1) cast to "int" is unnecessary:
u8 will be promoted to int before decrementing,
small positive numbers fit into "int", so their values won't be changed
during promotion.
Once everything is int including loop counters, signedness doesn't
matter: 32-bit operations will stay 32-bit operations.
But! Someone tried to make this loop smart by making everything of
the same type apparently in an attempt to optimise it.
Do the optimization, just differently.
Do the cast where it matters. :^)
2) frag size is unsigned entity and sum of fragments sizes is also
unsigned.
Make everything unsigned, leave no MOVSX instruction behind.
add/remove: 0/0 grow/shrink: 0/3 up/down: 0/-4 (-4)
function old new delta
skb_cow_data 835 834 -1
ip_do_fragment 2549 2548 -1
ip6_fragment 3130 3128 -2
Total: Before=154865032, After=154865028, chg -0.00%
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Length of a netlink attribute may be u16 but lengths of basic attributes
are much smaller, so small we can save 16 bytes of .rodata and pocket
change inside .text.
16-bit is worse on x86-64 than 8-bit because of operand size override prefix.
add/remove: 0/0 grow/shrink: 0/3 up/down: 0/-19 (-19)
function old new delta
validate_nla 418 417 -1
nla_policy_len 66 64 -2
nla_attr_minlen 32 16 -16
Total: Before=154865051, After=154865032, chg -0.00%
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
size_t is way too much for an integer not exceeding 64.
Space savings: 10 bytes!
add/remove: 0/0 grow/shrink: 0/3 up/down: 0/-10 (-10)
function old new delta
napi_consume_skb 165 163 -2
__kfree_skb_flush 56 53 -3
__kfree_skb_defer 97 92 -5
Total: Before=154865639, After=154865629, chg -0.00%
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
- 6 patches adding functionality to detect a WiFi interface under
other virtual interfaces, like VLANs. They introduce a cache for
the detected the WiFi configuration to avoid RTNL locking in
critical sections. Patches have been prepared by Marek Lindner
and Sven Eckelmann
- Enable automatic module loading for genl requests, by Sven Eckelmann
- Fix a potential race condition on interface removal. This is not
happening very often in practice, but requires bigger changes to fix,
so we are sending this to net-next. By Linus Luessing
-----BEGIN PGP SIGNATURE-----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=Pdlv
-----END PGP SIGNATURE-----
Merge tag 'batadv-next-for-davem-20161119' of git://git.open-mesh.org/linux-merge
Simon Wunderlich says:
====================
This feature patchset includes the following changes:
- 6 patches adding functionality to detect a WiFi interface under
other virtual interfaces, like VLANs. They introduce a cache for
the detected the WiFi configuration to avoid RTNL locking in
critical sections. Patches have been prepared by Marek Lindner
and Sven Eckelmann
- Enable automatic module loading for genl requests, by Sven Eckelmann
- Fix a potential race condition on interface removal. This is not
happening very often in practice, but requires bigger changes to fix,
so we are sending this to net-next. By Linus Luessing
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove static function __packet_rcv_vnet(), which only called
virtio_net_hdr_from_skb() and BUG()ged out if an error code was
returned. Instead, call virtio_net_hdr_from_skb() from the former
call sites of __packet_rcv_vnet() and actually use the error handling
code that is already there.
Signed-off-by: Jarno Rajahalme <jarno@ovn.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use the common virtio_net_hdr_to_skb() instead of open coding it.
Other call sites were changed by commit fd2a0437dc, but this one was
missed, maybe because it is split in two parts of the source code.
Interim comparisons of 'vnet_hdr->gso_type' still work as both the
vnet_hdr and skb notion of gso_type is zero when there is no gso.
Fixes: fd2a0437dc ("virtio_net: introduce virtio_net_hdr_{from,to}_skb")
Signed-off-by: Jarno Rajahalme <jarno@ovn.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
virtio_net_hdr_from_skb() clears the memory for the header, so there
is no point for the callers to do the same.
