David Ahern [Thu, 28 Mar 2019 03:53:50 +0000 (20:53 -0700)]
ipv6: Create init helper for fib6_nh
Similar to IPv4, consolidate the fib6_nh initialization into a helper.
As a new standalone function, add a cleanup path to put lwtstate on
error.
To avoid modifying fib6_config flags, move the reject check to a helper
that is invoked once by fib6_nh_init to reset the device and then
again in ip6_route_info_create to set the fib6_flags.
David Ahern [Thu, 28 Mar 2019 03:53:49 +0000 (20:53 -0700)]
ipv4: Create cleanup helper for fib_nh
Move the fib_nh cleanup code from free_fib_info_rcu into a new helper,
fib_nh_release. Move classid accounting into fib_nh_release which is
called per fib_nh to make accounting symmetrical with fib_nh_init.
Export the helper to allow for use with nexthop objects in the
future.
David Ahern [Thu, 28 Mar 2019 03:53:48 +0000 (20:53 -0700)]
ipv4: Create init helper for fib_nh
Consolidate the fib_nh initialization which is duplicated between
fib_create_info for single path and fib_get_nhs for multipath.
Export the helper to allow for use with nexthop objects in the
future.
David Ahern [Thu, 28 Mar 2019 03:53:47 +0000 (20:53 -0700)]
ipv4: Move IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN to helper
in_dev lookup followed by IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN check
is called in several places, some with the rcu lock and others with the
rtnl held.
Move the check to a helper similar to what IPv6 has. Since the helper
can be invoked from either context use rcu_dereference_rtnl to
dereference ip_ptr.
David Ahern [Thu, 28 Mar 2019 03:53:46 +0000 (20:53 -0700)]
ipv4: Define fib_get_nhs when CONFIG_IP_ROUTE_MULTIPATH is disabled
Define fib_get_nhs to return EINVAL when CONFIG_IP_ROUTE_MULTIPATH is
not enabled and remove the ifdef check for CONFIG_IP_ROUTE_MULTIPATH
in fib_create_info.
A new dir entry can be created in get_subdir and its 'header->parent' is
set to NULL. Only after insert_header success, it will be set to 'dir',
otherwise 'header->parent' is set to NULL and drop_sysctl_table is called.
However in err handling path of get_subdir, drop_sysctl_table also be
called on 'new->header' regardless its value of parent pointer. Then
put_links is called, which triggers NULL-ptr deref when access member of
header->parent.
In fact we have multiple error paths which call drop_sysctl_table() there,
upon failure on insert_links() we also call drop_sysctl_table().And even
in the successful case on __register_sysctl_table() we still always call
drop_sysctl_table().This patch fix it.
Lars Persson [Fri, 29 Mar 2019 03:44:28 +0000 (20:44 -0700)]
mm/migrate.c: add missing flush_dcache_page for non-mapped page migrate
Our MIPS 1004Kc SoCs were seeing random userspace crashes with SIGILL
and SIGSEGV that could not be traced back to a userspace code bug. They
had all the magic signs of an I/D cache coherency issue.
Now recently we noticed that the /proc/sys/vm/compact_memory interface
was quite efficient at provoking this class of userspace crashes.
Studying the code in mm/migrate.c there is a distinction made between
migrating a page that is mapped at the instant of migration and one that
is not mapped. Our problem turned out to be the non-mapped pages.
For the non-mapped page the code performs a copy of the page content and
all relevant meta-data of the page without doing the required D-cache
maintenance. This leaves dirty data in the D-cache of the CPU and on
the 1004K cores this data is not visible to the I-cache. A subsequent
page-fault that triggers a mapping of the page will happily serve the
process with potentially stale code.
What about ARM then, this bug should have seen greater exposure? Well
ARM became immune to this flaw back in 2010, see commit c01778001a4f
("ARM: 6379/1: Assume new page cache pages have dirty D-cache").
My proposed fix moves the D-cache maintenance inside move_to_new_page to
make it common for both cases.
Makoto report a below KASAN error: zram does out-of-bounds read. Because
strscpy copies from source up to count bytes unconditionally. It could
cause out-of-bounds read on next object in slab.
To prevent it, use strlcpy which checks source's length automatically.
BUG: KASAN: slab-out-of-bounds in strscpy+0x68/0x154
Read of size 8 at addr ffffffc0c3495a00 by task system_server/1314
..
Call trace:
strscpy+0x68/0x154
idle_store+0xc4/0x34c
dev_attr_store+0x50/0x6c
sysfs_kf_write+0x98/0xb4
kernfs_fop_write+0x198/0x260
__vfs_write+0x10c/0x338
vfs_write+0x114/0x238
SyS_write+0xc8/0x168
__sys_trace_return+0x0/0x4
Allocated by task 1314:
__kmalloc+0x280/0x318
kernfs_fop_write+0xac/0x260
__vfs_write+0x10c/0x338
vfs_write+0x114/0x238
SyS_write+0xc8/0x168
__sys_trace_return+0x0/0x4
The buggy address belongs to the object at ffffffc0c3495a00
which belongs to the cache kmalloc-128 of size 128
The buggy address is located 0 bytes inside of
128-byte region [ffffffc0c3495a00, ffffffc0c3495a80)
The buggy address belongs to the page:
page:ffffffbf030d2500 count:1 mapcount:0 mapping: (null) index:0x0 compound_mapcount: 0
flags: 0x4000000000010200(slab|head)
page dumped because: kasan: bad access detected
Memory state around the buggy address: ffffffc0c3495900: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ffffffc0c3495980: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
>ffffffc0c3495a00: 04 fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
^ ffffffc0c3495a80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc ffffffc0c3495b00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
Qian Cai [Fri, 29 Mar 2019 03:44:21 +0000 (20:44 -0700)]
mm/page_isolation.c: fix a wrong flag in set_migratetype_isolate()
Due to has_unmovable_pages() taking an incorrect irqsave flag instead of
the isolation flag in set_migratetype_isolate(), there are issues with
HWPOSION and error reporting where dump_page() is not called when there
is an unmovable page.
