lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZWRkN12fhENyN4PY@nanopsycho>
Date:   Mon, 27 Nov 2023 10:41:11 +0100
From:   Jiri Pirko <jiri@...nulli.us>
To:     Sachin Bahadur <sachin.bahadur@...el.com>
Cc:     intel-wired-lan@...ts.osuosl.org, netdev@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH iwl-net v4] ice: Block PF reinit if attached to bond

Mon, Nov 27, 2023 at 07:05:12AM CET, sachin.bahadur@...el.com wrote:
>PF interface part of Bond should not allow driver reinit via devlink. Bond
>config will be lost due to PF reinit. PF needs to be re-added to Bond
>after PF reinit. ice_devlink_reload_down is called before PF driver reinit.
>If PF is attached to bond, ice_devlink_reload_down returns error.
>
>Fixes: trailer
>Reviewed-by: Jacob Keller <jacob.e.keller@...el.com>
>Signed-off-by: Sachin Bahadur <sachin.bahadur@...el.com>
>---
> drivers/net/ethernet/intel/ice/ice_devlink.c | 4 ++++
> 1 file changed, 4 insertions(+)
>
>diff --git a/drivers/net/ethernet/intel/ice/ice_devlink.c b/drivers/net/ethernet/intel/ice/ice_devlink.c
>index f4e24d11ebd0..5fe88e949b09 100644
>--- a/drivers/net/ethernet/intel/ice/ice_devlink.c
>+++ b/drivers/net/ethernet/intel/ice/ice_devlink.c
>@@ -457,6 +457,10 @@ ice_devlink_reload_down(struct devlink *devlink, bool netns_change,
> 					   "Remove all VFs before doing reinit\n");
> 			return -EOPNOTSUPP;
> 		}
>+		if (pf->lag && pf->lag->bonded) {
>+			NL_SET_ERR_MSG_MOD(extack, "Remove all associated Bonds before doing reinit");

Nack. Remove the netdev during re-init, that would solve your issue.
Looks like some checks are needed to be added in devlink code to make
sure drivers behave properly. I'm on in.



>+			return -EBUSY;
>+		}
> 		ice_unload(pf);
> 		return 0;
> 	case DEVLINK_RELOAD_ACTION_FW_ACTIVATE:
>--
>2.25.1
>
>

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