drop in CPU utilization as the packets arrive at the correct CPUs and fewer calls to driver for programming on misses.
3. Larger hash tables reduces mis-steering due to more unique flow hashes, but still has clashes. However, with larger per-device rps_flow_cnt, old flows take more time to expire and new aRFS flows cannot be added if h/w limits are reached (rps_may_expire_flow() succeeds when 10*rps_flow_cnt pkts have been processed by this cpu that are not part of the flow).

---
 drivers/net/ethernet/intel/ice/ice_arfs.c | 48 +++++++++++++++++++++++
 1 file changed, 48 insertions(+)
drivers/net/ethernet/intel/ice/ice_arfs.c +48 -0
index 79074f041e39b..e3114c12e1293 100644
@@ -376,6 +376,50 @@ ice_arfs_is_perfect_flow_set(struct ice_hw *hw, __be16 l3_proto, u8 l4_proto)
 	return false;
 }
 
+/**
+ * ice_arfs_cmp - Check if aRFS filter matches this flow.
+ * @fltr_info: filter info of the saved ARFS entry.
+ * @fk: flow dissector keys.
+ * @n_proto:  One of htons(ETH_P_IP) or htons(ETH_P_IPV6).
+ * @ip_proto: One of IPPROTO_TCP or IPPROTO_UDP.
+ *
+ * Since this function assumes limited values for n_proto and ip_proto, it
+ * is meant to be called only from ice_rx_flow_steer().
+ *
+ * Return:
+ * * true	- fltr_info refers to the same flow as fk.
+ * * false	- fltr_info and fk refer to different flows.
+ */
+static bool
+ice_arfs_cmp(const struct ice_fdir_fltr *fltr_info, const struct flow_keys *fk,
+	     __be16 n_proto, u8 ip_proto)
+{
+	/* Determine if the filter is for IPv4 or IPv6 based on flow_type,
+	 * which is one of ICE_FLTR_PTYPE_NONF_IPV{4,6}_{TCP,UDP}.
+	 */
+	bool is_v4 = fltr_info->flow_type == ICE_FLTR_PTYPE_NONF_IPV4_TCP ||
+		     fltr_info->flow_type == ICE_FLTR_PTYPE_NONF_IPV4_UDP;
+
+	/* Following checks are arranged in the quickest and most discriminative
+	 * fields first for early failure.
+	 */
+	if (is_v4)
+		return n_proto == htons(ETH_P_IP) &&
+			fltr_info->ip.v4.src_port == fk->ports.src &&
+			fltr_info->ip.v4.dst_port == fk->ports.dst &&
+			fltr_info->ip.v4.src_ip == fk->addrs.v4addrs.src &&
+			fltr_info->ip.v4.dst_ip == fk->addrs.v4addrs.dst &&
+			fltr_info->ip.v4.proto == ip_proto;
+
+	return fltr_info->ip.v6.src_port == fk->ports.src &&
+		fltr_info->ip.v6.dst_port == fk->ports.dst &&
+		fltr_info->ip.v6.proto == ip_proto &&
+		!memcmp(&fltr_info->ip.v6.src_ip, &fk->addrs.v6addrs.src,
+			sizeof(struct in6_addr)) &&
+		!memcmp(&fltr_info->ip.v6.dst_ip, &fk->addrs.v6addrs.dst,
+			sizeof(struct in6_addr));
+}
+
 /**
  * ice_rx_flow_steer - steer the Rx flow to where application is being run
  * @netdev: ptr to the netdev being adjusted
@@ -447,6 +491,10 @@ ice_rx_flow_steer(struct net_device *netdev, const struct sk_buff *skb,
 			continue;
 
 		fltr_info = &arfs_entry->fltr_info;
+
+		if (!ice_arfs_cmp(fltr_info, &fk, n_proto, ip_proto))
+			continue;
+
 		ret = fltr_info->fltr_id;
 
 		if (fltr_info->q_index == rxq_idx ||
Signature · 1 line
2.39.5
author maintainer
#478 lore raw

PATCH 479/508 ptp: fix breakage after ptp_vclock_in_use() rework

In reply to Greg Kroah-Hartman

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

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

From: Vladimir Oltean <[email protected]>

[ Upstream commit 5ab73b010cad294851e558f1d4714a85c6f206c7 ]

What is broken
--------------

ptp4l, and any other application which calls clock_adjtime() on a physical clock, is greeted with error -EBUSY after commit 87f7ce260a3c ("ptp: remove ptp->n_vclocks check logic in ptp_vclock_in_use()").

Explanation for the breakage
----------------------------

The blamed commit was based on the false assumption that ptp_vclock_in_use() callers already test for n_vclocks prior to calling this function.

This is notably incorrect for the code path below, in which there is, in fact, no n_vclocks test:

ptp_clock_adjtime()
-> ptp_clock_freerun()
-> ptp_vclock_in_use()

The result is that any clock adjustment on any physical clock is now impossible. This is _despite_ there not being any vclock over this physical clock.

$ ptp4l -i eno0 -2 -P -m
ptp4l[58.425]: selected /dev/ptp0 as PTP clock
[   58.429749] ptp: physical clock is free running
ptp4l[58.431]: Failed to open /dev/ptp0: Device or resource busy
failed to create a clock
$ cat /sys/class/ptp/ptp0/n_vclocks
0

The patch makes the ptp_vclock_in_use() function say "if it's not a virtual clock, then this physical clock does have virtual clocks on top".

Then ptp_clock_freerun() uses this information to say "this physical clock has virtual clocks on top, so it must stay free-running".

Then ptp_clock_adjtime() uses this information to say "well, if this physical clock has to be free-running, I can't do it, return -EBUSY".

Simply put, ptp_vclock_in_use() cannot be simplified so as to remove the test whether vclocks are in use.

What did the blamed commit intend to fix
----------------------------------------

The blamed commit presents a lockdep warning stating "possible recursive locking detected", with the n_vclocks_store() and ptp_clock_unregister() functions involved.

The recursive locking seems this:
n_vclocks_store()
-> mutex_lock_interruptible(&ptp->n_vclocks_mux) // 1 -> device_for_each_child_reverse(..., unregister_vclock)
-> unregister_vclock()

      -> ptp_vclock_unregister()
         -> ptp_clock_unregister()
            -> ptp_vclock_in_use()
               -> mutex_lock_interruptible(&ptp->n_vclocks_mux) // 2

The issue can be triggered by creating and then deleting vclocks:

$ echo 2 > /sys/class/ptp/ptp0/n_vclocks
$ echo 0 > /sys/class/ptp/ptp0/n_vclocks

But note that in the original stack trace, the address of the first lock is different from the address of the second lock. This is because at step 1 marked above, &ptp->n_vclocks_mux is the lock of the parent (physical) PTP clock, and at step 2, the lock is of the child (virtual) PTP clock. They are different locks of different devices.

In this situation there is no real deadlock, the lockdep warning is caused by the fact that the mutexes have the same lock class on both the parent and the child. Functionally it is fine.

Proposed alternative solution
-----------------------------

We must reintroduce the body of ptp_vclock_in_use() mostly as it was structured prior to the blamed commit, but avoid the lockdep warning.

Based on the fact that vclocks cannot be nested on top of one another (ptp_is_attribute_visible() hides n_vclocks for virtual clocks), we already know that ptp->n_vclocks is zero for a virtual clock. And ptp->is_virtual_clock is a runtime invariant, established at ptp_clock_register() time and never changed. There is no need to serialize on any mutex in order to read ptp->is_virtual_clock, and we take advantage of that by moving it outside the lock.

Thus, virtual clocks do not need to acquire &ptp->n_vclocks_mux at all, and step 2 in the code walkthrough above can simply go away. We can simply return false to the question "ptp_vclock_in_use(a virtual clock)".

Other notes
-----------

Releasing &ptp->n_vclocks_mux before ptp_vclock_in_use() returns execution seems racy, because the returned value can become stale as soon as the function returns and before the return value is used (i.e. n_vclocks_store() can run any time). The locking requirement should somehow be transferred to the caller, to ensure a longer life time for the returned value, but this seems out of scope for this severe bug fix.

Because we are also fixing up the logic from the original commit, there is another Fixes: tag for that.

Fixes 87f7ce260a3c ("ptp: remove ptp->n_vclocks check logic in ptp_vclock_in_use()")
Fixes 73f37068d540 ("ptp: support ptp physical/virtual clocks conversion")
---
 drivers/ptp/ptp_private.h | 22 +++++++++++++++++++++-
 1 file changed, 21 insertions(+), 1 deletion(-)
drivers/ptp/ptp_private.h +21 -1
index d0eb4555720eb..a54124269c2f4 100644
@@ -89,7 +89,27 @@ static inline int queue_cnt(const struct timestamp_event_queue *q)
 /* Check if ptp virtual clock is in use */
 static inline bool ptp_vclock_in_use(struct ptp_clock *ptp)
 {
-	return !ptp->is_virtual_clock;
+	bool in_use = false;
+
+	/* Virtual clocks can't be stacked on top of virtual clocks.
+	 * Avoid acquiring the n_vclocks_mux on virtual clocks, to allow this
+	 * function to be called from code paths where the n_vclocks_mux of the
+	 * parent physical clock is already held. Functionally that's not an
+	 * issue, but lockdep would complain, because they have the same lock
+	 * class.
+	 */
+	if (ptp->is_virtual_clock)
+		return false;
+
+	if (mutex_lock_interruptible(&ptp->n_vclocks_mux))
+		return true;
+
+	if (ptp->n_vclocks)
+		in_use = true;
+
+	mutex_unlock(&ptp->n_vclocks_mux);
+
+	return in_use;
 }
 
 /* Check if ptp clock shall be free running */
Signature · 1 line
2.39.5
author maintainer
#479 lore raw

PATCH 480/508 ptp: allow reading of currently dialed frequency to succeed on free-running clocks

In reply to Greg Kroah-Hartman

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

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

From: Vladimir Oltean <[email protected]>

[ Upstream commit aa112cbc5f0ac6f3b44d829005bf34005d9fe9bb ]

There is a bug in ptp_clock_adjtime() which makes it refuse the operation even if we just want to read the current clock dialed frequency, not modify anything (tx->modes == 0). That should be possible even if the clock is free-running. For context, the kernel UAPI is the same for getting and setting the frequency of a POSIX clock.

For example, ptp4l errors out at clock_create() -> clockadj_get_freq() -> clock_adjtime() time, when it should logically only have failed on actual adjustments to the clock, aka if the clock was configured as slave. But in master mode it should work.

This was discovered when examining the issue described in the previous commit, where ptp_clock_freerun() returned true despite n_vclocks being zero.

Fixes 73f37068d540 ("ptp: support ptp physical/virtual clocks conversion")
---
 drivers/ptp/ptp_clock.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/ptp/ptp_clock.c +2 -1
index af66ed48dc006..642c939d4523c 100644
@@ -104,7 +104,8 @@ static int ptp_clock_adjtime(struct posix_clock *pc, struct __kernel_timex *tx)
 	struct ptp_clock_info *ops;
 	int err = -EOPNOTSUPP;
 
-	if (ptp_clock_freerun(ptp)) {
+	if (tx->modes & (ADJ_SETOFFSET | ADJ_FREQUENCY | ADJ_OFFSET) &&
+	    ptp_clock_freerun(ptp)) {
 		pr_err("ptp: physical clock is free running\n");
 		return -EBUSY;
 	}
Signature · 1 line
2.39.5
author maintainer
#480 lore raw

PATCH 481/508 wifi: carl9170: do not ping device which has failed to load firmware

In reply to Greg Kroah-Hartman

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

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

From: Dmitry Antipov <[email protected]>

[ Upstream commit 15d25307692312cec4b57052da73387f91a2e870 ]

Syzkaller reports [1, 2] crashes caused by an attempts to ping the device which has failed to load firmware. Since such a device doesn't pass 'ieee80211_register_hw()', an internal workqueue managed by 'ieee80211_queue_work()' is not yet created and an attempt to queue work on it causes null-ptr-deref.

