[RFC net v2 0/3] bnxt_en: Make RING FREE more robust

Joe Damato posted 3 patches 1 day, 20 hours ago
There is a newer version of this series
drivers/net/ethernet/broadcom/bnxt/bnxt.c     | 69 +++++++++++------
.../net/ethernet/broadcom/bnxt/bnxt_hwrm.c    | 75 ++++++++++++++-----
2 files changed, 102 insertions(+), 42 deletions(-)
[RFC net v2 0/3] bnxt_en: Make RING FREE more robust
Posted by Joe Damato 1 day, 20 hours ago
Greetings:

This is a follow up to the previous RFC (linked below), which takes a slightly
different approach, is now targeted at net instead of net-next, and has a new
subject line :)

On two production systems, I saw the following dmesg pattern:

  NETDEV WATCHDOG: transmit queue 0 timed out 6073 ms
  Resp cmpl intr err msg: 0x51                  x20
  hwrm_ring_free type 1 failed                  x12
  hwrm_ring_free type 2 failed                  x8
  AMD-Vi: IO_PAGE_FAULT  x3

This suggests that, for some currently unknown reason, TX completions stall
and the netdev watchdog fires. The driver asks FW to free the rings, this
times out, but the driver ignores the possible failure and frees ring memory.
Since the FW didn't respond to the ring free command, it is possible that the
FW is still DMAing to the memory which was freed.

This series tries to prevent this by:

 - Returning and checking ring free command return values
 - Examining the FW response if the ring free command times out. It is
   possible that, for some reason, the FW did complete the ring free but was
   unable to respond with an IRQ. This seems unlikely given what appears to be
   a use after free in dmesg, but worth logging just in case.
 - Lastly, disable the device to stop DMA before the driver frees ring memory,
   which should prevent any possible use after free.

Sending this as an RFC so that the Broadcom folks have some time to take a
look and test as needed.

Thanks,
Joe

v2:
  - No changes to patch 1
  - Patch 2 from v1 dropped
  - Patch 2 in the v2 now checks the response and logs state before giving up
  - Patch 3 in the v2 disables the device to stop DMA before freeing ring
    memory

RFCv1: https://lore.kernel.org/netdev/20260917233218.1160001-1-joe@dama.to/

Joe Damato (3):
  bnxt_en: return the RING_FREE status to callers
  bnxt_en: check HWRM response if completion never arrives
  bnxt_en: stop DMA before releasing rings the firmware did not free

 drivers/net/ethernet/broadcom/bnxt/bnxt.c     | 69 +++++++++++------
 .../net/ethernet/broadcom/bnxt/bnxt_hwrm.c    | 75 ++++++++++++++-----
 2 files changed, 102 insertions(+), 42 deletions(-)


base-commit: 17741334d00bf5ebd37f8c1c36bc9c146a351deb
-- 
2.53.0-Meta