Similar to the case for key_validate(), we should load the key ->expiry
once atomically in keyring_search_iterator(), since it can be changed
concurrently with the flags whenever the key semaphore isn't held.
Signed-off-by: Eric Biggers <[email protected]>
Signed-off-by: David Howells <[email protected]>
In key_validate(), load the flags and expiry time once atomically, since
these can change concurrently if key_validate() is called without the
key semaphore held. And we don't want to get inconsistent results if a
variable is referenced multiple times. For example, key->expiry was
referenced in both 'if (key->expiry)' and in 'if (now.tv_sec >=
key->expiry)', making it theoretically possible to see a spurious
EKEYEXPIRED while the expiration time was being removed, i.e. set to 0.
Signed-off-by: Eric Biggers <[email protected]>
Signed-off-by: David Howells <[email protected]>
Currently, when passed a key that already exists, add_key() will call the
key's ->update() method if such exists. But this is heavily broken in the
case where the key is uninstantiated because it doesn't call
__key_instantiate_and_link(). Consequently, it doesn't do most of the
things that are supposed to happen when the key is instantiated, such as
setting the instantiation state, clearing KEY_FLAG_USER_CONSTRUCT and
awakening tasks waiting on it, and incrementing key->user->nikeys.
It also never takes key_construction_mutex, which means that
->instantiate() can run concurrently with ->update() on the same key. In
the case of the "user" and "logon" key types this causes a memory leak, at
best. Maybe even worse, the ->update() methods of the "encrypted" and
"trusted" key types actually just dereference a NULL pointer when passed an
uninstantiated key.
Change key_create_or_update() to wait interruptibly for the key to finish
construction before continuing.
This patch only affects *uninstantiated* keys. For now we still allow a
negatively instantiated key to be updated (thereby positively
instantiating it), although that's broken too (the next patch fixes it)
and I'm not sure that anyone actually uses that functionality either.
Here is a simple reproducer for the bug using the "encrypted" key type
(requires CONFIG_ENCRYPTED_KEYS=y), though as noted above the bug
pertained to more than just the "encrypted" key type:
#include <stdlib.h>
#include <unistd.h>
#include <keyutils.h>
int main(void)
{
int ringid = keyctl_join_session_keyring(NULL);
if (fork()) {
for (;;) {
const char payload[] = "update user:foo 32";
usleep(rand() % 10000);
add_key("encrypted", "desc", payload, sizeof(payload), ringid);
keyctl_clear(ringid);
}
} else {
for (;;)
request_key("encrypted", "desc", "callout_info", ringid);
}
}
It causes:
BUG: unable to handle kernel NULL pointer dereference at 0000000000000018
IP: encrypted_update+0xb0/0x170
PGD 7a178067 P4D 7a178067 PUD 77269067 PMD 0
PREEMPT SMP
CPU: 0 PID: 340 Comm: reproduce Tainted: G D 4.14.0-rc1-00025-g428490e38b2e #796
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011
task: ffff8a467a39a340 task.stack: ffffb15c40770000
RIP: 0010:encrypted_update+0xb0/0x170
RSP: 0018:ffffb15c40773de8 EFLAGS: 00010246
RAX: 0000000000000000 RBX: ffff8a467a275b00 RCX: 0000000000000000
RDX: 0000000000000005 RSI: ffff8a467a275b14 RDI: ffffffffb742f303
RBP: ffffb15c40773e20 R08: 0000000000000000 R09: ffff8a467a275b17
R10: 0000000000000020 R11: 0000000000000000 R12: 0000000000000000
R13: 0000000000000000 R14: ffff8a4677057180 R15: ffff8a467a275b0f
FS: 00007f5d7fb08700(0000) GS:ffff8a467f200000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000018 CR3: 0000000077262005 CR4: 00000000001606f0
Call Trace:
key_create_or_update+0x2bc/0x460
SyS_add_key+0x10c/0x1d0
entry_SYSCALL_64_fastpath+0x1f/0xbe
RIP: 0033:0x7f5d7f211259
RSP: 002b:00007ffed03904c8 EFLAGS: 00000246 ORIG_RAX: 00000000000000f8
RAX: ffffffffffffffda RBX: 000000003b2a7955 RCX: 00007f5d7f211259
RDX: 00000000004009e4 RSI: 00000000004009ff RDI: 0000000000400a04
RBP: 0000000068db8bad R08: 000000003b2a7955 R09: 0000000000000004
R10: 000000000000001a R11: 0000000000000246 R12: 0000000000400868
R13: 00007ffed03905d0 R14: 0000000000000000 R15: 0000000000000000
Code: 77 28 e8 64 34 1f 00 45 31 c0 31 c9 48 8d 55 c8 48 89 df 48 8d 75 d0 e8 ff f9 ff ff 85 c0 41 89 c4 0f 88 84 00 00 00 4c 8b 7d c8 <49> 8b 75 18 4c 89 ff e8 24 f8 ff ff 85 c0 41 89 c4 78 6d 49 8b
RIP: encrypted_update+0xb0/0x170 RSP: ffffb15c40773de8
CR2: 0000000000000018
Cc: <[email protected]> # v2.6.12+
Reported-by: Eric Biggers <[email protected]>
Signed-off-by: David Howells <[email protected]>
cc: Eric Biggers <[email protected]>
Consolidate KEY_FLAG_INSTANTIATED, KEY_FLAG_NEGATIVE and the rejection
error into one field such that:
(1) The instantiation state can be modified/read atomically.
(2) The error can be accessed atomically with the state.
(3) The error isn't stored unioned with the payload pointers.
This deals with the problem that the state is spread over three different
objects (two bits and a separate variable) and reading or updating them
atomically isn't practical, given that not only can uninstantiated keys
change into instantiated or rejected keys, but rejected keys can also turn
into instantiated keys - and someone accessing the key might not be using
any locking.
The main side effect of this problem is that what was held in the payload
may change, depending on the state. For instance, you might observe the
key to be in the rejected state. You then read the cached error, but if
the key semaphore wasn't locked, the key might've become instantiated
between the two reads - and you might now have something in hand that isn't
actually an error code.
The state is now KEY_IS_UNINSTANTIATED, KEY_IS_POSITIVE or a negative error
code if the key is negatively instantiated. The key_is_instantiated()
function is replaced with key_is_positive() to avoid confusion as negative
keys are also 'instantiated'.
Additionally, barriering is included:
(1) Order payload-set before state-set during instantiation.
(2) Order state-read before payload-read when using the key.
Further separate barriering is necessary if RCU is being used to access the
payload content after reading the payload pointers.
Fixes: 146aa8b145 ("KEYS: Merge the type-specific data with the payload data")
Cc: [email protected] # v4.4+
Reported-by: Eric Biggers <[email protected]>
Signed-off-by: David Howells <[email protected]>
Reviewed-by: Eric Biggers <[email protected]>
For finding asymmetric key, the input id_0 and id_1 parameters can
not be NULL at the same time. This patch adds the BUG_ON checking
for id_0 and id_1.
Cc: David Howells <[email protected]>
Cc: Herbert Xu <[email protected]>
Cc: "David S. Miller" <[email protected]>
Signed-off-by: Chun-Yi Lee <[email protected]>
Signed-off-by: David Howells <[email protected]>
Fix the wrong index number when checking the existence of second
id in function of finding asymmetric key. The id_1 is the second
id that the index in array must be 1 but not 0.
Fixes: 9eb029893a (KEYS: Generalise x509_request_asymmetric_key())
Cc: David Howells <[email protected]>
Cc: Herbert Xu <[email protected]>
Cc: "David S. Miller" <[email protected]>
Signed-off-by: Chun-Yi Lee <[email protected]>
Signed-off-by: David Howells <[email protected]>
The recent rework introduced a possible randconfig build failure
when CONFIG_CRYPTO configured to only allow modules:
security/keys/big_key.o: In function `big_key_crypt':
big_key.c:(.text+0x29f): undefined reference to `crypto_aead_setkey'
security/keys/big_key.o: In function `big_key_init':
big_key.c:(.init.text+0x1a): undefined reference to `crypto_alloc_aead'
big_key.c:(.init.text+0x45): undefined reference to `crypto_aead_setauthsize'
big_key.c:(.init.text+0x77): undefined reference to `crypto_destroy_tfm'
crypto/gcm.o: In function `gcm_hash_crypt_remain_continue':
gcm.c:(.text+0x167): undefined reference to `crypto_ahash_finup'
crypto/gcm.o: In function `crypto_gcm_exit_tfm':
gcm.c:(.text+0x847): undefined reference to `crypto_destroy_tfm'
When we 'select CRYPTO' like the other users, we always get a
configuration that builds.
Fixes: 428490e38b ("security/keys: rewrite all of big_key crypto")
Signed-off-by: Arnd Bergmann <[email protected]>
Signed-off-by: David Howells <[email protected]>
The 'use' locking macros are no-ops if neither SMP or SND_DEBUG is
enabled. This might once have been OK in non-preemptible
configurations, but even in that case snd_seq_read() may sleep while
relying on a 'use' lock. So always use the proper implementations.
Cc: [email protected]
Signed-off-by: Ben Hutchings <[email protected]>
Signed-off-by: Takashi Iwai <[email protected]>
This reverts commit c91fc8519d.
That change caused a C6 and PC6 residency regression on large idle systems.
Users also complained about new output indicating jitter:
turbostat: cpu6 jitter 3794 9142
Signed-off-by: Len Brown <[email protected]>
Cc: 4.13+ <[email protected]> # v4.13+
Signed-off-by: Rafael J. Wysocki <[email protected]>
Otherwise we can get the following if the fck alias is missing:
Unable to handle kernel paging request at virtual address fffffffe
...
PC is at clk_get_rate+0x8/0x10
LR is at omap_i2c_probe+0x278/0x6ec
...
[<c056eb08>] (clk_get_rate) from [<c06f4f08>] (omap_i2c_probe+0x278/0x6ec)
[<c06f4f08>] (omap_i2c_probe) from [<c0610944>] (platform_drv_probe+0x50/0xb0)
[<c0610944>] (platform_drv_probe) from [<c060e900>] (driver_probe_device+0x264/0x2ec)
[<c060e900>] (driver_probe_device) from [<c060cda0>] (bus_for_each_drv+0x70/0xb8)
[<c060cda0>] (bus_for_each_drv) from [<c060e5b0>] (__device_attach+0xcc/0x13c)
[<c060e5b0>] (__device_attach) from [<c060db10>] (bus_probe_device+0x88/0x90)
[<c060db10>] (bus_probe_device) from [<c060df68>] (deferred_probe_work_func+0x4c/0x14c)
Signed-off-by: Tony Lindgren <[email protected]>
Signed-off-by: Wolfram Sang <[email protected]>
Pull SCSI fixes from James Bottomley:
"Four mostly error leg fixes and one more important regression in a
prior commit (the qla2xxx one)"
* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
scsi: fc: check for rport presence in fc_block_scsi_eh
scsi: qla2xxx: Fix uninitialized work element
scsi: libiscsi: fix shifting of DID_REQUEUE host byte
scsi: libfc: fix a deadlock in fc_rport_work
scsi: fixup kernel warning during rmmod()
Commit 7496946a8 ("tracing: Add samples of DECLARE_EVENT_CLASS() and
DEFINE_EVENT()") added template examples for all the events. It created a
DEFINE_EVENT_FN() example which reused the foo_bar_reg and foo_bar_unreg
functions.
