[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <D3QWDGUOJ4NF.1I16RTUH9Z45U@kernel.org>
Date: Tue, 27 Aug 2024 21:15:15 +0300
From: "Jarkko Sakkinen" <jarkko@...nel.org>
To: "Haitao Huang" <haitao.huang@...ux.intel.com>,
<dave.hansen@...ux.intel.com>, <kai.huang@...el.com>, <tj@...nel.org>,
<mkoutny@...e.com>, <chenridong@...wei.com>,
<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>, <tim.c.chen@...ux.intel.com>
Cc: <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>,
<chrisyan@...rosoft.com>
Subject: Re: [PATCH v16 08/16] x86/sgx: Encapsulate uses of the global LRU
On Wed Aug 21, 2024 at 4:53 AM EEST, Haitao Huang wrote:
> To support the per-cgroup reclamation, each cgroup will have its own
> "per-cgroup LRU" and EPC pages will be in its owner cgroup's LRU instead
> of the global LRU. Abstract the code that is directly working with the
> global LRU into functions reusable with per-cgroup LRUs.
>
> Currently the basic reclamation procedure, sgx_reclaim_pages() directly
> reclaims pages from the global LRU. Change it to take in an LRU.
>
> Note the global EPC reclamation will still be needed when the total EPC
> usage reaches the system capacity while usages of some cgroups are below
> their respective limits. Create a separate wrapper for the global
> reclamation, sgx_reclaim_pages_global(), passing in the global LRU to
> the new sgx_reclaim_pages() now. Later it will be revised to reclaim
> from multiple LRUs from all EPC cgroups instead of a single global LRU.
>
> Wrap the existing emptiness check of the global LRU with a helper so
> that it can be changed later to work with multiple LRUs when per-cgroup
> LRU comes to play.
>
> Also the per-cgroup EPC reclaim and global EPC reclaim will have
> different check on whether they should be done. Rename the existing
> sgx_should_reclaim() to sgx_should_reclaim_global() to separate the two
> cases.
>
> Co-developed-by: Sean Christopherson <sean.j.christopherson@...el.com>
> Signed-off-by: Sean Christopherson <sean.j.christopherson@...el.com>
> Co-developed-by: Kristen Carlson Accardi <kristen@...ux.intel.com>
> Signed-off-by: Kristen Carlson Accardi <kristen@...ux.intel.com>
> Signed-off-by: Haitao Huang <haitao.huang@...ux.intel.com>
Reviewed-by: Jarkko Sakkinen <jarkko@...nel.org>
BR, Jarkko
Powered by blists - more mailing lists