Qian Cai [Fri, 29 Mar 2019 03:44:16 +0000 (20:44 -0700)]
mm/memory_hotplug.c: fix notification in offline error path
When start_isolate_page_range() returned -EBUSY in __offline_pages(), it
calls memory_notify(MEM_CANCEL_OFFLINE, &arg) with an uninitialized
"arg". As the result, it triggers warnings below. Also, it is only
necessary to notify MEM_CANCEL_OFFLINE after MEM_GOING_OFFLINE.
Andrei Vagin [Fri, 29 Mar 2019 03:44:13 +0000 (20:44 -0700)]
ptrace: take into account saved_sigmask in PTRACE{GET,SET}SIGMASK
There are a few system calls (pselect, ppoll, etc) which replace a task
sigmask while they are running in a kernel-space
When a task calls one of these syscalls, the kernel saves a current
sigmask in task->saved_sigmask and sets a syscall sigmask.
On syscall-exit-stop, ptrace traps a task before restoring the
saved_sigmask, so PTRACE_GETSIGMASK returns the syscall sigmask and
PTRACE_SETSIGMASK does nothing, because its sigmask is replaced by
saved_sigmask, when the task returns to user-space.
This patch fixes this problem. PTRACE_GETSIGMASK returns saved_sigmask
if it's set. PTRACE_SETSIGMASK drops the TIF_RESTORE_SIGMASK flag.
Oscar Salvador [Fri, 29 Mar 2019 03:44:01 +0000 (20:44 -0700)]
mm/debug.c: fix __dump_page when mapping->host is not set
While debugging something, I added a dump_page() into do_swap_page(),
and I got the splat from below. The issue happens when dereferencing
mapping->host in __dump_page():
Swap address space does not contain an inode information, and so
mapping->host equals NULL.
Although the dump_page() call was added artificially into
do_swap_page(), I am not sure if we can hit this from any other path, so
it looks worth fixing it. We can easily do that by checking
mapping->host first.
Yang Shi [Fri, 29 Mar 2019 03:43:55 +0000 (20:43 -0700)]
mm: mempolicy: make mbind() return -EIO when MPOL_MF_STRICT is specified
When MPOL_MF_STRICT was specified and an existing page was already on a
node that does not follow the policy, mbind() should return -EIO. But
commit 6f4576e3687b ("mempolicy: apply page table walker on
queue_pages_range()") broke the rule.
And commit c8633798497c ("mm: mempolicy: mbind and migrate_pages support
thp migration") didn't return the correct value for THP mbind() too.
If MPOL_MF_STRICT is set, ignore vma_migratable() to make sure it
reaches queue_pages_to_pte_range() or queue_pages_pmd() to check if an
existing page was already on a node that does not follow the policy.
And, non-migratable vma may be used, return -EIO too if MPOL_MF_MOVE or
MPOL_MF_MOVE_ALL was specified.
Tested with https://github.com/metan-ucw/ltp/blob/master/testcases/kernel/syscalls/mbind/mbind02.c
>> mm/memory.c:3968:21: sparse: incorrect type in assignment (different
>> base types) @@ expected restricted vm_fault_t [usertype] ret @@
>> got e] ret @@
mm/memory.c:3968:21: expected restricted vm_fault_t [usertype] ret
mm/memory.c:3968:21: got int
This patch converts to return vm_fault_t type for hugetlb_fault() when
CONFIG_HUGETLB_PAGE=n.
Regarding the sparse warning, Luc said:
: This is the expected behaviour. The constant 0 is magic regarding bitwise
: types but ({ ...; 0; }) is not, it is just an ordinary expression of type
: 'int'.
:
: So, IMHO, Souptick's patch is the right thing to do.
Nicolas Boichat [Fri, 29 Mar 2019 03:43:46 +0000 (20:43 -0700)]
iommu/io-pgtable-arm-v7s: request DMA32 memory, and improve debugging
IOMMUs using ARMv7 short-descriptor format require page tables (level 1
and 2) to be allocated within the first 4GB of RAM, even on 64-bit
systems.
For level 1/2 pages, ensure GFP_DMA32 is used if CONFIG_ZONE_DMA32 is
defined (e.g. on arm64 platforms).
For level 2 pages, allocate a slab cache in SLAB_CACHE_DMA32. Note that
we do not explicitly pass GFP_DMA[32] to kmem_cache_zalloc, as this is
not strictly necessary, and would cause a warning in mm/sl*b.c, as we
did not update GFP_SLAB_BUG_MASK.
Also, print an error when the physical address does not fit in
32-bit, to make debugging easier in the future.
Nicolas Boichat [Fri, 29 Mar 2019 03:43:42 +0000 (20:43 -0700)]
mm: add support for kmem caches in DMA32 zone
Patch series "iommu/io-pgtable-arm-v7s: Use DMA32 zone for page tables",
v6.
