3.12.54-stable review

This is the start of the stable review cycle for the 3.12.54 release. There are 64 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 Mon Feb 15 14:54:53 CET 2016. Anything received after that time might be too late.

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

	http://kernel.org/pub/linux/kernel/people/jirislaby/stable-review/patch-3.12.54-rc1.xz

and the diffstat can be found below.

thanks,
js

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

Alex Deucher (1):
drm/radeon: cypress_dpm: Fix unused variable warning when
CONFIG_ACPI=n

Andrew Morton (1):
openrisc: fix CONFIG_UID16 setting

Andrey Ryabinin (1):
ipv6/addrlabel: fix ip6addrlbl_get()

Arnd Bergmann (2):
mISDN: avoid arch specific __builtin_return_address call
arm64: fix building without CONFIG_UID16

Behan Webster (1):
ARM: 8158/1: LLVMLinux: use static inline in ARM ftrace.h

Boqun Feng (2):
powerpc: Make value-returning atomics fully ordered
powerpc: Make {cmp}xchg* and their atomic_ versions fully ordered

Chunfeng Yun (1):
usb: xhci: fix config fail of FS hub behind a HS hub with MTT

Dan Carpenter (1):
USB: ipaq.c: fix a timeout loop

Dan Streetman (1):
xfrm: dst_entries_init() per-net dst_ops

David Henningsson (1):
ALSA: hda - Add inverted dmic for Packard Bell DOTS

Eric Dumazet (3):
ipv6: tcp: add rcu locking in tcp_v6_send_synack()
phonet: properly unshare skbs in phonet_rcv()
ipv6: update skb->csum when CE mark is propagated

Fabio Estevam (1):
drm: radeon: ni_dpm: Fix unused variable warning when CONFIG_ACPI=n

Florian Westphal (1):
connector: bump skb->users before callback invocation

Greg Kroah-Hartman (1):
xhci: fix placement of call to usb_disabled()

Guenter Roeck (1):
mn10300: Select CONFIG_HAVE_UID16 to fix build failure

Hannes Frederic Sowa (1):
bridge: Only call /sbin/bridge-stp for the initial network namespace

Herbert Xu (1):
crypto: algif_hash - Only export and import on sockets with data

Ido Schimmel (1):
team: Replace rcu_read_lock with a mutex in team_vlan_rx_kill_vid

Ioan-Adrian Ratiu (1):
HID: usbhid: fix recursive deadlock

Jiri Slaby (1):
x86: vvar, fix excessive gcc-6 DECLARE_VVAR warnings

Joe Perches (1):
compiler-gcc: integrate the various compiler-gcc[345].h files

John Blackwood (1):
arm64: Clear out any singlestep state on a ptrace detach operation

Kees Cook (1):
lkdtm: adjust recursion size to avoid warnings

Linus Torvalds (1):
vmstat: explicitly schedule per-cpu work on the CPU we need it to run
on

Marc Zyngier (1):
arm64: KVM: Fix AArch32 to AArch64 register mapping

Mario Kleiner (1):
ALSA: hda/realtek - Fix silent headphone output on MacPro 4,1 (v2)

Michael Neuling (1):
powerpc/tm: Block signal return setting invalid MSR state

Mike Snitzer (1):
dm btree: fix leak of bufio-backed block in btree_split_sibling error
path

Mikulas Patocka (1):
parisc iommu: fix panic due to trying to allocate too large region

Neal Cardwell (1):
tcp_yeah: don't set ssthresh below 2

Nicolas Boichat (2):
ALSA: seq: Fix snd_seq_call_port_info_ioctl in compat mode
ALSA: pcm: Fix snd_pcm_hw_params struct copy in compat mode

Nikesh Oswal (1):
ASoC: arizona: Fix bclk for sample rates that are multiple of 4kHz

Oliver Freyermuth (1):
USB: cp210x: add ID for ELV Marble Sound Board 1

Oliver Neukum (1):
xhci: refuse loading if nousb is used

Paul Bolle (1):
RDMA/cxgb4: Fix gcc warning on 32-bit arch

Richard Purdie (1):
HID: core: Avoid uninitialized buffer access

Sachin Pandhare (1):
ASoC: wm8962: correct addresses for HPF_C_0/1

Sasha Levin (1):
net: sctp: prevent writes to cookie_hmac_alg from accessing invalid
memory

Seth Jennings (1):
drivers/base/memory.c: prohibit offlining of memory blocks with
missing sections

Steven Noonan (1):
compiler/gcc4+: Remove inaccurate comment about 'asm goto' miscompiles

Takashi Iwai (8):
ALSA: rme96: Fix unexpected volume reset after rate changes
ALSA: seq: Fix missing NULL check at remove_events ioctl
ALSA: seq: Fix race at timer setup and close
ALSA: timer: Harden slave timer list handling
ALSA: timer: Fix race among timer ioctls
ALSA: timer: Fix double unlink of active_list
ALSA: hrtimer: Fix stall by hrtimer_cancel()
ALSA: control: Avoid kernel warnings from tlv ioctl with numid 0

Tejun Heo (1):
Revert "workqueue: make sure delayed work run in local cpu"

Ulrich Weigand (1):
scripts/recordmcount.pl: support data in text section on powerpc

Uwe Kleine-König (1):
ARM: 8160/1: drop warning about return_address not using unwind tables

Vijay Pandurangan (1):
veth: don’t modify ip_summed; doing so treats packets with bad
checksums as good.

Vinod Koul (1):
ASoC: compress: Fix compress device direction check

Will Deacon (1):
arm64: mm: ensure that the zero page is visible to the page table
walker

Xin Long (1):
sctp: sctp should release assoc when sctp_make_abort_user return NULL
in sctp_close

Xiong Zhang (1):
ALSA: hda - Set SKL+ hda controller power at freeze() and thaw()

Yang Shi (1):
arm64: restore bogomips information in /proc/cpuinfo

libin (1):
recordmcount: Fix endianness handling bug for nop_mcount

willy tarreau (1):
unix: properly account for FDs passed over unix sockets

 arch/arm/include/asm/ftrace.h           |   2 +-
 arch/arm/kernel/return_address.c        |   9 ---
 arch/arm64/include/asm/kvm_emulate.h    |   8 ++-
 arch/arm64/kernel/ptrace.c              |   6 ++
 arch/arm64/kernel/setup.c               |   4 ++
 arch/arm64/kvm/inject_fault.c           |   2 +-
 arch/arm64/mm/mmu.c                     |   3 +
 arch/mn10300/Kconfig                    |   4 +-
 arch/openrisc/Kconfig                   |   4 +-
 arch/powerpc/include/asm/cmpxchg.h      |  16 ++---
 arch/powerpc/include/asm/reg.h          |   1 +
 arch/powerpc/include/asm/synch.h        |   2 +-
 arch/powerpc/kernel/signal_32.c         |  14 ++--
 arch/powerpc/kernel/signal_64.c         |   4 ++
 arch/x86/include/asm/vvar.h             |   2 +-
 crypto/algif_hash.c                     |  12 +++-
 drivers/base/memory.c                   |   4 ++
 drivers/connector/connector.c           |  11 +--
 drivers/gpu/drm/radeon/cypress_dpm.c    |   2 +
 drivers/gpu/drm/radeon/ni_dpm.c         |   2 +-
 drivers/hid/hid-core.c                  |   2 +-
 drivers/hid/usbhid/hid-core.c           |   4 +-
 drivers/infiniband/hw/cxgb4/mem.c       |   2 +-
 drivers/isdn/hardware/mISDN/mISDNipac.c |  12 +++-
 drivers/isdn/hardware/mISDN/w6692.c     |   6 +-
 drivers/md/persistent-data/dm-btree.c   |   4 +-
 drivers/misc/lkdtm.c                    |  30 ++++++---
 drivers/net/team/team.c                 |   6 +-
 drivers/net/veth.c                      |   6 --
 drivers/parisc/iommu-helpers.h          |  15 +++--
 drivers/usb/host/xhci.c                 |  12 ++++
 drivers/usb/serial/cp210x.c             |   1 +
 drivers/usb/serial/ipaq.c               |   3 +-
 include/linux/compiler-gcc.h            | 114 ++++++++++++++++++++++++++++++--
 include/linux/compiler-gcc3.h           |  23 -------
 include/linux/compiler-gcc4.h           |  88 ------------------------
 include/linux/compiler-gcc5.h           |  66 ------------------
 include/linux/sched.h                   |   1 +
 include/linux/syscalls.h                |   2 +-
 include/linux/types.h                   |   2 +-
 include/net/inet_ecn.h                  |  19 +++++-
 kernel/workqueue.c                      |   8 +--
 mm/vmstat.c                             |   3 +-
 net/bridge/br_stp_if.c                  |   5 +-
 net/ipv4/tcp_yeah.c                     |   2 +-
 net/ipv4/xfrm4_policy.c                 |  46 ++++++++++---
 net/ipv6/addrlabel.c                    |   2 +-
 net/ipv6/tcp_ipv6.c                     |   2 +
 net/ipv6/xfrm6_mode_tunnel.c            |   2 +-
 net/ipv6/xfrm6_policy.c                 |  53 ++++++++++-----
 net/phonet/af_phonet.c                  |   4 ++
 net/sctp/sm_statefuns.c                 |   6 +-
 net/sctp/socket.c                       |   3 +-
 net/sctp/sysctl.c                       |   2 +-
 net/unix/af_unix.c                      |  24 +++++--
 net/unix/garbage.c                      |  16 +++--
 net/xfrm/xfrm_policy.c                  |  38 -----------
 scripts/recordmcount.h                  |   2 +-
 scripts/recordmcount.pl                 |   3 +-
 sound/core/control.c                    |   2 +
 sound/core/hrtimer.c                    |   3 +-
 sound/core/pcm_compat.c                 |  13 ++--
 sound/core/seq/seq_clientmgr.c          |   2 +-
 sound/core/seq/seq_compat.c             |   9 +--
 sound/core/seq/seq_queue.c              |   2 +
 sound/core/timer.c                      |  52 ++++++++++-----
 sound/pci/hda/hda_intel.c               |  34 ++++++++++
 sound/pci/hda/patch_realtek.c           |  12 +++-
 sound/pci/rme96.c                       |  41 +++++++-----
 sound/soc/codecs/arizona.c              |   2 +-
 sound/soc/codecs/wm8962.c               |   4 +-
 sound/soc/soc-compress.c                |  23 ++++++-
 72 files changed, 543 insertions(+), 407 deletions(-)
 delete mode 100644 include/linux/compiler-gcc3.h
 delete mode 100644 include/linux/compiler-gcc4.h
 delete mode 100644 include/linux/compiler-gcc5.h

2.7.1

From: Takashi Iwai <[email protected]>

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

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

commit a74a821624c0c75388a193337babd17a8c02c740 upstream.

rme96 driver needs to reset DAC depending on the sample rate, and this results in resetting to the max volume suddenly. It's because of the missing call of snd_rme96_apply_dac_volume().

However, calling this function right after the DAC reset still may not work, and we need some delay before this call. Since the DAC reset and the procedure after that are performed in the spinlock, we delay the DAC volume restore at the end after the spinlock.

---
 sound/pci/rme96.c | 41 ++++++++++++++++++++++++++---------------
 1 file changed, 26 insertions(+), 15 deletions(-)

index bb9ebc5543d7..2da24272e6a5 100644
@@ -744,10 +744,11 @@ snd_rme96_playback_setrate(struct rme96 *rme96,
 	{
 		/* change to/from double-speed: reset the DAC (if available) */
 		snd_rme96_reset_dac(rme96);
+		return 1; /* need to restore volume */
 	} else {
 		writel(rme96->wcreg, rme96->iobase + RME96_IO_CONTROL_REGISTER);
+		return 0;
 	}
-	return 0;
 }
 
 static int

@@ -985,6 +986,7 @@ snd_rme96_playback_hw_params(struct snd_pcm_substream *substream,
 	struct rme96 *rme96 = snd_pcm_substream_chip(substream);
 	struct snd_pcm_runtime *runtime = substream->runtime;
 	int err, rate, dummy;
+	bool apply_dac_volume = false;
 
 	runtime->dma_area = (void __force *)(rme96->iobase +
 					     RME96_IO_PLAY_BUFFER);

@@ -998,24 +1000,26 @@ snd_rme96_playback_hw_params(struct snd_pcm_substream *substream,
 	{
                 /* slave clock */
                 if ((int)params_rate(params) != rate) {
-			spin_unlock_irq(&rme96->lock);
-			return -EIO;                    
-                }
-	} else if ((err = snd_rme96_playback_setrate(rme96, params_rate(params))) < 0) {
-		spin_unlock_irq(&rme96->lock);
-		return err;
-	}
-	if ((err = snd_rme96_playback_setformat(rme96, params_format(params))) < 0) {
-		spin_unlock_irq(&rme96->lock);
-		return err;
+			err = -EIO;
+			goto error;
+		}
+	} else {
+		err = snd_rme96_playback_setrate(rme96, params_rate(params));
+		if (err < 0)
+			goto error;
+		apply_dac_volume = err > 0; /* need to restore volume later? */
 	}
+
+	err = snd_rme96_playback_setformat(rme96, params_format(params));
+	if (err < 0)
+		goto error;
 	snd_rme96_setframelog(rme96, params_channels(params), 1);
 	if (rme96->capture_periodsize != 0) {
 		if (params_period_size(params) << rme96->playback_frlog !=
 		    rme96->capture_periodsize)
 		{
-			spin_unlock_irq(&rme96->lock);
-			return -EBUSY;
+			err = -EBUSY;
+			goto error;
 		}
 	}
 	rme96->playback_periodsize =

@@ -1026,9 +1030,16 @@ snd_rme96_playback_hw_params(struct snd_pcm_substream *substream,
 		rme96->wcreg &= ~(RME96_WCR_PRO | RME96_WCR_DOLBY | RME96_WCR_EMP);
 		writel(rme96->wcreg |= rme96->wcreg_spdif_stream, rme96->iobase + RME96_IO_CONTROL_REGISTER);
 	}
+
+	err = 0;
+ error:
 	spin_unlock_irq(&rme96->lock);
-		
-	return 0;
+	if (apply_dac_volume) {
+		usleep_range(3000, 10000);
+		snd_rme96_apply_dac_volume(rme96);
+	}
+
+	return err;
 }
 
 static int

2.7.1

From: David Henningsson <[email protected]>

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

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

commit 02f6ff90400d055f08b0ba0b5f0707630b6faed7 upstream.

On the internal mic of the Packard Bell DOTS, one channel has an inverted signal. Add a quirk to fix this up.

---
 sound/pci/hda/patch_realtek.c | 1 +
 1 file changed, 1 insertion(+)

index 1ec93efc8253..6d9a42a16a16 100644
@@ -4956,6 +4956,7 @@ static const struct hda_fixup alc662_fixups[] = {
 static const struct snd_pci_quirk alc662_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x1019, 0x9087, "ECS", ALC662_FIXUP_ASUS_MODE2),
 	SND_PCI_QUIRK(0x1025, 0x022f, "Acer Aspire One", ALC662_FIXUP_INV_DMIC),
+	SND_PCI_QUIRK(0x1025, 0x0241, "Packard Bell DOTS", ALC662_FIXUP_INV_DMIC),
 	SND_PCI_QUIRK(0x1025, 0x0308, "Acer Aspire 8942G", ALC662_FIXUP_ASPIRE),
 	SND_PCI_QUIRK(0x1025, 0x031c, "Gateway NV79", ALC662_FIXUP_SKU_IGNORE),
 	SND_PCI_QUIRK(0x1025, 0x0349, "eMachines eM250", ALC662_FIXUP_INV_DMIC),

2.7.1

From: Xiong Zhang <[email protected]>

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

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

commit 3e6db33aaf1d42a30339f831ec4850570d6cc7a3 upstream.

It takes three minutes to enter into hibernation on some OEM SKL machines and we see many codec spurious response after thaw() opertion. This is because HDA is still in D0 state after freeze() call and pci_pm_freeze/pci_pm_freeze_noirq() don't set D3 hot in pci_bus driver. It seems bios still access HDA when system enter into freeze state, HDA will receive codec response interrupt immediately after thaw() call. Because of this unexpected interrupt, HDA enter into a abnormal state and slow down the system enter into hibernation.

