It turns out that the current version of gfs2_metadata_walker suffers
from multiple problems that can cause gfs2_hole_size to report an
incorrect size. This will confuse fiemap as well as lseek with the
SEEK_DATA flag.
Fix that by changing gfs2_hole_walker to compute the metapath to the
first data block after the hole (if any), and compute the hole size
based on that.
Fixes xfstest generic/490.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Reviewed-by: Bob Peterson <rpeterso@redhat.com> Cc: stable@vger.kernel.org # v4.18+ Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Since commit c66d4bd110a1f8 ("genirq/affinity: Add new callback for
(re)calculating interrupt sets"), irq_create_affinity_masks() returns
NULL in case of single vector. This change has caused regression on some
drivers, such as lpfc.
The problem is that single vector requests can happen in some generic cases:
1) kdump kernel
2) irq vectors resource is close to exhaustion.
If in that situation the affinity mask for a single vector is not created,
every caller has to handle the special case.
There is no reason why the mask cannot be created, so remove the check for
a single vector and create the mask.
Fixes: c66d4bd110a1f8 ("genirq/affinity: Add new callback for (re)calculating interrupt sets") Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: stable@vger.kernel.org Link: https://lkml.kernel.org/r/20190805011906.5020-1-ming.lei@redhat.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
KBUILD_CFLAGS is very carefully built up in the top level Makefile,
particularly when cross compiling or using different build tools.
Resetting KBUILD_CFLAGS via := assignment is an antipattern.
The comment above the reset mentions that -pg is problematic. Other
Makefiles use `CFLAGS_REMOVE_file.o = $(CC_FLAGS_FTRACE)` when
CONFIG_FUNCTION_TRACER is set. Prefer that pattern to wiping out all of
the important KBUILD_CFLAGS then manually having to re-add them. Seems
also that __stack_chk_fail references are generated when using
CONFIG_STACKPROTECTOR or CONFIG_STACKPROTECTOR_STRONG.
Fixes: 8fc5b4d4121c ("purgatory: core purgatory functionality") Reported-by: Vaibhav Rustagi <vaibhavrustagi@google.com> Suggested-by: Peter Zijlstra <peterz@infradead.org> Suggested-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Nick Desaulniers <ndesaulniers@google.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Tested-by: Vaibhav Rustagi <vaibhavrustagi@google.com> Cc: stable@vger.kernel.org Link: https://lkml.kernel.org/r/20190807221539.94583-2-ndesaulniers@google.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Implementing memcpy and memset in terms of __builtin_memcpy and
__builtin_memset is problematic.
GCC at -O2 will replace calls to the builtins with calls to memcpy and
memset (but will generate an inline implementation at -Os). Clang will
replace the builtins with these calls regardless of optimization level.
$ llvm-objdump -dr arch/x86/purgatory/string.o | tail
Such code results in infinite recursion at runtime. This is observed
when doing kexec.
Instead, reuse an implementation from arch/x86/boot/compressed/string.c.
This requires to implement a stub function for warn(). Also, Clang may
lower memcmp's that compare against 0 to bcmp's, so add a small definition,
too. See also: commit 5f074f3e192f ("lib/string.c: implement a basic bcmp")
There is an offset of 0x1990 bytes. The size of the qeth module is
151552 bytes (0x25000 in hex).
The location of the GOT/relocation table at the beginning of a module is
unique to s390.
commit 203d8a4aa6ed ("perf s390: Fix 'start' address of module's map")
adjusts the start address of a module in the map structures, but does
not adjust the size of the modules. This leads to overlapping of module
maps as this example shows:
The module qeth.ko has an adjusted start address modified to b3990, but
its size is unchanged and the module ends at 0x3ff800d8990. This end
address overlaps with the next modules start address of 0x3ff800d85a0.
When the size of the leading GOT/Relocation table stored in the
beginning of the text segment (0x1990 bytes) is subtracted from module
qeth end address, there are no overlaps anymore:
0x3ff800d8990 - 0x1990 = 0x0x3ff800d7000
which is the same as
0x3ff800b2000 + 0x25000 = 0x0x3ff800d7000.
To fix this issue, also adjust the modules size in function
arch__fix_module_text_start(). Add another function parameter named size
and reduce the size of the module when the text segment start address is
changed.
Threads synthesized from /proc have comms with a start time of zero, and
not marked as "exec". Currently, there can be 2 such comms. The first is
created by processing a synthesized fork event and is set to the
parent's comm string, and the second by processing a synthesized comm
event set to the thread's current comm string.
In the absence of an "exec" comm, thread__exec_comm() picks the last
(oldest) comm, which, in the case above, is the parent's comm string.
For a main thread, that is very probably wrong. Use the second-to-last
in that case.
This affects only db-export because it is the only user of
thread__exec_comm().
Example:
$ sudo perf record -a -o pt-a-sleep-1 -e intel_pt//u -- sleep 1
$ sudo chown ahunter pt-a-sleep-1
On s390 the kernel is located around memory address 0x200, 0x10000 or
0x100000, depending on linux version. Modules however start some- where
around 0x3ff xxxx xxxx.
This is different than x86 and produces a large gap for which histogram
allocation fails.
Fix this by detecting the kernel's last symbol and do no adjustment for
it. Introduce a weak function and handle s390 specifics.
Reported-by: Klaus Theurich <klaus.theurich@de.ibm.com> Signed-off-by: Thomas Richter <tmricht@linux.ibm.com> Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Hendrik Brueckner <brueckner@linux.ibm.com> Cc: Vasily Gorbik <gor@linux.ibm.com> Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/20190724122703.3996-2-tmricht@linux.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On x86-32 with PTI enabled, parts of the kernel page-tables are not shared
between processes. This can cause mappings in the vmalloc/ioremap area to
persist in some page-tables after the region is unmapped and released.
When the region is re-used the processes with the old mappings do not fault
in the new mappings but still access the old ones.
This causes undefined behavior, in reality often data corruption, kernel
oopses and panics and even spontaneous reboots.
Fix this problem by activly syncing unmaps in the vmalloc/ioremap area to
all page-tables in the system before the regions can be re-used.
References: https://bugzilla.suse.com/show_bug.cgi?id=1118689 Fixes: 5d72b4fba40ef ('x86, mm: support huge I/O mapping capability I/F') Signed-off-by: Joerg Roedel <jroedel@suse.de> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Dave Hansen <dave.hansen@linux.intel.com> Link: https://lkml.kernel.org/r/20190719184652.11391-4-joro@8bytes.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
With huge-page ioremap areas the unmappings also need to be synced between
all page-tables. Otherwise it can cause data corruption when a region is
unmapped and later re-used.
Make the vmalloc_sync_one() function ready to sync unmappings and make sure
vmalloc_sync_all() iterates over all page-tables even when an unmapped PMD
is found.
Fixes: 5d72b4fba40ef ('x86, mm: support huge I/O mapping capability I/F') Signed-off-by: Joerg Roedel <jroedel@suse.de> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Dave Hansen <dave.hansen@linux.intel.com> Link: https://lkml.kernel.org/r/20190719184652.11391-3-joro@8bytes.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Do not require a struct page for the mapped memory location because it
might not exist. This can happen when an ioremapped region is mapped with
2MB pages.
Fixes: 5d72b4fba40ef ('x86, mm: support huge I/O mapping capability I/F') Signed-off-by: Joerg Roedel <jroedel@suse.de> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Dave Hansen <dave.hansen@linux.intel.com> Link: https://lkml.kernel.org/r/20190719184652.11391-2-joro@8bytes.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The 2016 kabylake HP Spectre X360 (model number 13-w013dx) works much better
with psmouse.synaptics_intertouch=1 kernel parameter, so let's enable RMI4
mode automatically.
There are some new HP laptops with Elantech touchpad that don't support
multitouch.
Currently we use ETP_NEW_IC_SMBUS_HOST_NOTIFY() to check if SMBus is supported,
but in addition to firmware version, the bus type also informs us whether the IC
can support SMBus. To avoid breaking old ICs, we will only enable SMbus support
based the bus type on systems manufactured after 2018.
