linux-kernel - Re: [PATCH v6 01/12] cgroup/misc: Add per resource callbacks for CSS events

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]

Date: Fri, 5 Jan 2024 10:45:02 +0100
From: Michal Koutný <mkoutny@...e.com>
To: Haitao Huang <haitao.huang@...ux.intel.com>
Cc: jarkko@...nel.org, dave.hansen@...ux.intel.com, tj@...nel.org, 
	linux-kernel@...r.kernel.org, linux-sgx@...r.kernel.org, x86@...nel.org, 
	cgroups@...r.kernel.org, tglx@...utronix.de, mingo@...hat.com, bp@...en8.de, 
	hpa@...or.com, sohil.mehta@...el.com, zhiquan1.li@...el.com, 
	kristen@...ux.intel.com, seanjc@...gle.com, zhanb@...rosoft.com, anakrish@...rosoft.com, 
	mikko.ylinen@...ux.intel.com, yangjie@...rosoft.com
Subject: Re: [PATCH v6 01/12] cgroup/misc: Add per resource callbacks for CSS
 events

On Mon, Oct 30, 2023 at 11:20:02AM -0700, Haitao Huang <haitao.huang@...ux.intel.com> wrote:
> From: Kristen Carlson Accardi <kristen@...ux.intel.com>
> 
> The misc cgroup controller (subsystem) currently does not perform
> resource type specific action for Cgroups Subsystem State (CSS) events:
> the 'css_alloc' event when a cgroup is created and the 'css_free' event
> when a cgroup is destroyed.
> 
> Define callbacks for those events and allow resource providers to
> register the callbacks per resource type as needed. This will be
> utilized later by the EPC misc cgroup support implemented in the SGX
> driver.

I notice now that the callbacks are per resource and per cgroup.
I think that is an overkill that complicates misc_cg allocation code
with parent's callbacks while freeing is done by own callbacks.
It's possibly too much liberal to keep objects' lifecycle understandable
in the long term too. 

For some uniformity, I'd suggest struct blkcg_policy array in
block/blk-cgroup.c as the precedent. (Perhaps indexed with static enum
misc_res_type instead of dynamic index assignment as
blkcg_policy_registeer() does.)

I hope you don't really need per-cgroup callbacks :-)

Michal

Download attachment "signature.asc" of type "application/pgp-signature" (229 bytes)