Skip to content

L
linux
Switch branch/tag
  1. 25 Jan, 2016 33 commits
  3. 18 Jan, 2016 7 commits
    • Naoya Horiguchi's avatar
      mm: hugetlb: fix hugepage memory leak caused by wrong reserve count · 60d786c5
      Naoya Horiguchi authored 
      commit a88c7695 upstream.

When dequeue_huge_page_vma() in alloc_huge_page() fails, we fall back on
alloc_buddy_huge_page() to directly create a hugepage from the buddy
allocator.

In that case, however, if alloc_buddy_huge_page() succeeds we don't
decrement h->resv_huge_pages, which means that successful
hugetlb_fault() returns without releasing the reserve count.  As a
result, subsequent hugetlb_fault() might fail despite that there are
still free hugepages.

This patch simply adds decrementing code on that code path.

I reproduced this problem when testing v4.3 kernel in the following situation:
 - the test machine/VM is a NUMA system,
 - hugepage overcommiting is enabled,
 - most of hugepages are allocated and there's only one free hugepage
   which is on node 0 (for example),
 - another program, which calls set_mempolicy(MPOL_BIND) to bind itself to
   node 1, tries to allocate a hugepage,
 - the allocation should fail but the reserve count is still hold.
Signed-off-by: default avatarNaoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Hillf Danton <hillf.zj@alibaba-inc.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
[ luis: backported to 3.16:
  - use 'chg' instead of 'gbl_chg'
  - adjusted context ]
Signed-off-by: default avatarLuis Henriques <luis.henriques@canonical.com>
      60d786c5
    • Mikulas Patocka's avatar
      parisc iommu: fix panic due to trying to allocate too large region · 03dd7ad3
      Mikulas Patocka authored 
      commit e46e31a3 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).
Signed-off-by: default avatarMikulas Patocka <mpatocka@redhat.com>
Signed-off-by: default avatarHelge Deller <deller@gmx.de>
Signed-off-by: default avatarLuis Henriques <luis.henriques@canonical.com>
      03dd7ad3
    • Alan Stern's avatar
      USB: add quirk for devices with broken LPM · 23b42ce5
      Alan Stern authored 
      commit ad87e032 upstream.

Some USB device / host controller combinations seem to have problems
with Link Power Management.  For example, Steinar found that his xHCI
controller wouldn't handle bandwidth calculations correctly for two
video cards simultaneously when LPM was enabled, even though the bus
had plenty of bandwidth available.

This patch introduces a new quirk flag for devices that should remain
disabled for LPM, and creates quirk entries for Steinar's devices.
Signed-off-by: default avatarAlan Stern <stern@rowland.harvard.edu>
Reported-by: default avatarSteinar H. Gunderson <sgunderson@bigfoot.com>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@linuxfoundation.org>
[ luis: backported to 3.16: adjusted context ]
Signed-off-by: default avatarLuis Henriques <luis.henriques@canonical.com>
      23b42ce5
    • Mathias Nyman's avatar
      xhci: fix usb2 resume timing and races. · 30f8b511
      Mathias Nyman authored 
      commit f69115fd upstream.

According to USB 2 specs ports need to signal resume for at least 20ms,
in practice even longer, before moving to U0 state.
Both host and devices can initiate resume.

On device initiated resume, a port status interrupt with the port in resume
state in issued. The interrupt handler tags a resume_done[port]
timestamp with current time + USB_RESUME_TIMEOUT, and kick roothub timer.
Root hub timer requests for port status, finds the port in resume state,
checks if resume_done[port] timestamp passed, and set port to U0 state.

On host initiated resume, current code sets the port to resume state,
sleep 20ms, and finally sets the port to U0 state. This should also
be changed to work in a similar way as the device initiated resume, with
timestamp tagging, but that is not yet tested and will be a separate
fix later.

There are a few issues with this approach

1. A host initiated resume will also generate a resume event. The event
   handler will find the port in resume state, believe it's a device
   initiated resume, and act accordingly.

2. A port status request might cut the resume signalling short if a
   get_port_status request is handled during the host resume signalling.
   The port will be found in resume state. The timestamp is not set leading
   to time_after_eq(jiffies, timestamp) returning true, as timestamp = 0.
   get_port_status will proceed with moving the port to U0.

3. If an error, or anything else happens to the port during device
   initiated resume signalling it will leave all the device resume
   parameters hanging uncleared, preventing further suspend, returning
   -EBUSY, and cause the pm thread to busyloop trying to enter suspend.