Enabling both the TRACE_EVENT_FN() and DEFINE_EVENT_FN() example trace
events caused the foo_bar_reg to be called twice, creating the test thread
twice. The foo_bar_unreg would remove it only once, even if it was called
multiple times, leaving a thread existing when the module is unloaded,
causing an oops.
Add a ref count and allow foo_bar_reg() and foo_bar_unreg() be called by
multiple trace events.
Cc: [email protected]
Fixes: 7496946a8 ("tracing: Add samples of DECLARE_EVENT_CLASS() and DEFINE_EVENT()")
Signed-off-by: Steven Rostedt (VMware) <[email protected]>
The latest dtc warns about an extraneous cell in the interrupt
property of two of the iommu device nodes:
Warning (interrupts_property): interrupts size is (16), expected multiple of 12 in /iommu@ff373f00
Warning (interrupts_property): interrupts size is (16), expected multiple of 12 in /iommu@ff900800
This removes the typo.
Fixes: cede4c79de ("arm64: dts: rockchip: add rk3368 iommu nodes")
Fixes: 49c82f2b7c ("arm64: dts: rockchip: add rk3328 iommu nodes")
Signed-off-by: Arnd Bergmann <[email protected]>
Signed-off-by: Heiko Stuebner <[email protected]>
The vcc_sd or vcc_sdio used for IO voltage for sdmmc and sdio
interface on rk3399 platform have a limitation that it can't be
larger than 3.0v, otherwise it has a potential risk for the chip.
Correct all of them.
Fixes: 171582e00d ("arm64: dts: rockchip: add support for firefly-rk3399 board")
Fixes: 2c66fc34e9 ("arm64: dts: rockchip: add RK3399-Q7 (Puma) SoM")
Fixes: 8164a84cca ("arm64: dts: rockchip: Add support for rk3399 sapphire SOM")
Cc: [email protected]
Signed-off-by: Shawn Lin <[email protected]>
Tested-by: Klaus Goger <[email protected]>
Signed-off-by: Heiko Stuebner <[email protected]>
Currently we try to defer completion of async DIO to the process context
in case there are any mapped pages associated with the inode so that we
can invalidate the pages when the IO completes. However the check is racy
and the pages can be mapped afterwards. If this happens we might end up
calling invalidate_inode_pages2_range() in dio_complete() in interrupt
context which could sleep. This can be reproduced by generic/451.
Fix this by passing the information whether we can or can't invalidate
to the dio_complete(). Thanks Eryu Guan for reporting this and Jan Kara
for suggesting a fix.
Fixes: 332391a993 ("fs: Fix page cache inconsistency when mixing buffered and AIO DIO")
Reported-by: Eryu Guan <[email protected]>
Reviewed-by: Jan Kara <[email protected]>
Tested-by: Eryu Guan <[email protected]>
Signed-off-by: Lukas Czerner <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
Commit 3b2435192f ("MAINTAINERS: drop OMAP USB and MUSB
maintainership") switched the maintainer for musb module, but didn't
update the git tree location.
Delete the git tree information, since the current maintainer doesn't
have a public tree.
Reported-by: Jonathan Liu <[email protected]>
Signed-off-by: Bin Liu <[email protected]>
Acked-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
Thomas reported that 'perf buildid-list' gets a SEGFAULT due to NULL
pointer deref when he ran it on a data with namespace events. It was
because the buildid_id__mark_dso_hit_ops lacks the namespace event
handler and perf_too__fill_default() didn't set it.
Program received signal SIGSEGV, Segmentation fault.
0x0000000000000000 in ?? ()
Missing separate debuginfos, use: dnf debuginfo-install audit-libs-2.7.7-1.fc25.s390x bzip2-libs-1.0.6-21.fc25.s390x elfutils-libelf-0.169-1.fc25.s390x
+elfutils-libs-0.169-1.fc25.s390x libcap-ng-0.7.8-1.fc25.s390x numactl-libs-2.0.11-2.ibm.fc25.s390x openssl-libs-1.1.0e-1.1.ibm.fc25.s390x perl-libs-5.24.1-386.fc25.s390x
+python-libs-2.7.13-2.fc25.s390x slang-2.3.0-7.fc25.s390x xz-libs-5.2.3-2.fc25.s390x zlib-1.2.8-10.fc25.s390x
(gdb) where
#0 0x0000000000000000 in ?? ()
#1 0x00000000010fad6a in machines__deliver_event (machines=<optimized out>, machines@entry=0x2c6fd18,
evlist=<optimized out>, event=event@entry=0x3fffdf00470, sample=0x3ffffffe880, sample@entry=0x3ffffffe888,
tool=tool@entry=0x1312968 <build_id.mark_dso_hit_ops>, file_offset=1136) at util/session.c:1287
#2 0x00000000010fbf4e in perf_session__deliver_event (file_offset=1136, tool=0x1312968 <build_id.mark_dso_hit_ops>,
sample=0x3ffffffe888, event=0x3fffdf00470, session=0x2c6fc30) at util/session.c:1340
#3 perf_session__process_event (session=0x2c6fc30, session@entry=0x0, event=event@entry=0x3fffdf00470,
file_offset=file_offset@entry=1136) at util/session.c:1522
#4 0x00000000010fddde in __perf_session__process_events (file_size=11880, data_size=<optimized out>,
data_offset=<optimized out>, session=0x0) at util/session.c:1899
#5 perf_session__process_events (session=0x0, session@entry=0x2c6fc30) at util/session.c:1953
#6 0x000000000103b2ac in perf_session__list_build_ids (with_hits=<optimized out>, force=<optimized out>)
at builtin-buildid-list.c:83
#7 cmd_buildid_list (argc=<optimized out>, argv=<optimized out>) at builtin-buildid-list.c:115
#8 0x00000000010a026c in run_builtin (p=0x1311f78 <commands+24>, argc=argc@entry=2, argv=argv@entry=0x3fffffff3c0)
at perf.c:296
#9 0x000000000102bc00 in handle_internal_command (argv=<optimized out>, argc=2) at perf.c:348
#10 run_argv (argcp=<synthetic pointer>, argv=<synthetic pointer>) at perf.c:392
#11 main (argc=<optimized out>, argv=0x3fffffff3c0) at perf.c:536
(gdb)
Fix it by adding a stub event handler for namespace event.
