Summary: 246 instances, 218 unique

Text	Count
// TODO - verify this	3
// TODO: add support in fake	1
// TODO figure out why the following check fails	1
// TODO: deprecate this	1
// TODO: Look into adding context on which client is connected	1
// TODO: This routine goes away once we use common ACL infra to create	1
// TODO: Support programming dynamic tx_settings	1
// TODO: We may want to trap NDP on a per-port or per-VLAN basis.	1
// TODO: add qualifier in AclTableQualifier enum	1
// TODO (skhare) Simplify/fold into BcmRxPacket	1
// TODO: add more mappings + better error message	1
// TODO: rohitpuri	1
// TODO: fill out extra fields as needed	1
* TODO (allwync): get rid of getRouteTable after agent code with thrift	1
# TODO use rpmdev-newspec $FBOSS_BINS to create spec and edit it.	1
// TODO: We need a more robust mechanism for setting up the ethernet	1
// TODO: in future, FieldsT and ThrifT should be one type	1
# TODO Support SAI_PHY_IMPL(T86789543)	1
// XXX TODO: side-effect mode does NOT work with optionals	2
// TODO: adding retries adds tolerance in case the i2c bus is	1
// TODO infer the flow records size instead and remove the input param.	1
// TODO We probably need a timer to ensure we	1
// TODO: add else logic for other platforms	1
// TODO We currently do not need more than 32bit values as per	1
// TODO (ccpowers): remove this as soon as we've stopped using the accton xphy	1
// TODO: move this to a utility since it	1
// TODO: assert on All<IsSaiAttribute>	1
// TODO: support configuring LAG egress for mirror	1
* TODO: could change api to take Port SwitchState objects down the	1
// TODO: add truncate check	2
// TODO: Log command output to Scuba	1
// TODO we should add router id information to the packet	1
// TODO - just use thrift enum names	1
// TODO: We don't actually verify the checksum right now.	1
// TODO: Investigate why port_bitmap contains	1
// TODO: It might be nice to support duplicate address detection, and track	1
/* TODO t7153326: Following code is added for backward compatibility	2
// TODO: add ods counter for such dropped mpls packet	1
// TODO: replace with switch_config.QueueScheduling?	1
* TODO (allwync): get rid of getIpRoute after agent code with thrift	1
# TODO: There may be some way to approximate an availability	1
// TODO: Upgrade the fan_cpld through I2C	1
// TODO: Since this is an extended experiment, we resue the aoi_override flag	1
// TODO: Current warmboot state doesn't have such field yet	1
// TODO remove LC once fruid.json is fixed on Galaxy Linecards	1
// TODO: add support for turning off these fields via a command-line flag	1
// TODO: add rest of fields as we convert them to thrifty	1
// TODO: For now we are allowing v4 LLs to be programmed because they are	1
// TODO: add support for addRepeatedValue() in FSDB client-side library	1
* TODO get rid of this flag once configs have been updated to	1
TODO: I'd like to either extend this or add a variant that	1
// TODO: Eventually the spanning tree state should be part of the Port	1
* TODO (skhare)	1
// FIXME We should really forward this along unchanged.	1
// TODO: Look into requesting results that take a long time	1
# TODO: Break up complex function	1
// TODO: Maybe this can be configured?	1
// TODO: Use number of queues by querying SAI_SWITCH_ATTR_NUMBER_OF_QUEUES	1
// TODO: Will fully migrate I2CBusApi into TransceiverPlatformApi. Then we will	3
// TODO explore moving this check to a compile time check	2
// TODO: We could add an intfChanged() method to re-use the existing	1
// TODO: Stop doing this so we can use ECN	1
# TODO: Compile takes too long. Commenting until addressing it	1
// TODO: remove this check after another push	1
// TODO: block until repl loop is completed	1
// TODO (skhare)	1
// TODO: actually implement this	1
# TODO: On Windows we require the Python libraries as well.	1
// TODO Lookup interface route and confirm that it points to CPU nhop	1
// TODO: check the checksum	1
// TODO: use switch_config.PortState?	1
// TODO (srikrishnagopu): Use the read-only attribute	1
# TODO: We aren't extremely strict about error checking for one-value	1
// TODO: will check platform information (maybe fboss whoami) to	1
