> From: Kamalakshitha Aligeri [mailto:kamalakshitha.alig...@arm.com] > Sent: Friday, 24 February 2023 19.11 > > From: = Morten Brørup <m...@smartsharesystems.com>
This should be: From: Morten Brørup <m...@smartsharesystems.com> It could be fixed while merging. This is the only complaint in patchwork. > > Zero-copy access to mempool caches is beneficial for PMD performance, > and > must be provided by the mempool library to fix [Bug 1052] without a > performance regression. > > [Bug 1052]: https://bugs.dpdk.org/show_bug.cgi?id=1052 > > Bugzilla ID: 1052 > > Signed-off-by: Morten Brørup <m...@smartsharesystems.com> > Signed-off-by: Kamalakshitha Aligeri <kamalakshitha.alig...@arm.com> Can we get some reviews/acks on this please? I would like to see mempool zero-copy go into DPDK before 23.11. Please also note that the warnings and errors in patchwork regarding patch 2/2 are bogus or unrelated. > --- > v10: > * Added mempool test cases with zero-copy API's For the parts not provided by myself, i.e. the test cases: Acked-by: Morten Brørup <m...@smartsharesystems.com> [...] > diff --git a/lib/mempool/version.map b/lib/mempool/version.map > index dff2d1cb55..06cb83ad9d 100644 > --- a/lib/mempool/version.map > +++ b/lib/mempool/version.map > @@ -49,6 +49,15 @@ EXPERIMENTAL { > __rte_mempool_trace_get_contig_blocks; > __rte_mempool_trace_default_cache; > __rte_mempool_trace_cache_flush; > + __rte_mempool_trace_ops_populate; > + __rte_mempool_trace_ops_alloc; > + __rte_mempool_trace_ops_free; > + __rte_mempool_trace_set_ops_byname; > + > + # added in 23.03 Time is passing, so now this should be updated to 23.07 It could be fixed while merging. > + __rte_mempool_trace_cache_zc_put_bulk; > + __rte_mempool_trace_cache_zc_put_rewind; > + __rte_mempool_trace_cache_zc_get_bulk; > }; > > INTERNAL {