Committer testing:
Further clarifying, plain using 'perf buildid-list' will not end up in a
SEGFAULT when processing a perf.data file with namespace info:
# perf record -a --namespaces sleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 2.024 MB perf.data (1058 samples) ]
# perf buildid-list | wc -l
38
# perf buildid-list | head -5
e2a171c7b905826fc8494f0711ba76ab6abbd604 /lib/modules/4.14.0-rc3+/build/vmlinux
874840a02d8f8a31cedd605d0b8653145472ced3 /lib/modules/4.14.0-rc3+/kernel/arch/x86/kvm/kvm-intel.ko
ea7223776730cd8a22f320040aae4d54312984bc /lib/modules/4.14.0-rc3+/kernel/drivers/gpu/drm/i915/i915.ko
5961535e6732a8edb7f22b3f148bb2fa2e0be4b9 /lib/modules/4.14.0-rc3+/kernel/drivers/gpu/drm/drm.ko
f045f54aa78cf1931cc893f78b6cbc52c72a8cb1 /usr/lib64/libc-2.25.so
#
It is only when one asks for checking what of those entries actually had
samples, i.e. when we use either -H or --with-hits, that we will process
all the PERF_RECORD_ events, and since tools/perf/builtin-buildid-list.c
neither explicitely set a perf_tool.namespaces() callback nor the
default stub was set that we end up, when processing a
PERF_RECORD_NAMESPACE record, causing a SEGFAULT:
# perf buildid-list -H
Segmentation fault (core dumped)
^C
#
Reported-and-Tested-by: Thomas-Mich Richter <[email protected]>
Signed-off-by: Namhyung Kim <[email protected]>
Tested-by: Arnaldo Carvalho de Melo <[email protected]>
Cc: Hari Bathini <[email protected]>
Cc: Hendrik Brueckner <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Thomas-Mich Richter <[email protected]>
Fixes: f3b3614a28 ("perf tools: Add PERF_RECORD_NAMESPACES to include namespaces related info")
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
'perf record' had a '-l' option that meant "scale counter values" a very
long time ago, but it currently belongs to 'perf stat' as '-c'. So
remove it. I found this problem in the below case.
$ perf record -e cycles -l sleep 3
Error: unknown switch `l
Signed-off-by: Taeung Song <[email protected]>
Cc: Jiri Olsa <[email protected]>
Cc: Namhyung Kim <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
This keyboard doesn't implement Get String descriptors properly even
though string indexes are valid. What happens is that when requesting
for the String descriptor, the device disconnects and
reconnects. Without this quirk, this loop will continue forever.
Cc: Alan Stern <[email protected]>
Reported-by: Владимир Мартьянов <[email protected]>
Cc: stable <[email protected]>
Signed-off-by: Felipe Balbi <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
Pull media fixes from Mauro Carvalho Chehab:
"Core fixes:
- cec: Respond to unregistered initiators, when applicable
- dvb_frontend: only use kref after initialized
Driver-specific fixes:
- qcom, camss: Make function vfe_set_selection static
- qcom: VIDEO_QCOM_CAMSS should depend on HAS_DMA
- s5p-cec: add NACK detection support
- media: staging/imx: Fix uninitialized variable warning
- dib3000mc: i2c transfers over usb cannot be done from stack
- venus: init registered list on streamoff"
* tag 'media/v4.14-2' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media:
media: dvb_frontend: only use kref after initialized
media: platform: VIDEO_QCOM_CAMSS should depend on HAS_DMA
media: cec: Respond to unregistered initiators, when applicable
media: s5p-cec: add NACK detection support
media: staging/imx: Fix uninitialized variable warning
media: qcom: camss: Make function vfe_set_selection static
media: venus: init registered list on streamoff
media: dvb: i2c transfers over usb cannot be done from stack
This fixes a kernel oops when unloading the driver due to usb_put_phy
being called after usb_phy_generic_unregister when the device is
detached. Calling usb_phy_generic_unregister causes x->dev->driver to
be NULL in usb_put_phy and results in a NULL pointer dereference.
