Merge tag 'ceph-for-6.9-rc1' of https://github.com/ceph/ceph-client

Pull ceph updates from Ilya Dryomov:
 "A patch to minimize blockage when processing very large batches of
  dirty caps and two fixes to better handle EOF in the face of multiple
  clients performing reads and size-extending writes at the same time"

* tag 'ceph-for-6.9-rc1' of https://github.com/ceph/ceph-client:
  ceph: set correct cap mask for getattr request for read
  ceph: stop copying to iter at EOF on sync reads
  ceph: remove SLAB_MEM_SPREAD flag usage
  ceph: break the check delayed cap loop every 5s

2 files changed, 26 insertions, 13 deletions

diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 7fb4aae97412..55051ad09c19 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -4634,6 +4634,14 @@ unsigned long ceph_check_delayed_caps(struct ceph_mds_client *mdsc)
 			iput(inode);
 			spin_lock(&mdsc->cap_delay_lock);
 		}
+
+		/*
+		 * Make sure too many dirty caps or general
+		 * slowness doesn't block mdsc delayed work,
+		 * preventing send_renew_caps() from running.
+		 */
+		if (jiffies - loop_start >= 5 * HZ)
+			break;
 	}
 	spin_unlock(&mdsc->cap_delay_lock);
 	doutc(cl, "done\n");
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index abe8028d95bf..16873d07692f 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -1138,7 +1138,12 @@ ssize_t __ceph_sync_read(struct inode *inode, loff_t *ki_pos,
 		}
 
 		idx = 0;
-		left = ret > 0 ? ret : 0;
+		if (ret <= 0)
+			left = 0;
+		else if (off + ret > i_size)
+			left = i_size - off;
+		else
+			left = ret;
 		while (left > 0) {
 			size_t plen, copied;
 
@@ -1167,15 +1172,13 @@ ssize_t __ceph_sync_read(struct inode *inode, loff_t *ki_pos,
 	}
 
 	if (ret > 0) {
-		if (off > *ki_pos) {
-			if (off >= i_size) {
-				*retry_op = CHECK_EOF;
-				ret = i_size - *ki_pos;
-				*ki_pos = i_size;
-			} else {
-				ret = off - *ki_pos;
-				*ki_pos = off;
-			}
+		if (off >= i_size) {
+			*retry_op = CHECK_EOF;
+			ret = i_size - *ki_pos;
+			*ki_pos = i_size;
+		} else {
+			ret = off - *ki_pos;
+			*ki_pos = off;
 		}
 
 		if (last_objver)
@@ -2126,14 +2129,16 @@ again:
 		int statret;
 		struct page *page = NULL;
 		loff_t i_size;
+		int mask = CEPH_STAT_CAP_SIZE;
 		if (retry_op == READ_INLINE) {
 			page = __page_cache_alloc(GFP_KERNEL);
 			if (!page)
 				return -ENOMEM;
+
+			mask = CEPH_STAT_CAP_INLINE_DATA;
 		}
 
-		statret = __ceph_do_getattr(inode, page,
-					    CEPH_STAT_CAP_INLINE_DATA, !!page);
+		statret = __ceph_do_getattr(inode, page, mask, !!page);
 		if (statret < 0) {
 			if (page)
 				__free_page(page);
@@ -2174,7 +2179,7 @@ again:
 		/* hit EOF or hole? */
 		if (retry_op == CHECK_EOF && iocb->ki_pos < i_size &&
 		    ret < len) {
-			doutc(cl, "hit hole, ppos %lld < size %lld, reading more\n",
+			doutc(cl, "may hit hole, ppos %lld < size %lld, reading more\n",
 			      iocb->ki_pos, i_size);
 
 			read += ret;