From: Louis Peens <louis.peens@corigine.com>
To: Caleb Sander <csander@purestorage.com>
Cc: "Andrew Lunn" <andrew+netdev@lunn.ch>,
"AngeloGioacchino Del Regno"
<angelogioacchino.delregno@collabora.com>,
"Arthur Kiyanovski" <akiyano@amazon.com>,
"Brett Creeley" <brett.creeley@amd.com>,
"Broadcom internal kernel review list"
<bcm-kernel-feedback-list@broadcom.com>,
"Christophe Leroy" <christophe.leroy@csgroup.eu>,
"Claudiu Manoil" <claudiu.manoil@nxp.com>,
"David Arinzon" <darinzon@amazon.com>,
"David S. Miller" <davem@davemloft.net>,
"Doug Berger" <opendmb@gmail.com>,
"Eric Dumazet" <edumazet@google.com>,
"Eugenio Pérez" <eperezma@redhat.com>,
"Felix Fietkau" <nbd@nbd.name>,
"Florian Fainelli" <florian.fainelli@broadcom.com>,
"Geetha sowjanya" <gakula@marvell.com>,
hariprasad <hkelam@marvell.com>,
"Jakub Kicinski" <kuba@kernel.org>,
"Jason Wang" <jasowang@redhat.com>,
"Jonathan Corbet" <corbet@lwn.net>,
"Leon Romanovsky" <leon@kernel.org>,
"Lorenzo Bianconi" <lorenzo@kernel.org>,
"Mark Lee" <Mark-MC.Lee@mediatek.com>,
"Matthias Brugger" <matthias.bgg@gmail.com>,
"Michael Chan" <michael.chan@broadcom.com>,
"Michael S. Tsirkin" <mst@redhat.com>,
"Noam Dagan" <ndagan@amazon.com>,
"Paolo Abeni" <pabeni@redhat.com>,
"Przemek Kitszel" <przemyslaw.kitszel@intel.com>,
"Roy Pledge" <Roy.Pledge@nxp.com>,
"Saeed Bishara" <saeedb@amazon.com>,
"Saeed Mahameed" <saeedm@nvidia.com>,
"Sean Wang" <sean.wang@mediatek.com>,
"Shannon Nelson" <shannon.nelson@amd.com>,
"Shay Agroskin" <shayagr@amazon.com>,
"Simon Horman" <horms@kernel.org>,
"Subbaraya Sundeep" <sbhatta@marvell.com>,
"Sunil Goutham" <sgoutham@marvell.com>,
"Tal Gilboa" <talgi@nvidia.com>,
"Tariq Toukan" <tariqt@nvidia.com>,
"Tony Nguyen" <anthony.l.nguyen@intel.com>,
"Vladimir Oltean" <vladimir.oltean@nxp.com>,
"Xuan Zhuo" <xuanzhuo@linux.alibaba.com>,
intel-wired-lan@lists.osuosl.org,
linux-arm-kernel@lists.infradead.org, linux-doc@vger.kernel.org,
linux-kernel@vger.kernel.org, linux-mediatek@lists.infradead.org,
linuxppc-dev@lists.ozlabs.org, linux-rdma@vger.kernel.org,
netdev@vger.kernel.org, oss-drivers@corigine.com,
virtualization@lists.linux.dev
Subject: Re: [resend PATCH 2/2] dim: pass dim_sample to net_dim() by reference
Date: Fri, 1 Nov 2024 10:54:47 +0200 [thread overview]
Message-ID: <ZySXV46T4IE8YVqX@LouisNoVo> (raw)
In-Reply-To: <CADUfDZoba9hNOBU7TT+0K6BYiYzVkZ_awt751g6HBm+-cCZf8w@mail.gmail.com>
On Thu, Oct 31, 2024 at 10:19:55AM -0700, Caleb Sander wrote:
> [Some people who received this message don't often get email from csander@purestorage.com. Learn why this is important at https://aka.ms/LearnAboutSenderIdentification ]
>
> On Thu, Oct 31, 2024 at 5:49 AM Louis Peens <louis.peens@corigine.com> wrote:
> >
> > On Wed, Oct 30, 2024 at 06:23:26PM -0600, Caleb Sander Mateos wrote:
> > > net_dim() is currently passed a struct dim_sample argument by value.