Signed-off-by: Jarno Rajahalme <jarno@ovn.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
No point storing the return value of virtio_net_hdr_to_skb() or
virtio_net_hdr_from_skb() to a variable when the value is used only
once as a boolean in an immediately following if statement.
Signed-off-by: Jarno Rajahalme <jarno@ovn.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Allocate resources dynamically for Upper layer driver's (ULD) like
cxgbit, iw_cxgb4, cxgb4i and chcr. The resources allocated include Tx
queues which are allocated when ULD register with cxgb4 driver and freed
while un-registering. The Tx queues which are shared by ULD shall be
allocated by first registering driver and un-allocated by last
unregistering driver.
Signed-off-by: Atul Gupta <atul.gupta@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We obviously intended a bitwise AND here, not a logical one.
Fixes: 8c978d0592 ("liquidio CN23XX: Mailbox support")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Relocate mdix code to phy driver to be called at config_init().
Signed-off-by: Woojung Huh <woojung.huh@microchip.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Fainelli says:
====================
net: Enable COMPILE_TEST for Marvell & Freescale drivers
This patch series allows building the Freescale and Marvell Ethernet network
drivers with COMPILE_TEST.
Changes in v4:
- add proper HAS_DMA to fix build errors on m32r
- provide an inline stub for mvebu_mbus_get_dram_win_info
- added an additional patch to fix build errors with mv88e6xxx on m32r
Changes in v3:
- reorder patches to avoid introducing a build warning between commits
Changes in v2:
- rename register define clash when building for i386 (spotted by LKP)
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Some architectures may not define IRQ_DOMAIN (like m32r), fixes
undefined references to IRQ_DOMAIN functions.
Fixes: dc30c35be7 ("net: dsa: mv88e6xxx: Implement interrupt support.")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
All Marvell Ethernet drivers actually build fine with COMPILE_TEST with
a few warnings. We need to add a few HAS_DMA dependencies to fix linking
failures on problematic architectures like m32r.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In preparation for allowing CONFIG_MVNETA_BM to build with COMPILE_TEST,
provide an inline stub for mvebu_mbus_get_dram_win_info().
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There are only a handful of Freescale Ethernet drivers that don't
actually build with COMPILE_TEST:
* FEC, for which we would need to define a default register layout if no
supported architecture is defined
* UCC_GETH which depends on PowerPC cpm.h header (which could be moved
to a generic location)
* GIANFAR needs to depend on HAS_DMA to fix linking failures on some
architectures (like m32r)
We need to fix an unmet dependency to get there though:
warning: (FSL_XGMAC_MDIO) selects OF_MDIO which has unmet direct
dependencies (OF && PHYLIB)
which would result in CONFIG_OF_MDIO=[ym] without CONFIG_OF to be set.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
FS is a global symbol used by the x86 32-bit architecture, fixes builds
re-definitions:
>> drivers/net/ethernet/freescale/gianfar_ptp.c:75:0: warning: "FS"
>> redefined
#define FS (1<<28) /* FIPER start indication */
In file included from arch/x86/include/uapi/asm/ptrace.h:5:0,
from arch/x86/include/asm/ptrace.h:6,
from arch/x86/include/asm/math_emu.h:4,
from arch/x86/include/asm/processor.h:11,
from include/linux/mutex.h:19,
from include/linux/kernfs.h:13,
from include/linux/sysfs.h:15,
from include/linux/kobject.h:21,
from include/linux/device.h:17,
from
drivers/net/ethernet/freescale/gianfar_ptp.c:23:
arch/x86/include/uapi/asm/ptrace-abi.h:15:0: note: this is the
location of the previous definition
#define FS 9
Reported-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Update the connection type enumeration for backplane mode and return
an error when there is a mismatch between the mode and the connection
type.
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Allan W. Nielsen says:
====================
Adding PHY-Tunables and downshift support
(This is a re-post of the v3 patch set with a new cover letter - I was not
aware that the cover letters was used a commit comments in merge commits).
