3.10.93-stable review

This is the start of the stable review cycle for the 3.10.93 release. There are 24 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 Sun Nov 8 19:23:41 UTC 2015. Anything received after that time might be too late.

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

	kernel.org/pub/linux/kernel/v3.x/stable-review/patch-3.10.93-rc1.gz

and the diffstat can be found below.

thanks,

greg k-h

-------------
Pseudo-Shortlog of commits:

Greg Kroah-Hartman <[email protected]>

    Linux 3.10.93-rc1

Greg Kroah-Hartman <[email protected]>
xen: fix backport of previous kexec patch

Doron Tsur <[email protected]>
IB/cm: Fix rb-tree duplicate free and use-after-free

Dāvis Mosāns <[email protected]>
mvsas: Fix NULL pointer dereference in mvs_slot_task_free

Jes Sorensen <[email protected]>
md/raid10: submit_bio_wait() returns 0 on success

Jes Sorensen <[email protected]>
md/raid1: submit_bio_wait() returns 0 on success

Herbert Xu <[email protected]>
crypto: api - Only abort operations on fatal signal

Peter Zijlstra <[email protected]>
module: Fix locking in symbol_put_addr()

Cathy Avery <[email protected]>
xen-blkfront: check for null drvdata in blkback_changed (XenbusStateClosing)

Mathias Nyman <[email protected]>
xhci: handle no ping response error properly

Mike Snitzer <[email protected]>
dm btree: fix leak of bufio-backed block in btree_split_beneath error path

Joe Thornber <[email protected]>
dm btree remove: fix a bug when rebalancing nodes after removal

Will Deacon <[email protected]>
Revert "ARM64: unwind: Fix PC calculation"

Ilya Dryomov <[email protected]>
rbd: prevent kernel stack blow up on rbd map

Ilya Dryomov <[email protected]>
rbd: don't leak parent_spec in rbd_dev_probe_parent()

Ronny Hegewald <[email protected]>
rbd: require stable pages if message data CRCs are enabled

Ilia Mirkin <[email protected]>
drm/nouveau/gem: return only valid domain when there's only one

Jan Kara <[email protected]>
mm: make sendfile(2) killable

Paolo Bonzini <[email protected]>
x86/setup: Extend low identity map to cover whole kernel range

Charles Keepax <[email protected]>
ASoC: wm8904: Correct number of EQ registers

Vasant Hegde <[email protected]>
powerpc/rtas: Validate rtas.entry before calling enter_rtas()

Joerg Roedel <[email protected]>
iommu/amd: Don't clear DTE flags when modifying it

Johannes Berg <[email protected]>
iwlwifi: mvm: fix D3 firmware PN programming

Johannes Berg <[email protected]>
iwlwifi: dvm: fix D3 firmware PN programming

Felix Fietkau <[email protected]>
ath9k: declare required extra tx headroom

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

Diffstat:

 Makefile                                     |  4 +-
 arch/arm64/kernel/stacktrace.c               |  6 +--
 arch/powerpc/kernel/rtas.c                   |  3 ++
 arch/x86/kernel/setup.c                      |  8 ++++
 arch/x86/xen/enlighten.c                     |  6 +--
 crypto/ablkcipher.c                          |  2 +-
 crypto/algapi.c                              |  2 +-
 crypto/api.c                                 |  6 +--
 crypto/crypto_user.c                         |  2 +-
 drivers/block/rbd.c                          | 67 ++++++++++++++++------------
 drivers/block/xen-blkfront.c                 |  3 +-
 drivers/gpu/drm/nouveau/nouveau_gem.c        |  5 ++-
 drivers/infiniband/core/cm.c                 | 10 ++++-
 drivers/iommu/amd_iommu.c                    |  4 +-
 drivers/iommu/amd_iommu_types.h              |  1 +
 drivers/md/persistent-data/dm-btree-remove.c | 17 ++++---
 drivers/md/persistent-data/dm-btree.c        |  2 +-
 drivers/md/raid1.c                           |  2 +-
 drivers/md/raid10.c                          |  2 +-
 drivers/net/wireless/ath/ath9k/init.c        |  1 +
 drivers/net/wireless/iwlwifi/dvm/lib.c       |  2 +-
 drivers/net/wireless/iwlwifi/mvm/d3.c        | 12 ++---
 drivers/scsi/mvsas/mv_sas.c                  |  2 +
 drivers/usb/host/xhci-ring.c                 | 20 ++++++---
 include/sound/wm8904.h                       |  2 +-
 kernel/module.c                              |  8 +++-
 mm/filemap.c                                 |  9 ++--
 27 files changed, 130 insertions(+), 78 deletions(-)

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

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

From: Felix Fietkau <[email protected]>

commit 029cd0370241641eb70235d205aa0b90c84dce44 upstream.

ath9k inserts padding between the 802.11 header and the data area (to align it). Since it didn't declare this extra required headroom, this led to some nasty issues like randomly dropped packets in some setups.

---
 drivers/net/wireless/ath/ath9k/init.c |    1 +
 1 file changed, 1 insertion(+)

@@ -819,6 +819,7 @@ void ath9k_set_hw_capab(struct ath_softc
 	hw->max_rate_tries = 10;
 	hw->sta_data_size = sizeof(struct ath_node);
 	hw->vif_data_size = sizeof(struct ath_vif);
+	hw->extra_tx_headroom = 4;
 
 	hw->wiphy->available_antennas_rx = BIT(ah->caps.max_rxchains) - 1;
 	hw->wiphy->available_antennas_tx = BIT(ah->caps.max_txchains) - 1;

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

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

From: Johannes Berg <[email protected]>

commit 5bd166872d8f99f156fac191299d24f828bb2348 upstream.