[1] https://syzkaller.appspot.com/bug?extid=9a4aec827829942045ff
[2] https://syzkaller.appspot.com/bug?extid=0d8afba53e8fb2633217

Fixes e4a668c59080 ("carl9170: fix spurious restart due to high latency")
---
 drivers/net/wireless/ath/carl9170/usb.c | 19 +++++++++++++------
 1 file changed, 13 insertions(+), 6 deletions(-)
drivers/net/wireless/ath/carl9170/usb.c +13 -6
index a5265997b5767..debac4699687e 100644
@@ -438,14 +438,21 @@ static void carl9170_usb_rx_complete(struct urb *urb)
 
 		if (atomic_read(&ar->rx_anch_urbs) == 0) {
 			/*
-			 * The system is too slow to cope with
-			 * the enormous workload. We have simply
-			 * run out of active rx urbs and this
-			 * unfortunately leads to an unpredictable
-			 * device.
+			 * At this point, either the system is too slow to
+			 * cope with the enormous workload (so we have simply
+			 * run out of active rx urbs and this unfortunately
+			 * leads to an unpredictable device), or the device
+			 * is not fully functional after an unsuccessful
+			 * firmware loading attempts (so it doesn't pass
+			 * ieee80211_register_hw() and there is no internal
+			 * workqueue at all).
 			 */
 
-			ieee80211_queue_work(ar->hw, &ar->ping_work);
+			if (ar->registered)
+				ieee80211_queue_work(ar->hw, &ar->ping_work);
+			else
+				pr_warn_once("device %s is not registered\n",
+					     dev_name(&ar->udev->dev));
 		}
 	} else {
 		/*
Signature · 1 line
2.39.5
author maintainer
#481 lore raw

PATCH 482/508 mpls: Use rcu_dereference_rtnl() in mpls_route_input_rcu().

In reply to Greg Kroah-Hartman

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

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

From: Kuniyuki Iwashima <[email protected]>

[ Upstream commit 6dbb0d97c5096072c78a6abffe393584e57ae945 ]

As syzbot reported [0], mpls_route_input_rcu() can be called from mpls_getroute(), where is under RTNL.

net->mpls.platform_label is only updated under RTNL.

Let's use rcu_dereference_rtnl() in mpls_route_input_rcu() to silence the splat.

[0]:
WARNING: suspicious RCU usage 6.15.0-rc7-syzkaller-00082-g5cdb2c77c4c3 #0 Not tainted
----------------------------
net/mpls/af_mpls.c:84 suspicious rcu_dereference_check() usage!

other info that might help us debug this:

rcu_scheduler_active = 2, debug_locks = 1
1 lock held by syz.2.4451/17730:
#0: ffffffff9012a3e8 (rtnl_mutex){+.+.}-{4:4}, at: rtnl_lock net/core/rtnetlink.c:80 [inline] #0: ffffffff9012a3e8 (rtnl_mutex){+.+.}-{4:4}, at: rtnetlink_rcv_msg+0x371/0xe90 net/core/rtnetlink.c:6961

stack backtrace:
CPU: 1 UID: 0 PID: 17730 Comm: syz.2.4451 Not tainted 6.15.0-rc7-syzkaller-00082-g5cdb2c77c4c3 #0 PREEMPT(full)
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/07/2025
Call Trace:
 <TASK>
 __dump_stack lib/dump_stack.c:94 [inline]
 dump_stack_lvl+0x16c/0x1f0 lib/dump_stack.c:120
 lockdep_rcu_suspicious+0x166/0x260 kernel/locking/lockdep.c:6865
 mpls_route_input_rcu+0x1d4/0x200 net/mpls/af_mpls.c:84
 mpls_getroute+0x621/0x1ea0 net/mpls/af_mpls.c:2381
 rtnetlink_rcv_msg+0x3c9/0xe90 net/core/rtnetlink.c:6964
 netlink_rcv_skb+0x16d/0x440 net/netlink/af_netlink.c:2534
 netlink_unicast_kernel net/netlink/af_netlink.c:1313 [inline]
 netlink_unicast+0x53a/0x7f0 net/netlink/af_netlink.c:1339
 netlink_sendmsg+0x8d1/0xdd0 net/netlink/af_netlink.c:1883
 sock_sendmsg_nosec net/socket.c:712 [inline]
 __sock_sendmsg net/socket.c:727 [inline]
 ____sys_sendmsg+0xa98/0xc70 net/socket.c:2566
 ___sys_sendmsg+0x134/0x1d0 net/socket.c:2620
 __sys_sendmmsg+0x200/0x420 net/socket.c:2709
 __do_sys_sendmmsg net/socket.c:2736 [inline]
 __se_sys_sendmmsg net/socket.c:2733 [inline]
 __x64_sys_sendmmsg+0x9c/0x100 net/socket.c:2733
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0xcd/0x230 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f0a2818e969
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 a8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f0a28f52038 EFLAGS: 00000246 ORIG_RAX: 0000000000000133
RAX: ffffffffffffffda RBX: 00007f0a283b5fa0 RCX: 00007f0a2818e969
RDX: 0000000000000003 RSI: 0000200000000080 RDI: 0000000000000003
RBP: 00007f0a28210ab1 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 0000000000000000 R14: 00007f0a283b5fa0 R15: 00007ffce5e9f268
 </TASK>
---
 net/mpls/af_mpls.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
net/mpls/af_mpls.c +2 -2
index f1f43894efb8e..2d29d230f5698 100644
@@ -80,8 +80,8 @@ static struct mpls_route *mpls_route_input_rcu(struct net *net, unsigned index)
 
 	if (index < net->mpls.platform_labels) {
 		struct mpls_route __rcu **platform_label =
-			rcu_dereference(net->mpls.platform_label);
-		rt = rcu_dereference(platform_label[index]);
+			rcu_dereference_rtnl(net->mpls.platform_label);
+		rt = rcu_dereference_rtnl(platform_label[index]);
 	}
 	return rt;
 }
Signature · 1 line
2.39.5
author maintainer
#482 lore raw

PATCH 483/508 atm: atmtcp: Free invalid length skb in atmtcp_c_send().

In reply to Greg Kroah-Hartman

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

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

From: Kuniyuki Iwashima <[email protected]>

[ Upstream commit 2f370ae1fb6317985f3497b1bb80d457508ca2f7 ]

syzbot reported the splat below. [0]

vcc_sendmsg() copies data passed from userspace to skb and passes it to vcc->dev->ops->send().

atmtcp_c_send() accesses skb->data as struct atmtcp_hdr after checking if skb->len is 0, but it's not enough.

Also, when skb->len == 0, skb and sk (vcc) were leaked because dev_kfree_skb() is not called and sk_wmem_alloc adjustment is missing to revert atm_account_tx() in vcc_sendmsg(), which is expected to be done in atm_pop_raw().

Let's properly free skb with an invalid length in atmtcp_c_send().

[0]:

BUG: KMSAN: uninit-value in atmtcp_c_send+0x255/0xed0 drivers/atm/atmtcp.c:294
 atmtcp_c_send+0x255/0xed0 drivers/atm/atmtcp.c:294
 vcc_sendmsg+0xd7c/0xff0 net/atm/common.c:644
 sock_sendmsg_nosec net/socket.c:712 [inline]
 __sock_sendmsg+0x330/0x3d0 net/socket.c:727
 ____sys_sendmsg+0x7e0/0xd80 net/socket.c:2566
 ___sys_sendmsg+0x271/0x3b0 net/socket.c:2620
 __sys_sendmsg net/socket.c:2652 [inline]
 __do_sys_sendmsg net/socket.c:2657 [inline]
 __se_sys_sendmsg net/socket.c:2655 [inline]
 __x64_sys_sendmsg+0x211/0x3e0 net/socket.c:2655
 x64_sys_call+0x32fb/0x3db0 arch/x86/include/generated/asm/syscalls_64.h:47
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0xd9/0x210 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Uninit was created at:

 slab_post_alloc_hook mm/slub.c:4154 [inline]
 slab_alloc_node mm/slub.c:4197 [inline]
 kmem_cache_alloc_node_noprof+0x818/0xf00 mm/slub.c:4249
 kmalloc_reserve+0x13c/0x4b0 net/core/skbuff.c:579
 __alloc_skb+0x347/0x7d0 net/core/skbuff.c:670
 alloc_skb include/linux/skbuff.h:1336 [inline]
 vcc_sendmsg+0xb40/0xff0 net/atm/common.c:628
 sock_sendmsg_nosec net/socket.c:712 [inline]
 __sock_sendmsg+0x330/0x3d0 net/socket.c:727
 ____sys_sendmsg+0x7e0/0xd80 net/socket.c:2566
 ___sys_sendmsg+0x271/0x3b0 net/socket.c:2620
 __sys_sendmsg net/socket.c:2652 [inline]
 __do_sys_sendmsg net/socket.c:2657 [inline]
 __se_sys_sendmsg net/socket.c:2655 [inline]
 __x64_sys_sendmsg+0x211/0x3e0 net/socket.c:2655
 x64_sys_call+0x32fb/0x3db0 arch/x86/include/generated/asm/syscalls_64.h:47
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0xd9/0x210 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

CPU: 1 UID: 0 PID: 5798 Comm: syz-executor192 Not tainted 6.16.0-rc1-syzkaller-00010-g2c4a1f3fe03e #0 PREEMPT(undef)
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/07/2025
---
 drivers/atm/atmtcp.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/atm/atmtcp.c +3 -1
index 96bea1ab1eccf..ff558908897f3 100644
@@ -288,7 +288,9 @@ static int atmtcp_c_send(struct atm_vcc *vcc,struct sk_buff *skb)
 	struct sk_buff *new_skb;
 	int result = 0;
 
-	if (!skb->len) return 0;
+	if (skb->len < sizeof(struct atmtcp_hdr))
+		goto done;
+
 	dev = vcc->dev_data;
 	hdr = (struct atmtcp_hdr *) skb->data;
 	if (hdr->length == ATMTCP_HDR_MAGIC) {
Signature · 1 line
2.39.5
author maintainer
#483 lore raw

PATCH 484/508 tcp: fix tcp_packet_delayed() for tcp_is_non_sack_preventing_reopen() behavior

In reply to Greg Kroah-Hartman

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

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

From: Neal Cardwell <[email protected]>

[ Upstream commit d0fa59897e049e84432600e86df82aab3dce7aa5 ]

After the following commit from 2024:

commit e37ab7373696 ("tcp: fix to allow timestamp undo if no retransmits were sent")

...there was buggy behavior where TCP connections without SACK support could easily see erroneous undo events at the end of fast recovery or RTO recovery episodes. The erroneous undo events could cause those connections to suffer repeated loss recovery episodes and high retransmit rates.

The problem was an interaction between the non-SACK behavior on these connections and the undo logic. The problem is that, for non-SACK connections at the end of a loss recovery episode, if snd_una == high_seq, then tcp_is_non_sack_preventing_reopen() holds steady in CA_Recovery or CA_Loss, but clears tp->retrans_stamp to 0. Then upon the next ACK the "tcp: fix to allow timestamp undo if no retransmits were sent" logic saw the tp->retrans_stamp at 0 and erroneously concluded that no data was retransmitted, and erroneously performed an undo of the cwnd reduction, restoring cwnd immediately to the value it had before loss recovery. This caused an immediate burst of traffic and build-up of queues and likely another immediate loss recovery episode.

This commit fixes tcp_packet_delayed() to ignore zero retrans_stamp values for non-SACK connections when snd_una is at or above high_seq, because tcp_is_non_sack_preventing_reopen() clears retrans_stamp in this case, so it's not a valid signal that we can undo.

Note that the commit named in the Fixes footer restored long-present behavior from roughly 2005-2019, so apparently this bug was present for a while during that era, and this was simply not caught.

