[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAMdnO-JSPEkJZLMPL8BLy1DKFUtU4aC_JpaXfs=g5YJFBu4_RQ@mail.gmail.com>
Date: Mon, 26 Aug 2024 10:41:18 -0700
From: Jitendra Vegiraju <jitendra.vegiraju@...adcom.com>
To: Amit Singh Tomar <amitsinght@...vell.com>
Cc: netdev@...r.kernel.org, alexandre.torgue@...s.st.com, joabreu@...opsys.com,
davem@...emloft.net, edumazet@...gle.com, kuba@...nel.org, pabeni@...hat.com,
mcoquelin.stm32@...il.com, bcm-kernel-feedback-list@...adcom.com,
richardcochran@...il.com, ast@...nel.org, daniel@...earbox.net,
hawk@...nel.org, john.fastabend@...il.com, fancer.lancer@...il.com,
rmk+kernel@...linux.org.uk, ahalaney@...hat.com, xiaolei.wang@...driver.com,
rohan.g.thomas@...el.com, Jianheng.Zhang@...opsys.com,
linux-kernel@...r.kernel.org, linux-stm32@...md-mailman.stormreply.com,
linux-arm-kernel@...ts.infradead.org, bpf@...r.kernel.org, andrew@...n.ch,
linux@...linux.org.uk, horms@...nel.org, florian.fainelli@...adcom.com
Subject: Re: [net-next v4 2/5] net: stmmac: Add basic dw25gmac support to
stmmac core
Hi Amit,
On Thu, Aug 22, 2024 at 10:17 AM Amit Singh Tomar
<amitsinght@...vell.com> wrote:
>
> Hi,
>
> > The BCM8958x uses early adaptor version of DWC_xgmac version 4.00a for
> > ethernet MAC. The DW25GMAC introduced in this version adds new DMA
> > architecture called Hyper-DMA (HDMA) for virtualization scalability.
> > This is realized by decoupling physical DMA channels(PDMA) from potentially
> > large number of virtual DMA channels (VDMA). The VDMAs are software
> > abastractions that map to PDMAs for frame transmission and reception.
> You should either run ./scripts/checkpatch.pl --strict --codespell
> --patch or use :set spell in vi to check for spelling mistakes
Thank you, I will do that next time.
> > +static u32 decode_vdma_count(u32 regval)
> > +{
> > + /* compressed encoding for vdma count
> > + * regval: VDMA count
> > + * 0-15 : 1 - 16
> > + * 16-19 : 20, 24, 28, 32
> > + * 20-23 : 40, 48, 56, 64
> > + * 24-27 : 80, 96, 112, 128
> > + */
> > + if (regval < 16)
> > + return regval + 1;
> > + return (4 << ((regval - 16) / 4)) * ((regval % 4) + 5);
> Is there a potential for regval to be out of bounds (regval > 27) that
> needed to be handled properly?
The IP core documentation we have only defines support upto 128 VDMAs.
The same formula should for higher values (bound by bitmask).
> > diff --git a/drivers/net/ethernet/stmicro/stmmac/dw25gmac.h b/drivers/net/ethernet/stmicro/stmmac/dw25gmac.h
> > new file mode 100644
> > index 000000000000..c7fdf6624fea
> > --- /dev/null
> > +++ b/drivers/net/ethernet/stmicro/stmmac/dw25gmac.h
> > @@ -0,0 +1,90 @@
> > +/* SPDX-License-Identifier: GPL-2.0-only */
> > +/* Copyright (c) 2024 Broadcom Corporation
> > + * DW25GMAC definitions.
> > + */
> > +
> > +#define XXVGMAC_ADDR_OFFSET GENMASK(14, 8)
> > +#define XXVGMAC_AUTO_INCR GENMASK(5, 4)
> > +#define XXVGMAC_CMD_TYPE BIT(1)
> > +#define XXVGMAC_OB BIT(0)
> > +#define XXVGMAC_DMA_CH_IND_DATA 0X00003084
> nit: lower case please, 0x00003084.
Thanks, will fix it.
Powered by blists - more mailing lists