The code to send the RX PN data (for each TID) to the firmware has a devastating bug: it overwrites the data for TID 0 with all the TID data, leaving the remaining TIDs zeroed. This will allow replays to actually be accepted by the firmware, which could allow waking up the system.

---
 drivers/net/wireless/iwlwifi/dvm/lib.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -1023,7 +1023,7 @@ static void iwlagn_wowlan_program_keys(s
 			u8 *pn = seq.ccmp.pn;
 
 			ieee80211_get_key_rx_seq(key, i, &seq);
-			aes_sc->pn = cpu_to_le64(
+			aes_sc[i].pn = cpu_to_le64(
 					(u64)pn[5] |
 					((u64)pn[4] << 8) |
 					((u64)pn[3] << 16) |

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

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

From: Johannes Berg <[email protected]>

commit 2cf5eb3ab7bb7f2e3a70edcef236cd62c87db030 upstream.

The code to send the RX PN data (for each TID) to the firmware has a devastating bug: it overwrites the data for TID 0 with all the TID data, leaving the remaining TIDs zeroed. This will allow replays to actually be accepted by the firmware, which could allow waking up the system.

---
 drivers/net/wireless/iwlwifi/mvm/d3.c |   12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

@@ -295,12 +295,12 @@ static void iwl_mvm_wowlan_program_keys(
 			u8 *pn = seq.ccmp.pn;
 
 			ieee80211_get_key_rx_seq(key, i, &seq);
-			aes_sc->pn = cpu_to_le64((u64)pn[5] |
-						 ((u64)pn[4] << 8) |
-						 ((u64)pn[3] << 16) |
-						 ((u64)pn[2] << 24) |
-						 ((u64)pn[1] << 32) |
-						 ((u64)pn[0] << 40));
+			aes_sc[i].pn = cpu_to_le64((u64)pn[5] |
+						   ((u64)pn[4] << 8) |
+						   ((u64)pn[3] << 16) |
+						   ((u64)pn[2] << 24) |
+						   ((u64)pn[1] << 32) |
+						   ((u64)pn[0] << 40));
 		}
 		data->use_rsc_tsc = true;
 		break;

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

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

From: Joerg Roedel <[email protected]>

commit cbf3ccd09d683abf1cacd36e3640872ee912d99b upstream.

During device assignment/deassignment the flags in the DTE get lost, which might cause spurious faults, for example when the device tries to access the system management range. Fix this by not clearing the flags with the rest of the DTE.

---
 drivers/iommu/amd_iommu.c       |    4 ++--
 drivers/iommu/amd_iommu_types.h |    1 +
 2 files changed, 3 insertions(+), 2 deletions(-)

@@ -2106,8 +2106,8 @@ static void set_dte_entry(u16 devid, str
 static void clear_dte_entry(u16 devid)
 {
 	/* remove entry from the device table seen by the hardware */
-	amd_iommu_dev_table[devid].data[0] = IOMMU_PTE_P | IOMMU_PTE_TV;
-	amd_iommu_dev_table[devid].data[1] = 0;
+	amd_iommu_dev_table[devid].data[0]  = IOMMU_PTE_P | IOMMU_PTE_TV;
+	amd_iommu_dev_table[devid].data[1] &= DTE_FLAG_MASK;
 
 	amd_iommu_apply_erratum_63(devid);
 }

@@ -282,6 +282,7 @@
 #define IOMMU_PTE_IR (1ULL << 61)
 #define IOMMU_PTE_IW (1ULL << 62)
 
+#define DTE_FLAG_MASK	(0x3ffULL << 32)
 #define DTE_FLAG_IOTLB	(0x01UL << 32)
 #define DTE_FLAG_GV	(0x01ULL << 55)
 #define DTE_GLX_SHIFT	(56)

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

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

From: Vasant Hegde <[email protected]>

commit 8832317f662c06f5c06e638f57bfe89a71c9b266 upstream.

Currently we do not validate rtas.entry before calling enter_rtas(). This leads to a kernel oops when user space calls rtas system call on a powernv platform (see below). This patch adds code to validate rtas.entry before making enter_rtas() call.

  Oops: Exception in kernel mode, sig: 4 [#1]
  SMP NR_CPUS=1024 NUMA PowerNV
  task: c000000004294b80 ti: c0000007e1a78000 task.ti: c0000007e1a78000
  NIP: 0000000000000000 LR: 0000000000009c14 CTR: c000000000423140
  REGS: c0000007e1a7b920 TRAP: 0e40   Not tainted  (3.18.17-340.el7_1.pkvm3_1_0.2400.1.ppc64le)
  MSR: 1000000000081000 <HV,ME>  CR: 00000000  XER: 00000000
  CFAR: c000000000009c0c SOFTE: 0
  NIP [0000000000000000]           (null)
  LR [0000000000009c14] 0x9c14
  Call Trace:
  [c0000007e1a7bba0] [c00000000041a7f4] avc_has_perm_noaudit+0x54/0x110 (unreliable)
  [c0000007e1a7bd80] [c00000000002ddc0] ppc_rtas+0x150/0x2d0
  [c0000007e1a7be30] [c000000000009358] syscall_exit+0x0/0x98

---
 arch/powerpc/kernel/rtas.c |    3 +++
 1 file changed, 3 insertions(+)

@@ -1041,6 +1041,9 @@ asmlinkage int ppc_rtas(struct rtas_args
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
+	if (!rtas.entry)
+		return -EINVAL;
+
 	if (copy_from_user(&args, uargs, 3 * sizeof(u32)) != 0)
 		return -EFAULT;
 

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

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

From: Charles Keepax <[email protected]>

commit 97aff2c03a1e4d343266adadb52313613efb027f upstream.

