Only when RXD advanced layout is enabled, in some cases
(e.g. ip fragments), the checksum of entire packet will be
calculated and filled in the least significant 16 bits of
the unused addr field.
So refactor out the handling of RX completion checksum: adjust
the location of the checksum in RX descriptor, and use ptype table
to identify whether this kind of checksum is calculated.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, the driver gets packet type by parsing the
L3_ID/L4_ID/OL3_ID/OL4_ID from RX descriptor, it's
time-consuming.
Now some new devices support RXD advanced layout, which combines
previous OL3_ID/OL4_ID to 8bit ptype field, so the driver gets
packet type by looking up only one table, and L3_ID/L4_ID become
reserved fields.
Considering compatibility, the firmware will report capability of
RXD advanced layout, the driver will identify and enable it by
default. This patch provides basic function: identify and enable
the RXD advanced layout, and refactor out hns3_rx_checksum() by
using ptype table to handle RX checksum if supported.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Yunsheng Lin says:
====================
ix packet stuck problem for lockless qdisc
This patchset fixes the packet stuck problem mentioned in [1].
Patch 1: Add STATE_MISSED flag to fix packet stuck problem.
Patch 2: Fix a tx_action rescheduling problem after STATE_MISSED
flag is added in patch 1.
Patch 3: Fix the significantly higher CPU consumption problem when
multiple threads are competing on a saturated outgoing
device.
V8: Change function name as suggested by Jakub and fix some typo
in patch 3, adjust commit log in patch 2, and add Acked-by
from Jakub.
V7: Fix netif_tx_wake_queue() data race noted by Jakub.
V6: Some performance optimization in patch 1 suggested by Jakub
and drop NET_XMIT_DROP checking in patch 3.
V5: add patch 3 to fix the problem reported by Michal Kubecek.
V4: Change STATE_NEED_RESCHEDULE to STATE_MISSED and add patch 2.
[1]. https://lkml.org/lkml/2019/10/9/42
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
The netdev qeueue might be stopped when byte queue limit has
reached or tx hw ring is full, net_tx_action() may still be
rescheduled if STATE_MISSED is set, which consumes unnecessary
cpu without dequeuing and transmiting any skb because the
netdev queue is stopped, see qdisc_run_end().
This patch fixes it by checking the netdev queue state before
calling qdisc_run() and clearing STATE_MISSED if netdev queue is
stopped during qdisc_run(), the net_tx_action() is rescheduled
again when netdev qeueue is restarted, see netif_tx_wake_queue().
As there is time window between netif_xmit_frozen_or_stopped()
checking and STATE_MISSED clearing, between which STATE_MISSED
may set by net_tx_action() scheduled by netif_tx_wake_queue(),
so set the STATE_MISSED again if netdev queue is restarted.
Fixes: 6b3ba9146f ("net: sched: allow qdiscs to handle locking")
Reported-by: Michal Kubecek <mkubecek@suse.cz>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently qdisc_run() checks the STATE_DEACTIVATED of lockless
qdisc before calling __qdisc_run(), which ultimately clear the
STATE_MISSED when all the skb is dequeued. If STATE_DEACTIVATED
is set before clearing STATE_MISSED, there may be rescheduling
of net_tx_action() at the end of qdisc_run_end(), see below:
CPU0(net_tx_atcion) CPU1(__dev_xmit_skb) CPU2(dev_deactivate)
. . .
. set STATE_MISSED .
. __netif_schedule() .
. . set STATE_DEACTIVATED
. . qdisc_reset()
. . .
.<--------------- . synchronize_net()
clear __QDISC_STATE_SCHED | . .
. | . .
. | . some_qdisc_is_busy()
. | . return *false*
. | . .
test STATE_DEACTIVATED | . .
__qdisc_run() *not* called | . .
. | . .
test STATE_MISS | . .
__netif_schedule()--------| . .
. . .
. . .