This is a followup to the discussion in [1], [2].
IOMMUs using ARMv7 short-descriptor format require page tables (level 1
and 2) to be allocated within the first 4GB of RAM, even on 64-bit
systems.
For L1 tables that are bigger than a page, we can just use
__get_free_pages with GFP_DMA32 (on arm64 systems only, arm would still
use GFP_DMA).
For L2 tables that only take 1KB, it would be a waste to allocate a full
page, so we considered 3 approaches:
1. This series, adding support for GFP_DMA32 slab caches.
2. genalloc, which requires pre-allocating the maximum number of L2 page
tables (4096, so 4MB of memory).
3. page_frag, which is not very memory-efficient as it is unable to reuse
freed fragments until the whole page is freed. [3]
This series is the most memory-efficient approach.
stable@ note:
We confirmed that this is a regression, and IOMMU errors happen on 4.19
and linux-next/master on MT8173 (elm, Acer Chromebook R13). The issue
most likely starts from commit ad67f5a6545f ("arm64: replace ZONE_DMA
with ZONE_DMA32"), i.e. 4.15, and presumably breaks a number of Mediatek
platforms (and maybe others?).
IOMMUs using ARMv7 short-descriptor format require page tables to be
allocated within the first 4GB of RAM, even on 64-bit systems. On arm64,
this is done by passing GFP_DMA32 flag to memory allocation functions.
For IOMMU L2 tables that only take 1KB, it would be a waste to allocate
a full page using get_free_pages, so we considered 3 approaches:
1. This patch, adding support for GFP_DMA32 slab caches.
2. genalloc, which requires pre-allocating the maximum number of L2
page tables (4096, so 4MB of memory).
3. page_frag, which is not very memory-efficient as it is unable
to reuse freed fragments until the whole page is freed.
This change makes it possible to create a custom cache in DMA32 zone using
kmem_cache_create, then allocate memory using kmem_cache_alloc.
We do not create a DMA32 kmalloc cache array, as there are currently no
users of kmalloc(..., GFP_DMA32). These calls will continue to trigger a
warning, as we keep GFP_DMA32 in GFP_SLAB_BUG_MASK.
This implies that calls to kmem_cache_*alloc on a SLAB_CACHE_DMA32
kmem_cache must _not_ use GFP_DMA32 (it is anyway redundant and
unnecessary).
Darrick J. Wong [Fri, 29 Mar 2019 03:43:38 +0000 (20:43 -0700)]
ocfs2: fix inode bh swapping mixup in ocfs2_reflink_inodes_lock
ocfs2_reflink_inodes_lock() can swap the inode1/inode2 variables so that
we always grab cluster locks in order of increasing inode number.
Unfortunately, we forget to swap the inode record buffer head pointers
when we've done this, which leads to incorrect bookkeepping when we're
trying to make the two inodes have the same refcount tree.
This has the effect of causing filesystem shutdowns if you're trying to
reflink data from inode 100 into inode 97, where inode 100 already has a
refcount tree attached and inode 97 doesn't. The reflink code decides
to copy the refcount tree pointer from 100 to 97, but uses inode 97's
inode record to open the tree root (which it doesn't have) and blows up.
This issue causes filesystem shutdowns and metadata corruption!
Qian Cai [Fri, 29 Mar 2019 03:43:34 +0000 (20:43 -0700)]
mm/hotplug: fix offline undo_isolate_page_range()
Commit f1dd2cd13c4b ("mm, memory_hotplug: do not associate hotadded
memory to zones until online") introduced move_pfn_range_to_zone() which
calls memmap_init_zone() during onlining a memory block.
memmap_init_zone() will reset pagetype flags and makes migrate type to
be MOVABLE.
However, in __offline_pages(), it also call undo_isolate_page_range()
after offline_isolated_pages() to do the same thing. Due to commit 2ce13640b3f4 ("mm: __first_valid_page skip over offline pages") changed
__first_valid_page() to skip offline pages, undo_isolate_page_range()
here just waste CPU cycles looping around the offlining PFN range while
doing nothing, because __first_valid_page() will return NULL as
offline_isolated_pages() has already marked all memory sections within
the pfn range as offline via offline_mem_sections().
Also, after calling the "useless" undo_isolate_page_range() here, it
reaches the point of no returning by notifying MEM_OFFLINE. Those pages
will be marked as MIGRATE_MOVABLE again once onlining. The only thing
left to do is to decrease the number of isolated pageblocks zone counter
which would make some paths of the page allocation slower that the above
commit introduced.
Even if alloc_contig_range() can be used to isolate 16GB-hugetlb pages
on ppc64, an "int" should still be enough to represent the number of
pageblocks there. Fix an incorrect comment along the way.
Tetsuo Handa [Fri, 29 Mar 2019 03:43:30 +0000 (20:43 -0700)]
fs/open.c: allow opening only regular files during execve()
syzbot is hitting lockdep warning [1] due to trying to open a fifo
during an execve() operation. But we don't need to open non regular
files during an execve() operation, for all files which we will need are
the executable file itself and the interpreter programs like /bin/sh and
ld-linux.so.2 .
Since the manpage for execve(2) says that execve() returns EACCES when
the file or a script interpreter is not a regular file, and the manpage
for uselib(2) says that uselib() can return EACCES, and we use
FMODE_EXEC when opening for execve()/uselib(), we can bail out if a non
regular file is requested with FMODE_EXEC set.
