The code attempts to use scattered_message::append_static() to gain zero
copy, claiming that _output protects the data's lifetime, but that's a
false claim - it will remain in the transmit queue waiting for an ACK
while we process (and respond to) the next request.
Fix by just writing the output naturally. We may zero-copy support later,
though it is dubious it will help thrift, as the code is full of copies
and atomics.
sstring's std::string conversion uses c_str() to construct the value,
but the conversion is broken if the value contains NUL - both sstring and
std::string can contain NULs, but C strings use them as a terminator.
Fix by using the size+length std::string constructor.
size of the sstring _ascii_prefix should also be added when computing
item footprint. Without this change, reclaimer would end up evicting
more than needed.
Signed-off-by: Raphael S. Carvalho <raphaelsc@cloudius-systems.com>
This patchset extends the database and thrift layers sufficiently to
pass the cassandra-stress insert test in thrift mode. I see 70krps
on my desktop (posix stack), likely limited by the client.
Using sstring can lead to confusion with UTF8 strings.
The Java byte type is signed, so make bytes' internal type be signed as
well (even though Cassandra tries to treat it as unsigned).
While we should use int8_t, sstring is not perfectly compatible with this
yet, so add a FIXME and use char instead.
This patchset adds a simple metadata and data store, and wires the
system_add_keyspace RPC to set it up.
With this, cassandra-stress is able to initialize the database (but not
store anything in it).
Simplistic database using std::map<> to hold rows, and boost::any to
hold values.
Supports:
- multiple key spaces
- multiple column families
- a few data types
Does not support:
- container data types
- secondary indexes
- composites
- validators
The byte-order functions were changed not to do in-place conversions,
but they still accept non-const inputs, although they do not modify them.
This can make them harder to use in some cases.
Fix by marking the inputs const.
Provide a function that maps packet's rss hash to a cpu that should handle
it. This function is needed to find appropriate src port for outgoing
tcp/udp connection. Use this function to forward de-fragmented ip packet
to avoid one extra hop too.
Add a space after the "Checking link status" to prevent it from
merging with "done" if the link is up immediatelly.
For instance this is going to be the case for a VF
of a PF with already established link (e.g. on AWS).
Signed-off-by: Vlad Zolotarov <vladz@cloudius-systems.com>
We assume that if Rx IPv4, TCP and UDP checksum offload features are suported then
they are supported or not supported all together. The same is about the Tx UDP and TCP
checksum offload.
Add the assert that check this assumption.
Signed-off-by: Vlad Zolotarov <vladz@cloudius-systems.com>
Even if port has a single queue we still want the RSS feature to be
available in order to make HW calculate RSS hash for us.
Signed-off-by: Vlad Zolotarov <vladz@cloudius-systems.com>
DPDK 1.8 provides per-device default Tx and Rx queues configurations in the output
of rte_eth_dev_info_get(). Use them instead of ixgbe tuned hardcoded values.
Signed-off-by: Vlad Zolotarov <vladz@cloudius-systems.com>
- Rename: init_port() -> init_port_start().
- Added a function init_port_fini() that has a code originally found flat in
init_local_queue().
- Moved the link state check to init_port_fini() since the link state should
be checked after the port has been started.
Signed-off-by: Vlad Zolotarov <vladz@cloudius-systems.com>
