public inbox for linux-rdma@vger.kernel.org
 help / color / mirror / Atom feed
* Handling RDMA_CM_EVENT_DEVICE_REMOVAL in rdma-core
@ 2025-01-25  0:30 Rita Han (she/her)
  2025-01-26 12:53 ` Leon Romanovsky
  0 siblings, 1 reply; 2+ messages in thread
From: Rita Han (she/her) @ 2025-01-25  0:30 UTC (permalink / raw)
  To: linux-rdma; +Cc: Matthew Carlis

Hi all,

I am currently working on validating PCIe DPC above Mellanox devices,
and I've encountered issues where the device does not recover after
injecting DPC above it with ENOMEM returned from rdma_create_qp. It
seems that the rdma-core library does not support hotplugging or
handle the removal of InfiniBand devices
(RDMA_CM_EVENT_DEVICE_REMOVAL).

Are there any new patches that address this issue? If not, what would
be the correct approach to handle this event or recover the device?
Would it require us to free/destroy and reinitialize the device?

Best regards,
Rita Han

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: Handling RDMA_CM_EVENT_DEVICE_REMOVAL in rdma-core
  2025-01-25  0:30 Handling RDMA_CM_EVENT_DEVICE_REMOVAL in rdma-core Rita Han (she/her)
@ 2025-01-26 12:53 ` Leon Romanovsky
  0 siblings, 0 replies; 2+ messages in thread
From: Leon Romanovsky @ 2025-01-26 12:53 UTC (permalink / raw)
  To: Rita Han (she/her); +Cc: linux-rdma, Matthew Carlis

On Fri, Jan 24, 2025 at 04:30:33PM -0800, Rita Han (she/her) wrote:
> Hi all,
> 
> I am currently working on validating PCIe DPC above Mellanox devices,
> and I've encountered issues where the device does not recover after
> injecting DPC above it with ENOMEM returned from rdma_create_qp. It
> seems that the rdma-core library does not support hotplugging or
> handle the removal of InfiniBand devices
> (RDMA_CM_EVENT_DEVICE_REMOVAL).

I was under impression that rdma-core supports it.
RDMA-CM disassociate support - https://github.com/linux-rdma/rdma-core/pull/750

Thanks

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2025-01-26 12:53 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-01-25  0:30 Handling RDMA_CM_EVENT_DEVICE_REMOVAL in rdma-core Rita Han (she/her)
2025-01-26 12:53 ` Leon Romanovsky

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox