[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <aDMU2rjaFCyrFALd@alpha.franken.de>
Date: Sun, 25 May 2025 15:02:18 +0200
From: Thomas Bogendoerfer <tsbogend@...ha.franken.de>
To: Gregory CLEMENT <gregory.clement@...tlin.com>
Cc: Huacai Chen <chenhuacai@...nel.org>,
Jiaxun Yang <jiaxun.yang@...goat.com>,
Vladimir Kondratiev <vladimir.kondratiev@...ileye.com>,
Théo Lebrun <theo.lebrun@...tlin.com>,
Tawfik Bayouk <tawfik.bayouk@...ileye.com>,
Thomas Petazzoni <thomas.petazzoni@...tlin.com>,
linux-mips@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] MIPS: SMP: Move the AP sync point before the calibration
delay
On Fri, May 23, 2025 at 09:58:15AM +0200, Gregory CLEMENT wrote:
> In the calibration delay process, some resources are shared, so it's
> better to move it after the parallel execution part. Thanks to the
> patch optimizing CPU delay calibration, this change has no impact on
> the boot time improvements gained from CPU parallel boot.
>
> Signed-off-by: Gregory CLEMENT <gregory.clement@...tlin.com>
> ---
> Hello,
>
> After a thorough review, as reported in [1], the CPU delay calibration
> is the last potential issue area. However, I believe that with this
> patch [2] applied, the source of concurrency will disappear.
>
> Gregory
>
> [1]: https://lore.kernel.org/linux-mips/87frgvokga.fsf@BLaptop.bootlin.com/
> [2] :https://lore.kernel.org/linux-mips/20250520-smp_calib-v1-1-cd04f0a78648@bootlin.com/
> ---
> arch/mips/kernel/smp.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c
> index 7901b59d8f60eddefc020cf2a137716af963f09e..4868e79f3b30e9d80fe6390785b297c35d8c02a3 100644
> --- a/arch/mips/kernel/smp.c
> +++ b/arch/mips/kernel/smp.c
> @@ -371,12 +371,12 @@ asmlinkage void start_secondary(void)
> * to an option instead of something based on .cputype
> */
>
> - calibrate_delay();
> - cpu_data[cpu].udelay_val = loops_per_jiffy;
> -
> #ifdef CONFIG_HOTPLUG_PARALLEL
> cpuhp_ap_sync_alive();
> #endif
> + calibrate_delay();
> + cpu_data[cpu].udelay_val = loops_per_jiffy;
> +
> set_cpu_sibling_map(cpu);
> set_cpu_core_map(cpu);
>
>
> ---
> base-commit: faefb0a59c5914b7b8f737e2ec5c82822e5bc4c7
> change-id: 20250522-hotplug-paralell-fix2-1bffa02a9529
applied to mips-next
Thomas.
--
Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
good idea. [ RFC1925, 2.3 ]
Powered by blists - more mailing lists