> > > struct dim_sample is 24 bytes. Since this is greater 16 bytes, x86-64
> > > passes it on the stack. All callers have already initialized dim_sample
> > > on the stack, so passing it by value requires pushing a duplicated copy
> > > to the stack. Either witing to the stack and immediately reading it, or
> > > perhaps dereferencing addresses relative to the stack pointer in a chain
> > > of push instructions, seems to perform quite poorly.
> > >
> > > In a heavy TCP workload, mlx5e_handle_rx_dim() consumes 3% of CPU time,
> > > 94% of which is attributed to the first push instruction to copy
> > > dim_sample on the stack for the call to net_dim():
> > > // Call ktime_get()
> > > 0.26 |4ead2: call 4ead7 <mlx5e_handle_rx_dim+0x47>
> > > // Pass the address of struct dim in %rdi
> > > |4ead7: lea 0x3d0(%rbx),%rdi
> > > // Set dim_sample.pkt_ctr
> > > |4eade: mov %r13d,0x8(%rsp)
> > > // Set dim_sample.byte_ctr
> > > |4eae3: mov %r12d,0xc(%rsp)
> > > // Set dim_sample.event_ctr
> > > 0.15 |4eae8: mov %bp,0x10(%rsp)
> > > // Duplicate dim_sample on the stack
> > > 94.16 |4eaed: push 0x10(%rsp)
> > > 2.79 |4eaf1: push 0x10(%rsp)
> > > 0.07 |4eaf5: push %rax
> > > // Call net_dim()
> > > 0.21 |4eaf6: call 4eafb <mlx5e_handle_rx_dim+0x6b>
> > >
> > > To allow the caller to reuse the struct dim_sample already on the stack,
> > > pass the struct dim_sample by reference to net_dim().
> > >
> > > Signed-off-by: Caleb Sander Mateos <csander@purestorage.com>
> > > ---
> > > Documentation/networking/net_dim.rst | 2 +-
> > > drivers/net/ethernet/amazon/ena/ena_netdev.c | 2 +-
> > > drivers/net/ethernet/broadcom/bcmsysport.c | 2 +-
> > > drivers/net/ethernet/broadcom/bnxt/bnxt.c | 4 ++--
> > > drivers/net/ethernet/broadcom/genet/bcmgenet.c | 2 +-
> > > drivers/net/ethernet/freescale/enetc/enetc.c | 2 +-
> > > drivers/net/ethernet/hisilicon/hns3/hns3_enet.c | 4 ++--
> > > drivers/net/ethernet/intel/ice/ice_txrx.c | 4 ++--
> > > drivers/net/ethernet/intel/idpf/idpf_txrx.c | 4 ++--
> > > drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c | 2 +-
> > > drivers/net/ethernet/mediatek/mtk_eth_soc.c | 4 ++--
> > > drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c | 4 ++--
> > > drivers/net/ethernet/netronome/nfp/nfd3/dp.c | 4 ++--
> > > drivers/net/ethernet/netronome/nfp/nfdk/dp.c | 4 ++--
> > > drivers/net/ethernet/pensando/ionic/ionic_txrx.c | 2 +-
> > > drivers/net/virtio_net.c | 2 +-
> > > drivers/soc/fsl/dpio/dpio-service.c | 2 +-
> > > include/linux/dim.h | 2 +-
> > > lib/dim/net_dim.c | 10 +++++-----
> > > 19 files changed, 31 insertions(+), 31 deletions(-)
> > >
> > --- snip --
> >
> > > diff --git a/drivers/net/ethernet/netronome/nfp/nfd3/dp.c b/drivers/net/ethernet/netronome/nfp/nfd3/dp.c
