From: AKASHI Takahiro <takahiro.akashi@linaro.org>
To: u-boot@lists.denx.de
Subject: [U-Boot] [PATCH v6 7/7] cmd: efidebug: add memmap command
Date: Wed, 20 Feb 2019 09:53:25 +0900 [thread overview]
Message-ID: <20190220005323.GD20286@linaro.org> (raw)
In-Reply-To: <ccd3b8a4-700c-d361-f944-477dcbadef58@gmx.de>
On Tue, Feb 19, 2019 at 08:11:05PM +0100, Heinrich Schuchardt wrote:
> On 1/24/19 12:04 PM, AKASHI Takahiro wrote:
> > "memmap" command prints uefi-specific memory map information.
> > => efi memmap
> > Type Start End Attributes
> > ================ ================ ================ ==========
> > CONVENTIONAL 0000000040000000-000000007de27000 WB
> > RUNTIME DATA 000000007de27000-000000007de28000 WB|RT
> > RESERVED 000000007de28000-000000007de2a000 WB
> > RUNTIME DATA 000000007de2a000-000000007de2b000 WB|RT
> > RESERVED 000000007de2b000-000000007de2c000 WB
> > RUNTIME DATA 000000007de2c000-000000007de2d000 WB|RT
> > LOADER DATA 000000007de2d000-000000007ff37000 WB
> > RUNTIME CODE 000000007ff37000-000000007ff38000 WB|RT
> > LOADER DATA 000000007ff38000-0000000080000000 WB
>
> On qemu_arm_defconfig I got this ouptut:
>
> => efidebug memmap
> Type Start End Attributes
> ================ ======== ======== ==========
> CONVENTIONAL 0000000040000000-000000007ddf8000 WB
> BOOT DATA 000000007ddf8000-000000007ddfd000 WB
> RUNTIME DATA 000000007ddfd000-000000007ddfe000 WB|RT
> RESERVED 000000007ddfe000-000000007ddff000 WB
> RUNTIME DATA 000000007ddff000-000000007de00000 WB|RT
> LOADER DATA 000000007de00000-000000007ff42000 WB
> RUNTIME CODE 000000007ff42000-000000007ff43000 WB|RT
> LOADER DATA 000000007ff43000-0000000080000000 WB
>
> Something is wrong with you format codes.
>
> >
> > Signed-off-by: AKASHI Takahiro <takahiro.akashi@linaro.org>
> > ---
> > cmd/efidebug.c | 100 ++++++++++++++++++++++++++++++++++++++++++++++++-
> > 1 file changed, 99 insertions(+), 1 deletion(-)
> >
> > diff --git a/cmd/efidebug.c b/cmd/efidebug.c
> > index 81ab3654f746..39398669e18f 100644
> > --- a/cmd/efidebug.c
> > +++ b/cmd/efidebug.c
> > @@ -309,6 +309,100 @@ static int do_efi_show_images(cmd_tbl_t *cmdtp, int flag,
> > return CMD_RET_SUCCESS;
> > }
> >
> > +static const char * const efi_mem_type_string[] = {
> > + [EFI_RESERVED_MEMORY_TYPE] = "RESERVED",
> > + [EFI_LOADER_CODE] = "LOADER CODE",
> > + [EFI_LOADER_DATA] = "LOADER DATA",
> > + [EFI_BOOT_SERVICES_CODE] = "BOOT CODE",
> > + [EFI_BOOT_SERVICES_DATA] = "BOOT DATA",
> > + [EFI_RUNTIME_SERVICES_CODE] = "RUNTIME CODE",
> > + [EFI_RUNTIME_SERVICES_DATA] = "RUNTIME DATA",
> > + [EFI_CONVENTIONAL_MEMORY] = "CONVENTIONAL",
> > + [EFI_UNUSABLE_MEMORY] = "UNUSABLE MEM",
> > + [EFI_ACPI_RECLAIM_MEMORY] = "ACPI RECLAIM MEM",
> > + [EFI_ACPI_MEMORY_NVS] = "ACPI NVS",
> > + [EFI_MMAP_IO] = "IO",