There are 24 EQ registers not 25, I suspect this bug came about because the registers start at EQ1 not zero. The bug is relatively harmless as the extra register written is an unused one.

---
 include/sound/wm8904.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -119,7 +119,7 @@
 #define WM8904_MIC_REGS  2
 #define WM8904_GPIO_REGS 4
 #define WM8904_DRC_REGS  4
-#define WM8904_EQ_REGS   25
+#define WM8904_EQ_REGS   24
 
 /**
  * DRC configurations are specified with a label and a set of register

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

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

From: Paolo Bonzini <[email protected]>

commit f5f3497cad8c8416a74b9aaceb127908755d020a upstream.

On 32-bit systems, the initial_page_table is reused by efi_call_phys_prolog as an identity map to call SetVirtualAddressMap. efi_call_phys_prolog takes care of converting the current CPU's GDT to a physical address too.

For PAE kernels the identity mapping is achieved by aliasing the first PDPE for the kernel memory mapping into the first PDPE of initial_page_table. This makes the EFI stub's trick "just work".

However, for non-PAE kernels there is no guarantee that the identity mapping in the initial_page_table extends as far as the GDT; in this case, accesses to the GDT will cause a page fault (which quickly becomes a triple fault). Fix this by copying the kernel mappings from swapper_pg_dir to initial_page_table twice, both at PAGE_OFFSET and at identity mapping.

For some reason, this is only reproducible with QEMU's dynamic translation mode, and not for example with KVM. However, even under KVM one can clearly see that the page table is bogus:

    $ qemu-system-i386 -pflash OVMF.fd -M q35 vmlinuz0 -s -S -daemonize
    $ gdb
    (gdb) target remote localhost:1234
    (gdb) hb *0x02858f6f
    Hardware assisted breakpoint 1 at 0x2858f6f
    (gdb) c
    Continuing.

    Breakpoint 1, 0x02858f6f in ?? ()
    (gdb) monitor info registers
    ...
    GDT=     0724e000 000000ff
    IDT=     fffbb000 000007ff
    CR0=0005003b CR2=ff896000 CR3=032b7000 CR4=00000690
    ...

The page directory is sane:

    (gdb) x/4wx 0x32b7000
    0x32b7000:	0x03398063	0x03399063	0x0339a063	0x0339b063
    (gdb) x/4wx 0x3398000
    0x3398000:	0x00000163	0x00001163	0x00002163	0x00003163
    (gdb) x/4wx 0x3399000
    0x3399000:	0x00400003	0x00401003	0x00402003	0x00403003

but our particular page directory entry is empty:

    (gdb) x/1wx 0x32b7000 + (0x724e000 >> 22) * 4
    0x32b7070:	0x00000000

[ It appears that you can skate past this issue if you don't receive
any interrupts while the bogus GDT pointer is loaded, or if you avoid reloading the segment registers in general.

Andy Lutomirski provides some additional insight:

"AFAICT it's entirely permissible for the GDTR and/or LDT
descriptor to point to unmapped memory. Any attempt to use them (segment loads, interrupts, IRET, etc) will try to access that memory as if the access came from CPL 0 and, if the access fails, will generate a valid page fault with CR2 pointing into the GDT or LDT."

Up until commit 23a0d4e8fa6d ("efi: Disable interrupts around EFI calls, not in the epilog/prolog calls") interrupts were disabled around the prolog and epilog calls, and the functional GDT was re-installed before interrupts were re-enabled.

Which explains why no one has hit this issue until now. ]

---
 arch/x86/kernel/setup.c |    8 ++++++++
 1 file changed, 8 insertions(+)

@@ -1156,6 +1156,14 @@ void __init setup_arch(char **cmdline_p)
 	clone_pgd_range(initial_page_table + KERNEL_PGD_BOUNDARY,
 			swapper_pg_dir     + KERNEL_PGD_BOUNDARY,
 			KERNEL_PGD_PTRS);
+
+	/*
+	 * sync back low identity map too.  It is used for example
+	 * in the 32-bit EFI stub.
+	 */
+	clone_pgd_range(initial_page_table,
+			swapper_pg_dir     + KERNEL_PGD_BOUNDARY,
+			KERNEL_PGD_PTRS);
 #endif
 
 	tboot_probe();

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

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

From: Jan Kara <[email protected]>

commit 296291cdd1629c308114504b850dc343eabc2782 upstream.

Currently a simple program below issues a sendfile(2) system call which takes about 62 days to complete in my test KVM instance.

        int fd;
        off_t off = 0;

        fd = open("file", O_RDWR | O_TRUNC | O_SYNC | O_CREAT, 0644);
        ftruncate(fd, 2);
        lseek(fd, 0, SEEK_END);
        sendfile(fd, fd, &off, 0xfffffff);

Now you should not ask kernel to do a stupid stuff like copying 256MB in 2-byte chunks and call fsync(2) after each chunk but if you do, sysadmin should have a way to stop you.

We actually do have a check for fatal_signal_pending() in generic_perform_write() which triggers in this path however because we always succeed in writing something before the check is done, we return value > 0 from generic_perform_write() and thus the information about signal gets lost.

Fix the problem by doing the signal check before writing anything. That way generic_perform_write() returns -EINTR, the error gets propagated up and the sendfile loop terminates early.

---
 mm/filemap.c |    9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

@@ -2340,6 +2340,11 @@ again:
 			break;
 		}
 
+		if (fatal_signal_pending(current)) {
+			status = -EINTR;
+			break;
+		}
+
 		status = a_ops->write_begin(file, mapping, pos, bytes, flags,
 						&page, &fsdata);
 		if (unlikely(status))

@@ -2380,10 +2385,6 @@ again:
 		written += copied;
 
 		balance_dirty_pages_ratelimited(mapping);
-		if (fatal_signal_pending(current)) {
-			status = -EINTR;
-			break;
-		}
 	} while (iov_iter_count(i));
 
 	return written ? written : status;

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

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

From: Ilia Mirkin <[email protected]>

commit 2a6c521bb41ce862e43db46f52e7681d33e8d771 upstream.