__qdisc_run() is not called by net_tx_atcion() in CPU0 because
CPU2 has set STATE_DEACTIVATED flag during dev_deactivate(), and
STATE_MISSED is only cleared in __qdisc_run(), __netif_schedule
is called at the end of qdisc_run_end(), causing tx action
rescheduling problem.
qdisc_run() called by net_tx_action() runs in the softirq context,
which should has the same semantic as the qdisc_run() called by
__dev_xmit_skb() protected by rcu_read_lock_bh(). And there is a
synchronize_net() between STATE_DEACTIVATED flag being set and
qdisc_reset()/some_qdisc_is_busy in dev_deactivate(), we can safely
bail out for the deactived lockless qdisc in net_tx_action(), and
qdisc_reset() will reset all skb not dequeued yet.
So add the rcu_read_lock() explicitly to protect the qdisc_run()
and do the STATE_DEACTIVATED checking in net_tx_action() before
calling qdisc_run_begin(). Another option is to do the checking in
the qdisc_run_end(), but it will add unnecessary overhead for
non-tx_action case, because __dev_queue_xmit() will not see qdisc
with STATE_DEACTIVATED after synchronize_net(), the qdisc with
STATE_DEACTIVATED can only be seen by net_tx_action() because of
__netif_schedule().
The STATE_DEACTIVATED checking in qdisc_run() is to avoid race
between net_tx_action() and qdisc_reset(), see:
commit d518d2ed86 ("net/sched: fix race between deactivation
and dequeue for NOLOCK qdisc"). As the bailout added above for
deactived lockless qdisc in net_tx_action() provides better
protection for the race without calling qdisc_run() at all, so
remove the STATE_DEACTIVATED checking in qdisc_run().
After qdisc_reset(), there is no skb in qdisc to be dequeued, so
clear the STATE_MISSED in dev_reset_queue() too.
Fixes: 6b3ba9146f ("net: sched: allow qdiscs to handle locking")
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
V8: Clearing STATE_MISSED before calling __netif_schedule() has
avoid the endless rescheduling problem, but there may still
be a unnecessary rescheduling, so adjust the commit log.
Signed-off-by: David S. Miller <davem@davemloft.net>
Lockless qdisc has below concurrent problem:
cpu0 cpu1
. .
q->enqueue .
. .
qdisc_run_begin() .
. .
dequeue_skb() .
. .
sch_direct_xmit() .
. .
. q->enqueue
. qdisc_run_begin()
. return and do nothing
. .
qdisc_run_end() .
cpu1 enqueue a skb without calling __qdisc_run() because cpu0
has not released the lock yet and spin_trylock() return false
for cpu1 in qdisc_run_begin(), and cpu0 do not see the skb
enqueued by cpu1 when calling dequeue_skb() because cpu1 may
enqueue the skb after cpu0 calling dequeue_skb() and before
cpu0 calling qdisc_run_end().
Lockless qdisc has below another concurrent problem when
tx_action is involved:
cpu0(serving tx_action) cpu1 cpu2
. . .
. q->enqueue .
. qdisc_run_begin() .
. dequeue_skb() .
. . q->enqueue
. . .
. sch_direct_xmit() .
. . qdisc_run_begin()
. . return and do nothing
. . .
clear __QDISC_STATE_SCHED . .
qdisc_run_begin() . .
return and do nothing . .
. . .
. qdisc_run_end() .
This patch fixes the above data race by:
1. If the first spin_trylock() return false and STATE_MISSED is
not set, set STATE_MISSED and retry another spin_trylock() in
case other CPU may not see STATE_MISSED after it releases the
lock.
2. reschedule if STATE_MISSED is set after the lock is released
at the end of qdisc_run_end().
For tx_action case, STATE_MISSED is also set when cpu1 is at the
end if qdisc_run_end(), so tx_action will be rescheduled again
to dequeue the skb enqueued by cpu2.
Clear STATE_MISSED before retrying a dequeuing when dequeuing
returns NULL in order to reduce the overhead of the second
spin_trylock() and __netif_schedule() calling.
Also clear the STATE_MISSED before calling __netif_schedule()
at the end of qdisc_run_end() to avoid doing another round of
dequeuing in the pfifo_fast_dequeue().
The performance impact of this patch, tested using pktgen and
dummy netdev with pfifo_fast qdisc attached:
threads without+this_patch with+this_patch delta
1 2.61Mpps 2.60Mpps -0.3%
2 3.97Mpps 3.82Mpps -3.7%
4 5.62Mpps 5.59Mpps -0.5%
8 2.78Mpps 2.77Mpps -0.3%
16 2.22Mpps 2.22Mpps -0.0%
Fixes: 6b3ba9146f ("net: sched: allow qdiscs to handle locking")
Acked-by: Jakub Kicinski <kuba@kernel.org>
Tested-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In tls_sw_splice_read, checkout MSG_* is inappropriate, should use
SPLICE_*, update tls_wait_data to accept nonblock arguments instead
of flags for recvmsg and splice.
Fixes: c46234ebb4 ("tls: RX path for ktls")
Signed-off-by: Jim Ma <majinjing3@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The result of container_of() operations is never NULL unless the embedded
element is the first element of the structure. This is not the case here.
The NULL check is therefore unnecessary and misleading. Remove it.
This change was made automatically with the following Coccinelle script.
@@
type t;
identifier v;
statement s;
@@
<+...
(
t v = container_of(...);
|
v = container_of(...);
)
...
when != v
- if (\( !v \| v == NULL \) ) s
...+>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Using netif_info() before the net_device is registered results in ugly
messages like the following:
sfc 0000:01:00.1 (unnamed net_device) (uninitialized): Solarflare NIC detected
Therefore use pci_info() et al until net_device is registered.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull tracing fix from Steven Rostedt:
"Fix trace_check_vprintf() for %.*s
The sanity check of all strings being read from the ring buffer to
make sure they are in safe memory space did not account for the %.*s
notation having another parameter to process (the length).
Add that to the check"
* tag 'trace-v5.13-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
tracing: Handle %.*s in trace_check_vprintf()
drm/i915 fixes for v5.13-rc2:
- Fix active callback alignment annotations and subsequent crashes
- Retract link training strategy to slow and wide, again
- Avoid division by zero on gen2
- Use correct width reads for C0DRB3/C1DRB3 registers
- Fix double free in pdp allocation failure path
- Fix HDMI 2.1 PCON downstream caps check
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/87a6oxu9ao.fsf@intel.com
Pull arm64 fixes from Catalin Marinas:
"Fixes and cpucaps.h automatic generation:
- Generate cpucaps.h at build time rather than carrying lots of
#defines. Merged at -rc1 to avoid some conflicts during the merge
window.
- Initialise RGSR_EL1.SEED in __cpu_setup() as it may be left as 0
out of reset and the IRG instruction would not function as expected
if only the architected pseudorandom number generator is
implemented.
- Fix potential race condition in __sync_icache_dcache() where the
PG_dcache_clean page flag is set before the actual cache
maintenance.
- Fix header include in BTI kselftests"
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64: Fix race condition on PG_dcache_clean in __sync_icache_dcache()
arm64: tools: Add __ASM_CPUCAPS_H to the endif in cpucaps.h
arm64: mte: initialize RGSR_EL1.SEED in __cpu_setup
kselftest/arm64: Add missing stddef.h include to BTI tests
arm64: Generate cpucaps.h
Pull f2fs fixes from Jaegeuk Kim:
"This fixes some critical bugs such as memory leak in compression
flows, kernel panic when handling errors, and swapon failure due to
newly added condition check"
* tag 'f2fs-5.13-rc1-fix' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs:
f2fs: return EINVAL for hole cases in swap file
f2fs: avoid swapon failure by giving a warning first
f2fs: compress: fix to assign cc.cluster_idx correctly
f2fs: compress: fix race condition of overwrite vs truncate
f2fs: compress: fix to free compress page correctly
f2fs: support iflag change given the mask
f2fs: avoid null pointer access when handling IPU error
Pablo Neira Ayuso says:
====================
Netfilter fixes for net
The following patchset contains Netfilter fixes for net:
1) Remove the flowtable hardware refresh state, fall back to the
existing hardware pending state instead, from Roi Dayan.
2) Fix crash in pipapo avx2 lookup when FPU is in used from user
context, from Stefano Brivio.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
The br_ip6_multicast_add_router() prototype is defined only when
CONFIG_IPV6 is enabled, but the function is always referenced, so there
is this build error with CONFIG_IPV6 not defined:
net/bridge/br_multicast.c: In function ‘__br_multicast_enable_port’:
net/bridge/br_multicast.c:1743:3: error: implicit declaration of function ‘br_ip6_multicast_add_router’; did you mean ‘br_ip4_multicast_add_router’? [-Werror=implicit-function-declaration]
1743 | br_ip6_multicast_add_router(br, port);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~
| br_ip4_multicast_add_router
net/bridge/br_multicast.c: At top level:
net/bridge/br_multicast.c:2804:13: warning: conflicting types for ‘br_ip6_multicast_add_router’
2804 | static void br_ip6_multicast_add_router(struct net_bridge *br,
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~
net/bridge/br_multicast.c:2804:13: error: static declaration of ‘br_ip6_multicast_add_router’ follows non-static declaration
net/bridge/br_multicast.c:1743:3: note: previous implicit declaration of ‘br_ip6_multicast_add_router’ was here
1743 | br_ip6_multicast_add_router(br, port);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~
Fix this build error by moving the definition out of the #ifdef.
Fixes: a3c02e769e ("net: bridge: mcast: split multicast router state for IPv4 and IPv6")
Signed-off-by: Matteo Croce <mcroce@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull drm fixes from Dave Airlie:
"Not much here, mostly amdgpu fixes, with a couple of radeon, and a
cosmetic vc4.
Two MAINTAINERS file updates also.
amdgpu:
- Fixes for flexible array conversions
- Fix sysfs attribute init
- Harvesting fixes
- VCN CG/PG fixes for Picasso
radeon:
- Fixes for flexible array conversions
- Fix for flickering on Oland with multiple 4K displays
vc4:
- drop unused function"
* tag 'drm-fixes-2021-05-14' of git://anongit.freedesktop.org/drm/drm:
drm/amdgpu: update vcn1.0 Non-DPG suspend sequence
drm/amdgpu: set vcn mgcg flag for picasso
drm/radeon/dpm: Disable sclk switching on Oland when two 4K 60Hz monitors are connected
drm/amdgpu: update the method for harvest IP for specific SKU
drm/amdgpu: add judgement when add ip blocks (v2)
drm/amd/display: Initialize attribute for hdcp_srm sysfs file
drm/amd/pm: Fix out-of-bounds bug
drm/radeon/si_dpm: Fix SMU power state load
drm/radeon/ni_dpm: Fix booting bug
MAINTAINERS: Update address for Emma Anholt
MAINTAINERS: Update my e-mail
drm/vc4: remove unused function
drm/ttm: Do not add non-system domain BO into swap list
br_multicast_is_router takes two arguments when bridge IGMP is enabled
and just one when it's disabled, fix the stub to take two as well.
Fixes: 1a3065a268 ("net: bridge: mcast: prepare is-router function for mcast router split")
Signed-off-by: Nikolay Aleksandrov <nikolay@nvidia.com>
Acked-by: Linus Lüssing <linus.luessing@c0d3.blue>
Signed-off-by: David S. Miller <davem@davemloft.net>
To ensure that instructions are observable in a new mapping, the arm64
set_pte_at() implementation cleans the D-cache and invalidates the
I-cache to the PoU. As an optimisation, this is only done on executable
mappings and the PG_dcache_clean page flag is set to avoid future cache
maintenance on the same page.
When two different processes map the same page (e.g. private executable
file or shared mapping) there's a potential race on checking and setting
PG_dcache_clean via set_pte_at() -> __sync_icache_dcache(). While on the
fault paths the page is locked (PG_locked), mprotect() does not take the
page lock. The result is that one process may see the PG_dcache_clean
flag set but the I/D cache maintenance not yet performed.
Avoid test_and_set_bit(PG_dcache_clean) in favour of separate test_bit()
and set_bit(). In the rare event of a race, the cache maintenance is
done twice.
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Cc: <stable@vger.kernel.org>
Cc: Will Deacon <will@kernel.org>
Cc: Steven Price <steven.price@arm.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Acked-by: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20210514095001.13236-1-catalin.marinas@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
If a tag set is shared across request queues (e.g. SCSI LUNs) then the
block layer core keeps track of the number of active request queues in
tags->active_queues. blk_mq_tag_busy() and blk_mq_tag_idle() update that
atomic counter if the hctx flag BLK_MQ_F_TAG_QUEUE_SHARED is set. Make
sure that blk_mq_exit_queue() calls blk_mq_tag_idle() before that flag is
cleared by blk_mq_del_queue_tag_set().
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Hannes Reinecke <hare@suse.com>
Fixes: 0d2602ca30 ("blk-mq: improve support for shared tags maps")
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20210513171529.7977-1-bvanassche@acm.org
Signed-off-by: Jens Axboe <axboe@kernel.dk>
In case of shared sbitmap, request won't be held in plug list any more
sine commit 32bc15afed ("blk-mq: Facilitate a shared sbitmap per
tagset"), this way makes request merge from flush plug list & batching
submission not possible, so cause performance regression.
Yanhui reports performance regression when running sequential IO
test(libaio, 16 jobs, 8 depth for each job) in VM, and the VM disk
is emulated with image stored on xfs/megaraid_sas.
Fix the issue by recovering original behavior to allow to hold request
in plug list.
Cc: Yanhui Ma <yama@redhat.com>
Cc: John Garry <john.garry@huawei.com>
Cc: Bart Van Assche <bvanassche@acm.org>
Cc: kashyap.desai@broadcom.com
Fixes: 32bc15afed ("blk-mq: Facilitate a shared sbitmap per tagset")
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20210514022052.1047665-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
xen_swiotlb_init calls swiotlb_late_init_with_tbl, which fails with
-ENOMEM if the swiotlb has already been initialized.
Add an explicit check io_tlb_default_mem != NULL at the beginning of
xen_swiotlb_init. If the swiotlb is already initialized print a warning
and return -EEXIST.
On x86, the error propagates.
On ARM, we don't actually need a special swiotlb buffer (yet), any
buffer would do. So ignore the error and continue.
CC: boris.ostrovsky@oracle.com
CC: jgross@suse.com
Signed-off-by: Stefano Stabellini <stefano.stabellini@xilinx.com>
Reviewed-by: Boris Ostrovsky <boris.ostrvsky@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Link: https://lore.kernel.org/r/20210512201823.1963-3-sstabellini@kernel.org
Signed-off-by: Juergen Gross <jgross@suse.com>
Commit ef84928cff ("uio/uio_pci_generic: use device-managed function
equivalents") was able to simplify various error paths thanks to no
longer having to clean up on the way out. Some error paths were dropped,
others were simplified. In one of those simplifications, the return
value was accidentally changed from -ENODEV to -ENOMEM. Restore the old
return value.
Fixes: ef84928cff ("uio/uio_pci_generic: use device-managed function equivalents")
Cc: stable <stable@vger.kernel.org>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Martin Ågren <martin.agren@gmail.com>
Link: https://lore.kernel.org/r/20210422192240.1136373-1-martin.agren@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Mackie d.2 has an extension card for IEEE 1394 communication, which uses
BridgeCo DM1000 ASIC. On the other hand, Mackie d.4 Pro has built-in
function for IEEE 1394 communication by Oxford Semiconductor OXFW971,
according to schematic diagram available in Mackie website. Although I
misunderstood that Mackie d.2 Pro would be also a model with OXFW971,
it's wrong. Mackie d.2 Pro is a model which includes the extension card
as factory settings.