This series add support for PHY tunables, and uses this facility to
configure downshifting. The downshifting mechanism is implemented for MSCC
phys.
This series tries to address the comments provided back in mid October when
this feature was posted along with fast-link-failure. Fast-link-failure has
been separated out, but we would like to pick continue on that if/when we
agree on how the phy-tunables and downshifting should be done.
The proposed generic interface is similar to
ETHTOOL_GTUNABLE/ETHTOOL_STUNABLE, it uses the same type
(ethtool_tunable/tunable_type_id) but a new enum (phy_tunable_id) is added
to reflect the PHY tunable.
The implementation just call the newly added function pointers in
get_tunable/set_tunable phy_device structure.
To configure downshifting, the ethtool_tunable structure is used. 'id' must
be set to 'ETHTOOL_PHY_DOWNSHIFT', 'type_id' must be set to
'ETHTOOL_TUNABLE_U8' and 'data' value configure the amount of downshift
re-tries.
If configured to DOWNSHIFT_DEV_DISABLE, then downshift is disabled If
configured to DOWNSHIFT_DEV_DEFAULT_COUNT, then it is up to the device to
choose a device-specific re-try count.
Tested on Beaglebone Black with VSC 8531 PHY.
Change set:
v0:
- Link Speed downshift and Fast Link failure-2 features coded by using
Device tree.
v1:
- Split the Downshift and FLF2 features in different set of patches.
- Removed DT access and implemented IOCTL access suggested by Andrew.
- Added function pointers in get_tunable/set_tunable phy_device structure
v2:
- Added trace message with a hist is printed when downshifting clould not
be eanbled with the requested count
- (ethtool) Syntax is changed from "--set-phy-tunable downshift on|off|%d"
to "--set-phy-tunable [downshift on|off [count N]]" - as requested by
Andrew.
v3:
- Fixed Spelling in "net: phy: Add downshift get/set support in Microsemi
PHYs driver"
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Implements the phy tunable function pointers and implement downshift
functionality for MSCC PHYs.
Signed-off-by: Raju Lakkaraju <Raju.Lakkaraju@microsemi.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Allan W. Nielsen <allan.nielsen@microsemi.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Adding validation support for the ETHTOOL_PHY_DOWNSHIFT. Functional
implementation needs to be done in the individual PHY drivers.
Signed-off-by: Raju Lakkaraju <Raju.Lakkaraju@microsemi.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Allan W. Nielsen <allan.nielsen@microsemi.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
For operation in cabling environments that are incompatible with
1000BASE-T, PHY device may provide an automatic link speed downshift
operation. When enabled, the device automatically changes its 1000BASE-T
auto-negotiation to the next slower speed after a configured number of
failed attempts at 1000BASE-T. This feature is useful in setting up in
networks using older cable installations that include only pairs A and B,
and not pairs C and D.
Signed-off-by: Raju Lakkaraju <Raju.Lakkaraju@microsemi.com>
Signed-off-by: Allan W. Nielsen <allan.nielsen@microsemi.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Adding get_tunable/set_tunable function pointer to the phy_driver
structure, and uses these function pointers to implement the
ETHTOOL_PHY_GTUNABLE/ETHTOOL_PHY_STUNABLE ioctls.
Signed-off-by: Raju Lakkaraju <Raju.Lakkaraju@microsemi.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Allan W. Nielsen <allan.nielsen@microsemi.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Defines a generic API to get/set phy tunables. The API is using the
existing ethtool_tunable/tunable_type_id types which is already being used
for mac level tunables.
Signed-off-by: Raju Lakkaraju <Raju.Lakkaraju@microsemi.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Allan W. Nielsen <allan.nielsen@microsemi.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Saeed Mahameed says:
====================
Mellanox 100G mlx5 update 2016-11-15
This series contains four humble mlx5 features.
