net: hns3: add protect when handling mac addr list
It used netdev->uc and netdev->mc list in function
hns3_recover_hw_addr() and hns3_remove_hw_addr().
We should add protect for them.
Fixes: f05e21097121 ("net: hns3: Clear mac vlan table entries when unload driver or function reset") Signed-off-by: Jian Shen <[email protected]> Signed-off-by: Peng Li <[email protected]> Signed-off-by: Huazhong Tan <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Huazhong Tan [Thu, 4 Apr 2019 08:17:54 +0000 (16:17 +0800)]
net: hns3: simplify hclgevf_cmd_csq_clean
csq is used as a ring buffer, the value of the desc will be replaced
in next use. This patch removes the unnecessary memset, and just
updates the next_to_clean.
Yunsheng Lin [Thu, 4 Apr 2019 08:17:53 +0000 (16:17 +0800)]
net: hns3: minor optimization for datapath
This patch adds a likely case for hns3_fill_desc and
limits the local variables' scope as much as possible,
also avoid div operation when the tqp_vector->num_tqps
is one.
Peng Li [Thu, 4 Apr 2019 08:17:51 +0000 (16:17 +0800)]
net: hns3: return 0 and print warning when hit duplicate MAC
When set 2 same MAC to different function of one port, IMP
will return error as the later one may modify the origin one.
This will cause bond fail for 2 VFs of one port.
Driver just print warning and return 0 with this patch, so
if set same MAC address, it will return 0 but do not really
configure HW.
net: hns3: modify the VF network port media type acquisition method
Method for obtaining the media type of the VF network port periodically,
regular tasks will not run until the network port UP. When the network
port is DOWN, the network port cannot obtain the media type.
Modifies the media type obtained when initializing the VF network port.
Yonglong Liu [Thu, 4 Apr 2019 08:17:49 +0000 (16:17 +0800)]
net: hns3: reduce resources use in kdump kernel
When the kdump kernel started, the HNS3 driver fail to register:
[14.753340] hns3 0000:7d:00.0: Alloc umv space failed, want 512, get 0
[14.795034] hns3 0000:7d:00.0: add uc mac address fail, ret =-22.
By default, the HNS3 driver will use about 512M memory, but
usually the reserved memory of kdump kernel is 576M, so the HNS3
driver fail to register. This patch reduces the memory use in
kdump kernel to about 16M.
And when the kdump kernel starts, we must clear ucast mac address
first to avoid add fail.
Peng Li [Thu, 4 Apr 2019 08:17:48 +0000 (16:17 +0800)]
net: hns3: check 1000M half for hns3_ethtool_ops.set_link_ksettings
Hip08 SOC does not support 1000M half, this patch adds 1000M half
check for hns3_ethtool_ops.set_link_ksettings, so the user can not
set 1000M half by ethtool.
Qian Cai [Thu, 4 Apr 2019 10:54:41 +0000 (11:54 +0100)]
mm/compaction.c: abort search if isolation fails
Running LTP oom01 in a tight loop or memory stress testing put the system
in a low-memory situation could triggers random memory corruption like
page flag corruption below due to in fast_isolate_freepages(), if
isolation fails, next_search_order() does not abort the search immediately
could lead to improper accesses.
He bisected it down to e332f741a8dd ("mm, compaction: be selective about
what pageblocks to clear skip hints"). The problem is that the patch in
question was sloppy with respect to the handling of zone boundaries. In
some instances, it was possible for PFNs outside of a zone to be examined
and if those were not properly initialised or poisoned then it would
trigger the VM_BUG_ON. This patch corrects the zone boundary issues when
resetting pageblock skip hints and Mikhail reported that the bug did not
trigger after 30 hours of testing.
Merge tag '5.1-rc3-smb3-fixes' of git://git.samba.org/sfrench/cifs-2.6
Pull smb3 fixes from Steve French:
"Four smb3 fixes for stable:
- fix an open path where we had an unitialized structure
- fix for snapshot (previous version) enumeration
- allow reconnect timeout on handles to be configurable to better
handle network or server crash
- correctly handle lack of file_all_info structure"
* tag '5.1-rc3-smb3-fixes' of git://git.samba.org/sfrench/cifs-2.6:
cifs: a smb2_validate_and_copy_iov failure does not mean the handle is invalid.
SMB3: Allow persistent handle timeout to be configurable on mount
smb3: Fix enumerating snapshots to Azure
cifs: fix kref underflow in close_shroot()
David S. Miller [Thu, 4 Apr 2019 04:50:20 +0000 (21:50 -0700)]
Merge branch 'net-More-movement-to-fib_nh_common'
David Ahern says:
====================
net: More movement to fib_nh_common
Second set of three with the end goal of enabling IPv6 gateways with IPv4
routes.
This set moves:
- the ipv4 tracepoint to take a fib_nh_common and updates it to handle
a v6 gateway.
- consolidates route notifications to use the same fill functions
for both ipv4 and ipv6
v4
- enhanced the commit message for patches 1 and 2
v3
- comments from Martin:
+ renamed FIB_RES_NH to FIB_RES_NHC
+ removed family check from fib_result_prefsrc
+ in fib_nexthop_info, renamed nexthop arg to nhc and dropped for_ipv4 arg
v2
- dropped patches moving cached routes and exception buckets to
fib_nh_common. The goal is allowing a fib6_nh to be used with an
IPv4 route. The hold up is the need for separate exception buckets -
one for v6 routes and one for v4 routes. When all of the nexthop patches
are in, adding a secondi exception bucket pushes IPv6 fib6_info
allocations over 256 which means fib6_info allocations roll up to 512.
