Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Luis Chamberlain <mcgrof@kernel.org>
To:     david@redhat.com, patches@lists.linux.dev,
        linux-modules@vger.kernel.org, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org, pmladek@suse.com,
        petr.pavlu@suse.com, prarit@redhat.com,
        torvalds@linux-foundation.org, gregkh@linuxfoundation.org,
        rafael@kernel.org
Cc:     christophe.leroy@csgroup.eu, tglx@linutronix.de,
        peterz@infradead.org, song@kernel.org, rppt@kernel.org,
        dave@stgolabs.net, willy@infradead.org, vbabka@suse.cz,
        mhocko@suse.com, dave.hansen@linux.intel.com,
        colin.i.king@gmail.com, jim.cromie@gmail.com,
        catalin.marinas@arm.com, jbaron@akamai.com,
        rick.p.edgecombe@intel.com, mcgrof@kernel.org
Subject: [PATCH v2 2/2] modules/kmod: replace implementation with a sempahore
Date:   Wed,  5 Apr 2023 13:35:05 -0700
Message-Id: <20230405203505.1343562-3-mcgrof@kernel.org>
In-Reply-To: <20230405203505.1343562-1-mcgrof@kernel.org>
References: <20230405203505.1343562-1-mcgrof@kernel.org>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Sender: Luis Chamberlain <mcgrof@infradead.org>
Precedence: bulk

Simplfy the concurrency delimiter we user for kmod with the semaphore.
I had used the kmod strategy to try to implement a similar concurrency
delimiter for the kernel_read*() calls from the finit_module() path
so to reduce vmalloc() memory pressure. That effort didn't provid yet
conclusive results, but one thing that did became clear is we can use
the suggested alternative solution with semaphores which Linus hinted
at instead of using the atomic / wait strategy.

I've stress tested this with kmod test 0008:

time /data/linux-next/tools/testing/selftests/kmod/kmod.sh -t 0008

And I get only a *slight* delay. That delay however is small, a few
seconds for a full test loop run that runs 150 times, for about ~30-40
seconds. The small delay is worth the simplfication IMHO.

Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
---
 kernel/module/kmod.c | 26 +++++++-------------------
 1 file changed, 7 insertions(+), 19 deletions(-)

diff --git a/kernel/module/kmod.c b/kernel/module/kmod.c
index b717134ebe17..925eb85b8346 100644
--- a/kernel/module/kmod.c
+++ b/kernel/module/kmod.c
@@ -40,8 +40,7 @@
  * effect. Systems like these are very unlikely if modules are enabled.
  */
 #define MAX_KMOD_CONCURRENT 50
-static atomic_t kmod_concurrent_max = ATOMIC_INIT(MAX_KMOD_CONCURRENT);
-static DECLARE_WAIT_QUEUE_HEAD(kmod_wq);
+static DEFINE_SEMAPHORE(kmod_concurrent_max, MAX_KMOD_CONCURRENT);
 
 /*
  * This is a restriction on having *all* MAX_KMOD_CONCURRENT threads
@@ -148,29 +147,18 @@ int __request_module(bool wait, const char *fmt, ...)
 	if (ret)
 		return ret;
 
-	if (atomic_dec_if_positive(&kmod_concurrent_max) < 0) {
-		pr_warn_ratelimited("request_module: kmod_concurrent_max (%u) close to 0 (max_modprobes: %u), for module %s, throttling...",
-				    atomic_read(&kmod_concurrent_max),
-				    MAX_KMOD_CONCURRENT, module_name);
-		ret = wait_event_killable_timeout(kmod_wq,
-						  atomic_dec_if_positive(&kmod_concurrent_max) >= 0,
-						  MAX_KMOD_ALL_BUSY_TIMEOUT * HZ);
-		if (!ret) {
-			pr_warn_ratelimited("request_module: modprobe %s cannot be processed, kmod busy with %d threads for more than %d seconds now",
-					    module_name, MAX_KMOD_CONCURRENT, MAX_KMOD_ALL_BUSY_TIMEOUT);
-			return -ETIME;
-		} else if (ret == -ERESTARTSYS) {
-			pr_warn_ratelimited("request_module: sigkill sent for modprobe %s, giving up", module_name);
-			return ret;
-		}
+	ret = down_timeout(&kmod_concurrent_max, MAX_KMOD_ALL_BUSY_TIMEOUT);
+	if (ret) {
+		pr_warn_ratelimited("request_module: modprobe %s cannot be processed, kmod busy with %d threads for more than %d seconds now",
+				    module_name, MAX_KMOD_CONCURRENT, MAX_KMOD_ALL_BUSY_TIMEOUT);
+		return ret;
 	}
 
 	trace_module_request(module_name, wait, _RET_IP_);
 
 	ret = call_modprobe(module_name, wait ? UMH_WAIT_PROC : UMH_WAIT_EXEC);
 
-	atomic_inc(&kmod_concurrent_max);
-	wake_up(&kmod_wq);
+	up(&kmod_concurrent_max);
 
 	return ret;
 }
-- 
2.39.2