[U-Boot] [PATCH v6 13/16] arm: socfpga: Add SPL support for Arria 10
Dalon Westergreen
dalon.westergreen at linux.intel.com
Thu Apr 20 20:00:03 UTC 2017
On Thu, 2017-04-20 at 07:12 -0700, Dalon Westergreen wrote:
> On Wed, 2017-04-19 at 23:58 -0500, Dinh Nguyen wrote:
> >
> > On Wed, Apr 19, 2017 at 6:21 PM, Dalon Westergreen
> > <dalon.westergreen at linux.intel.com> wrote:
> > >
> > >
> > > On Wed, 2017-04-19 at 13:54 -0700, Dalon Westergreen wrote:
> > > >
> > > >
> > > > On Wed, 2017-04-19 at 15:44 -0500, Dinh Nguyen wrote:
> > > > >
> > > > >
> > > > >
> > > > > Really including Dalon
> > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > On Wed, Apr 19, 2017 at 3:26 PM, Dinh Nguyen <dinguyen at kernel.org>
> > > > > > wrote:
> > > > > > CC: Dalon Westergreen
> > > > > >
> > > > > > On 04/19/2017 02:49 PM, Dinh Nguyen wrote:
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > On 04/19/2017 04:29 AM, Ley Foon Tan wrote:
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > Add SPL support for Arria 10.
> > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > > Signed-off-by: Tien Fong Chee <tien.fong.chee at intel.com>
> > > > > > > > > > > > Signed-off-by: Ley Foon Tan <ley.foon.tan at intel.com>
> > > > > > > > ---
> > > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > >
> > > > > > > > > > > > arch/arm/mach-socfpga/spl.c | 72
> > > > +++++++++++++++++++++++++++++++++++++++++----
> > > > >
> > > > >
> > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > 1 file changed, 67 insertions(+), 5 deletions(-)
> > > > > > > >
> > > > > > > > diff --git a/arch/arm/mach-socfpga/spl.c b/arch/arm/mach-
> > > > > > > > socfpga/spl.c
> > > > > > > > index 0064fc8..f4a3cdd 100644
> > > > > > > > --- a/arch/arm/mach-socfpga/spl.c
> > > > > > > > +++ b/arch/arm/mach-socfpga/spl.c
> > > > > > > > @@ -19,23 +19,32 @@
> > > > > > > > #include <asm/arch/sdram.h>
> > > > > > > > #include <asm/arch/scu.h>
> > > > > > > > #include <asm/arch/nic301.h>
> > > > > > > > +#include <asm/sections.h>
> > > > > > > > +#include <fdtdec.h>
> > > > > > > > +#include <watchdog.h>
> > > > > > > > +#if defined(CONFIG_TARGET_SOCFPGA_ARRIA10)
> > > > > > > > +#include <asm/arch/pinmux.h>
> > > > > > > > +#endif
> > > > > > > >
> > > > > > > > DECLARE_GLOBAL_DATA_PTR;
> > > > > > > >
> > > > > > > > +#if defined(CONFIG_TARGET_SOCFPGA_GEN5)
> > > > > > > > static struct pl310_regs *const pl310 =
> > > > > > > > (struct pl310_regs *)CONFIG_SYS_PL310_BASE;
> > > > > > > > static struct scu_registers *scu_regs =
> > > > > > > > (struct scu_registers *)SOCFPGA_MPUSCU_ADDRESS;
> > > > > > > > static struct nic301_registers *nic301_regs =
> > > > > > > > (struct nic301_registers *)SOCFPGA_L3REGS_ADDRESS;
> > > > > > > > -static struct socfpga_system_manager *sysmgr_regs =
> > > > > > > > +#endif
> > > > > > > > +
> > > > > > > > +static const struct socfpga_system_manager *sysmgr_regs =
> > > > > > > > (struct socfpga_system_manager *)SOCFPGA_SYSMGR_ADDRESS;
> > > > > > > >
> > > > > > > > u32 spl_boot_device(void)
> > > > > > > > {
> > > > > > > > const u32 bsel = readl(&sysmgr_regs->bootinfo);
> > > > > > > >
> > > > > > > > - switch (bsel & 0x7) {
> > > > > > > > + switch (SYSMGR_GET_BOOTINFO_BSEL(bsel)) {
> > > > > > > > case 0x1: /* FPGA (HPS2FPGA Bridge) */
> > > > > > > > return BOOT_DEVICE_RAM;
> > > > > > > > case 0x2: /* NAND Flash (1.8V) */