Lastly, let's consolidate all checks into elantech_use_host_notify() and use it
to determine whether to use PS/2 or SMBus.
Commit 89e524c04fa9 ("loop: Fix mount(2) failure due to race with
LOOP_SET_FD") converted blkdev_get() to use the new helpers for
finishing claiming of a block device. However the conversion botched the
error handling in blkdev_get() and thus the bdev has been marked as held
even in case __blkdev_get() returned error. This led to occasional
warnings with block/001 test from blktests like:
kernel: WARNING: CPU: 5 PID: 907 at fs/block_dev.c:1899 __blkdev_put+0x396/0x3a0
Correct the error handling.
CC: stable@vger.kernel.org Fixes: 89e524c04fa9 ("loop: Fix mount(2) failure due to race with LOOP_SET_FD") Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@kernel.dk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We have set the mmc_host.max_seg_size to 8M, but the dma max segment
size of PCI device is set to 64K by default in function pci_device_add().
The mmc_host.max_seg_size is used to set the max segment size of
the blk queue. Then this mismatch will trigger a calltrace like below
when a bigger than 64K segment request arrives at mmc dev. So we should
consider the limitation of the cvm_mmc_host when setting the
mmc_host.max_seg_size.
DMA-API: thunderx_mmc 0000:01:01.4: mapping sg segment longer than device claims to support [len=131072] [max=65536]
WARNING: CPU: 6 PID: 238 at kernel/dma/debug.c:1221 debug_dma_map_sg+0x2b8/0x350
Modules linked in:
CPU: 6 PID: 238 Comm: kworker/6:1H Not tainted 5.3.0-rc1-next-20190724-yocto-standard+ #62
Hardware name: Marvell OcteonTX CN96XX board (DT)
Workqueue: kblockd blk_mq_run_work_fn
pstate: 80c00009 (Nzcv daif +PAN +UAO)
pc : debug_dma_map_sg+0x2b8/0x350
lr : debug_dma_map_sg+0x2b8/0x350
sp : ffff00001770f9e0
x29: ffff00001770f9e0 x28: ffffffff00000000
x27: 00000000ffffffff x26: ffff800bc2c73180
x25: ffff000010e83700 x24: 0000000000000002
x23: 0000000000000001 x22: 0000000000000001
x21: 0000000000000000 x20: ffff800bc48ba0b0
x19: ffff800bc97e8c00 x18: ffffffffffffffff
x17: 0000000000000000 x16: 0000000000000000
x15: ffff000010e835c8 x14: 6874207265676e6f
x13: 6c20746e656d6765 x12: 7320677320676e69
x11: 7070616d203a342e x10: 31303a31303a3030
x9 : 303020636d6d5f78 x8 : 35363d78616d5b20
x7 : 00000000000002fd x6 : ffff000010fd57dc
x5 : 0000000000000000 x4 : ffff0000106c61f0
x3 : 00000000ffffffff x2 : 0000800bee060000
x1 : 7010678df3041a00 x0 : 0000000000000000
Call trace:
debug_dma_map_sg+0x2b8/0x350
cvm_mmc_request+0x3c4/0x988
__mmc_start_request+0x9c/0x1f8
mmc_start_request+0x7c/0xb0
mmc_blk_mq_issue_rq+0x5c4/0x7b8
mmc_mq_queue_rq+0x11c/0x278
blk_mq_dispatch_rq_list+0xb0/0x568
blk_mq_do_dispatch_sched+0x6c/0x108
blk_mq_sched_dispatch_requests+0x110/0x1b8
__blk_mq_run_hw_queue+0xb0/0x118
blk_mq_run_work_fn+0x28/0x38
process_one_work+0x210/0x490
worker_thread+0x48/0x458
kthread+0x130/0x138
ret_from_fork+0x10/0x1c
In sound_insert_unit(), the controlling structure 's' is allocated through
kmalloc(). Then it is added to the sound driver list by invoking
__sound_insert_unit(). Later on, if __register_chrdev() fails, 's' is
removed from the list through __sound_remove_unit(). If 'index' is not less
than 0, -EBUSY is returned to indicate the error. However, 's' is not
deallocated on this execution path, leading to a memory leak bug.
To fix the above issue, free 's' before -EBUSY is returned.
We have to drop the mutex before we close() upon disconnect()
as close() needs the lock. This is safe to do by dropping the
mutex as intfdata is already set to NULL, so open() will fail.
Fixes: 03f36e885fc26 ("USB: open disconnect race in iowarrior") Reported-by: syzbot+a64a382964bf6c71a9c0@syzkaller.appspotmail.com Cc: stable <stable@vger.kernel.org> Signed-off-by: Oliver Neukum <oneukum@suse.com> Link: https://lore.kernel.org/r/20190808092728.23417-1-oneukum@suse.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Upon an error within proc_do_submiturb(), dec_usb_memory_use_count()
gets called once by the error handling tail and again by free_async().
Remove the first call.
Commit daaef255dc96 ("driver: platform: Support parsing GpioInt 0 in
platform_get_irq()") broke the Embedded Controller driver on most LPC
Chromebooks (i.e., most x86 Chromebooks), because cros_ec_lpc expects
platform_get_irq() to return -ENXIO for non-existent IRQs.
Unfortunately, acpi_dev_gpio_irq_get() doesn't follow this convention
and returns -ENOENT instead. So we get this error from cros_ec_lpc:
couldn't retrieve IRQ number (-2)
I see a variety of drivers that treat -ENXIO specially, so rather than
fix all of them, let's fix up the API to restore its previous behavior.
and the result is that the bug landed and remains unfixed.
I differ from the v3 patch by:
* allowing for ret==0, even though acpi_dev_gpio_irq_get() specifically
documents (and enforces) that 0 is not a valid return value (noted on
the v3 review)
* adding a small comment
Reported-by: Brian Norris <briannorris@chromium.org> Reported-by: Salvatore Bellizzi <salvatore.bellizzi@linux.seppia.net> Cc: Enrico Granata <egranata@chromium.org> Cc: <stable@vger.kernel.org> Fixes: daaef255dc96 ("driver: platform: Support parsing GpioInt 0 in platform_get_irq()") Signed-off-by: Brian Norris <briannorris@chromium.org> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Enrico Granata <egranata@google.com> Link: https://lore.kernel.org/r/20190729204954.25510-1-briannorris@chromium.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
AES GCM input buffers for decryption contain AAD+CTEXT+TAG. Only
decrypt the ciphertext, and use the tag for comparison.
Fixes: 36cf515b9bbe2 ("crypto: ccp - Enable support for AES GCM on v5 CCPs") Cc: <stable@vger.kernel.org> Signed-off-by: Gary R Hook <gary.hook@amd.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
AES GCM encryption allows for authsize values of 4, 8, and 12-16 bytes.
Validate the requested authsize, and retain it to save in the request
context.
Fixes: 36cf515b9bbe2 ("crypto: ccp - Enable support for AES GCM on v5 CCPs") Cc: <stable@vger.kernel.org> Signed-off-by: Gary R Hook <gary.hook@amd.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A plaintext or ciphertext length of 0 is allowed in AES, in which case
no encryption occurs. Ensure that we don't clean up data structures
that were never allocated.
Fixes: 36cf515b9bbe2 ("crypto: ccp - Enable support for AES GCM on v5 CCPs") Cc: <stable@vger.kernel.org> Signed-off-by: Gary R Hook <gary.hook@amd.com> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Typically gpiod_set_value calls would assert the reset line and
then release it using the symantics of:
gpiod_set_value(par->gpio.reset, 0);
... delay
gpiod_set_value(par->gpio.reset, 1);
And the gpio binding would specify the polarity.
Prior to conversion to gpiod calls the polarity in the DT
was ignored and assumed to be active low. Fix it so that
DT polarity is respected.
