mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Will Deacon <will@kernel.org>
To: Yeoreum Yun <yeoreum.yun@arm.com>
Cc: linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org, linux-rt-devel@lists.linux.dev,
	catalin.marinas@arm.com, ryan.roberts@arm.com,
	akpm@linux-oundation.org, david@kernel.org,
	kevin.brodsky@arm.com, quic_zhenhuah@quicinc.com,
	dev.jain@arm.com, yang@os.amperecomputing.com,
	chaitanyas.prakash@arm.com, bigeasy@linutronix.de,
	clrkwllms@kernel.org, rostedt@goodmis.org,
	lorenzo.stoakes@oracle.com, ardb@kernel.org, jackmanb@google.com,
	vbabka@suse.cz, mhocko@suse.com
Subject: Re: [PATCH v5 3/3] arm64: mmu: avoid allocating pages while installing ng-mapping for KPTI
Date: Mon, 19 Jan 2026 17:31:42 +0000	[thread overview]
Message-ID: <aW5qfh_bsGNRd0Vd@willie-the-truck> (raw)
In-Reply-To: <20260105202328.2418990-4-yeoreum.yun@arm.com>

On Mon, Jan 05, 2026 at 08:23:28PM +0000, Yeoreum Yun wrote:
> The current __kpti_install_ng_mappings() allocates a temporary PGD
> while installing the NG mapping for KPTI under stop_machine(),
> using GFP_ATOMIC.
> 
> This is fine in the non-PREEMPT_RT case. However, it becomes a problem
> under PREEMPT_RT because generic memory allocation/free APIs
> (e.g., pgtable_alloc(), __get_free_pages(), etc.) cannot be invoked
> in a non-preemptible context, except for the *_nolock() variants.
> These generic allocators may sleep due to their use of spin_lock().
> 
> In other words, calling __get_free_pages(), even with GFP_ATOMIC,
> is not allowed in __kpti_install_ng_mappings(), which is executed by
> the stopper thread where preemption is disabled under PREEMPT_RT.
> 
> To address this, preallocate the page needed for the temporary PGD
> before invoking __kpti_install_ng_mappings() via stop_machine().
> 
> Fixes: 47546a1912fc ("arm64: mm: install KPTI nG mappings with MMU enabled")
> Signed-off-by: Yeoreum Yun <yeoreum.yun@arm.com>
> Reviewed-by: Ryan Roberts <ryan.roberts@arm.com>
> ---
>  arch/arm64/mm/mmu.c | 21 ++++++++++++---------
>  1 file changed, 12 insertions(+), 9 deletions(-)
> 
> diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c
> index 120874a2d35b..6ea5b80ab54f 100644
> --- a/arch/arm64/mm/mmu.c
> +++ b/arch/arm64/mm/mmu.c
> @@ -1360,7 +1360,7 @@ static phys_addr_t __init kpti_ng_pgd_alloc(enum pgtable_type type)
>  	return kpti_ng_temp_alloc;
>  }
>  
> -static int __init __kpti_install_ng_mappings(void *__unused)
> +static int __init __kpti_install_ng_mappings(void *data)
>  {
>  	typedef void (kpti_remap_fn)(int, int, phys_addr_t, unsigned long);
>  	extern kpti_remap_fn idmap_kpti_install_ng_mappings;
> @@ -1368,10 +1368,9 @@ static int __init __kpti_install_ng_mappings(void *__unused)
>  
>  	int cpu = smp_processor_id();
>  	int levels = CONFIG_PGTABLE_LEVELS;
> -	int order = order_base_2(levels);
>  	u64 kpti_ng_temp_pgd_pa = 0;
>  	pgd_t *kpti_ng_temp_pgd;
> -	u64 alloc = 0;
> +	u64 alloc = *(u64 *)data;
>  
>  	if (levels == 5 && !pgtable_l5_enabled())
>  		levels = 4;
> @@ -1382,8 +1381,6 @@ static int __init __kpti_install_ng_mappings(void *__unused)
>  
>  	if (!cpu) {
>  		int ret;
> -
> -		alloc = __get_free_pages(GFP_ATOMIC | __GFP_ZERO, order);
>  		kpti_ng_temp_pgd = (pgd_t *)(alloc + (levels - 1) * PAGE_SIZE);
>  		kpti_ng_temp_alloc = kpti_ng_temp_pgd_pa = __pa(kpti_ng_temp_pgd);
>  
> @@ -1414,16 +1411,16 @@ static int __init __kpti_install_ng_mappings(void *__unused)
>  	remap_fn(cpu, num_online_cpus(), kpti_ng_temp_pgd_pa, KPTI_NG_TEMP_VA);
>  	cpu_uninstall_idmap();
>  
> -	if (!cpu) {
> -		free_pages(alloc, order);
> +	if (!cpu)
>  		arm64_use_ng_mappings = true;
> -	}
>  
>  	return 0;
>  }
>  
>  void __init kpti_install_ng_mappings(void)
>  {
> +	int order = order_base_2(CONFIG_PGTABLE_LEVELS);
> +	u64 alloc;
>  	/* Check whether KPTI is going to be used */
>  	if (!arm64_kernel_unmapped_at_el0())
>  		return;
> @@ -1436,8 +1433,14 @@ void __init kpti_install_ng_mappings(void)
>  	if (arm64_use_ng_mappings)
>  		return;
>  
> +	alloc = __get_free_pages(GFP_KERNEL | __GFP_ZERO, order);
> +	if (!alloc)
> +		panic("Failed to alloc page tables\n");