Since this deadlock followed by khungtaskd warnings is trivially
reproducible by a local unprivileged user, and syzbot's frequent crash
due to this deadlock defers finding other bugs, let's workaround this
deadlock until we get a chance to find a better solution.
Qian Cai [Fri, 29 Mar 2019 03:43:23 +0000 (20:43 -0700)]
mm/debug.c: add a cast to u64 for atomic64_read()
atomic64_read() on ppc64le returns "long int", so fix the same way as
commit d549f545e690 ("drm/virtio: use %llu format string form
atomic64_t") by adding a cast to u64, which makes it work on all arches.
In file included from ./include/linux/printk.h:7,
from ./include/linux/kernel.h:15,
from mm/debug.c:9:
mm/debug.c: In function 'dump_mm':
./include/linux/kern_levels.h:5:18: warning: format '%llx' expects argument of type 'long long unsigned int', but argument 19 has type 'long int' [-Wformat=]
#define KERN_SOH "A" /* ASCII Start Of Header */
^~~~~~
./include/linux/kern_levels.h:8:20: note: in expansion of macro
'KERN_SOH'
#define KERN_EMERG KERN_SOH "0" /* system is unusable */
^~~~~~~~
./include/linux/printk.h:297:9: note: in expansion of macro 'KERN_EMERG'
printk(KERN_EMERG pr_fmt(fmt), ##__VA_ARGS__)
^~~~~~~~~~
mm/debug.c:133:2: note: in expansion of macro 'pr_emerg'
pr_emerg("mm %px mmap %px seqnum %llu task_size %lu"
^~~~~~~~
mm/debug.c:140:17: note: format string is defined here
"pinned_vm %llx data_vm %lx exec_vm %lx stack_vm %lx"
~~~^
%lx
Qian Cai [Fri, 29 Mar 2019 03:43:15 +0000 (20:43 -0700)]
kasan: fix variable 'tag' set but not used warning
set_tag() compiles away when CONFIG_KASAN_SW_TAGS=n, so make
arch_kasan_set_tag() a static inline function to fix warnings below.
mm/kasan/common.c: In function '__kasan_kmalloc':
mm/kasan/common.c:475:5: warning: variable 'tag' set but not used [-Wunused-but-set-variable]
u8 tag;
^~~
By default, cells in DT are 32-bit in size. The driver reads "ti,mode"
using the function of_property_read_u8() which causes the value to be
read incorrectly in little-endian architectures if the size is not
specified.
Make it explicit in the binding documentation that this prorperty must
be set as a 8-bit value.
Guenter Roeck [Wed, 20 Mar 2019 17:32:58 +0000 (10:32 -0700)]
hwmon: (ntc_thermistor) Fix temperature type reporting
Commit 7cc7de93fad4 ("hwmon: (ntc_thermistor) Convert to new hwmon API")
converted the driver to use the new hwmon API, but introduced a subtle
error: The temperature type is no longer reported as temp1_type, but as
temp2_type.
Fixes: 7cc7de93fad4 ("hwmon: (ntc_thermistor) Convert to new hwmon API") Signed-off-by: Guenter Roeck <[email protected]>
Eddie James [Tue, 19 Mar 2019 21:01:58 +0000 (16:01 -0500)]
hwmon: (occ) Fix power sensor indexing
In the case of power sensor version 0xA0, the sensor indexing overlapped
with the "caps" power sensors, resulting in probe failure and kernel
warnings. Fix this by specifying the next index for each power sensor
version.
Gao Xiang [Thu, 28 Mar 2019 20:14:58 +0000 (04:14 +0800)]
staging: erofs: keep corrupted fs from crashing kernel in erofs_readdir()
After commit 419d6efc50e9, kernel cannot be crashed in the namei
path. However, corrupted nameoff can do harm in the process of
readdir for scenerios without dm-verity as well. Fix it now.
Joerg Roedel [Thu, 28 Mar 2019 10:44:59 +0000 (11:44 +0100)]
iommu/amd: Reserve exclusion range in iova-domain
If a device has an exclusion range specified in the IVRS
table, this region needs to be reserved in the iova-domain
of that device. This hasn't happened until now and can cause
data corruption on data transfered with these devices.
Treat exclusion ranges as reserved regions in the iommu-core
to fix the problem.
Fixes: be2a022c0dd0 ('x86, AMD IOMMU: add functions to parse IOMMU memory mapping requirements for devices') Signed-off-by: Joerg Roedel <[email protected]> Reviewed-by: Gary R Hook <[email protected]>
Changbin Du [Mon, 25 Mar 2019 15:16:47 +0000 (15:16 +0000)]
kconfig/[mn]conf: handle backspace (^H) key
Backspace is not working on some terminal emulators which do not send the
key code defined by terminfo. Terminals either send '^H' (8) or '^?' (127).
But currently only '^?' is handled. Let's also handle '^H' for those
terminals.
Merge tag 'fixes-for-v5.1-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/balbi/usb into usb-linus
Felipe writes:
usb: fixes for v5.1-rc2
One deadlock fix on f_hid. NET2280 got a fix on its dequeue
implementation and a fix for overrun of OUT messages.
DWC3 learned about another Intel product: Comment Lake PCH.
NET2272 got a similar fix to NET2280 on its dequeue implementation.