---
 net/ipv4/tcp_input.c | 37 +++++++++++++++++++++++++------------
 1 file changed, 25 insertions(+), 12 deletions(-)
net/ipv4/tcp_input.c +25 -12
index 2888aaae0d76f..222b829f33f42 100644
@@ -2441,20 +2441,33 @@ static inline bool tcp_packet_delayed(const struct tcp_sock *tp)
 {
 	const struct sock *sk = (const struct sock *)tp;
 
-	if (tp->retrans_stamp &&
-	    tcp_tsopt_ecr_before(tp, tp->retrans_stamp))
-		return true;  /* got echoed TS before first retransmission */
-
-	/* Check if nothing was retransmitted (retrans_stamp==0), which may
-	 * happen in fast recovery due to TSQ. But we ignore zero retrans_stamp
-	 * in TCP_SYN_SENT, since when we set FLAG_SYN_ACKED we also clear
-	 * retrans_stamp even if we had retransmitted the SYN.
+	/* Received an echoed timestamp before the first retransmission? */
+	if (tp->retrans_stamp)
+		return tcp_tsopt_ecr_before(tp, tp->retrans_stamp);
+
+	/* We set tp->retrans_stamp upon the first retransmission of a loss
+	 * recovery episode, so normally if tp->retrans_stamp is 0 then no
+	 * retransmission has happened yet (likely due to TSQ, which can cause
+	 * fast retransmits to be delayed). So if snd_una advanced while
+	 * (tp->retrans_stamp is 0 then apparently a packet was merely delayed,
+	 * not lost. But there are exceptions where we retransmit but then
+	 * clear tp->retrans_stamp, so we check for those exceptions.
 	 */
-	if (!tp->retrans_stamp &&	   /* no record of a retransmit/SYN? */
-	    sk->sk_state != TCP_SYN_SENT)  /* not the FLAG_SYN_ACKED case? */
-		return true;  /* nothing was retransmitted */
 
-	return false;
+	/* (1) For non-SACK connections, tcp_is_non_sack_preventing_reopen()
+	 * clears tp->retrans_stamp when snd_una == high_seq.
+	 */
+	if (!tcp_is_sack(tp) && !before(tp->snd_una, tp->high_seq))
+		return false;
+
+	/* (2) In TCP_SYN_SENT tcp_clean_rtx_queue() clears tp->retrans_stamp
+	 * when setting FLAG_SYN_ACKED is set, even if the SYN was
+	 * retransmitted.
+	 */
+	if (sk->sk_state == TCP_SYN_SENT)
+		return false;
+
+	return true;	/* tp->retrans_stamp is zero; no retransmit yet */
 }
 
 /* Undo procedures. */
Signature · 1 line
2.39.5
author maintainer
#484 lore raw

PATCH 485/508 tipc: fix null-ptr-deref when acquiring remote ip of ethernet bearer

In reply to Greg Kroah-Hartman

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

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

From: Haixia Qu <[email protected]>

[ Upstream commit f82727adcf2992822e12198792af450a76ebd5ef ]

The reproduction steps:
1. create a tun interface
2. enable l2 bearer
3. TIPC_NL_UDP_GET_REMOTEIP with media name set to tun

tipc: Started in network mode
tipc: Node identity 8af312d38a21, cluster identity 4711
tipc: Enabled bearer <eth:syz_tun>, priority 1

Oops: general protection fault
KASAN: null-ptr-deref in range
CPU: 1 UID: 1000 PID: 559 Comm: poc Not tainted 6.16.0-rc1+ #117 PREEMPT
Hardware name: QEMU Ubuntu 24.04 PC
RIP: 0010:tipc_udp_nl_dump_remoteip+0x4a4/0x8f0

the ub was in fact a struct dev.

when bid != 0 && skip_cnt != 0, bearer_list[bid] may be NULL or other media when other thread changes it.

fix this by checking media_id.

Fixes 832629ca5c313 ("tipc: add UDP remoteip dump to netlink API")
---
 net/tipc/udp_media.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
net/tipc/udp_media.c +2 -2
index f5bd75d931c1b..e1305d159834b 100644
@@ -489,7 +489,7 @@ int tipc_udp_nl_dump_remoteip(struct sk_buff *skb, struct netlink_callback *cb)
 
 		rtnl_lock();
 		b = tipc_bearer_find(net, bname);
-		if (!b) {
+		if (!b || b->bcast_addr.media_id != TIPC_MEDIA_TYPE_UDP) {
 			rtnl_unlock();
 			return -EINVAL;
 		}
@@ -500,7 +500,7 @@ int tipc_udp_nl_dump_remoteip(struct sk_buff *skb, struct netlink_callback *cb)
 
 		rtnl_lock();
 		b = rtnl_dereference(tn->bearer_list[bid]);
-		if (!b) {
+		if (!b || b->bcast_addr.media_id != TIPC_MEDIA_TYPE_UDP) {
 			rtnl_unlock();
 			return -EINVAL;
 		}
Signature · 1 line
2.39.5
author maintainer
#485 lore raw

PATCH 486/508 tcp: fix passive TFO socket having invalid NAPI ID

In reply to Greg Kroah-Hartman

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

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

From: David Wei <[email protected]>

[ Upstream commit dbe0ca8da1f62b6252e7be6337209f4d86d4a914 ]

There is a bug with passive TFO sockets returning an invalid NAPI ID 0 from SO_INCOMING_NAPI_ID. Normally this is not an issue, but zero copy receive relies on a correct NAPI ID to process sockets on the right queue.

Fix by adding a sk_mark_napi_id_set().

Fixes e5907459ce7e ("tcp: Record Rx hash and NAPI ID in tcp_child_process")
---
 net/ipv4/tcp_fastopen.c | 3 +++
 1 file changed, 3 insertions(+)
net/ipv4/tcp_fastopen.c +3 -0
index cb01c770d8cf5..cbce1306bb08c 100644
@@ -3,6 +3,7 @@
 #include <linux/tcp.h>
 #include <linux/rcupdate.h>
 #include <net/tcp.h>
+#include <net/busy_poll.h>
 
 void tcp_fastopen_init_key_once(struct net *net)
 {
@@ -279,6 +280,8 @@ static struct sock *tcp_fastopen_create_child(struct sock *sk,
 
 	refcount_set(&req->rsk_refcnt, 2);
 
+	sk_mark_napi_id_set(child, skb);
+
 	/* Now finish processing the fastopen child socket. */
 	tcp_init_transfer(child, BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB, skb);
 
Signature · 1 line
2.39.5
author maintainer
#486 lore raw

PATCH 487/508 net: microchip: lan743x: Reduce PTP timeout on HW failure

In reply to Greg Kroah-Hartman

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

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

From: Rengarajan S <[email protected]>

[ Upstream commit b1de3c0df7abc41dc41862c0b08386411f2799d7 ]

The PTP_CMD_CTL is a self clearing register which controls the PTP clock values. In the current implementation driver waits for a duration of 20 sec in case of HW failure to clear the PTP_CMD_CTL register bit. This timeout of 20 sec is very long to recognize a HW failure, as it is typically cleared in one clock(<16ns). Hence reducing the timeout to 1 sec would be sufficient to conclude if there is any HW failure observed. The usleep_range will sleep somewhere between 1 msec to 20 msec for each iteration. By setting the PTP_CMD_CTL_TIMEOUT_CNT to 50 the max timeout is extended to 1 sec.

Stable-dep-of: e353b0854d3a ("net: lan743x: fix potential out-of-bounds write in lan743x_ptp_io_event_clock_get()")

---
 drivers/net/ethernet/microchip/lan743x_ptp.c | 2 +-
 drivers/net/ethernet/microchip/lan743x_ptp.h | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)
drivers/net/ethernet/microchip/lan743x_ptp.c +1 -1
index da3ea905adbb8..7ace578551a09 100644
@@ -58,7 +58,7 @@ int lan743x_gpio_init(struct lan743x_adapter *adapter)
 static void lan743x_ptp_wait_till_cmd_done(struct lan743x_adapter *adapter,
 					   u32 bit_mask)
 {
-	int timeout = 1000;
+	int timeout = PTP_CMD_CTL_TIMEOUT_CNT;
 	u32 data = 0;
 
 	while (timeout &&
drivers/net/ethernet/microchip/lan743x_ptp.h +1 -0
index e26d4eff71336..0d29914cd4606 100644
@@ -21,6 +21,7 @@
 #define LAN743X_PTP_N_EXTTS		4
 #define LAN743X_PTP_N_PPS		0
 #define PCI11X1X_PTP_IO_MAX_CHANNELS	8
+#define PTP_CMD_CTL_TIMEOUT_CNT		50
 
 struct lan743x_adapter;
 
Signature · 1 line
2.39.5
author maintainer
#487 lore raw

PATCH 488/508 net: lan743x: fix potential out-of-bounds write in lan743x_ptp_io_event_clock_get()

In reply to Greg Kroah-Hartman

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

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

From: Alexey Kodanev <[email protected]>

[ Upstream commit e353b0854d3a1a31cb061df8d022fbfea53a0f24 ]

Before calling lan743x_ptp_io_event_clock_get(), the 'channel' value is checked against the maximum value of PCI11X1X_PTP_IO_MAX_CHANNELS(8). This seems correct and aligns with the PTP interrupt status register (PTP_INT_STS) specifications.

However, lan743x_ptp_io_event_clock_get() writes to ptp->extts[] with only LAN743X_PTP_N_EXTTS(4) elements, using channel as an index:

    lan743x_ptp_io_event_clock_get(..., u8 channel,...)
    {
        ...
        /* Update Local timestamp */
        extts = &ptp->extts[channel];
        extts->ts.tv_sec = sec;
        ...
    }

To avoid an out-of-bounds write and utilize all the supported GPIO inputs, set LAN743X_PTP_N_EXTTS to 8.

Detected using the static analysis tool - Svace.

Fixes 60942c397af6 ("net: lan743x: Add support for PTP-IO Event Input External Timestamp (extts)")
---
 drivers/net/ethernet/microchip/lan743x_ptp.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/microchip/lan743x_ptp.h +2 -2
index 0d29914cd4606..225e8232474d7 100644
@@ -18,9 +18,9 @@
  */
 #define LAN743X_PTP_N_EVENT_CHAN	2
 #define LAN743X_PTP_N_PEROUT		LAN743X_PTP_N_EVENT_CHAN
-#define LAN743X_PTP_N_EXTTS		4
-#define LAN743X_PTP_N_PPS		0
 #define PCI11X1X_PTP_IO_MAX_CHANNELS	8
+#define LAN743X_PTP_N_EXTTS		PCI11X1X_PTP_IO_MAX_CHANNELS
+#define LAN743X_PTP_N_PPS		0
 #define PTP_CMD_CTL_TIMEOUT_CNT		50
 
 struct lan743x_adapter;
Signature · 1 line
2.39.5
author maintainer
#488 lore raw

PATCH 489/508 calipso: Fix null-ptr-deref in calipso_req_{set,del}attr().

In reply to Greg Kroah-Hartman

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

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

From: Kuniyuki Iwashima <[email protected]>

[ Upstream commit 10876da918fa1aec0227fb4c67647513447f53a9 ]

syzkaller reported a null-ptr-deref in sock_omalloc() while allocating a CALIPSO option. [0]

The NULL is of struct sock, which was fetched by sk_to_full_sk() in calipso_req_setattr().

Since commit a1a5344ddbe8 ("tcp: avoid two atomic ops for syncookies"), reqsk->rsk_listener could be NULL when SYN Cookie is returned to its client, as hinted by the leading SYN Cookie log.

Here are 3 options to fix the bug:

1) Return 0 in calipso_req_setattr()
2) Return an error in calipso_req_setattr()
3) Alaways set rsk_listener

1) is no go as it bypasses LSM, but 2) effectively disables SYN Cookie for CALIPSO. 3) is also no go as there have been many efforts to reduce atomic ops and make TCP robust against DDoS. See also commit 3b24d854cb35 ("tcp/dccp: do not touch listener sk_refcnt under synflood").