> > + [EFI_MMAP_IO_PORT] = "IO PORT",
> > + [EFI_PAL_CODE] = "PAL",
> > +};
> > +
> > +static const struct efi_mem_attrs {
> > + const u64 bit;
> > + const char *text;
> > +} efi_mem_attrs[] = {
> > + {EFI_MEMORY_UC, "UC"},
> > + {EFI_MEMORY_UC, "UC"},
> > + {EFI_MEMORY_WC, "WC"},
> > + {EFI_MEMORY_WT, "WT"},
> > + {EFI_MEMORY_WB, "WB"},
> > + {EFI_MEMORY_UCE, "UCE"},
> > + {EFI_MEMORY_WP, "WP"},
> > + {EFI_MEMORY_RP, "RP"},
> > + {EFI_MEMORY_XP, "WP"},
> > + {EFI_MEMORY_NV, "NV"},
> > + {EFI_MEMORY_MORE_RELIABLE, "REL"},
> > + {EFI_MEMORY_RO, "RO"},
> > + {EFI_MEMORY_RUNTIME, "RT"},
> > +};
> > +
> > +static void print_memory_attributes(u64 attributes)
> > +{
> > + int sep, i;
> > +
> > + for (sep = 0, i = 0; i < ARRAY_SIZE(efi_mem_attrs); i++)
> > + if (attributes & efi_mem_attrs[i].bit) {
> > + if (sep) {
> > + putc('|');
> > + } else {
> > + putc(' ');
> > + sep = 1;
> > + }
> > + puts(efi_mem_attrs[i].text);
> > + }
> > +}
> > +
> > +static int do_efi_show_memmap(cmd_tbl_t *cmdtp, int flag,
> > + int argc, char * const argv[])
> > +{
> > + struct efi_mem_desc *memmap = NULL, *map;
> > + efi_uintn_t map_size = 0;
> > + const char *type;
> > + int i;
> > + efi_status_t ret;
> > +
> > + ret = efi_get_memory_map(&map_size, memmap, NULL, NULL, NULL);
> > + if (ret == EFI_BUFFER_TOO_SMALL) {
> > + memmap = malloc(map_size);
> > + if (!memmap)
> > + return CMD_RET_FAILURE;
> > + ret = efi_get_memory_map(&map_size, memmap, NULL, NULL, NULL);
> > + }
> > + if (ret != EFI_SUCCESS) {
> > + free(memmap);
> > + return CMD_RET_FAILURE;
> > + }
> > +
> > + printf("Type Start%.*s End%.*s Attributes\n",
> > + EFI_HANDLE_WIDTH - 5, spc, EFI_HANDLE_WIDTH - 3, spc);
> > + printf("================ %.*s %.*s ==========\n",
> > + EFI_HANDLE_WIDTH, sep, EFI_HANDLE_WIDTH, sep);
> > + for (i = 0, map = memmap; i < map_size / sizeof(*map); map++, i++) {
> > + if (map->type < EFI_MAX_MEMORY_TYPE)
> > + type = efi_mem_type_string[map->type];
> > + else
> > + type = "(unknown)";
> > + printf("%-16s %016llx-%016llx", type, map->physical_start,
>
> Width 16 is ok on 64bit systems but not on 32bit ones.
It sounds reasonable, but the reality is not so trivial.
In struct efi_mem_desc, physical_start is defined as efi_physical_addr_t,
and efi_physical_addr_t is defined as u64 whatever the arch is.
So how do we know the system has 64-bit address space?
There is a config, CONFIG_PHYS_64BIT, but it is never defined on x86(_64).
So workable but ugly solution to meet your requirement would be
a)
if (sizeof(phys_addr_t) >= 8)
printf("%-16s %016llx-%016llx", ...);
else
printf("%-16s %08llx-%08llx", ...);
or
b)
printf("%-16s %p-%p", type, (void *)map->physical_start, ...);
(I don't think "void *" always reflects *physical* bit width
on LPAE arch though.)
Which do you like better?