In this patch, we put HDA into D3 hot state in azx_freeze_noirq() and put HDA into D0 state in azx_thaw_noirq().

V2: Only apply this fix to SKL+
Fix compile error when CONFIG_PM_SLEEP isn't defined

[Yet another fix for CONFIG_PM_SLEEP ifdef and the additional comment
by tiwai]

---
 sound/pci/hda/hda_intel.c | 34 ++++++++++++++++++++++++++++++++++
 1 file changed, 34 insertions(+)

index baf12f1a2820..6a5e36dc23e5 100644
@@ -1180,6 +1180,36 @@ static unsigned int azx_get_response(struct hda_bus *bus,
 		return azx_rirb_get_response(bus, addr);
 }
 
+#ifdef CONFIG_PM_SLEEP
+/* put codec down to D3 at hibernation for Intel SKL+;
+ * otherwise BIOS may still access the codec and screw up the driver
+ */
+#define IS_SKL(pci) ((pci)->vendor == 0x8086 && (pci)->device == 0xa170)
+#define IS_SKL_LP(pci) ((pci)->vendor == 0x8086 && (pci)->device == 0x9d70)
+#define IS_BXT(pci) ((pci)->vendor == 0x8086 && (pci)->device == 0x5a98)
+#define IS_SKL_PLUS(pci) (IS_SKL(pci) || IS_SKL_LP(pci) || IS_BXT(pci))
+
+static int azx_freeze_noirq(struct device *dev)
+{
+	struct pci_dev *pci = to_pci_dev(dev);
+
+	if (IS_SKL_PLUS(pci))
+		pci_set_power_state(pci, PCI_D3hot);
+
+	return 0;
+}
+
+static int azx_thaw_noirq(struct device *dev)
+{
+	struct pci_dev *pci = to_pci_dev(dev);
+
+	if (IS_SKL_PLUS(pci))
+		pci_set_power_state(pci, PCI_D0);
+
+	return 0;
+}
+#endif /* CONFIG_PM_SLEEP */
+
 #ifdef CONFIG_PM
 static void azx_power_notify(struct hda_bus *bus, bool power_up);
 #endif

@@ -3139,6 +3169,10 @@ static int azx_runtime_idle(struct device *dev)
 #ifdef CONFIG_PM
 static const struct dev_pm_ops azx_pm = {
 	SET_SYSTEM_SLEEP_PM_OPS(azx_suspend, azx_resume)
+#ifdef CONFIG_PM_SLEEP
+	.freeze_noirq = azx_freeze_noirq,
+	.thaw_noirq = azx_thaw_noirq,
+#endif
 	SET_RUNTIME_PM_OPS(azx_runtime_suspend, azx_runtime_resume, azx_runtime_idle)
 };
 

2.7.1

From: Mario Kleiner <[email protected]>

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

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

commit 9f660a1c43890c2cdd1f423fd73654e7ca08fe56 upstream.

Without this patch, internal speaker and line-out work, but front headphone output jack stays silent on the Mac Pro 4,1.

This code path also gets executed on the MacPro 5,1 due to identical codec SSID, but i don't know if it has any positive or adverse effects there or not.

(v2) Implement feedback from Takashi Iwai: Reuse
alc889_fixup_mbp_vref and just add a new nid 0x19 for the MacPro 4,1.

---
 sound/pci/hda/patch_realtek.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

index 6d9a42a16a16..1dc0702ff818 100644
@@ -1775,6 +1775,7 @@ enum {
 	ALC889_FIXUP_MBA11_VREF,
 	ALC889_FIXUP_MBA21_VREF,
 	ALC889_FIXUP_MP11_VREF,
+	ALC889_FIXUP_MP41_VREF,
 	ALC882_FIXUP_INV_DMIC,
 	ALC882_FIXUP_NO_PRIMARY_HP,
 	ALC887_FIXUP_ASUS_BASS,

@@ -1861,7 +1862,7 @@ static void alc889_fixup_mbp_vref(struct hda_codec *codec,
 				  const struct hda_fixup *fix, int action)
 {
 	struct alc_spec *spec = codec->spec;
-	static hda_nid_t nids[2] = { 0x14, 0x15 };
+	static hda_nid_t nids[3] = { 0x14, 0x15, 0x19 };
 	int i;
 
 	if (action != HDA_FIXUP_ACT_INIT)

@@ -2137,6 +2138,12 @@ static const struct hda_fixup alc882_fixups[] = {
 		.chained = true,
 		.chain_id = ALC885_FIXUP_MACPRO_GPIO,
 	},
+	[ALC889_FIXUP_MP41_VREF] = {
+		.type = HDA_FIXUP_FUNC,
+		.v.func = alc889_fixup_mbp_vref,
+		.chained = true,
+		.chain_id = ALC885_FIXUP_MACPRO_GPIO,
+	},
 	[ALC882_FIXUP_INV_DMIC] = {
 		.type = HDA_FIXUP_FUNC,
 		.v.func = alc_fixup_inv_dmic_0x12,

@@ -2209,7 +2216,7 @@ static const struct snd_pci_quirk alc882_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x106b, 0x3f00, "Macbook 5,1", ALC889_FIXUP_IMAC91_VREF),
 	SND_PCI_QUIRK(0x106b, 0x4000, "MacbookPro 5,1", ALC889_FIXUP_IMAC91_VREF),
 	SND_PCI_QUIRK(0x106b, 0x4100, "Macmini 3,1", ALC889_FIXUP_IMAC91_VREF),
-	SND_PCI_QUIRK(0x106b, 0x4200, "Mac Pro 5,1", ALC885_FIXUP_MACPRO_GPIO),
+	SND_PCI_QUIRK(0x106b, 0x4200, "Mac Pro 4,1/5,1", ALC889_FIXUP_MP41_VREF),
 	SND_PCI_QUIRK(0x106b, 0x4300, "iMac 9,1", ALC889_FIXUP_IMAC91_VREF),
 	SND_PCI_QUIRK(0x106b, 0x4600, "MacbookPro 5,2", ALC889_FIXUP_IMAC91_VREF),
 	SND_PCI_QUIRK(0x106b, 0x4900, "iMac 9,1 Aluminum", ALC889_FIXUP_IMAC91_VREF),

2.7.1

From: Takashi Iwai <[email protected]>

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

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

commit 030e2c78d3a91dd0d27fef37e91950dde333eba1 upstream.

snd_seq_ioctl_remove_events() calls snd_seq_fifo_clear() unconditionally even if there is no FIFO assigned, and this leads to an Oops due to NULL dereference. The fix is just to add a proper NULL check.

---
 sound/core/seq/seq_clientmgr.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index 4dc6bae80e15..ecfbf5f39d38 100644
@@ -1950,7 +1950,7 @@ static int snd_seq_ioctl_remove_events(struct snd_seq_client *client,
 		 * No restrictions so for a user client we can clear
 		 * the whole fifo
 		 */
-		if (client->type == USER_CLIENT)
+		if (client->type == USER_CLIENT && client->data.user.fifo)
 			snd_seq_fifo_clear(client->data.user.fifo);
 	}
 

2.7.1

From: Takashi Iwai <[email protected]>

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

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

commit 3567eb6af614dac436c4b16a8d426f9faed639b3 upstream.

ALSA sequencer code has an open race between the timer setup ioctl and the close of the client. This was triggered by syzkaller fuzzer, and a use-after-free was caught there as a result.

This patch papers over it by adding a proper queue->timer_mutex lock around the timer-related calls in the relevant code path.

---
 sound/core/seq/seq_queue.c | 2 ++
 1 file changed, 2 insertions(+)

index f9077361c119..4c9aa462de9b 100644
@@ -144,8 +144,10 @@ static struct snd_seq_queue *queue_new(int owner, int locked)
 static void queue_delete(struct snd_seq_queue *q)
 {
 	/* stop and release the timer */
+	mutex_lock(&q->timer_mutex);
 	snd_seq_timer_stop(q->timer);
 	snd_seq_timer_close(q);
+	mutex_unlock(&q->timer_mutex);
 	/* wait until access free */
 	snd_use_lock_sync(&q->use_lock);
 	/* release resources... */

2.7.1

From: Takashi Iwai <[email protected]>

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

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

commit b5a663aa426f4884c71cd8580adae73f33570f0d upstream.

A slave timer instance might be still accessible in a racy way while operating the master instance as it lacks of locking. Since the master operation is mostly protected with timer->lock, we should cope with it while changing the slave instance, too. Also, some linked lists (active_list and ack_list) of slave instances aren't unlinked immediately at stopping or closing, and this may lead to unexpected accesses.

This patch tries to address these issues. It adds spin lock of timer->lock (either from master or slave, which is equivalent) in a few places. For avoiding a deadlock, we ensure that the global slave_active_lock is always locked at first before each timer lock.

Also, ack and active_list of slave instances are properly unlinked at snd_timer_stop() and snd_timer_close().

Last but not least, remove the superfluous call of _snd_timer_stop() at removing slave links. This is a noop, and calling it may confuse readers wrt locking. Further cleanup will follow in a later patch.

Actually we've got reports of use-after-free by syzkaller fuzzer, and this hopefully fixes these issues.

---
 sound/core/timer.c | 18 ++++++++++++++----
 1 file changed, 14 insertions(+), 4 deletions(-)

index 6ddcf06f52f9..38a137d6b04f 100644
@@ -215,11 +215,13 @@ static void snd_timer_check_master(struct snd_timer_instance *master)
 		    slave->slave_id == master->slave_id) {
 			list_move_tail(&slave->open_list, &master->slave_list_head);
 			spin_lock_irq(&slave_active_lock);
+			spin_lock(&master->timer->lock);
 			slave->master = master;
 			slave->timer = master->timer;
 			if (slave->flags & SNDRV_TIMER_IFLG_RUNNING)
 				list_add_tail(&slave->active_list,
 					      &master->slave_active_head);
+			spin_unlock(&master->timer->lock);
 			spin_unlock_irq(&slave_active_lock);
 		}
 	}

@@ -345,15 +347,18 @@ int snd_timer_close(struct snd_timer_instance *timeri)
 		    timer->hw.close)
 			timer->hw.close(timer);
 		/* remove slave links */
+		spin_lock_irq(&slave_active_lock);
+		spin_lock(&timer->lock);
 		list_for_each_entry_safe(slave, tmp, &timeri->slave_list_head,
 					 open_list) {
-			spin_lock_irq(&slave_active_lock);
-			_snd_timer_stop(slave, 1, SNDRV_TIMER_EVENT_RESOLUTION);
 			list_move_tail(&slave->open_list, &snd_timer_slave_list);
 			slave->master = NULL;
 			slave->timer = NULL;
-			spin_unlock_irq(&slave_active_lock);
+			list_del_init(&slave->ack_list);
+			list_del_init(&slave->active_list);
 		}
+		spin_unlock(&timer->lock);
+		spin_unlock_irq(&slave_active_lock);
 		mutex_unlock(&register_mutex);
 	}
  out:

@@ -440,9 +445,12 @@ static int snd_timer_start_slave(struct snd_timer_instance *timeri)
 
 	spin_lock_irqsave(&slave_active_lock, flags);
 	timeri->flags |= SNDRV_TIMER_IFLG_RUNNING;
-	if (timeri->master)
+	if (timeri->master && timeri->timer) {
+		spin_lock(&timeri->timer->lock);
 		list_add_tail(&timeri->active_list,
 			      &timeri->master->slave_active_head);
+		spin_unlock(&timeri->timer->lock);
+	}
 	spin_unlock_irqrestore(&slave_active_lock, flags);
 	return 1; /* delayed start */
 }

@@ -488,6 +496,8 @@ static int _snd_timer_stop(struct snd_timer_instance * timeri,
 		if (!keep_flag) {
 			spin_lock_irqsave(&slave_active_lock, flags);
 			timeri->flags &= ~SNDRV_TIMER_IFLG_RUNNING;
+			list_del_init(&timeri->ack_list);
+			list_del_init(&timeri->active_list);
 			spin_unlock_irqrestore(&slave_active_lock, flags);
 		}
 		goto __end;

2.7.1

From: Takashi Iwai <[email protected]>

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

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

commit af368027a49a751d6ff4ee9e3f9961f35bb4fede upstream.

ALSA timer ioctls have an open race and this may lead to a use-after-free of timer instance object. A simplistic fix is to make each ioctl exclusive. We have already tread_sem for controlling the tread, and extend this as a global mutex to be applied to each ioctl.

The downside is, of course, the worse concurrency. But these ioctls aren't to be parallel accessible, in anyway, so it should be fine to serialize there.

---
 sound/core/timer.c | 32 +++++++++++++++++++-------------
 1 file changed, 19 insertions(+), 13 deletions(-)

index 38a137d6b04f..1d5461719e31 100644
@@ -73,7 +73,7 @@ struct snd_timer_user {
 	struct timespec tstamp;		/* trigger tstamp */
 	wait_queue_head_t qchange_sleep;
 	struct fasync_struct *fasync;
-	struct mutex tread_sem;
+	struct mutex ioctl_lock;
 };
 
 /* list of timers */

@@ -1266,7 +1266,7 @@ static int snd_timer_user_open(struct inode *inode, struct file *file)
 		return -ENOMEM;
 	spin_lock_init(&tu->qlock);
 	init_waitqueue_head(&tu->qchange_sleep);
-	mutex_init(&tu->tread_sem);
+	mutex_init(&tu->ioctl_lock);
 	tu->ticks = 1;
 	tu->queue_size = 128;
 	tu->queue = kmalloc(tu->queue_size * sizeof(struct snd_timer_read),

@@ -1286,8 +1286,10 @@ static int snd_timer_user_release(struct inode *inode, struct file *file)
 	if (file->private_data) {
 		tu = file->private_data;
 		file->private_data = NULL;
+		mutex_lock(&tu->ioctl_lock);
 		if (tu->timeri)
 			snd_timer_close(tu->timeri);
+		mutex_unlock(&tu->ioctl_lock);
 		kfree(tu->queue);
 		kfree(tu->tqueue);
 		kfree(tu);

@@ -1525,7 +1527,6 @@ static int snd_timer_user_tselect(struct file *file,
 	int err = 0;
 
 	tu = file->private_data;
-	mutex_lock(&tu->tread_sem);
 	if (tu->timeri) {
 		snd_timer_close(tu->timeri);
 		tu->timeri = NULL;

@@ -1569,7 +1570,6 @@ static int snd_timer_user_tselect(struct file *file,
 	}
 
       __err:
-      	mutex_unlock(&tu->tread_sem);
 	return err;
 }
 

@@ -1782,7 +1782,7 @@ enum {
 	SNDRV_TIMER_IOCTL_PAUSE_OLD = _IO('T', 0x23),
 };
 
-static long snd_timer_user_ioctl(struct file *file, unsigned int cmd,
+static long __snd_timer_user_ioctl(struct file *file, unsigned int cmd,
 				 unsigned long arg)
 {
 	struct snd_timer_user *tu;

@@ -1799,17 +1799,11 @@ static long snd_timer_user_ioctl(struct file *file, unsigned int cmd,
 	{
 		int xarg;
 
-		mutex_lock(&tu->tread_sem);
-		if (tu->timeri)	{	/* too late */
-			mutex_unlock(&tu->tread_sem);
+		if (tu->timeri)	/* too late */
 			return -EBUSY;
-		}
-		if (get_user(xarg, p)) {
-			mutex_unlock(&tu->tread_sem);
+		if (get_user(xarg, p))
 			return -EFAULT;
-		}
 		tu->tread = xarg ? 1 : 0;
-		mutex_unlock(&tu->tread_sem);
 		return 0;
 	}
 	case SNDRV_TIMER_IOCTL_GINFO:

@@ -1842,6 +1836,18 @@ static long snd_timer_user_ioctl(struct file *file, unsigned int cmd,
 	return -ENOTTY;
 }
 
+static long snd_timer_user_ioctl(struct file *file, unsigned int cmd,
+				 unsigned long arg)
+{
+	struct snd_timer_user *tu = file->private_data;
+	long ret;
+
+	mutex_lock(&tu->ioctl_lock);
+	ret = __snd_timer_user_ioctl(file, cmd, arg);
+	mutex_unlock(&tu->ioctl_lock);
+	return ret;
+}
+
 static int snd_timer_user_fasync(int fd, struct file * file, int on)
 {
 	struct snd_timer_user *tu;

2.7.1

From: Takashi Iwai <[email protected]>

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

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

commit ee8413b01045c74340aa13ad5bdf905de32be736 upstream.