As of the blamed commit, SYN Cookie already did not need refcounting, and no one has stumbled on the bug for 9 years, so no CALIPSO user will care about SYN Cookie.

Let's return an error in calipso_req_setattr() and calipso_req_delattr() in the SYN Cookie case.

This can be reproduced by [1] on Fedora and now connect() of nc times out.

[0]:
TCP: request_sock_TCPv6: Possible SYN flooding on port [::]:20002. Sending cookies.

Oops: general protection fault, probably for non-canonical address 0xdffffc0000000006: 0000 [#1] PREEMPT SMP KASAN NOPTI
KASAN: null-ptr-deref in range [0x0000000000000030-0x0000000000000037]
CPU: 3 UID: 0 PID: 12262 Comm: syz.1.2611 Not tainted 6.14.0 #2
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.3-0-ga6ed6b701f0a-prebuilt.qemu.org 04/01/2014
RIP: 0010:read_pnet include/net/net_namespace.h:406 [inline]
RIP: 0010:sock_net include/net/sock.h:655 [inline]
RIP: 0010:sock_kmalloc+0x35/0x170 net/core/sock.c:2806
Code: 89 d5 41 54 55 89 f5 53 48 89 fb e8 25 e3 c6 fd e8 f0 91 e3 00 48 8d 7b 30 48 b8 00 00 00 00 00 fc ff df 48 89 fa 48 c1 ea 03 <80> 3c 02 00 0f 85 26 01 00 00 48 b8 00 00 00 00 00 fc ff df 4c 8b
RSP: 0018:ffff88811af89038 EFLAGS: 00010216
RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffff888105266400
RDX: 0000000000000006 RSI: ffff88800c890000 RDI: 0000000000000030
RBP: 0000000000000050 R08: 0000000000000000 R09: ffff88810526640e
R10: ffffed1020a4cc81 R11: ffff88810526640f R12: 0000000000000000
R13: 0000000000000820 R14: ffff888105266400 R15: 0000000000000050
FS:  00007f0653a07640(0000) GS:ffff88811af80000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f863ba096f4 CR3: 00000000163c0005 CR4: 0000000000770ef0
PKRU: 80000000
Call Trace:
 <IRQ>
 ipv6_renew_options+0x279/0x950 net/ipv6/exthdrs.c:1288
 calipso_req_setattr+0x181/0x340 net/ipv6/calipso.c:1204
 calipso_req_setattr+0x56/0x80 net/netlabel/netlabel_calipso.c:597
 netlbl_req_setattr+0x18a/0x440 net/netlabel/netlabel_kapi.c:1249
 selinux_netlbl_inet_conn_request+0x1fb/0x320 security/selinux/netlabel.c:342
 selinux_inet_conn_request+0x1eb/0x2c0 security/selinux/hooks.c:5551
 security_inet_conn_request+0x50/0xa0 security/security.c:4945
 tcp_v6_route_req+0x22c/0x550 net/ipv6/tcp_ipv6.c:825
 tcp_conn_request+0xec8/0x2b70 net/ipv4/tcp_input.c:7275
 tcp_v6_conn_request+0x1e3/0x440 net/ipv6/tcp_ipv6.c:1328
 tcp_rcv_state_process+0xafa/0x52b0 net/ipv4/tcp_input.c:6781
 tcp_v6_do_rcv+0x8a6/0x1a40 net/ipv6/tcp_ipv6.c:1667
 tcp_v6_rcv+0x505e/0x5b50 net/ipv6/tcp_ipv6.c:1904
 ip6_protocol_deliver_rcu+0x17c/0x1da0 net/ipv6/ip6_input.c:436
 ip6_input_finish+0x103/0x180 net/ipv6/ip6_input.c:480
 NF_HOOK include/linux/netfilter.h:314 [inline]
 NF_HOOK include/linux/netfilter.h:308 [inline]
 ip6_input+0x13c/0x6b0 net/ipv6/ip6_input.c:491
 dst_input include/net/dst.h:469 [inline]
 ip6_rcv_finish net/ipv6/ip6_input.c:79 [inline]
 ip6_rcv_finish+0xb6/0x490 net/ipv6/ip6_input.c:69
 NF_HOOK include/linux/netfilter.h:314 [inline]
 NF_HOOK include/linux/netfilter.h:308 [inline]
 ipv6_rcv+0xf9/0x490 net/ipv6/ip6_input.c:309
 __netif_receive_skb_one_core+0x12e/0x1f0 net/core/dev.c:5896
 __netif_receive_skb+0x1d/0x170 net/core/dev.c:6009
 process_backlog+0x41e/0x13b0 net/core/dev.c:6357
 __napi_poll+0xbd/0x710 net/core/dev.c:7191
 napi_poll net/core/dev.c:7260 [inline]
 net_rx_action+0x9de/0xde0 net/core/dev.c:7382
 handle_softirqs+0x19a/0x770 kernel/softirq.c:561
 do_softirq.part.0+0x36/0x70 kernel/softirq.c:462
 </IRQ>
 <TASK>
 do_softirq arch/x86/include/asm/preempt.h:26 [inline]
 __local_bh_enable_ip+0xf1/0x110 kernel/softirq.c:389
 local_bh_enable include/linux/bottom_half.h:33 [inline]
 rcu_read_unlock_bh include/linux/rcupdate.h:919 [inline]
 __dev_queue_xmit+0xc2a/0x3c40 net/core/dev.c:4679
 dev_queue_xmit include/linux/netdevice.h:3313 [inline]
 neigh_hh_output include/net/neighbour.h:523 [inline]
 neigh_output include/net/neighbour.h:537 [inline]
 ip6_finish_output2+0xd69/0x1f80 net/ipv6/ip6_output.c:141
 __ip6_finish_output net/ipv6/ip6_output.c:215 [inline]
 ip6_finish_output+0x5dc/0xd60 net/ipv6/ip6_output.c:226
 NF_HOOK_COND include/linux/netfilter.h:303 [inline]
 ip6_output+0x24b/0x8d0 net/ipv6/ip6_output.c:247
 dst_output include/net/dst.h:459 [inline]
 NF_HOOK include/linux/netfilter.h:314 [inline]
 NF_HOOK include/linux/netfilter.h:308 [inline]
 ip6_xmit+0xbbc/0x20d0 net/ipv6/ip6_output.c:366
 inet6_csk_xmit+0x39a/0x720 net/ipv6/inet6_connection_sock.c:135
 __tcp_transmit_skb+0x1a7b/0x3b40 net/ipv4/tcp_output.c:1471
 tcp_transmit_skb net/ipv4/tcp_output.c:1489 [inline]
 tcp_send_syn_data net/ipv4/tcp_output.c:4059 [inline]
 tcp_connect+0x1c0c/0x4510 net/ipv4/tcp_output.c:4148
 tcp_v6_connect+0x156c/0x2080 net/ipv6/tcp_ipv6.c:333
 __inet_stream_connect+0x3a7/0xed0 net/ipv4/af_inet.c:677
 tcp_sendmsg_fastopen+0x3e2/0x710 net/ipv4/tcp.c:1039
 tcp_sendmsg_locked+0x1e82/0x3570 net/ipv4/tcp.c:1091
 tcp_sendmsg+0x2f/0x50 net/ipv4/tcp.c:1358
 inet6_sendmsg+0xb9/0x150 net/ipv6/af_inet6.c:659
 sock_sendmsg_nosec net/socket.c:718 [inline]
 __sock_sendmsg+0xf4/0x2a0 net/socket.c:733
 __sys_sendto+0x29a/0x390 net/socket.c:2187
 __do_sys_sendto net/socket.c:2194 [inline]
 __se_sys_sendto net/socket.c:2190 [inline]
 __x64_sys_sendto+0xe1/0x1c0 net/socket.c:2190
 do_syscall_x64 arch/x86/entry/common.c:52 [inline]
 do_syscall_64+0xc3/0x1d0 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f06553c47ed
Code: 02 b8 ff ff ff ff c3 66 0f 1f 44 00 00 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 a8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f0653a06fc8 EFLAGS: 00000246 ORIG_RAX: 000000000000002c
RAX: ffffffffffffffda RBX: 00007f0655605fa0 RCX: 00007f06553c47ed
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 000000000000000b
RBP: 00007f065545db38 R08: 0000200000000140 R09: 000000000000001c
R10: f7384d4ea84b01bd R11: 0000000000000246 R12: 0000000000000000
R13: 00007f0655605fac R14: 00007f0655606038 R15: 00007f06539e7000
 </TASK>
Modules linked in:

[1]:
dnf install -y selinux-policy-targeted policycoreutils netlabel_tools procps-ng nmap-ncat mount -t selinuxfs none /sys/fs/selinux
load_policy
netlabelctl calipso add pass doi:1
netlabelctl map del default
netlabelctl map add default address:::1 protocol:calipso,1
sysctl net.ipv4.tcp_syncookies=2
nc -l ::1 80 &
nc ::1 80

---
 net/ipv6/calipso.c | 8 ++++++++
 1 file changed, 8 insertions(+)
net/ipv6/calipso.c +8 -0
index c07e3da08d2a8..24666291c54a8 100644
@@ -1210,6 +1210,10 @@ static int calipso_req_setattr(struct request_sock *req,
 	struct ipv6_opt_hdr *old, *new;
 	struct sock *sk = sk_to_full_sk(req_to_sk(req));
 
+	/* sk is NULL for SYN+ACK w/ SYN Cookie */
+	if (!sk)
+		return -ENOMEM;
+
 	if (req_inet->ipv6_opt && req_inet->ipv6_opt->hopopt)
 		old = req_inet->ipv6_opt->hopopt;
 	else
@@ -1250,6 +1254,10 @@ static void calipso_req_delattr(struct request_sock *req)
 	struct ipv6_txoptions *txopts;
 	struct sock *sk = sk_to_full_sk(req_to_sk(req));
 
+	/* sk is NULL for SYN+ACK w/ SYN Cookie */
+	if (!sk)
+		return;
+
 	if (!req_inet->ipv6_opt || !req_inet->ipv6_opt->hopopt)
 		return;
 
Signature · 1 line
2.39.5
author maintainer
#489 lore raw

PATCH 490/508 net: atm: add lec_mutex

In reply to Greg Kroah-Hartman

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

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

From: Eric Dumazet <[email protected]>

[ Upstream commit d13a3824bfd2b4774b671a75cf766a16637a0e67 ]

syzbot found its way in net/atm/lec.c, and found an error path in lecd_attach() could leave a dangling pointer in dev_lec[].

Add a mutex to protect dev_lecp[] uses from lecd_attach(), lec_vcc_attach() and lec_mcast_attach().

Following patch will use this mutex for /proc/net/atm/lec.