Fixes: c440eee1a7a1 ("Staging: fbtft: Switch to the gpio descriptor interface") Reviewed-by: Nicolas Saenz Julienne <nsaenzjulienne@suse.de> Tested-by: Nicolas Saenz Julienne <nsaenzjulienne@suse.de> Tested-by: Jan Sebastian Götte <linux@jaseg.net> Signed-off-by: Phil Reid <preid@electromag.com.au> Cc: stable <stable@vger.kernel.org> Link: https://lore.kernel.org/r/1563236677-5045-3-git-send-email-preid@electromag.com.au Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Conversion to use gpio descriptors broke all gpio lookups as
devm_gpiod_get_index was converted to use dev->driver->name for
the gpio name lookup. Fix this by using the name param. In
addition gpiod_get post-fixes the -gpios to the name so that
shouldn't be included in the call. However this then breaks the
of_find_property call to see if the gpio entry exists as all
fbtft treats all gpios as optional. So use devm_gpiod_get_index_optional
instead which achieves the same thing and is simpler.
Nishad confirmed the changes where only ever compile tested.
Fixes: c440eee1a7a1 ("Staging: fbtft: Switch to the gpio descriptor interface") Reviewed-by: Nicolas Saenz Julienne <nsaenzjulienne@suse.de> Tested-by: Nicolas Saenz Julienne <nsaenzjulienne@suse.de> Tested-by: Jan Sebastian Götte <linux@jaseg.net> Signed-off-by: Phil Reid <preid@electromag.com.au> Cc: stable <stable@vger.kernel.org> Link: https://lore.kernel.org/r/1563236677-5045-2-git-send-email-preid@electromag.com.au Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
syzbot found that a thread can stall for minutes inside
ion_system_heap_allocate() after that thread was killed by SIGKILL [1].
Let's check for SIGKILL before doing memory allocation.
In sysfs_show() case-branches ATTR_KERNEL_HIB_PAGE_TABLE_SIZE and
ATTR_KERNEL_HIB_SIMPLE_PAGE_TABLE_SIZE do the same. It looks like
copy-paste mistake.
Driver only supports 3-axis gyro and/or 3-axis accel.
For icm20602, temp data is mandatory for all configurations.
Fix all single and double axis configurations (almost never used) and more
importantly fix 3-axis gyro and 6-axis accel+gyro buffer on icm20602 when
temp data is not enabled.
The SADC component can run at up to 8 MHz on JZ4725B, but is fed
a 12 MHz input clock (EXT). Divide it by two to get 6 MHz, then
set up another divider to match, to produce a 10us clock.
If the clock dividers are left on their power-on defaults (a divider
of 1), the SADC mostly works, but will occasionally produce erroneous
readings. This led to button presses being detected out of nowhere on
the RS90 every few minutes. With this change, no ghost button presses
were logged in almost a day worth of testing.
The ADCLK register for configuring clock dividers doesn't exist on
JZ4740, so avoid writing it there.
A function has been introduced rather than a flag because there is a lot
of variation between the ADCLK registers on JZ47xx SoCs, both in
the internal layout of the register and in the frequency range
supported by the SADC. So this solution should make it easier
to add support for other JZ47xx SoCs later.
Fixes: 1a78daea107d ("iio: adc: probe should set clock divider") Signed-off-by: Maarten ter Huurne <maarten@treewalker.org> Signed-off-by: Artur Rojek <contact@artur-rojek.eu> Cc: <Stable@vger.kernel.org> Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit c2bf1fc212f7 ("PCI: Add missing link delays required by the PCIe
spec") turned out causing issues with some systems either by making them
unresponsive or slowing down runtime and system wide resume of PCIe
devices. While root cause for the unresponsiveness is still under
investigation given the amount of issues reported better to revert it
for now.
Commit 6935224da248 ("spi: bcm2835: enable support of 3-wire mode")
added 3-wire support to the BCM2835 SPI driver by setting the REN bit
(Read Enable) in the CS register when receiving data. The REN bit puts
the transmitter in high-impedance state. The driver recognizes that
data is to be received by checking whether the rx_buf of a transfer is
non-NULL.
Commit 3ecd37edaa2a ("spi: bcm2835: enable dma modes for transfers
meeting certain conditions") subsequently broke 3-wire support because
it set the SPI_MASTER_MUST_RX flag which causes spi_map_msg() to replace
rx_buf with a dummy buffer if it is NULL. As a result, rx_buf is
*always* non-NULL if DMA is enabled.
Reinstate 3-wire support by not only checking whether rx_buf is non-NULL,
but also checking that it is not the dummy buffer.
Fixes: 3ecd37edaa2a ("spi: bcm2835: enable dma modes for transfers meeting certain conditions") Reported-by: Nuno Sá <nuno.sa@analog.com> Signed-off-by: Lukas Wunner <lukas@wunner.de> Cc: stable@vger.kernel.org # v4.2+ Cc: Martin Sperl <kernel@martin.sperl.org> Acked-by: Stefan Wahren <wahrenst@gmx.net> Link: https://lore.kernel.org/r/328318841455e505370ef8ecad97b646c033dc8a.1562148527.git.lukas@wunner.de Signed-off-by: Mark Brown <broonie@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Revert this for now, it has been reported multiple times that it
completely breaks connectivity on various devices.
Cc: stable@vger.kernel.org Fixes: 8dbb000ee73b ("mac80211: set NETIF_F_LLTX when using intermediate tx queues") Reported-by: Jean Delvare <jdelvare@suse.de> Reported-by: Peter Lebbing <peter@digitalbrains.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A single 32-bit PSR2 training pattern field follows the sixteen element
array of PSR table entries in the VBT spec. But, we incorrectly define
this PSR2 field for each of the PSR table entries. As a result, the PSR1
training pattern duration for any panel_type != 0 will be parsed
incorrectly. Secondly, PSR2 training pattern durations for VBTs with bdb
version >= 226 will also be wrong.
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Cc: stable@vger.kernel.org Cc: stable@vger.kernel.org #v5.2 Fixes: 88a0d9606aff ("drm/i915/vbt: Parse and use the new field with PSR2 TP2/3 wakeup time")
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=111088
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=204183 Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Reviewed-by: José Roberto de Souza <jose.souza@intel.com> Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Tested-by: François Guerraz <kubrick@fgv6.net> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20190717223451.2595-1-dhinakaran.pandiyan@intel.com
(cherry picked from commit b5ea9c9337007d6e700280c8a60b4e10d070fb53) Signed-off-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Support for handling the PPPOEIOCSFWD ioctl in compat mode was added in
linux-2.5.69 along with hundreds of other commands, but was always broken
sincen only the structure is compatible, but the command number is not,
due to the size being sizeof(size_t), or at first sizeof(sizeof((struct
sockaddr_pppox)), which is different on 64-bit architectures.
Guillaume Nault adds:
And the implementation was broken until 2016 (see 29e73269aa4d ("pppoe:
fix reference counting in PPPoE proxy")), and nobody ever noticed. I
should probably have removed this ioctl entirely instead of fixing it.
Clearly, it has never been used.
Fix it by adding a compat_ioctl handler for all pppoe variants that
translates the command number and then calls the regular ioctl function.
All other ioctl commands handled by pppoe are compatible between 32-bit
and 64-bit, and require compat_ptr() conversion.
This should apply to all stable kernels.
Acked-by: Guillaume Nault <g.nault@alphalink.fr> Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Speed translation is performed based on legacy or extended PTYS
register. Translate speed with respect to:
1) Capability bit of extended PTYS table.
2) User request:
a) When auto-negotiation is turned on, inspect advertisement whether it
contains extended link modes.
b) When auto-negotiation is turned off, speed > 100Gbps (maximal
speed supported in legacy mode).
With both conditions fulfilled translation is done with extended PTYS
table otherwise use legacy PTYS table.
Without this patch 25/50/100 Gbps speed cannot be set, since try to
configure in extended mode but read from legacy mode.
New flow table type RDMA_RX was added but the MLX5_CAP_FLOW_TABLE_TYPE
didn't handle this new flow table type.
This means that MLX5_CAP_FLOW_TABLE_TYPE returns an empty capability to
this flow table type.
Update both the macro and the maximum supported flow table type to
RDMA_RX.
Fixes: d83eb50e29de ("net/mlx5: Add support in RDMA RX steering") Signed-off-by: Maor Gottlieb <maorg@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In commit e891ce1dd2a5 ("mlxsw: spectrum_buffers: Reduce pool size on
Spectrum-2"), pool size was reduced to mitigate a problem in port buffer
usage of ports split four ways. It turns out that this work around does not
solve the issue, and a further reduction is required.