ALSA timer instance object has a couple of linked lists and they are unlinked unconditionally at snd_timer_stop(). Meanwhile snd_timer_interrupt() unlinks it, but it calls list_del() which leaves the element list itself unchanged. This ends up with unlinking twice, and it was caught by syzkaller fuzzer.

The fix is to use list_del_init() variant properly there, too.

---
 sound/core/timer.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index 1d5461719e31..4e436fe53afa 100644
@@ -703,7 +703,7 @@ void snd_timer_interrupt(struct snd_timer * timer, unsigned long ticks_left)
 		} else {
 			ti->flags &= ~SNDRV_TIMER_IFLG_RUNNING;
 			if (--timer->running)
-				list_del(&ti->active_list);
+				list_del_init(&ti->active_list);
 		}
 		if ((timer->hw.flags & SNDRV_TIMER_HW_TASKLET) ||
 		    (ti->flags & SNDRV_TIMER_IFLG_FAST))

2.7.1

From: Nicolas Boichat <[email protected]>

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

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

commit 9586495dc3011a80602329094e746dbce16cb1f1 upstream.

This reverts one hunk of
commit ef44a1ec6eee ("ALSA: sound/core: use memdup_user()"), which replaced a number of kmalloc followed by memcpy with memdup calls.

In this case, we are copying from a struct snd_seq_port_info32 to a struct snd_seq_port_info, but the latter is 4 bytes longer than the 32-bit version, so we need to separate kmalloc and copy calls.

---
 sound/core/seq/seq_compat.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

index 81f7c109dc46..65175902a68a 100644
@@ -49,11 +49,12 @@ static int snd_seq_call_port_info_ioctl(struct snd_seq_client *client, unsigned
 	struct snd_seq_port_info *data;
 	mm_segment_t fs;
 
-	data = memdup_user(data32, sizeof(*data32));
-	if (IS_ERR(data))
-		return PTR_ERR(data);
+	data = kmalloc(sizeof(*data), GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
 
-	if (get_user(data->flags, &data32->flags) ||
+	if (copy_from_user(data, data32, sizeof(*data32)) ||
+	    get_user(data->flags, &data32->flags) ||
 	    get_user(data->time_queue, &data32->time_queue))
 		goto error;
 	data->kernel = NULL;

2.7.1

From: Nicolas Boichat <[email protected]>

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

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

commit 43c54b8c7cfe22f868a751ba8a59abf1724160b1 upstream.

This reverts one hunk of
commit ef44a1ec6eee ("ALSA: sound/core: use memdup_user()"), which replaced a number of kmalloc followed by memcpy with memdup calls.

In this case, we are copying from a struct snd_pcm_hw_params32 to a struct snd_pcm_hw_params, but the latter is 4 bytes longer than the 32-bit version, so we need to separate kmalloc and copy calls.

This actually leads to an out-of-bounds memory access later on in sound/soc/soc-pcm.c:soc_pcm_hw_params() (detected using KASan).

---
 sound/core/pcm_compat.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

index c4ac3c1e19af..1bb1a43c7d03 100644
@@ -236,10 +236,15 @@ static int snd_pcm_ioctl_hw_params_compat(struct snd_pcm_substream *substream,
 	if (! (runtime = substream->runtime))
 		return -ENOTTY;
 
-	/* only fifo_size is different, so just copy all */
-	data = memdup_user(data32, sizeof(*data32));
-	if (IS_ERR(data))
-		return PTR_ERR(data);
+	data = kmalloc(sizeof(*data), GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
+	/* only fifo_size (RO from userspace) is different, so just copy all */
+	if (copy_from_user(data, data32, sizeof(*data32))) {
+		err = -EFAULT;
+		goto error;
+	}
 
 	if (refine)
 		err = snd_pcm_hw_refine(substream, data);

2.7.1

From: Takashi Iwai <[email protected]>

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

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

commit 2ba1fe7a06d3624f9a7586d672b55f08f7c670f3 upstream.

hrtimer_cancel() waits for the completion from the callback, thus it must not be called inside the callback itself. This was already a problem in the past with ALSA hrtimer driver, and the early commit [fcfdebe70759: ALSA: hrtimer - Fix lock-up] tried to address it.

However, the previous fix is still insufficient: it may still cause a lockup when the ALSA timer instance reprograms itself in its callback. Then it invokes the start function even in snd_timer_interrupt() that is called in hrtimer callback itself, results in a CPU stall. This is no hypothetical problem but actually triggered by syzkaller fuzzer.

This patch tries to fix the issue again. Now we call hrtimer_try_to_cancel() at both start and stop functions so that it won't fall into a deadlock, yet giving some chance to cancel the queue if the functions have been called outside the callback. The proper hrtimer_cancel() is called in anyway at closing, so this should be enough.

---
 sound/core/hrtimer.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

index b8b31c433d64..14d483d6b3b0 100644
@@ -90,7 +90,7 @@ static int snd_hrtimer_start(struct snd_timer *t)
 	struct snd_hrtimer *stime = t->private_data;
 
 	atomic_set(&stime->running, 0);
-	hrtimer_cancel(&stime->hrt);
+	hrtimer_try_to_cancel(&stime->hrt);
 	hrtimer_start(&stime->hrt, ns_to_ktime(t->sticks * resolution),
 		      HRTIMER_MODE_REL);
 	atomic_set(&stime->running, 1);

@@ -101,6 +101,7 @@ static int snd_hrtimer_stop(struct snd_timer *t)
 {
 	struct snd_hrtimer *stime = t->private_data;
 	atomic_set(&stime->running, 0);
+	hrtimer_try_to_cancel(&stime->hrt);
 	return 0;
 }
 

2.7.1

From: Takashi Iwai <[email protected]>

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

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

commit c0bcdbdff3ff73a54161fca3cb8b6cdbd0bb8762 upstream.

When a TLV ioctl with numid zero is handled, the driver may spew a kernel warning with a stack trace at each call. The check was intended obviously only for a kernel driver, but not for a user interaction. Let's fix it.

This was spotted by syzkaller fuzzer.

---
 sound/core/control.c | 2 ++
 1 file changed, 2 insertions(+)

index f2082a35b890..3fcead61f0ef 100644
@@ -1325,6 +1325,8 @@ static int snd_ctl_tlv_ioctl(struct snd_ctl_file *file,
 		return -EFAULT;
 	if (tlv.length < sizeof(unsigned int) * 2)
 		return -EINVAL;
+	if (!tlv.numid)
+		return -EINVAL;
 	down_read(&card->controls_rwsem);
 	kctl = snd_ctl_find_numid(card, tlv.numid);
 	if (kctl == NULL) {

2.7.1

From: Sachin Pandhare <[email protected]>

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

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

commit e9f96bc53c1b959859599cb30ce6fd4fbb4448c2 upstream.

R17408 (4400h) HPF_C_1
R17409 (4401h) HPF_C_0
17048 -> 17408 (0x4400)
17049 -> 17409 (0x4401)

---
 sound/soc/codecs/wm8962.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

index ea16dc456352..c2cd83d8ed97 100644
@@ -364,8 +364,8 @@ static struct reg_default wm8962_reg[] = {
 	{ 16924, 0x0059 },   /* R16924 - HDBASS_PG_1 */
 	{ 16925, 0x999A },   /* R16925 - HDBASS_PG_0 */
 
-	{ 17048, 0x0083 },   /* R17408 - HPF_C_1 */
-	{ 17049, 0x98AD },   /* R17409 - HPF_C_0 */
+	{ 17408, 0x0083 },   /* R17408 - HPF_C_1 */
+	{ 17409, 0x98AD },   /* R17409 - HPF_C_0 */
 
 	{ 17920, 0x007F },   /* R17920 - ADCL_RETUNE_C1_1 */
 	{ 17921, 0xFFFF },   /* R17921 - ADCL_RETUNE_C1_0 */

2.7.1

From: Nikesh Oswal <[email protected]>

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

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

commit e73694d871867cae8471d2350ce89acb38bc2b63 upstream.

For a sample rate of 12kHz the bclk was taken from the 44.1kHz table as we test for a multiple of 8kHz. This patch fixes this issue by testing for multiples of 4kHz instead.

---
 sound/soc/codecs/arizona.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index f38ed4d225ca..26a5925b5c51 100644
@@ -1120,7 +1120,7 @@ static int arizona_hw_params(struct snd_pcm_substream *substream,
 	int chan_limit = arizona->pdata.max_channels_clocked[dai->id - 1];
 	int bclk, lrclk, wl, frame, bclk_target;
 
-	if (params_rate(params) % 8000)
+	if (params_rate(params) % 4000)
 		rates = &arizona_44k1_bclk_rates[0];
 	else
 		rates = &arizona_48k_bclk_rates[0];

2.7.1

From: Vinod Koul <[email protected]>

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

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

commit a1068045883ed4a18363a4ebad0c3d55e473b716 upstream.

The detection of direction for compress was only taking into account codec capabilities and not CPU ones. Fix this by checking the CPU side capabilities as well

---
 sound/soc/soc-compress.c | 23 ++++++++++++++++++++---
 1 file changed, 20 insertions(+), 3 deletions(-)

index 53c9ecdd119f..2868a17ff9a8 100644
@@ -385,17 +385,34 @@ int soc_new_compress(struct snd_soc_pcm_runtime *rtd, int num)
 	struct snd_compr *compr;
 	char new_name[64];
 	int ret = 0, direction = 0;
+	int playback = 0, capture = 0;
 
 	/* check client and interface hw capabilities */
 	snprintf(new_name, sizeof(new_name), "%s %s-%d",
 			rtd->dai_link->stream_name, codec_dai->name, num);
 
 	if (codec_dai->driver->playback.channels_min)
+		playback = 1;
+	if (codec_dai->driver->capture.channels_min)
+		capture = 1;
+
+	capture = capture && cpu_dai->driver->capture.channels_min;
+	playback = playback && cpu_dai->driver->playback.channels_min;
+
+	/*
+	 * Compress devices are unidirectional so only one of the directions
+	 * should be set, check for that (xor)
+	 */
+	if (playback + capture != 1) {
+		dev_err(rtd->card->dev, "Invalid direction for compress P %d, C %d\n",
+				playback, capture);
+		return -EINVAL;
+	}
+
+	if(playback)
 		direction = SND_COMPRESS_PLAYBACK;
-	else if (codec_dai->driver->capture.channels_min)
-		direction = SND_COMPRESS_CAPTURE;
 	else
-		return -EINVAL;
+		direction = SND_COMPRESS_CAPTURE;
 
 	compr = kzalloc(sizeof(*compr), GFP_KERNEL);
 	if (compr == NULL) {

2.7.1

From: Chunfeng Yun <[email protected]>

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

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

commit 096b110a3dd3c868e4610937c80d2e3f3357c1a9 upstream.

if a full speed hub connects to a high speed hub which supports MTT, the MTT field of its slot context will be set to 1 when xHCI driver setups an xHCI virtual device in xhci_setup_addressable_virt_dev(); once usb core fetch its hub descriptor, and need to update the xHC's internal data structures for the device, the HUB field of its slot context will be set to 1 too, meanwhile MTT is also set before, this will cause configure endpoint command fail, so in the case, we should clear MTT to 0 for full speed hub according to section 6.2.2

---
 drivers/usb/host/xhci.c | 8 ++++++++
 1 file changed, 8 insertions(+)

index 3d98a3a82c79..47b25542a99c 100644
@@ -4696,8 +4696,16 @@ int xhci_update_hub_device(struct usb_hcd *hcd, struct usb_device *hdev,
 	ctrl_ctx->add_flags |= cpu_to_le32(SLOT_FLAG);
 	slot_ctx = xhci_get_slot_ctx(xhci, config_cmd->in_ctx);
 	slot_ctx->dev_info |= cpu_to_le32(DEV_HUB);
+	/*
+	 * refer to section 6.2.2: MTT should be 0 for full speed hub,
+	 * but it may be already set to 1 when setup an xHCI virtual
+	 * device, so clear it anyway.
+	 */
 	if (tt->multi)
 		slot_ctx->dev_info |= cpu_to_le32(DEV_MTT);
+	else if (hdev->speed == USB_SPEED_FULL)
+		slot_ctx->dev_info &= cpu_to_le32(~DEV_MTT);
+
 	if (xhci->hci_version > 0x95) {
 		xhci_dbg(xhci, "xHCI version %x needs hub "
 				"TT think time and number of ports\n",

2.7.1

From: Dan Carpenter <[email protected]>

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

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

commit abdc9a3b4bac97add99e1d77dc6d28623afe682b upstream.

The code expects the loop to end with "retries" set to zero but, because it is a post-op, it will end set to -1. I have fixed this by moving the decrement inside the loop.

---
 drivers/usb/serial/ipaq.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

index 76c9a847da5d..e03900e8c667 100644
@@ -532,7 +532,8 @@ static int ipaq_open(struct tty_struct *tty,
 	 * through. Since this has a reasonably high failure rate, we retry
 	 * several times.
 	 */
-	while (retries--) {
+	while (retries) {
+		retries--;
 		result = usb_control_msg(serial->dev,
 				usb_sndctrlpipe(serial->dev, 0), 0x22, 0x21,
 				0x1, 0, NULL, 0, 100);

2.7.1

From: Oliver Freyermuth <[email protected]>

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

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

commit f7d7f59ab124748156ea551edf789994f05da342 upstream.

Add the USB device ID for ELV Marble Sound Board 1.

---
 drivers/usb/serial/cp210x.c | 1 +
 1 file changed, 1 insertion(+)

index 3597be0a5ae4..9a3c0f76db8c 100644
@@ -160,6 +160,7 @@ static const struct usb_device_id id_table[] = {
 	{ USB_DEVICE(0x17F4, 0xAAAA) }, /* Wavesense Jazz blood glucose meter */
 	{ USB_DEVICE(0x1843, 0x0200) }, /* Vaisala USB Instrument Cable */
 	{ USB_DEVICE(0x18EF, 0xE00F) }, /* ELV USB-I2C-Interface */
+	{ USB_DEVICE(0x18EF, 0xE025) }, /* ELV Marble Sound Board 1 */
 	{ USB_DEVICE(0x1ADB, 0x0001) }, /* Schweitzer Engineering C662 Cable */
 	{ USB_DEVICE(0x1B1C, 0x1C00) }, /* Corsair USB Dongle */
 	{ USB_DEVICE(0x1BA4, 0x0002) },	/* Silicon Labs 358x factory default */

2.7.1

From: Oliver Neukum <[email protected]>

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

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

commit 1eaf35e4dd592c59041bc1ed3248c46326da1f5f upstream.

The module should fail to load.

---
 drivers/usb/host/xhci.c | 4 ++++
 1 file changed, 4 insertions(+)

index 47b25542a99c..6888e9a956d4 100644
@@ -4891,6 +4891,10 @@ static int __init xhci_hcd_init(void)
 	BUILD_BUG_ON(sizeof(struct xhci_intr_reg) != 8*32/8);
 	/* xhci_run_regs has eight fields and embeds 128 xhci_intr_regs */
 	BUILD_BUG_ON(sizeof(struct xhci_run_regs) != (8+8*128)*32/8);
+
+	if (usb_disabled())
+		return -ENODEV;
+
 	return 0;
 unreg_pci:
 	xhci_unregister_pci();

2.7.1

This check needs to go at the top of the function, before the call to xhci_register_pci().

Ben.