* tag 'fixes-for-v5.1-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/balbi/usb:
USB: gadget: f_hid: fix deadlock in f_hidg_write()
usb: gadget: net2272: Fix net2272_dequeue()
usb: gadget: net2280: Fix net2280_dequeue()
usb: gadget: net2280: Fix overrun of OUT messages
usb: dwc3: pci: add support for Comet Lake PCH ID
Anup Patel [Fri, 22 Mar 2019 10:04:44 +0000 (10:04 +0000)]
RISC-V: Fix FIXMAP_TOP to avoid overlap with VMALLOC area
The FIXMAP area overlaps with VMALLOC area in Linux-5.1-rc1 hence we get
below warning in Linux RISC-V 32bit kernel. This warning does not show-up
in Linux RISC-V 64bit kernel due to large VMALLOC area.
powerpc/pseries/mce: Fix misleading print for TLB mutlihit
On pseries, TLB multihit are reported as D-Cache Multihit. This is because
the wrongly populated mc_err_types[] array. Per PAPR, TLB error type is 0x04
and mc_err_types[4] points to "D-Cache" instead of "TLB" string. Fixup the
mc_err_types[] array.
Yan Zhao [Wed, 27 Mar 2019 04:55:45 +0000 (00:55 -0400)]
drm/i915/gvt: do not let pin count of shadow mm go negative
shadow mm's pin count got increased in workload preparation phase, which
is after workload scanning.
it will get decreased in complete_current_workload() anyway after
workload completion.
Sometimes, if a workload meets a scanning error, its shadow mm pin count
will not get increased but will get decreased in the end.
This patch lets shadow mm's pin count not go below 0.
Linus Walleij [Fri, 29 Mar 2019 02:04:47 +0000 (03:04 +0100)]
Merge tag 'gpio-v5.1-rc3-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux into fixes
gpio fixes for v5.1-rc3
- fix for a potential NULL-pointer dereference in the aspeed driver
- revert of the commit using the new gpio_set_config() when setting
debaunce and transitory state config as it caused a regression in
the aspeed driver
- two fixes for gpio-mockup for debugfs problems introduced in the
last merge window
Dave Airlie [Fri, 29 Mar 2019 00:38:25 +0000 (10:38 +1000)]
Merge tag 'drm-intel-fixes-2019-03-28' of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes
drm/i915 fixes for v5.2-rc3:
- fix mmap range checks
- fix gvt ppgtt mm LRU list access races
- fix selftest error pointer check
- fix a macro definition (pre-emptive for potential further backports)
- fix one AML SKU ULX status
net: core: netif_receive_skb_list: unlist skb before passing to pt->func
__netif_receive_skb_list_ptype() leaves skb->next poisoned before passing
it to pt_prev->func handler, what may produce (in certain cases, e.g. DSA
setup) crashes like:
Fix this by pulling skb off the sublist and zeroing skb->next pointer
before calling ptype callback.
Fixes: 88eb1944e18c ("net: core: propagate SKB lists through packet_type lookup") Reviewed-by: Edward Cree <[email protected]> Signed-off-by: Alexander Lobakin <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Petr Machata [Thu, 28 Mar 2019 12:12:27 +0000 (12:12 +0000)]
selftests: mlxsw: Add a new test for strict priority
Test that when strict priority is configured on a system, the
higher-priority traffic does actually win all the available bandwidth.
The test uses a similar approach to qos_mc_aware.sh to run and account
the traffic.
This test runs two streams of traffic from two independent ports to
create congestion on one egress port. It is necessary to configure the
shared buffer thresholds correctly, to make sure that there is traffic
from both streams in the shared buffer. Only then can the test actually
test prioritization among these streams.
Without this configuration, it is possible, that one of the streams
takes all of port-pool quota, and the other stream is not even admitted,
thus invalidating the result.
On Spectrum-1, this is not a problem, because MC traffic uses a separate
pool. But for Spectrum-2, MC and UC share the same pool, and the correct
configuration is important.
Ido Schimmel [Thu, 28 Mar 2019 12:12:19 +0000 (12:12 +0000)]
selftests: forwarding: Add reverse path forwarding (RPF) test cases
In case a packet is routed using a multicast route whose specified
ingress interface does not match the interface from which the packet was
received, the packet is dropped.
Add IPv4 and IPv6 test cases for above mentioned scenario.
Mao Wenan [Thu, 28 Mar 2019 09:10:56 +0000 (17:10 +0800)]
net: rds: force to destroy connection if t_sock is NULL in rds_tcp_kill_sock().
When it is to cleanup net namespace, rds_tcp_exit_net() will call
rds_tcp_kill_sock(), if t_sock is NULL, it will not call
rds_conn_destroy(), rds_conn_path_destroy() and rds_tcp_conn_free() to free
connection, and the worker cp_conn_w is not stopped, afterwards the net is freed in
net_drop_ns(); While cp_conn_w rds_connect_worker() will call rds_tcp_conn_path_connect()
and reference 'net' which has already been freed.
In rds_tcp_conn_path_connect(), rds_tcp_set_callbacks() will set t_sock = sock before
sock->ops->connect, but if connect() is failed, it will call
rds_tcp_restore_callbacks() and set t_sock = NULL, if connect is always
failed, rds_connect_worker() will try to reconnect all the time, so
rds_tcp_kill_sock() will never to cancel worker cp_conn_w and free the
connections.