// TODO:(shankaran) - Try to reconnect when there is a failure.	1
//   at the SwitchState level seamlessly. TODO: test what happens to	1
// TODO: Once BMC devices are provisioned with a non self-signed cert	1
// TODO: use VLAN_ETH_ZLEN when it is available in /linux/if_vlan.h	1
// TODO (skhare) Use ResourceGenerator to create this map, where the number	1
// TODO: Initialize the LEDs.  The LED handling code isn't open source yet,	1
/* TODO: Mimicking TH3 size here, figure out the defaults for TH4*/	2
// TODO: We should probably change this to store raw pointers.	1
* TODO - replace getNeighborEntryTable as getNeighborTable	1
// TODO: we discovered a resource leak on this call when it	1
5: i64 speedMbps; // TODO: i32 (someone is optimistic about port speeds)	1
# system packages are installed.  TODO: For users not defined in the	1
// TODO: clean up the EthHdr code and use EthHdr here	1
# TODO: Make this function less complex	1
// TODO - ideally we shouldn't care about whether	1
* TODO: Maintain a second index by name. Typically we have only a	2
// TODO verify that created state here is internally consistent t4155406	1
# TODO Try to call qsfp_service to handle xphy prbs logic first	3
# TODO: generate a main executable that will invoke Python3	1
// TODO:	1
// TODO: what to put in argc/argv?!?	1
// TODO throw error if shared bytes is set. We don't handle that in SAI	1
* TODO: We should switch from NodeMap to PrefixMap when the new PrefixMap	1
// TODO: Figure out the best way to handle errors here.	1
// TODO: Can we have reason code to a queue mapping that does not have	1
// TODO skhare	3
// TODO: only vrf 0 now	1
// TODO fill these in when we have debug counter support in SAI	1
// TODO interpret string source.deviceType to support ORv3 devices.	1
// TODO reconfigure the QCM parameters in hardware	1
int dataAddress; // TODO: rename to page	1
// TODO: fill in ClientInformation, or get rid of it in the API	1
// TODO - figure out why getRegexCounters fails for OSS	1
* TODO - make HwWriteBehavior granularity be per swtich	1
// TODO: Look into adding timeout for try_connect	1
* TODO: Consider breaking down update into add, del, syncClient	1
// TODO: should be case insensitive	1
// TODO: implement output	1
// TODO: explore if the parsing library provides a better way to implement	1
// TODO: Also check to see if this is the broadcast address for one of the	1
// TODO: To be implemented if the function is used by fboss	4
// TODO: In the future it would be nice to support allocating buf_ in a DMA	1
* TODO - Convert mask to mask len for efficient storage/lookup	1
// TODO: Also check to see if this is the broadcast address for one of the	1
// TODO: store speed as enum, not string?	1
// TODO support options	1
// TODO: other sanity checks (i.e. packet length, checksum...)	1
# TODO: process the subdirectories of data_dir_name	1
* TODO: Add support for Type 2 and Type 3 PDUs.	1
# TODO: Introduce some sort of defaults config that can select by	1
// TODO: use switch_config.PortSpeed?	1
// TODO: For other APIs, create new API wrappers and invoke	1
# TODO: Make this more accurate.	1
# TODO: Our caller still won't be able to distinguish between an empty	1
* TODO: Add support for IPv6 extension headers.	1
// TODO: replace with switch_config.MMUScalingFactor?	1
// TODO: make it possible to generate a PortStatus struct solely	1
// TODO assert for non empty response once this thrift API is implemented	1
// TODO: add support for options	1
// TODO (sgwang)	5
// TODO: do FDB entries associated with this port need to be removed	1
// TODO: path extensions here like regex support	1
// TODO: This should eventually be done as part of applyConfig()	1
// its 256 bytes wasted. but worth it. TODO use a	1
// TODO: sort here?	1
// TODO: outQueueLen_ only exports the average queue length over the last	1
// TODO: check the reason of punt, for now, assume it is for	1
// TODO - compute used resource stats from internal data structures and	1
// TODO: separate config and operational state	1
// TODO: handle labels punted to cpu	1
// TODO: add version for Coro once coro based thrift server is standardized	1
// TODO: migrate all cable types	1
#ifdef IS_OSS /* TODO: remove once OSS support added */	1
# TODO (rsunkad) re-enable this library for use with linking with libmain	1
// TODO: Do we need to bother handling this yet?	1
// TODO: This function contains high-level logic for how to apply the	1
* TODO (srikrishnagopu) : Remove after rolling out AFE TRIM (S249471)	1
// TODO (xiangzhu) Will further optimizing the performance, possibly	1
// TODO: ideally, we need to store this packet until the ARP is done and	1
// TODO: FSDB needs to support count() method on stats	1
// TODO: Test if this is true when rx pause is enabled	1
// TODO:	1
// FIXME: implement this	1
// TODO: read fpga version register and set up version_	1
// TODO: replace with switch_config.StreamType?	1
* TODO: see if we can stop doing this at SwSwitch layre	1
// TODO remove FAB once fruid.json is fixed on Galaxy fabric cards	1
// TODO: remove this in favor of state::NeighborState in switch_state.thrift	1
// TODO: store admin state as enum, not string?	1
// TODO: It would be nicer to use a monotonic clock, but unfortunately	1
// TODO: T70244223. Clarifying the right LED color for uplinks	1
// TODO: add a constructor that doesn't require string processing	1
# TODO: audit that both source paths have the same inode number	1
// TODO: does this really mean the cable is too long?	1
* TODO: bcm_rx_*	1
return false; // TODO: Return unknown instead	3
// TODO: explore using atomic exchanges on shared_ptr to get rid of lock.	1
// TODO: Customize	1
// TODO: remove this after we ensure exclusive access to cp2112 chip	2
// TODO: Add rate limiting so we don't generate too many requests for the	1
# TODO: In the future it would be nice to support pre-compiling the source	1
// TODO: this will deprecate port speed and port fec	1
// TODO MAYBE: parameterize (this? whole API?) by IP?	1
// TODO - change to AgentConfig once SwSwitch can pass us that	1
// TODO: support TCP segment	1
// TODO: use a static factory method when we add more versions	1
std::nullopt, // Part Number : TODO: Read and cache tcvrPartNumber	1
// TODO: implement this	2
// TODO:	1
// FIXME : Stoo skipping for Tajo once T79717530 resolved	1
// TODO (skhare) Find OpenNSA method for this	1
# TODO: we could just go ahead and build it here, but I	1
# TODO Future work planned	1
// TODO:(shankaran) - when healthcheck fails remove the sak from set.	1
// TODO: explore performance improvement if we make all this parallel.	1
// TODO (aeckert): it would be nicer if the BcmPortGroup wrote its	1
// TODO once hw shell access is supported	1
// TODO: it might be worth splitting up limits for ecmp/ucmp	1
// TODO: We may want to trap ARP on a per-port or per-VLAN basis.	1
// TODO: replace with pulling name from platform port	1
* TODO: Not much code here yet.	1
// TODO: there may be a better way of determining this rather than	1
# TODO: `mkscratch` doesn't currently know how best to place things on	1
// TODO: Add file lock to prevent multiple instace of fw-util from running	1
// TODO: if needed, we can adjust send buffer size, TUNSETSNDBUF	1
* TODO: The ingress and egress mirror is cached in port handle	1
// TODO: For other APIs, create new API wrappers and invoke wrappedApi()	1
// TODO - figure out a way to send full agent config	1
// TODO: prob need to handle ret code?	1
// TODO: Will deprecate the optional fields in PlatformConfig and start using	1
* TODO: Scheduling params require using portGport instead of queueGport due	2
// TODO: Add SaiSflowMirrorTraits here.	1
// TODO: figure out why we can't use bottom_border on the header instead	1
// TODO: use switch_config.PortPause?	1
// TODO: support Preemphasis once it is also used	1
// TODO: remove all migration along with old ser/des after next disruptive push	2
// TODO: Check platform specific termination criteria	1
// TODO: Look into how we can add timeout in future diffs	1
// TODO (ccpowers): remove this once all platforms have their own PortStats type	1
// TODO (skhare) refactor and replace with right OpenNSA API	1
# TODO: Ideally, this function would inspect the current cgroup for any	1
// TODO: add counter to keep track of spurious rx packet	2
// TODO: assume vrf 0 now	1
// TODO: we now generate one station entry per interface, even if all	1