From: Michael Opdenacker <michael.opdenacker@rootcommit.com>
To: Gyorgy Sarvari <skandigraun@gmail.com>
Cc: michael.opdenacker@rootcommit.com, Khem Raj <raj.khem@gmail.com>,
OE-core <openembedded-core@lists.openembedded.org>
Subject: Re: [OE-core] linux-yocto.inc: remove riscv-isa-clear.scc from KERNEL_FEATURES_RISCV?
Date: Fri, 28 Nov 2025 13:40:30 +0000 (UTC) [thread overview]
Message-ID: <4cf47893-8d84-4a4f-8301-a8dcaf9eb510@rootcommit.com> (raw)
In-Reply-To: <7e4c03b7-8f5c-4e92-91b2-1535e5533f8d@rootcommit.com>
On 11/27/25 12:03, Michael Opdenacker wrote:
> Hi Gyorgy
>
> On 11/27/25 11:46, Gyorgy Sarvari wrote:
>>
>> On 11/27/25 11:20, Michael Opdenacker via lists.openembedded.org wrote:
>>> Greetings,
>>>
>>> I'm getting the same issue trying a master build on several RISC-V
>>> boards:
>>>
>>> ERROR: linux-milkv-duo-dev-1.0-r0 do_kernel_metadata: Feature
>>> 'arch/riscv/tunes/riscv-isa-clear.scc' not found, this will cause
>>> configuration failures.
>>> ERROR: linux-milkv-duo-dev-1.0-r0 do_kernel_metadata: Check the SRC_URI
>>> for meta-data repositories or directories that may be missing
>>> ERROR: linux-milkv-duo-dev-1.0-r0 do_kernel_metadata: Set
>>> KERNEL_DANGLING_FEATURES_WARN_ONLY to ignore this issue
>>>
>>> OK, I can set
>>> KERNEL_DANGLING_FEATURES_WARN_ONLY = "1"
>>> but is this a clean solution?
>>>
>>> Where are all the arch/riscv/tunes/riscv-isa-*.scc files supposed to
>>> be?
>>> I didn't manage to find them anywhere...
>> Do you have the yocto-kernel-cache[1] repo in your SRC_URI? The scc
>> files come from there
>>
>> [1]:
>> https://git.yoctoproject.org/yocto-kernel-cache/tree/arch/riscv/tunes
>
> Ho, thanks! That was exactly what I needed.
> I will add these to my kernel recipes.
Interestingly, the Linux 6.18-rc7 kernel on my OrangePi RV2 board hates
the change.
It's not completely crashing, but many devices get initialized with the
same kind of error:
"device non-coherent but no non-coherent operations supported"
0.111641] ------------[ cut here ]------------
[ 0.111652] spacemit,k1-ccu d4090000.clock-controller: device
non-coherent but no non-coherent operations supported
[ 0.111659] WARNING: CPU: 0 PID: 1 at
/usr/src/kernel/arch/riscv/mm/dma-noncoherent.c:139
arch_setup_dma_ops+0x96/0xfc
[ 0.111673] Modules linked in:
[ 0.111682] CPU: 0 UID: 0 PID: 1 Comm: swapper/0 Tainted: G S 6.18.0-rc7
#1 NONE
[ 0.111690] Tainted: [S]=CPU_OUT_OF_SPEC
[ 0.111693] Hardware name: OrangePi RV2 (DT)
[ 0.111696] epc : arch_setup_dma_ops+0x96/0xfc
[ 0.111703] ra : arch_setup_dma_ops+0x96/0xfc
[ 0.111709] epc : ffffffff8001ea02 ra : ffffffff8001ea02 sp :
ffffffc600043b70
[ 0.111713] gp : ffffffff81a1ae08 tp : ffffffd700230000 t0 :
ffffffff81829520
[ 0.111717] t1 : ffffffff8181c228 t2 : 2d2d2d2d2d2d2d2d s0 :
ffffffc600043b90
[ 0.111720] s1 : ffffffd700b79c10 a0 : 0000000000000067 a1 :
00000000fffff264
[ 0.111724] a2 : 0000000000000000 a3 : 00000000fffff264 a4 :
0000000000000000
[ 0.111728] a5 : 0000000000000000 a6 : 0000000000000004 a7 :
0000000000000000
[ 0.111731] s2 : 0000000000000000 s3 : 0000000000000000 s4 :
0000000000000000
[ 0.111734] s5 : ffffffff80e30800 s6 : 0000000000000008 s7 :
0000000000000000
[ 0.111738] s8 : 0000000000000143 s9 : 0000000000000000 s10:
0000000000000000
[ 0.111742] s11: 0000000000000000 t3 : 0000000000000070 t4 :
00007fffffffffff
[ 0.111744] t5 : 00007fffffffffff t6 : ffffffff81a3de07
[ 0.111747] status: 0000000200000120 badaddr: 0000000000000000 cause:
0000000000000003
[ 0.111752] [<ffffffff8001ea02>] arch_setup_dma_ops+0x96/0xfc
[ 0.111761] [<ffffffff80918f30>] of_dma_configure_id+0x94/0x208
[ 0.111770] [<ffffffff80720050>] platform_dma_configure+0xa4/0xac
[ 0.111778] [<ffffffff8071d8cc>] really_probe+0x54/0x22c
[ 0.111785] [<ffffffff8071db00>] __driver_probe_device+0x5c/0xd4
[ 0.111791] [<ffffffff8071dc3e>] driver_probe_device+0x2e/0xf4
[ 0.111796] [<ffffffff8071de66>] __driver_attach+0x6e/0x14c
[ 0.111802] [<ffffffff8071bae0>] bus_for_each_dev+0x60/0xb0
[ 0.111807] [<ffffffff8071d432>] driver_attach+0x1a/0x24
[ 0.111813] [<ffffffff8071cc9a>] bus_add_driver+0xca/0x1d8
[ 0.111820] [<ffffffff8071ebea>] driver_register+0x3e/0xdc
[ 0.111827] [<ffffffff8071fde4>] __platform_driver_register+0x1c/0x24
[ 0.111834] [<ffffffff80c2ee0a>] k1_ccu_driver_init+0x1a/0x24
[ 0.111843] [<ffffffff8000f8aa>] do_one_initcall+0x4e/0x1b8
[ 0.111849] [<ffffffff80c012ac>] kernel_init_freeable+0x264/0x2e4
[ 0.111858] [<ffffffff80b761b8>] kernel_init+0x1c/0x144
[ 0.111862] [<ffffffff800111b6>] ret_from_fork_kernel+0xe/0x60
[ 0.111870] [<ffffffff80b8081e>] ret_from_fork_kernel_asm+0x16/0x18
[ 0.111877] ---[ end trace 0000000000000000 ]---
Is this expected?
Cheers
Michael.
--
Michael Opdenacker
Root Commit
Yocto Project and OpenEmbedded Training course - Learn by doing:
https://rootcommit.com/training/yocto/
prev parent reply other threads:[~2025-11-28 13:40 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-11-27 10:20 linux-yocto.inc: remove riscv-isa-clear.scc from KERNEL_FEATURES_RISCV? Michael Opdenacker
2025-11-27 10:46 ` [OE-core] " Gyorgy Sarvari
2025-11-27 11:03 ` Michael Opdenacker
2025-11-28 13:40 ` Michael Opdenacker [this message]
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=4cf47893-8d84-4a4f-8301-a8dcaf9eb510@rootcommit.com \
--to=michael.opdenacker@rootcommit.com \
--cc=openembedded-core@lists.openembedded.org \
--cc=raj.khem@gmail.com \
--cc=skandigraun@gmail.com \
/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