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:   =?UTF-8?q?Christoph=20B=C3=B6hmwalder?= 
        <christoph.boehmwalder@linbit.com>
To:     Jens Axboe <axboe@kernel.dk>
Cc:     drbd-dev@lists.linbit.com, linux-kernel@vger.kernel.org,
        Lars Ellenberg <lars.ellenberg@linbit.com>,
        Philipp Reisner <philipp.reisner@linbit.com>,
        linux-block@vger.kernel.org,
        =?UTF-8?q?Christoph=20B=C3=B6hmwalder?= 
        <christoph.boehmwalder@linbit.com>
Subject: [PATCH 4/5] drbd: introduce drbd_ratelimit()
Date:   Thu,  1 Dec 2022 12:03:49 +0100
Message-Id: <20221201110349.1282687-5-christoph.boehmwalder@linbit.com>
In-Reply-To: <20221201110349.1282687-1-christoph.boehmwalder@linbit.com>
References: <20221201110349.1282687-1-christoph.boehmwalder@linbit.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Precedence: bulk

Use call site specific ratelimit instead of one single static global.
Also ratelimit ASSERTION messages generated by expect().

Originally-from: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Christoph Böhmwalder <christoph.boehmwalder@linbit.com>
---
 drivers/block/drbd/drbd_actlog.c           |  2 +-
 drivers/block/drbd/drbd_bitmap.c           |  6 +++---
 drivers/block/drbd/drbd_int.h              |  2 +-
 drivers/block/drbd/drbd_main.c             |  2 +-
 drivers/block/drbd/drbd_polymorph_printk.h | 10 +++++++++-
 drivers/block/drbd/drbd_receiver.c         |  4 ++--
 drivers/block/drbd/drbd_req.c              |  6 +++---
 drivers/block/drbd/drbd_worker.c           | 12 ++++++------
 8 files changed, 26 insertions(+), 18 deletions(-)

diff --git a/drivers/block/drbd/drbd_actlog.c b/drivers/block/drbd/drbd_actlog.c
index 5db147f3c02d..87d93012289f 100644
--- a/drivers/block/drbd/drbd_actlog.c
+++ b/drivers/block/drbd/drbd_actlog.c
@@ -1143,7 +1143,7 @@ void drbd_rs_complete_io(struct drbd_device *device, sector_t sector)
 	bm_ext = e ? lc_entry(e, struct bm_extent, lce) : NULL;
 	if (!bm_ext) {
 		spin_unlock_irqrestore(&device->al_lock, flags);
-		if (__ratelimit(&drbd_ratelimit_state))
+		if (drbd_ratelimit())
 			drbd_err(device, "drbd_rs_complete_io() called, but extent not found\n");
 		return;
 	}
diff --git a/drivers/block/drbd/drbd_bitmap.c b/drivers/block/drbd/drbd_bitmap.c
index b90a5c1003fc..8e6534a90e12 100644
--- a/drivers/block/drbd/drbd_bitmap.c
+++ b/drivers/block/drbd/drbd_bitmap.c
@@ -113,7 +113,7 @@ struct drbd_bitmap {
 static void __bm_print_lock_info(struct drbd_device *device, const char *func)
 {
 	struct drbd_bitmap *b = device->bitmap;
-	if (!__ratelimit(&drbd_ratelimit_state))
+	if (!drbd_ratelimit())
 		return;
 	drbd_err(device, "FIXME %s[%d] in %s, bitmap locked for '%s' by %s[%d]\n",
 		 current->comm, task_pid_nr(current),
@@ -952,7 +952,7 @@ static void drbd_bm_endio(struct bio *bio)
 		bm_set_page_io_err(b->bm_pages[idx]);
 		/* Not identical to on disk version of it.
 		 * Is BM_PAGE_IO_ERROR enough? */
-		if (__ratelimit(&drbd_ratelimit_state))
+		if (drbd_ratelimit())
 			drbd_err(device, "IO ERROR %d on bitmap page idx %u\n",
 					bio->bi_status, idx);
 	} else {
@@ -1013,7 +1013,7 @@ static void bm_page_io_async(struct drbd_bm_aio_ctx *ctx, int page_nr) __must_ho
 		else
 			len = PAGE_SIZE;
 	} else {
-		if (__ratelimit(&drbd_ratelimit_state)) {
+		if (drbd_ratelimit()) {
 			drbd_err(device, "Invalid offset during on-disk bitmap access: "
 				 "page idx %u, sector %llu\n", page_nr, on_disk_sector);
 		}
diff --git a/drivers/block/drbd/drbd_int.h b/drivers/block/drbd/drbd_int.h
index 1734a7f8a096..ae713338aa46 100644
--- a/drivers/block/drbd/drbd_int.h
+++ b/drivers/block/drbd/drbd_int.h
@@ -1658,7 +1658,7 @@ static inline void __drbd_chk_io_error_(struct drbd_device *device,
 	switch (ep) {
 	case EP_PASS_ON: /* FIXME would this be better named "Ignore"? */
 		if (df == DRBD_READ_ERROR || df == DRBD_WRITE_ERROR) {
-			if (__ratelimit(&drbd_ratelimit_state))
+			if (drbd_ratelimit())
 				drbd_err(device, "Local IO failed in %s.\n", where);
 			if (device->state.disk > D_INCONSISTENT)
 				_drbd_set_state(_NS(device, disk, D_INCONSISTENT), CS_HARD, NULL);
diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c
index e02db1dccab1..acfbba3c0f21 100644
--- a/drivers/block/drbd/drbd_main.c
+++ b/drivers/block/drbd/drbd_main.c
@@ -3767,7 +3767,7 @@ _drbd_insert_fault(struct drbd_device *device, unsigned int type)
 	if (ret) {
 		drbd_fault_count++;
 