Hence, deferring the patches until some data mining can be done to keep
the allocations at 256.
====================
David Ahern [Tue, 2 Apr 2019 21:11:58 +0000 (14:11 -0700)]
ipv6: Flip to fib_nexthop_info
Export fib_nexthop_info and fib_add_nexthop for use by IPv6 code.
Remove rt6_nexthop_info and rt6_add_nexthop in favor of the IPv4
versions. Update fib_nexthop_info for IPv6 linkdown check and
RTA_GATEWAY for AF_INET6.
David Ahern [Tue, 2 Apr 2019 21:11:57 +0000 (14:11 -0700)]
ipv4: Change fib_nexthop_info and fib_add_nexthop to take fib_nh_common
With the exception of the nexthop weight, the nexthop attributes used by
fib_nexthop_info and fib_add_nexthop come from the fib_nh_common struct.
Update both to use it and change fib_nexthop_info to check the family
as needed.
nexthop weight comes from the common struct for existing use cases, but
for nexthop groups the weight is outside of the fib_nh_common to allow
the same nexthop definition to be used in multiple groups with different
weights.
David Ahern [Tue, 2 Apr 2019 21:11:56 +0000 (14:11 -0700)]
ipv4: Refactor nexthop attributes in fib_dump_info
Similar to ipv6, move addition of nexthop attributes to dump
message into helpers that are called for both single path and
multipath routes. Align the new helpers to the IPv6 variant
which most notably means computing the flags argument based on
settings in nh_flags.
The RTA_FLOW argument is unique to IPv4, so it is appended after
the new fib_nexthop_info helper. The intent of a later patch is to
make both fib_nexthop_info and fib_add_nexthop usable for both IPv4
and IPv6. This patch is stepping stone in that direction.
David Ahern [Tue, 2 Apr 2019 21:11:54 +0000 (14:11 -0700)]
ipv4: Update fib_table_lookup tracepoint to take common nexthop
Update fib_table_lookup tracepoint to take a fib_nh_common struct and
dump the v6 gateway address if the nexthop uses it.
Over the years saddr has not proven useful and the output of the
tracepoint produces very long lines. Since saddr is not part of
fib_nh_common, drop it. If it needs to be added later, fib_nh which
contains saddr can be obtained from a fib_nh_common via container_of.
net: phy: fix autoneg mismatch case in genphy_read_status
The original patch didn't consider the case that autoneg process
finishes successfully but both link partners have no mode in common.
In this case there's no link, nevertheless we may be interested in
what the link partner advertised.
Like phydev->link we set phydev->autoneg_complete in
genphy_update_link() and use the stored value in genphy_read_status().
This way we don't have to read register BMSR again.
Fixes: b6163f194c69 ("net: phy: improve genphy_read_status") Signed-off-by: Heiner Kallweit <[email protected]> Reviewed-by: Andrew Lunn <[email protected]> Signed-off-by: David S. Miller <[email protected]>
This series implements the suspend/resume callbacks in the marvell10g
PHY driver.
Thanks,
Antoine
Since v3:
- Use the new phy_set/clear_bits_mmd() instead of phy_modify_mmd().
- Use VEND2.f001.11 to power down the port instead of the per-mode
LPOWER.
Since v2:
- Removed the third patch, setting the PHY in low power by default, as
the change was controversial.
- Rebased on the latest net-next.
Since v1:
- Fixed a mix up in the patches where two implementations of the
suspend/resume callbacks were kept in the driver.
- Rebased on the latest net-next.
====================
This patch adds the suspend/resume callbacks for Marvell 10G PHYs. The
three PCS (base-t, base-r and 1000base-x) are set in low power (the PCS
are powered down) when the PHY isn't used.
Junwei Hu [Tue, 2 Apr 2019 11:38:04 +0000 (19:38 +0800)]
ipv6: Fix dangling pointer when ipv6 fragment
At the beginning of ip6_fragment func, the prevhdr pointer is
obtained in the ip6_find_1stfragopt func.
However, all the pointers pointing into skb header may change
when calling skb_checksum_help func with
skb->ip_summed = CHECKSUM_PARTIAL condition.
The prevhdr pointe will be dangling if it is not reloaded after
calling __skb_linearize func in skb_checksum_help func.
Here, I add a variable, nexthdr_offset, to evaluate the offset,
which does not changes even after calling __skb_linearize func.
net-gro: Fix GRO flush when receiving a GSO packet.
Currently we may merge incorrectly a received GSO packet
or a packet with frag_list into a packet sitting in the
gro_hash list. skb_segment() may crash case because
the assumptions on the skb layout are not met.
The correct behaviour would be to flush the packet in the
gro_hash list and send the received GSO packet directly
afterwards. Commit d61d072e87c8e ("net-gro: avoid reorders")
sets NAPI_GRO_CB(skb)->flush in this case, but this is not
checked before merging. This patch makes sure to check this
flag and to not merge in that case.