Why are you adding this panic?

Will

  reply	other threads:[~2026-01-19 17:31 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-01-05 20:23 [PATCH v5 0/3] fix wrong usage of memory allocation APIs under PREEMPT_RT in arm64 Yeoreum Yun
2026-01-05 20:23 ` [PATCH v5 1/3] arm64: mmu: introduce pgtable_alloc_t Yeoreum Yun
2026-01-05 20:23 ` [PATCH v5 2/3] arm64: mmu: avoid allocating pages while splitting the linear mapping Yeoreum Yun
2026-01-19 17:28   ` Will Deacon
2026-01-19 21:24     ` Yeoreum Yun
2026-01-20  8:56       ` Ryan Roberts
2026-01-20  9:29         ` Yeoreum Yun
2026-01-20 10:40           ` Ryan Roberts
2026-01-20 10:54             ` Yeoreum Yun
2026-01-20 15:53             ` Will Deacon
2026-01-20 16:16               ` Yeoreum Yun
2026-01-20 16:22               ` Ryan Roberts
2026-01-20 16:31                 ` Yeoreum Yun
2026-01-20 17:35                   ` Ryan Roberts
2026-01-20 17:49                     ` Yeoreum Yun
2026-01-21  0:12                     ` Yang Shi
2026-01-21  8:32                       ` Yeoreum Yun
2026-01-21 10:20                         ` Ryan Roberts
2026-01-21 11:30                           ` Yeoreum Yun
2026-01-21 22:57                           ` Yang Shi
2026-01-22  7:42                             ` Yeoreum Yun
2026-01-22 13:47                               ` Ryan Roberts
2026-01-20 22:24           ` Yang Shi
2026-01-20 23:01             ` Yeoreum Yun
2026-01-21  0:43               ` Yang Shi
2026-01-21  8:15                 ` Yeoreum Yun
2026-01-05 20:23 ` [PATCH v5 3/3] arm64: mmu: avoid allocating pages while installing ng-mapping for KPTI Yeoreum Yun
2026-01-19 17:31   ` Will Deacon [this message]
2026-01-19 21:30     ` Yeoreum Yun
2026-01-20 11:44       ` Will Deacon
2026-01-20 15:30         ` Yeoreum Yun

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=aW5qfh_bsGNRd0Vd@willie-the-truck \
    --to=will@kernel.org \
    --cc=akpm@linux-oundation.org \
    --cc=ardb@kernel.org \
    --cc=bigeasy@linutronix.de \
    --cc=catalin.marinas@arm.com \
    --cc=chaitanyas.prakash@arm.com \
    --cc=clrkwllms@kernel.org \
    --cc=david@kernel.org \
    --cc=dev.jain@arm.com \
    --cc=jackmanb@google.com \
    --cc=kevin.brodsky@arm.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-rt-devel@lists.linux.dev \
    --cc=lorenzo.stoakes@oracle.com \
    --cc=mhocko@suse.com \
    --cc=quic_zhenhuah@quicinc.com \
    --cc=rostedt@goodmis.org \
    --cc=ryan.roberts@arm.com \
    --cc=vbabka@suse.cz \
    --cc=yang@os.amperecomputing.com \
    --cc=yeoreum.yun@arm.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

all inboxes | Powered by JetHome®