Hi,
I think we need to suppress the following checkpatch CHECK level warning. When
an API, e.g. odp_shm_addr() defines NULL as a return value, an application
should be able to do ...
addr = odp_shm_addr(shm);
if (addr == NULL)
.. instead of ...
if (!addr)
... which seems to be the new favorite style of checkpatch. In theory, NULL can
be something else than 0 (in practice it's nearly always 0), but it's defined
exactly for this use case and we should be able use it (in API and apps).
-Petri
v2-0002-api-thread-added-thread-count-max.patch has no obvious style problems
and is ready for submission.
CHECK: Comparison to NULL could be written "!gbl_args->rx_stats"
#259: FILE: test/performance/odp_pktio_perf.c:1020:
+ if (gbl_args->rx_stats == NULL)
CHECK: Comparison to NULL could be written "!gbl_args->tx_stats"
#270: FILE: test/performance/odp_pktio_perf.c:1031:
+ if (gbl_args->tx_stats == NULL)
+ shm = odp_shm_reserve("test_globals.rx_stats",
+ gbl_args->rx_stats_size,
+ ODP_CACHE_LINE_SIZE, 0);
+
+ gbl_args->rx_stats = odp_shm_addr(shm);
+
+ if (gbl_args->rx_stats == NULL)
+ LOG_ABORT("Shared memory reserve failed.\n");
_______________________________________________
lng-odp mailing list
[email protected]
https://lists.linaro.org/mailman/listinfo/lng-odp