On nv50+, we restrict the valid domains to just the one where the buffer was originally created. However after the buffer is evicted to system memory, we might move it back to a different domain that was not originally valid. When sharing the buffer and retrieving its GEM_INFO data, we still want the domain that will be valid for this buffer in a pushbuf, not the one where it currently happens to be.

This resolves fdo#92504 and several others. These are due to suspend evicting all buffers, making it more likely that they temporarily end up in the wrong place.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92504

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

@@ -177,11 +177,12 @@ nouveau_gem_info(struct drm_file *file_p
 	struct nouveau_bo *nvbo = nouveau_gem_object(gem);
 	struct nouveau_vma *vma;
 
-	if (nvbo->bo.mem.mem_type == TTM_PL_TT)
+	if (is_power_of_2(nvbo->valid_domains))
+		rep->domain = nvbo->valid_domains;
+	else if (nvbo->bo.mem.mem_type == TTM_PL_TT)
 		rep->domain = NOUVEAU_GEM_DOMAIN_GART;
 	else
 		rep->domain = NOUVEAU_GEM_DOMAIN_VRAM;
-
 	rep->offset = nvbo->bo.offset;
 	if (cli->base.vm) {
 		vma = nouveau_bo_vma_find(nvbo, cli->base.vm);

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

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

From: Ronny Hegewald <[email protected]>

commit bae818ee1577c27356093901a0ea48f672eda514 upstream.

rbd requires stable pages, as it performs a crc of the page data before they are send to the OSDs.

But since kernel 3.9 (patch 1d1d1a767206fbe5d4c69493b7e6d2a8d08cc0a0 "mm: only enforce stable page writes if the backing device requires it") it is not assumed anymore that block devices require stable pages.

This patch sets the necessary flag to get stable pages back for rbd.

In a ceph installation that provides multiple ext4 formatted rbd devices "bad crc" messages appeared regularly (ca 1 message every 1-2 minutes on every OSD that provided the data for the rbd) in the OSD-logs before this patch. After this patch this messages are pretty much gone (only ca 1-2 / month / OSD).

---
 drivers/block/rbd.c |    3 +++
 1 file changed, 3 insertions(+)

@@ -3453,6 +3453,9 @@ static int rbd_init_disk(struct rbd_devi
 	blk_queue_io_opt(q, segment_size);
 
 	blk_queue_merge_bvec(q, rbd_merge_bvec);
+	if (!ceph_test_opt(rbd_dev->rbd_client->client, NOCRC))
+		q->backing_dev_info.capabilities |= BDI_CAP_STABLE_WRITES;
+
 	disk->queue = q;
 
 	q->queuedata = rbd_dev;

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

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

From: Ilya Dryomov <[email protected]>

commit 1f2c6651f69c14d0d3a9cfbda44ea101b02160ba upstream.

Currently we leak parent_spec and trigger a "parent reference underflow" warning if rbd_dev_create() in rbd_dev_probe_parent() fails. The problem is we take the !parent out_err branch and that only drops refcounts; parent_spec that would've been freed had we called rbd_dev_unparent() remains and triggers rbd_warn() in rbd_dev_parent_put() - at that point we have parent_spec != NULL and parent_ref == 0, so counter ends up being -1 after the decrement.

Redo rbd_dev_probe_parent() to fix this.

---
 drivers/block/rbd.c |   35 +++++++++++++++--------------------
 1 file changed, 15 insertions(+), 20 deletions(-)

@@ -4834,41 +4834,36 @@ out_err:
 static int rbd_dev_probe_parent(struct rbd_device *rbd_dev)
 {
 	struct rbd_device *parent = NULL;
-	struct rbd_spec *parent_spec;
-	struct rbd_client *rbdc;
 	int ret;
 
 	if (!rbd_dev->parent_spec)
 		return 0;
-	/*
-	 * We need to pass a reference to the client and the parent
-	 * spec when creating the parent rbd_dev.  Images related by
-	 * parent/child relationships always share both.
-	 */
-	parent_spec = rbd_spec_get(rbd_dev->parent_spec);
-	rbdc = __rbd_get_client(rbd_dev->rbd_client);
 
-	ret = -ENOMEM;
-	parent = rbd_dev_create(rbdc, parent_spec);
-	if (!parent)
+	parent = rbd_dev_create(rbd_dev->rbd_client, rbd_dev->parent_spec);
+	if (!parent) {
+		ret = -ENOMEM;
 		goto out_err;
+	}
+
+	/*
+	 * Images related by parent/child relationships always share
+	 * rbd_client and spec/parent_spec, so bump their refcounts.
+	 */
+	__rbd_get_client(rbd_dev->rbd_client);
+	rbd_spec_get(rbd_dev->parent_spec);
 
 	ret = rbd_dev_image_probe(parent, false);
 	if (ret < 0)
 		goto out_err;
+
 	rbd_dev->parent = parent;
 	atomic_set(&rbd_dev->parent_ref, 1);
-
 	return 0;
+
 out_err:
-	if (parent) {
-		rbd_dev_unparent(rbd_dev);
+	rbd_dev_unparent(rbd_dev);
+	if (parent)
 		rbd_dev_destroy(parent);
-	} else {
-		rbd_put_client(rbdc);
-		rbd_spec_put(parent_spec);
-	}
-
 	return ret;
 }
 

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

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

From: Ilya Dryomov <[email protected]>

commit 6d69bb536bac0d403d83db1ca841444981b280cd upstream.

