From: "Clément Leger" <cleger@kalray.eu>
To: Ahmad Fatoum <a.fatoum@pengutronix.de>
Cc: Barebox List <barebox@lists.infradead.org>
Subject: Re: [PATCH 4/4] kvx: add support for elf loading using bootm
Date: Mon, 29 Jun 2020 10:23:26 +0200 (CEST) [thread overview]
Message-ID: <2007203678.240919.1593419006986.JavaMail.zimbra@kalray.eu> (raw)
In-Reply-To: <40a7f463-db1b-1241-13e5-a8a454393c75@pengutronix.de>
Hi Ahmad
----- On 29 Jun, 2020, at 10:15, Ahmad Fatoum a.fatoum@pengutronix.de wrote:
> Hello,
>
> On 6/24/20 9:15 AM, Clément Leger wrote:
>>> You already flushed out the dcache contents to the point of unification.
>>> What do you gain by invalidating it?
>>
>> Indeed, this is not strictly necessary, it just allow to boot the elf without a
>> dirty D-cache. But actually, the booted program should invalidate its D-cache
>> before fetching any data to avoid using cached data. I can remove it since
>> this is only for some buggy programs.
>
> Why would use of cached data hurt here? If you keep the MMU throughout, you
> should only need to invalidate the data cache when doing DMA. Am I missing
> something?
You are actually right, since barebox loaded the program by itself, the D-cache
contains the expected data even for the loaded program.
BTW, MMU is not enabled yet in barebox.
>
> Either way, I've no preference (except for naming it the same as on ARM,
> see previous mail).
Sorry, I missed the points you made on my previous patchset, my mailer marked
all the mail has read :/. I will make another more through review.
Regards,
Clément
>
> Cheers
> Ahmad
>
>>
>>>
>>>> +
>>>> + /**
>>>> + * Parameters passing
>>>> + * r0: boot magic
>>>> + * r1: device tree pointer
>>>> + */
>>>> + entry(LINUX_BOOT_PARAM_MAGIC, (void *) fdt_load_addr);
>>>> +
>>>> + /* should never return ! */
>>>> + panic("Returned from boot program !\n");
>>>> +
>>>> + return -EINVAL;
>>>> +}
>>>> +
>>>> +static int do_boot_elf(struct image_data *data, struct elf_image *elf)
>>>> +{
>>>> + int ret;
>>>> + void *fdt;
>>>> + boot_func_entry entry;
>>>> + unsigned long load_addr, initrd_address;
>>>> +
>>>> + /* load initrd after the elf */
>>>> + load_addr = PAGE_ALIGN((unsigned long) elf->high_addr);
>>>> + if (bootm_has_initrd(data)) {
>>>> + if (data->initrd_address != UIMAGE_INVALID_ADDRESS)
>>>> + initrd_address = data->initrd_address;
>>>> + else
>>>> + initrd_address = load_addr;
>>>> +
>>>> + printf("Loading initrd at 0x%lx\n", initrd_address);
>>>> + ret = bootm_load_initrd(data, initrd_address);
>>>> + if (ret) {
>>>> + printf("Failed to load initrd\n");
>>>> + return ret;
>>>> + }
>>>> +
>>>> + if (data->initrd_address == UIMAGE_INVALID_ADDRESS) {
>>>> + load_addr += resource_size(data->initrd_res);
>>>> + load_addr = PAGE_ALIGN(load_addr);
>>>> + }
>>>> + }
>>>> +
>>>> + fdt = bootm_get_devicetree(data);
>>>> + if (IS_ERR(fdt)) {
>>>> + printf("Failed to load dtb\n");
>>>> + return PTR_ERR(fdt);
>>>> + }
>>>> +
>>>> + printf("Loading device tree at %lx\n", load_addr);
>>>> + /* load device tree after the initrd if any */
>>>> + ret = bootm_load_devicetree(data, fdt, load_addr);
>>>> + if (ret) {
>>>> + printf("Failed to load device tree: %d\n", ret);
>>>> + goto err_free_fdt;
>>>> + }
>>>> +
>>>> + entry = (boot_func_entry) data->os_address;
>>>> +
>>>> + ret = do_boot_entry(data, entry, fdt);
>>>> +
>>>> +err_free_fdt:
>>>> + free(fdt);
>>>> +
>>>> + return ret;
>>>> +}
>>>> +
>>>> +static int do_bootm_elf(struct image_data *data)
>>>> +{
>>>> + int ret;
>>>> +
>>>> + ret = bootm_load_os(data, data->os_address);
>>>> + if (ret)
>>>> + return ret;
>>>> +
>>>> + return do_boot_elf(data, data->elf);
>>>> +}
>>>> +
>>>> +static struct image_handler elf_handler = {
>>>> + .name = "ELF",
>>>> + .bootm = do_bootm_elf,
>>>> + .filetype = filetype_elf,
>>>> +};
>>>> +
>>>> +static struct binfmt_hook binfmt_elf_hook = {
>>>> + .type = filetype_elf,
>>>> + .exec = "bootm",
>>>> +};
>>>> +
>>>> +static int kvx_register_image_handler(void)
>>>> +{
>>>> + register_image_handler(&elf_handler);
>>>> +
>>>> + binfmt_register(&binfmt_elf_hook);
>>>> +
>>>> + return 0;
>>>> +}
>>>> +
>>>> +late_initcall(kvx_register_image_handler);
>>>>
>>>
>>> --
>>> Pengutronix e.K. | |
>>> Steuerwalder Str. 21 | http://www.pengutronix.de/ |
>>> 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
>>> Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
>>
>
> --
> Pengutronix e.K. | |
> Steuerwalder Str. 21 | http://www.pengutronix.de/ |
> 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
> Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox
next prev parent reply other threads:[~2020-06-29 8:23 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-06-23 19:35 [PATCH 0/4] kvx: add elf bootm support Clement Leger
2020-06-23 19:35 ` [PATCH 1/4] common: bootm: allow letting IH_ARCH undefined Clement Leger
2020-06-23 19:35 ` [PATCH 2/4] common: Kconfig: remove MIPS dependency Clement Leger
2020-06-24 6:06 ` Ahmad Fatoum
2020-06-23 19:35 ` [PATCH 3/4] kvx: add D-cache inval and I-cache sync Clement Leger
2020-06-24 6:12 ` Ahmad Fatoum
2020-06-23 19:35 ` [PATCH 4/4] kvx: add support for elf loading using bootm Clement Leger
2020-06-24 6:17 ` Ahmad Fatoum
2020-06-24 7:15 ` Clément Leger
2020-06-29 8:15 ` Ahmad Fatoum
2020-06-29 8:23 ` Clément Leger [this message]
2020-06-29 8:29 ` Ahmad Fatoum
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=2007203678.240919.1593419006986.JavaMail.zimbra@kalray.eu \
--to=cleger@kalray.eu \
--cc=a.fatoum@pengutronix.de \
--cc=barebox@lists.infradead.org \
/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