Therefore, the condition !tc->t_sock is not needed if it is going to do
cleanup_net->rds_tcp_exit_net->rds_tcp_kill_sock, because tc->t_sock is always
NULL, and there is on other path to cancel cp_conn_w and free
connection. So this patch is to fix this.
rds_tcp_kill_sock():
...
if (net != c_net || !tc->t_sock)
... Acked-by: Santosh Shilimkar <[email protected]>
==================================================================
BUG: KASAN: use-after-free in inet_create+0xbcc/0xd28
net/ipv4/af_inet.c:340
Read of size 4 at addr ffff8003496a4684 by task kworker/u8:4/3721
Fixes: 467fa15356ac("RDS-TCP: Support multiple RDS-TCP listen endpoints, one per netns.") Reported-by: Hulk Robot <[email protected]> Signed-off-by: Mao Wenan <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Andrea Righi [Thu, 28 Mar 2019 06:36:00 +0000 (07:36 +0100)]
openvswitch: fix flow actions reallocation
The flow action buffer can be resized if it's not big enough to contain
all the requested flow actions. However, this resize doesn't take into
account the new requested size, the buffer is only increased by a factor
of 2x. This might be not enough to contain the new data, causing a
buffer overflow, for example:
====================
nfp: fix retcode and disable netpoll on representors
This series avoids a potential crash on nfp representor devices
when netpoll is in use. If transmitting the frame through underlying
vNIC fails we'd return an error code (by passing on error code from
__dev_queue_xmit()) and cause double free in netpoll code.
Fix the error code and disable netpoll on reprs altogether.
IRQ-safety of locking the queues and calling __dev_queue_xmit()
is questionable.
Big thanks to John Hurley for debugging and narrowing down
the trace log after I gave up! :)
====================
Jakub Kicinski [Wed, 27 Mar 2019 18:38:39 +0000 (11:38 -0700)]
nfp: disable netpoll on representors
NFP reprs are software device on top of the PF's vNIC.
The comment above __dev_queue_xmit() sayeth:
When calling this method, interrupts MUST be enabled. This is because
the BH enable code must have IRQs enabled so that it will not deadlock.
For netconsole we can't guarantee IRQ state, let's just
disable netpoll on representors to be on the safe side.
When the initial implementation of NFP reprs was added by the
commit 5de73ee46704 ("nfp: general representor implementation")
.ndo_poll_controller was required for netpoll to be enabled.
Jakub Kicinski [Wed, 27 Mar 2019 18:38:38 +0000 (11:38 -0700)]
nfp: validate the return code from dev_queue_xmit()
dev_queue_xmit() may return error codes as well as netdev_tx_t,
and it always consumes the skb. Make sure we always return a
correct netdev_tx_t value.
Some PHYs will use the 2500BaseX PHY_INTERFACE_MODE when being linked
with a partner using 2.5GBaseT.
Since we can't autonegotiate this speed between the MAC and the PHY, we
need to have the proper comphy support enabled, to make sure we can
safely advertise 2.5G and 1G in BaseT and be able to switch between both
corresponding PHY interface modes. This is now possible since comphy
support was added to this driver.
This commit adds the 2500BaseT mode to the list of supported modes when
using 2500BaseX, and was tested on a setup with an Armada385 and a 88E2010 PHY, both with and without the comphy node in the DT.
Eric Dumazet [Wed, 27 Mar 2019 15:21:30 +0000 (08:21 -0700)]
netns: provide pure entropy for net_hash_mix()
net_hash_mix() currently uses kernel address of a struct net,
and is used in many places that could be used to reveal this
address to a patient attacker, thus defeating KASLR, for
the typical case (initial net namespace, &init_net is
not dynamically allocated)
I believe the original implementation tried to avoid spending
too many cycles in this function, but security comes first.
Yi-Hung Wei [Tue, 26 Mar 2019 18:31:14 +0000 (11:31 -0700)]
openvswitch: Add timeout support to ct action
Add support for fine-grain timeout support to conntrack action.
The new OVS_CT_ATTR_TIMEOUT attribute of the conntrack action
specifies a timeout to be associated with this connection.
If no timeout is specified, it acts as is, that is the default
timeout for the connection will be automatically applied.
Example usage:
$ nfct timeout add timeout_1 inet tcp syn_sent 100 established 200
$ ovs-ofctl add-flow br0 in_port=1,ip,tcp,action=ct(commit,timeout=timeout_1)
Yi-Hung Wei [Tue, 26 Mar 2019 18:31:13 +0000 (11:31 -0700)]
netfilter: Export nf_ct_{set,destroy}_timeout()
This patch exports nf_ct_set_timeout() and nf_ct_destroy_timeout().
The two functions are derived from xt_ct_destroy_timeout() and
xt_ct_set_timeout() in xt_CT.c, and moved to nf_conntrack_timeout.c
without any functional change.
It would be useful for other users (i.e. OVS) that utilizes the
finer-grain conntrack timeout feature.
