linux-kernel - [PATCH v2 Cover Letter] thunderbolt: Disable Gen 4 Recovery on Asymmetric Transitions

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-next>] [day] [month] [year] [list]

Message-ID: <20250130095704.10779-1-rahimi.mhmmd@gmail.com>
Date: Thu, 30 Jan 2025 09:51:06 +0000
From: Mohammad Rahimi <rahimi.mhmmd@...il.com>
To: mika.westerberg@...ux.intel.com
Cc: andreas.noever@...il.com,
	michael.jamet@...el.com,
	YehezkelShB@...il.com,
	linux-usb@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	Mohammad Rahimi <rahimi.mhmmd@...il.com>
Subject: [PATCH v2 Cover Letter] thunderbolt: Disable Gen 4 Recovery on Asymmetric Transitions

Dear Mika,

In this revision, I have decided not to define the following functions,
despite our initial agreement in v1:

int tb_enable_link_recovery(struct tb_switch *sw);
int tb_disable_link_recovery(struct tb_switch *sw);

This decision was based on the following considerations:

1. If a write operation fails, there is no clear way to determine which port
the failure occurred on, the upstream port of the switch or its link partner.

2. To restore the exact Link Recovery state after an asymmetric transition, I
would need to track two boolean parameters and pass them to
tb_disable_link_recovery(). This would expose port state logic to
tb_configure_asym(), ending up with the same design we already have.

Additionally, I have chosen to ignore the return value from
usb4_port_link_recovery_enable() in tb_configure_asym(). At that stage, we
cannot take meaningful action based on the return value. Moreover, re-enabling
is performed immediately after tb_switch_set_link_width() to avoid introducing
unnecessary conditional branches.

Please let me know if you have any concerns or suggestions.

Best regards,
Mohammad Rahimi