Mapping an image with a long parent chain (e.g. image foo, whose parent is bar, whose parent is baz, etc) currently leads to a kernel stack overflow, due to the following recursion in the reply path:

rbd_osd_req_callback()

    rbd_obj_request_complete()
      rbd_img_obj_callback()
        rbd_img_parent_read_callback()
          rbd_obj_request_complete()
            ...

Limit the parent chain to 16 images, which is ~5K worth of stack. When the above recursion is eliminated, this limit can be lifted.

---
 drivers/block/rbd.c |   29 +++++++++++++++++++++--------
 1 file changed, 21 insertions(+), 8 deletions(-)

@@ -93,6 +93,8 @@ static int atomic_dec_return_safe(atomic
 
 #define RBD_MINORS_PER_MAJOR	256		/* max minors per blkdev */
 
+#define RBD_MAX_PARENT_CHAIN_LEN	16
+
 #define RBD_SNAP_DEV_NAME_PREFIX	"snap_"
 #define RBD_MAX_SNAP_NAME_LEN	\
 			(NAME_MAX - (sizeof (RBD_SNAP_DEV_NAME_PREFIX) - 1))

@@ -394,7 +396,7 @@ static ssize_t rbd_add(struct bus_type *
 		       size_t count);
 static ssize_t rbd_remove(struct bus_type *bus, const char *buf,
 			  size_t count);
-static int rbd_dev_image_probe(struct rbd_device *rbd_dev, bool mapping);
+static int rbd_dev_image_probe(struct rbd_device *rbd_dev, int depth);
 static void rbd_spec_put(struct rbd_spec *spec);
 
 static struct bus_attribute rbd_bus_attrs[] = {

@@ -4831,7 +4833,12 @@ out_err:
 	return ret;
 }
 
-static int rbd_dev_probe_parent(struct rbd_device *rbd_dev)
+/*
+ * @depth is rbd_dev_image_probe() -> rbd_dev_probe_parent() ->
+ * rbd_dev_image_probe() recursion depth, which means it's also the
+ * length of the already discovered part of the parent chain.
+ */
+static int rbd_dev_probe_parent(struct rbd_device *rbd_dev, int depth)
 {
 	struct rbd_device *parent = NULL;
 	int ret;

@@ -4839,6 +4846,12 @@ static int rbd_dev_probe_parent(struct r
 	if (!rbd_dev->parent_spec)
 		return 0;
 
+	if (++depth > RBD_MAX_PARENT_CHAIN_LEN) {
+		pr_info("parent chain is too long (%d)\n", depth);
+		ret = -EINVAL;
+		goto out_err;
+	}
+
 	parent = rbd_dev_create(rbd_dev->rbd_client, rbd_dev->parent_spec);
 	if (!parent) {
 		ret = -ENOMEM;

@@ -4852,7 +4865,7 @@ static int rbd_dev_probe_parent(struct r
 	__rbd_get_client(rbd_dev->rbd_client);
 	rbd_spec_get(rbd_dev->parent_spec);
 
-	ret = rbd_dev_image_probe(parent, false);
+	ret = rbd_dev_image_probe(parent, depth);
 	if (ret < 0)
 		goto out_err;
 

@@ -4969,7 +4982,7 @@ static void rbd_dev_image_release(struct
  * parent), initiate a watch on its header object before using that
  * object to get detailed information about the rbd image.
  */
-static int rbd_dev_image_probe(struct rbd_device *rbd_dev, bool mapping)
+static int rbd_dev_image_probe(struct rbd_device *rbd_dev, int depth)
 {
 	int ret;
 	int tmp;

@@ -4990,7 +5003,7 @@ static int rbd_dev_image_probe(struct rb
 	if (ret)
 		goto err_out_format;
 
-	if (mapping) {
+	if (!depth) {
 		ret = rbd_dev_header_watch_sync(rbd_dev, true);
 		if (ret)
 			goto out_header_name;

@@ -5007,7 +5020,7 @@ static int rbd_dev_image_probe(struct rb
 	if (ret)
 		goto err_out_probe;
 
-	ret = rbd_dev_probe_parent(rbd_dev);
+	ret = rbd_dev_probe_parent(rbd_dev, depth);
 	if (ret)
 		goto err_out_probe;
 

@@ -5018,7 +5031,7 @@ static int rbd_dev_image_probe(struct rb
 err_out_probe:
 	rbd_dev_unprobe(rbd_dev);
 err_out_watch:
-	if (mapping) {
+	if (!depth) {
 		tmp = rbd_dev_header_watch_sync(rbd_dev, false);
 		if (tmp)
 			rbd_warn(rbd_dev, "unable to tear down "

@@ -5089,7 +5102,7 @@ static ssize_t rbd_add(struct bus_type *
 	rbdc = NULL;		/* rbd_dev now owns this */
 	spec = NULL;		/* rbd_dev now owns this */
 
-	rc = rbd_dev_image_probe(rbd_dev, true);
+	rc = rbd_dev_image_probe(rbd_dev, 0);
 	if (rc < 0)
 		goto err_out_rbd_dev;
 

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

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

From: Will Deacon <[email protected]>

commit 9702970c7bd3e2d6fecb642a190269131d4ac16c upstream.

This reverts commit e306dfd06fcb44d21c80acb8e5a88d55f3d1cf63.

With this patch applied, we were the only architecture making this sort of adjustment to the PC calculation in the unwinder. This causes problems for ftrace, where the PC values are matched against the contents of the stack frames in the callchain and fail to match any records after the address adjustment.