Ben Hutchings
Life is what happens to you while you're busy making other plans.
                                                               - John Lennon

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

Oh yes, this is done later in 3.12.54 in:

commit b48d054283a2c2d60a100bf7772bf308042c57ac

Author: Greg Kroah-Hartman <[email protected]>
Date:   Sun Jan 31 11:11:58 2016 -0800

xhci: fix placement of call to usb_disabled()

thanks,
- --
js
suse labs
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQIcBAEBCAAGBQJWwuKNAAoJEL0lsQQGtHBJijIP/iZ6yp7A99iQi2ZqPvsHe+xC Cqip1AXpPaA727acGmlhC+B55jFzWQyEGFYOq7/Q3pZ8uxp9iVDfa6RsyfPqCbrF ItadjVKhjTc4RW005sjqUJCh/7fg776tr7mcf+2vgrF/TexSi/OrOOOcgsBVp3vj iZFnr2mKPC4nLhDx6iOIqYCSEGUASgdwQW3NuhnfOaVRM7lRBLl6Isbg24CN+Kra V9/x1xxNl00eBpMEU6G9bav847KWdOLik6qW0emuy3Bjzc34bVRH74m20ZVcJSEI 5bG9BFGNbikQdasRoWr/H8OSnh1YwZ38daazPsIdDqyut4VKkQ5wfvxP+/c/KUsG rYvfVTVaJWXb9gdODDleVIWRaZ1Va/0q5uzDrQi4KGV62dw9Pb0JJnDcuvWKmtx/ N0Fjjl7L1RoqNGklGPtqrYA2B8h/dqPlU0nOxwMBaafjeds+KK1k5ZWLlJUH3K4a CAFeabEjVCaysrHZ768OXGAH/BGreAUTsYL4l5VapQZwy/EggV5oeFlJjkIRZDJG kGQpGe63Bl/8+fFtx8UtW4d0LbtvgfSCIOZBCmyd/HIODODv53Exehm7vU409Dyd 9XsVJUhO/ANCHu/NSc+c9bL4f3u5Exd2uLoq2jeMricOZzat3ysVHFrTXqB1tO0f D14IaCz004DOckjJzJQa
=Wtvm
-----END PGP SIGNATURE-----

From: Behan Webster <[email protected]>

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

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

commit aeea3592a13bf12861943e44fc48f1f270941f8d upstream.

With compilers which follow the C99 standard (like modern versions of gcc and clang), "extern inline" does the wrong thing (emits code for an externally linkable version of the inline function). In this case using static inline and removing the NULL version of return_address in return_address.c does the right thing.

---
 arch/arm/include/asm/ftrace.h    | 2 +-
 arch/arm/kernel/return_address.c | 5 -----
 2 files changed, 1 insertion(+), 6 deletions(-)

index f89515adac60..2bb8cac28b9e 100644
@@ -45,7 +45,7 @@ void *return_address(unsigned int);
 
 #else
 
-extern inline void *return_address(unsigned int level)
+static inline void *return_address(unsigned int level)
 {
 	return NULL;
 }

index fafedd86885d..f6aa84d5b93c 100644
@@ -63,11 +63,6 @@ void *return_address(unsigned int level)
 #warning "TODO: return_address should use unwind tables"
 #endif
 
-void *return_address(unsigned int level)
-{
-	return NULL;
-}
-
 #endif /* if defined(CONFIG_FRAME_POINTER) && !defined(CONFIG_ARM_UNWIND) / else */
 
 EXPORT_SYMBOL_GPL(return_address);

2.7.1

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

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

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

commit e16343c47e4276f5ebc77ca16feb5e50ca1918f9 upstream.

The warning was introduced in 2009 (commit 4bf1fa5a34aa ([ARM] 5613/1:
implement CALLER_ADDRESSx)). The only "problem" here is that CALLER_ADDRESSx for x > 1 returns NULL which doesn't do much harm.

The drawback of implementing a fix (i.e. use unwind tables to implement CALLER_ADDRESSx) is that much of the unwinder code would need to be marked as not traceable.

---
 arch/arm/kernel/return_address.c | 4 ----
 1 file changed, 4 deletions(-)

index f6aa84d5b93c..98ea4b7eb406 100644
@@ -59,10 +59,6 @@ void *return_address(unsigned int level)
 
 #else /* if defined(CONFIG_FRAME_POINTER) && !defined(CONFIG_ARM_UNWIND) */
 
-#if defined(CONFIG_ARM_UNWIND)
-#warning "TODO: return_address should use unwind tables"
-#endif
-
 #endif /* if defined(CONFIG_FRAME_POINTER) && !defined(CONFIG_ARM_UNWIND) / else */
 
 EXPORT_SYMBOL_GPL(return_address);

2.7.1

From: Alex Deucher <[email protected]>

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

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

commit 936b32643cb03dcb34ef5dab81970229b1cc2a33 upstream.

With CONFIG_ACPI=n the following build warning is seen:

drivers/gpu/drm/radeon/cypress_dpm.c:302:31: warning: unused variable 'eg_pi' [-Wunused-variable]

Protect eg_pi with CONFIG_ACPI.

Based on patch from: Fabio Estevam <[email protected]> but doesn't mix allocation and code.

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

index 7143783fb237..b5a00771f27a 100644
@@ -299,7 +299,9 @@ void cypress_program_response_times(struct radeon_device *rdev)
 static int cypress_pcie_performance_request(struct radeon_device *rdev,
 					    u8 perf_req, bool advertise)
 {
+#if defined(CONFIG_ACPI)
 	struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
+#endif
 	u32 tmp;
 
 	udelay(10);

2.7.1

From: Fabio Estevam <[email protected]>

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

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

commit 71473dbd66d4a61be4cd531ef96b4877f6ee3941 upstream.

With CONFIG_ACPI=n the following build warning is seen:

drivers/gpu/drm/radeon/ni_dpm.c:3448:31: warning: unused variable 'eg_pi' [-Wunused-variable]

Move the definition of eg_pi inside the CONFIG_ACPI 'if' block.

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

index 85f36e702595..89bdc12adebb 100644
@@ -3445,9 +3445,9 @@ static int ni_enable_smc_cac(struct radeon_device *rdev,
 static int ni_pcie_performance_request(struct radeon_device *rdev,
 				       u8 perf_req, bool advertise)
 {
+#if defined(CONFIG_ACPI)
 	struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
 
-#if defined(CONFIG_ACPI)
 	if ((perf_req == PCIE_PERF_REQ_PECI_GEN1) ||
             (perf_req == PCIE_PERF_REQ_PECI_GEN2)) {
 		if (eg_pi->pcie_performance_request_registered == false)

2.7.1

From: Kees Cook <[email protected]>

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

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

commit 7d196ac303652588c60350f0a581d71e2e7b1a50 upstream.

When CONFIG_FRAME_WARN is set low (e.g. some ARM builds), the hard-coded stack buffer size used for kernel stack over run testing triggers build warnings. Instead, avoid the warning by recalcuating the buffer size and recursion count needed to trigger the test. Also uses the recursion counter indirectly to avoid changing the parameter during the test.

---
 drivers/misc/lkdtm.c | 30 ++++++++++++++++++++----------
 1 file changed, 20 insertions(+), 10 deletions(-)

index 9cbd0370ca44..482344242f94 100644
@@ -49,8 +49,19 @@
 #include <linux/ide.h>
 #endif
 
+/*
+ * Make sure our attempts to over run the kernel stack doesn't trigger
+ * a compiler warning when CONFIG_FRAME_WARN is set. Then make sure we
+ * recurse past the end of THREAD_SIZE by default.
+ */
+#if defined(CONFIG_FRAME_WARN) && (CONFIG_FRAME_WARN > 0)
+#define REC_STACK_SIZE (CONFIG_FRAME_WARN / 2)
+#else
+#define REC_STACK_SIZE (THREAD_SIZE / 8)
+#endif
+#define REC_NUM_DEFAULT ((THREAD_SIZE / REC_STACK_SIZE) * 2)
+
 #define DEFAULT_COUNT 10
-#define REC_NUM_DEFAULT 10
 #define EXEC_SIZE 64
 
 enum cname {

@@ -140,8 +151,7 @@ static DEFINE_SPINLOCK(lock_me_up);
 static u8 data_area[EXEC_SIZE];
 
 module_param(recur_count, int, 0644);
-MODULE_PARM_DESC(recur_count, " Recursion level for the stack overflow test, "\
-				 "default is 10");
+MODULE_PARM_DESC(recur_count, " Recursion level for the stack overflow test");
 module_param(cpoint_name, charp, 0444);
 MODULE_PARM_DESC(cpoint_name, " Crash Point, where kernel is to be crashed");
 module_param(cpoint_type, charp, 0444);

@@ -280,16 +290,16 @@ static int lkdtm_parse_commandline(void)
 	return -EINVAL;
 }
 
-static int recursive_loop(int a)
+static int recursive_loop(int remaining)
 {
-	char buf[1024];
+	char buf[REC_STACK_SIZE];
 
-	memset(buf,0xFF,1024);
-	recur_count--;
-	if (!recur_count)
+	/* Make sure compiler does not optimize this away. */
+	memset(buf, (remaining & 0xff) | 0x1, REC_STACK_SIZE);
+	if (!remaining)
 		return 0;
 	else
-        	return recursive_loop(a);
+		return recursive_loop(remaining - 1);
 }
 
 static void do_nothing(void)

@@ -333,7 +343,7 @@ static void lkdtm_do_action(enum ctype which)
 			;
 		break;
 	case CT_OVERFLOW:
-		(void) recursive_loop(0);
+		(void) recursive_loop(recur_count);
 		break;
 	case CT_CORRUPT_STACK:
 		corrupt_stack();

2.7.1

From: Paul Bolle <[email protected]>

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

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

commit 298589b1cb626adf4beba6dd8e3cd4b64e8799be upstream.

Building mem.o for 32 bits x86 triggers a GCC warning:

drivers/infiniband/hw/cxgb4/mem.c: In function '_c4iw_write_mem_dma_aligned':
drivers/infiniband/hw/cxgb4/mem.c:79:25: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]

Silence that warning by casting "&wr_wait" to unsigned long before casting it to __be64. That's what _c4iw_write_mem_inline() already does.

---
 drivers/infiniband/hw/cxgb4/mem.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index 4cb8eb24497c..a80503b3795c 100644
@@ -76,7 +76,7 @@ static int _c4iw_write_mem_dma_aligned(struct c4iw_rdev *rdev, u32 addr,
 	INIT_ULPTX_WR(req, wr_len, 0, 0);
 	req->wr.wr_hi = cpu_to_be32(FW_WR_OP(FW_ULPTX_WR) |
 			(wait ? FW_WR_COMPL(1) : 0));
-	req->wr.wr_lo = wait ? (__force __be64)&wr_wait : 0;
+	req->wr.wr_lo = wait ? (__force __be64)(unsigned long) &wr_wait : 0L;
 	req->wr.wr_mid = cpu_to_be32(FW_WR_LEN16(DIV_ROUND_UP(wr_len, 16)));
 	req->cmd = cpu_to_be32(ULPTX_CMD(ULP_TX_MEM_WRITE));
 	req->cmd |= cpu_to_be32(V_T5_ULP_MEMIO_ORDER(1));

2.7.1

From: Arnd Bergmann <[email protected]>

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

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

commit 3e7a8716e20b759eec0ad88145255bb33174f0c8 upstream.

Not all architectures are able to call __builtin_return_address(). On ARM, the mISDN code produces this warning:

hardware/mISDN/w6692.c: In function 'w6692_dctrl':
hardware/mISDN/w6692.c:1181:75: warning: unsupported argument to '__builtin_return_address'
pr_debug("%s: %s dev(%d) open from %p\n", card->name, __func__,

                                                                           ^
hardware/mISDN/mISDNipac.c: In function 'open_dchannel':
hardware/mISDN/mISDNipac.c:759:75: warning: unsupported argument to '__builtin_return_address'
  pr_debug("%s: %s dev(%d) open from %p\n", isac->name, __func__,
                                                                           ^

In a lot of cases, this is relatively easy to work around by passing the value of __builtin_return_address(0) from the callers into the functions that want it. One exception is the indirect 'open' function call in struct isac_hw. While it would be possible to fix this as well, this patch only addresses the other callers properly and lets this one return the direct parent function, which should be good enough.

---
 drivers/isdn/hardware/mISDN/mISDNipac.c | 12 +++++++++---
 drivers/isdn/hardware/mISDN/w6692.c     |  6 +++---
 2 files changed, 12 insertions(+), 6 deletions(-)

index ccd7d851be26..a77eea594b69 100644
@@ -754,10 +754,10 @@ dbusy_timer_handler(struct isac_hw *isac)
 }
 
 static int
-open_dchannel(struct isac_hw *isac, struct channel_req *rq)
+open_dchannel_caller(struct isac_hw *isac, struct channel_req *rq, void *caller)
 {
 	pr_debug("%s: %s dev(%d) open from %p\n", isac->name, __func__,
-		 isac->dch.dev.id, __builtin_return_address(1));
+		 isac->dch.dev.id, caller);
 	if (rq->protocol != ISDN_P_TE_S0)
 		return -EINVAL;
 	if (rq->adr.channel == 1)

@@ -771,6 +771,12 @@ open_dchannel(struct isac_hw *isac, struct channel_req *rq)
 	return 0;
 }
 
+static int
+open_dchannel(struct isac_hw *isac, struct channel_req *rq)
+{
+	return open_dchannel_caller(isac, rq, __builtin_return_address(0));
+}
+
 static const char *ISACVer[] =
 {"2086/2186 V1.1", "2085 B1", "2085 B2",
  "2085 V2.3"};

@@ -1548,7 +1554,7 @@ ipac_dctrl(struct mISDNchannel *ch, u32 cmd, void *arg)
 	case OPEN_CHANNEL:
 		rq = arg;
 		if (rq->protocol == ISDN_P_TE_S0)
-			err = open_dchannel(isac, rq);
+			err = open_dchannel_caller(isac, rq, __builtin_return_address(0));
 		else
 			err = open_bchannel(ipac, rq);
 		if (err)

index de69f6828c76..741675525b53 100644
@@ -1176,10 +1176,10 @@ w6692_l1callback(struct dchannel *dch, u32 cmd)
 }
 
 static int
-open_dchannel(struct w6692_hw *card, struct channel_req *rq)
+open_dchannel(struct w6692_hw *card, struct channel_req *rq, void *caller)
 {
 	pr_debug("%s: %s dev(%d) open from %p\n", card->name, __func__,
-		 card->dch.dev.id, __builtin_return_address(1));
+		 card->dch.dev.id, caller);
 	if (rq->protocol != ISDN_P_TE_S0)
 		return -EINVAL;
 	if (rq->adr.channel == 1)

@@ -1207,7 +1207,7 @@ w6692_dctrl(struct mISDNchannel *ch, u32 cmd, void *arg)
 	case OPEN_CHANNEL:
 		rq = arg;
 		if (rq->protocol == ISDN_P_TE_S0)
-			err = open_dchannel(card, rq);
+			err = open_dchannel(card, rq, __builtin_return_address(0));
 		else
 			err = open_bchannel(card, rq);
 		if (err)

2.7.1

From: Vijay Pandurangan <[email protected]>

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

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

[ Upstream commit ce8c839b74e3017996fad4e1b7ba2e2625ede82f ]

Packets that arrive from real hardware devices have ip_summed == CHECKSUM_UNNECESSARY if the hardware verified the checksums, or CHECKSUM_NONE if the packet is bad or it was unable to verify it. The current version of veth will replace CHECKSUM_NONE with CHECKSUM_UNNECESSARY, which causes corrupt packets routed from hardware to a veth device to be delivered to the application. This caused applications at Twitter to receive corrupt data when network hardware was corrupting packets.

We believe this was added as an optimization to skip computing and verifying checksums for communication between containers. However, locally generated packets have ip_summed == CHECKSUM_PARTIAL, so the code as written does nothing for them. As far as we can tell, after removing this code, these packets are transmitted from one stack to another unmodified (tcpdump shows invalid checksums on both sides, as expected), and they are delivered correctly to applications. We didn’t test every possible network configuration, but we tried a few common ones such as bridging containers, using NAT between the host and a container, and routing from hardware devices to containers. We have effectively deployed this in production at Twitter (by disabling RX checksum offloading on veth devices).