This commit fixes entries in Kconfig and comment in ALSA OXFW driver.
Cc: <stable@vger.kernel.org>
Fixes: fd6f4b0dc1 ("ALSA: bebob: Add skelton for BeBoB based devices")
Fixes: ec4dba5053 ("ALSA: oxfw: Add support for Behringer/Mackie devices")
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Link: https://lore.kernel.org/r/20210513125652.110249-3-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Alesis iO 26 FireWire has two pairs of digital optical interface. It
delivers PCM frames from the interfaces by second isochronous packet
streaming. Although both of the interfaces are available at 44.1/48.0
kHz, first one of them is only available at 88.2/96.0 kHz. It reduces
the number of PCM samples to 4 in Multi Bit Linear Audio data channel
of data blocks on the second isochronous packet streaming.
This commit fixes hardcoded stream formats.
Cc: <stable@vger.kernel.org>
Fixes: 28b208f600 ("ALSA: dice: add parameters of stream formats for models produced by Alesis")
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Link: https://lore.kernel.org/r/20210513125652.110249-2-o-takashi@sakamocchi.jp
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Some interrupt handlers have an "extra" that saves 1 or 2
registers (r14, r15) in the paca save area and makes them available to
use by the handler.
The change to always save nvgprs in exception handlers lead to some
interrupt handlers saving those scratch r14 / r15 registers into the
interrupt frame's GPR saves, which get restored on interrupt exit.
Fix this by always reloading those scratch registers from paca before
the EXCEPTION_COMMON that saves nvgprs.
Fixes: 4228b2c3d2 ("powerpc/64e/interrupt: always save nvgprs on interrupt")
Reported-by: Christian Zigotzky <chzigotzky@xenosoft.de>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Tested-by: Christian Zigotzky <chzigotzky@xenosoft.de>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20210514044008.1955783-1-npiggin@gmail.com
The stf entry barrier fallback is unsafe to execute in a semi-patched
state, which can happen when enabling/disabling the mitigation with
strict kernel RWX enabled and using the hash MMU.
See the previous commit for more details.
Fix it by changing the order in which we patch the instructions.
Note the stf barrier fallback is only used on Power6 or earlier.
Fixes: bd573a8131 ("powerpc/mm/64s: Allow STRICT_KERNEL_RWX again")
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20210513140800.1391706-2-mpe@ellerman.id.au
The entry flush mitigation can be enabled/disabled at runtime. When this
happens it results in the kernel patching its own instructions to
enable/disable the mitigation sequence.
With strict kernel RWX enabled instruction patching happens via a
secondary mapping of the kernel text, so that we don't have to make the
primary mapping writable. With the hash MMU this leads to a hash fault,
which causes us to execute the exception entry which contains the entry
flush mitigation.
This means we end up executing the entry flush in a semi-patched state,
ie. after we have patched the first instruction but before we patch the
second or third instruction of the sequence.
On machines with updated firmware the entry flush is a series of special
nops, and it's safe to to execute in a semi-patched state.
However when using the fallback flush the sequence is mflr/branch/mtlr,
and so it's not safe to execute if we have patched out the mflr but not
the other two instructions. Doing so leads to us corrputing LR, leading
to an oops, for example:
# echo 0 > /sys/kernel/debug/powerpc/entry_flush
kernel tried to execute exec-protected page (c000000002971000) - exploit attempt? (uid: 0)
BUG: Unable to handle kernel instruction fetch
Faulting instruction address: 0xc000000002971000
Oops: Kernel access of bad area, sig: 11 [#1]
LE PAGE_SIZE=64K MMU=Hash SMP NR_CPUS=2048 NUMA pSeries
CPU: 0 PID: 2215 Comm: bash Not tainted 5.13.0-rc1-00010-gda3bb206c9ce #1
NIP: c000000002971000 LR: c000000002971000 CTR: c000000000120c40
REGS: c000000013243840 TRAP: 0400 Not tainted (5.13.0-rc1-00010-gda3bb206c9ce)
MSR: 8000000010009033 <SF,EE,ME,IR,DR,RI,LE> CR: 48428482 XER: 00000000
...