> > > index d215efc6cad0..f1c6c47564b1 100644
> > > --- a/drivers/net/ethernet/netronome/nfp/nfd3/dp.c
> > > +++ b/drivers/net/ethernet/netronome/nfp/nfd3/dp.c
> > > @@ -1177,11 +1177,11 @@ int nfp_nfd3_poll(struct napi_struct *napi, int budget)
> > > pkts = r_vec->rx_pkts;
> > > bytes = r_vec->rx_bytes;
> > > } while (u64_stats_fetch_retry(&r_vec->rx_sync, start));
> > >
> > > dim_update_sample(r_vec->event_ctr, pkts, bytes, &dim_sample);
> > > - net_dim(&r_vec->rx_dim, dim_sample);
> > > + net_dim(&r_vec->rx_dim, &dim_sample);
> > > }
> > >
> > > if (r_vec->nfp_net->tx_coalesce_adapt_on && r_vec->tx_ring) {
> > > struct dim_sample dim_sample = {};
> > > unsigned int start;
> > > @@ -1192,11 +1192,11 @@ int nfp_nfd3_poll(struct napi_struct *napi, int budget)
> > > pkts = r_vec->tx_pkts;
> > > bytes = r_vec->tx_bytes;
> > > } while (u64_stats_fetch_retry(&r_vec->tx_sync, start));
> > >
> > > dim_update_sample(r_vec->event_ctr, pkts, bytes, &dim_sample);
> > > - net_dim(&r_vec->tx_dim, dim_sample);
> > > + net_dim(&r_vec->tx_dim, &dim_sample);
> > > }
> > >
> > > return pkts_polled;
> > > }
> > >
> > > diff --git a/drivers/net/ethernet/netronome/nfp/nfdk/dp.c b/drivers/net/ethernet/netronome/nfp/nfdk/dp.c
> > > index dae5af7d1845..ebeb6ab4465c 100644
> > > --- a/drivers/net/ethernet/netronome/nfp/nfdk/dp.c
> > > +++ b/drivers/net/ethernet/netronome/nfp/nfdk/dp.c
> > > @@ -1287,11 +1287,11 @@ int nfp_nfdk_poll(struct napi_struct *napi, int budget)
> > > pkts = r_vec->rx_pkts;
> > > bytes = r_vec->rx_bytes;
> > > } while (u64_stats_fetch_retry(&r_vec->rx_sync, start));
> > >
> > > dim_update_sample(r_vec->event_ctr, pkts, bytes, &dim_sample);
> > > - net_dim(&r_vec->rx_dim, dim_sample);
> > > + net_dim(&r_vec->rx_dim, &dim_sample);
> > > }
> > >
> > > if (r_vec->nfp_net->tx_coalesce_adapt_on && r_vec->tx_ring) {
> > > struct dim_sample dim_sample = {};
> > > unsigned int start;
> > > @@ -1302,11 +1302,11 @@ int nfp_nfdk_poll(struct napi_struct *napi, int budget)
> > > pkts = r_vec->tx_pkts;
> > > bytes = r_vec->tx_bytes;
> > > } while (u64_stats_fetch_retry(&r_vec->tx_sync, start));
> > >
> > > dim_update_sample(r_vec->event_ctr, pkts, bytes, &dim_sample);
> > > - net_dim(&r_vec->tx_dim, dim_sample);
> > > + net_dim(&r_vec->tx_dim, &dim_sample);
> > > }
> > >
> > > return pkts_polled;
> > > }
> > --- snip ---
> >
> > Hi Caleb. Looks like a fair enough update to me in general, but I am not an
> > expert on 'dim'. For the corresponding nfp driver changes feel free to add:
> >
> > Signed-off-by: Louis Peens <louis.peens@corigine.com>
>
> Hi Louis,
> Thanks for the review. Did you mean "Reviewed-by"? If there was a
> change you were suggesting, I missed it.