This code dates back to the first version of the driver, commit <e314dbdc1c0dc6a548ecf> ("[NET]: Virtual ethernet device driver"), so I suspect this bug occurred mostly because the driver API has evolved significantly since then. Commit <0b7967503dc97864f283a> ("net/veth: Fix packet checksumming") (in December 2010) fixed this for packets that get created locally and sent to hardware devices, by not changing CHECKSUM_PARTIAL. However, the same issue still occurs for packets coming in from hardware devices.

---
 drivers/net/veth.c | 6 ------
 1 file changed, 6 deletions(-)

index 61c4044f644e..917abeae77ad 100644
@@ -116,12 +116,6 @@ static netdev_tx_t veth_xmit(struct sk_buff *skb, struct net_device *dev)
 		kfree_skb(skb);
 		goto drop;
 	}
-	/* don't change ip_summed == CHECKSUM_PARTIAL, as that
-	 * will cause bad checksum on forwarded packets
-	 */
-	if (skb->ip_summed == CHECKSUM_NONE &&
-	    rcv->features & NETIF_F_RXCSUM)
-		skb->ip_summed = CHECKSUM_UNNECESSARY;
 
 	if (likely(dev_forward_skb(rcv, skb) == NET_RX_SUCCESS)) {
 		struct pcpu_vstats *stats = this_cpu_ptr(dev->vstats);

2.7.1

From: Andrey Ryabinin <[email protected]>

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

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

[ Upstream commit e459dfeeb64008b2d23bdf600f03b3605dbb8152 ]

ip6addrlbl_get() has never worked. If ip6addrlbl_hold() succeeded, ip6addrlbl_get() will exit with '-ESRCH'. If ip6addrlbl_hold() failed, ip6addrlbl_get() will use about to be free ip6addrlbl_entry pointer.

Fix this by inverting ip6addrlbl_hold() check.

---
 net/ipv6/addrlabel.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index b30ad3741b46..d5c918975c8c 100644
@@ -558,7 +558,7 @@ static int ip6addrlbl_get(struct sk_buff *in_skb, struct nlmsghdr* nlh)
 
 	rcu_read_lock();
 	p = __ipv6_addr_label(net, addr, ipv6_addr_type(addr), ifal->ifal_index);
-	if (p && ip6addrlbl_hold(p))
+	if (p && !ip6addrlbl_hold(p))
 		p = NULL;
 	lseq = ip6addrlbl_table.seq;
 	rcu_read_unlock();

2.7.1

From: Xin Long <[email protected]>

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

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

[ Upstream commit 068d8bd338e855286aea54e70d1c101569284b21 ]

In sctp_close, sctp_make_abort_user may return NULL because of memory allocation failure. If this happens, it will bypass any state change and never free the assoc. The assoc has no chance to be freed and it will be kept in memory with the state it had even after the socket is closed by sctp_close().

So if sctp_make_abort_user fails to allocate memory, we should abort the asoc via sctp_primitive_ABORT as well. Just like the annotation in sctp_sf_cookie_wait_prm_abort and sctp_sf_do_9_1_prm_abort said, "Even if we can't send the ABORT due to low memory delete the TCB. This is a departure from our typical NOMEM handling".

But then the chunk is NULL (low memory) and the SCTP_CMD_REPLY cmd would dereference the chunk pointer, and system crash. So we should add SCTP_CMD_REPLY cmd only when the chunk is not NULL, just like other places where it adds SCTP_CMD_REPLY cmd.

---
 net/sctp/sm_statefuns.c | 6 ++++--
 net/sctp/socket.c       | 3 +--
 2 files changed, 5 insertions(+), 4 deletions(-)

index bf12098bbe1c..63a116c31a8b 100644
@@ -4835,7 +4835,8 @@ sctp_disposition_t sctp_sf_do_9_1_prm_abort(
 
 	retval = SCTP_DISPOSITION_CONSUME;
 
-	sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(abort));
+	if (abort)
+		sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(abort));
 
 	/* Even if we can't send the ABORT due to low memory delete the
 	 * TCB.  This is a departure from our typical NOMEM handling.

@@ -4972,7 +4973,8 @@ sctp_disposition_t sctp_sf_cookie_wait_prm_abort(
 			SCTP_TO(SCTP_EVENT_TIMEOUT_T1_INIT));
 	retval = SCTP_DISPOSITION_CONSUME;
 
-	sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(abort));
+	if (abort)
+		sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(abort));
 
 	sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE,
 			SCTP_STATE(SCTP_STATE_CLOSED));

index e2b1da09dc79..9c47fbc5de0c 100644
@@ -1518,8 +1518,7 @@ static void sctp_close(struct sock *sk, long timeout)
 			struct sctp_chunk *chunk;
 
 			chunk = sctp_make_abort_user(asoc, NULL, 0);
-			if (chunk)
-				sctp_primitive_ABORT(net, asoc, chunk);
+			sctp_primitive_ABORT(net, asoc, chunk);
 		} else
 			sctp_primitive_SHUTDOWN(net, asoc, NULL);
 	}

2.7.1

From: Florian Westphal <[email protected]>

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

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

[ Upstream commit 55285bf09427c5abf43ee1d54e892f352092b1f1 ]

Dmitry reports memleak with syskaller program.
Problem is that connector bumps skb usecount but might not invoke callback.

So move skb_get to where we invoke the callback.

---
 drivers/connector/connector.c | 11 +++--------
 1 file changed, 3 insertions(+), 8 deletions(-)

index fb1bad083aa9..12e737e0a5af 100644
@@ -160,26 +160,21 @@ static int cn_call_callback(struct sk_buff *skb)
  *
  * It checks skb, netlink header and msg sizes, and calls callback helper.
  */
-static void cn_rx_skb(struct sk_buff *__skb)
+static void cn_rx_skb(struct sk_buff *skb)
 {
 	struct nlmsghdr *nlh;
-	struct sk_buff *skb;
 	int len, err;
 
-	skb = skb_get(__skb);
-
 	if (skb->len >= NLMSG_HDRLEN) {
 		nlh = nlmsg_hdr(skb);
 		len = nlmsg_len(nlh);
 
 		if (len < (int)sizeof(struct cn_msg) ||
 		    skb->len < nlh->nlmsg_len ||
-		    len > CONNECTOR_MAX_MSG_SIZE) {
-			kfree_skb(skb);
+		    len > CONNECTOR_MAX_MSG_SIZE)
 			return;
-		}
 
-		err = cn_call_callback(skb);
+		err = cn_call_callback(skb_get(skb));
 		if (err < 0)
 			kfree_skb(skb);
 	}

2.7.1

From: willy tarreau <[email protected]>

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

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

[ Upstream commit 712f4aad406bb1ed67f3f98d04c044191f0ff593 ]

It is possible for a process to allocate and accumulate far more FDs than the process' limit by sending them over a unix socket then closing them to keep the process' fd count low.

This change addresses this problem by keeping track of the number of FDs in flight per user and preventing non-privileged processes from having more FDs in flight than their configured FD limit.

Mitigates: CVE-2013-4312 (Linux 2.0+)

---
 include/linux/sched.h |  1 +
 net/unix/af_unix.c    | 24 ++++++++++++++++++++----
 net/unix/garbage.c    | 16 ++++++++++++----
 3 files changed, 33 insertions(+), 8 deletions(-)

index a4d7d19fc338..3ecea51ea060 100644
@@ -664,6 +664,7 @@ struct user_struct {
 	unsigned long mq_bytes;	/* How many bytes can be allocated to mqueue? */
 #endif
 	unsigned long locked_shm; /* How many pages of mlocked shm ? */
+	unsigned long unix_inflight;	/* How many files in flight in unix sockets */
 
 #ifdef CONFIG_KEYS
 	struct key *uid_keyring;	/* UID specific keyring */

index 31b88dcb0f01..e6b021327c3a 100644
@@ -1484,6 +1484,21 @@ static void unix_destruct_scm(struct sk_buff *skb)
 	sock_wfree(skb);
 }
 
+/*
+ * The "user->unix_inflight" variable is protected by the garbage
+ * collection lock, and we just read it locklessly here. If you go
+ * over the limit, there might be a tiny race in actually noticing
+ * it across threads. Tough.
+ */
+static inline bool too_many_unix_fds(struct task_struct *p)
+{
+	struct user_struct *user = current_user();
+
+	if (unlikely(user->unix_inflight > task_rlimit(p, RLIMIT_NOFILE)))
+		return !capable(CAP_SYS_RESOURCE) && !capable(CAP_SYS_ADMIN);
+	return false;
+}
+
 #define MAX_RECURSION_LEVEL 4
 
 static int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb)

@@ -1492,6 +1507,9 @@ static int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb)
 	unsigned char max_level = 0;
 	int unix_sock_count = 0;
 
+	if (too_many_unix_fds(current))
+		return -ETOOMANYREFS;
+
 	for (i = scm->fp->count - 1; i >= 0; i--) {
 		struct sock *sk = unix_get_socket(scm->fp->fp[i]);
 

@@ -1513,10 +1531,8 @@ static int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb)
 	if (!UNIXCB(skb).fp)
 		return -ENOMEM;
 
-	if (unix_sock_count) {
-		for (i = scm->fp->count - 1; i >= 0; i--)
-			unix_inflight(scm->fp->fp[i]);
-	}
+	for (i = scm->fp->count - 1; i >= 0; i--)
+		unix_inflight(scm->fp->fp[i]);
 	return max_level;
 }
 

index 9bc73f87f64a..06730fe6ad9d 100644
@@ -125,9 +125,12 @@ struct sock *unix_get_socket(struct file *filp)
 void unix_inflight(struct file *fp)
 {
 	struct sock *s = unix_get_socket(fp);
+
+	spin_lock(&unix_gc_lock);
+
 	if (s) {
 		struct unix_sock *u = unix_sk(s);
-		spin_lock(&unix_gc_lock);
+
 		if (atomic_long_inc_return(&u->inflight) == 1) {
 			BUG_ON(!list_empty(&u->link));
 			list_add_tail(&u->link, &gc_inflight_list);

@@ -135,22 +138,27 @@ void unix_inflight(struct file *fp)
 			BUG_ON(list_empty(&u->link));
 		}
 		unix_tot_inflight++;
-		spin_unlock(&unix_gc_lock);
 	}
+	fp->f_cred->user->unix_inflight++;
+	spin_unlock(&unix_gc_lock);
 }
 
 void unix_notinflight(struct file *fp)
 {
 	struct sock *s = unix_get_socket(fp);
+
+	spin_lock(&unix_gc_lock);
+
 	if (s) {
 		struct unix_sock *u = unix_sk(s);
-		spin_lock(&unix_gc_lock);
+
 		BUG_ON(list_empty(&u->link));
 		if (atomic_long_dec_and_test(&u->inflight))
 			list_del_init(&u->link);
 		unix_tot_inflight--;
-		spin_unlock(&unix_gc_lock);
 	}
+	fp->f_cred->user->unix_inflight--;
+	spin_unlock(&unix_gc_lock);
 }
 
 static void scan_inflight(struct sock *x, void (*func)(struct unix_sock *),

2.7.1

Hi Jiri,

A possible issue was reported regarding this patch, and Hannes implemented a fix that's not yet in mainline. I guess it's preferable to postpone this patch for now.

Thanks,
Willy

Hi,

yes definitely. Thanks for noting.

For reference:
http://article.gmane.org/gmane.linux.kernel/2142236

thanks,

js
suse labs

Am 12.02.2016 um 08:57 schrieb Jiri Slaby:

Yes and no: the above mentioned patch looks innocent now after more bisecting, but there is <https://patchwork.ozlabs.org/patch/577653/> as a folow-up to the FD-accounting.

Better read the full thread:
<http://thread.gmane.org/gmane.linux.kernel/2142236>; the suspected bad patch is
unix: avoid use-after-free in ep_remove_wait_queue

Philipp

It was not the only issue reported, as I remember there is a possibility that for processes having to pass FDs from one socket to another (eg:
dbus), the wrong user could be credited. I don't remember the exact detail but since the fix is pending and the current patch fixes an issue which is as old as kernel 2.0 or so, there's no need to rush it into stable kernels.

Willy

From: Hannes Frederic Sowa <[email protected]>

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

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

[ Upstream commit ff62198553e43cdffa9d539f6165d3e83f8a42bc ]

[I stole this patch from Eric Biederman. He wrote:]

[Hannes: changed patch using netns_eq]

---
 net/bridge/br_stp_if.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

index 886f6d6dc48a..3995a66c3e4e 100644
@@ -128,7 +128,10 @@ static void br_stp_start(struct net_bridge *br)
 	char *argv[] = { BR_STP_PROG, br->dev->name, "start", NULL };
 	char *envp[] = { NULL };
 
-	r = call_usermodehelper(BR_STP_PROG, argv, envp, UMH_WAIT_PROC);
+	if (net_eq(dev_net(br->dev), &init_net))
+		r = call_usermodehelper(BR_STP_PROG, argv, envp, UMH_WAIT_PROC);
+	else
+		r = -ENOENT;
 
 	spin_lock_bh(&br->lock);
 

2.7.1

From: Sasha Levin <[email protected]>

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

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

[ Upstream commit 320f1a4a175e7cd5d3f006f92b4d4d3e2cbb7bb5 ]

proc_dostring() needs an initialized destination string, while the one provided in proc_sctp_do_hmac_alg() contains stack garbage.

Thus, writing to cookie_hmac_alg would strlen() that garbage and end up accessing invalid memory.

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

index 968355f0de60..596aa3c5321c 100644
@@ -306,7 +306,7 @@ static int proc_sctp_do_hmac_alg(struct ctl_table *ctl,
 	struct ctl_table tbl;
 	bool changed = false;
 	char *none = "none";
-	char tmp[8];
+	char tmp[8] = {0};
 	int ret;
 
 	memset(&tbl, 0, sizeof(struct ctl_table));

2.7.1

From: Eric Dumazet <[email protected]>

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

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

[ Upstream commit 3e4006f0b86a5ae5eb0e8215f9a9e1db24506977 ]

When first SYNACK is sent, we already hold rcu_read_lock(), but this is not true if a SYNACK is retransmitted, as a timer (soft) interrupt does not hold rcu_read_lock()

---
 net/ipv6/tcp_ipv6.c | 2 ++
 1 file changed, 2 insertions(+)

index 90004c6e3bff..7138ee87e07c 100644
@@ -484,8 +484,10 @@ static int tcp_v6_send_synack(struct sock *sk, struct dst_entry *dst,
 
 		fl6->daddr = treq->rmt_addr;
 		skb_set_queue_mapping(skb, queue_mapping);
+		rcu_read_lock();
 		err = ip6_xmit(sk, skb, fl6, rcu_dereference(np->opt),
 			       np->tclass);
+		rcu_read_unlock();
 		err = net_xmit_eval(err);
 	}
 

2.7.1

From: Neal Cardwell <[email protected]>

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

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

[ Upstream commit 83d15e70c4d8909d722c0d64747d8fb42e38a48f ]

For tcp_yeah, use an ssthresh floor of 2, the same floor used by Reno and CUBIC, per RFC 5681 (equation 4).

tcp_yeah_ssthresh() was sometimes returning a 0 or negative ssthresh value if the intended reduction is as big or bigger than the current cwnd. Congestion control modules should never return a zero or negative ssthresh. A zero ssthresh generally results in a zero cwnd, causing the connection to stall. A negative ssthresh value will be interpreted as a u32 and will set a target cwnd for PRR near 4 billion.

Oleksandr Natalenko reported that a system using tcp_yeah with ECN could see a warning about a prior_cwnd of 0 in tcp_cwnd_reduction(). Testing verified that this was due to tcp_yeah_ssthresh() misbehaving in this way.