Fixes: d61d072e87c8e ("net-gro: avoid reorders") Signed-off-by: Steffen Klassert <[email protected]> Signed-off-by: David S. Miller <[email protected]>
net: dsa: microchip: mark expected switch fall-through
In preparation to enabling -Wimplicit-fallthrough, mark switch cases
where we are expecting to fall through.
This patch fixes the following warning:
drivers/net/dsa/microchip/ksz9477.c: In function ‘ksz9477_get_interface’:
drivers/net/dsa/microchip/ksz9477.c:1145:6: warning: this statement may fall through [-Wimplicit-fallthrough=]
if (gbit)
^
drivers/net/dsa/microchip/ksz9477.c:1147:2: note: here
case 0:
^~~~
Warning level 3 was used: -Wimplicit-fallthrough=3
This patch is part of the ongoing efforts to enable
-Wimplicit-fallthrough.
In preparation to enabling -Wimplicit-fallthrough, mark switch cases
where we are expecting to fall through.
This patch fixes the following warning:
net/rxrpc/local_object.c: In function ‘rxrpc_open_socket’:
net/rxrpc/local_object.c:175:6: warning: this statement may fall through [-Wimplicit-fallthrough=]
if (ret < 0) {
^
net/rxrpc/local_object.c:184:2: note: here
case AF_INET:
^~~~
Warning level 3 was used: -Wimplicit-fallthrough=3
Currently, GCC is expecting to find the fall-through annotations
at the very bottom of the case and on its own line. That's why
I had to add the annotation, although the intentional fall-through
is already mentioned in a few lines above.
This patch is part of the ongoing efforts to enable
-Wimplicit-fallthrough.
Chris Wilson [Tue, 5 Feb 2019 20:30:33 +0000 (20:30 +0000)]
drm/i915/gvt: Fix kerneldoc typo for intel_vgpu_emulate_hotplug
drivers/gpu/drm/i915/gvt/display.c:457: warning: Function parameter or member 'connected' not described in 'intel_vgpu_emulate_hotplug'
drivers/gpu/drm/i915/gvt/display.c:457: warning: Excess function parameter 'conncted' description in 'intel_vgpu_emulate_hotplug'
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/hid/hid
Pull HID fixes from Jiri Kosina:
- build dependency fix for hid-asus from Arnd Bergmann
- addition of omitted mapping of _ASSISTANT key from Dmitry Torokhov
- race condition fix in hid-debug inftastructure from He, Bo
- fixed support for devices with big maximum report size from Kai-Heng
Feng
- deadlock fix in hid-steam from Rodrigo Rivas Costa
- quite a few device-specific quirks
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/hid/hid:
HID: input: add mapping for Assistant key
HID: i2c-hid: Disable runtime PM on Synaptics touchpad
HID: quirks: Fix keyboard + touchpad on Lenovo Miix 630
HID: logitech: Handle 0 scroll events for the m560
HID: debug: fix race condition with between rdesc_show() and device removal
HID: logitech: check the return value of create_singlethread_workqueue
HID: Increase maximum report size allowed by hid_field_extract()
HID: steam: fix deadlock with input devices.
HID: uclogic: remove redudant duplicated null check on ver_ptr
HID: quirks: Drop misused kernel-doc annotation
HID: hid-asus: select CONFIG_POWER_SUPPLY
HID: quirks: use correct format chars in dbg_hid
Daniel Borkmann [Wed, 3 Apr 2019 14:49:49 +0000 (16:49 +0200)]
Merge branch 'bpf-flow-dissector-fixes'
Stanislav Fomichev says:
====================
This patch series fixes the existing BPF flow dissector API to
support calling BPF progs from the eth_get_headlen context (the
support itself will be added in bpf-next tree).
The summary of the changes:
* fix VLAN handling in bpf_flow.c, we don't need to peek back and look
at skb->vlan_present; add selftests
* pass and use flow_keys->n_proto instead of skb->protocol
* fix clamping of flow_keys->nhoff for packets with nhoff > 0
* prohibit access to most of the __sk_buff fields from BPF flow
dissector progs; only data/data_end/flow_keys are allowed (all input
is now passed via flow_keys)
* finally, document BPF flow dissector program environment
====================
flow_dissector: allow access only to a subset of __sk_buff fields
Use whitelist instead of a blacklist and allow only a small set of
fields that might be relevant in the context of flow dissector:
* data
* data_end
* flow_keys
This is required for the eth_get_headlen case where we have only a
chunk of data to dissect (i.e. trying to read the other skb fields
doesn't make sense).
Note, that it is a breaking API change! However, we've provided
flow_keys->n_proto as a substitute for skb->protocol; and there is
no need to manually handle skb->vlan_present. So even if we
break somebody, the migration is trivial. Unfortunately, we can't
support eth_get_headlen use-case without those breaking changes.
flow_dissector: fix clamping of BPF flow_keys for non-zero nhoff
Don't allow BPF program to set flow_keys->nhoff to less than initial
value. We currently don't read the value afterwards in anything but
the tests, but it's still a good practice to return consistent
values to the test programs.
selftests/bpf: fix vlan handling in flow dissector program
When we tail call PROG(VLAN) from parse_eth_proto we don't need to peek
back to handle vlan proto because we didn't adjust nhoff/thoff yet. Use
flow_keys->n_proto, that we set in parse_eth_proto instead and
properly increment nhoff as well.