Cc: [email protected] # v4.3+
Signed-off-by: Jonathan Liu <[email protected]>
Signed-off-by: Bin Liu <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
The sunxi musb has a bug where sometimes it will generate a babble
error on device disconnect instead of a disconnect IRQ. When this
happens the musb controller switches from host mode to device mode
(it clears MUSB_DEVCTL_HM/MUSB_DEVCTL_SESSION and sets
MUSB_DEVCTL_BDEVICE) and gets stuck in this state.
The babble error is misdetected as a bus reset because MUSB_DEVCTL_HM
was cleared.
To fix this, use is_host_active() rather than (devctl & MUSB_DEVCTL_HM)
to detect babble error so that sunxi musb babble recovery can handle it
by restoring the mode. This information is provided by the driver logic
and does not rely on register contents.
Cc: [email protected] # v4.1+
Signed-off-by: Jonathan Liu <[email protected]>
Signed-off-by: Bin Liu <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
Currently, the number of channels is set to 15 but in the case of DA8xx,
the number of channels is 4.
Update the driver to configure the number of channels at runtime.
Cc: [email protected] # v4.12+
Signed-off-by: Alexandre Bailon <[email protected]>
Tested-by: Sekhar Nori <[email protected]>
Signed-off-by: Bin Liu <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
The way to configure the DMA mode on DA8xx is different from DSPS.
Add a new function to configure DMA mode on DA8xx and use a callback
to call the right function based on the platform.
Cc: [email protected] # v4.12+
Signed-off-by: Alexandre Bailon <[email protected]>
Tested-by: Sekhar Nori <[email protected]>
Signed-off-by: Bin Liu <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
The DA8xx and DSPS platforms don't use the same address for few registers.
On Da8xx, this is causing some issues (e.g. teardown that doesn't work).
Configure the address of the register during the init and use them instead
of constants.
Cc: [email protected] # v4.12+
Reported-by: [email protected]
Signed-off-by: Alexandre Bailon <[email protected]>
Tested-by: Sekhar Nori <[email protected]>
Signed-off-by: Bin Liu <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
The musb delayed irq work was never flushed on suspend, something which
since 4.9 can lead to an external abort if the work is scheduled after
the grandparent's clock has been disabled:
PM: Suspending system (mem)
PM: suspend of devices complete after 125.224 msecs
PM: suspend devices took 0.132 seconds
PM: late suspend of devices complete after 7.423 msecs
PM: noirq suspend of devices complete after 7.083 msecs
suspend debug: Waiting for 5 second(s).
Unhandled fault: external abort on non-linefetch (0x1008) at 0xd0262c60
...
[<c054880c>] (musb_default_readb) from [<c0547b5c>] (musb_irq_work+0x48/0x220)
[<c0547b5c>] (musb_irq_work) from [<c014f8a4>] (process_one_work+0x1f4/0x758)
[<c014f8a4>] (process_one_work) from [<c014fe5c>] (worker_thread+0x54/0x514)
[<c014fe5c>] (worker_thread) from [<c015704c>] (kthread+0x128/0x158)
[<c015704c>] (kthread) from [<c0109330>] (ret_from_fork+0x14/0x24)
Commit 2bff3916fd ("usb: musb: Fix PM for hub disconnect") started
scheduling musb_irq_work with a delay of up to a second and with
retries thereby making this easy to trigger, for example, by suspending
shortly after a disconnect.
Note that we set a flag to prevent the irq work from rescheduling itself
during suspend and instead process a disconnect immediately. This takes
care of the case where we are disconnected shortly before suspending.
However, when in host mode, a disconnect while suspended will still
go unnoticed and thus prevent the controller from runtime suspending
upon resume as the session bit is always set. This will need to be
addressed separately.
Fixes: 550a7375fe ("USB: Add MUSB and TUSB support")
Fixes: 467d5c9807 ("usb: musb: Implement session bit based runtime PM for musb-core")
Fixes: 2bff3916fd ("usb: musb: Fix PM for hub disconnect")
Cc: stable <[email protected]> # 4.9
Cc: Felipe Balbi <[email protected]>
Cc: Tony Lindgren <[email protected]>
Signed-off-by: Johan Hovold <[email protected]>
Tested-by: Tony Lindgren <[email protected]>
Signed-off-by: Bin Liu <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
The current session-bit quirk implementation does not prevent the retry
counter from underflowing, something which could break runtime PM and
keep the device active for a very long time (about 2^32 seconds) after a
disconnect.
This notably breaks the B-device timeout case, but could potentially
cause problems also when the controller is operating as an A-device.
Fixes: 2bff3916fd ("usb: musb: Fix PM for hub disconnect")
Cc: stable <[email protected]> # 4.9
Cc: Tony Lindgren <[email protected]>
Signed-off-by: Johan Hovold <[email protected]>
Tested-by: Tony Lindgren <[email protected]>
Signed-off-by: Bin Liu <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
Elatec TWN3 has the union descriptor on data interface. This results in
failure to bind the device to the driver with the following log:
usb 1-1.2: new full speed USB device using streamplug-ehci and address 4
usb 1-1.2: New USB device found, idVendor=09d8, idProduct=0320
usb 1-1.2: New USB device strings: Mfr=1, Product=2, SerialNumber=0
usb 1-1.2: Product: RFID Device (COM)
usb 1-1.2: Manufacturer: OEM
cdc_acm 1-1.2:1.0: Zero length descriptor references
cdc_acm: probe of 1-1.2:1.0 failed with error -22
Adding the NO_UNION_NORMAL quirk for the device fixes the issue.
