[PATCH v3 0/6] flow: Introduce flow_epoll_set() to centralize epoll operations

[Laurent Vivier <lvivier@redhat.com>]

Currently, each protocol handler (TCP, TCP splice, ICMP/ping, UDP) has its
own code to add or modify file descriptors in epoll.  This leads to
duplicated boilerplate across icmp.c, tcp.c, tcp_splice.c, and udp_flow.c,
with each setting up epoll_ref unions and calling epoll_ctl() with
flow-type-specific details.

This series introduces flow_epoll_set() in flow.c to handle epoll
operations for all flow types in a unified way.  The function derives
the epoll type from the flow type via a mapping array, providing a
simpler interface where the caller only needs to pass the flow, command,
events, file descriptor, and side index.

The API is:

    int flow_epoll_set(const struct flow_common *f, int command,
                       uint32_t events, int fd, unsigned int sidei);

This centralized approach will be essential for queue pair migration in
the upcoming multithreading work, where flows need to be moved between
different epoll instances owned by different threads.

Preparatory patches:
- Patch 1 removes dead timer update code in tcp_epoll_ctl()
- Patch 2 refactors TCP timer creation to use epoll_add() helper
- Patch 3 removes unneeded epoll_ref indirection in udp_flow
- Patch 4 refactors udp_flow_sock() to assign the socket internally
- Patch 5 refactors tcp_splice_conn_epoll_events() to per-side computation

Core patch:
- Patch 6 introduces flow_epoll_set() used by all protocol handlers

Changes in v3:
- Added patch 2: "tcp: cleanup timer creation" refactoring tcp_timer_ctl()
  to use epoll_add() helper with improved error handling
- Removed epoll type parameter from flow_epoll_set(): the type is now
  derived from the flow type via a new flow_epoll[] mapping array
- Simplified function signature to flow_epoll_set(f, command, events, fd, sidei)
- TCP timer no longer uses flow_epoll_set(), uses epoll_add() directly
- Added Reviewed-by from David Gibson to patches 1 and 3
- Series now 6 patches (was 5)

Changes in v2:
- Added patch 1: Remove dead timer update in tcp_epoll_ctl() since flow
  table compaction was eliminated
- Added patch 2: Remove unnecessary epoll_ref indirection in udp_flow.c
- Dropped v1 patch 6 (compute events inside flow_epoll_set())
- Dropped v1 patch 7 (retrieve fd from flow structure)
- Added epoll_ctl() command parameter (EPOLL_CTL_ADD/EPOLL_CTL_MOD)
- Callers must set epollfd via flow_epollid_set() before calling
  flow_epoll_set()
- Removed FLOW_EPOLL_TIMER_ADD/FLOW_EPOLL_TIMER_MOD constants
- Reduced series from 7 to 5 patches

Laurent Vivier (6):
  tcp: remove timer update in tcp_epoll_ctl()
  tcp: cleanup timer creation
  udp_flow: remove unneeded epoll_ref indirection
  udp_flow: Assign socket to flow inside udp_flow_sock()
  tcp_splice: Refactor tcp_splice_conn_epoll_events() to per-side
    computation
  flow: Introduce flow_epoll_set() to centralize epoll operations

 flow.c       |  37 +++++++++++++++++
 flow.h       |   2 +
 icmp.c       |  10 ++---
 tcp.c        |  88 ++++++++++++++++++++--------------------
 tcp_splice.c | 111 +++++++++++++++++++++++----------------------------
 udp_flow.c   |  18 +++------
 6 files changed, 140 insertions(+), 126 deletions(-)

-- 
2.52.0