Also, always use skb->protocol and don't look at skb->vlan_present.
skb->vlan_present indicates that vlan information is stored out-of-band
in skb->vlan_{tci,proto} and vlan header is already pulled from skb.
That means, skb->vlan_present == true is not relevant for BPF flow
dissector.
Add simple test cases with VLAN tagged frames:
* single vlan for ipv4
* double vlan for ipv6
Maxime Ripard [Fri, 22 Mar 2019 09:16:50 +0000 (10:16 +0100)]
mfd: sun6i-prcm: Allow to compile with COMPILE_TEST
Since this driver only has a dependency on ARCH_SUNXI just because it
doesn't make any sense to run it on something else, we can definitely
enable it through COMPILE_TEST as well to get some build coverage.
Merge tag 'pidfd-fixes-v5.1-rc3' of gitolite.kernel.org:pub/scm/linux/kernel/git/brauner/linux
Pull pidfd fix from Christian Brauner:
"This should be an uncontroversial fix for pidfd_send_signal() by Jann
to better align it's behavior with other signal sending functions:
In one of the early versions of the patchset it was suggested to not
unconditionally error out when a signal with SI_USER is sent to a
non-current task (cf. [1]).
Instead, pidfd_send_signal() currently silently changes this to a
regular kill signal. While this is technically fine, the semantics are
weird since the kernel just silently converts a user's request behind
their back and also no other signal sending function allows to do
this. It gets more hairy when we introduce sending signals to a
specific thread soon.
So let's align pidfd_send_signal() with all the other signal sending
functions and error out when SI_USER signals are sent to a non-current
task"
* tag 'pidfd-fixes-v5.1-rc3' of gitolite.kernel.org:pub/scm/linux/kernel/git/brauner/linux:
signal: don't silently convert SI_USER signals to non-current pidfd
Le Ma [Mon, 1 Apr 2019 10:08:30 +0000 (18:08 +0800)]
drm/amdgpu: remove unnecessary rlc reset function on gfx9
The rlc reset function is not necessary during gfx9 initialization/resume phase.
And this function would even cause rlc fw loading failed on some gfx9 ASIC.
Remove this function safely with verification well on Vega/Raven platform.
David S. Miller [Tue, 2 Apr 2019 20:27:11 +0000 (13:27 -0700)]
Merge branch '40GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-queue
Jeff Kirsher says:
====================
Intel Wired LAN Driver Fixes 2019-04-01
This series contains two fixes for XDP in the i40e driver.
Björn provides both fixes, first moving a function out of the header and
into the main.c file. Second fixes a regression introduced in an
earlier patch that removed umem from the VSI. This caused an issue
because the setup code would try to enable AF_XDP zero copy
unconditionally, as long as there was a umem placed in the netdev
receive structure.
====================
Action tunnel_key doesn't have a metadata/tunnel for release(decap) action.
Drivers do not dereference entry->tunnel pointer for that action type, so
this behavior doesn't result in a crash at the moment. However, this needs
to be corrected as a preparation for updating hardware offloads API to not
rely on rtnl lock, for which flow_action code will copy the tunnel data to
temporary buffer to prevent concurrent action overwrite from
invalidating/freeing it.
Fixes: 3a7b68617de7 ("cls_api: add translator to flow_action representation") Signed-off-by: Vlad Buslov <[email protected]> Signed-off-by: David S. Miller <[email protected]>
The device type for ip6 tunnels is set to
ARPHRD_TUNNEL6. However, the ip4ip6_err function
is expecting the device type of the tunnel to be
ARPHRD_TUNNEL. Since the device types do not
match, the function exits and the ICMP error
packet is not sent to the originating host. Note
that the device type for IPv4 tunnels is set to
ARPHRD_TUNNEL.
Fix is to expect a tunnel device type of
ARPHRD_TUNNEL6 instead. Now the tunnel device
type matches and the ICMP error packet is sent
to the originating host.
This series improves genphy_c45_read_lpa:
- Use clause 45 standard register / bit to detect link partner autoneg
capability.
- Consider that lpa register values may be invalid if "autoneg complete"
bit isn't set.
====================
Heiner Kallweit [Sun, 31 Mar 2019 17:54:07 +0000 (19:54 +0200)]
net: phy: deal properly with autoneg incomplete in genphy_c45_read_lpa
The link partner advertisement registers are not guaranteed to contain
valid values if autoneg is incomplete. Therefore, if
MDIO_AN_STAT1_COMPLETE isn't set, let's clear all link partner
capability bits. This also avoids unnecessary register reads if link
is down and phylib is in polling mode.
Heiner Kallweit [Sun, 31 Mar 2019 17:52:51 +0000 (19:52 +0200)]
net: phy: use c45 standard to detect link partner autoneg capability
Currently mii_lpa_mod_linkmode_lpa_t() checks bit LPA_LPACK to detect
whether link partner supports autoneg. This doesn't work correctly at
least on Aquantia AQCS109 when it negotiates 1000Base-T2 mode.
The "link partner is autoneg-capable" bit as defined by clause 45 is
set however. Better let's switch in general to use the clause 45
standard for link partner autoneg detection.