Thus reduce the size of pool 0 by another 2.7 MiB, and round down to the
whole number of cells.
Fixes: e891ce1dd2a5 ("mlxsw: spectrum_buffers: Reduce pool size on Spectrum-2") Signed-off-by: Petr Machata <petrm@mellanox.com> Signed-off-by: Ido Schimmel <idosch@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently there are two error return paths that leak memory allocated
to fib_work. Fix this by kfree'ing fib_work before returning.
Addresses-Coverity: ("Resource leak") Fixes: 19a9d136f198 ("ipv4: Flag fib_info with a fib_nh using IPv6 gateway") Fixes: dbcc4fa718ee ("rocker: Fail attempts to use routes with nexthop objects") Signed-off-by: Colin Ian King <colin.king@canonical.com> Reviewed-by: David Ahern <dsahern@gmail.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
FASTOPEN is not possible with SMC. sendmsg() with msg_flag MSG_FASTOPEN
triggers a fallback to TCP if the socket is in state SMC_INIT.
But if a nonblocking connect is already started, fallback to TCP
is no longer possible, even though the socket may still be in state
SMC_INIT.
And if a nonblocking connect is already started, a listen() call
does not make sense.
Reported-by: syzbot+bd8cc73d665590a1fcad@syzkaller.appspotmail.com Fixes: 50717a37db032 ("net/smc: nonblocking connect rework") Signed-off-by: Ursula Braun <ubraun@linux.ibm.com> Signed-off-by: Karsten Graul <kgraul@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In phy_start_aneg() autoneg is started, and immediately after that
link and autoneg status are read. As reported in [0] it can happen that
at time of this read the PHY has reset the "aneg complete" bit but not
yet the "link up" bit, what can result in a false link-up detection.
To fix this don't report link as up if we're in aneg mode and PHY
doesn't signal "aneg complete".
There is a race condition for an established connection that is being closed
by the guest: the refcnt is 4 at the end of hvs_release() (Note: here the
'remove_sock' is false):
1 for the initial value;
1 for the sk being in the bound list;
1 for the sk being in the connected list;
1 for the delayed close_work.
After hvs_release() finishes, __vsock_release() -> sock_put(sk) *may*
decrease the refcnt to 3.
Concurrently, hvs_close_connection() runs in another thread:
calls vsock_remove_sock() to decrease the refcnt by 2;
call sock_put() to decrease the refcnt to 0, and free the sk;
next, the "release_sock(sk)" may hang due to use-after-free.
In the above, after hvs_release() finishes, if hvs_close_connection() runs
faster than "__vsock_release() -> sock_put(sk)", then there is not any issue,
because at the beginning of hvs_close_connection(), the refcnt is still 4.
The issue can be resolved if an extra reference is taken when the
connection is established.
Fixes: a9eeb998c28d ("hv_sock: Add support for delayed close") Signed-off-by: Dexuan Cui <decui@microsoft.com> Reviewed-by: Sunil Muthuswamy <sunilmut@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When generic-XDP was moved to a later processing step by commit 458bf2f224f0 ("net: core: support XDP generic on stacked devices.")
a regression was introduced when using bpf_xdp_adjust_head.
The issue is that after this commit the skb->network_header is now
changed prior to calling generic XDP and not after. Thus, if the header
is changed by XDP (via bpf_xdp_adjust_head), then skb->network_header
also need to be updated again. Fix by calling skb_reset_network_header().
Fixes: 458bf2f224f0 ("net: core: support XDP generic on stacked devices.") Reported-by: Brandon Cazander <brandon.cazander@multapplied.net> Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Given the increasing number of BPF selftests, it makes sense to
reduce the time to execute these tests. The ping parameters are
adjusted to reduce the time from measures 9 sec to approx 2.8 sec.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In-order to test both native-XDP (xdpdrv) and generic-XDP (xdpgeneric)
create two wrapper test scripts, that start the test_xdp_vlan.sh script
with these modes.
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Change BPF selftest test_xdp_vlan.sh to (default) use generic XDP.
This selftest was created together with a fix for generic XDP, in commit 297249569932 ("net: fix generic XDP to handle if eth header was
mangled"). And was suppose to catch if generic XDP was broken again.
The tests are using veth and assumed that veth driver didn't support
native driver XDP, thus it used the (ip link set) 'xdp' attach that fell
back to generic-XDP. But veth gained native-XDP support in 948d4f214fde
("veth: Add driver XDP"), which caused this test script to use
native-XDP.
Fixes: 948d4f214fde ("veth: Add driver XDP") Fixes: 97396ff0bc2d ("selftests/bpf: add XDP selftests for modifying and popping VLAN headers") Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It was reported that after resuming from suspend network fails with
error "do_IRQ: 3.38 No irq handler for vector", see [0]. Enabling WoL
can work around the issue, but the only actual fix is to disable MSI.
So let's mimic the behavior of the vendor driver and disable MSI on
all chip versions before RTL8168d.
This patch prevents a race between user invoked cached counters
query and a neighbor last usage updater.
The cached flow counter stats can be queried by calling
"mlx5_fc_query_cached" which provides the number of bytes and
packets that passed via this flow since the last time this counter
was queried.
It does so by reducting the last saved stats from the current, cached
stats and then updating the last saved stats with the cached stats.
It also provide the lastuse value for that flow.
Since "mlx5e_tc_update_neigh_used_value" needs to retrieve the
last usage time of encapsulation flows, it calls the flow counter
query method periodically and async to user queries of the flow counter
using cls_flower.
This call is causing the driver to update the last reported bytes and
packets from the cache and therefore, future user queries of the flow
stats will return lower than expected number for bytes and packets
since the last saved stats in the driver was updated async to the last
saved stats in cls_flower.
This causes wrong stats presentation of encapsulation flows to user.
Since the neighbor usage updater only needs the lastuse stats from the
cached counter, the fix is to use a dedicated lastuse query call that
returns the lastuse value without synching between the cached stats and
the last saved stats.
Fixes: f6dfb4c3f216 ("net/mlx5e: Update neighbour 'used' state using HW flow rules counters") Signed-off-by: Ariel Levkovich <lariel@mellanox.com> Reviewed-by: Roi Dayan <roid@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
- v1 -> v2: Move skb_set_owner_w to __tun_build_skb to reduce patch size
Small packets going out of a tap device go through an optimized code
path that uses build_skb() rather than sock_alloc_send_pskb(). The
latter calls skb_set_owner_w(), but the small packet code path does not.
The net effect is that small packets are not owned by the userland
application's socket (e.g. QEMU), while large packets are.
This can be seen with a TCP session, where packets are not owned when
the window size is small enough (around PAGE_SIZE), while they are once
the window grows (note that this requires the host to support virtio
tso for the guest to offload segmentation).
All this leads to inconsistent behaviour in the kernel, especially on
netfilter modules that uses sk->socket (e.g. xt_owner).
Fixes: 66ccbc9c87c2 ("tap: use build_skb() for small packet") Signed-off-by: Alexis Bauvin <abauvin@scaleway.com> Acked-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The reason is that the skb list tipc_socket::mc_method.deferredq only
is initialized for connectionless sockets, while nothing stops arriving
multicast messages from being filtered by connection oriented sockets,
with subsequent access to the said list.
We fix this by initializing the list unconditionally at socket creation.
This eliminates the crash, while the message still is dropped further
down in tipc_sk_filter_rcv() as it should be.
Reported-by: Li Shuang <shuali@redhat.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Reviewed-by: Xin Long <lucien.xin@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 2753ca5d9009 ("tipc: fix uninit-value in tipc_nl_compat_doit")
broke older tipc tools that use compat interface (e.g. tipc-config from
tipcutils package):
% tipc-config -p
operation not supported
The commit started to reject TIPC netlink compat messages that do not
have attributes. It is too restrictive because some of such messages are
valid (they don't need any arguments):
Fix two reset-gpio sanity checks which were never converted to use
gpio_is_valid(), and make sure to use -EINVAL to indicate a missing
reset line also for the UART-driver module parameter and for the USB
driver.