`lsusb -v` of the device:
Bus 001 Device 003: ID 09d8:0320
Device Descriptor:
bLength 18
bDescriptorType 1
bcdUSB 2.00
bDeviceClass 2 Communications
bDeviceSubClass 0
bDeviceProtocol 0
bMaxPacketSize0 32
idVendor 0x09d8
idProduct 0x0320
bcdDevice 3.00
iManufacturer 1 OEM
iProduct 2 RFID Device (COM)
iSerial 0
bNumConfigurations 1
Configuration Descriptor:
bLength 9
bDescriptorType 2
wTotalLength 67
bNumInterfaces 2
bConfigurationValue 1
iConfiguration 0
bmAttributes 0x80
(Bus Powered)
MaxPower 250mA
Interface Descriptor:
bLength 9
bDescriptorType 4
bInterfaceNumber 0
bAlternateSetting 0
bNumEndpoints 1
bInterfaceClass 2 Communications
bInterfaceSubClass 2 Abstract (modem)
bInterfaceProtocol 1 AT-commands (v.25ter)
iInterface 0
Endpoint Descriptor:
bLength 7
bDescriptorType 5
bEndpointAddress 0x83 EP 3 IN
bmAttributes 3
Transfer Type Interrupt
Synch Type None
Usage Type Data
wMaxPacketSize 0x0020 1x 32 bytes
bInterval 2
Interface Descriptor:
bLength 9
bDescriptorType 4
bInterfaceNumber 1
bAlternateSetting 0
bNumEndpoints 2
bInterfaceClass 10 CDC Data
bInterfaceSubClass 0 Unused
bInterfaceProtocol 0
iInterface 0
Endpoint Descriptor:
bLength 7
bDescriptorType 5
bEndpointAddress 0x02 EP 2 OUT
bmAttributes 2
Transfer Type Bulk
Synch Type None
Usage Type Data
wMaxPacketSize 0x0020 1x 32 bytes
bInterval 0
Endpoint Descriptor:
bLength 7
bDescriptorType 5
bEndpointAddress 0x81 EP 1 IN
bmAttributes 2
Transfer Type Bulk
Synch Type None
Usage Type Data
wMaxPacketSize 0x0020 1x 32 bytes
bInterval 0
CDC Header:
bcdCDC 1.10
CDC Call Management:
bmCapabilities 0x03
call management
use DataInterface
bDataInterface 1
CDC ACM:
bmCapabilities 0x06
sends break
line coding and serial state
CDC Union:
bMasterInterface 0
bSlaveInterface 1
Device Status: 0x0000
(Bus Powered)
Signed-off-by: Maksim Salau <[email protected]>
Acked-by: Oliver Neukum <[email protected]>
Cc: stable <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
Taking the uurb->buffer_length userspace passes in as a maximum for the
actual urbs transfer_buffer_length causes 2 serious issues:
1) It breaks isochronous support for all userspace apps using libusb,
as existing libusb versions pass in 0 for uurb->buffer_length,
relying on the kernel using the lenghts of the usbdevfs_iso_packet_desc
descriptors passed in added together as buffer length.
This for example causes redirection of USB audio and Webcam's into
virtual machines using qemu-kvm to no longer work. This is a userspace
ABI break and as such must be reverted.
Note that the original commit does not protect other users / the
kernels memory, it only stops the userspace process making the call
from shooting itself in the foot.
2) It may cause the kernel to program host controllers to DMA over random
memory. Just as the devio code used to only look at the iso_packet_desc
lenghts, the host drivers do the same, relying on the submitter of the
urbs to make sure the entire buffer is large enough and not checking
transfer_buffer_length.
But the "USB: devio: Don't corrupt user memory" commit now takes the
userspace provided uurb->buffer_length for the buffer-size while copying
over the user-provided iso_packet_desc lengths 1:1, allowing the user
to specify a small buffer size while programming the host controller to
dma a lot more data.
(Atleast the ohci, uhci, xhci and fhci drivers do not check
transfer_buffer_length for isoc transfers.)
This reverts commit fa1ed74eb1 ("USB: devio: Don't corrupt user memory")
fixing both these issues.
Cc: Dan Carpenter <[email protected]>
Cc: [email protected]
Signed-off-by: Hans de Goede <[email protected]>
Acked-by: Alan Stern <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
Kishon writes:
phy: for 4.14 -rc
*) Handle error return values in rockchip-typec and tegra-xusb
*) Fix MUX error check and ioremap_resource error check in mvebu-cp110-comphy
*) Fix NULL pointer dereference error in phy-mtk-tphy
*) Make sure pipe selector is not set to incompatible value
*) Fix flaky aux channel communication with rockchip-typec PHY
*) Fix DP monitors detection issue in rockchip-typec PHY
Signed-off-by: Kishon Vijay Abraham I <[email protected]>
xhci_stop_device() calls xhci_queue_stop_endpoint() multiple times
without checking the return value. xhci_queue_stop_endpoint() can
return error if the HC is already halted or unable to queue commands.
This can cause a deadlock condition as xhci_stop_device() would
end up waiting indefinitely for a completion for the command that
didn't get queued. Fix this by checking the return value and bailing
out of xhci_stop_device() in case of error. This patch happens to fix
potential memory leaks of the allocated command structures as well.
Fixes: c311e391a7 ("xhci: rework command timeout and cancellation,")
Cc: <[email protected]>
Signed-off-by: Mayank Rana <[email protected]>
Signed-off-by: Jack Pham <[email protected]>
Signed-off-by: Mathias Nyman <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
KASAN reported use-after-free bug when xhci host controller died:
[ 176.952537] BUG: KASAN: use-after-free in xhci_handle_command_timeout+0x68/0x224
[ 176.960846] Write of size 4 at addr ffffffc0cbb01608 by task kworker/3:3/1680
...