David S. Miller [Tue, 2 Apr 2019 20:09:56 +0000 (13:09 -0700)]
Merge branch 'genphy_read_abilities'
Heiner Kallweit says:
====================
net: phy: add and use new function genphy_read_abilities
Similar to genphy_c45_pma_read_abilities() add a function to dynamically
detect the abilities of a Clause 22 PHY. This is mainly copied from
genphy_config_init(). Main benefit is that PHY drivers no longer have
to specify whether they support GBit or not (provided they keep to
the C22 standard).
====================
Heiner Kallweit [Wed, 27 Mar 2019 20:59:33 +0000 (21:59 +0100)]
net: phy: use genphy_read_abilities in genphy driver
Currently the genphy driver populates phydev->supported like this:
First all possible feature bits are set, then genphy_config_init()
reads the available features from the chip and remove all unsupported
features from phydev->supported. This can be simplified by using
genphy_read_abilities().
Heiner Kallweit [Wed, 27 Mar 2019 20:58:44 +0000 (21:58 +0100)]
net: phy: add genphy_read_abilities
Similar to genphy_c45_pma_read_abilities() add a function to dynamically
detect the abilities of a Clause 22 PHY. This is mainly copied from
genphy_config_init().
Dan Carpenter [Tue, 2 Apr 2019 06:53:14 +0000 (09:53 +0300)]
openvswitch: use after free in __ovs_ct_free_action()
We free "ct_info->ct" and then use it on the next line when we pass it
to nf_ct_destroy_timeout(). This patch swaps the order to avoid the use
after free.
Fixes: 06bd2bdf19d2 ("openvswitch: Add timeout support to ct action") Signed-off-by: Dan Carpenter <[email protected]> Acked-by: Yi-Hung Wei <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Jon Maxwell [Tue, 2 Apr 2019 05:07:56 +0000 (16:07 +1100)]
tg3: allow ethtool -p to work for NICs in down state
Make tg3 behave like other drivers and let "ethtool -p" identify the
NIC even when it's in the DOWN state. Before this patch it would get an
error as follows if the NIC was down:
# ip link set down dev em4
# ethtool -p em4
Cannot identify NIC: Resource temporarily unavailable
With this patch ethtool identify works regardless of whether the NIC is up
or down as it does for other drivers.
Andreas Kemnade [Sat, 23 Feb 2019 11:47:54 +0000 (12:47 +0100)]
mfd: twl-core: Disable IRQ while suspended
Since commit 6e2bd956936 ("i2c: omap: Use noirq system sleep pm ops to idle device for suspend")
on gta04 we have handle_twl4030_pih() called in situations where pm_runtime_get()
in i2c-omap.c returns -EACCES.
This happens when we wakeup via something behing twl4030 (powerbutton or rtc
alarm). This goes on for minutes until the system is finally resumed.
Disable the irq on suspend and enable it on resume to avoid
having i2c access problems when the irq registers are checked.
Fixes: 6e2bd956936 ("i2c: omap: Use noirq system sleep pm ops to idle device for suspend") Signed-off-by: Andreas Kemnade <[email protected]> Tested-by: Tony Lindgren <[email protected]> Signed-off-by: Lee Jones <[email protected]>
macsec: add noinline tag to avoid a frame size warning
seen with debug config:
drivers/net/macsec.c: In function 'dump_secy':
drivers/net/macsec.c:2597: warning: the frame size of 2216 bytes is larger
than 2048 bytes [-Wframe-larger-than=]
just mark it with noinline_for_stack, this is netlink dump code.
v2: use 'static noinline_for_stack int' consistently
First, the xmit_more flag conceptually doesn't fit into the skb, as
xmit_more is not a property related to the skb.
Its only a hint to the driver that the stack is about to transmit another
packet immediately.
Second, it was only done this way to not have to pass another argument
to ndo_start_xmit().
We can place xmit_more in the softnet data, next to the device recursion.
The recursion counter is already written to on each transmit. The "more"
indicator is placed right next to it.
Drivers can use the netdev_xmit_more() helper instead of skb->xmit_more
to check the "more packets coming" hint.
skb->xmit_more is retained (but always 0) to not cause build breakage.
This change takes care of the simple s/skb->xmit_more/netdev_xmit_more()/
conversions. Remaining drivers are converted in the next patches.
Heiner Kallweit [Sun, 31 Mar 2019 15:42:24 +0000 (17:42 +0200)]
net: phy: aquantia: add SGMII statistics
The AQR107 family has SGMII statistics counters. Let's expose them to
ethtool. To interpret the counters correctly one has to be aware that
rx on SGMII side is tx on ethernet side. The counters are populated
by the chip in 100Mbps/1Gbps mode only.
v2:
- add constant AQR107_SGMII_STAT_SZ
- add struct aqr107_priv to be prepared for more private data fields
- let aqr107_get_stat() return U64_MAX in case of an error
Xin Long [Sun, 31 Mar 2019 09:03:02 +0000 (17:03 +0800)]
net: use rcu_dereference_protected to fetch sk_dst_cache in sk_destruct
As Eric noticed, in .sk_destruct, sk->sk_dst_cache update is prevented, and
no barrier is needed for this. So change to use rcu_dereference_protected()
instead of rcu_dereference_check() to fetch sk_dst_cache in there.
v1->v2:
- no change, repost after net-next is open.