This specifically prevents the UART and USB drivers from incidentally
trying to request and use gpio 0, and also avoids triggering a WARN() in
gpio_to_desc() during probe when no valid reset line has been specified.
The setsockopts options TCP_NODELAY and TCP_CORK may schedule the
tx worker. Make sure the socket is not yet moved into SMC_CLOSED
state (for instance by a shutdown SHUT_RDWR call).
Reported-by: syzbot+92209502e7aab127c75f@syzkaller.appspotmail.com Reported-by: syzbot+b972214bb803a343f4fe@syzkaller.appspotmail.com Fixes: 01d2f7e2cdd31 ("net/smc: sockopts TCP_NODELAY and TCP_CORK") Signed-off-by: Ursula Braun <ubraun@linux.ibm.com> Signed-off-by: Karsten Graul <kgraul@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently init call of all actions (except ipt) init their 'parm'
structure as a direct pointer to nla data in skb. This leads to race
condition when some of the filter actions were initialized successfully
(and were assigned with idr action index that was written directly
into nla data), but then were deleted and retried (due to following
action module missing or classifier-initiated retry), in which case
action init code tries to insert action to idr with index that was
assigned on previous iteration. During retry the index can be reused
by another action that was inserted concurrently, which causes
unintended action sharing between filters.
To fix described race condition, save action idr index to temporary
stack-allocated variable instead on nla data.
Fixes: 0190c1d452a9 ("net: sched: atomically check-allocate action") Signed-off-by: Dmytro Linkin <dmitrolin@mellanox.com> Signed-off-by: Vlad Buslov <vladbu@mellanox.com> Acked-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add get_fill_size() routine used to calculate the action size
when building a batch of events.
Fixes: c7e2b9689 ("sched: introduce vlan action") Signed-off-by: Roman Mashak <mrv@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The udp_ip4_ind bit is set only for IPv4 UDP non-fragmented packets
so that the hardware can flip the checksum to 0xFFFF if the computed
checksum is 0 per RFC768.
However, this bit had to be set for IPv6 UDP non fragmented packets
as well per hardware requirements. Otherwise, IPv6 UDP packets
with computed checksum as 0 were transmitted by hardware and were
dropped in the network.
In addition to setting this bit for IPv6 UDP, the field is also
appropriately renamed to udp_ind as part of this change.
Fixes: 5eb5f8608ef1 ("net: qualcomm: rmnet: Add support for TX checksum offload") Cc: Sean Tranchetti <stranche@codeaurora.org> Signed-off-by: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In phylink_parse_fixedlink() the pl->link_config.advertising bits are AND
with pl->supported, pl->supported is zeroed and only the speed/duplex
modes and MII bits are set.
So pl->link_config.advertising always loses the flow control/pause bits.
By setting Pause and Asym_Pause bits in pl->supported, the flow control
work again when devicetree "pause" is set in fixes-link node and the MAC
advertise that is supports pause.
Results with this patch.
Legend:
- DT = 'Pause' is set in the fixed-link in devicetree.
- validate() = ‘Yes’ means phylink_set(mask, Pause) is set in the
validate().
- flow = results reported my link is Up line.
+-----+------------+-------+
| DT | validate() | flow |
+-----+------------+-------+
| Yes | Yes | rx/tx |
| No | Yes | off |
| Yes | No | off |
+-----+------------+-------+
Fixes: 9525ae83959b ("phylink: add phylink infrastructure") Signed-off-by: René van Dorst <opensource@vdorst.com> Acked-by: Russell King <rmk+kernel@armlinux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
SFP modules connected using the SGMII interface have their own PHYs which
are handled by the struct phylink's phydev field. On the other hand, for
the modules connected using 1000Base-X interface that field is not set.
Since commit ce0aa27ff3f6 ("sfp: add sfp-bus to bridge between network
devices and sfp cages") phylink_start() ends up setting the phydev field
using the sfp-bus infrastructure, which eventually calls phy_start() on it,
and then calling phy_start() again on the same phydev from phylink_start()
itself. Similar call sequence holds for phylink_stop(), only in the reverse
order. This results in WARNs during network interface bringup and shutdown
when a copper SFP module is connected, as phy_start() and phy_stop() are
called twice in a row for the same phy_device:
% ip link set up dev eth0
------------[ cut here ]------------
called from state UP
WARNING: CPU: 1 PID: 155 at drivers/net/phy/phy.c:895 phy_start+0x74/0xc0
Modules linked in:
CPU: 1 PID: 155 Comm: backend Not tainted 5.2.0+ #1
NIP: c0227bf0 LR: c0227bf0 CTR: c004d224
REGS: df547720 TRAP: 0700 Not tainted (5.2.0+)
MSR: 00029000 <CE,EE,ME> CR: 24002822 XER: 00000000
% ip link set down dev eth0
------------[ cut here ]------------
called from state HALTED
WARNING: CPU: 1 PID: 184 at drivers/net/phy/phy.c:858 phy_stop+0x3c/0x88
SFP modules with the 1000Base-X interface are not affected.
Place explicit calls to phy_start() and phy_stop() before enabling or after
disabling an attached SFP module, where phydev is not yet set (or is
already unset), so they will be made only from the inside of sfp-bus, if
needed.
Fixes: 217962615662 ("net: phy: warn if phy_start is called from invalid state") Signed-off-by: Arseny Solokha <asolokha@kb.kras.ru> Acked-by: Russell King <rmk+kernel@armlinux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It is perfectly ok to not have an gpio attached to the fixed-link node. So
the driver should not throw an error message when the gpio is missing.
Fixes: 5468e82f7034 ("net: phy: fixed-phy: Drop GPIO from fixed_phy_add()") Signed-off-by: Hubert Feurstein <h.feurstein@gmail.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When lag is active, which is controlled by the bonded mlx5e netdev, mlx5
interface unregestering must happen in the reverse order where rdma is
unregistered (unloaded) first, to guarantee all references to the lag
context in hardware is removed, then remove mlx5e netdev interface which
will cleanup the lag context from hardware.
Without this fix during destroy of LAG interface, we observed following
errors:
* mlx5_cmd_check:752:(pid 12556): DESTROY_LAG(0x843) op_mod(0x0) failed,
status bad parameter(0x3), syndrome (0xe4ac33)
* mlx5_cmd_check:752:(pid 12556): DESTROY_LAG(0x843) op_mod(0x0) failed,
status bad parameter(0x3), syndrome (0xa5aee8).
Fixes: a31208b1e11d ("net/mlx5_core: New init and exit flow for mlx5_core") Reviewed-by: Parav Pandit <parav@mellanox.com> Reviewed-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Mark Zhang <markz@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The commit 069d11465a80 ("net/mlx5e: RX, Enhance legacy Receive Queue
memory scheme") introduced an undefined behaviour below due to
"frag->last_in_page" is only initialized in mlx5e_init_frags_partition()
when,
if (next_frag.offset + frag_info[f].frag_stride > PAGE_SIZE)
or after bailed out the loop,
for (i = 0; i < mlx5_wq_cyc_get_size(&rq->wqe.wq); i++)
As the result, there could be some "frag" have uninitialized
value of "last_in_page".
Later, get_frag() obtains those "frag" and check "frag->last_in_page" in
mlx5e_put_rx_frag() and triggers the error during boot. Fix it by always
initializing "frag->last_in_page" to "false" in
mlx5e_init_frags_partition().