Thanks,
-Takahiro Akashi
> Best regards
>
> Heinrich
>
> > + map->physical_start + map->num_pages * EFI_PAGE_SIZE);
> > +
> > + print_memory_attributes(map->attribute);
> > + putc('\n');
> > + }
> > +
> > + free(memmap);
> > +
> > + return CMD_RET_SUCCESS;
> > +}
> > +
> > static int do_efi_boot_add(cmd_tbl_t *cmdtp, int flag,
> > int argc, char * const argv[])
> > {
> > @@ -708,6 +802,8 @@ static cmd_tbl_t cmd_efidebug_sub[] = {
> > "", ""),
> > U_BOOT_CMD_MKENT(images, CONFIG_SYS_MAXARGS, 1, do_efi_show_images,
> > "", ""),
> > + U_BOOT_CMD_MKENT(memmap, CONFIG_SYS_MAXARGS, 1, do_efi_show_memmap,
> > + "", ""),
> > };
> >
> > /* Interpreter command to configure UEFI environment */
> > @@ -761,7 +857,9 @@ static char efidebug_help_text[] =
> > "efidebug dh\n"
> > " - show uefi handles\n"
> > "efidebug images\n"
> > - " - show loaded images\n";
> > + " - show loaded images\n"
> > + "efidebug memmap\n"
> > + " - show uefi memory map\n";
> > #endif
> >
> > U_BOOT_CMD(
> >
>
next prev parent reply other threads:[~2019-02-20 0:53 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-24 11:04 [U-Boot] [PATCH v6 0/7] cmd: add efidebug for efi environment AKASHI Takahiro
2019-01-24 11:04 ` [U-Boot] [PATCH v6 1/7] cmd: env: add "-e" option for handling UEFI variables AKASHI Takahiro
2019-01-25 12:42 ` Alexander Graf
2019-01-29 3:22 ` AKASHI Takahiro
2019-01-24 11:04 ` [U-Boot] [PATCH v6 2/7] cmd: add efidebug command AKASHI Takahiro
2019-02-19 19:32 ` Heinrich Schuchardt
2019-02-20 4:58 ` AKASHI Takahiro
2019-02-20 6:40 ` Heinrich Schuchardt
2019-01-24 11:04 ` [U-Boot] [PATCH v6 3/7] cmd: efidebug: add devices command AKASHI Takahiro
2019-02-19 19:46 ` Heinrich Schuchardt
2019-02-19 19:53 ` Heinrich Schuchardt
2019-02-20 2:20 ` AKASHI Takahiro
2019-02-20 6:49 ` Heinrich Schuchardt
2019-02-20 7:42 ` AKASHI Takahiro
2019-01-24 11:04 ` [U-Boot] [PATCH v6 4/7] cmd: efidebug: add drivers command AKASHI Takahiro
2019-02-19 19:49 ` Heinrich Schuchardt
2019-02-20 2:22 ` AKASHI Takahiro
2019-01-24 11:04 ` [U-Boot] [PATCH v6 5/7] cmd: efidebug: add dh command AKASHI Takahiro
2019-01-24 11:04 ` [U-Boot] [PATCH v6 6/7] cmd: efidebug: add images command AKASHI Takahiro
2019-01-29 15:35 ` Alexander Graf
2019-01-30 0:00 ` AKASHI Takahiro
2019-02-19 19:51 ` Heinrich Schuchardt
2019-01-24 11:04 ` [U-Boot] [PATCH v6 7/7] cmd: efidebug: add memmap command AKASHI Takahiro
2019-02-19 19:11 ` Heinrich Schuchardt
2019-02-20 0:53 ` AKASHI Takahiro [this message]
2019-02-20 6:53 ` Heinrich Schuchardt
2019-02-20 7:45 ` AKASHI Takahiro
2019-01-25 11:56 ` [U-Boot] [PATCH v6 0/7] cmd: add efidebug for efi environment Alexander Graf
2019-01-29 3:07 ` AKASHI Takahiro
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=20190220005323.GD20286@linaro.org \
--to=takahiro.akashi@linaro.org \
--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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.