---
 net/ipv4/tcp_yeah.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index 05c3b6f0e8e1..bf8321d6f2ef 100644
@@ -222,7 +222,7 @@ static u32 tcp_yeah_ssthresh(struct sock *sk) {
 	yeah->fast_count = 0;
 	yeah->reno_count = max(yeah->reno_count>>1, 2U);
 
-	return tp->snd_cwnd - reduction;
+	return max_t(int, tp->snd_cwnd - reduction, 2);
 }
 
 static struct tcp_congestion_ops tcp_yeah __read_mostly = {

2.7.1

From: Eric Dumazet <[email protected]>

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

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

[ Upstream commit 7aaed57c5c2890634cfadf725173c7c68ea4cb4f ]

Ivaylo Dimitrov reported a regression caused by commit 7866a621043f ("dev: add per net_device packet type chains").

skb->dev becomes NULL and we crash in __netif_receive_skb_core().

Before above commit, different kind of bugs or corruptions could happen without major crash.

But the root cause is that phonet_rcv() can queue skb without checking if skb is shared or not.

Many thanks to Ivaylo Dimitrov for his help, diagnosis and tests.

---
 net/phonet/af_phonet.c | 4 ++++
 1 file changed, 4 insertions(+)

index 5a940dbd74a3..f0229223bf91 100644
@@ -377,6 +377,10 @@ static int phonet_rcv(struct sk_buff *skb, struct net_device *dev,
 	struct sockaddr_pn sa;
 	u16 len;
 
+	skb = skb_share_check(skb, GFP_ATOMIC);
+	if (!skb)
+		return NET_RX_DROP;
+
 	/* check we have at least a full Phonet header */
 	if (!pskb_pull(skb, sizeof(struct phonethdr)))
 		goto out;

2.7.1

From: Eric Dumazet <[email protected]>

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

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

[ Upstream commit 34ae6a1aa0540f0f781dd265366036355fdc8930 ]

When a tunnel decapsulates the outer header, it has to comply with RFC 6080 and eventually propagate CE mark into inner header.

It turns out IP6_ECN_set_ce() does not correctly update skb->csum for CHECKSUM_COMPLETE packets, triggering infamous "hw csum failure" messages and stack traces.

---
 include/net/inet_ecn.h       | 19 ++++++++++++++++---
 net/ipv6/xfrm6_mode_tunnel.c |  2 +-
 2 files changed, 17 insertions(+), 4 deletions(-)

index 3bd22795c3e2..194723c2e7bb 100644
@@ -111,11 +111,24 @@ static inline void ipv4_copy_dscp(unsigned int dscp, struct iphdr *inner)
 
 struct ipv6hdr;
 
-static inline int IP6_ECN_set_ce(struct ipv6hdr *iph)
+/* Note:
+ * IP_ECN_set_ce() has to tweak IPV4 checksum when setting CE,
+ * meaning both changes have no effect on skb->csum if/when CHECKSUM_COMPLETE
+ * In IPv6 case, no checksum compensates the change in IPv6 header,
+ * so we have to update skb->csum.
+ */
+static inline int IP6_ECN_set_ce(struct sk_buff *skb, struct ipv6hdr *iph)
 {
+	__be32 from, to;
+
 	if (INET_ECN_is_not_ect(ipv6_get_dsfield(iph)))
 		return 0;
-	*(__be32*)iph |= htonl(INET_ECN_CE << 20);
+
+	from = *(__be32 *)iph;
+	to = from | htonl(INET_ECN_CE << 20);
+	*(__be32 *)iph = to;
+	if (skb->ip_summed == CHECKSUM_COMPLETE)
+		skb->csum = csum_add(csum_sub(skb->csum, from), to);
 	return 1;
 }
 

@@ -142,7 +155,7 @@ static inline int INET_ECN_set_ce(struct sk_buff *skb)
 	case cpu_to_be16(ETH_P_IPV6):
 		if (skb_network_header(skb) + sizeof(struct ipv6hdr) <=
 		    skb_tail_pointer(skb))
-			return IP6_ECN_set_ce(ipv6_hdr(skb));
+			return IP6_ECN_set_ce(skb, ipv6_hdr(skb));
 		break;
 	}
 

index 4770d515c2c8..d43c9babc2b0 100644
@@ -24,7 +24,7 @@ static inline void ipip6_ecn_decapsulate(struct sk_buff *skb)
 	struct ipv6hdr *inner_iph = ipipv6_hdr(skb);
 
 	if (INET_ECN_is_ce(ipv6_get_dsfield(outer_iph)))
-		IP6_ECN_set_ce(inner_iph);
+		IP6_ECN_set_ce(skb, inner_iph);
 }
 
 /* Add encapsulation header.

2.7.1

From: Ido Schimmel <[email protected]>

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

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

[ Upstream commit 60a6531bfe49555581ccd65f66a350cc5693fcde ]

We can't be within an RCU read-side critical section when deleting VLANs, as underlying drivers might sleep during the hardware operation. Therefore, replace the RCU critical section with a mutex. This is consistent with team_vlan_rx_add_vid.

---
 drivers/net/team/team.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

index 020581ddfdd3..3059b8c3825f 100644
@@ -1831,10 +1831,10 @@ static int team_vlan_rx_kill_vid(struct net_device *dev, __be16 proto, u16 vid)
 	struct team *team = netdev_priv(dev);
 	struct team_port *port;
 
-	rcu_read_lock();
-	list_for_each_entry_rcu(port, &team->port_list, list)
+	mutex_lock(&team->lock);
+	list_for_each_entry(port, &team->port_list, list)
 		vlan_vid_del(port->dev, proto, vid);
-	rcu_read_unlock();
+	mutex_unlock(&team->lock);
 
 	return 0;
 }

2.7.1

From: Dan Streetman <[email protected]>

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

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

[ Upstream commit a8a572a6b5f2a79280d6e302cb3c1cb1fbaeb3e8 ]

Remove the dst_entries_init/destroy calls for xfrm4 and xfrm6 dst_ops templates; their dst_entries counters will never be used. Move the xfrm dst_ops initialization from the common xfrm/xfrm_policy.c to xfrm4/xfrm4_policy.c and xfrm6/xfrm6_policy.c, and call dst_entries_init and dst_entries_destroy for each net namespace.

The ipv4 and ipv6 xfrms each create dst_ops template, and perform dst_entries_init on the templates. The template values are copied to each net namespace's xfrm.xfrm*_dst_ops. The problem there is the dst_ops pcpuc_entries field is a percpu counter and cannot be used correctly by simply copying it to another object.

The result of this is a very subtle bug; changes to the dst entries counter from one net namespace may sometimes get applied to a different net namespace dst entries counter. This is because of how the percpu counter works; it has a main count field as well as a pointer to the percpu variables. Each net namespace maintains its own main count variable, but all point to one set of percpu variables. When any net namespace happens to change one of the percpu variables to outside its small batch range, its count is moved to the net namespace's main count variable. So with multiple net namespaces operating concurrently, the dst_ops entries counter can stray from the actual value that it should be; if counts are consistently moved from one net namespace to another (which my testing showed is likely), then one net namespace winds up with a negative dst_ops count while another winds up with a continually increasing count, eventually reaching its gc_thresh limit, which causes all new traffic on the net namespace to fail with -ENOBUFS.

---
 net/ipv4/xfrm4_policy.c | 46 +++++++++++++++++++++++++++++++++---------
 net/ipv6/xfrm6_policy.c | 53 +++++++++++++++++++++++++++++++++++--------------
 net/xfrm/xfrm_policy.c  | 38 -----------------------------------
 3 files changed, 75 insertions(+), 62 deletions(-)

index adf998322bd2..bb201660bd8a 100644
@@ -230,7 +230,7 @@ static void xfrm4_dst_ifdown(struct dst_entry *dst, struct net_device *dev,
 	xfrm_dst_ifdown(dst, dev);
 }
 
-static struct dst_ops xfrm4_dst_ops = {
+static struct dst_ops xfrm4_dst_ops_template = {
 	.family =		AF_INET,
 	.protocol =		cpu_to_be16(ETH_P_IP),
 	.gc =			xfrm4_garbage_collect,

@@ -245,7 +245,7 @@ static struct dst_ops xfrm4_dst_ops = {
 
 static struct xfrm_policy_afinfo xfrm4_policy_afinfo = {
 	.family = 		AF_INET,
-	.dst_ops =		&xfrm4_dst_ops,
+	.dst_ops =		&xfrm4_dst_ops_template,
 	.dst_lookup =		xfrm4_dst_lookup,
 	.get_saddr =		xfrm4_get_saddr,
 	.decode_session =	_decode_session4,

@@ -267,7 +267,7 @@ static struct ctl_table xfrm4_policy_table[] = {
 	{ }
 };
 
-static int __net_init xfrm4_net_init(struct net *net)
+static int __net_init xfrm4_net_sysctl_init(struct net *net)
 {
 	struct ctl_table *table;
 	struct ctl_table_header *hdr;

@@ -295,7 +295,7 @@ err_alloc:
 	return -ENOMEM;
 }
 
-static void __net_exit xfrm4_net_exit(struct net *net)
+static void __net_exit xfrm4_net_sysctl_exit(struct net *net)
 {
 	struct ctl_table *table;
 

@@ -307,12 +307,44 @@ static void __net_exit xfrm4_net_exit(struct net *net)
 	if (!net_eq(net, &init_net))
 		kfree(table);
 }
+#else /* CONFIG_SYSCTL */
+static int inline xfrm4_net_sysctl_init(struct net *net)
+{
+	return 0;
+}
+
+static void inline xfrm4_net_sysctl_exit(struct net *net)
+{
+}
+#endif
+
+static int __net_init xfrm4_net_init(struct net *net)
+{
+	int ret;
+
+	memcpy(&net->xfrm.xfrm4_dst_ops, &xfrm4_dst_ops_template,
+	       sizeof(xfrm4_dst_ops_template));
+	ret = dst_entries_init(&net->xfrm.xfrm4_dst_ops);
+	if (ret)
+		return ret;
+
+	ret = xfrm4_net_sysctl_init(net);
+	if (ret)
+		dst_entries_destroy(&net->xfrm.xfrm4_dst_ops);
+
+	return ret;
+}
+
+static void __net_exit xfrm4_net_exit(struct net *net)
+{
+	xfrm4_net_sysctl_exit(net);
+	dst_entries_destroy(&net->xfrm.xfrm4_dst_ops);
+}
 
 static struct pernet_operations __net_initdata xfrm4_net_ops = {
 	.init	= xfrm4_net_init,
 	.exit	= xfrm4_net_exit,
 };
-#endif
 
 static void __init xfrm4_policy_init(void)
 {

@@ -321,12 +353,8 @@ static void __init xfrm4_policy_init(void)
 
 void __init xfrm4_init(void)
 {
-	dst_entries_init(&xfrm4_dst_ops);
-
 	xfrm4_state_init();
 	xfrm4_policy_init();
-#ifdef CONFIG_SYSCTL
 	register_pernet_subsys(&xfrm4_net_ops);
-#endif
 }
 

index 57674ddc683d..5606e994f56e 100644
@@ -2724,7 +2724,6 @@ static struct neighbour *xfrm_neigh_lookup(const struct dst_entry *dst,
 
 int xfrm_policy_register_afinfo(struct xfrm_policy_afinfo *afinfo)
 {
-	struct net *net;
 	int err = 0;
 	if (unlikely(afinfo == NULL))
 		return -EINVAL;

@@ -2755,26 +2754,6 @@ int xfrm_policy_register_afinfo(struct xfrm_policy_afinfo *afinfo)
 	}
 	spin_unlock(&xfrm_policy_afinfo_lock);
 
-	rtnl_lock();
-	for_each_net(net) {
-		struct dst_ops *xfrm_dst_ops;
-
-		switch (afinfo->family) {
-		case AF_INET:
-			xfrm_dst_ops = &net->xfrm.xfrm4_dst_ops;
-			break;
-#if IS_ENABLED(CONFIG_IPV6)
-		case AF_INET6:
-			xfrm_dst_ops = &net->xfrm.xfrm6_dst_ops;
-			break;
-#endif
-		default:
-			BUG();
-		}
-		*xfrm_dst_ops = *afinfo->dst_ops;
-	}
-	rtnl_unlock();
-
 	return err;
 }
 EXPORT_SYMBOL(xfrm_policy_register_afinfo);

@@ -2810,22 +2789,6 @@ int xfrm_policy_unregister_afinfo(struct xfrm_policy_afinfo *afinfo)
 }
 EXPORT_SYMBOL(xfrm_policy_unregister_afinfo);
 
-static void __net_init xfrm_dst_ops_init(struct net *net)
-{
-	struct xfrm_policy_afinfo *afinfo;
-
-	rcu_read_lock();
-	afinfo = rcu_dereference(xfrm_policy_afinfo[AF_INET]);
-	if (afinfo)
-		net->xfrm.xfrm4_dst_ops = *afinfo->dst_ops;
-#if IS_ENABLED(CONFIG_IPV6)
-	afinfo = rcu_dereference(xfrm_policy_afinfo[AF_INET6]);
-	if (afinfo)
-		net->xfrm.xfrm6_dst_ops = *afinfo->dst_ops;
-#endif
-	rcu_read_unlock();
-}
-
 static int xfrm_dev_event(struct notifier_block *this, unsigned long event, void *ptr)
 {
 	struct net_device *dev = netdev_notifier_info_to_dev(ptr);

@@ -2971,7 +2934,6 @@ static int __net_init xfrm_net_init(struct net *net)
 	rv = xfrm_policy_init(net);
 	if (rv < 0)
 		goto out_policy;
-	xfrm_dst_ops_init(net);
 	rv = xfrm_sysctl_init(net);
 	if (rv < 0)
 		goto out_sysctl;

2.7.1

From: Michael Neuling <[email protected]>

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

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

commit d2b9d2a5ad5ef04ff978c9923d19730cb05efd55 upstream.

Currently we allow both the MSR T and S bits to be set by userspace on a signal return. Unfortunately this is a reserved configuration and will cause a TM Bad Thing exception if attempted (via rfid).

This patch checks for this case in both the 32 and 64 bit signals code. If both T and S are set, we mark the context as invalid.

Found using a syscall fuzzer.

---
 arch/powerpc/include/asm/reg.h  |  1 +
 arch/powerpc/kernel/signal_32.c | 14 +++++++++-----
 arch/powerpc/kernel/signal_64.c |  4 ++++
 3 files changed, 14 insertions(+), 5 deletions(-)

index 390e09872b77..3ce6b7b5ca19 100644
@@ -108,6 +108,7 @@
 #define MSR_TS_T	__MASK(MSR_TS_T_LG)	/*  Transaction Transactional */
 #define MSR_TS_MASK	(MSR_TS_T | MSR_TS_S)   /* Transaction State bits */
 #define MSR_TM_ACTIVE(x) (((x) & MSR_TS_MASK) != 0) /* Transaction active? */
+#define MSR_TM_RESV(x) (((x) & MSR_TS_MASK) == MSR_TS_MASK) /* Reserved */
 #define MSR_TM_TRANSACTIONAL(x)	(((x) & MSR_TS_MASK) == MSR_TS_T)
 #define MSR_TM_SUSPENDED(x)	(((x) & MSR_TS_MASK) == MSR_TS_S)
 

index 7fce77b89f6d..3678e5097c59 100644
@@ -867,6 +867,15 @@ static long restore_tm_user_regs(struct pt_regs *regs,
 		return 1;
 #endif /* CONFIG_SPE */
 
+	/* Get the top half of the MSR from the user context */
+	if (__get_user(msr_hi, &tm_sr->mc_gregs[PT_MSR]))
+		return 1;
+	msr_hi <<= 32;
+	/* If TM bits are set to the reserved value, it's an invalid context */
+	if (MSR_TM_RESV(msr_hi))
+		return 1;
+	/* Pull in the MSR TM bits from the user context */
+	regs->msr = (regs->msr & ~MSR_TS_MASK) | (msr_hi & MSR_TS_MASK);
 	/* Now, recheckpoint.  This loads up all of the checkpointed (older)
 	 * registers, including FP and V[S]Rs.  After recheckpointing, the
 	 * transactional versions should be loaded.

