linux-kernel - [PATCH 1/2] uprobes: kill xol_area->slot

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20241001142458.GA13629@redhat.com>
Date: Tue, 1 Oct 2024 16:24:59 +0200
From: Oleg Nesterov <oleg@...hat.com>
To: Andrii Nakryiko <andrii@...nel.org>, Jiri Olsa <jolsa@...nel.org>,
	Masami Hiramatsu <mhiramat@...nel.org>,
	Peter Zijlstra <peterz@...radead.org>
Cc: Liao Chang <liaochang1@...wei.com>, linux-kernel@...r.kernel.org,
	linux-trace-kernel@...r.kernel.org
Subject: [PATCH 1/2] uprobes: kill xol_area->slot_count

Add the new helper, xol_get_slot_nr() which does
find_first_zero_bit() + test_and_set_bit().

xol_take_insn_slot() can wait for the "xol_get_slot_nr() < UINSNS_PER_PAGE"
event instead of "area->slot_count < UINSNS_PER_PAGE".

So we can kill area->slot_count and avoid atomic_inc() + atomic_dec(), this
simplifies the code and can slightly improve the performance.

Signed-off-by: Oleg Nesterov <oleg@...hat.com>
---
 kernel/events/uprobes.c | 29 +++++++++++++++--------------
 1 file changed, 15 insertions(+), 14 deletions(-)

diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c
index 20c58b6ee1ad..38d91e10d455 100644
--- a/kernel/events/uprobes.c
+++ b/kernel/events/uprobes.c
@@ -100,7 +100,6 @@ static LIST_HEAD(delayed_uprobe_list);
  */
 struct xol_area {
 	wait_queue_head_t 		wq;		/* if all slots are busy */
-	atomic_t 			slot_count;	/* number of in-use slots */
 	unsigned long 			*bitmap;	/* 0 = free slot */
 
 	struct page			*page;
@@ -1559,7 +1558,6 @@ static struct xol_area *__create_xol_area(unsigned long vaddr)
 	init_waitqueue_head(&area->wq);
 	/* Reserve the 1st slot for get_trampoline_vaddr() */
 	set_bit(0, area->bitmap);
-	atomic_set(&area->slot_count, 1);
 	insns = arch_uprobe_trampoline(&insns_size);
 	arch_uprobe_copy_ixol(area->page, 0, insns, insns_size);
 
@@ -1632,24 +1630,28 @@ void uprobe_dup_mmap(struct mm_struct *oldmm, struct mm_struct *newmm)
 	}
 }
 
+static unsigned long xol_get_slot_nr(struct xol_area *area)
+{
+	unsigned long slot_nr;
+
+	slot_nr = find_first_zero_bit(area->bitmap, UINSNS_PER_PAGE);
+	if (slot_nr < UINSNS_PER_PAGE) {
+		if (!test_and_set_bit(slot_nr, area->bitmap))
+			return slot_nr;
+	}
+
+	return UINSNS_PER_PAGE;
+}
+
 /*
  *  - search for a free slot.
  */
 static unsigned long xol_take_insn_slot(struct xol_area *area)
 {
-	unsigned int slot_nr;
+	unsigned long slot_nr;
 
-	for (;;) {
-		slot_nr = find_first_zero_bit(area->bitmap, UINSNS_PER_PAGE);
-		if (slot_nr < UINSNS_PER_PAGE) {
-			if (!test_and_set_bit(slot_nr, area->bitmap))
-				break;
-			continue;
-		}
-		wait_event(area->wq, (atomic_read(&area->slot_count) < UINSNS_PER_PAGE));
-	}
+	wait_event(area->wq, (slot_nr = xol_get_slot_nr(area)) < UINSNS_PER_PAGE);
 
-	atomic_inc(&area->slot_count);
 	return area->vaddr + slot_nr * UPROBE_XOL_SLOT_BYTES;
 }
 
@@ -1685,7 +1687,6 @@ static void xol_free_insn_slot(struct uprobe_task *utask)
 
 	slot_nr = offset / UPROBE_XOL_SLOT_BYTES;
 	clear_bit(slot_nr, area->bitmap);
-	atomic_dec(&area->slot_count);
 	smp_mb__after_atomic(); /* pairs with prepare_to_wait() */
 	if (waitqueue_active(&area->wq))
 		wake_up(&area->wq);
-- 
2.25.1.362.g51ebf55