Merge tag 'erofs-for-6.8-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs

Pull erofs fixes from Gao Xiang:

 - fix an infinite loop issue of sub-page compressed data support found
   with lengthy stress tests on a 64k-page arm64 VM

 - optimize the temporary buffer allocation for low-memory scenarios,
   which can reduce 20.21% on average under a heavy multi-app launch
   benchmark workload

 - get rid of unnecessary GFP_NOFS

* tag 'erofs-for-6.8-rc3-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs:
  erofs: relaxed temporary buffers allocation on readahead
  erofs: fix infinite loop due to a race of filling compressed_bvecs
  erofs: get rid of unneeded GFP_NOFS

8 files changed, 87 insertions, 63 deletions

diff --git a/fs/erofs/compress.h b/fs/erofs/compress.h
index 279933e007d2..7cc5841577b2 100644
--- a/fs/erofs/compress.h
+++ b/fs/erofs/compress.h
@@ -11,13 +11,12 @@
 struct z_erofs_decompress_req {
 	struct super_block *sb;
 	struct page **in, **out;
-
 	unsigned short pageofs_in, pageofs_out;
 	unsigned int inputsize, outputsize;
 
-	/* indicate the algorithm will be used for decompression */
-	unsigned int alg;
+	unsigned int alg;       /* the algorithm for decompression */
 	bool inplace_io, partial_decoding, fillgaps;
+	gfp_t gfp;      /* allocation flags for extra temporary buffers */
 };
 
 struct z_erofs_decompressor {
diff --git a/fs/erofs/decompressor.c b/fs/erofs/decompressor.c
index 072ef6a66823..d4cee95af14c 100644
--- a/fs/erofs/decompressor.c
+++ b/fs/erofs/decompressor.c
@@ -111,8 +111,9 @@ static int z_erofs_lz4_prepare_dstpages(struct z_erofs_lz4_decompress_ctx *ctx,
 			victim = availables[--top];
 			get_page(victim);
 		} else {
-			victim = erofs_allocpage(pagepool,
-						 GFP_KERNEL | __GFP_NOFAIL);
+			victim = erofs_allocpage(pagepool, rq->gfp);
+			if (!victim)
+				return -ENOMEM;
 			set_page_private(victim, Z_EROFS_SHORTLIVED_PAGE);
 		}
 		rq->out[i] = victim;
diff --git a/fs/erofs/decompressor_deflate.c b/fs/erofs/decompressor_deflate.c
index 4a64a9c91dd3..b98872058abe 100644
--- a/fs/erofs/decompressor_deflate.c
+++ b/fs/erofs/decompressor_deflate.c
@@ -95,7 +95,7 @@ int z_erofs_load_deflate_config(struct super_block *sb,
 }
 
 int z_erofs_deflate_decompress(struct z_erofs_decompress_req *rq,
-			       struct page **pagepool)
+			       struct page **pgpl)
 {
 	const unsigned int nrpages_out =
 		PAGE_ALIGN(rq->pageofs_out + rq->outputsize) >> PAGE_SHIFT;
@@ -158,8 +158,12 @@ again:
 			strm->z.avail_out = min_t(u32, outsz, PAGE_SIZE - pofs);
 			outsz -= strm->z.avail_out;
 			if (!rq->out[no]) {
-				rq->out[no] = erofs_allocpage(pagepool,
-						GFP_KERNEL | __GFP_NOFAIL);
+				rq->out[no] = erofs_allocpage(pgpl, rq->gfp);
+				if (!rq->out[no]) {
+					kout = NULL;
+					err = -ENOMEM;
+					break;
+				}
 				set_page_private(rq->out[no],
 						 Z_EROFS_SHORTLIVED_PAGE);
 			}
@@ -211,8 +215,11 @@ again:
 
 			DBG_BUGON(erofs_page_is_managed(EROFS_SB(sb),
 							rq->in[j]));
-			tmppage = erofs_allocpage(pagepool,
-						  GFP_KERNEL | __GFP_NOFAIL);
+			tmppage = erofs_allocpage(pgpl, rq->gfp);
+			if (!tmppage) {
+				err = -ENOMEM;
+				goto failed;
+			}
 			set_page_private(tmppage, Z_EROFS_SHORTLIVED_PAGE);
 			copy_highpage(tmppage, rq->in[j]);
 			rq->in[j] = tmppage;
@@ -230,7 +237,7 @@ again:
 			break;
 		}
 	}
-
+failed:
 	if (zlib_inflateEnd(&strm->z) != Z_OK && !err)
 		err = -EIO;
 	if (kout)
diff --git a/fs/erofs/decompressor_lzma.c b/fs/erofs/decompressor_lzma.c
index 2dd14f99c1dc..6ca357d83cfa 100644
--- a/fs/erofs/decompressor_lzma.c
+++ b/fs/erofs/decompressor_lzma.c
@@ -148,7 +148,7 @@ again:
 }
 