Xin Long [Sun, 31 Mar 2019 08:58:15 +0000 (16:58 +0800)]
sctp: initialize _pad of sockaddr_in before copying to user memory
Syzbot report a kernel-infoleak:
BUG: KMSAN: kernel-infoleak in _copy_to_user+0x16b/0x1f0 lib/usercopy.c:32
Call Trace:
_copy_to_user+0x16b/0x1f0 lib/usercopy.c:32
copy_to_user include/linux/uaccess.h:174 [inline]
sctp_getsockopt_peer_addrs net/sctp/socket.c:5911 [inline]
sctp_getsockopt+0x1668e/0x17f70 net/sctp/socket.c:7562
...
Uninit was stored to memory at:
sctp_transport_init net/sctp/transport.c:61 [inline]
sctp_transport_new+0x16d/0x9a0 net/sctp/transport.c:115
sctp_assoc_add_peer+0x532/0x1f70 net/sctp/associola.c:637
sctp_process_param net/sctp/sm_make_chunk.c:2548 [inline]
sctp_process_init+0x1a1b/0x3ed0 net/sctp/sm_make_chunk.c:2361
...
Bytes 8-15 of 16 are uninitialized
It was caused by that th _pad field (the 8-15 bytes) of a v4 addr (saved in
struct sockaddr_in) wasn't initialized, but directly copied to user memory
in sctp_getsockopt_peer_addrs().
So fix it by calling memset(addr->v4.sin_zero, 0, 8) to initialize _pad of
sockaddr_in before copying it to user memory in sctp_v4_addr_to_user(), as
sctp_v6_addr_to_user() does.
Heiner Kallweit [Sat, 30 Mar 2019 09:22:45 +0000 (10:22 +0100)]
net: phy: improve genphy_read_status
This patch improves few aspects of genphy_read_status():
- Don't initialize lpagb, it's not needed.
- Move initializing phydev->speed et al before the if clause.
- In auto-neg case, skip populating lp_advertising if we
don't have a link. This avoids quite some unnecessary
MDIO reads in case of phylib polling mode.
====================
nfp: flower improvement and SFF module EEPROM
The first patch in this series from Pieter improves the
handling of mangle actions in TC flower offload. These
used to be sent down to the driver in groups, but after
Pablo N's patches they are split out causing suboptimal
expression.
The ramaining two patches from Dirk add support for reading
SFF module EEPROM data.
====================
Now that the NSP provides the ability to read from the SFF modules'
EEPROM, we can use this interface to implement the ethtool callback.
If the NSP only provides partial data, we log the event from within
the driver but pass a success code to ethtool to prevent it from
discarding the partial data.
nfp: flower: reduce action list size by coalescing mangle actions
With the introduction of flow_action_for_each pedit actions are no
longer grouped together, instead pedit actions are broken out per
32 byte word. This results in an inefficient use of the action list
that is pushed to hardware where each 32 byte word becomes its own
action. Therefore we combine groups of 32 byte word before sending
the action list to hardware.
====================
nfp: flower: fix matching and pushing vlan CFI bit
This patch clears up some confusion around the meaning of bit 12
for FW messages related to VLAN and flower offload.
Pieter says:
It fixes issues with matching, pushing and popping vlan tags.
We replace the vlan CFI bit with a vlan present bit that
indicates the presence of a vlan tag. We also no longer set
the CFI when pushing vlan tags.
====================
nfp: flower: remove vlan CFI bit from push vlan action
We no longer set CFI when pushing vlan tags, therefore we remove
the CFI bit from push vlan.
Fixes: 1a1e586f54bf ("nfp: add basic action capabilities to flower offloads") Signed-off-by: Pieter Jansen van Vuuren <[email protected]> Signed-off-by: Louis Peens <[email protected]> Reviewed-by: Jakub Kicinski <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Replace vlan CFI bit with a vlan present bit that indicates the
presence of a vlan tag. Previously the driver incorrectly assumed
that an vlan id of 0 is not matchable, therefore we indicate vlan
presence with a vlan present bit.
One of the more common cases of allocation size calculations is finding
the size of a structure that has a zero-sized array at the end, along
with memory for some number of elements for that array. For example:
The use of zero-sized array causes undefined behaviour when it is not
the last member in a structure. As it happens to be in this case.
Also, the current code makes use of a language extension to the C90
standard, but the preferred mechanism to declare variable-length
types such as this one is a flexible array member, introduced in
C99:
struct foo {
int stuff;
struct boo array[];
};
By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last. Which is beneficial
to cultivate a high-quality code.
Fixes: e48f129c2f20 ("[SCSI] cxgb3i: convert cdev->l2opt to use rcu to prevent NULL dereference") Signed-off-by: Gustavo A. R. Silva <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Jiri Slaby [Fri, 29 Mar 2019 11:19:46 +0000 (12:19 +0100)]
kcm: switch order of device registration to fix a crash
When kcm is loaded while many processes try to create a KCM socket, a
crash occurs:
BUG: unable to handle kernel NULL pointer dereference at 000000000000000e
IP: mutex_lock+0x27/0x40 kernel/locking/mutex.c:240
PGD 8000000016ef2067 P4D 8000000016ef2067 PUD 3d6e9067 PMD 0
Oops: 0002 [#1] SMP KASAN PTI
CPU: 0 PID: 7005 Comm: syz-executor.5 Not tainted 4.12.14-396-default #1 SLE15-SP1 (unreleased)
RIP: 0010:mutex_lock+0x27/0x40 kernel/locking/mutex.c:240
RSP: 0018:ffff88000d487a00 EFLAGS: 00010246
RAX: 0000000000000000 RBX: 000000000000000e RCX: 1ffff100082b0719
...
