Linux-mm Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Ard Biesheuvel <ardb+git@google.com>
To: linux-kernel@vger.kernel.org
Cc: "Ard Biesheuvel" <ardb@kernel.org>,
	"Catalin Marinas" <catalin.marinas@arm.com>,
	"Will Deacon" <will@kernel.org>,
	"Steven Rostedt" <rostedt@goodmis.org>,
	"Masami Hiramatsu" <mhiramat@kernel.org>,
	"Mark Rutland" <mark.rutland@arm.com>,
	"Andrew Morton" <akpm@linux-foundation.org>,
	"Mike Rapoport" <rppt@kernel.org>,
	"Luis Chamberlain" <mcgrof@kernel.org>,
	"Petr Pavlu" <petr.pavlu@suse.com>,
	"Daniel Gomez" <da.gomez@kernel.org>,
	"Sami Tolvanen" <samitolvanen@google.com>,
	"Aaron Tomlin" <atomlin@atomlin.com>,
	"Adrian Barnaś" <abarnas@google.com>,
	"Ryan Roberts" <ryan.roberts@arm.com>,
	"Kevin Brodsky" <kevin.brodsky@arm.com>,
	linux-arm-kernel@lists.infradead.org,
	linux-trace-kernel@vger.kernel.org, linux-mm@kvack.org,
	linux-modules@vger.kernel.org
Subject: [RFC PATCH 4/9] module: Allocate MOD_INIT_TEXT from the MOD_TEXT ROX allocation
Date: Sat, 22 Aug 2026 15:53:26 +0200	[thread overview]
Message-ID: <20260822135323.795946-15-ardb+git@google.com> (raw)
In-Reply-To: <20260822135323.795946-11-ardb+git@google.com>

From: Ard Biesheuvel <ardb@kernel.org>

When execmem ROX caches are used for module text and inittext, place
them adjacently in memory, by allocating space for both initially, and
splitting off the space for MOD_INIT_TEXT as needed.

This avoids the corner case on arm64, where .init.text being placed far
from .text results in a lot of complexity wrt indirect branches and PLTs
that we'd prefer to avoid.

Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
---
 kernel/module/main.c | 22 ++++++++++++++++----
 1 file changed, 18 insertions(+), 4 deletions(-)

diff --git a/kernel/module/main.c b/kernel/module/main.c
index 46dd8d25a605..2d6213496359 100644
--- a/kernel/module/main.c
+++ b/kernel/module/main.c
@@ -1342,7 +1342,7 @@ static int module_memory_alloc(struct module *mod, enum mod_mem_type type)
 {
 	unsigned int size = PAGE_ALIGN(mod->mem[type].size);
 	enum execmem_type execmem_type;
-	void *ptr;
+	void *ptr = NULL;
 
 	mod->mem[type].size = size;
 
@@ -1351,11 +1351,25 @@ static int module_memory_alloc(struct module *mod, enum mod_mem_type type)
 	else
 		execmem_type = EXECMEM_MODULE_TEXT;
 
-	ptr = execmem_alloc_rw(execmem_type, size);
+	bool is_rox = execmem_is_rox(execmem_type);
+	if (is_rox) {
+		/*
+		 * Special case for MOD_TEXT / MOD_INIT_TEXT: allocate the
+		 * latter by splitting off required space from the former
+		 * so that they are always placed close together.
+		 */
+		if (type == MOD_TEXT)
+			size += PAGE_ALIGN(mod->mem[MOD_INIT_TEXT].size);
+		else if (type == MOD_INIT_TEXT)
+			ptr = execmem_split(mod->mem[MOD_TEXT].base, size);
+	}
+
+	if (!ptr)
+		ptr = execmem_alloc_rw(execmem_type, size);
 	if (!ptr)
 		return -ENOMEM;
 
-	mod->mem[type].is_rox = execmem_is_rox(execmem_type);
+	mod->mem[type].is_rox = is_rox;
 
 	/*
 	 * The pointer to these blocks of memory are stored on the module
@@ -1368,7 +1382,7 @@ static int module_memory_alloc(struct module *mod, enum mod_mem_type type)
 	 * *do* eventually get freed, but let's just keep things simple
 	 * and avoid *any* false positives.
 	 */
-	if (!mod->mem[type].is_rox)
+	if (!is_rox)
 		kmemleak_not_leak(ptr);
 
 	memset(ptr, 0, size);
-- 
2.55.0.860.g4b6b3295ed-goog



  parent reply	other threads:[~2026-08-22 13:54 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-08-22 13:53 [RFC PATCH 0/9] arm64: Allocate .text and .init.text together Ard Biesheuvel
2026-08-22 13:53 ` [RFC PATCH 1/9] mm: execmem: Add API to split an existing execmem cache allocation Ard Biesheuvel
2026-08-22 13:53 ` [RFC PATCH 2/9] mm: execmem: Allow huge vmappings to be avoided for execmem caches Ard Biesheuvel
2026-08-22 13:53 ` [RFC PATCH 3/9] module: Place MOD_TEXT before MOD_INIT_TEXT in enumeration Ard Biesheuvel
2026-08-22 13:53 ` Ard Biesheuvel [this message]
2026-08-22 13:53 ` [RFC PATCH 5/9] arm64: mm: Permit permissions changes on huge vmappings Ard Biesheuvel
2026-08-23 16:52   ` Adrian Barnaś
2026-08-22 13:53 ` [RFC PATCH 6/9] arm64: Enable the execmem ROX cache for module text Ard Biesheuvel
2026-08-23 16:46   ` Adrian Barnaś
2026-08-22 13:53 ` [RFC PATCH 7/9] arm64: ftrace: Revert "fix unreachable PLT for ftrace_caller ..." Ard Biesheuvel
2026-08-22 13:53 ` [RFC PATCH 8/9] arm64: module: Combine init and core PLT entries again Ard Biesheuvel
2026-08-22 13:53 ` [RFC PATCH 9/9] arm64: ftrace: Simplify PLT handling Ard Biesheuvel

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=20260822135323.795946-15-ardb+git@google.com \
    --to=ardb+git@google.com \
    --cc=abarnas@google.com \
    --cc=akpm@linux-foundation.org \
    --cc=ardb@kernel.org \
    --cc=atomlin@atomlin.com \
    --cc=catalin.marinas@arm.com \
    --cc=da.gomez@kernel.org \
    --cc=kevin.brodsky@arm.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-modules@vger.kernel.org \
    --cc=linux-trace-kernel@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=mcgrof@kernel.org \
    --cc=mhiramat@kernel.org \
    --cc=petr.pavlu@suse.com \
    --cc=rostedt@goodmis.org \
    --cc=rppt@kernel.org \
    --cc=ryan.roberts@arm.com \
    --cc=samitolvanen@google.com \
    --cc=will@kernel.org \
    /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