NIP 0xc000000002971000
LR 0xc000000002971000
Call Trace:
do_patch_instruction+0xc4/0x340 (unreliable)
do_entry_flush_fixups+0x100/0x3b0
entry_flush_set+0x50/0xe0
simple_attr_write+0x160/0x1a0
full_proxy_write+0x8c/0x110
vfs_write+0xf0/0x340
ksys_write+0x84/0x140
system_call_exception+0x164/0x2d0
system_call_common+0xec/0x278
The simplest fix is to change the order in which we patch the
instructions, so that the sequence is always safe to execute. For the
non-fallback flushes it doesn't matter what order we patch in.
Fixes: bd573a8131 ("powerpc/mm/64s: Allow STRICT_KERNEL_RWX again")
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20210513140800.1391706-1-mpe@ellerman.id.au
The entry flush mitigation can be enabled/disabled at runtime via a
debugfs file (entry_flush), which causes the kernel to patch itself to
enable/disable the relevant mitigations.
However depending on which mitigation we're using, it may not be safe to
do that patching while other CPUs are active. For example the following
crash:
sleeper[15639]: segfault (11) at c000000000004c20 nip c000000000004c20 lr c000000000004c20
Shows that we returned to userspace with a corrupted LR that points into
the kernel, due to executing the partially patched call to the fallback
entry flush (ie. we missed the LR restore).
Fix it by doing the patching under stop machine. The CPUs that aren't
doing the patching will be spinning in the core of the stop machine
logic. That is currently sufficient for our purposes, because none of
the patching we do is to that code or anywhere in the vicinity.
Fixes: f79643787e ("powerpc/64s: flush L1D on kernel entry")
Cc: stable@vger.kernel.org # v5.10+
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20210506044959.1298123-2-mpe@ellerman.id.au
The STF (store-to-load forwarding) barrier mitigation can be
enabled/disabled at runtime via a debugfs file (stf_barrier), which
causes the kernel to patch itself to enable/disable the relevant
mitigations.
However depending on which mitigation we're using, it may not be safe to
do that patching while other CPUs are active. For example the following
crash:
User access of kernel address (c00000003fff5af0) - exploit attempt? (uid: 0)
segfault (11) at c00000003fff5af0 nip 7fff8ad12198 lr 7fff8ad121f8 code 1
code: 40820128 e93c00d0 e9290058 7c292840 40810058 38600000 4bfd9a81 e8410018
code: 2c030006 41810154 3860ffb6 e9210098 <e94d8ff0> 7d295279 39400000 40820a3c
Shows that we returned to userspace without restoring the user r13
value, due to executing the partially patched STF exit code.
Fix it by doing the patching under stop machine. The CPUs that aren't
doing the patching will be spinning in the core of the stop machine
logic. That is currently sufficient for our purposes, because none of
the patching we do is to that code or anywhere in the vicinity.
Fixes: a048a07d7f ("powerpc/64s: Add support for a store forwarding barrier at kernel entry/exit")
Cc: stable@vger.kernel.org # v4.17+
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20210506044959.1298123-1-mpe@ellerman.id.au
Static maps never really worked with libbpf, because all such maps were always
silently resolved to the very first map. Detect static maps (both legacy and
BTF-defined) and report user-friendly error.
Tested locally by switching few maps (legacy and BTF-defined) in selftests to
static ones and verifying that now libbpf rejects them loudly.
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/bpf/20210513233643.194711-2-andrii@kernel.org
