netdev - Re: [PATCH iwl-net] idpf: enable WB_ON

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] [day] [month] [year] [list]

Date: Fri, 15 Dec 2023 20:04:52 +0100
From: Michal Kubiak <michal.kubiak@...el.com>
To: Brett Creeley <bcreeley@....com>
CC: <intel-wired-lan@...ts.osuosl.org>, <aleksander.lobakin@...el.com>,
	<larysa.zaremba@...el.com>, <alan.brady@...el.com>, <joshua.a.hay@...el.com>,
	<emil.s.tantilov@...el.com>, <maciej.fijalkowski@...el.com>,
	<netdev@...r.kernel.org>, Przemek Kitszel <przemyslaw.kitszel@...el.com>
Subject: Re: [PATCH iwl-net] idpf: enable WB_ON_ITR

On Fri, Dec 15, 2023 at 10:01:36AM -0800, Brett Creeley wrote:
> 

[...]

> 
> > 
> > diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.h b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
> > index df76493faa75..50761c2d9f3b 100644
> > --- a/drivers/net/ethernet/intel/idpf/idpf_txrx.h
> > +++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
> > @@ -495,9 +495,11 @@ struct idpf_vec_regs {
> >   struct idpf_intr_reg {
> >          void __iomem *dyn_ctl;
> >          u32 dyn_ctl_intena_m;
> > +       u32 dyn_ctl_intena_msk_m;
> >          u32 dyn_ctl_itridx_s;
> >          u32 dyn_ctl_itridx_m;
> >          u32 dyn_ctl_intrvl_s;
> > +       u32 dyn_ctl_wb_on_itr_m;
> >          void __iomem *rx_itr;
> >          void __iomem *tx_itr;
> >          void __iomem *icr_ena;
> > @@ -534,6 +536,7 @@ struct idpf_q_vector {
> >          struct napi_struct napi;
> >          u16 v_idx;
> >          struct idpf_intr_reg intr_reg;
> > +       bool wb_on_itr;
> 
> Not sure if this was considered, but it might be best to put this before
> intr_reg so it's on the same cacheline as intr_reg.
> 

Good point! It wasn't considered before.
I have just confirmed with pahole that it is worth putting that flag
before the register structure in terms of cacheline.

Thank you for your suggestion. I will reorder this.

Thanks,
Michal