Linus Torvalds [Thu, 28 Mar 2019 20:29:09 +0000 (13:29 -0700)]
Merge tag 'pci-v5.1-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci
Pull PCI fixes from Bjorn Helgaas:
"PCI fixes:
- Clear level-triggered interrupts for the bandwidth notification
supported added for v5.1 (Alexandru Gagniuc)
- Clear bandwidth notification interrupts before enabling them (Lukas
Wunner)
- Report post-enumeration bandwidth changes only once for
multi-function devices (Lukas Wunner)"
* tag 'pci-v5.1-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
PCI/LINK: Deduplicate bandwidth reports for multi-function devices
PCI/LINK: Clear bandwidth notification interrupt before enabling it
PCI/LINK: Supply IRQ handler so level-triggered IRQs are acked
David S. Miller [Thu, 28 Mar 2019 19:59:54 +0000 (12:59 -0700)]
Merge branch '40GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-queue
Jeff Kirsher says:
====================
Intel Wired LAN Driver Fixes 2019-03-26
This series contains updates to igb, ixgbe, i40e and fm10k.
Jake fixes an issue with PTP in i40e where a previous commit resulted
in a regression where the driver would interpret small negative
adjustments as large positive additions, resulting in incorrect
behavior.
Arvind Sankar fixes an issue in igb where a previous commit would cause
a warning in the PCI pm core and resulted in pci_pm_runtime_suspend
would not call pci_save_state or pci_finish_runtime_suspend.
Ivan Vecera fixes MDIO bus registration with ixgbe, where the driver was
ignoring errors returned when registering and would leave the pointer in
a NULL state which triggered a BUG when un-registering.
Stefan Assmann fixes the check for Wake-On-LAN for i40e, which only
supports magic packet.
Yue Haibing fixes a potential NULL pointer de-reference in fm10k by
adding a simple check if the value is NULL.
====================
please apply the following patchset to net-next. This reworks the control
IO code in qeth so that we no longer need to poll for cmd completion,
and refactors the IDX setup code to also use this improved IO path.
====================
Julian Wiedmann [Thu, 28 Mar 2019 15:39:28 +0000 (16:39 +0100)]
s390/qeth: send IDX cmds via qeth_send_control_data()
This converts the IDX code to use qeth_send_control_data(), replacing
a bunch of duplicated IO code and unbounded waits. It also allows the
IDX sequence to benefit from the improved timeout & notify
infrastructure, so that we can eliminate the DOWN -> ACTIVATING -> UP
transition in the channel state machine.
The patch looks rather big, but most of it is a straight-forward
conversion of the old IDX cmd setup & callbacks to the new model.
Julian Wiedmann [Thu, 28 Mar 2019 15:39:27 +0000 (16:39 +0100)]
s390/qeth: use callback to finalize cmd
To avoid concurrency issues, some parts of the cmd setup are delayed
until qeth_send_control_data() holds the IO channel's irq_pending
"lock". Rather than hard-coding those setup steps for each cmd type,
have the cmd provide a callback. This will make it easier to also issue
IDX commands via qeth_send_control_data().
Julian Wiedmann [Thu, 28 Mar 2019 15:39:25 +0000 (16:39 +0100)]
s390/qeth: clarify default cmd callback
Current code makes it look like qeth_send_control_data_cb() is some
sort of default callback for all cmds. But in practice, it is only used
for half of the cmd buffers we issue.
Reduce the confusion by only setting this callback for cmds that
actually want it, and while at it give the callback a name that matches
the established naming scheme.
Julian Wiedmann [Thu, 28 Mar 2019 15:39:22 +0000 (16:39 +0100)]
s390/qeth: defer IPv6 address notifier events
The inet6addr_chain is atomic. So instead of starting the cmd IO for
SETIP / DELIP straight from the notifier callback, run it from a
workqueue. This is the last step towards removal of cmd IO completion
polling.
Julian Wiedmann [Thu, 28 Mar 2019 15:39:21 +0000 (16:39 +0100)]
s390/qeth: add wrapper for IP table access
Extract a little helper, so that high-level callers can manipulate the
IP table without worrying about the locking. This will make it easier
to convert the code to a different locking primitive later on.
Julian Wiedmann [Thu, 28 Mar 2019 15:39:20 +0000 (16:39 +0100)]
s390/qeth: remove locking for RX modeset cache
The L2 and L3 .ndo_set_rx_mode callbacks maintain an address cache
to decide which addresses have changed since the last modeset.
When the card is set offline, qeth_l?_stop_card() drains this cache.
This happens only after 1) the net_device has been detached, and
2) any pending RX modeset has completed. Consequently we can access the
cache lock-free.
Julian Wiedmann [Thu, 28 Mar 2019 15:39:19 +0000 (16:39 +0100)]
s390/qeth: defer RX modesetting
.ndo_set_rx_mode gets called in process context, but while holding the
addr_list spinlock. Which means we currently can't sleep while
re-programming the HW, and need to poll for IO completion. That's bad,
in particular since receiving the cmd response can fail silently and
we're then polling until the timeout hits.
As a first step towards eliminating the IO completion polling, run the
RX modeset from a work element and only take the addr_list lock while
updating the RX mode address cache.
===================
net: call for phys_port_name into devlink directly if possible
phys_port_name may be assembled by a helper in devlink. It is currently
the case only for mlxsw driver. Benefit from the get_devlink_port ndo
and call into devlink directly from dev_get_phys_port_name(). That saves
the trip to the driver, simplifies the code and makes it similar to
recently introduced ethtool-devlink compat helpers.
Move bnxt, partly nfp and dsa to let devlink core generate the name too.