-		if (__ratelimit(&drbd_ratelimit_state))
+		if (drbd_ratelimit())
 			drbd_warn(device, "***Simulating %s failure\n",
 				_drbd_fault_str(type));
 	}
diff --git a/drivers/block/drbd/drbd_polymorph_printk.h b/drivers/block/drbd/drbd_polymorph_printk.h
index 46cda9dd9af8..914f2a3c0e2e 100644
--- a/drivers/block/drbd/drbd_polymorph_printk.h
+++ b/drivers/block/drbd/drbd_polymorph_printk.h
@@ -110,6 +110,14 @@ void drbd_dyn_dbg_with_wrong_object_type(void);
 	drbd_printk(KERN_INFO, device, fmt, ## args)
 
 
+#define drbd_ratelimit() \
+({						\
+	static DEFINE_RATELIMIT_STATE(_rs,	\
+		DEFAULT_RATELIMIT_INTERVAL,	\
+		DEFAULT_RATELIMIT_BURST);	\
+	__ratelimit(&_rs);			\
+})
+
 #define D_ASSERT(x, exp)							\
 	do {									\
 		if (!(exp))							\
@@ -124,7 +132,7 @@ void drbd_dyn_dbg_with_wrong_object_type(void);
  */
 #define expect(exp) ({								\
 		bool _bool = (exp);						\
-		if (!_bool)							\
+		if (!_bool && drbd_ratelimit())					\
 			drbd_err(device, "ASSERTION %s FAILED in %s\n",		\
 				#exp, __func__);				\
 		_bool;								\
diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c
index e045fb55f3bf..afbe0df3ff75 100644
--- a/drivers/block/drbd/drbd_receiver.c
+++ b/drivers/block/drbd/drbd_receiver.c
@@ -2165,7 +2165,7 @@ static int receive_RSDataReply(struct drbd_connection *connection, struct packet
 		 * or in drbd_peer_request_endio. */
 		err = recv_resync_read(peer_device, sector, pi);
 	} else {
-		if (__ratelimit(&drbd_ratelimit_state))
+		if (drbd_ratelimit())
 			drbd_err(device, "Can not write resync data to local disk.\n");
 
 		err = drbd_drain_block(peer_device, pi->size);
@@ -2846,7 +2846,7 @@ static int receive_DataRequest(struct drbd_connection *connection, struct packet
 		default:
 			BUG();
 		}
-		if (verb && __ratelimit(&drbd_ratelimit_state))
+		if (verb && drbd_ratelimit())
 			drbd_err(device, "Can not satisfy peer's read request, "
 			    "no local data.\n");
 
