Re: [Qemu-devel] [PATCH for-2.9 24/30] aspeed: use first SPI flash as a boot ROM

["mar.krzeminski" <mar.krzeminski@gmail.com>]

Hi Cedric,

it looks like good idea for now to handle boot from flash.
As I understand you are trying to omit bootrom code in Qemu model?
This could lead you to some hacks in device models (eg SMC).

W dniu 29.11.2016 o 16:44, Cédric Le Goater pisze:
> Fill a ROM region with the flash content to support U-Boot. This is a
> little hacky but until we can boot from a MMIO region, it seems
> difficult to do anything else.
>
> Signed-off-by: Cédric Le Goater <clg@kaod.org>
> Reviewed-by: Joel Stanley <joel@jms.id.au>
> Reviewed-by: Andrew Jeffery <andrew@aj.id.au>
> ---
>   hw/arm/aspeed.c | 41 +++++++++++++++++++++++++++++++++++++++++
>   1 file changed, 41 insertions(+)
>
> diff --git a/hw/arm/aspeed.c b/hw/arm/aspeed.c
> index 40c13838fb2d..a92c2f1c362b 100644
> --- a/hw/arm/aspeed.c
> +++ b/hw/arm/aspeed.c
> @@ -20,6 +20,8 @@
>   #include "qemu/log.h"
>   #include "sysemu/block-backend.h"
>   #include "sysemu/blockdev.h"
> +#include "hw/loader.h"
> +#include "qemu/error-report.h"
>   
>   static struct arm_boot_info aspeed_board_binfo = {
>       .board_id = -1, /* device-tree-only board */
> @@ -104,6 +106,28 @@ static const AspeedBoardConfig aspeed_boards[] = {
>       },
>   };
>   
> +#define FIRMWARE_ADDR 0x0
> +
> +static void write_boot_rom(DriveInfo *dinfo, hwaddr addr, size_t rom_size,
> +                           Error **errp)
> +{
> +    BlockBackend *blk = blk_by_legacy_dinfo(dinfo);
> +    uint8_t *storage;
> +
> +    if (rom_size > blk_getlength(blk)) {
> +        rom_size = blk_getlength(blk);
> +    }
I was not able to attach smaller file as m25p80 storage.
> +
> +    storage = g_new0(uint8_t, rom_size);
> +    if (blk_pread(blk, 0, storage, rom_size) < 0) {
> +        error_setg(errp, "failed to read the initial flash content");
> +        return;
> +    }
> +
> +    rom_add_blob_fixed("aspeed.boot_rom", storage, rom_size, addr);
> +    g_free(storage);
> +}
> +
>   static void aspeed_board_init_flashes(AspeedSMCState *s, const char *flashtype,
>                                         Error **errp)
>   {
> @@ -135,6 +159,7 @@ static void aspeed_board_init(MachineState *machine,
>   {
>       AspeedBoardState *bmc;
>       AspeedSoCClass *sc;
> +    DriveInfo *drive0 = drive_get(IF_MTD, 0, 0);
>   
>       bmc = g_new0(AspeedBoardState, 1);
>       object_initialize(&bmc->soc, (sizeof(bmc->soc)), cfg->soc_name);
> @@ -168,6 +193,22 @@ static void aspeed_board_init(MachineState *machine,
>       aspeed_board_init_flashes(&bmc->soc.fmc, cfg->fmc_model, &error_abort);
>       aspeed_board_init_flashes(&bmc->soc.spi[0], cfg->spi_model, &error_abort);
>   
> +    /* Install first FMC flash content as a boot rom. */
> +    if (drive0) {
> +        AspeedSMCFlash *fl = &bmc->soc.fmc.flashes[0];
> +        MemoryRegion *boot_rom = g_new(MemoryRegion, 1);
> +
> +        /*
> +         * create a ROM region using the default mapping window size of
> +         * the flash module.
> +         */
> +        memory_region_init_rom(boot_rom, OBJECT(bmc), "aspeed.boot_rom",
> +                               fl->size, &error_abort);
> +        memory_region_add_subregion(get_system_memory(), FIRMWARE_ADDR,
> +                                    boot_rom);
> +        write_boot_rom(drive0, FIRMWARE_ADDR, fl->size, &error_abort);
Is it possible that fl->size will be bigger than segment size?
I think max_size here should be segment size in smc.

Thanks,
Marcin
> +    }
> +
>       aspeed_board_binfo.kernel_filename = machine->kernel_filename;
>       aspeed_board_binfo.initrd_filename = machine->initrd_filename;
>       aspeed_board_binfo.kernel_cmdline = machine->kernel_cmdline;