 int z_erofs_lzma_decompress(struct z_erofs_decompress_req *rq,
-			    struct page **pagepool)
+			    struct page **pgpl)
 {
 	const unsigned int nrpages_out =
 		PAGE_ALIGN(rq->pageofs_out + rq->outputsize) >> PAGE_SHIFT;
@@ -215,8 +215,11 @@ again:
 						   PAGE_SIZE - pageofs);
 			outlen -= strm->buf.out_size;
 			if (!rq->out[no] && rq->fillgaps) {	/* deduped */
-				rq->out[no] = erofs_allocpage(pagepool,
-						GFP_KERNEL | __GFP_NOFAIL);
+				rq->out[no] = erofs_allocpage(pgpl, rq->gfp);
+				if (!rq->out[no]) {
+					err = -ENOMEM;
+					break;
+				}
 				set_page_private(rq->out[no],
 						 Z_EROFS_SHORTLIVED_PAGE);
 			}
@@ -258,8 +261,11 @@ again:
 
 			DBG_BUGON(erofs_page_is_managed(EROFS_SB(rq->sb),
 							rq->in[j]));
-			tmppage = erofs_allocpage(pagepool,
-						  GFP_KERNEL | __GFP_NOFAIL);
+			tmppage = erofs_allocpage(pgpl, rq->gfp);
+			if (!tmppage) {
+				err = -ENOMEM;
+				goto failed;
+			}
 			set_page_private(tmppage, Z_EROFS_SHORTLIVED_PAGE);
 			copy_highpage(tmppage, rq->in[j]);
 			rq->in[j] = tmppage;
@@ -277,6 +283,7 @@ again:
 			break;
 		}
 	}
+failed:
 	if (no < nrpages_out && strm->buf.out)
 		kunmap(rq->out[no]);
 	if (ni < nrpages_in)
diff --git a/fs/erofs/fscache.c b/fs/erofs/fscache.c
index bc12030393b2..5ff90026fd43 100644
--- a/fs/erofs/fscache.c
+++ b/fs/erofs/fscache.c
@@ -459,7 +459,7 @@ static struct erofs_fscache *erofs_fscache_acquire_cookie(struct super_block *sb
 
 	inode->i_size = OFFSET_MAX;
 	inode->i_mapping->a_ops = &erofs_fscache_meta_aops;
-	mapping_set_gfp_mask(inode->i_mapping, GFP_NOFS);
+	mapping_set_gfp_mask(inode->i_mapping, GFP_KERNEL);
 	inode->i_blkbits = EROFS_SB(sb)->blkszbits;
 	inode->i_private = ctx;
 
diff --git a/fs/erofs/inode.c b/fs/erofs/inode.c
index 3d616dea55dc..36e638e8b53a 100644
--- a/fs/erofs/inode.c
+++ b/fs/erofs/inode.c
@@ -60,7 +60,7 @@ static void *erofs_read_inode(struct erofs_buf *buf,
 		} else {
 			const unsigned int gotten = sb->s_blocksize - *ofs;
 
-			copied = kmalloc(vi->inode_isize, GFP_NOFS);
+			copied = kmalloc(vi->inode_isize, GFP_KERNEL);
 			if (!copied) {
 				err = -ENOMEM;
 				goto err_out;
diff --git a/fs/erofs/utils.c b/fs/erofs/utils.c
index 5dea308764b4..e146d09151af 100644
--- a/fs/erofs/utils.c
+++ b/fs/erofs/utils.c
@@ -81,7 +81,7 @@ struct erofs_workgroup *erofs_insert_workgroup(struct super_block *sb,
 repeat:
 	xa_lock(&sbi->managed_pslots);
 	pre = __xa_cmpxchg(&sbi->managed_pslots, grp->index,
-			   NULL, grp, GFP_NOFS);
+			   NULL, grp, GFP_KERNEL);
 	if (pre) {
 		if (xa_is_err(pre)) {
 			pre = ERR_PTR(xa_err(pre));
diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c
index 692c0c39be63..ff0aa72b0db3 100644
--- a/fs/erofs/zdata.c
+++ b/fs/erofs/zdata.c
@@ -82,6 +82,9 @@ struct z_erofs_pcluster {
 	/* L: indicate several pageofs_outs or not */
 	bool multibases;
 