[ 177.180644] Freed by task 0:
[ 177.183882] kasan_slab_free+0x90/0x15c
[ 177.188194] kfree+0x114/0x28c
[ 177.191630] xhci_cleanup_command_queue+0xc8/0xf8
[ 177.196916] xhci_hc_died+0x84/0x358
Problem here is that when the cmd_timer fired, it would try to access
current_cmd while the command queue is already freed by xhci_hc_died().
Cleanup current_cmd in xhci_cleanup_command_queue() to avoid that.
Fixes: d9f11ba9f1 ("xhci: Rework how we handle unresponsive or hoptlug removed hosts")
Cc: <[email protected]> # v4.12+
Signed-off-by: Jeffy Chen <[email protected]>
Signed-off-by: Mathias Nyman <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
Many USB 3.1 capable hosts never updated the Serial Bus Release Number
(SBRN) register to USB 3.1 from USB 3.0
xhci driver identified USB 3.1 capable hosts based on this SBRN register,
which according to specs "contains the release of the Universal Serial
Bus Specification with which this Universal Serial Bus Host Controller
module is compliant." but still in october 2017 gives USB 3.0 as
the only possible option.
Make an additional check for USB 3.1 support and enable it if the xHCI
supported protocol capablity lists USB 3.1 capable ports.
Cc: <[email protected]> # v4.6+
Signed-off-by: Mathias Nyman <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
The mount i_version flag is not enabled in the new sb_flags. This patch
adds the missing SB_I_VERSION flag.
Fixes: e462ec5 "VFS: Differentiate mount flags (MS_*) from internal
superblock flags"
Signed-off-by: Mimi Zohar <[email protected]>
Signed-off-by: Al Viro <[email protected]>
- Fix potential use-after-free issue in suspend/resume
by cleanning up drvdata at unbind.
- Fix potential NULL pointer dereference issue in suspend/resume
by setting drm_dev after checking if drm_dev is null or not.
* tag 'exynos-drm-fixes-for-v4.14-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/daeinki/drm-exynos:
drm/exynos: Clear drvdata after component unbind
drm/exynos: Fix potential NULL pointer dereference in suspend/resume paths
HDMI Mode selection on CNL is on CFGCR0 for that PLL, not
on in a global CTRL1 as it was on SKL.
The original patch addressed this difference, but leaving behind
this single entry here. So we were checking the wrong bits during
the PLL initialization and consequently avoiding the CFGCR1 setup
during HDMI initialization. Luckly when only HDMI was in use BIOS
had already setup this for us. But the dual display with hot plug
were messed up.
Fixes: a927c927de ("drm/i915/cnl: Initialize PLLs")
Cc: Paulo Zanoni <[email protected]>
Cc: Manasi Navare <[email protected]>
Cc: Kahola, Mika <[email protected]>
Signed-off-by: Rodrigo Vivi <[email protected]>
Reviewed-by: James Ausmus <[email protected]>
Reviewed-by: Manasi Navare <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
(cherry picked from commit 614ee07acf)
Signed-off-by: Rodrigo Vivi <[email protected]>
On PLL Enable sequence we need to "Configure DPCLKA_CFGCR0 to turn on
the clock for the DDI and map the DPLL to the DDI"
So we first do the map and then we unset DDI_CLK_OFF to turn the clock
on. We do this in 2 separated steps.
However, on this second step where we should only unset the off bit we are
also unmapping the ddi from the pll. So we end up using the pll 0
for almost everything. Consequently breaking cases with more than one
display.
Fixes: 555e38d273 ("drm/i915/cnl: DDI - PLL mapping")
Cc: Paulo Zanoni <[email protected]>
Cc: Manasi Navare <[email protected]>
Cc: Kahola, Mika <[email protected]>
Signed-off-by: Rodrigo Vivi <[email protected]>
Reviewed-by: James Ausmus <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
(cherry picked from commit 87145d95c3)
Signed-off-by: Rodrigo Vivi <[email protected]>
In the full-ppgtt world, we can fill the GGTT full of context objects.
These context objects are currently implicitly tracked by the requests
that pin them i.e. they are only unpinned when the request is completed
and retired, but we do not have the link from the vma to the request
(anymore). In order to unpin those contexts, we have to issue another
request and wait upon the switch to the kernel context.
The bug during eviction was that we assumed that a full GGTT meant we
would have requests on the GGTT timeline, and so we missed situations
where those requests where merely in flight (and when even they have not
yet been submitted to hw yet). The fix employed here is to change the
already-is-idle test to no look at the execution timeline, but count the
outstanding requests and then check that we have switched to the kernel
context. Erring on the side of overkill here just means that we stall a
little longer than may be strictly required, but we only expect to hit
this path in extreme corner cases where returning an erroneous error is
worse than the delay.
v2: Logical inversion when swapping over branches.
Fixes: 80b204bce8 ("drm/i915: Enable multiple timelines")
Signed-off-by: Chris Wilson <[email protected]>
Cc: Tvrtko Ursulin <[email protected]>
Cc: Joonas Lahtinen <[email protected]>
Reviewed-by: Tvrtko Ursulin <[email protected]>
Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
(cherry picked from commit 55b4f1ce2f)
Signed-off-by: Rodrigo Vivi <[email protected]>
Johannes Berg says:
====================
Just a single fix, for a WoWLAN-related part of CVE-2017-13080.
