From: Oleksij Rempel <linux@rempel-privat.de>
To: Ahmad Fatoum <a.fatoum@pengutronix.de>,
Antony Pavlov <antonynpavlov@gmail.com>,
Oleksij Rempel <o.rempel@pengutronix.de>,
Michael Grzeschik <mgr@pengutronix.de>
Cc: barebox@lists.infradead.org
Subject: Re: MIPS RELOCATABLE: [PATCH 14/20] RISC-V: erizo: migrate to PBL
Date: Mon, 15 Mar 2021 18:03:55 +0100 [thread overview]
Message-ID: <2da18c76-a0f2-5ccc-f4bc-ff201be60ad3@rempel-privat.de> (raw)
In-Reply-To: <14b14f6b-a8ae-1b8b-c06a-73d909a4a9f7@pengutronix.de>
Am 15.03.21 um 16:46 schrieb Ahmad Fatoum:
> Hi,
>
> On 15.03.21 16:37, Antony Pavlov wrote:
>> On Mon, 15 Mar 2021 13:40:10 +0100
>> Ahmad Fatoum <a.fatoum@pengutronix.de> wrote:
>>> Without having dug into the MIPS code, the main point of CONFIG_RELOCATABLE
>>> is that PBL can check for it to decide where to copy barebox proper to.
>>>
>>> This is very relevant if there's no TEXT_BASE that works for all boards
>>> that you need to support with your multi-image barebox.
>>
>> So CONFIG_RELOCATABLE can be very helpful.
>>
>> At the moment MIPS barebox is always relocatable.
>> Sometimes it's useful to run barebox from fixed TEXT_BASE (e.g. during gdb debug session),
>> so this fixed TEXT_BASE behaviour can be introduced for CONFIG_RELOCATABLE=n case.
>
> With the pbl break stuff, you could in theory already flip a config option
> and have a gdb script that computes your load address from the address of
> your breakpoint. That would be nice to have as a ready-to-use script.
>
> I recall Oleksij and Michael did something in that regard.
>
> Benefit is that the break is an actual machine code nop when inactive,
> so binary size/layout shouldn't change, unlike with flipping CONFIG_RELOCATABLE.
> For RISC-V, I see no good reason to allow CONFIG_RELOCATABLE=n right now.
>
> Cheers,
> Ahmad
>
If i remember it correctly, this two patches are needed, to make
relocatable barebox debugable with openocd and gdb
https://git.pengutronix.de/cgit/barebox/commit/?id=c37cc2d725a496551113f4ebb1d116823eb0dbd1
https://git.pengutronix.de/cgit/barebox/commit/?id=d71573a006a2fc2b83c076c556b88166f4e3dec3
For the MIPS and RISC-V we would need similar helpers.
Regards,
Oleksij
--
Regards,
Oleksij
_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox
next prev parent reply other threads:[~2021-03-15 17:05 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-14 12:27 [PATCH 00/20] RISC-V: rework for PBL, VIRT and 64-Bit support Ahmad Fatoum
2021-03-14 12:27 ` [PATCH 01/20] partitions: don't allocate dma capable memory Ahmad Fatoum
2021-03-14 12:27 ` [PATCH 02/20] images: make BOARD_ARM_GENERIC_DT available for other arches Ahmad Fatoum
2021-03-14 12:27 ` [PATCH 03/20] ARM: rename CONFIG_ARM_USE_COMPRESSED_DTB to CONFIG_USE_COMPRESSED_DTB Ahmad Fatoum
2021-03-14 12:27 ` [PATCH 04/20] ARM: aarch64: ommit unused label in assembly Ahmad Fatoum
2021-03-14 12:27 ` [PATCH 05/20] serial: virtio-console: depend on, but don't select VIRTIO Ahmad Fatoum
2021-03-14 12:27 ` [PATCH 06/20] filetype: detect RISC-V Linux kernel image Ahmad Fatoum
2021-03-14 12:27 ` [PATCH 07/20] asm: unaligned: don't do unaligned accesses Ahmad Fatoum
2021-03-14 12:27 ` [PATCH 08/20] RISC-V: debug_ll: ns16550: align C access size with assembly's Ahmad Fatoum
2021-03-14 12:27 ` [PATCH 09/20] RISC-V: drop duplicate or unneeded cflags Ahmad Fatoum
2021-03-14 12:27 ` [PATCH 10/20] RISC-V: add cacheless HAS_DMA support Ahmad Fatoum
2021-03-14 12:27 ` [PATCH 11/20] RISC-V: erizo: move to arch/riscv/boards/erizo Ahmad Fatoum
2021-03-14 12:27 ` [PATCH 12/20] RISC-V: import Linux' optimized string functions Ahmad Fatoum
2021-03-14 12:27 ` [PATCH 13/20] RISC-V: implement PBL and relocation support Ahmad Fatoum
2021-03-15 7:43 ` Ahmad Fatoum
2021-03-14 12:27 ` [PATCH 14/20] RISC-V: erizo: migrate to PBL Ahmad Fatoum
2021-03-15 8:43 ` MIPS RELOCATABLE: " Antony Pavlov
2021-03-15 11:45 ` Oleksij Rempel
2021-03-15 12:08 ` Antony Pavlov
2021-03-15 12:40 ` Ahmad Fatoum
2021-03-15 15:37 ` Antony Pavlov
2021-03-15 15:46 ` Ahmad Fatoum
2021-03-15 17:03 ` Oleksij Rempel [this message]
2021-03-14 12:27 ` [PATCH 15/20] RISC-V: support symbol names in barebox image Ahmad Fatoum
2021-03-14 12:28 ` [PATCH 16/20] RISC-V: add 64-bit support Ahmad Fatoum
2021-03-14 12:51 ` Rouven Czerwinski
2021-03-14 13:03 ` [PATCH 1/2] fixup! " Ahmad Fatoum
2021-03-14 13:03 ` [PATCH 2/2] " Ahmad Fatoum
2021-03-15 8:38 ` Antony Pavlov
2021-03-14 12:28 ` [PATCH 17/20] RISC-V: add generic DT image Ahmad Fatoum
2021-03-14 12:28 ` [PATCH 18/20] clocksource: add driver for RISC-V CLINT timer Ahmad Fatoum
2021-03-14 12:28 ` [PATCH 19/20] power: reset: add drivers for generic syscon reset and poweroff Ahmad Fatoum
2021-03-14 12:28 ` [PATCH 20/20] RISC-V: add Qemu virt support Ahmad Fatoum
2021-03-14 16:30 ` [PATCH] fixup! " Ahmad Fatoum
2021-03-15 8:22 ` [PATCH 00/20] RISC-V: rework for PBL, VIRT and 64-Bit support Antony Pavlov
2021-03-15 10:30 ` 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=2da18c76-a0f2-5ccc-f4bc-ff201be60ad3@rempel-privat.de \
--to=linux@rempel-privat.de \
--cc=a.fatoum@pengutronix.de \
--cc=antonynpavlov@gmail.com \
--cc=barebox@lists.infradead.org \
--cc=mgr@pengutronix.de \
--cc=o.rempel@pengutronix.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