From: Heiko Stuebner <heiko@sntech.de>
To: u-boot@lists.denx.de
Subject: [U-Boot] [PATCH 1/3] rockchip: rk3399: add tpl support
Date: Thu, 25 Apr 2019 11:00:10 +0200 [thread overview]
Message-ID: <11199372.bADmE1qTQZ@phil> (raw)
In-Reply-To: <CAPnjgZ13MQEUdu8tWf+3UETdWKAvxbvdWC2Wei91dTroNmJCFg@mail.gmail.com>
Hi Simon,
Am Donnerstag, 25. April 2019, 00:26:28 CEST schrieb Simon Glass:
> On Thu, 18 Apr 2019 at 10:47, Heiko Stübner <heiko@sntech.de> wrote:
> >
> > Am Donnerstag, 18. April 2019, 08:35:49 CEST schrieb Philipp Tomsich:
> > > Simon,
> > >
> > > > On 18.04.2019, at 06:32, Simon Glass <sjg@chromium.org> wrote:
> > > >
> > > > Hi Kever,
> > > >
> > > > On Mon, 1 Apr 2019 at 02:21, Kever Yang <kever.yang@rock-chips.com> wrote:
> > > >>
> > > >> Rockchip platform suppose to use TPL(run in SRAM) as dram init and
> > > >> SPL(run in DDR SDRAM) as pre-loader, so that the SPL would not be
> > > >> limited by SRAM size.
> > > >> This patch add rk3399-board-tpl.c and its common configs.
> > > >
> > > > So this means that TPL inits SDRAM? That seems strange to me...why
> > > > have SPL at all, then? What is SPL supported to do on RK3399
> > > > platforms?
> > >
> > > For TPL->SPL implementation on Rockchip, we generally rely on the BootROM
> > > to load the TPL stage (to SRAM) and the SPL stage (to the start of DRAM). The
> > > BootROM usually limits the size of the SPL stage, so we can’t use a full U-Boot
> > > already.
> > >
> > > Even more constricting are the TPL size constraints (as this seems to have originally
> > > only been intended to be size-optimized DRAM init code) on some devices.
> >
> > and maybe for a broader functionality description, something needs to
> > load things like ATF via a FIT image from a mmc for example.
> > (which needs FIT code and also real mmc drivers).
> >
> > And as we experienced on rk3288 as well recently, SRAM may be very well
> > too small to accomplish that, so TPL needs to init the sdram separately
> > to make room for the code needed to get ATF up and running, which must
> > be done before jumping into proper uboot.
>
> I have to ask what the boot ROM is buying us? U-Boot TPL/SPL can load
> programs itself without the boot ROM and its size limit. Why not just
> stop using the boot ROM on these devices?
The problem is more that the sram is to small to hold mmc + fit + everything
else. So we need at least on some socs
bootrom -> tpl (in sram) -> sdram init -> bootrom -> spl (in ddr)
(1) -> load fit with u-boot and atf binaries -> jump to ATF -> jump to uboot
or in falcon mode for vyasa for example
(2) -> load linux directly -> jump to Linux
Heiko
next prev parent reply other threads:[~2019-04-25 9:00 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-04-01 9:20 [U-Boot] [PATCH 1/3] rockchip: rk3399: add tpl support Kever Yang
2019-04-01 9:20 ` [U-Boot] [PATCH 2/3] rockchip: ram: rk3399: update for TPL Kever Yang
2019-04-29 9:32 ` Andy Yan
2019-04-01 9:20 ` [U-Boot] [PATCH 3/3] rockchip: rk3399: update defconfig " Kever Yang
2019-04-29 9:36 ` Andy Yan
2019-04-18 4:32 ` [U-Boot] [PATCH 1/3] rockchip: rk3399: add tpl support Simon Glass
2019-04-18 6:35 ` Philipp Tomsich
2019-04-18 16:47 ` Heiko Stübner
2019-04-24 22:26 ` Simon Glass
2019-04-25 9:00 ` Heiko Stuebner [this message]
2019-04-26 1:07 ` Simon Glass
2019-04-18 7:07 ` [U-Boot] [PATCH 1/3] rockchip: rk3399: add tpl support【请注意,邮件由sjg@google.com代发】 Kever Yang
2019-04-18 7:34 ` Jagan Teki
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=11199372.bADmE1qTQZ@phil \
--to=heiko@sntech.de \
--cc=u-boot@lists.denx.de \
/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