linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Yeoreum Yun <yeoreum.yun@arm.com>
To: akpm@linux-foundation.org, david@kernel.org,
	lorenzo.stoakes@oracle.com, Liam.Howlett@oracle.com,
	vbabka@suse.cz, rppt@kernel.org, surenb@google.com,
	mhocko@suse.com, ast@kernel.org, daniel@iogearbox.net,
	andrii@kernel.org, martin.lau@linux.dev, eddyz87@gmail.com,
	song@kernel.org, yonghong.song@linux.dev,
	john.fastabend@gmail.com, kpsingh@kernel.org, sdf@fomichev.me,
	haoluo@google.com, jolsa@kernel.org, jackmanb@google.com,
	hannes@cmpxchg.org, ziy@nvidia.com, bigeasy@linutronix.de,
	clrkwllms@kernel.org, rostedt@goodmis.org,
	catalin.marinas@arm.com, will@kernel.org, ryan.roberts@arm.com,
	kevin.brodsky@arm.com, dev.jain@arm.com,
	yang@os.amperecomputing.com
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
	bpf@vger.kernel.org, linux-rt-devel@lists.linux.dev,
	linux-arm-kernel@lists.infradead.org,
	Yeoreum Yun <yeoreum.yun@arm.com>
Subject: [PATCH 2/2] arm64: mmu: use pagetable_alloc_nolock() while stop_machine()
Date: Fri, 12 Dec 2025 16:18:32 +0000	[thread overview]
Message-ID: <20251212161832.2067134-3-yeoreum.yun@arm.com> (raw)
In-Reply-To: <20251212161832.2067134-1-yeoreum.yun@arm.com>

linear_map_split_to_ptes() and __kpti_install_ng_mappings()
are called as callback of stop_machine().
That means these functions context are preemption disabled.

Unfortunately, under PREEMPT_RT, the pagetable_alloc() or
__get_free_pages() couldn't be called in this context
since spin lock that becomes sleepable on RT,
potentially causing a sleep during page allocation.

To address this, pagetable_alloc_nolock().

Signed-off-by: Yeoreum Yun <yeoreum.yun@arm.com>
---
 arch/arm64/mm/mmu.c | 23 ++++++++++++++++++-----
 1 file changed, 18 insertions(+), 5 deletions(-)

diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c
index 2ba01dc8ef82..0e98606d8c4c 100644
--- a/arch/arm64/mm/mmu.c
+++ b/arch/arm64/mm/mmu.c
@@ -475,10 +475,15 @@ static void __create_pgd_mapping(pgd_t *pgdir, phys_addr_t phys,
 static phys_addr_t __pgd_pgtable_alloc(struct mm_struct *mm, gfp_t gfp,
 				       enum pgtable_type pgtable_type)
 {
-	/* Page is zeroed by init_clear_pgtable() so don't duplicate effort. */
-	struct ptdesc *ptdesc = pagetable_alloc(gfp & ~__GFP_ZERO, 0);
+	struct ptdesc *ptdesc;
 	phys_addr_t pa;
 
+	/* Page is zeroed by init_clear_pgtable() so don't duplicate effort. */
+	if (gfpflags_allow_spinning(gfp))
+		ptdesc  = pagetable_alloc(gfp & ~__GFP_ZERO, 0);
+	else
+		ptdesc  = pagetable_alloc_nolock(gfp & ~__GFP_ZERO, 0);
+
 	if (!ptdesc)
 		return INVALID_PHYS_ADDR;
 
@@ -869,6 +874,7 @@ static int __init linear_map_split_to_ptes(void *__unused)
 		unsigned long kstart = (unsigned long)lm_alias(_stext);
 		unsigned long kend = (unsigned long)lm_alias(__init_begin);
 		int ret;
+		gfp_t gfp = IS_ENABLED(CONFIG_PREEMPT_RT) ? __GFP_HIGH : GFP_ATOMIC;
 
 		/*
 		 * Wait for all secondary CPUs to be put into the waiting area.
@@ -881,9 +887,9 @@ static int __init linear_map_split_to_ptes(void *__unused)
 		 * PTE. The kernel alias remains static throughout runtime so
 		 * can continue to be safely mapped with large mappings.
 		 */
-		ret = range_split_to_ptes(lstart, kstart, GFP_ATOMIC);
+		ret = range_split_to_ptes(lstart, kstart, gfp);
 		if (!ret)
-			ret = range_split_to_ptes(kend, lend, GFP_ATOMIC);
+			ret = range_split_to_ptes(kend, lend, gfp);
 		if (ret)
 			panic("Failed to split linear map\n");
 		flush_tlb_kernel_range(lstart, lend);
@@ -1207,7 +1213,14 @@ static int __init __kpti_install_ng_mappings(void *__unused)
 	remap_fn = (void *)__pa_symbol(idmap_kpti_install_ng_mappings);
 