CR2: 000000000000000e CR3: 000000004b1bc003 CR4: 0000000000060ef0
Call Trace:
kcm_create+0x600/0xbf0 [kcm]
__sock_create+0x324/0x750 net/socket.c:1272
...
This is due to race between sock_create and unfinished
register_pernet_device. kcm_create tries to do "net_generic(net,
kcm_net_id)". but kcm_net_id is not initialized yet.
So switch the order of the two to close the race.
This can be reproduced with mutiple processes doing socket(PF_KCM, ...)
and one process doing module removal.
Fixes: ab7ac4eb9832 ("kcm: Kernel Connection Multiplexor module") Reviewed-by: Michal Kubecek <[email protected]> Signed-off-by: Jiri Slaby <[email protected]> Signed-off-by: David S. Miller <[email protected]>
====================
net: sched: fix stats accounting for child NOLOCK qdiscs
Currently, stats accounting for NOLOCK qdisc enslaved to classful (lock)
qdiscs is buggy. Per CPU values are ignored in most places, as a result,
stats dump in the above scenario always report 0 length backlog and parent
backlog len is not updated correctly on NOLOCK qdisc removal.
The first patch address stats dumping, and the second one child qdisc removal.
I'm targeting the net tree as this is a bugfix, but it could be moved to
net-next due to the relatively large diffstat.
====================
Paolo Abeni [Thu, 28 Mar 2019 15:53:13 +0000 (16:53 +0100)]
net: sched: introduce and use qdisc tree flush/purge helpers
The same code to flush qdisc tree and purge the qdisc queue
is duplicated in many places and in most cases it does not
respect NOLOCK qdisc: the global backlog len is used and the
per CPU values are ignored.
This change addresses the above, factoring-out the relevant
code and using the helpers introduced by the previous patch
to fetch the correct backlog len.
Fixes: c5ad119fb6c0 ("net: sched: pfifo_fast use skb_array") Signed-off-by: Paolo Abeni <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Paolo Abeni [Thu, 28 Mar 2019 15:53:12 +0000 (16:53 +0100)]
net: sched: introduce and use qstats read helpers
Classful qdiscs can't access directly the child qdiscs backlog
length: if such qdisc is NOLOCK, per CPU values should be
accounted instead.
Most qdiscs no not respect the above. As a result, qstats fetching
for most classful qdisc is currently incorrect: if the child qdisc is
NOLOCK, it always reports 0 len backlog.
This change introduces a pair of helpers to safely fetch
both backlog and qlen and use them in stats class dumping
functions, fixing the above issue and cleaning a bit the code.
DRR needs also to access the child qdisc queue length, so it
needs custom handling.
Fixes: c5ad119fb6c0 ("net: sched: pfifo_fast use skb_array") Signed-off-by: Paolo Abeni <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Nicolas Dichtel [Thu, 28 Mar 2019 09:35:06 +0000 (10:35 +0100)]
net/sched: fix ->get helper of the matchall cls
It returned always NULL, thus it was never possible to get the filter.
Example:
$ ip link add foo type dummy
$ ip link add bar type dummy
$ tc qdisc add dev foo clsact
$ tc filter add dev foo protocol all pref 1 ingress handle 1234 \
matchall action mirred ingress mirror dev bar
Before the patch:
$ tc filter get dev foo protocol all pref 1 ingress handle 1234 matchall
Error: Specified filter handle not found.
We have an error talking to the kernel
After:
$ tc filter get dev foo protocol all pref 1 ingress handle 1234 matchall
filter ingress protocol all pref 1 matchall chain 0 handle 0x4d2
not_in_hw
action order 1: mirred (Ingress Mirror to device bar) pipe
index 1 ref 1 bind 1
Jann Horn [Sat, 30 Mar 2019 02:12:32 +0000 (03:12 +0100)]
signal: don't silently convert SI_USER signals to non-current pidfd
The current sys_pidfd_send_signal() silently turns signals with explicit
SI_USER context that are sent to non-current tasks into signals with
kernel-generated siginfo.
This is unlike do_rt_sigqueueinfo(), which returns -EPERM in this case.
If a user actually wants to send a signal with kernel-provided siginfo,
they can do that with pidfd_send_signal(pidfd, sig, NULL, 0); so allowing
this case is unnecessary.
Instead of silently replacing the siginfo, just bail out with an error;
this is consistent with other interfaces and avoids special-casing behavior
based on security checks.
Steve French [Fri, 29 Mar 2019 21:31:07 +0000 (16:31 -0500)]
SMB3: Allow persistent handle timeout to be configurable on mount
Reconnecting after server or network failure can be improved
(to maintain availability and protect data integrity) by allowing
the client to choose the default persistent (or resilient)
handle timeout in some use cases. Today we default to 0 which lets
the server pick the default timeout (usually 120 seconds) but this
can be problematic for some workloads. Add the new mount parameter
to cifs.ko for SMB3 mounts "handletimeout" which enables the user
to override the default handle timeout for persistent (mount
option "persistenthandles") or resilient handles (mount option
"resilienthandles"). Maximum allowed is 16 minutes (960000 ms).