Hi - sorry, I do still manage to mix up when to use signed-off-by and
reviewed-by. I did not suggest any changes no, and since the main focus of the
patch is not the nfp driver I can see in hindsight that Reviewed-by: may make
more sense. So updated:
Reviewed-by: Louis Peens <louis.peens@corigine.com>
>
> Best,
> Caleb
next prev parent reply other threads:[~2024-11-01 8:55 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-10-31 0:23 [resend PATCH 1/2] dim: make dim_calc_stats() inputs const pointers Caleb Sander Mateos
2024-10-31 0:23 ` [resend PATCH 2/2] dim: pass dim_sample to net_dim() by reference Caleb Sander Mateos
2024-10-31 12:49 ` Louis Peens
2024-10-31 17:19 ` Caleb Sander
2024-11-01 8:54 ` Louis Peens [this message]
2024-11-01 14:11 ` Vladimir Oltean
2024-10-31 16:48 ` Vladimir Oltean
2024-10-31 16:57 ` Nelson, Shannon
2024-10-31 17:17 ` Florian Fainelli
2024-10-31 18:28 ` Kiyanovski, Arthur
2024-11-03 20:21 ` Jakub Kicinski
2024-11-03 22:50 ` Caleb Sander
2024-11-04 2:52 ` Xuan Zhuo
2024-10-31 16:44 ` [resend PATCH 1/2] dim: make dim_calc_stats() inputs const pointers Vladimir Oltean
2024-10-31 17:15 ` Florian Fainelli
2024-10-31 18:22 ` Kiyanovski, Arthur
2024-11-04 2:45 ` Xuan Zhuo
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ZySXV46T4IE8YVqX@LouisNoVo \
--to=louis.peens@corigine.com \
--cc=Mark-MC.Lee@mediatek.com \
--cc=Roy.Pledge@nxp.com \
--cc=akiyano@amazon.com \
--cc=andrew+netdev@lunn.ch \
--cc=angelogioacchino.delregno@collabora.com \
--cc=anthony.l.nguyen@intel.com \
--cc=bcm-kernel-feedback-list@broadcom.com \
--cc=brett.creeley@amd.com \
--cc=christophe.leroy@csgroup.eu \
--cc=claudiu.manoil@nxp.com \
--cc=corbet@lwn.net \
--cc=csander@purestorage.com \
--cc=darinzon@amazon.com \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=eperezma@redhat.com \
--cc=florian.fainelli@broadcom.com \
--cc=gakula@marvell.com \
--cc=hkelam@marvell.com \
--cc=horms@kernel.org \
--cc=intel-wired-lan@lists.osuosl.org \
--cc=jasowang@redhat.com \
--cc=kuba@kernel.org \
--cc=leon@kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mediatek@lists.infradead.org \
--cc=linux-rdma@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=lorenzo@kernel.org \
--cc=matthias.bgg@gmail.com \
--cc=michael.chan@broadcom.com \
--cc=mst@redhat.com \
--cc=nbd@nbd.name \
--cc=ndagan@amazon.com \
--cc=netdev@vger.kernel.org \
--cc=opendmb@gmail.com \
--cc=oss-drivers@corigine.com \
--cc=pabeni@redhat.com \
--cc=przemyslaw.kitszel@intel.com \
--cc=saeedb@amazon.com \
--cc=saeedm@nvidia.com \
--cc=sbhatta@marvell.com \
--cc=sean.wang@mediatek.com \
--cc=sgoutham@marvell.com \
--cc=shannon.nelson@amd.com \
--cc=shayagr@amazon.com \
--cc=talgi@nvidia.com \
--cc=tariqt@nvidia.com \
--cc=virtualization@lists.linux.dev \
--cc=vladimir.oltean@nxp.com \
--cc=xuanzhuo@linux.alibaba.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).