====================
Signed-off-by: David S. Miller <[email protected]>
When changing dev tx_queue_len via netlink or net-sysfs,
a NETDEV_CHANGE_TX_QUEUE_LEN event notification will be
called.
But dev_ioctl missed this event notification, which could
cause no userspace notification would be sent.
Signed-off-by: Xin Long <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
Commit 7091d8c '(net/sched: cls_flower: Add offload support using egress
Hardware device') made sure (when fl_hw_replace_filter is called) to put
the egress_dev mark on persisent structure instance. Hence, following calls
into the HW driver for stats and deletion will note it and act accordingly.
With commit de4784ca03 this property is lost and hence when called,
the HW driver failes to operate (stats, delete) on the offloaded flow.
Fix it by setting the egress_dev flag whenever the ingress device is
different from the hw device since this is exactly the condition under
which we're calling into the HW driver through the egress port net-device.
Fixes: de4784ca03 ('net: sched: get rid of struct tc_to_netdev')
Signed-off-by: Or Gerlitz <[email protected]>
Signed-off-by: Roi Dayan <[email protected]>
Acked-by: Jiri Pirko <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
register_netdevice() could fail early when we have an invalid
dev name, in which case ->ndo_uninit() is not called. For tun
device, this is a problem because a timer etc. are already
initialized and it expects ->ndo_uninit() to clean them up.
We could move these initializations into a ->ndo_init() so
that register_netdevice() knows better, however this is still
complicated due to the logic in tun_detach().
Therefore, I choose to just call dev_get_valid_name() before
register_netdevice(), which is quicker and much easier to audit.
And for this specific case, it is already enough.
Fixes: 96442e4242 ("tuntap: choose the txq based on rxq")
Reported-by: Dmitry Alexeev <[email protected]>
Cc: Jason Wang <[email protected]>
Cc: "Michael S. Tsirkin" <[email protected]>
Signed-off-by: Cong Wang <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
RFC791 specifies the minimum MTU to be 68, while xen-net{front|back}
drivers use a minimum value of 0.
When set MTU to 0~67 with xen_net{front|back} driver, the network
will become unreachable immediately, the guest can no longer be pinged.
xen_net{front|back} should not allow the user to set this value which causes
network problems.
Reported-by: Chen Shi <[email protected]>
Signed-off-by: Mohammed Gamal <[email protected]>
Acked-by: Wei Liu <[email protected]>
Reviewed-by: Boris Ostrovsky <[email protected]>
Signed-off-by: Boris Ostrovsky <[email protected]>
IFLA_IFALIAS is defined as NLA_STRING. It means that the minimal length of
the attribute is 1 ("\0"). However, to remove an alias, the attribute
length must be 0 (see dev_set_alias()).
Let's define the type to NLA_BINARY to allow 0-length string, so that the
alias can be removed.
Example:
$ ip l s dummy0 alias foo
$ ip l l dev dummy0
5: dummy0: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
link/ether ae:20:30:4f:a7:f3 brd ff:ff:ff:ff:ff:ff
alias foo
Before the patch:
$ ip l s dummy0 alias ""
RTNETLINK answers: Numerical result out of range
After the patch:
$ ip l s dummy0 alias ""
$ ip l l dev dummy0
5: dummy0: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
link/ether ae:20:30:4f:a7:f3 brd ff:ff:ff:ff:ff:ff
CC: Oliver Hartkopp <[email protected]>
CC: Stephen Hemminger <[email protected]>
Fixes: 96ca4a2cc1 ("net: remove ifalias on empty given alias")
Reported-by: Julien FLoret <[email protected]>
Signed-off-by: Nicolas Dichtel <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
Xin Long says:
====================
rtnetlink: a bunch of fixes for userspace notifications in changing dev properties
Whenever any property of a link, address, route, etc. changes by whatever way,
kernel should notify the programs that listen for such events in userspace.
The patchet "rtnetlink: Cleanup user notifications for netdev events" tried to
fix a redundant notifications issue, but it also introduced a side effect.
After that, user notifications could only be sent when changing dev properties
via netlink api. As it removed some events process in rtnetlink_event where
the notifications was sent to users.
It resulted in no notification generated when dev properties are changed via
other ways, like ioctl, sysfs, etc. It may cause some user programs doesn't
work as expected because of the missing notifications.
This patchset will fix it by bringing some of these netdev events back and
also fix the old redundant notifications issue with a proper way.
====================
Signed-off-by: David S. Miller <[email protected]>
NETDEV_CHANGE_TX_QUEUE_LEN event process in rtnetlink_event would
send a notification for userspace and tx_queue_len's setting in
do_setlink would trigger NETDEV_CHANGE_TX_QUEUE_LEN.
So it shouldn't set DO_SETLINK_NOTIFY status for this change to
send a notification any more.
Signed-off-by: Xin Long <[email protected]>
Acked-by: David Ahern <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
The check 'status & DO_SETLINK_NOTIFY' in do_setlink doesn't really
work after status & DO_SETLINK_MODIFIED, as:
DO_SETLINK_MODIFIED 0x1
DO_SETLINK_NOTIFY 0x3
Considering that notifications are suppposed to be sent only when
status have the flag DO_SETLINK_NOTIFY, the right check would be:
(status & DO_SETLINK_NOTIFY) == DO_SETLINK_NOTIFY
This would avoid lots of duplicated notifications when setting some
properties of a link.
Fixes: ba9989069f ("rtnl/do_setlink(): notify when a netdev is modified")
Signed-off-by: Xin Long <[email protected]>
Acked-by: David Ahern <[email protected]>
Acked-by: Nicolas Dichtel <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
Eric Biggers