UBSAN: Undefined behaviour in
drivers/net/ethernet/mellanox/mlx5/core/en_rx.c:325:12
load of value 170 is not a valid value for type 'bool' (aka '_Bool')
Call trace:
dump_backtrace+0x0/0x264
show_stack+0x20/0x2c
dump_stack+0xb0/0x104
__ubsan_handle_load_invalid_value+0x104/0x128
mlx5e_handle_rx_cqe+0x8e8/0x12cc [mlx5_core]
mlx5e_poll_rx_cq+0xca8/0x1a94 [mlx5_core]
mlx5e_napi_poll+0x17c/0xa30 [mlx5_core]
net_rx_action+0x248/0x940
__do_softirq+0x350/0x7b8
irq_exit+0x200/0x26c
__handle_domain_irq+0xc8/0x128
gic_handle_irq+0x138/0x228
el1_irq+0xb8/0x140
arch_cpu_idle+0x1a4/0x348
do_idle+0x114/0x1b0
cpu_startup_entry+0x24/0x28
rest_init+0x1ac/0x1dc
arch_call_rest_init+0x10/0x18
start_kernel+0x4d4/0x57c
Fixes: 069d11465a80 ("net/mlx5e: RX, Enhance legacy Receive Queue memory scheme") Signed-off-by: Qian Cai <cai@lca.pw> Reviewed-by: Tariq Toukan <tariqt@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit aca51397d014 ("netns: Fix arbitrary net_device-s corruptions
on net_ns stop.") introduced a possibility to hit a BUG in case device
is returning back to init_net and two following conditions are met:
1) dev->ifindex value is used in a name of another "dev%d"
device in init_net.
2) dev->name is used by another device in init_net.
Under real life circumstances this is hard to get. Therefore this has
been present happily for over 10 years. To reproduce:
$ ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: dummy0: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN group default qlen 1000
link/ether 86:89:3f:86:61:29 brd ff:ff:ff:ff:ff:ff
3: enp0s2: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
link/ether 52:54:00:12:34:56 brd ff:ff:ff:ff:ff:ff
$ ip netns add ns1
$ ip -n ns1 link add dummy1ns1 type dummy
$ ip -n ns1 link add dummy2ns1 type dummy
$ ip link set enp0s2 netns ns1
$ ip -n ns1 link set enp0s2 name dummy0
[ 100.858894] virtio_net virtio0 dummy0: renamed from enp0s2
$ ip link add dev4 type dummy
$ ip -n ns1 a
1: lo: <LOOPBACK> mtu 65536 qdisc noop state DOWN group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: dummy1ns1: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN group default qlen 1000
link/ether 16:63:4c:38:3e:ff brd ff:ff:ff:ff:ff:ff
3: dummy2ns1: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN group default qlen 1000
link/ether aa:9e:86:dd:6b:5d brd ff:ff:ff:ff:ff:ff
4: dummy0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN group default qlen 1000
link/ether 52:54:00:12:34:56 brd ff:ff:ff:ff:ff:ff
$ ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: dummy0: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN group default qlen 1000
link/ether 86:89:3f:86:61:29 brd ff:ff:ff:ff:ff:ff
4: dev4: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN group default qlen 1000
link/ether 5a:e1:4a:b6:ec:f8 brd ff:ff:ff:ff:ff:ff
$ ip netns del ns1
[ 158.717795] default_device_exit: failed to move dummy0 to init_net: -17
[ 158.719316] ------------[ cut here ]------------
[ 158.720591] kernel BUG at net/core/dev.c:9824!
[ 158.722260] invalid opcode: 0000 [#1] SMP KASAN PTI
[ 158.723728] CPU: 0 PID: 56 Comm: kworker/u2:1 Not tainted 5.3.0-rc1+ #18
[ 158.725422] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-2.fc30 04/01/2014
[ 158.727508] Workqueue: netns cleanup_net
[ 158.728915] RIP: 0010:default_device_exit.cold+0x1d/0x1f
[ 158.730683] Code: 84 e8 18 c9 3e fe 0f 0b e9 70 90 ff ff e8 36 e4 52 fe 89 d9 4c 89 e2 48 c7 c6 80 d6 25 84 48 c7 c7 20 c0 25 84 e8 f4 c8 3e
[ 158.736854] RSP: 0018:ffff8880347e7b90 EFLAGS: 00010282
[ 158.738752] RAX: 000000000000003b RBX: 00000000ffffffef RCX: 0000000000000000
[ 158.741369] RDX: 0000000000000000 RSI: ffffffff8128013d RDI: ffffed10068fcf64
[ 158.743418] RBP: ffff888033550170 R08: 000000000000003b R09: fffffbfff0b94b9c
[ 158.745626] R10: fffffbfff0b94b9b R11: ffffffff85ca5cdf R12: ffff888032f28000
[ 158.748405] R13: dffffc0000000000 R14: ffff8880335501b8 R15: 1ffff110068fcf72
[ 158.750638] FS: 0000000000000000(0000) GS:ffff888036000000(0000) knlGS:0000000000000000
[ 158.752944] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 158.755245] CR2: 00007fe8b45d21d0 CR3: 00000000340b4005 CR4: 0000000000360ef0
[ 158.757654] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 158.760012] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 158.762758] Call Trace:
[ 158.763882] ? dev_change_net_namespace+0xbb0/0xbb0
[ 158.766148] ? devlink_nl_cmd_set_doit+0x520/0x520
[ 158.768034] ? dev_change_net_namespace+0xbb0/0xbb0
[ 158.769870] ops_exit_list.isra.0+0xa8/0x150
[ 158.771544] cleanup_net+0x446/0x8f0
[ 158.772945] ? unregister_pernet_operations+0x4a0/0x4a0
[ 158.775294] process_one_work+0xa1a/0x1740
[ 158.776896] ? pwq_dec_nr_in_flight+0x310/0x310
[ 158.779143] ? do_raw_spin_lock+0x11b/0x280
[ 158.780848] worker_thread+0x9e/0x1060
[ 158.782500] ? process_one_work+0x1740/0x1740
[ 158.784454] kthread+0x31b/0x420
[ 158.786082] ? __kthread_create_on_node+0x3f0/0x3f0
[ 158.788286] ret_from_fork+0x3a/0x50
[ 158.789871] ---[ end trace defd6c657c71f936 ]---
[ 158.792273] RIP: 0010:default_device_exit.cold+0x1d/0x1f
[ 158.795478] Code: 84 e8 18 c9 3e fe 0f 0b e9 70 90 ff ff e8 36 e4 52 fe 89 d9 4c 89 e2 48 c7 c6 80 d6 25 84 48 c7 c7 20 c0 25 84 e8 f4 c8 3e
[ 158.804854] RSP: 0018:ffff8880347e7b90 EFLAGS: 00010282
[ 158.807865] RAX: 000000000000003b RBX: 00000000ffffffef RCX: 0000000000000000
[ 158.811794] RDX: 0000000000000000 RSI: ffffffff8128013d RDI: ffffed10068fcf64
[ 158.816652] RBP: ffff888033550170 R08: 000000000000003b R09: fffffbfff0b94b9c
[ 158.820930] R10: fffffbfff0b94b9b R11: ffffffff85ca5cdf R12: ffff888032f28000
[ 158.825113] R13: dffffc0000000000 R14: ffff8880335501b8 R15: 1ffff110068fcf72
[ 158.829899] FS: 0000000000000000(0000) GS:ffff888036000000(0000) knlGS:0000000000000000
[ 158.834923] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 158.838164] CR2: 00007fe8b45d21d0 CR3: 00000000340b4005 CR4: 0000000000360ef0
[ 158.841917] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 158.845149] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Fix this by checking if a device with the same name exists in init_net
and fallback to original code - dev%d to allocate name - in case it does.
This was found using syzkaller.
Fixes: aca51397d014 ("netns: Fix arbitrary net_device-s corruptions on net_ns stop.") Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Most of the bridge device's vlan init bugs come from the fact that its
default pvid is created at the wrong time, way too early in ndo_init()
before the device is even assigned an ifindex. It introduces a bug when the
bridge's dev_addr is added as fdb during the initial default pvid creation
the notification has ifindex/NDA_MASTER both equal to 0 (see example below)
which really makes no sense for user-space[0] and is wrong.
Usually user-space software would ignore such entries, but they are
actually valid and will eventually have all necessary attributes.
It makes much more sense to send a notification *after* the device has
registered and has a proper ifindex allocated rather than before when
there's a chance that the registration might still fail or to receive
it with ifindex/NDA_MASTER == 0. Note that we can remove the fdb flush
from br_vlan_flush() since that case can no longer happen. At
NETDEV_REGISTER br->default_pvid is always == 1 as it's initialized by
br_vlan_init() before that and at NETDEV_UNREGISTER it can be anything
depending why it was called (if called due to NETDEV_REGISTER error
it'll still be == 1, otherwise it could be any value changed during the
device life time).