@@ -876,11 +885,6 @@ static long restore_tm_user_regs(struct pt_regs *regs,
 	current->thread.tm_texasr |= TEXASR_FS;
 	/* This loads the checkpointed FP/VEC state, if used */
 	tm_recheckpoint(&current->thread, msr);
-	/* Get the top half of the MSR */
-	if (__get_user(msr_hi, &tm_sr->mc_gregs[PT_MSR]))
-		return 1;
-	/* Pull in MSR TM from user context */
-	regs->msr = (regs->msr & ~MSR_TS_MASK) | ((msr_hi<<32) & MSR_TS_MASK);
 
 	/* This loads the speculative FP/VEC state, if used */
 	if (msr & MSR_FP) {

index 4456779dba1c..1c43da49fb1c 100644
@@ -420,6 +420,10 @@ static long restore_tm_sigcontexts(struct pt_regs *regs,
 
 	/* get MSR separately, transfer the LE bit if doing signal return */
 	err |= __get_user(msr, &sc->gp_regs[PT_MSR]);
+	/* Don't allow reserved mode. */
+	if (MSR_TM_RESV(msr))
+		return -EINVAL;
+
 	/* pull in MSR TM from user context */
 	regs->msr = (regs->msr & ~MSR_TS_MASK) | (msr & MSR_TS_MASK);
 

2.7.1

From: Boqun Feng <[email protected]>

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

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

commit 49e9cf3f0c04bf76ffa59242254110309554861d upstream.

According to memory-barriers.txt:

Which mean these operations should be fully ordered. However on PPC, PPC_ATOMIC_ENTRY_BARRIER is the barrier before the actual operation, which is currently "lwsync" if SMP=y. The leading "lwsync" can not guarantee fully ordered atomics, according to Paul Mckenney:

https://lkml.org/lkml/2015/10/14/970

To fix this, we define PPC_ATOMIC_ENTRY_BARRIER as "sync" to guarantee the fully-ordered semantics.

This also makes futex atomics fully ordered, which can avoid possible memory ordering problems if userspace code relies on futex system call for fully ordered semantics.

---
 arch/powerpc/include/asm/synch.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index e682a7143edb..c50868681f9e 100644
@@ -44,7 +44,7 @@ static inline void isync(void)
 	MAKE_LWSYNC_SECTION_ENTRY(97, __lwsync_fixup);
 #define PPC_ACQUIRE_BARRIER	 "\n" stringify_in_c(__PPC_ACQUIRE_BARRIER)
 #define PPC_RELEASE_BARRIER	 stringify_in_c(LWSYNC) "\n"
-#define PPC_ATOMIC_ENTRY_BARRIER "\n" stringify_in_c(LWSYNC) "\n"
+#define PPC_ATOMIC_ENTRY_BARRIER "\n" stringify_in_c(sync) "\n"
 #define PPC_ATOMIC_EXIT_BARRIER	 "\n" stringify_in_c(sync) "\n"
 #else
 #define PPC_ACQUIRE_BARRIER

2.7.1

From: Boqun Feng <[email protected]>

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

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

commit 81d7a3294de7e9828310bbf986a67246b13fa01e upstream.

According to memory-barriers.txt, xchg*, cmpxchg* and their atomic_ versions all need to be fully ordered, however they are now just RELEASE+ACQUIRE, which are not fully ordered.

So also replace PPC_RELEASE_BARRIER and PPC_ACQUIRE_BARRIER with PPC_ATOMIC_ENTRY_BARRIER and PPC_ATOMIC_EXIT_BARRIER in __{cmp,}xchg_{u32,u64} respectively to guarantee fully ordered semantics of atomic{,64}_{cmp,}xchg() and {cmp,}xchg(), as a complement of commit b97021f85517 ("powerpc: Fix atomic_xxx_return barrier semantics")

This patch depends on patch "powerpc: Make value-returning atomics fully ordered" for PPC_ATOMIC_ENTRY_BARRIER definition.

---
 arch/powerpc/include/asm/cmpxchg.h | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

index e245aab7f191..95b515113186 100644
@@ -18,12 +18,12 @@ __xchg_u32(volatile void *p, unsigned long val)
 	unsigned long prev;
 
 	__asm__ __volatile__(
-	PPC_RELEASE_BARRIER
+	PPC_ATOMIC_ENTRY_BARRIER
 "1:	lwarx	%0,0,%2 \n"
 	PPC405_ERR77(0,%2)
 "	stwcx.	%3,0,%2 \n\
 	bne-	1b"
-	PPC_ACQUIRE_BARRIER
+	PPC_ATOMIC_EXIT_BARRIER
 	: "=&r" (prev), "+m" (*(volatile unsigned int *)p)
 	: "r" (p), "r" (val)
 	: "cc", "memory");

@@ -61,12 +61,12 @@ __xchg_u64(volatile void *p, unsigned long val)
 	unsigned long prev;
 
 	__asm__ __volatile__(
-	PPC_RELEASE_BARRIER
+	PPC_ATOMIC_ENTRY_BARRIER
 "1:	ldarx	%0,0,%2 \n"
 	PPC405_ERR77(0,%2)
 "	stdcx.	%3,0,%2 \n\
 	bne-	1b"
-	PPC_ACQUIRE_BARRIER
+	PPC_ATOMIC_EXIT_BARRIER
 	: "=&r" (prev), "+m" (*(volatile unsigned long *)p)
 	: "r" (p), "r" (val)
 	: "cc", "memory");

@@ -152,14 +152,14 @@ __cmpxchg_u32(volatile unsigned int *p, unsigned long old, unsigned long new)
 	unsigned int prev;
 
 	__asm__ __volatile__ (
-	PPC_RELEASE_BARRIER
+	PPC_ATOMIC_ENTRY_BARRIER
 "1:	lwarx	%0,0,%2		# __cmpxchg_u32\n\
 	cmpw	0,%0,%3\n\
 	bne-	2f\n"
 	PPC405_ERR77(0,%2)
 "	stwcx.	%4,0,%2\n\
 	bne-	1b"
-	PPC_ACQUIRE_BARRIER
+	PPC_ATOMIC_EXIT_BARRIER
 	"\n\
 2:"
 	: "=&r" (prev), "+m" (*p)

@@ -198,13 +198,13 @@ __cmpxchg_u64(volatile unsigned long *p, unsigned long old, unsigned long new)
 	unsigned long prev;
 
 	__asm__ __volatile__ (
-	PPC_RELEASE_BARRIER
+	PPC_ATOMIC_ENTRY_BARRIER
 "1:	ldarx	%0,0,%2		# __cmpxchg_u64\n\
 	cmpd	0,%0,%3\n\
 	bne-	2f\n\
 	stdcx.	%4,0,%2\n\
 	bne-	1b"
-	PPC_ACQUIRE_BARRIER
+	PPC_ATOMIC_EXIT_BARRIER
 	"\n\
 2:"
 	: "=&r" (prev), "+m" (*p)

2.7.1

From: Ulrich Weigand <[email protected]>

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

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

commit 2e50c4bef77511b42cc226865d6bc568fa7f8769 upstream.

If a text section starts out with a data blob before the first function start label, disassembly parsing doing in recordmcount.pl gets confused on powerpc, leading to creation of corrupted module objects.

This was not a problem so far since the compiler would never create such text sections. However, this has changed with a recent change in GCC 6 to support distances of > 2GB between a function and its assoicated TOC in the ELFv2 ABI, exposing this problem.

There is already code in recordmcount.pl to handle such data blobs on the sparc64 platform. This patch uses the same method to handle those on powerpc as well.

---
 scripts/recordmcount.pl | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

index a27134fc3f76..efe7567d6225 100755
@@ -265,7 +265,8 @@ if ($arch eq "x86_64") {
 
 } elsif ($arch eq "powerpc") {
     $local_regex = "^[0-9a-fA-F]+\\s+t\\s+(\\.?\\S+)";
-    $function_regex = "^([0-9a-fA-F]+)\\s+<(\\.?.*?)>:";
+    # See comment in the sparc64 section for why we use '\w'.
+    $function_regex = "^([0-9a-fA-F]+)\\s+<(\\.?\\w*?)>:";
     $mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\s\\.?_mcount\$";
 
     if ($bits == 64) {

2.7.1

From: Marc Zyngier <[email protected]>

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

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

commit c0f0963464c24e034b858441205455bf2a5d93ad upstream.

When running a 32bit guest under a 64bit hypervisor, the ARMv8 architecture defines a mapping of the 32bit registers in the 64bit space. This includes banked registers that are being demultiplexed over the 64bit ones.

On exceptions caused by an operation involving a 32bit register, the HW exposes the register number in the ESR_EL2 register. It was so far understood that SW had to distinguish between AArch32 and AArch64 accesses (based on the current AArch32 mode and register number).

It turns out that I misinterpreted the ARM ARM, and the clue is in D1.20.1: "For some exceptions, the exception syndrome given in the ESR_ELx identifies one or more register numbers from the issued instruction that generated the exception. Where the exception is taken from an Exception level using AArch32 these register numbers give the AArch64 view of the register."

Which means that the HW is already giving us the translated version, and that we shouldn't try to interpret it at all (for example, doing an MMIO operation from the IRQ mode using the LR register leads to very unexpected behaviours).

The fix is thus not to perform a call to vcpu_reg32() at all from vcpu_reg(), and use whatever register number is supplied directly. The only case we need to find out about the mapping is when we actively generate a register access, which only occurs when injecting a fault in a guest.

---
 arch/arm64/include/asm/kvm_emulate.h | 8 +++++---
 arch/arm64/kvm/inject_fault.c        | 2 +-
 2 files changed, 6 insertions(+), 4 deletions(-)

index 2b01e2bdb7ef..a79bcce9b66d 100644
@@ -86,11 +86,13 @@ static inline void vcpu_set_thumb(struct kvm_vcpu *vcpu)
 	*vcpu_cpsr(vcpu) |= COMPAT_PSR_T_BIT;
 }
 
+/*
+ * vcpu_reg should always be passed a register number coming from a
+ * read of ESR_EL2. Otherwise, it may give the wrong result on AArch32
+ * with banked registers.
+ */
 static inline unsigned long *vcpu_reg(const struct kvm_vcpu *vcpu, u8 reg_num)
 {
-	if (vcpu_mode_is_32bit(vcpu))
-		return vcpu_reg32(vcpu, reg_num);
-
 	return (unsigned long *)&vcpu_gp_regs(vcpu)->regs.regs[reg_num];
 }
 

index 86825f8883de..f527a37ac979 100644
@@ -48,7 +48,7 @@ static void prepare_fault32(struct kvm_vcpu *vcpu, u32 mode, u32 vect_offset)
 
 	/* Note: These now point to the banked copies */
 	*vcpu_spsr(vcpu) = new_spsr_value;
-	*vcpu_reg(vcpu, 14) = *vcpu_pc(vcpu) + return_offset;
+	*vcpu_reg32(vcpu, 14) = *vcpu_pc(vcpu) + return_offset;
 
 	/* Branch to exception vector */
 	if (sctlr & (1 << 13))

2.7.1

From: Arnd Bergmann <[email protected]>

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

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

commit fbc416ff86183e2203cdf975e2881d7c164b0271 upstream.

As reported by Michal Simek, building an ARM64 kernel with CONFIG_UID16 disabled currently fails because the system call table still needs to reference the individual function entry points that are provided by kernel/sys_ni.c in this case, and the declarations are hidden inside of #ifdef CONFIG_UID16:

arch/arm64/include/asm/unistd32.h:57:8: error: 'sys_lchown16' undeclared here (not in a function)
__SYSCALL(__NR_lchown, sys_lchown16)

I believe this problem only exists on ARM64, because older architectures tend to not need declarations when their system call table is built in assembly code, while newer architectures tend to not need UID16 support. ARM64 only uses these system calls for compatibility with 32-bit ARM binaries.

This changes the CONFIG_UID16 check into CONFIG_HAVE_UID16, which is set unconditionally on ARM64 with CONFIG_COMPAT, so we see the declarations whenever we need them, but otherwise the behavior is unchanged.

---
 include/linux/syscalls.h | 2 +-
 include/linux/types.h    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

index 7fac04e7ff6e..f662a3719a1b 100644
@@ -499,7 +499,7 @@ asmlinkage long sys_chown(const char __user *filename,
 asmlinkage long sys_lchown(const char __user *filename,
 				uid_t user, gid_t group);
 asmlinkage long sys_fchown(unsigned int fd, uid_t user, gid_t group);
-#ifdef CONFIG_UID16
+#ifdef CONFIG_HAVE_UID16
 asmlinkage long sys_chown16(const char __user *filename,
 				old_uid_t user, old_gid_t group);
 asmlinkage long sys_lchown16(const char __user *filename,

index 4d118ba11349..83db8e5974dc 100644
@@ -35,7 +35,7 @@ typedef __kernel_gid16_t        gid16_t;
 
 typedef unsigned long		uintptr_t;
 
-#ifdef CONFIG_UID16
+#ifdef CONFIG_HAVE_UID16
 /* This is defined by include/asm-{arch}/posix_types.h */
 typedef __kernel_old_uid_t	old_uid_t;
 typedef __kernel_old_gid_t	old_gid_t;

2.7.1

From: John Blackwood <[email protected]>

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

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

commit 5db4fd8c52810bd9740c1240ebf89223b171aa70 upstream.

Make sure to clear out any ptrace singlestep state when a ptrace(2) PTRACE_DETACH call is made on arm64 systems.

Otherwise, the previously ptraced task will die off with a SIGTRAP signal if the debugger just previously singlestepped the ptraced task.

---
 arch/arm64/kernel/ptrace.c | 6 ++++++
 1 file changed, 6 insertions(+)

index ee79a1a6e965..9b9d651446ba 100644
@@ -51,6 +51,12 @@
  */
 void ptrace_disable(struct task_struct *child)
 {
+	/*
+	 * This would be better off in core code, but PTRACE_DETACH has
+	 * grown its fair share of arch-specific worts and changing it
+	 * is likely to cause regressions on obscure architectures.
+	 */
+	user_disable_single_step(child);
 }
 
 #ifdef CONFIG_HAVE_HW_BREAKPOINT

2.7.1

From: Will Deacon <[email protected]>

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

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

commit 32d6397805d00573ce1fa55f408ce2bca15b0ad3 upstream.

In paging_init, we allocate the zero page, memset it to zero and then point TTBR0 to it in order to avoid speculative fetches through the identity mapping.

In order to guarantee that the freshly zeroed page is indeed visible to the page table walker, we need to execute a dsb instruction prior to writing the TTBR.

---
 arch/arm64/mm/mmu.c | 3 +++
 1 file changed, 3 insertions(+)

index f8dc7e8fce6f..84ddb372fbc6 100644
@@ -374,6 +374,9 @@ void __init paging_init(void)
 
 	empty_zero_page = virt_to_page(zero_page);
 
+	/* Ensure the zero page is visible to the page table walker */
+	dsb();
+
 	/*
 	 * TTBR0 is only used for the identity mapping at this stage. Make it
 	 * point to zero page to avoid speculatively fetching new entries.

2.7.1

From: Mikulas Patocka <[email protected]>

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

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

commit e46e31a3696ae2d66f32c207df3969613726e636 upstream.

When using the Promise TX2+ SATA controller on PA-RISC, the system often crashes with kernel panic, for example just writing data with the dd utility will make it crash.