Fix this by using the existing resuming_ports bitfield to indicate that
resume signalling timing is taken care of.
Check if the resume_done[port] is set before using it for timestamp
comparison, and also clear out any resume signalling related variables
if port is not in U0 or Resume state

This issue was discovered when a PM thread busylooped, trying to runtime
suspend the xhci USB 2 roothub on a Dell XPS
Reported-by: default avatarDaniel J Blueman <daniel@quora.org>
Tested-by: default avatarDaniel J Blueman <daniel@quora.org>
Signed-off-by: default avatarMathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: default avatarLuis Henriques <luis.henriques@canonical.com>
      30f8b511
    • Kirill A. Shutemov's avatar
      vgaarb: fix signal handling in vga_get() · ccebddfb
      Kirill A. Shutemov authored 
      commit 9f5bd308 upstream.

There are few defects in vga_get() related to signal hadning:

  - we shouldn't check for pending signals for TASK_UNINTERRUPTIBLE
    case;

  - if we found pending signal we must remove ourself from wait queue
    and change task state back to running;

  - -ERESTARTSYS is more appropriate, I guess.
Signed-off-by: default avatarKirill A. Shutemov <kirill@shutemov.name>
Reviewed-by: default avatarDavid Herrmann <dh.herrmann@gmail.com>
Signed-off-by: default avatarDave Airlie <airlied@redhat.com>
Signed-off-by: default avatarLuis Henriques <luis.henriques@canonical.com>
      ccebddfb
    • Joe Thornber's avatar
      dm btree: fix bufio buffer leaks in dm_btree_del() error path · a94064d4
      Joe Thornber authored 
      commit ed8b45a3 upstream.

If dm_btree_del()'s call to push_frame() fails, e.g. due to
btree_node_validator finding invalid metadata, the dm_btree_del() error
path must unlock all frames (which have active dm-bufio buffers) that
were pushed onto the del_stack.

Otherwise, dm_bufio_client_destroy() will BUG_ON() because dm-bufio
buffers have leaked, e.g.:
  device-mapper: bufio: leaked buffer 3, hold count 1, list 0
Signed-off-by: default avatarJoe Thornber <ejt@redhat.com>
Signed-off-by: default avatarMike Snitzer <snitzer@redhat.com>
Signed-off-by: default avatarLuis Henriques <luis.henriques@canonical.com>
      a94064d4
    • Jan Stancek's avatar
      ipmi: move timer init to before irq is setup · 35f6d738
      Jan Stancek authored 
      commit 27f972d3 upstream.

We encountered a panic on boot in ipmi_si on a dell per320 due to an
uninitialized timer as follows.

static int smi_start_processing(void       *send_info,
                                ipmi_smi_t intf)
{
        /* Try to claim any interrupts. */
        if (new_smi->irq_setup)
                new_smi->irq_setup(new_smi);

 --> IRQ arrives here and irq handler tries to modify uninitialized timer

    which triggers BUG_ON(!timer->function) in __mod_timer().

 Call Trace:
   <IRQ>
   [<ffffffffa0532617>] start_new_msg+0x47/0x80 [ipmi_si]
   [<ffffffffa053269e>] start_check_enables+0x4e/0x60 [ipmi_si]
   [<ffffffffa0532bd8>] smi_event_handler+0x1e8/0x640 [ipmi_si]
   [<ffffffff810f5584>] ? __rcu_process_callbacks+0x54/0x350
   [<ffffffffa053327c>] si_irq_handler+0x3c/0x60 [ipmi_si]
   [<ffffffff810efaf0>] handle_IRQ_event+0x60/0x170
   [<ffffffff810f245e>] handle_edge_irq+0xde/0x180
   [<ffffffff8100fc59>] handle_irq+0x49/0xa0
   [<ffffffff8154643c>] do_IRQ+0x6c/0xf0
   [<ffffffff8100ba53>] ret_from_intr+0x0/0x11

        /* Set up the timer that drives the interface. */
        setup_timer(&new_smi->si_timer, smi_timeout, (long)new_smi);

The following patch fixes the problem.

To: Openipmi-developer@lists.sourceforge.net
To: Corey Minyard <minyard@acm.org>
CC: linux-kernel@vger.kernel.org
Signed-off-by: default avatarJan Stancek <jstancek@redhat.com>
Signed-off-by: default avatarTony Camuso <tcamuso@redhat.com>
Signed-off-by: default avatarCorey Minyard <cminyard@mvista.com>
Signed-off-by: default avatarLuis Henriques <luis.henriques@canonical.com>
      35f6d738
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7