linux-kernel - [PATCH 3/3] iommufd: Use atomic_long_try_cmpxchg() in incr_user_locked

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: <20240522082729.971123-3-ubizjak@gmail.com>
Date: Wed, 22 May 2024 10:26:49 +0200
From: Uros Bizjak <ubizjak@...il.com>
To: iommu@...ts.linux.dev,
	linux-kernel@...r.kernel.org
Cc: Uros Bizjak <ubizjak@...il.com>,
	Jason Gunthorpe <jgg@...pe.ca>,
	Kevin Tian <kevin.tian@...el.com>,
	Joerg Roedel <joro@...tes.org>,
	Will Deacon <will@...nel.org>,
	Robin Murphy <robin.murphy@....com>
Subject: [PATCH 3/3] iommufd: Use atomic_long_try_cmpxchg() in incr_user_locked_vm()

Use atomic_long_try_cmpxchg() instead of
atomic_long_cmpxchg (*ptr, old, new) != old in incr_user_locked_vm().
cmpxchg returns success in ZF flag, so this change saves a compare
after cmpxchg (and related move instruction in front of cmpxchg).

Also, atomic_long_try_cmpxchg() implicitly assigns old *ptr
value to "old" when cmpxchg fails. There is no need to re-read
the value in the loop.

Signed-off-by: Uros Bizjak <ubizjak@...il.com>
Cc: Jason Gunthorpe <jgg@...pe.ca>
Cc: Kevin Tian <kevin.tian@...el.com>
Cc: Joerg Roedel <joro@...tes.org>
Cc: Will Deacon <will@...nel.org>
Cc: Robin Murphy <robin.murphy@....com>
---
 drivers/iommu/iommufd/pages.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/iommu/iommufd/pages.c b/drivers/iommu/iommufd/pages.c
index 528f356238b3..117f644a0c5b 100644
--- a/drivers/iommu/iommufd/pages.c
+++ b/drivers/iommu/iommufd/pages.c
@@ -809,13 +809,14 @@ static int incr_user_locked_vm(struct iopt_pages *pages, unsigned long npages)
 
 	lock_limit = task_rlimit(pages->source_task, RLIMIT_MEMLOCK) >>
 		     PAGE_SHIFT;
+
+	cur_pages = atomic_long_read(&pages->source_user->locked_vm);
 	do {
-		cur_pages = atomic_long_read(&pages->source_user->locked_vm);
 		new_pages = cur_pages + npages;
 		if (new_pages > lock_limit)
 			return -ENOMEM;
-	} while (atomic_long_cmpxchg(&pages->source_user->locked_vm, cur_pages,
-				     new_pages) != cur_pages);
+	} while (!atomic_long_try_cmpxchg(&pages->source_user->locked_vm,
+					  &cur_pages, new_pages));
 	return 0;
 }
 
-- 
2.45.1