Whilst there has been some effort to change ftrace to workaround this, those patches are not yet ready for mainline and, since we're the odd architecture in this regard, let's just step in line with other architectures (like arch/arm/) for now.

index 407991bf79f5..ccb6078ed9f2 100644
@@ -48,11 +48,7 @@ int notrace unwind_frame(struct stackframe *frame)
 
 	frame->sp = fp + 0x10;
 	frame->fp = *(unsigned long *)(fp);
-	/*
-	 * -4 here because we care about the PC at time of bl,
-	 * not where the return will go.
-	 */
-	frame->pc = *(unsigned long *)(fp + 8) - 4;
+	frame->pc = *(unsigned long *)(fp + 8);
 
 	return 0;
 }

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

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

From: Joe Thornber <[email protected]>

commit 2871c69e025e8bc507651d5a9cf81a8a7da9d24b upstream.

Commit 4c7e309340ff ("dm btree remove: fix bug in redistribute3") wasn't a complete fix for redistribute3().

The redistribute3 function takes 3 btree nodes and shares out the entries evenly between them. If the three nodes in total contained (MAX_ENTRIES * 3) - 1 entries between them then this was erroneously getting rebalanced as (MAX_ENTRIES - 1) on the left and right, and (MAX_ENTRIES + 1) in the center.

Fix this issue by being more careful about calculating the target number of entries for the left and right nodes.

Unit tested in userspace using this program:
https://github.com/jthornber/redistribute3-test/blob/master/redistribute3_t.c

---
 drivers/md/persistent-data/dm-btree-remove.c |   17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

@@ -301,11 +301,16 @@ static void redistribute3(struct dm_btre
 {
 	int s;
 	uint32_t max_entries = le32_to_cpu(left->header.max_entries);
-	unsigned target = (nr_left + nr_center + nr_right) / 3;
-	BUG_ON(target > max_entries);
+	unsigned total = nr_left + nr_center + nr_right;
+	unsigned target_right = total / 3;
+	unsigned remainder = (target_right * 3) != total;
+	unsigned target_left = target_right + remainder;
+
+	BUG_ON(target_left > max_entries);
+	BUG_ON(target_right > max_entries);
 
 	if (nr_left < nr_right) {
-		s = nr_left - target;
+		s = nr_left - target_left;
 
 		if (s < 0 && nr_center < -s) {
 			/* not enough in central node */

@@ -316,10 +321,10 @@ static void redistribute3(struct dm_btre
 		} else
 			shift(left, center, s);
 
-		shift(center, right, target - nr_right);
+		shift(center, right, target_right - nr_right);
 
 	} else {
-		s = target - nr_right;
+		s = target_right - nr_right;
 		if (s > 0 && nr_center < s) {
 			/* not enough in central node */
 			shift(center, right, nr_center);

@@ -329,7 +334,7 @@ static void redistribute3(struct dm_btre
 		} else
 			shift(center, right, s);
 
-		shift(left, center, nr_left - target);
+		shift(left, center, nr_left - target_left);
 	}
 
 	*key_ptr(parent, c->index) = center->keys[0];

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

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

From: Mike Snitzer <[email protected]>

commit 4dcb8b57df3593dcb20481d9d6cf79d1dc1534be upstream.

btree_split_beneath()'s error path had an outstanding FIXME that speaks directly to the potential for _not_ cleaning up a previously allocated bufio-backed block.

Fix this by releasing the previously allocated bufio block using unlock_block().

---
 drivers/md/persistent-data/dm-btree.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -507,7 +507,7 @@ static int btree_split_beneath(struct sh
 
 	r = new_block(s->info, &right);
 	if (r < 0) {
-		/* FIXME: put left */
+		unlock_block(s->info, left);
 		return r;
 	}
 

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

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

From: Mathias Nyman <[email protected]>

commit 3b4739b8951d650becbcd855d7d6f18ac98a9a85 upstream.

If a host fails to wake up a isochronous SuperSpeed device from U1/U2 in time for a isoch transfer it will generate a "No ping response error" Host will then move to the next transfer descriptor.

Handle this case in the same way as missed service errors, tag the current TD as skipped and handle it on the next transfer event.

---
 drivers/usb/host/xhci-ring.c |   20 +++++++++++++++-----
 1 file changed, 15 insertions(+), 5 deletions(-)

@@ -2348,6 +2348,7 @@ static int handle_tx_event(struct xhci_h
 	u32 trb_comp_code;
 	int ret = 0;
 	int td_num = 0;
+	bool handling_skipped_tds = false;
 
 	slot_id = TRB_TO_SLOT_ID(le32_to_cpu(event->flags));
 	xdev = xhci->devs[slot_id];

@@ -2481,6 +2482,10 @@ static int handle_tx_event(struct xhci_h
 		ep->skip = true;
 		xhci_dbg(xhci, "Miss service interval error, set skip flag\n");
 		goto cleanup;
+	case COMP_PING_ERR:
+		ep->skip = true;
+		xhci_dbg(xhci, "No Ping response error, Skip one Isoc TD\n");
+		goto cleanup;
 	default:
 		if (xhci_is_vendor_info_code(xhci, trb_comp_code)) {
 			status = 0;

@@ -2612,13 +2617,18 @@ static int handle_tx_event(struct xhci_h
 						 ep, &status);
 
 cleanup:
+
+
+		handling_skipped_tds = ep->skip &&
+			trb_comp_code != COMP_MISSED_INT &&
+			trb_comp_code != COMP_PING_ERR;
+
 		/*
-		 * Do not update event ring dequeue pointer if ep->skip is set.
-		 * Will roll back to continue process missed tds.
+		 * Do not update event ring dequeue pointer if we're in a loop
+		 * processing missed tds.
 		 */
-		if (trb_comp_code == COMP_MISSED_INT || !ep->skip) {
+		if (!handling_skipped_tds)
 			inc_deq(xhci, xhci->event_ring);
-		}
 
 		if (ret) {
 			urb = td->urb;

@@ -2662,7 +2672,7 @@ cleanup:
 	 * Process them as short transfer until reach the td pointed by
 	 * the event.
 	 */
-	} while (ep->skip && trb_comp_code != COMP_MISSED_INT);
+	} while (handling_skipped_tds);
 
 	return 0;
 }

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

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

From: Cathy Avery <[email protected]>

commit a54c8f0f2d7df525ff997e2afe71866a1a013064 upstream.

xen-blkfront will crash if the check to talk_to_blkback() in blkback_changed()(XenbusStateInitWait) returns an error. The driver data is freed and info is set to NULL. Later during the close process via talk_to_blkback's call to xenbus_dev_fatal() the null pointer is passed to and dereference in blkfront_closing.