diff --git a/drivers/block/drbd/drbd_req.c b/drivers/block/drbd/drbd_req.c
index ced15557197a..eb14ec8ec04c 100644
--- a/drivers/block/drbd/drbd_req.c
+++ b/drivers/block/drbd/drbd_req.c
@@ -144,7 +144,7 @@ void drbd_req_destroy(struct kref *kref)
 			if (get_ldev_if_state(device, D_FAILED)) {
 				drbd_al_complete_io(device, &req->i);
 				put_ldev(device);
-			} else if (__ratelimit(&drbd_ratelimit_state)) {
+			} else if (drbd_ratelimit()) {
 				drbd_warn(device, "Should have called drbd_al_complete_io(, %llu, %u), "
 					 "but my Disk seems to have failed :(\n",
 					 (unsigned long long) req->i.sector, req->i.size);
@@ -518,7 +518,7 @@ static void mod_rq_state(struct drbd_request *req, struct bio_and_error *m,
 
 static void drbd_report_io_error(struct drbd_device *device, struct drbd_request *req)
 {
-	if (!__ratelimit(&drbd_ratelimit_state))
+	if (!drbd_ratelimit())
 		return;
 
 	drbd_warn(device, "local %s IO error sector %llu+%u on %pg\n",
@@ -1402,7 +1402,7 @@ static void drbd_send_and_submit(struct drbd_device *device, struct drbd_request
 		submit_private_bio = true;
 	} else if (no_remote) {
 nodata:
-		if (__ratelimit(&drbd_ratelimit_state))
+		if (drbd_ratelimit())
 			drbd_err(device, "IO ERROR: neither local nor remote data, sector %llu+%u\n",
 					(unsigned long long)req->i.sector, req->i.size >> 9);
 		/* A write may have been queued for send_oos, however.
diff --git a/drivers/block/drbd/drbd_worker.c b/drivers/block/drbd/drbd_worker.c
index 3df033bfccf8..f46738040d6b 100644
--- a/drivers/block/drbd/drbd_worker.c
+++ b/drivers/block/drbd/drbd_worker.c
@@ -176,7 +176,7 @@ void drbd_peer_request_endio(struct bio *bio)
 	bool is_discard = bio_op(bio) == REQ_OP_WRITE_ZEROES ||
 			  bio_op(bio) == REQ_OP_DISCARD;
 
-	if (bio->bi_status && __ratelimit(&drbd_ratelimit_state))
+	if (bio->bi_status && drbd_ratelimit())
 		drbd_warn(device, "%s: error=%d s=%llus\n",
 				is_write ? (is_discard ? "discard" : "write")
 					: "read", bio->bi_status,
@@ -240,7 +240,7 @@ void drbd_request_endio(struct bio *bio)
 	 * though we still will complain noisily about it.
 	 */
 	if (unlikely(req->rq_state & RQ_LOCAL_ABORTED)) {
-		if (__ratelimit(&drbd_ratelimit_state))
+		if (drbd_ratelimit())
 			drbd_emerg(device, "delayed completion of aborted local request; disk-timeout may be too aggressive\n");
 
 		if (!bio->bi_status)
@@ -1062,7 +1062,7 @@ int w_e_end_data_req(struct drbd_work *w, int cancel)
 	if (likely((peer_req->flags & EE_WAS_ERROR) == 0)) {
 		err = drbd_send_block(peer_device, P_DATA_REPLY, peer_req);
 	} else {
-		if (__ratelimit(&drbd_ratelimit_state))
+		if (drbd_ratelimit())
 			drbd_err(device, "Sending NegDReply. sector=%llus.\n",
 			    (unsigned long long)peer_req->i.sector);
 
@@ -1135,13 +1135,13 @@ int w_e_end_rsdata_req(struct drbd_work *w, int cancel)
 			else
 				err = drbd_send_block(peer_device, P_RS_DATA_REPLY, peer_req);
 		} else {
-			if (__ratelimit(&drbd_ratelimit_state))
+			if (drbd_ratelimit())
 				drbd_err(device, "Not sending RSDataReply, "
 				    "partner DISKLESS!\n");
 			err = 0;
 		}
 	} else {
-		if (__ratelimit(&drbd_ratelimit_state))
+		if (drbd_ratelimit())
 			drbd_err(device, "Sending NegRSDReply. sector %llus.\n",
 			    (unsigned long long)peer_req->i.sector);
 
@@ -1212,7 +1212,7 @@ int w_e_end_csum_rs_req(struct drbd_work *w, int cancel)
 		}
 	} else {
 		err = drbd_send_ack(peer_device, P_NEG_RS_DREPLY, peer_req);
-		if (__ratelimit(&drbd_ratelimit_state))
+		if (drbd_ratelimit())
 			drbd_err(device, "Sending NegDReply. I guess it gets messy.\n");
 	}
 
-- 
2.38.1