 	if (!cpu) {
-		alloc = __get_free_pages(GFP_ATOMIC | __GFP_ZERO, order);
+		if (IS_ENABLED(CONFIG_PREEMPT_RT))
+			alloc = (u64) pagetable_alloc_nolock(__GFP_HIGH | __GFP_ZERO, order);
+		else
+			alloc = __get_free_pages(GFP_ATOMIC | __GFP_ZERO, order);
+
+		if (!alloc)
+			panic("Failed to alloc kpti_ng_pgd\n");
+
 		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);
 
-- 
LEVI:{C3F47F37-75D8-414A-A8BA-3980EC8A46D7}



  parent reply	other threads:[~2025-12-12 16:18 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-12-12 16:18 [PATCH 0/2] introduce pagetable_alloc_nolock() Yeoreum Yun
2025-12-12 16:18 ` [PATCH 1/2] mm: " Yeoreum Yun
2025-12-12 16:18 ` Yeoreum Yun [this message]
2025-12-13  7:05   ` [PATCH 2/2] arm64: mmu: use pagetable_alloc_nolock() while stop_machine() Brendan Jackman
2025-12-14  9:13     ` Yeoreum Yun
2025-12-15  9:22       ` Brendan Jackman
2025-12-15  9:34         ` Yeoreum Yun
2025-12-15  9:55           ` Brendan Jackman
2025-12-15 10:06             ` Yeoreum Yun
2025-12-16 10:10               ` Brendan Jackman
2025-12-16 11:03                 ` Yeoreum Yun
2025-12-16 11:26                   ` Brendan Jackman
2025-12-16 12:01                     ` Yeoreum Yun
2025-12-16 12:39                       ` Brendan Jackman
2025-12-16 13:25                         ` Yeoreum Yun
2025-12-18  9:30   ` Michal Hocko
2025-12-18  9:36     ` Yeoreum Yun
2025-12-18 12:02       ` Ryan Roberts
2025-12-18 12:17         ` Michal Hocko
2025-12-18 12:24           ` Yeoreum Yun
2025-12-16 15:11 ` [PATCH 0/2] introduce pagetable_alloc_nolock() Ryan Roberts
2025-12-16 16:52   ` Yeoreum Yun
2025-12-17  9:34     ` Ryan Roberts
2025-12-17 10:48       ` Yeoreum Yun
2025-12-17 12:04         ` Ryan Roberts
2025-12-17 12:52           ` Yeoreum Yun
2025-12-17 13:15             ` Vlastimil Babka
2025-12-17 13:35               ` Brendan Jackman
2025-12-17 13:56                 ` Yeoreum Yun
2025-12-17 15:10                 ` Vlastimil Babka
2025-12-17 17:19                   ` Brendan Jackman
2025-12-18  7:47                     ` Vlastimil Babka
2025-12-18  7:52                   ` David Hildenbrand (Red Hat)
2025-12-23 22:59           ` Yang Shi
2025-12-24  7:00             ` 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=20251212161832.2067134-3-yeoreum.yun@arm.com \
    --to=yeoreum.yun@arm.com \
    --cc=Liam.Howlett@oracle.com \
    --cc=akpm@linux-foundation.org \
    --cc=andrii@kernel.org \
    --cc=ast@kernel.org \
    --cc=bigeasy@linutronix.de \
    --cc=bpf@vger.kernel.org \
    --cc=catalin.marinas@arm.com \
    --cc=clrkwllms@kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=david@kernel.org \
    --cc=dev.jain@arm.com \
    --cc=eddyz87@gmail.com \
    --cc=hannes@cmpxchg.org \
    --cc=haoluo@google.com \
    --cc=jackmanb@google.com \
    --cc=john.fastabend@gmail.com \
    --cc=jolsa@kernel.org \
    --cc=kevin.brodsky@arm.com \
    --cc=kpsingh@kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-rt-devel@lists.linux.dev \
    --cc=lorenzo.stoakes@oracle.com \
    --cc=martin.lau@linux.dev \
    --cc=mhocko@suse.com \
    --cc=rostedt@goodmis.org \
    --cc=rppt@kernel.org \
    --cc=ryan.roberts@arm.com \
    --cc=sdf@fomichev.me \
    --cc=song@kernel.org \
    --cc=surenb@google.com \
    --cc=vbabka@suse.cz \
    --cc=will@kernel.org \
    --cc=yang@os.amperecomputing.com \
    --cc=yonghong.song@linux.dev \
    --cc=ziy@nvidia.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