> > > > > > > > @@ -68,6 +77,7 @@ u32 spl_boot_mode(const u32 boot_device)
> > > > > > > > }
> > > > > > > > #endif
> > > > > > > >
> > > > > > > > +#if defined(CONFIG_TARGET_SOCFPGA_GEN5)
> > > > > > > > static void socfpga_nic301_slave_ns(void)
> > > > > > > > {
> > > > > > > > writel(0x1, &nic301_regs->lwhps2fpgaregs);
> > > > > > > > @@ -85,6 +95,7 @@ void board_init_f(ulong dummy)
> > > > > > > > #endif
> > > > > > > > unsigned long sdram_size;
> > > > > > > > unsigned long reg;
> > > > > > > > + int ret;
> > > > > > > >
> > > > > > > > /*
> > > > > > > > * First C code to run. Clear fake OCRAM ECC first as SBE
> > > > > > > > @@ -117,7 +128,11 @@ void board_init_f(ulong dummy)
> > > > > > > > /* Put everything into reset but L4WD0. */
> > > > > > > > socfpga_per_reset_all();
> > > > > > > > /* Put FPGA bridges into reset too. */
> > > > > > > > - socfpga_bridges_reset(1);
> > > > > > > > + ret = socfpga_bridges_reset(1);
> > > > > > > > + if (ret) {
> > > > > > > > + printf("socfpga_bridges_reset() failed: %d\n",
> > > > > > > > ret);
> > > > > > > > + hang();
> > > > > > > > + }
> > > > > > > >
> > > > > > > > socfpga_per_reset(SOCFPGA_RESET(SDR), 0);
> > > > > > > > socfpga_per_reset(SOCFPGA_RESET(UART0), 0);
> > > > > > > > @@ -148,7 +163,11 @@ void board_init_f(ulong dummy)
> > > > > > > >
> > > > > > > > /* De-assert reset for peripherals and bridges based on
> > > > > > > > handoff
> > > > > > > > */
> > > > > > > > reset_deassert_peripherals_handoff();
> > > > > > > > - socfpga_bridges_reset(0);
> > > > > > > > + ret = socfpga_bridges_reset(0);
> > > > > > > > + if (ret) {
> > > > > > > > + printf("socfpga_bridges_reset() failed: %d\n",
> > > > > > > > ret);
> > > > > > > > + hang();
> > > > > > >
> > > > > > > If you keep this patch the way it is, this will cause the Atlas
> > > > > > > board to
> > > > > > > hang here.
> > > > > > >
> > > > > >
> > > > > > Hi Dalon,
> > > > > >
> > > > > > Can you check this patch? On the Atlas board, I'm seeing the call to
> > > > > > socfpga_bridges_reset(0) fail because fpgamgr_test_fpga_ready() is
> > > > > > failing, because is_fpgamgr_initdone_high() is returning 0.
> > > > > >
> > > > > > Dinh
> > > >
> > > > i saw it the first go round. i will test this out this afternoon.
> > > >
> > > > --dalon
> > >
> > > i see the exact same thing. for the c5/a5 devices there is no reason for
> > > init
> > > done to be high since fpga configuration isnt required to boot.
> >
> > Okay, good! It doesn't happen on the C5 devkit though.
> >
> > Ley Foon, did you test this on the Arria5?
> >
> > Dinh
>
> Odd, init done should not be high without the FPGA configured. i will try it
> on
> the c5 board as well to try and figure out why init done is high.
>
> --dalon
I ran this on the c5 socdk and noticed the same behavior as Dinh, then realized
the board was setup to configure the fpga independently of the processor. When
i change the MSELs on the board to ensure the FPGA is unconfigured the code
fails in the same location where init done is not high. It is definitely an
expected use case in c5/a5 that uboot can run without the fpga configured at
all.
--dalon
> _______________________________________________
> U-Boot mailing list
> U-Boot at lists.denx.de
> https://lists.denx.de/listinfo/u-boot
More information about the U-Boot
mailing list