From Gal,
- Add the support for PCIe statistics and expose them in ethtool
From Huy,
- Add the support for port module events reporting and statistics
- Add the support for driver version setting into FW (for display purposes only)
From Mohamad,
- Extended the command interface cache flexibility
This series was generated against commit
6a02f5eb6a ("Merge branch 'mlxsw-i2c")
V2:
- Changed plain "unsigned" to "unsigned int"
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch exposes two groups of PCIe counters:
- Performance counters.
- Timers and states counters.
Queried with ethtool -S <devname>.
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add the needed infrastructure for future use of MPCNT register.
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If driver_version capability bit is enabled, set driver version
to firmware after the init HCA command, for display purposes.
Example of driver version: "Linux,mlx5_core,3.0-1"
Signed-off-by: Huy Nguyen <huyn@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add driver_version capability bit is enabled, and set driver
version command in mlx5_ifc firmware header. The only purpose
of this command is to store a driver version/OS string in FW
to be reported and displayed in various management systems,
such as IPMI/BMC.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Huy Nguyen <huyn@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add port module event counters to ethtool -S command
Signed-off-by: Huy Nguyen <huyn@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
For each asynchronous port module event:
1. print with ratelimit to the dmesg log
2. increment the corresponding event counter
Signed-off-by: Huy Nguyen <huyn@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add hardware structures and constants definitions needed for module
events support.
Signed-off-by: Huy Nguyen <huyn@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add more cache command size sets and more entries for each set based on
the current commands set different sizes and commands frequency.
Fixes: e126ba97db ('mlx5: Add driver for Mellanox Connect-IB adapters')
Signed-off-by: Mohamad Haj Yahia <mohamad@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Edward Cree says:
====================
sfc: Firmware-Assisted TSO version 2
The firmware on 8000 series SFC NICs supports a new TSO API ("FATSOv2"), and
7000 series NICs will also support this in an imminent release. This series
adds driver support for this TSO implementation.
The series also removes SWTSO, as it's now equivalent to GSO. This does not
actually remove very much code, because SWTSO was grotesquely intertwingled
with FATSOv1, which will also be removed once 7000 series supports FATSOv2.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
It gives no advantage over GSO now that xmit_more exists. If we find
ourselves unable to handle a TSO skb (because our TXQ doesn't have a
TSOv2 context and the NIC doesn't support TSOv1), hand it back to GSO.
Also do that if the TSO handler fails with EINVAL for any other reason.
As Falcon-architecture NICs don't support any firmware-assisted TSO,
they no longer advertise TSO feature flags at all.
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If we fail to init the TXQ because of insufficient TSOv2 contexts,
try again with TSOv2 disabled.
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add support for FATSOv2 to the driver. FATSOv2 offloads far more of the task
of TCP segmentation to the firmware, such that we now just pass a single
super-packet to the NIC. This means TSO has a great deal in common with a
normal DMA transmit, apart from adding a couple of option descriptors.
NIC-specific checks have been moved off the fast path and in to
initialisation where possible.
This also moves FATSOv1/SWTSO to a new file (tx_tso.c). The end of transmit
and some error handling is now outside TSO, since it is common with other
code.
Signed-off-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Make struct pernet_operations::id unsigned.
There are 2 reasons to do so:
1)
This field is really an index into an zero based array and
thus is unsigned entity. Using negative value is out-of-bound
access by definition.
2)
On x86_64 unsigned 32-bit data which are mixed with pointers
via array indexing or offsets added or subtracted to pointers
are preffered to signed 32-bit data.
"int" being used as an array index needs to be sign-extended
to 64-bit before being used.
void f(long *p, int i)
{
g(p[i]);
}
roughly translates to
movsx rsi, esi
mov rdi, [rsi+...]
call g
MOVSX is 3 byte instruction which isn't necessary if the variable is
unsigned because x86_64 is zero extending by default.
Now, there is net_generic() function which, you guessed it right, uses
"int" as an array index:
static inline void *net_generic(const struct net *net, int id)
{
...
ptr = ng->ptr[id - 1];
...
}
And this function is used a lot, so those sign extensions add up.