---
 drivers/block/xen-blkfront.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -1590,7 +1590,8 @@ static void blkback_changed(struct xenbu
 			break;
 		/* Missed the backend's Closing state -- fallthrough */
 	case XenbusStateClosing:
-		blkfront_closing(info);
+		if (info)
+			blkfront_closing(info);
 		break;
 	}
 }

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

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

From: Peter Zijlstra <[email protected]>

commit 275d7d44d802ef271a42dc87ac091a495ba72fc5 upstream.

Poma (on the way to another bug) reported an assertion triggering:

  [<ffffffff81150529>] module_assert_mutex_or_preempt+0x49/0x90
  [<ffffffff81150822>] __module_address+0x32/0x150
  [<ffffffff81150956>] __module_text_address+0x16/0x70
  [<ffffffff81150f19>] symbol_put_addr+0x29/0x40
  [<ffffffffa04b77ad>] dvb_frontend_detach+0x7d/0x90 [dvb_core]

Laura Abbott <[email protected]> produced a patch which lead us to inspect symbol_put_addr(). This function has a comment claiming it doesn't need to disable preemption around the module lookup because it holds a reference to the module it wants to find, which therefore cannot go away.

This is wrong (and a false optimization too, preempt_disable() is really rather cheap, and I doubt any of this is on uber critical paths, otherwise it would've retained a pointer to the actual module anyway and avoided the second lookup).

While its true that the module cannot go away while we hold a reference on it, the data structure we do the lookup in very much _CAN_ change while we do the lookup. Therefore fix the comment and add the required preempt_disable().

---
 kernel/module.c |    8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

@@ -942,11 +942,15 @@ void symbol_put_addr(void *addr)
 	if (core_kernel_text(a))
 		return;
 
-	/* module_text_address is safe here: we're supposed to have reference
-	 * to module from symbol_get, so it can't go away. */
+	/*
+	 * Even though we hold a reference on the module; we still need to
+	 * disable preemption in order to safely traverse the data structure.
+	 */
+	preempt_disable();
 	modaddr = __module_text_address(a);
 	BUG_ON(!modaddr);
 	module_put(modaddr);
+	preempt_enable();
 }
 EXPORT_SYMBOL_GPL(symbol_put_addr);
 

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

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

From: Herbert Xu <[email protected]>

commit 3fc89adb9fa4beff31374a4bf50b3d099d88ae83 upstream.

Currently a number of Crypto API operations may fail when a signal occurs. This causes nasty problems as the caller of those operations are often not in a good position to restart the operation.

In fact there is currently no need for those operations to be interrupted by user signals at all. All we need is for them to be killable.

This patch replaces the relevant calls of signal_pending with fatal_signal_pending, and wait_for_completion_interruptible with wait_for_completion_killable, respectively.

---
 crypto/ablkcipher.c  |    2 +-
 crypto/algapi.c      |    2 +-
 crypto/api.c         |    6 +++---
 crypto/crypto_user.c |    2 +-
 4 files changed, 6 insertions(+), 6 deletions(-)

@@ -700,7 +700,7 @@ struct crypto_ablkcipher *crypto_alloc_a
 err:
 		if (err != -EAGAIN)
 			break;
-		if (signal_pending(current)) {
+		if (fatal_signal_pending(current)) {
 			err = -EINTR;
 			break;
 		}

@@ -325,7 +325,7 @@ static void crypto_wait_for_test(struct
 		crypto_alg_tested(larval->alg.cra_driver_name, 0);
 	}
 
-	err = wait_for_completion_interruptible(&larval->completion);
+	err = wait_for_completion_killable(&larval->completion);
 	WARN_ON(err);
 
 out:

@@ -172,7 +172,7 @@ static struct crypto_alg *crypto_larval_
 	struct crypto_larval *larval = (void *)alg;
 	long timeout;
 
-	timeout = wait_for_completion_interruptible_timeout(
+	timeout = wait_for_completion_killable_timeout(
 		&larval->completion, 60 * HZ);
 
 	alg = larval->adult;

@@ -435,7 +435,7 @@ struct crypto_tfm *crypto_alloc_base(con
 err:
 		if (err != -EAGAIN)
 			break;
-		if (signal_pending(current)) {
+		if (fatal_signal_pending(current)) {
 			err = -EINTR;
 			break;
 		}

@@ -552,7 +552,7 @@ void *crypto_alloc_tfm(const char *alg_n
 err:
 		if (err != -EAGAIN)
 			break;
-		if (signal_pending(current)) {
+		if (fatal_signal_pending(current)) {
 			err = -EINTR;
 			break;
 		}

@@ -361,7 +361,7 @@ static struct crypto_alg *crypto_user_ae
 		err = PTR_ERR(alg);
 		if (err != -EAGAIN)
 			break;
-		if (signal_pending(current)) {
+		if (fatal_signal_pending(current)) {
 			err = -EINTR;
 			break;
 		}

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

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

From: Jes Sorensen <[email protected]>

commit 203d27b0226a05202438ddb39ef0ef1acb14a759 upstream.