BUG: KASAN: slab-use-after-free in lecd_attach net/atm/lec.c:751 [inline]
BUG: KASAN: slab-use-after-free in lane_ioctl+0x2224/0x23e0 net/atm/lec.c:1008
Read of size 8 at addr ffff88807c7b8e68 by task syz.1.17/6142

CPU: 1 UID: 0 PID: 6142 Comm: syz.1.17 Not tainted 6.16.0-rc1-syzkaller-00239-g08215f5486ec #0 PREEMPT(full)
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/07/2025
Call Trace:
 <TASK>
  __dump_stack lib/dump_stack.c:94 [inline]
  dump_stack_lvl+0x116/0x1f0 lib/dump_stack.c:120
  print_address_description mm/kasan/report.c:408 [inline]
  print_report+0xcd/0x680 mm/kasan/report.c:521
  kasan_report+0xe0/0x110 mm/kasan/report.c:634
  lecd_attach net/atm/lec.c:751 [inline]
  lane_ioctl+0x2224/0x23e0 net/atm/lec.c:1008
  do_vcc_ioctl+0x12c/0x930 net/atm/ioctl.c:159
  sock_do_ioctl+0x118/0x280 net/socket.c:1190
  sock_ioctl+0x227/0x6b0 net/socket.c:1311
  vfs_ioctl fs/ioctl.c:51 [inline]
  __do_sys_ioctl fs/ioctl.c:907 [inline]
  __se_sys_ioctl fs/ioctl.c:893 [inline]
  __x64_sys_ioctl+0x18e/0x210 fs/ioctl.c:893
  do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
  do_syscall_64+0xcd/0x4c0 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
 </TASK>

Allocated by task 6132:
  kasan_save_stack+0x33/0x60 mm/kasan/common.c:47
  kasan_save_track+0x14/0x30 mm/kasan/common.c:68
  poison_kmalloc_redzone mm/kasan/common.c:377 [inline]
  __kasan_kmalloc+0xaa/0xb0 mm/kasan/common.c:394
  kasan_kmalloc include/linux/kasan.h:260 [inline]
  __do_kmalloc_node mm/slub.c:4328 [inline]
  __kvmalloc_node_noprof+0x27b/0x620 mm/slub.c:5015
  alloc_netdev_mqs+0xd2/0x1570 net/core/dev.c:11711
  lecd_attach net/atm/lec.c:737 [inline]
  lane_ioctl+0x17db/0x23e0 net/atm/lec.c:1008
  do_vcc_ioctl+0x12c/0x930 net/atm/ioctl.c:159
  sock_do_ioctl+0x118/0x280 net/socket.c:1190
  sock_ioctl+0x227/0x6b0 net/socket.c:1311
  vfs_ioctl fs/ioctl.c:51 [inline]
  __do_sys_ioctl fs/ioctl.c:907 [inline]
  __se_sys_ioctl fs/ioctl.c:893 [inline]
  __x64_sys_ioctl+0x18e/0x210 fs/ioctl.c:893
  do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
  do_syscall_64+0xcd/0x4c0 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f

Freed by task 6132:
  kasan_save_stack+0x33/0x60 mm/kasan/common.c:47
  kasan_save_track+0x14/0x30 mm/kasan/common.c:68
  kasan_save_free_info+0x3b/0x60 mm/kasan/generic.c:576
  poison_slab_object mm/kasan/common.c:247 [inline]
  __kasan_slab_free+0x51/0x70 mm/kasan/common.c:264
  kasan_slab_free include/linux/kasan.h:233 [inline]
  slab_free_hook mm/slub.c:2381 [inline]
  slab_free mm/slub.c:4643 [inline]
  kfree+0x2b4/0x4d0 mm/slub.c:4842
  free_netdev+0x6c5/0x910 net/core/dev.c:11892
  lecd_attach net/atm/lec.c:744 [inline]
  lane_ioctl+0x1ce8/0x23e0 net/atm/lec.c:1008
  do_vcc_ioctl+0x12c/0x930 net/atm/ioctl.c:159
  sock_do_ioctl+0x118/0x280 net/socket.c:1190
  sock_ioctl+0x227/0x6b0 net/socket.c:1311
  vfs_ioctl fs/ioctl.c:51 [inline]
  __do_sys_ioctl fs/ioctl.c:907 [inline]
  __se_sys_ioctl fs/ioctl.c:893 [inline]
  __x64_sys_ioctl+0x18e/0x210 fs/ioctl.c:893
---
 net/atm/lec.c | 7 +++++++
 1 file changed, 7 insertions(+)
net/atm/lec.c +7 -0
index ac3cfc1ae5102..d4ac1488eca6f 100644
@@ -124,6 +124,7 @@ static unsigned char bus_mac[ETH_ALEN] = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };
 
 /* Device structures */
 static struct net_device *dev_lec[MAX_LEC_ITF];
+static DEFINE_MUTEX(lec_mutex);
 
 #if IS_ENABLED(CONFIG_BRIDGE)
 static void lec_handle_bridge(struct sk_buff *skb, struct net_device *dev)
@@ -685,6 +686,7 @@ static int lec_vcc_attach(struct atm_vcc *vcc, void __user *arg)
 	int bytes_left;
 	struct atmlec_ioc ioc_data;
 
+	lockdep_assert_held(&lec_mutex);
 	/* Lecd must be up in this case */
 	bytes_left = copy_from_user(&ioc_data, arg, sizeof(struct atmlec_ioc));
 	if (bytes_left != 0)
@@ -710,6 +712,7 @@ static int lec_vcc_attach(struct atm_vcc *vcc, void __user *arg)
 
 static int lec_mcast_attach(struct atm_vcc *vcc, int arg)
 {
+	lockdep_assert_held(&lec_mutex);
 	if (arg < 0 || arg >= MAX_LEC_ITF)
 		return -EINVAL;
 	arg = array_index_nospec(arg, MAX_LEC_ITF);
@@ -725,6 +728,7 @@ static int lecd_attach(struct atm_vcc *vcc, int arg)
 	int i;
 	struct lec_priv *priv;
 
+	lockdep_assert_held(&lec_mutex);
 	if (arg < 0)
 		arg = 0;
 	if (arg >= MAX_LEC_ITF)
@@ -742,6 +746,7 @@ static int lecd_attach(struct atm_vcc *vcc, int arg)
 		snprintf(dev_lec[i]->name, IFNAMSIZ, "lec%d", i);
 		if (register_netdev(dev_lec[i])) {
 			free_netdev(dev_lec[i]);
+			dev_lec[i] = NULL;
 			return -EINVAL;
 		}
 
@@ -1003,6 +1008,7 @@ static int lane_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 		return -ENOIOCTLCMD;
 	}
 
+	mutex_lock(&lec_mutex);
 	switch (cmd) {
 	case ATMLEC_CTRL:
 		err = lecd_attach(vcc, (int)arg);
@@ -1017,6 +1023,7 @@ static int lane_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
 		break;
 	}
 
+	mutex_unlock(&lec_mutex);
 	return err;
 }
 
Signature · 1 line
2.39.5
author maintainer
#490 lore raw

PATCH 491/508 net: atm: fix /proc/net/atm/lec handling

In reply to Greg Kroah-Hartman

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

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

From: Eric Dumazet <[email protected]>

[ Upstream commit d03b79f459c7935cff830d98373474f440bd03ae ]

/proc/net/atm/lec must ensure safety against dev_lec[] changes.

It appears it had dev_put() calls without prior dev_hold(), leading to imbalance and UAF.

---
 net/atm/lec.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
net/atm/lec.c +3 -2
index d4ac1488eca6f..b7fa48a9b7205 100644
@@ -909,7 +909,6 @@ static void *lec_itf_walk(struct lec_state *state, loff_t *l)
 	v = (dev && netdev_priv(dev)) ?
 		lec_priv_walk(state, l, netdev_priv(dev)) : NULL;
 	if (!v && dev) {
-		dev_put(dev);
 		/* Partial state reset for the next time we get called */
 		dev = NULL;
 	}
@@ -933,6 +932,7 @@ static void *lec_seq_start(struct seq_file *seq, loff_t *pos)
 {
 	struct lec_state *state = seq->private;
 
+	mutex_lock(&lec_mutex);
 	state->itf = 0;
 	state->dev = NULL;
 	state->locked = NULL;
@@ -950,8 +950,9 @@ static void lec_seq_stop(struct seq_file *seq, void *v)
 	if (state->dev) {
 		spin_unlock_irqrestore(&state->locked->lec_arp_lock,
 				       state->flags);
-		dev_put(state->dev);
+		state->dev = NULL;
 	}
+	mutex_unlock(&lec_mutex);
 }
 
 static void *lec_seq_next(struct seq_file *seq, void *v, loff_t *pos)
Signature · 1 line
2.39.5
author maintainer
#491 lore raw

PATCH 492/508 dt-bindings: i2c: nvidia,tegra20-i2c: Specify the required properties

In reply to Greg Kroah-Hartman

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

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

From: Akhil R <[email protected]>

commit 903cc7096db22f889d48e2cee8840709ce04fdac upstream.

Specify the properties which are essential and which are not for the Tegra I2C driver to function correctly. This was not added correctly when the TXT binding was converted to yaml. All the existing DT nodes have these properties already and hence this does not break the ABI.

dmas and dma-names which were specified as a must in the TXT binding is now made optional since the driver can work in PIO mode if dmas are missing.

Fixes f10a9b722f80 ("dt-bindings: i2c: tegra: Convert to json-schema”)
Cc <[email protected]> # v5.17+
---
 Documentation/devicetree/bindings/i2c/nvidia,tegra20-i2c.yaml |   24 +++++++++-
 1 file changed, 23 insertions(+), 1 deletion(-)
Documentation/devicetree/bindings/i2c/nvidia,tegra20-i2c.yaml +23 -1
@@ -103,7 +103,10 @@ properties:
 
   resets:
     items:
-      - description: module reset
+      - description:
+          Module reset. This property is optional for controllers in Tegra194,
+          Tegra234 etc where an internal software reset is available as an
+          alternative.
 
   reset-names:
     items:
@@ -119,6 +122,13 @@ properties:
       - const: rx
       - const: tx
 
+required:
+  - compatible
+  - reg
+  - interrupts
+  - clocks
+  - clock-names
+
 allOf:
   - $ref: /schemas/i2c/i2c-controller.yaml
   - if:
@@ -172,6 +182,18 @@ allOf:
           items:
             - description: phandle to the VENC power domain
 
+  - if:
+      not:
+        properties:
+          compatible:
+            contains:
+              enum:
+                - nvidia,tegra194-i2c
+    then:
+      required:
+        - resets
+        - reset-names
+
 unevaluatedProperties: false
 
 examples:
author maintainer
#492 lore raw

PATCH 493/508 platform/x86: ideapad-laptop: add missing Ideapad Pro 5 fn keys

In reply to Greg Kroah-Hartman

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

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

From: Renato Caldas <[email protected]>

commit 36e66be874a7ea9d28fb9757629899a8449b8748 upstream.

The scancodes for the Mic Mute and Airplane keys on the Ideapad Pro 5 (14AHP9 at least, probably the other variants too) are different and were not being picked up by the driver. This adds them to the keymap.

Apart from what is already supported, the remaining fn keys are unfortunately producing windows-specific key-combos.

---
 drivers/platform/x86/ideapad-laptop.c |    3 +++
 1 file changed, 3 insertions(+)
drivers/platform/x86/ideapad-laptop.c +3 -0
@@ -1191,6 +1191,9 @@ static const struct key_entry ideapad_ke
 	{ KE_KEY,	0x27 | IDEAPAD_WMI_KEY, { KEY_HELP } },
 	/* Refresh Rate Toggle */
 	{ KE_KEY,	0x0a | IDEAPAD_WMI_KEY, { KEY_DISPLAYTOGGLE } },
+	/* Specific to some newer models */
+	{ KE_KEY,	0x3e | IDEAPAD_WMI_KEY, { KEY_MICMUTE } },
+	{ KE_KEY,	0x3f | IDEAPAD_WMI_KEY, { KEY_RFKILL } },
 
 	{ KE_END },
 };
author maintainer
#493 lore raw

PATCH 494/508 ARM: dts: am335x-bone-common: Add GPIO PHY reset on revision C3 board

In reply to Greg Kroah-Hartman

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

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

From: Shengyu Qu <[email protected]>

commit 623cef652768860bd5f205fb7b741be278585fba upstream.

This patch adds ethernet PHY reset GPIO config for Beaglebone Black series boards with revision C3. This fixes a random phy startup failure bug discussed at [1]. The GPIO pin used for reset is not used on older revisions, so it is ok to apply to all board revisions. The reset timing was discussed and tested at [2].