Patch snipes ~1730 bytes on allyesconfig kernel (without all junk
messing with code generation):
add/remove: 0/0 grow/shrink: 70/598 up/down: 396/-2126 (-1730)
Unfortunately some functions actually grow bigger.
This is a semmingly random artefact of code generation with register
allocator being used differently. gcc decides that some variable
needs to live in new r8+ registers and every access now requires REX
prefix. Or it is shifted into r12, so [r12+0] addressing mode has to be
used which is longer than [r8]
However, overall balance is in negative direction:
add/remove: 0/0 grow/shrink: 70/598 up/down: 396/-2126 (-1730)
function old new delta
nfsd4_lock 3886 3959 +73
tipc_link_build_proto_msg 1096 1140 +44
mac80211_hwsim_new_radio 2776 2808 +32
tipc_mon_rcv 1032 1058 +26
svcauth_gss_legacy_init 1413 1429 +16
tipc_bcbase_select_primary 379 392 +13
nfsd4_exchange_id 1247 1260 +13
nfsd4_setclientid_confirm 782 793 +11
...
put_client_renew_locked 494 480 -14
ip_set_sockfn_get 730 716 -14
geneve_sock_add 829 813 -16
nfsd4_sequence_done 721 703 -18
nlmclnt_lookup_host 708 686 -22
nfsd4_lockt 1085 1063 -22
nfs_get_client 1077 1050 -27
tcf_bpf_init 1106 1076 -30
nfsd4_encode_fattr 5997 5930 -67
Total: Before=154856051, After=154854321, chg -0.00%
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
UDP busy polling is restricted to connected UDP sockets.
This is because sk_busy_loop() only takes care of one NAPI context.
There are cases where it could be extended.
1) Some hosts receive traffic on a single NIC, with one RX queue.
2) Some applications use SO_REUSEPORT and associated BPF filter
to split the incoming traffic on one UDP socket per RX
queue/thread/cpu
3) Some UDP sockets are used to send/receive traffic for one flow, but
they do not bother with connect()
This patch records the napi_id of first received skb, giving more
reach to busy polling.
Tested:
lpaa23:~# echo 70 >/proc/sys/net/core/busy_read
lpaa24:~# echo 70 >/proc/sys/net/core/busy_read
lpaa23:~# for f in `seq 1 10`; do ./super_netperf 1 -H lpaa24 -t UDP_RR -l 5; done
Before patch :
27867 28870 37324 41060 41215
36764 36838 44455 41282 43843
After patch :
73920 73213 70147 74845 71697
68315 68028 75219 70082 73707
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Sowmini Varadhan says:
====================
RDS: TCP: HA/Failover fixes
This series contains a set of fixes for bugs exposed when
we ran the following in a loop between a test machine pair:
while (1); do
# modprobe rds-tcp on test nodes
# run rds-stress in bi-dir mode between test machine pair
# modprobe -r rds-tcp on test nodes
done
rds-stress in bi-dir mode will cause both nodes to initiate
RDS-TCP connections at almost the same instant, exposing the
bugs fixed in this series.
Without the fixes, rds-stress reports sporadic packet drops,
and packets arriving out of sequence. After the fixes,we have
been able to run the test overnight, without any issues.
Each patch has a detailed description of the root-cause fixed
by the patch.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
When 2 RDS peers initiate an RDS-TCP connection simultaneously,
there is a potential for "duelling syns" on either/both sides.
See commit 241b271952 ("RDS-TCP: Reset tcp callbacks if re-using an
outgoing socket in rds_tcp_accept_one()") for a description of this
condition, and the arbitration logic which ensures that the
numerically large IP address in the TCP connection is bound to the
RDS_TCP_PORT ("canonical ordering").