For the demonstration below a small change to iproute2 for printing all fdb
notifications is added, because it contained a workaround not to show
entries with ifindex == 0.
Command executed while monitoring: $ ip l add br0 type bridge
Before (both ifindex and master == 0):
$ bridge monitor fdb
36:7e:8a:b3:56:ba dev * vlan 1 master * permanent
After (proper br0 ifindex):
$ bridge monitor fdb
e6:2a:ae:7a:b7:48 dev br0 vlan 1 master br0 permanent
v4: move only the default pvid init/deinit to NETDEV_REGISTER/UNREGISTER
v3: send the correct v2 patch with all changes (stub should return 0)
v2: on error in br_vlan_init set br->vlgrp to NULL and return 0 in
the br_vlan_bridge_event stub when bridge vlans are disabled
Reported-by: michael-dev <michael-dev@fami-braun.de> Fixes: 5be5a2df40f0 ("bridge: Add filtering support for default_pvid") Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Acked-by: Roopa Prabhu <roopa@cumulusnetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When permanent entries were introduced by the commit below, they were
exempt from timing out and thus igmp leave wouldn't affect them unless
fast leave was enabled on the port which was added before permanent
entries existed. It shouldn't matter if fast leave is enabled or not
if the user added a permanent entry it shouldn't be deleted on igmp
leave.
Before:
$ echo 1 > /sys/class/net/eth4/brport/multicast_fast_leave
$ bridge mdb add dev br0 port eth4 grp 229.1.1.1 permanent
$ bridge mdb show
dev br0 port eth4 grp 229.1.1.1 permanent
< join and leave 229.1.1.1 on eth4 >
$ bridge mdb show
$
After:
$ echo 1 > /sys/class/net/eth4/brport/multicast_fast_leave
$ bridge mdb add dev br0 port eth4 grp 229.1.1.1 permanent
$ bridge mdb show
dev br0 port eth4 grp 229.1.1.1 permanent
< join and leave 229.1.1.1 on eth4 >
$ bridge mdb show
dev br0 port eth4 grp 229.1.1.1 permanent
Fixes: ccb1c31a7a87 ("bridge: add flags to distinguish permanent mdb entires") Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On initialization failure we have to delete the local fdb which was
inserted due to the default pvid creation. This problem has been present
since the inception of default_pvid. Note that currently there are 2 cases:
1) in br_dev_init() when br_multicast_init() fails
2) if register_netdevice() fails after calling ndo_init()
This patch takes care of both since br_vlan_flush() is called on both
occasions. Also the new fdb delete would be a no-op on normal bridge
device destruction since the local fdb would've been already flushed by
br_dev_delete(). This is not an issue for ports since nbp_vlan_init() is
called last when adding a port thus nothing can fail after it.
Reported-by: syzbot+88533dc8b582309bf3ee@syzkaller.appspotmail.com Fixes: 5be5a2df40f0 ("bridge: Add filtering support for default_pvid") Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The MTU change code can call napi_disable() with the device already down,
leading to a deadlock. Also, lot of code is duplicated unnecessarily.
Rework mvpp2_change_mtu() to avoid the deadlock and remove duplicated code.
Fixes: 3f518509dedc ("ethernet: Add new driver for Marvell Armada 375 network unit") Signed-off-by: Matteo Croce <mcroce@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
mvpp2 uses a delayed workqueue to gather traffic statistics.
On module removal the workqueue can be destroyed before calling
cancel_delayed_work_sync() on its works.
Fix it by moving the destroy_workqueue() call after mvpp2_port_remove().
Also remove an unneeded call to flush_workqueue()
We need the same checks introduced by commit cb9f1b783850
("ip: validate header length on virtual device xmit") for
ipip tunnel.
Fixes: cb9f1b783850b ("ip: validate header length on virtual device xmit") Signed-off-by: Haishuang Yan <yanhaishuang@cmss.chinamobile.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
ip4ip6/ip6ip6 tunnels run iptunnel_handle_offloads on xmit which
can cause a possible use-after-free accessing iph/ipv6h pointer
since the packet will be 'uncloned' running pskb_expand_head if
it is a cloned gso skb.
Fixes: 0e9a709560db ("ip6_tunnel, ip6_gre: fix setting of DSCP on encapsulated packets") Signed-off-by: Haishuang Yan <yanhaishuang@cmss.chinamobile.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Since ip6_tnl_parse_tlv_enc_lim() can call pskb_may_pull()
which may change skb->data, so we need to re-load ipv6h at
the right place.
Fixes: 898b29798e36 ("ip6_gre: Refactor ip6gre xmit codes") Cc: William Tu <u9012063@gmail.com> Signed-off-by: Haishuang Yan <yanhaishuang@cmss.chinamobile.com> Acked-by: William Tu <u9012063@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Orion5.x systems are still using machine files and not device-tree.
Commit 96cb4342382290c9 ("net: mvmdio: allow up to three clocks to be
specified for orion-mdio") has replaced devm_clk_get() with of_clk_get(),
leading to a oops at boot and not working network, as reported in
https://lists.debian.org/debian-arm/2019/07/msg00088.html and possibly in
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=908712.
Link: https://lists.debian.org/debian-arm/2019/07/msg00088.html Fixes: 96cb4342382290c9 ("net: mvmdio: allow up to three clocks to be specified for orion-mdio") Signed-off-by: Arnaud Patard <arnaud.patard@rtp-net.org> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 3968d38917eb ("bnx2x: Fix Multi-Cos.") which enabled multi-cos
feature after prolonged time in driver added some regression causing
numerous issues (sudden reboots, tx timeout etc.) reported by customers.
We plan to backout this commit and submit proper fix once we have root
cause of issues reported with this feature enabled.
Fixes: 3968d38917eb ("bnx2x: Fix Multi-Cos.") Signed-off-by: Sudarsana Reddy Kalluru <skalluru@marvell.com> Signed-off-by: Manish Chopra <manishc@marvell.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
board is controlled by user-space, hence leading to a potential
exploitation of the Spectre variant 1 vulnerability.
This issue was detected with the help of Smatch:
drivers/atm/iphase.c:2765 ia_ioctl() warn: potential spectre issue 'ia_dev' [r] (local cap)
drivers/atm/iphase.c:2774 ia_ioctl() warn: possible spectre second half. 'iadev'
drivers/atm/iphase.c:2782 ia_ioctl() warn: possible spectre second half. 'iadev'
drivers/atm/iphase.c:2816 ia_ioctl() warn: possible spectre second half. 'iadev'
drivers/atm/iphase.c:2823 ia_ioctl() warn: possible spectre second half. 'iadev'
drivers/atm/iphase.c:2830 ia_ioctl() warn: potential spectre issue '_ia_dev' [r] (local cap)
drivers/atm/iphase.c:2845 ia_ioctl() warn: possible spectre second half. 'iadev'
drivers/atm/iphase.c:2856 ia_ioctl() warn: possible spectre second half. 'iadev'
Fix this by sanitizing board before using it to index ia_dev and _ia_dev
Notice that given that speculation windows are large, the policy is
to kill the speculation on the first load and not worry if it can be
completed with a dependent load/store [1].
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The PixArt OEM mice are known for disconnecting every minute in
runlevel 1 or 3 if they are not always polled. So add quirk
ALWAYS_POLL for this one as well.
Jonathan Teh (@jonathan-teh) reported and tested the quirk.
Reference: https://github.com/sriemer/fix-linux-mouse/issues/15
The recent USB core code performs sanity checks for the given pipe and
EP types, and it can be hit by manipulated USB descriptors by syzbot.
For making syzbot happier, this patch introduces a local helper for a
sanity check in the driver side and calls it at each place before the
message handling, so that we can avoid the WARNING splats.
A multithreaded namespace creation/destruction stress test currently
deadlocks with the following lockup signature:
INFO: task ndctl:2924 blocked for more than 122 seconds.
