On Tue, Jun 11, 2024 at 03:06:01PM +0200, Greg Kroah-Hartman wrote:
On Mon, Jun 10, 2024 at 08:26:10AM +0100, Conor Dooley wrote:
On Mon, Jun 10, 2024 at 08:28:29AM +0200, Greg Kroah-Hartman wrote:
On Sun, Jun 09, 2024 at 11:21:55PM -0700, Ron Economos wrote:
On 6/9/24 12:34 PM, Pavel Machek wrote:
Hi!
This is the start of the stable review cycle for the 6.6.33 release. There are 741 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.
6.6 seems to have build problem on risc-v:
arch/riscv/kernel/suspend.c:14:66: error: 'RISCV_ISA_EXT_XLINUXENVCFG' undeclared (first use in this function); did you mean 'RISCV_ISA_EXT_ZIFENCEI'? 694 14 | if (riscv_cpu_has_extension_unlikely(smp_processor_id(), RISCV_ISA_EXT_XLINUXENVCFG)) 695 | ^~~~~~~~~~~~~~~~~~~~~~~~~~ 696 | RISCV_ISA_EXT_ZIFENCEI
https://gitlab.com/cip-project/cip-testing/linux-stable-rc-ci/-/jobs/7053222... https://gitlab.com/cip-project/cip-testing/linux-stable-rc-ci/-/pipelines/13...
No problems detected on 6.8-stable and 6.1-stable.
Best regards, Pavel
I'm seeing the same thing here. Somehow some extra patches got slipped in between rc1 and rc2. The new patches for RISC-V are:
Samuel Holland samuel.holland@sifive.com riscv: Save/restore envcfg CSR during CPU suspend
commit 88b55a586b87994a33e0285c9e8881485e9b77ea
Samuel Holland samuel.holland@sifive.com riscv: Fix enabling cbo.zero when running in M-mode
commit 8c6e096cf527d65e693bfbf00aa6791149c58552
The first patch "riscv: Save/restore envcfg CSR during CPU suspend" causes the build failure.
Yes, these were added because they were marked as fixes for other commits in the series. I'll unwind them all now as something is going wrong...
Really we should just backport this envcfg handling to stable, this isn't the first (and won't be the last) issue it'll cause. I'll put a backport of it on my todo list cos I think last time around it couldn't be cherrypicked.
Thanks, I've dropped almost all riscv patches from this queue now. If they want to be added back, please send working backports :)
I went to take a look at this, but since 6.8 is now EOL, I dunno if I actually need to do anything here? These were needed because you had applied "RISC-V: Enable cbo.zero in usermode", but that's a feature, not a fix, so dropping that makes these changes unneeded. IIRC the previous time that there was an envcfg related build failure it was on the requested backport to 6.7+ in the envcfg addition in "riscv: Add a custom ISA extension for the [ms]envcfg CSR", and an assertion failed because of a definition for the maximum number of ISA extensions was larger in 6.9 than 6.7 and the patch depended on that.
For 6.6, I don't think envcfg is needed, unless there was some other reason that you backported "RISC-V: Enable cbo.zero in usermode".
Cheers, Conor.