[1] https://forum.digikey.com/t/ethernet-device-is-not-detecting-on-ubuntu-20-04-lts-on-bbg/19948
[2] https://forum.beagleboard.org/t/recognizing-a-beaglebone-black-rev-c3-board/31249/

Message-ID: <TY3P286MB26113797A3B2EC7E0348BBB2980FA@TY3P286MB2611.JPNP286.PROD.OUTLOOK.COM>

---
 arch/arm/boot/dts/am335x-bone-common.dtsi |    8 ++++++++
 1 file changed, 8 insertions(+)
arch/arm/boot/dts/am335x-bone-common.dtsi +8 -0
@@ -145,6 +145,8 @@
 			/* MDIO */
 			AM33XX_PADCONF(AM335X_PIN_MDIO, PIN_INPUT_PULLUP | SLEWCTRL_FAST, MUX_MODE0)
 			AM33XX_PADCONF(AM335X_PIN_MDC, PIN_OUTPUT_PULLUP, MUX_MODE0)
+			/* Added to support GPIO controlled PHY reset */
+			AM33XX_PADCONF(AM335X_PIN_UART0_CTSN, PIN_OUTPUT_PULLUP, MUX_MODE7)
 		>;
 	};
 
@@ -153,6 +155,8 @@
 			/* MDIO reset value */
 			AM33XX_PADCONF(AM335X_PIN_MDIO, PIN_INPUT_PULLDOWN, MUX_MODE7)
 			AM33XX_PADCONF(AM335X_PIN_MDC, PIN_INPUT_PULLDOWN, MUX_MODE7)
+			/* Added to support GPIO controlled PHY reset */
+			AM33XX_PADCONF(AM335X_PIN_UART0_CTSN, PIN_INPUT_PULLDOWN, MUX_MODE7)
 		>;
 	};
 
@@ -377,6 +381,10 @@
 
 	ethphy0: ethernet-phy@0 {
 		reg = <0>;
+		/* Support GPIO reset on revision C3 boards */
+		reset-gpios = <&gpio1 8 GPIO_ACTIVE_LOW>;
+		reset-assert-us = <300>;
+		reset-deassert-us = <6500>;
 	};
 };
 
author maintainer
#494 lore raw

PATCH 495/508 ARM: dts: am335x-bone-common: Increase MDIO reset deassert time

In reply to Greg Kroah-Hartman

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

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

From: Colin Foster <[email protected]>

commit b9bf5612610aa7e38d58fee16f489814db251c01 upstream.

Prior to commit df16c1c51d81 ("net: phy: mdio_device: Reset device only when necessary") MDIO reset deasserts were performed twice during boot. Now that the second deassert is no longer performed, device probe failures happen due to the change in timing with the following error message:

SMSC LAN8710/LAN8720: probe of 4a101000.mdio:00 failed with error -5

Restore the original effective timing, which resolves the probe failures.

---
 arch/arm/boot/dts/am335x-bone-common.dtsi |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/boot/dts/am335x-bone-common.dtsi +1 -1
@@ -384,7 +384,7 @@
 		/* Support GPIO reset on revision C3 boards */
 		reset-gpios = <&gpio1 8 GPIO_ACTIVE_LOW>;
 		reset-assert-us = <300>;
-		reset-deassert-us = <6500>;
+		reset-deassert-us = <13000>;
 	};
 };
 
author maintainer
#495 lore raw

PATCH 496/508 ARM: dts: am335x-bone-common: Increase MDIO reset deassert delay to 50ms

In reply to Greg Kroah-Hartman

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

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

From: Geert Uytterhoeven <[email protected]>

commit 929d8490f8790164f5f63671c1c58d6c50411cb2 upstream.

Commit b9bf5612610aa7e3 ("ARM: dts: am335x-bone-common: Increase MDIO reset deassert time") already increased the MDIO reset deassert delay from 6.5 to 13 ms, but this may still cause Ethernet PHY probe failures:

    SMSC LAN8710/LAN8720 4a101000.mdio:00: probe with driver SMSC LAN8710/LAN8720 failed with error -5

On BeagleBone Black Rev. C3, ETH_RESETn is controlled by an open-drain AND gate. It is pulled high by a 10K resistor, and has a 4.7µF capacitor to ground, giving an RC time constant of 47ms. As it takes 0.7RC to charge the capacitor above the threshold voltage of a CMOS input (VDD/2), the delay should be at least 33ms. Considering the typical tolerance of 20% on capacitors, 40ms would be safer. Add an additional safety margin and settle for 50ms.

---
 arch/arm/boot/dts/am335x-bone-common.dtsi |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/boot/dts/am335x-bone-common.dtsi +1 -1
@@ -384,7 +384,7 @@
 		/* Support GPIO reset on revision C3 boards */
 		reset-gpios = <&gpio1 8 GPIO_ACTIVE_LOW>;
 		reset-assert-us = <300>;
-		reset-deassert-us = <13000>;
+		reset-deassert-us = <50000>;
 	};
 };
 
author maintainer
#496 lore raw

PATCH 497/508 arm64: dts: ti: k3-j721e-sk: Add DT nodes for power regulators

In reply to Greg Kroah-Hartman

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

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

From: Yemike Abhilash Chandra <[email protected]>

commit 97b67cc102dc2cc8aa39a569c22a196e21af5a21 upstream.

Add device tree nodes for two power regulators on the J721E SK board.
vsys_5v0: A fixed regulator representing the 5V supply output from the LM61460 and vdd_sd_dv: A GPIO-controlled TLV71033 regulator.

J721E-SK schematics: https://www.ti.com/lit/zip/sprr438

---
 arch/arm64/boot/dts/ti/k3-j721e-sk.dts |   31 +++++++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)
arch/arm64/boot/dts/ti/k3-j721e-sk.dts +31 -0
@@ -175,6 +175,17 @@
 		regulator-boot-on;
 	};
 
+	vsys_5v0: fixedregulator-vsys5v0 {
+		/* Output of LM61460 */
+		compatible = "regulator-fixed";
+		regulator-name = "vsys_5v0";
+		regulator-min-microvolt = <5000000>;
+		regulator-max-microvolt = <5000000>;
+		vin-supply = <&vusb_main>;
+		regulator-always-on;
+		regulator-boot-on;
+	};
+
 	vdd_mmc1: fixedregulator-sd {
 		compatible = "regulator-fixed";
 		pinctrl-names = "default";
@@ -202,6 +213,20 @@
 			 <3300000 0x1>;
 	};
 
+	vdd_sd_dv: gpio-regulator-TLV71033 {
+		compatible = "regulator-gpio";
+		pinctrl-names = "default";
+		pinctrl-0 = <&vdd_sd_dv_pins_default>;
+		regulator-name = "tlv71033";
+		regulator-min-microvolt = <1800000>;
+		regulator-max-microvolt = <3300000>;
+		regulator-boot-on;
+		vin-supply = <&vsys_5v0>;
+		gpios = <&main_gpio0 118 GPIO_ACTIVE_HIGH>;
+		states = <1800000 0x0>,
+			 <3300000 0x1>;
+	};
+
 	dp_pwr_3v3: fixedregulator-dp-prw {
 		compatible = "regulator-fixed";
 		regulator-name = "dp-pwr";
@@ -455,6 +480,12 @@
 		>;
 	};
 
+	vdd_sd_dv_pins_default: vdd-sd-dv-default-pins {
+		pinctrl-single,pins = <
+			J721E_IOPAD(0x1dc, PIN_OUTPUT, 7) /* (Y1) SPI1_CLK.GPIO0_118 */
+		>;
+	};
+
 	wkup_i2c0_pins_default: wkup-i2c0-pins-default {
 		pinctrl-single,pins = <
 			J721E_WKUP_IOPAD(0xf8, PIN_INPUT_PULLUP, 0) /* (J25) WKUP_I2C0_SCL */
author maintainer
#497 lore raw

PATCH 498/508 arm64: dts: imx8mm: Drop sd-vsel-gpios from i.MX8M Mini Verdin SoM

In reply to Greg Kroah-Hartman

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

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

From: Marek Vasut <[email protected]>

commit 8bad8c923f217d238ba4f1a6d19d761e53bfbd26 upstream.

The VSELECT pin is configured as MX8MM_IOMUXC_GPIO1_IO04_USDHC2_VSELECT and not as a GPIO, drop the bogus sd-vsel-gpios property as the eSDHC block handles the VSELECT pin on its own.

---
 arch/arm64/boot/dts/freescale/imx8mm-verdin.dtsi |    1 -
 1 file changed, 1 deletion(-)
arch/arm64/boot/dts/freescale/imx8mm-verdin.dtsi +0 -1
@@ -372,7 +372,6 @@
 		pinctrl-names = "default";
 		pinctrl-0 = <&pinctrl_pmic>;
 		reg = <0x25>;
-		sd-vsel-gpios = <&gpio1 4 GPIO_ACTIVE_HIGH>;
 
 		/*
 		 * The bootloader is expected to switch on the I2C level shifter for the TLA2024 ADC
author maintainer
#498 lore raw

PATCH 499/508 serial: sh-sci: Increment the runtime usage counter for the earlycon device

In reply to Greg Kroah-Hartman

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

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

From: Claudiu Beznea <[email protected]>

commit 651dee03696e1dfde6d9a7e8664bbdcd9a10ea7f upstream.

In the sh-sci driver, serial ports are mapped to the sci_ports[] array, with earlycon mapped at index zero.

The uart_add_one_port() function eventually calls __device_attach(), which, in turn, calls pm_request_idle(). The identified code path is as follows:

uart_add_one_port() ->
serial_ctrl_register_port() ->

    serial_core_register_port() ->
      serial_core_port_device_add() ->
        serial_base_port_add() ->
          device_add() ->
            bus_probe_device() ->
              device_initial_probe() ->
                __device_attach() ->
                  // ...
                  if (dev->p->dead) {
                    // ...
                  } else if (dev->driver) {
                    // ...
                  } else {
                    // ...
                    pm_request_idle(dev);
                    // ...
                  }

The earlycon device clocks are enabled by the bootloader. However, the pm_request_idle() call in __device_attach() disables the SCI port clocks while earlycon is still active.

The earlycon write function, serial_console_write(), calls sci_poll_put_char() via serial_console_putchar(). If the SCI port clocks are disabled, writing to earlycon may sometimes cause the SR.TDFE bit to remain unset indefinitely, causing the while loop in sci_poll_put_char() to never exit. On single-core SoCs, this can result in the system being blocked during boot when this issue occurs.

To resolve this, increment the runtime PM usage counter for the earlycon SCI device before registering the UART port.

Fixes 0b0cced19ab1 ("serial: sh-sci: Add CONFIG_SERIAL_EARLYCON support")
---
 drivers/tty/serial/sh-sci.c |   16 ++++++++++++++++
 1 file changed, 16 insertions(+)
drivers/tty/serial/sh-sci.c +16 -0
@@ -3352,6 +3352,22 @@ static int sci_probe_single(struct platf
 
 	if (sci_uart_earlycon && sci_ports[0].port.mapbase == sci_res->start) {
 		/*
+		 * In case:
+		 * - this is the earlycon port (mapped on index 0 in sci_ports[]) and
+		 * - it now maps to an alias other than zero and
+		 * - the earlycon is still alive (e.g., "earlycon keep_bootcon" is
+		 *   available in bootargs)
+		 *
+		 * we need to avoid disabling clocks and PM domains through the runtime
+		 * PM APIs called in __device_attach(). For this, increment the runtime
+		 * PM reference counter (the clocks and PM domains were already enabled
+		 * by the bootloader). Otherwise the earlycon may access the HW when it
+		 * has no clocks enabled leading to failures (infinite loop in
+		 * sci_poll_put_char()).
+		 */
+		pm_runtime_get_noresume(&dev->dev);
+
+		/*
 		 * Skip cleanup the sci_port[0] in early_console_exit(), this
 		 * port is the same as the earlycon one.
 		 */
author maintainer
#499 lore raw

PATCH 500/508 Revert "cpufreq: tegra186: Share policy per cluster"

In reply to Greg Kroah-Hartman

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

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

From: Jon Hunter <[email protected]>

This reverts commit 89172666228de1cefcacf5bc6f61c6281751d2ed which is upstream commit be4ae8c19492cd6d5de61ccb34ffb3f5ede5eec8.