Units for the timeout are expressed in milliseconds. See
section 2.2.14.2.12 and 2.2.31.3 of the MS-SMB2 protocol
specification for more information.
Steve French [Fri, 29 Mar 2019 03:32:49 +0000 (22:32 -0500)]
smb3: Fix enumerating snapshots to Azure
Some servers (see MS-SMB2 protocol specification
section 3.3.5.15.1) expect that the FSCTL enumerate snapshots
is done twice, with the first query having EXACTLY the minimum
size response buffer requested (16 bytes) which refreshes
the snapshot list (otherwise that and subsequent queries get
an empty list returned). So had to add code to set
the maximum response size differently for the first snapshot
query (which gets the size needed for the second query which
contains the actual list of snapshots).
Ronnie Sahlberg [Thu, 28 Mar 2019 01:20:02 +0000 (11:20 +1000)]
cifs: fix kref underflow in close_shroot()
Fix a bug where we used to not initialize the cached fid structure at all
in open_shroot() if the open was successful but we did not get a lease.
This would leave the structure uninitialized and later when we close the handle
we would in close_shroot() try to kref_put() an uninitialized refcount.
Fix this by always initializing this structure if the open was successful
but only do the extra get() if we got a lease.
This extra get() is only used to hold the structure until we get a lease
break from the server at which point we will kref_put() it during lease
processing.
Björn Töpel [Tue, 12 Feb 2019 08:52:05 +0000 (09:52 +0100)]
i40e: add tracking of AF_XDP ZC state for each queue pair
In commit f3fef2b6e1cc ("i40e: Remove umem from VSI") a regression was
introduced; When the VSI was reset, the setup code would try to enable
AF_XDP ZC unconditionally (as long as there was a umem placed in the
netdev._rx struct). Here, we add a bitmap to the VSI that tracks if a
certain queue pair has been "zero-copy enabled" via the ndo_bpf. The
bitmap is used in i40e_xsk_umem, and enables zero-copy if and only if
XDP is enabled, the corresponding qid in the bitmap is set and the
umem is non-NULL.
vrf: check accept_source_route on the original netdevice
Configuration check to accept source route IP options should be made on
the incoming netdevice when the skb->dev is an l3mdev master. The route
lookup for the source route next hop also needs the incoming netdev.
v2->v3:
- Simplify by passing the original netdevice down the stack (per David
Ahern).
Chris Wilson [Fri, 29 Mar 2019 16:51:52 +0000 (16:51 +0000)]
drm/i915: Always backoff after a drm_modeset_lock() deadlock
If drm_modeset_lock() reports a deadlock it sets the ctx->contexted
field and insists that the caller calls drm_modeset_backoff() or else it
generates a WARN on cleanup.
Eric Dumazet [Mon, 1 Apr 2019 10:09:20 +0000 (03:09 -0700)]
tcp: fix tcp_inet6_sk() for 32bit kernels
It turns out that struct ipv6_pinfo is not located as we think.
inet6_sk_generic() and tcp_inet6_sk() disagree on 32bit kernels by 4-bytes,
because struct tcp_sock has 8-bytes alignment,
but ipv6_pinfo size is not a multiple of 8.
sizeof(struct ipv6_pinfo): 116 (not padded to 8)
I actually first coded tcp_inet6_sk() as this patch does, but thought
that "container_of(tcp_sk(sk), struct tcp6_sock, tcp)" was cleaner.
As Julian told me : Nobody should use tcp6_sock.inet6
directly, it should be accessed via tcp_inet6_sk() or inet6_sk().
This happened when we added the first u64 field in struct tcp_sock.
Dust Li [Mon, 1 Apr 2019 08:04:53 +0000 (16:04 +0800)]
tcp: fix a potential NULL pointer dereference in tcp_sk_exit
When tcp_sk_init() failed in inet_ctl_sock_create(),
'net->ipv4.tcp_congestion_control' will be left
uninitialized, but tcp_sk_exit() hasn't check for
that.
This patch add checking on 'net->ipv4.tcp_congestion_control'
in tcp_sk_exit() to prevent NULL-ptr dereference.
Fixes: 6670e1524477 ("tcp: Namespace-ify sysctl_tcp_default_congestion_control") Signed-off-by: Dust Li <[email protected]> Signed-off-by: Eric Dumazet <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Merge branch 'work.aio' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull aio race fixes and cleanups from Al Viro.
The aio code had more issues with error handling and races with the aio
completing at just the right (wrong) time along with freeing the file
descriptor when another thread closes the file.
Just a couple of these commits are the actual fixes: the others are
cleanups to either make the fixes simpler, or to make the code legible
and understandable enough that we hope there's no more fundamental races
hiding.
* 'work.aio' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
aio: move sanity checks and request allocation to io_submit_one()
deal with get_reqs_available() in aio_get_req() itself
aio: move dropping ->ki_eventfd into iocb_destroy()
make aio_read()/aio_write() return int
Fix aio_poll() races
aio: store event at final iocb_put()
aio: keep io_event in aio_kiocb
aio: fold lookup_kiocb() into its sole caller
pin iocb through aio.
projects / linux.git / log