linux-kernel - Re: [Intel-wired-lan] [PATCH 2/2] igb: Make DMA faster when CPU is active on the PCIe link

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: <CAAd53p4h1-SJROvUghPYbBnh2Z9nRtgfNEagE4X6XtBwNg8JOg@mail.gmail.com>
Date:   Fri, 20 May 2022 10:45:33 +0800
From:   Kai-Heng Feng <kai.heng.feng@...onical.com>
To:     Paul Menzel <pmenzel@...gen.mpg.de>
Cc:     jesse.brandeburg@...el.com, anthony.l.nguyen@...el.com,
        netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
        Eric Dumazet <edumazet@...gle.com>,
        intel-wired-lan@...ts.osuosl.org, Jakub Kicinski <kuba@...nel.org>,
        Paolo Abeni <pabeni@...hat.com>,
        "David S. Miller" <davem@...emloft.net>
Subject: Re: [Intel-wired-lan] [PATCH 2/2] igb: Make DMA faster when CPU is
 active on the PCIe link

On Thu, May 12, 2022 at 10:55 AM Kai-Heng Feng
<kai.heng.feng@...onical.com> wrote:
>
> Hi Paul,
>
> On Wed, May 11, 2022 at 8:49 PM Paul Menzel <pmenzel@...gen.mpg.de> wrote:
> >
> > Dear Kai-Hang,
> >
> >
> > Thank you for the patch.
> >
> >
> > Am 11.05.22 um 14:28 schrieb Kai-Heng Feng:
> > > We found Intel I210 can only achieve ~750Mbps Tx speed on some
> > > platforms. The RR2DCDELAY shows around 0x2xxx DMA delay, which will be
> >
> > Please give an example platform, where it works and where it does not.
>
> The platform is about but not yet hit the market yet, so I can't disclose it.
> They are Intel Alder Lake based.
>
> >
> > How did you test transfer speed?
>
> Iperf.
>
> >
> > > significantly lower when 1) ASPM is disabled or 2) SoC package c-state
> > > stays above PC3. When the RR2DCDELAY is around 0x1xxx the Tx speed can
> > > reach to ~950Mbps.
> > >
> > > According to the I210 datasheet "8.26.1 PCIe Misc. Register - PCIEMISC",
> > > "DMA Idle Indication" doesn't seem to tie to DMA coalesce anymore, so
> > > set it to 1b for "DMA is considered idle when there is no Rx or Tx AND
> > > when there are no TLPs indicating that CPU is active detected on the
> > > PCIe link (such as the host executes CSR or Configuration register read
> > > or write operation)" and performing Tx should also fall under "active
> > > CPU on PCIe link" case.
> > >
> > > In addition to that, commit b6e0c419f040 ("igb: Move DMA Coalescing init
> > > code to separate function.") seems to wrongly changed from enabling
> > > E1000_PCIEMISC_LX_DECISION to disabling it, also fix that.
> >
> > Please split this into a separate commit with Fixes tag, and maybe the
> > commit author in Cc.
>
> I don't see the need to split to separate commit as both require the
> same change.
>
> I will add the "Fixes" tag once the igb maintainers reviewed the patch.

A gentle ping...

Please let me know if this is a proper fix so I can send v2.

Kai-Heng

>
> Kai-Heng
>
> >
> >
> > Kind regards,
> >
> > Paul
> >
> >
> > > Signed-off-by: Kai-Heng Feng <kai.heng.feng@...onical.com>
> > > ---
> > >   drivers/net/ethernet/intel/igb/igb_main.c | 12 +++++-------
> > >   1 file changed, 5 insertions(+), 7 deletions(-)
> > >
> > > diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
> > > index 34b33b21e0dcd..eca797dded429 100644
> > > --- a/drivers/net/ethernet/intel/igb/igb_main.c
> > > +++ b/drivers/net/ethernet/intel/igb/igb_main.c
> > > @@ -9897,11 +9897,10 @@ static void igb_init_dmac(struct igb_adapter *adapter, u32 pba)
> > >       struct e1000_hw *hw = &adapter->hw;
> > >       u32 dmac_thr;
> > >       u16 hwm;
> > > +     u32 reg;
> > >
> > >       if (hw->mac.type > e1000_82580) {
> > >               if (adapter->flags & IGB_FLAG_DMAC) {
> > > -                     u32 reg;
> > > -
> > >                       /* force threshold to 0. */
> > >                       wr32(E1000_DMCTXTH, 0);
> > >
> > > @@ -9934,7 +9933,6 @@ static void igb_init_dmac(struct igb_adapter *adapter, u32 pba)
> > >                       /* Disable BMC-to-OS Watchdog Enable */
> > >                       if (hw->mac.type != e1000_i354)
> > >                               reg &= ~E1000_DMACR_DC_BMC2OSW_EN;
> > > -
> > >                       wr32(E1000_DMACR, reg);
> > >
> > >                       /* no lower threshold to disable
> > > @@ -9951,12 +9949,12 @@ static void igb_init_dmac(struct igb_adapter *adapter, u32 pba)
> > >                        */
> > >                       wr32(E1000_DMCTXTH, (IGB_MIN_TXPBSIZE -
> > >                            (IGB_TX_BUF_4096 + adapter->max_frame_size)) >> 6);
> > > +             }
> > >
> > > -                     /* make low power state decision controlled
> > > -                      * by DMA coal
> > > -                      */
> > > +             if (hw->mac.type >= e1000_i210 ||
> > > +                 (adapter->flags & IGB_FLAG_DMAC)) {
> > >                       reg = rd32(E1000_PCIEMISC);
> > > -                     reg &= ~E1000_PCIEMISC_LX_DECISION;
> > > +                     reg |= E1000_PCIEMISC_LX_DECISION;
> > >                       wr32(E1000_PCIEMISC, reg);
> > >               } /* endif adapter->dmac is not disabled */
> > >       } else if (hw->mac.type == e1000_82580) {