This commit is causing a suspend regression on Tegra186 Jetson TX2 with Linux v6.12.y kernels. This is not seen with Linux v6.15 that includes this change but indicates that there are there changes missing. Therefore, revert this change.

Fixes 89172666228d ("cpufreq: tegra186: Share policy per cluster")
---
 drivers/cpufreq/tegra186-cpufreq.c |    7 -------
 1 file changed, 7 deletions(-)
drivers/cpufreq/tegra186-cpufreq.c +0 -7
@@ -73,18 +73,11 @@ static int tegra186_cpufreq_init(struct
 {
 	struct tegra186_cpufreq_data *data = cpufreq_get_driver_data();
 	unsigned int cluster = data->cpus[policy->cpu].bpmp_cluster_id;
-	u32 cpu;
 
 	policy->freq_table = data->clusters[cluster].table;
 	policy->cpuinfo.transition_latency = 300 * 1000;
 	policy->driver_data = NULL;
 
-	/* set same policy for all cpus in a cluster */
-	for (cpu = 0; cpu < ARRAY_SIZE(tegra186_cpus); cpu++) {
-		if (data->cpus[cpu].bpmp_cluster_id == cluster)
-			cpumask_set_cpu(cpu, policy->cpus);
-	}
-
 	return 0;
 }
 
author maintainer
#500 lore raw

PATCH 501/508 smb: client: fix first command failure during re-negotiation

In reply to Greg Kroah-Hartman

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

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

From: zhangjian <[email protected]>

commit 34331d7beed7576acfc98e991c39738b96162499 upstream.

after fabc4ed200f9, server_unresponsive add a condition to check whether client need to reconnect depending on server->lstrp. When client failed to reconnect for some time and abort connection, server->lstrp is updated for the last time. In the following scene, server->lstrp is too old. This cause next command failure in re-negotiation rather than waiting for re-negotiation done.

1. mount -t cifs -o username=Everyone,echo_internal=10 //$server_ip/export /mnt
2. ssh $server_ip "echo b > /proc/sysrq-trigger &"
3. ls /mnt
4. sleep 21s
5. ssh $server_ip "service firewalld stop"
6. ls # return EHOSTDOWN

If the interval between 5 and 6 is too small, 6 may trigger sending negotiation request. Before backgrounding cifsd thread try to receive negotiation response from server in cifs_readv_from_socket, server_unresponsive may trigger cifs_reconnect which cause 6 to be failed:

ls thread
----------------
smb2_negotiate

    server->tcpStatus = CifsInNegotiate
    compound_send_recv
      wait_for_compound_request

cifsd thread
----------------
cifs_readv_from_socket
server_unresponsive
server->tcpStatus == CifsInNegotiate && jiffies > server->lstrp + 20s
cifs_reconnect
cifs_abort_connection: mid_state = MID_RETRY_NEEDED

ls thread
----------------
cifs_sync_mid_result return EAGAIN
smb2_negotiate return EHOSTDOWN

Though server->lstrp means last server response time, it is updated in cifs_abort_connection and cifs_get_tcp_session. We can also update server->lstrp before switching into CifsInNegotiate state to avoid failure in 6.

Fixes 7ccc1465465d ("smb: client: fix hang in wait_for_response() for negproto")
---
 fs/smb/client/connect.c |    1 +
 1 file changed, 1 insertion(+)
fs/smb/client/connect.c +1 -0
@@ -4217,6 +4217,7 @@ retry:
 		return 0;
 	}
 
+	server->lstrp = jiffies;
 	server->tcpStatus = CifsInNegotiate;
 	spin_unlock(&server->srv_lock);
 
author maintainer
#501 lore raw

PATCH 502/508 platform/loongarch: laptop: Add backlight power control support

In reply to Greg Kroah-Hartman

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

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

From: Yao Zi <[email protected]>

commit 53c762b47f726e4079a1f06f684bce2fc0d56fba upstream.

loongson_laptop_turn_{on,off}_backlight() are designed for controlling the power of the backlight, but they aren't really used in the driver previously.

Unify these two functions since they only differ in arguments passed to ACPI method, and wire up loongson_laptop_backlight_update() to update the power state of the backlight as well. Tested on the TongFang L860-T2 Loongson-3A5000 laptop.

Fixes 6246ed09111f ("LoongArch: Add ACPI-based generic laptop driver")
---
 drivers/platform/loongarch/loongson-laptop.c |   73 +++++++++++++--------------
 1 file changed, 37 insertions(+), 36 deletions(-)
drivers/platform/loongarch/loongson-laptop.c +37 -36
@@ -56,8 +56,7 @@ static struct input_dev *generic_inputde
 static acpi_handle hotkey_handle;
 static struct key_entry hotkey_keycode_map[GENERIC_HOTKEY_MAP_MAX];
 
-int loongson_laptop_turn_on_backlight(void);
-int loongson_laptop_turn_off_backlight(void);
+static bool bl_powered;
 static int loongson_laptop_backlight_update(struct backlight_device *bd);
 
 /* 2. ACPI Helpers and device model */
@@ -354,16 +353,42 @@ static int ec_backlight_level(u8 level)
 	return level;
 }
 
+static int ec_backlight_set_power(bool state)
+{
+	int status;
+	union acpi_object arg0 = { ACPI_TYPE_INTEGER };
+	struct acpi_object_list args = { 1, &arg0 };
+
+	arg0.integer.value = state;
+	status = acpi_evaluate_object(NULL, "\\BLSW", &args, NULL);
+	if (ACPI_FAILURE(status)) {
+		pr_info("Loongson lvds error: 0x%x\n", status);
+		return -EIO;
+	}
+
+	return 0;
+}
+
 static int loongson_laptop_backlight_update(struct backlight_device *bd)
 {
-	int lvl = ec_backlight_level(bd->props.brightness);
+	bool target_powered = !backlight_is_blank(bd);
+	int ret = 0, lvl = ec_backlight_level(bd->props.brightness);
 
 	if (lvl < 0)
 		return -EIO;
+
 	if (ec_set_brightness(lvl))
 		return -EIO;
 
-	return 0;
+	if (target_powered != bl_powered) {
+		ret = ec_backlight_set_power(target_powered);
+		if (ret < 0)
+			return ret;
+
+		bl_powered = target_powered;
+	}
+
+	return ret;
 }
 
 static int loongson_laptop_get_brightness(struct backlight_device *bd)
@@ -384,7 +409,7 @@ static const struct backlight_ops backli
 
 static int laptop_backlight_register(void)
 {
-	int status = 0;
+	int status = 0, ret;
 	struct backlight_properties props;
 
 	memset(&props, 0, sizeof(props));
@@ -392,44 +417,20 @@ static int laptop_backlight_register(voi
 	if (!acpi_evalf(hotkey_handle, &status, "ECLL", "d"))
 		return -EIO;
 
+	ret = ec_backlight_set_power(true);
+	if (ret)
+		return ret;
+
+	bl_powered = true;
+
 	props.max_brightness = status;
 	props.brightness = ec_get_brightness();
+	props.power = FB_BLANK_UNBLANK;
 	props.type = BACKLIGHT_PLATFORM;
 
 	backlight_device_register("loongson_laptop",
 				NULL, NULL, &backlight_laptop_ops, &props);
 
-	return 0;
-}
-
-int loongson_laptop_turn_on_backlight(void)
-{
-	int status;
-	union acpi_object arg0 = { ACPI_TYPE_INTEGER };
-	struct acpi_object_list args = { 1, &arg0 };
-
-	arg0.integer.value = 1;
-	status = acpi_evaluate_object(NULL, "\\BLSW", &args, NULL);
-	if (ACPI_FAILURE(status)) {
-		pr_info("Loongson lvds error: 0x%x\n", status);
-		return -ENODEV;
-	}
-
-	return 0;
-}
-
-int loongson_laptop_turn_off_backlight(void)
-{
-	int status;
-	union acpi_object arg0 = { ACPI_TYPE_INTEGER };
-	struct acpi_object_list args = { 1, &arg0 };
-
-	arg0.integer.value = 0;
-	status = acpi_evaluate_object(NULL, "\\BLSW", &args, NULL);
-	if (ACPI_FAILURE(status)) {
-		pr_info("Loongson lvds error: 0x%x\n", status);
-		return -ENODEV;
-	}
 
 	return 0;
 }
author maintainer
#502 lore raw

PATCH 503/508 s390/pci: Fix __pcilg_mio_inuser() inline assembly

In reply to Greg Kroah-Hartman

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

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

From: Heiko Carstens <[email protected]>

commit c4abe6234246c75cdc43326415d9cff88b7cf06c upstream.

Use "a" constraint for the shift operand of the __pcilg_mio_inuser() inline assembly. The used "d" constraint allows the compiler to use any general purpose register for the shift operand, including register zero.

If register zero is used this my result in incorrect code generation:

 8f6:   a7 0a ff f8             ahi     %r0,-8
 8fa:   eb 32 00 00 00 0c       srlg    %r3,%r2,0  <----

If register zero is selected to contain the shift value, the srlg instruction ignores the contents of the register and always shifts zero bits. Therefore use the "a" constraint which does not permit to select register zero.

Fixes f058599e22d5 ("s390/pci: Fix s390_mmio_read/write with MIO")
---
 arch/s390/pci/pci_mmio.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/s390/pci/pci_mmio.c +1 -1
@@ -223,7 +223,7 @@ static inline int __pcilg_mio_inuser(
 		[ioaddr_len] "+&d" (ioaddr_len.pair),
 		[cc] "+d" (cc), [val] "=d" (val),
 		[dst] "+a" (dst), [cnt] "+d" (cnt), [tmp] "=d" (tmp),
-		[shift] "+d" (shift)
+		[shift] "+a" (shift)
 		:: "cc", "memory");
 
 	/* did we write everything to the user space buffer? */
author maintainer
#503 lore raw

PATCH 504/508 perf: Fix sample vs do_exit()

In reply to Greg Kroah-Hartman

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

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

From: Peter Zijlstra <[email protected]>

[ Upstream commit 4f6fc782128355931527cefe3eb45338abd8ab39 ]

Baisheng Gao reported an ARM64 crash, which Mark decoded as being a synchronous external abort -- most likely due to trying to access MMIO in bad ways.

The crash further shows perf trying to do a user stack sample while in exit_mmap()'s tlb_finish_mmu() -- i.e. while tearing down the address space it is trying to access.

It turns out that we stop perf after we tear down the userspace mm; a receipie for disaster, since perf likes to access userspace for various reasons.

Flip this order by moving up where we stop perf in do_exit().

Additionally, harden PERF_SAMPLE_CALLCHAIN and PERF_SAMPLE_STACK_USER to abort when the current task does not have an mm (exit_mm() makes sure to set current->mm = NULL; before commencing with the actual teardown). Such that CPU wide events don't trip on this same problem.

Fixes c5ebcedb566e ("perf: Add ability to attach user stack dump to sample")
---
 kernel/events/core.c |  7 +++++++
 kernel/exit.c        | 17 +++++++++--------
 2 files changed, 16 insertions(+), 8 deletions(-)
kernel/events/core.c +7 -0
index 25df8971a56ed..2761db0365ddc 100644
@@ -6800,6 +6800,10 @@ perf_sample_ustack_size(u16 stack_size, u16 header_size,
 	if (!regs)
 		return 0;
 
+	/* No mm, no stack, no dump. */
+	if (!current->mm)
+		return 0;
+
 	/*
 	 * Check if we fit in with the requested stack size into the:
 	 * - TASK_SIZE
@@ -7485,6 +7489,9 @@ perf_callchain(struct perf_event *event, struct pt_regs *regs)
 	const u32 max_stack = event->attr.sample_max_stack;
 	struct perf_callchain_entry *callchain;
 
+	if (!current->mm)
+		user = false;
+
 	if (!kernel && !user)
 		return &__empty_callchain;
 
kernel/exit.c +9 -8
index 156283b3c1bf6..10af26a4ab010 100644
@@ -855,6 +855,15 @@ void __noreturn do_exit(long code)
 	tsk->exit_code = code;
 	taskstats_exit(tsk, group_dead);
 
+	/*
+	 * Since sampling can touch ->mm, make sure to stop everything before we
+	 * tear it down.
+	 *
+	 * Also flushes inherited counters to the parent - before the parent
+	 * gets woken up by child-exit notifications.
+	 */
+	perf_event_exit_task(tsk);
+
 	exit_mm();
 
 	if (group_dead)
@@ -871,14 +880,6 @@ void __noreturn do_exit(long code)
 	exit_task_work(tsk);
 	exit_thread(tsk);
 
-	/*
-	 * Flush inherited counters to the parent - before the parent
-	 * gets woken up by child-exit notifications.
-	 *
-	 * because of cgroup mode, must be called before cgroup_exit()
-	 */
-	perf_event_exit_task(tsk);
-
 	sched_autogroup_exit_task(tsk);
 	cgroup_exit(tsk);
 
Signature · 1 line
2.39.5
author maintainer
#504 lore raw

PATCH 505/508 arm64/ptrace: Fix stack-out-of-bounds read in regs_get_kernel_stack_nth()

In reply to Greg Kroah-Hartman

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

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

From: Tengda Wu <[email protected]>

[ Upstream commit 39dfc971e42d886e7df01371cd1bef505076d84c ]

KASAN reports a stack-out-of-bounds read in regs_get_kernel_stack_nth().

Call Trace:
[   97.283505] BUG: KASAN: stack-out-of-bounds in regs_get_kernel_stack_nth+0xa8/0xc8
[   97.284677] Read of size 8 at addr ffff800089277c10 by task 1.sh/2550
[   97.285732]
[   97.286067] CPU: 7 PID: 2550 Comm: 1.sh Not tainted 6.6.0+ #11
[   97.287032] Hardware name: linux,dummy-virt (DT)
[   97.287815] Call trace:
[   97.288279]  dump_backtrace+0xa0/0x128
[   97.288946]  show_stack+0x20/0x38
[   97.289551]  dump_stack_lvl+0x78/0xc8
[   97.290203]  print_address_description.constprop.0+0x84/0x3c8
[   97.291159]  print_report+0xb0/0x280
[   97.291792]  kasan_report+0x84/0xd0
[   97.292421]  __asan_load8+0x9c/0xc0
[   97.293042]  regs_get_kernel_stack_nth+0xa8/0xc8
[   97.293835]  process_fetch_insn+0x770/0xa30
[   97.294562]  kprobe_trace_func+0x254/0x3b0
[   97.295271]  kprobe_dispatcher+0x98/0xe0
[   97.295955]  kprobe_breakpoint_handler+0x1b0/0x210
[   97.296774]  call_break_hook+0xc4/0x100
[   97.297451]  brk_handler+0x24/0x78
[   97.298073]  do_debug_exception+0xac/0x178
[   97.298785]  el1_dbg+0x70/0x90
[   97.299344]  el1h_64_sync_handler+0xcc/0xe8
[   97.300066]  el1h_64_sync+0x78/0x80
[   97.300699]  kernel_clone+0x0/0x500
[   97.301331]  __arm64_sys_clone+0x70/0x90
[   97.302084]  invoke_syscall+0x68/0x198
[   97.302746]  el0_svc_common.constprop.0+0x11c/0x150
[   97.303569]  do_el0_svc+0x38/0x50
[   97.304164]  el0_svc+0x44/0x1d8
[   97.304749]  el0t_64_sync_handler+0x100/0x130
[   97.305500]  el0t_64_sync+0x188/0x190
[   97.306151]
[   97.306475] The buggy address belongs to stack of task 1.sh/2550
[   97.307461]  and is located at offset 0 in frame:
[   97.308257]  __se_sys_clone+0x0/0x138
[   97.308910]
[   97.309241] This frame has 1 object:
[   97.309873]  [48, 184) 'args'
[   97.309876]
[   97.310749] The buggy address belongs to the virtual mapping at
[   97.310749]  [ffff800089270000, ffff800089279000) created by:
[   97.310749]  dup_task_struct+0xc0/0x2e8
[   97.313347]
[   97.313674] The buggy address belongs to the physical page:
[   97.314604] page: refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x14f69a
[   97.315885] flags: 0x15ffffe00000000(node=1|zone=2|lastcpupid=0xfffff)
[   97.316957] raw: 015ffffe00000000 0000000000000000 dead000000000122 0000000000000000
[   97.318207] raw: 0000000000000000 0000000000000000 00000001ffffffff 0000000000000000
[   97.319445] page dumped because: kasan: bad access detected
[   97.320371]
[   97.320694] Memory state around the buggy address:
[   97.321511]  ffff800089277b00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[   97.322681]  ffff800089277b80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[   97.323846] >ffff800089277c00: 00 00 f1 f1 f1 f1 f1 f1 00 00 00 00 00 00 00 00
[   97.325023]                          ^
[   97.325683]  ffff800089277c80: 00 00 00 00 00 00 00 00 00 f3 f3 f3 f3 f3 f3 f3
[   97.326856]  ffff800089277d00: f3 f3 00 00 00 00 00 00 00 00 00 00 00 00 00 00

This issue seems to be related to the behavior of some gcc compilers and was also fixed on the s390 architecture before:

commit d93a855c31b7 ("s390/ptrace: Avoid KASAN false positives in regs_get_kernel_stack_nth()")

As described in that commit, regs_get_kernel_stack_nth() has confirmed that addr is on the stack, so reading the value at *addr should be allowed. Use READ_ONCE_NOCHECK() helper to silence the KASAN check for this case.

Fixes 0a8ea52c3eb1 ("arm64: Add HAVE_REGS_AND_STACK_ACCESS_API feature")
[will: Use '*addr' as the argument to READ_ONCE_NOCHECK()]
---
 arch/arm64/kernel/ptrace.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/kernel/ptrace.c +1 -1
index 2f1f86d916129..3025d14c253de 100644
@@ -139,7 +139,7 @@ unsigned long regs_get_kernel_stack_nth(struct pt_regs *regs, unsigned int n)
 
 	addr += n;
 	if (regs_within_kernel_stack(regs, (unsigned long)addr))
-		return *addr;
+		return READ_ONCE_NOCHECK(*addr);
 	else
 		return 0;
 }
Signature · 1 line
2.39.5
author maintainer
#505 lore raw

PATCH 506/508 scsi: elx: efct: Fix memory leak in efct_hw_parse_filter()

In reply to Greg Kroah-Hartman

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

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

From: Vitaliy Shevtsov <[email protected]>

[ Upstream commit 2a8a5a5dd06eef580f9818567773fd75057cb875 ]

strsep() modifies the address of the pointer passed to it so that it no longer points to the original address. This means kfree() gets the wrong pointer.

Fix this by passing unmodified pointer returned from kstrdup() to kfree().

Found by Linux Verification Center (linuxtesting.org) with Svace.

Fixes 4df84e846624 ("scsi: elx: efct: Driver initialization routines")
---
 drivers/scsi/elx/efct/efct_hw.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/scsi/elx/efct/efct_hw.c +3 -2
index 5a5525054d71c..5b079b8b7a082 100644
@@ -1120,7 +1120,7 @@ int
 efct_hw_parse_filter(struct efct_hw *hw, void *value)
 {
 	int rc = 0;
-	char *p = NULL;
+	char *p = NULL, *pp = NULL;
 	char *token;
 	u32 idx = 0;
 
@@ -1132,6 +1132,7 @@ efct_hw_parse_filter(struct efct_hw *hw, void *value)
 		efc_log_err(hw->os, "p is NULL\n");
 		return -ENOMEM;
 	}
+	pp = p;
 
 	idx = 0;
 	while ((token = strsep(&p, ",")) && *token) {
@@ -1144,7 +1145,7 @@ efct_hw_parse_filter(struct efct_hw *hw, void *value)
 		if (idx == ARRAY_SIZE(hw->config.filter_def))
 			break;
 	}
-	kfree(p);
+	kfree(pp);
 
 	return rc;
 }
Signature · 1 line
2.39.5
author maintainer
#506 lore raw

PATCH 507/508 RISC-V: KVM: Fix the size parameter check in SBI SFENCE calls

In reply to Greg Kroah-Hartman

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

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

From: Anup Patel <[email protected]>

[ Upstream commit 6aba0cb5bba6141158d5449f2cf53187b7f755f9 ]

As-per the SBI specification, an SBI remote fence operation applies to the entire address space if either:
1) start_addr and size are both 0
2) size is equal to 2^XLEN-1

Quoted 1 line

From the above, only #1 is checked by SBI SFENCE calls so fix the

size parameter check in SBI SFENCE calls to cover #2 as well.

Fixes 13acfec2dbcc ("RISC-V: KVM: Add remote HFENCE functions based on VCPU requests")
---
 arch/riscv/kvm/vcpu_sbi_replace.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/riscv/kvm/vcpu_sbi_replace.c +2 -2
index 4c034d8a606a1..304d29383e6c0 100644
@@ -91,14 +91,14 @@ static int kvm_sbi_ext_rfence_handler(struct kvm_vcpu *vcpu, struct kvm_run *run
 		kvm_riscv_fence_i(vcpu->kvm, hbase, hmask);
 		break;
 	case SBI_EXT_RFENCE_REMOTE_SFENCE_VMA:
-		if (cp->a2 == 0 && cp->a3 == 0)
+		if ((cp->a2 == 0 && cp->a3 == 0) || cp->a3 == -1UL)
 			kvm_riscv_hfence_vvma_all(vcpu->kvm, hbase, hmask);
 		else
 			kvm_riscv_hfence_vvma_gva(vcpu->kvm, hbase, hmask,
 						  cp->a2, cp->a3, PAGE_SHIFT);
 		break;
 	case SBI_EXT_RFENCE_REMOTE_SFENCE_VMA_ASID:
-		if (cp->a2 == 0 && cp->a3 == 0)
+		if ((cp->a2 == 0 && cp->a3 == 0) || cp->a3 == -1UL)
 			kvm_riscv_hfence_vvma_asid_all(vcpu->kvm,
 						       hbase, hmask, cp->a4);
 		else
Signature · 1 line
2.39.5
author maintainer
#507 lore raw

PATCH 508/508 RISC-V: KVM: Dont treat SBI HFENCE calls as NOPs

In reply to Greg Kroah-Hartman

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

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

From: Anup Patel <[email protected]>

[ Upstream commit 2e7be162996640bbe3b6da694cc064c511b8a5d9 ]

The SBI specification clearly states that SBI HFENCE calls should return SBI_ERR_NOT_SUPPORTED when one of the target hart doesn’t support hypervisor extension (aka nested virtualization in-case of KVM RISC-V).

Fixes c7fa3c48de86 ("RISC-V: KVM: Treat SBI HFENCE calls as NOPs")
---
 arch/riscv/kvm/vcpu_sbi_replace.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/riscv/kvm/vcpu_sbi_replace.c +2 -2
index 304d29383e6c0..1b8b3fa5f74ab 100644
@@ -113,9 +113,9 @@ static int kvm_sbi_ext_rfence_handler(struct kvm_vcpu *vcpu, struct kvm_run *run
 	case SBI_EXT_RFENCE_REMOTE_HFENCE_VVMA_ASID:
 		/*
 		 * Until nested virtualization is implemented, the
-		 * SBI HFENCE calls should be treated as NOPs
+		 * SBI HFENCE calls should return not supported
+		 * hence fallthrough.
 		 */
-		break;
 	default:
 		ret = -EOPNOTSUPP;
 	}
Signature · 1 line
2.39.5
Tested-by on the v1 series
#508 lore raw
In reply to Greg Kroah-Hartman
Quoted 9 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 6.1.142 release. There are 508 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed, 25 Jun 2025 13:05:52 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 8 lines
	https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.142-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

thanks,

greg k-h

Built and booted successfully on RISC-V RV64 (HiFive Unmatched).