The rds_connection should not be marked as RDS_CONN_UP until the
arbitration logic has converged for the following reason. The sender
may start transmitting RDS datagrams as soon as RDS_CONN_UP is set,
and since the sender removes all datagrams from the rds_connection's
cp_retrans queue based on TCP acks. If the TCP ack was sent from
a tcp socket that got reset as part of duel aribitration (but
before data was delivered to the receivers RDS socket layer),
the sender may end up prematurely freeing the datagram, and
the datagram is no longer reliably deliverable.
This patch remedies that condition by making sure that, upon
receipt of 3WH completion state change notification of TCP_ESTABLISHED
in rds_tcp_state_change, we mark the rds_connection as RDS_CONN_UP
if, and only if, the IP addresses and ports for the connection are
canonically ordered. In all other cases, rds_tcp_state_change will
force an rds_conn_path_drop(), and rds_queue_reconnect() on
both peers will restart the connection to ensure canonical ordering.
A side-effect of enforcing this condition in rds_tcp_state_change()
is that rds_tcp_accept_one_path() can now be refactored for simplicity.
It is also no longer possible to encounter an RDS_CONN_UP connection in
the arbitration logic in rds_tcp_accept_one().
Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The RDS transport has to be able to distinguish between
two types of failure events:
(a) when the transport fails (e.g., TCP connection reset)
but the RDS socket/connection layer on both sides stays
the same
(b) when the peer's RDS layer itself resets (e.g., due to module
reload or machine reboot at the peer)
In case (a) both sides must reconnect and continue the RDS messaging
without any message loss or disruption to the message sequence numbers,
and this is achieved by rds_send_path_reset().
In case (b) we should reset all rds_connection state to the
new incarnation of the peer. Examples of state that needs to
be reset are next expected rx sequence number from, or messages to be
retransmitted to, the new incarnation of the peer.
To achieve this, the RDS handshake probe added as part of
commit 5916e2c155 ("RDS: TCP: Enable multipath RDS for TCP")
is enhanced so that sender and receiver of the RDS ping-probe
will add a generation number as part of the RDS_EXTHDR_GEN_NUM
extension header. Each peer stores local and remote generation
numbers as part of each rds_connection. Changes in generation
number will be detected via incoming handshake probe ping
request or response and will allow the receiver to reset rds_connection
state.
Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As noted in rds_recv_incoming() sequence numbers on data packets
can decreas for the failover case, and the Rx path is equipped
to recover from this, if the RDS_FLAG_RETRANSMITTED is set
on the rds header of an incoming message with a suspect sequence
number.
The RDS_FLAG_RETRANSMITTED is predicated on the RDS_FLAG_RETRANSMITTED
flag in the rds_message, so make sure the flag is set on messages
queued for retransmission.
Signed-off-by: Sowmini Varadhan <sowmini.varadhan@oracle.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As sugested by Joe Perches, we could replace all
if (netif_msg_type(priv)) dev_xxx(priv->devices, ...)
by the simpler macro netif_xxx(priv, hw, priv->dev, ...)
Signed-off-by: Corentin Labbe <clabbe.montjoie@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some printing have the function name hardcoded.
It is better to use __func__ instead.
Signed-off-by: Corentin Labbe <clabbe.montjoie@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The stmmac driver use lots of pr_xxx functions to print information.
This is bad since we cannot know which device logs the information.
(moreover if two stmmac device are present)
Furthermore, it seems that it assumes wrongly that all logs will always
be subsequent by using a dev_xxx then some indented pr_xxx like this:
kernel: sun7i-dwmac 1c50000.ethernet: no reset control found
kernel: Ring mode enabled
kernel: No HW DMA feature register supported
kernel: Normal descriptors
kernel: TX Checksum insertion supported
So this patch replace all pr_xxx by their netdev_xxx counterpart.
Excepts for some printing where netdev "cause" unpretty output like:
sun7i-dwmac 1c50000.ethernet (unnamed net_device) (uninitialized): no reset control found
In those case, I keep dev_xxx.
In the same time I remove some "stmmac:" print since
this will be a duplicate with that dev_xxx displays.
Signed-off-by: Corentin Labbe <clabbe.montjoie@gmail.com>
Acked-by: Giuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