===================
Jiri Pirko [Thu, 28 Mar 2019 12:56:46 +0000 (13:56 +0100)]
net: devlink: add warning for ndo_get_phys_port_name set when not needed
Currently if the driver registers devlink port instance, it should set
the devlink port attributes as well. Then the devlink core is able to
obtain physical port name itself, no need for driver to implement
the ndo. Once all drivers will implement devlink port registration,
this ndo should be removed. This warning guides new
drivers to do things as they should be done.
Jiri Pirko [Thu, 28 Mar 2019 12:56:43 +0000 (13:56 +0100)]
dsa: implement ndo_get_devlink_port
In order for devlink compat functions to work, implement
ndo_get_devlink_port. Legacy slaves does not have devlink port instances
created for themselves.
Jiri Pirko [Thu, 28 Mar 2019 12:56:41 +0000 (13:56 +0100)]
bnxt: implement ndo_get_devlink_port
In order for devlink compat functions to work, implement
ndo_get_devlink_port. Legacy slaves does not have devlink port instances
created for themselves.
Introduce devlink_compat_phys_port_name_get() helper that
gets the physical port name for specified netdevice
according to devlink port attributes.
Call this helper from dev_get_phys_port_name()
in case ndo_get_phys_port_name is not defined.
Jiri Pirko [Thu, 28 Mar 2019 12:56:36 +0000 (13:56 +0100)]
net: replace ndo_get_devlink with ndo_get_devlink_port
Follow-up patch is going to need a devlink port instance according to
a netdev. Devlink port instance should be always available when devlink
is used. So change the recently introduced ndo_get_devlink to
ndo_get_devlink_port. With that, adjust the wrapper for the only
user to get devlink pointer.
Jiri Pirko [Thu, 28 Mar 2019 12:56:35 +0000 (13:56 +0100)]
nfp: register devlink port before netdev
Change the init/fini flow and register devlink port instance before
netdev. Now it is needed for correct behavior of phys_port_name
generation, but in general it makes sense to register devlink port
first.
Adrian Hunter [Wed, 27 Mar 2019 07:28:26 +0000 (09:28 +0200)]
perf scripts python: exported-sql-viewer.py: Fix python3 support
Unlike python2, python3 strings are not compatible with byte strings.
That results in disassembly not working for the branches reports. Fixup
those places overlooked in the port to python3.
Paolo Bonzini [Thu, 28 Mar 2019 18:07:30 +0000 (19:07 +0100)]
Merge tag 'kvmarm-fixes-for-5.1' of git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm into kvm-master
KVM/ARM fixes for 5.1
- Fix THP handling in the presence of pre-existing PTEs
- Honor request for PTE mappings even when THPs are available
- GICv4 performance improvement
- Take the srcu lock when writing to guest-controlled ITS data structures
- Reset the virtual PMU in preemptible context
- Various cleanups
pyside version 1 fails to handle python3 large integers in some cases,
resulting in Qt getting into a never-ending loop. This affects:
samples Table
samples_view Table
All branches Report
Selected branches Report
Wei Li [Thu, 28 Feb 2019 09:20:03 +0000 (17:20 +0800)]
perf machine: Update kernel map address and re-order properly
Since commit 1fb87b8e9599 ("perf machine: Don't search for active kernel
start in __machine__create_kernel_maps"), the __machine__create_kernel_maps()
just create a map what start and end are both zero. Though the address will be
updated later, the order of map in the rbtree may be incorrect.
The commit ee05d21791db ("perf machine: Set main kernel end address properly")
fixed the logic in machine__create_kernel_maps(), but it's still wrong in
function machine__process_kernel_mmap_event().
To reproduce this issue, we need an environment which the module address
is before the kernel text segment. I tested it on an aarch64 machine with
kernel 4.19.25:
[root@localhost hulk]# grep _stext /proc/kallsyms ffff000008081000 T _stext
[root@localhost hulk]# grep _etext /proc/kallsyms ffff000009780000 R _etext
[root@localhost hulk]# tail /proc/modules
hisi_sas_v2_hw 77824 0 - Live 0xffff00000191d000
nvme_core 126976 7 nvme, Live 0xffff0000018b6000
mdio 20480 1 ixgbe, Live 0xffff0000018ab000
hisi_sas_main 106496 1 hisi_sas_v2_hw, Live 0xffff000001861000
hns_mdio 20480 2 - Live 0xffff000001822000
hnae 28672 3 hns_dsaf,hns_enet_drv, Live 0xffff000001815000
dm_mirror 40960 0 - Live 0xffff000001804000
dm_region_hash 32768 1 dm_mirror, Live 0xffff0000017f5000
dm_log 32768 2 dm_mirror,dm_region_hash, Live 0xffff0000017e7000
dm_mod 315392 17 dm_mirror,dm_log, Live 0xffff000001780000
[root@localhost hulk]#
Before fix:
[root@localhost bin]# perf record sleep 3
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.011 MB perf.data (9 samples) ]
[root@localhost bin]# perf buildid-list -i perf.data 4c4e46c971ca935f781e603a09b52a92e8bdfee8 [vdso]
[root@localhost bin]# perf buildid-list -i perf.data -H 0000000000000000000000000000000000000000 /proc/kcore
[root@localhost bin]#
Warning: Kernel ABI header at 'tools/arch/powerpc/include/uapi/asm/kvm.h' differs from latest version at 'arch/powerpc/include/uapi/asm/kvm.h'
diff -u tools/arch/powerpc/include/uapi/asm/kvm.h arch/powerpc/include/uapi/asm/kvm.h
projects / linux.git / log