This was introduced with 9e882242c6193ae6f416f2d8d8db0d9126bd996b which changed the return value of submit_bio_wait() to return != 0 on error, but didn't update the caller accordingly.

---
 drivers/md/raid1.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -2147,7 +2147,7 @@ static int narrow_write_error(struct r1b
 		md_trim_bio(wbio, sector - r1_bio->sector, sectors);
 		wbio->bi_sector += rdev->data_offset;
 		wbio->bi_bdev = rdev->bdev;
-		if (submit_bio_wait(WRITE, wbio) == 0)
+		if (submit_bio_wait(WRITE, wbio) < 0)
 			/* failure! */
 			ok = rdev_set_badblocks(rdev, sector,
 						sectors, 0)

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

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

From: Jes Sorensen <[email protected]>

commit 681ab4696062f5aa939c9e04d058732306a97176 upstream.

This was introduced with 9e882242c6193ae6f416f2d8d8db0d9126bd996b which changed the return value of submit_bio_wait() to return != 0 on error, but didn't update the caller accordingly.

---
 drivers/md/raid10.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -2597,7 +2597,7 @@ static int narrow_write_error(struct r10
 				   choose_data_offset(r10_bio, rdev) +
 				   (sector - r10_bio->sector));
 		wbio->bi_bdev = rdev->bdev;
-		if (submit_bio_wait(WRITE, wbio) == 0)
+		if (submit_bio_wait(WRITE, wbio) < 0)
 			/* Failure! */
 			ok = rdev_set_badblocks(rdev, sector,
 						sectors, 0)

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

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

From: Doron Tsur <[email protected]>

commit 0ca81a2840f77855bbad1b9f172c545c4dc9e6a4 upstream.

ib_send_cm_sidr_rep could sometimes erase the node from the sidr (depending on errors in the process). Since ib_send_cm_sidr_rep is called both from cm_sidr_req_handler and cm_destroy_id, cm_id_priv could be either erased from the rb_tree twice or not erased at all. Fixing that by making sure it's erased only once before freeing cm_id_priv.

---
 drivers/infiniband/core/cm.c |   10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

@@ -857,6 +857,11 @@ retest:
 	case IB_CM_SIDR_REQ_RCVD:
 		spin_unlock_irq(&cm_id_priv->lock);
 		cm_reject_sidr_req(cm_id_priv, IB_SIDR_REJECT);
+		spin_lock_irq(&cm.lock);
+		if (!RB_EMPTY_NODE(&cm_id_priv->sidr_id_node))
+			rb_erase(&cm_id_priv->sidr_id_node,
+				 &cm.remote_sidr_table);
+		spin_unlock_irq(&cm.lock);
 		break;
 	case IB_CM_REQ_SENT:
 		ib_cancel_mad(cm_id_priv->av.port->mad_agent, cm_id_priv->msg);

@@ -3093,7 +3098,10 @@ int ib_send_cm_sidr_rep(struct ib_cm_id
 	spin_unlock_irqrestore(&cm_id_priv->lock, flags);
 
 	spin_lock_irqsave(&cm.lock, flags);
-	rb_erase(&cm_id_priv->sidr_id_node, &cm.remote_sidr_table);
+	if (!RB_EMPTY_NODE(&cm_id_priv->sidr_id_node)) {
+		rb_erase(&cm_id_priv->sidr_id_node, &cm.remote_sidr_table);
+		RB_CLEAR_NODE(&cm_id_priv->sidr_id_node);
+	}
 	spin_unlock_irqrestore(&cm.lock, flags);
 	return 0;
 

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

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

Fixes the backport of 0b34a166f291d255755be46e43ed5497cdd194f2 upstream

Commit 0b34a166f291d255755be46e43ed5497cdd194f2 "x86/xen: Support kexec/kdump in HVM guests by doing a soft reset" has been added to the 4.2-stable tree" needed to correct the CONFIG variable, as CONFIG_KEXEC_CORE only showed up in 4.3.

---
 arch/x86/xen/enlighten.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

@@ -33,7 +33,7 @@
 #include <linux/memblock.h>
 #include <linux/edd.h>
 
-#ifdef CONFIG_KEXEC_CORE
+#ifdef CONFIG_KEXEC
 #include <linux/kexec.h>
 #endif
 

@@ -1748,7 +1748,7 @@ static struct notifier_block xen_hvm_cpu
 	.notifier_call	= xen_hvm_cpu_notify,
 };
 
-#ifdef CONFIG_KEXEC_CORE
+#ifdef CONFIG_KEXEC
 static void xen_hvm_shutdown(void)
 {
 	native_machine_shutdown();

@@ -1777,7 +1777,7 @@ static void __init xen_hvm_guest_init(vo
 	x86_init.irqs.intr_init = xen_init_IRQ;
 	xen_hvm_init_time_ops();
 	xen_hvm_init_mmu_ops();
-#ifdef CONFIG_KEXEC_CORE
+#ifdef CONFIG_KEXEC
 	machine_ops.shutdown = xen_hvm_shutdown;
 	machine_ops.crash_shutdown = xen_hvm_crash_shutdown;
 #endif

Build results:

	total: 121 pass: 121 fail: 0
Qemu test results:
	total: 71 pass: 71 fail: 0

Details are available at http://server.roeck-us.net:8010/builders.

Guenter

Compiled and booted on my test system. No dmesg regressions.

thanks,
-- Shuah

Shuah Khan
Sr. Linux Kernel Developer
Open Source Innovation Group
Samsung Research America (Silicon Valley)
[email protected] | (970) 217-8978