Kernel panic - not syncing: drivers/parisc/sba_iommu.c: I/O MMU @ 000000000000a000 is out of mapping resources

CPU: 0 PID: 18442 Comm: mkspadfs Not tainted 4.4.0-rc2 #2
Backtrace:
 [<000000004021497c>] show_stack+0x14/0x20
 [<0000000040410bf0>] dump_stack+0x88/0x100
 [<000000004023978c>] panic+0x124/0x360
 [<0000000040452c18>] sba_alloc_range+0x698/0x6a0
 [<0000000040453150>] sba_map_sg+0x260/0x5b8
 [<000000000c18dbb4>] ata_qc_issue+0x264/0x4a8 [libata]
 [<000000000c19535c>] ata_scsi_translate+0xe4/0x220 [libata]
 [<000000000c19a93c>] ata_scsi_queuecmd+0xbc/0x320 [libata]
 [<0000000040499bbc>] scsi_dispatch_cmd+0xfc/0x130
 [<000000004049da34>] scsi_request_fn+0x6e4/0x970
 [<00000000403e95a8>] __blk_run_queue+0x40/0x60
 [<00000000403e9d8c>] blk_run_queue+0x3c/0x68
 [<000000004049a534>] scsi_run_queue+0x2a4/0x360
 [<000000004049be68>] scsi_end_request+0x1a8/0x238
 [<000000004049de84>] scsi_io_completion+0xfc/0x688
 [<0000000040493c74>] scsi_finish_command+0x17c/0x1d0

The cause of the crash is not exhaustion of the IOMMU space, there is plenty of free pages. The function sba_alloc_range is called with size 0x11000, thus the pages_needed variable is 0x11. The function sba_search_bitmap is called with bits_wanted 0x11 and boundary size is 0x10 (because dma_get_seg_boundary(dev) returns 0xffff).

The function sba_search_bitmap attempts to allocate 17 pages that must not cross 16-page boundary - it can't satisfy this requirement (iommu_is_span_boundary always returns true) and fails even if there are many free entries in the IOMMU space.

How did it happen that we try to allocate 17 pages that don't cross 16-page boundary? The cause is in the function iommu_coalesce_chunks. This function tries to coalesce adjacent entries in the scatterlist. The function does several checks if it may coalesce one entry with the next, one of those checks is this:

	if (startsg->length + dma_len > max_seg_size)
		break;

When it finishes coalescing adjacent entries, it allocates the mapping:

sg_dma_len(contig_sg) = dma_len;
dma_len = ALIGN(dma_len + dma_offset, IOVP_SIZE);
sg_dma_address(contig_sg) =

	PIDE_FLAG
	| (iommu_alloc_range(ioc, dev, dma_len) << IOVP_SHIFT)
	| dma_offset;

It is possible that (startsg->length + dma_len > max_seg_size) is false (we are just near the 0x10000 max_seg_size boundary), so the funcion decides to coalesce this entry with the next entry. When the coalescing succeeds, the function performs

	dma_len = ALIGN(dma_len + dma_offset, IOVP_SIZE);

And now, because of non-zero dma_offset, dma_len is greater than 0x10000. iommu_alloc_range (a pointer to sba_alloc_range) is called and it attempts to allocate 17 pages for a device that must not cross 16-page boundary.

To fix the bug, we must make sure that dma_len after addition of dma_offset and alignment doesn't cross the segment boundary. I.e. change

	if (startsg->length + dma_len > max_seg_size)
		break;
to
	if (ALIGN(dma_len + dma_offset + startsg->length, IOVP_SIZE) > max_seg_size)
		break;

This patch makes this change (it precalculates max_seg_boundary at the beginning of the function iommu_coalesce_chunks). I also added a check that the mapping length doesn't exceed dma_get_seg_boundary(dev) (it is not needed for Promise TX2+ SATA, but it may be needed for other devices that have dma_get_seg_boundary lower than dma_get_max_seg_size).

---
 drivers/parisc/iommu-helpers.h | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

index 8c33491b21fe..c6aa38883466 100644
@@ -104,7 +104,11 @@ iommu_coalesce_chunks(struct ioc *ioc, struct device *dev,
 	struct scatterlist *contig_sg;	   /* contig chunk head */
 	unsigned long dma_offset, dma_len; /* start/len of DMA stream */
 	unsigned int n_mappings = 0;
-	unsigned int max_seg_size = dma_get_max_seg_size(dev);
+	unsigned int max_seg_size = min(dma_get_max_seg_size(dev),
+					(unsigned)DMA_CHUNK_SIZE);
+	unsigned int max_seg_boundary = dma_get_seg_boundary(dev) + 1;
+	if (max_seg_boundary)	/* check if the addition above didn't overflow */
+		max_seg_size = min(max_seg_size, max_seg_boundary);
 
 	while (nents > 0) {
 

@@ -139,14 +143,11 @@ iommu_coalesce_chunks(struct ioc *ioc, struct device *dev,
 
 			/*
 			** First make sure current dma stream won't
-			** exceed DMA_CHUNK_SIZE if we coalesce the
+			** exceed max_seg_size if we coalesce the
 			** next entry.
 			*/   
-			if(unlikely(ALIGN(dma_len + dma_offset + startsg->length,
-					    IOVP_SIZE) > DMA_CHUNK_SIZE))
-				break;
-
-			if (startsg->length + dma_len > max_seg_size)
+			if (unlikely(ALIGN(dma_len + dma_offset + startsg->length, IOVP_SIZE) >
+				     max_seg_size))
 				break;
 
 			/*

2.7.1

From: Richard Purdie <[email protected]>

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

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

commit 79b568b9d0c7c5d81932f4486d50b38efdd6da6d upstream.

hid_connect adds various strings to the buffer but they're all conditional. You can find circumstances where nothing would be written to it but the kernel will still print the supposedly empty buffer with printk. This leads to corruption on the console/in the logs.

Ensure buf is initialized to an empty string.

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

index 85b0da8c33f4..7ca1b4a97a14 100644
@@ -1514,7 +1514,7 @@ int hid_connect(struct hid_device *hdev, unsigned int connect_mask)
 		"Multi-Axis Controller"
 	};
 	const char *type, *bus;
-	char buf[64];
+	char buf[64] = "";
 	unsigned int i;
 	int len;
 	int ret;

2.7.1

From: Guenter Roeck <[email protected]>

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

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

commit c86576ea114a9a881cf7328dc7181052070ca311 upstream.

mn10300 builds fail with

fs/stat.c: In function 'cp_old_stat':
fs/stat.c:163:2: error: 'old_uid_t' undeclared

ipc/util.c: In function 'ipc64_perm_to_ipc_perm':
ipc/util.c:540:2: error: 'old_uid_t' undeclared

Select CONFIG_HAVE_UID16 and remove local definition of CONFIG_UID16 to fix the problem.

---
 arch/mn10300/Kconfig | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

index 6aaa1607001a..c61bf144b8f2 100644
@@ -1,6 +1,7 @@
 config MN10300
 	def_bool y
 	select HAVE_OPROFILE
+	select HAVE_UID16
 	select GENERIC_IRQ_SHOW
 	select ARCH_WANT_IPC_PARSE_VERSION
 	select HAVE_ARCH_TRACEHOOK

@@ -37,9 +38,6 @@ config HIGHMEM
 config NUMA
 	def_bool n
 
-config UID16
-	def_bool y
-
 config RWSEM_GENERIC_SPINLOCK
 	def_bool y
 

2.7.1

From: Yang Shi <[email protected]>

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

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

commit 92e788b749862ebe9920360513a718e5dd4da7a9 upstream.

As previously reported, some userspace applications depend on bogomips showed by /proc/cpuinfo. Although there is much less legacy impact on aarch64 than arm, it does break libvirt.

This patch reverts commit 326b16db9f69 ("arm64: delay: don't bother reporting bogomips in /proc/cpuinfo"), but with some tweak due to context change and without the pr_info().

---
 arch/arm64/kernel/setup.c | 4 ++++
 1 file changed, 4 insertions(+)

index 24bf1563c3bd..59411c933393 100644
@@ -365,6 +365,10 @@ static int c_show(struct seq_file *m, void *v)
 		seq_printf(m, "processor\t: %d\n", i);
 #endif
 
+		seq_printf(m, "BogoMIPS\t: %lu.%02lu\n",
+			   loops_per_jiffy / (500000UL/HZ),
+			   loops_per_jiffy / (5000UL/HZ) % 100);
+
 		/*
 		 * Dump out the common processor features in a single line.
 		 * Userspace should read the hwcaps with getauxval(AT_HWCAP)

2.7.1

From: Steven Noonan <[email protected]>

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

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

commit 5631b8fba640a4ab2f8a954f63a603fa34eda96b upstream.

The bug referenced by the comment in this commit was not completely fixed in GCC 4.8.2, as I mentioned in a thread back in February:

https://lkml.org/lkml/2014/2/12/797

The conclusion at that time was to make the quirk unconditional until the bug could be found and fixed in GCC. Unfortunately, when I submitted the patch (commit a9f18034) I left a comment in that claimed the bug was fixed in GCC 4.8.2+.

This comment is inaccurate, and should be removed.

---
 include/linux/compiler-gcc4.h | 1 -
 include/linux/compiler-gcc5.h | 1 -
 2 files changed, 2 deletions(-)

index 2507fd2a1eb4..d1a558239b1a 100644
@@ -71,7 +71,6 @@
  *   http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58670
  *
  * Work it around via a compiler barrier quirk suggested by Jakub Jelinek.
- * Fixed in GCC 4.8.2 and later versions.
  *
  * (asm goto is automatically volatile - the naming reflects this.)
  */

index cdd1cc202d51..c8c565952548 100644
@@ -53,7 +53,6 @@
  *   http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58670
  *
  * Work it around via a compiler barrier quirk suggested by Jakub Jelinek.
- * Fixed in GCC 4.8.2 and later versions.
  *
  * (asm goto is automatically volatile - the naming reflects this.)
  */

2.7.1

From: Joe Perches <[email protected]>

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

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

commit cb984d101b30eb7478d32df56a0023e4603cba7f upstream.

As gcc major version numbers are going to advance rather rapidly in the future, there's no real value in separate files for each compiler version.

Deduplicate some of the macros #defined in each file too.

Neaten comments using normal kernel commenting style.

---
 include/linux/compiler-gcc.h  | 114 ++++++++++++++++++++++++++++++++++++++++--
 include/linux/compiler-gcc3.h |  23 ---------
 include/linux/compiler-gcc4.h |  87 --------------------------------
 include/linux/compiler-gcc5.h |  65 ------------------------
 4 files changed, 110 insertions(+), 179 deletions(-)
 delete mode 100644 include/linux/compiler-gcc3.h
 delete mode 100644 include/linux/compiler-gcc4.h
 delete mode 100644 include/linux/compiler-gcc5.h

index 02ae99e8e6d3..65856c3599b4 100644
@@ -100,10 +100,116 @@
 #define __maybe_unused			__attribute__((unused))
 #define __always_unused			__attribute__((unused))
 
-#define __gcc_header(x) #x
-#define _gcc_header(x) __gcc_header(linux/compiler-gcc##x.h)
-#define gcc_header(x) _gcc_header(x)
-#include gcc_header(__GNUC__)
+/* gcc version specific checks */
+
+#if GCC_VERSION < 30200
+# error Sorry, your compiler is too old - please upgrade it.
+#endif
+
+#if GCC_VERSION < 30300
+# define __used			__attribute__((__unused__))
+#else
+# define __used			__attribute__((__used__))
+#endif
+
+#ifdef CONFIG_GCOV_KERNEL
+# if GCC_VERSION < 30400
+#   error "GCOV profiling support for gcc versions below 3.4 not included"
+# endif /* __GNUC_MINOR__ */
+#endif /* CONFIG_GCOV_KERNEL */
+
+#if GCC_VERSION >= 30400
+#define __must_check		__attribute__((warn_unused_result))
+#endif
+
+#if GCC_VERSION >= 40000
+
+/* GCC 4.1.[01] miscompiles __weak */
+#ifdef __KERNEL__
+# if GCC_VERSION >= 40100 &&  GCC_VERSION <= 40101
+#  error Your version of gcc miscompiles the __weak directive
+# endif
+#endif
+
+#define __used			__attribute__((__used__))
+#define __compiler_offsetof(a, b)					\
+	__builtin_offsetof(a, b)
+
+#if GCC_VERSION >= 40100 && GCC_VERSION < 40600
+# define __compiletime_object_size(obj) __builtin_object_size(obj, 0)
+#endif
+
+#if GCC_VERSION >= 40300
+/* Mark functions as cold. gcc will assume any path leading to a call
+ * to them will be unlikely.  This means a lot of manual unlikely()s
+ * are unnecessary now for any paths leading to the usual suspects
+ * like BUG(), printk(), panic() etc. [but let's keep them for now for
+ * older compilers]
+ *
+ * Early snapshots of gcc 4.3 don't support this and we can't detect this
+ * in the preprocessor, but we can live with this because they're unreleased.
+ * Maketime probing would be overkill here.
+ *
+ * gcc also has a __attribute__((__hot__)) to move hot functions into
+ * a special section, but I don't see any sense in this right now in
+ * the kernel context
+ */
+#define __cold			__attribute__((__cold__))
+
+#define __UNIQUE_ID(prefix) __PASTE(__PASTE(__UNIQUE_ID_, prefix), __COUNTER__)
+
+#ifndef __CHECKER__
+# define __compiletime_warning(message) __attribute__((warning(message)))
+# define __compiletime_error(message) __attribute__((error(message)))
+#endif /* __CHECKER__ */
+#endif /* GCC_VERSION >= 40300 */
+
+#if GCC_VERSION >= 40500
+/*
+ * Mark a position in code as unreachable.  This can be used to
+ * suppress control flow warnings after asm blocks that transfer
+ * control elsewhere.
+ *
+ * Early snapshots of gcc 4.5 don't support this and we can't detect
+ * this in the preprocessor, but we can live with this because they're
+ * unreleased.  Really, we need to have autoconf for the kernel.
+ */
+#define unreachable() __builtin_unreachable()
+
+/* Mark a function definition as prohibited from being cloned. */
+#define __noclone	__attribute__((__noclone__))
+
+#endif /* GCC_VERSION >= 40500 */
+
+#if GCC_VERSION >= 40600
+/*
+ * Tell the optimizer that something else uses this function or variable.
+ */
+#define __visible	__attribute__((externally_visible))
+#endif
+
+/*
+ * GCC 'asm goto' miscompiles certain code sequences:
+ *
+ *   http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58670
+ *
+ * Work it around via a compiler barrier quirk suggested by Jakub Jelinek.
+ *
+ * (asm goto is automatically volatile - the naming reflects this.)
+ */
+#define asm_volatile_goto(x...)	do { asm goto(x); asm (""); } while (0)
+
+#ifdef CONFIG_ARCH_USE_BUILTIN_BSWAP
+#if GCC_VERSION >= 40400
+#define __HAVE_BUILTIN_BSWAP32__
+#define __HAVE_BUILTIN_BSWAP64__
+#endif
+#if GCC_VERSION >= 40800 || (defined(__powerpc__) && GCC_VERSION >= 40600)
+#define __HAVE_BUILTIN_BSWAP16__
+#endif
+#endif /* CONFIG_ARCH_USE_BUILTIN_BSWAP */
+
+#endif	/* gcc version >= 40000 specific checks */
 
 #if !defined(__noclone)
 #define __noclone	/* not needed */

deleted file mode 100644
index 7d89febe4d79..000000000000
@@ -1,23 +0,0 @@
-#ifndef __LINUX_COMPILER_H
-#error "Please don't include <linux/compiler-gcc3.h> directly, include <linux/compiler.h> instead."
-#endif
-
-#if GCC_VERSION < 30200
-# error Sorry, your compiler is too old - please upgrade it.
-#endif
-
-#if GCC_VERSION >= 30300
-# define __used			__attribute__((__used__))
-#else
-# define __used			__attribute__((__unused__))
-#endif
-
-#if GCC_VERSION >= 30400
-#define __must_check		__attribute__((warn_unused_result))
-#endif
-
-#ifdef CONFIG_GCOV_KERNEL
-# if GCC_VERSION < 30400
-#   error "GCOV profiling support for gcc versions below 3.4 not included"
-# endif /* __GNUC_MINOR__ */
-#endif /* CONFIG_GCOV_KERNEL */

2.7.1

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

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

On 3.12, with gcc-6, I see a lot of:
arch/x86/include/asm/vvar.h:33:28: warning: ‘vvaraddr_jiffies’ defined but not used [-Wunused-const-variable]

  static type const * const vvaraddr_ ## name =   \
                            ^

arch/x86/include/asm/vvar.h:46:1: note: in expansion of macro ‘DECLARE_VVAR’
DECLARE_VVAR(0, volatile unsigned long, jiffies)
^~~~~~~~~~~~