Tainted: G OE 5.2.0-rc4+ #3382
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
ndctl D 0 2924 1176 0x00000000
Call Trace:
? __schedule+0x27e/0x780
schedule+0x30/0xb0
wait_nvdimm_bus_probe_idle+0x8a/0xd0 [libnvdimm]
? finish_wait+0x80/0x80
uuid_store+0xe6/0x2e0 [libnvdimm]
kernfs_fop_write+0xf0/0x1a0
vfs_write+0xb7/0x1b0
ksys_write+0x5c/0xd0
do_syscall_64+0x60/0x240
INFO: task ndctl:2923 blocked for more than 122 seconds.
Tainted: G OE 5.2.0-rc4+ #3382
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
ndctl D 0 2923 1175 0x00000000
Call Trace:
? __schedule+0x27e/0x780
? __mutex_lock+0x489/0x910
schedule+0x30/0xb0
schedule_preempt_disabled+0x11/0x20
__mutex_lock+0x48e/0x910
? nvdimm_namespace_common_probe+0x95/0x4d0 [libnvdimm]
? __lock_acquire+0x23f/0x1710
? nvdimm_namespace_common_probe+0x95/0x4d0 [libnvdimm]
nvdimm_namespace_common_probe+0x95/0x4d0 [libnvdimm]
__dax_pmem_probe+0x5e/0x210 [dax_pmem_core]
? nvdimm_bus_probe+0x1d0/0x2c0 [libnvdimm]
dax_pmem_probe+0xc/0x20 [dax_pmem]
nvdimm_bus_probe+0x90/0x2c0 [libnvdimm]
really_probe+0xef/0x390
driver_probe_device+0xb4/0x100
In this sequence an 'nd_dax' device is being probed and trying to take
the lock on its backing namespace to validate that the 'nd_dax' device
indeed has exclusive access to the backing namespace. Meanwhile, another
thread is trying to update the uuid property of that same backing
namespace. So one thread is in the probe path trying to acquire the
lock, and the other thread has acquired the lock and tries to flush the
probe path.
Fix this deadlock by not holding the namespace device_lock over the
wait_nvdimm_bus_probe_idle() synchronization step. In turn this requires
the device_lock to be held on entry to wait_nvdimm_bus_probe_idle() and
subsequently dropped internally to wait_nvdimm_bus_probe_idle().
In preparation for not holding a lock over the execution of nd_ioctl(),
update the implementation to allow multiple threads to be attempting
ioctls at the same time. The bus lock still prevents multiple in-flight
->ndctl() invocations from corrupting each other's state, but static
global staging buffers are moved to the heap.
Gcc-9 complains for a memset across pointer boundaries, which happens as
the code tries to allocate a flexible array on the stack. Turns out we
cannot do this without relying on gcc-isms, so with this patch we'll embed
the fc_rport_priv structure into fcoe_rport, can use the normal
'container_of' outcast, and will only have to do a memset over one
structure.
Signed-off-by: Hannes Reinecke <hare@suse.de> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On all current Atom processors, instructions that use a segment register
value (e.g. a load or store) will not speculatively execute before the
last writer of that segment retires. Thus they will not use a
speculatively written segment value.
That means on ATOMs there is no speculation through SWAPGS, so the SWAPGS
entry paths can be excluded from the extra LFENCE if PTI is disabled.
Create a separate bug flag for the through SWAPGS speculation and mark all
out-of-order ATOMs and AMD/HYGON CPUs as not affected. The in-order ATOMs
are excluded from the whole mitigation mess anyway.
Reported-by: Andrew Cooper <andrew.cooper3@citrix.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Tyler Hicks <tyhicks@canonical.com> Reviewed-by: Josh Poimboeuf <jpoimboe@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Somehow the swapgs mitigation entry code patch ended up with a JMPQ
instruction instead of JMP, where only the short jump is needed. Some
assembler versions apparently fail to optimize JMPQ into a two-byte JMP
when possible, instead always using a 7-byte JMP with relocation. For
some reason that makes the entry code explode with a #GP during boot.
The previous commit added macro calls in the entry code which mitigate the
Spectre v1 swapgs issue if the X86_FEATURE_FENCE_SWAPGS_* features are
enabled. Enable those features where applicable.
The mitigations may be disabled with "nospectre_v1" or "mitigations=off".
There are different features which can affect the risk of attack:
- When FSGSBASE is enabled, unprivileged users are able to place any
value in GS, using the wrgsbase instruction. This means they can
write a GS value which points to any value in kernel space, which can
be useful with the following gadget in an interrupt/exception/NMI
handler:
if (coming from user space)
swapgs
mov %gs:<percpu_offset>, %reg1
// dependent load or store based on the value of %reg
// for example: mov %(reg1), %reg2
If an interrupt is coming from user space, and the entry code
speculatively skips the swapgs (due to user branch mistraining), it
may speculatively execute the GS-based load and a subsequent dependent
load or store, exposing the kernel data to an L1 side channel leak.
Note that, on Intel, a similar attack exists in the above gadget when
coming from kernel space, if the swapgs gets speculatively executed to
switch back to the user GS. On AMD, this variant isn't possible
because swapgs is serializing with respect to future GS-based
accesses.
NOTE: The FSGSBASE patch set hasn't been merged yet, so the above case
doesn't exist quite yet.
- When FSGSBASE is disabled, the issue is mitigated somewhat because
unprivileged users must use prctl(ARCH_SET_GS) to set GS, which
restricts GS values to user space addresses only. That means the
gadget would need an additional step, since the target kernel address
needs to be read from user space first. Something like:
if (coming from user space)
swapgs
mov %gs:<percpu_offset>, %reg1
mov (%reg1), %reg2
// dependent load or store based on the value of %reg2
// for example: mov %(reg2), %reg3
It's difficult to audit for this gadget in all the handlers, so while
there are no known instances of it, it's entirely possible that it
exists somewhere (or could be introduced in the future). Without
tooling to analyze all such code paths, consider it vulnerable.
Effects of SMAP on the !FSGSBASE case:
- If SMAP is enabled, and the CPU reports RDCL_NO (i.e., not
susceptible to Meltdown), the kernel is prevented from speculatively
reading user space memory, even L1 cached values. This effectively
disables the !FSGSBASE attack vector.
- If SMAP is enabled, but the CPU *is* susceptible to Meltdown, SMAP
still prevents the kernel from speculatively reading user space
memory. But it does *not* prevent the kernel from reading the
user value from L1, if it has already been cached. This is probably
only a small hurdle for an attacker to overcome.
Thanks to Dave Hansen for contributing the speculative_smap() function.
Thanks to Andrew Cooper for providing the inside scoop on whether swapgs
is serializing on AMD.
[ tglx: Fixed the USER fence decision and polished the comment as suggested
by Dave Hansen ]
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Dave Hansen <dave.hansen@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Spectre v1 isn't only about array bounds checks. It can affect any
conditional checks. The kernel entry code interrupt, exception, and NMI
handlers all have conditional swapgs checks. Those may be problematic in
the context of Spectre v1, as kernel code can speculatively run with a user
GS.
For example:
if (coming from user space)
swapgs
mov %gs:<percpu_offset>, %reg
mov (%reg), %reg1
When coming from user space, the CPU can speculatively skip the swapgs, and
then do a speculative percpu load using the user GS value. So the user can
speculatively force a read of any kernel value. If a gadget exists which
uses the percpu value as an address in another load/store, then the
contents of the kernel value may become visible via an L1 side channel
attack.
A similar attack exists when coming from kernel space. The CPU can
speculatively do the swapgs, causing the user GS to get used for the rest
of the speculative window.
The mitigation is similar to a traditional Spectre v1 mitigation, except:
a) index masking isn't possible; because the index (percpu offset)
isn't user-controlled; and
b) an lfence is needed in both the "from user" swapgs path and the
"from kernel" non-swapgs path (because of the two attacks described
above).
The user entry swapgs paths already have SWITCH_TO_KERNEL_CR3, which has a
CR3 write when PTI is enabled. Since CR3 writes are serializing, the
lfences can be skipped in those cases.
On the other hand, the kernel entry swapgs paths don't depend on PTI.
To avoid unnecessary lfences for the user entry case, create two separate
features for alternative patching:
projects / users / dwmw2 / linux.git / log