+	/* L: whether extra buffer allocations are best-effort */
+	bool besteffort;
+
 	/* A: compressed bvecs (can be cached or inplaced pages) */
 	struct z_erofs_bvec compressed_bvecs[];
 };
@@ -230,7 +233,7 @@ static int z_erofs_bvec_enqueue(struct z_erofs_bvec_iter *iter,
 		struct page *nextpage = *candidate_bvpage;
 
 		if (!nextpage) {
-			nextpage = erofs_allocpage(pagepool, GFP_NOFS);
+			nextpage = erofs_allocpage(pagepool, GFP_KERNEL);
 			if (!nextpage)
 				return -ENOMEM;
 			set_page_private(nextpage, Z_EROFS_SHORTLIVED_PAGE);
@@ -302,7 +305,7 @@ static struct z_erofs_pcluster *z_erofs_alloc_pcluster(unsigned int size)
 		if (nrpages > pcs->maxpages)
 			continue;
 
-		pcl = kmem_cache_zalloc(pcs->slab, GFP_NOFS);
+		pcl = kmem_cache_zalloc(pcs->slab, GFP_KERNEL);
 		if (!pcl)
 			return ERR_PTR(-ENOMEM);
 		pcl->pclustersize = size;
@@ -563,21 +566,19 @@ static void z_erofs_bind_cache(struct z_erofs_decompress_frontend *fe)
 			__GFP_NOMEMALLOC | __GFP_NORETRY | __GFP_NOWARN;
 	unsigned int i;
 
-	if (i_blocksize(fe->inode) != PAGE_SIZE)
-		return;
-	if (fe->mode < Z_EROFS_PCLUSTER_FOLLOWED)
+	if (i_blocksize(fe->inode) != PAGE_SIZE ||
+	    fe->mode < Z_EROFS_PCLUSTER_FOLLOWED)
 		return;
 
 	for (i = 0; i < pclusterpages; ++i) {
 		struct page *page, *newpage;
 		void *t;	/* mark pages just found for debugging */
 
-		/* the compressed page was loaded before */
+		/* Inaccurate check w/o locking to avoid unneeded lookups */
 		if (READ_ONCE(pcl->compressed_bvecs[i].page))
 			continue;
 
 		page = find_get_page(mc, pcl->obj.index + i);
-
 		if (page) {
 			t = (void *)((unsigned long)page | 1);
 			newpage = NULL;
@@ -597,9 +598,13 @@ static void z_erofs_bind_cache(struct z_erofs_decompress_frontend *fe)
 			set_page_private(newpage, Z_EROFS_PREALLOCATED_PAGE);
 			t = (void *)((unsigned long)newpage | 1);
 		}
-
-		if (!cmpxchg_relaxed(&pcl->compressed_bvecs[i].page, NULL, t))
+		spin_lock(&pcl->obj.lockref.lock);
+		if (!pcl->compressed_bvecs[i].page) {
+			pcl->compressed_bvecs[i].page = t;
+			spin_unlock(&pcl->obj.lockref.lock);
 			continue;
+		}
+		spin_unlock(&pcl->obj.lockref.lock);
 
 		if (page)
 			put_page(page);
@@ -694,7 +699,7 @@ static void z_erofs_cache_invalidate_folio(struct folio *folio,
 	DBG_BUGON(stop > folio_size(folio) || stop < length);
 
 	if (offset == 0 && stop == folio_size(folio))
-		while (!z_erofs_cache_release_folio(folio, GFP_NOFS))
+		while (!z_erofs_cache_release_folio(folio, 0))
 			cond_resched();
 }
 
@@ -713,36 +718,30 @@ int erofs_init_managed_cache(struct super_block *sb)
 	set_nlink(inode, 1);
 	inode->i_size = OFFSET_MAX;
 	inode->i_mapping->a_ops = &z_erofs_cache_aops;
-	mapping_set_gfp_mask(inode->i_mapping, GFP_NOFS);
+	mapping_set_gfp_mask(inode->i_mapping, GFP_KERNEL);
 	EROFS_SB(sb)->managed_cache = inode;
 	return 0;
 }
 
-static bool z_erofs_try_inplace_io(struct z_erofs_decompress_frontend *fe,
-				   struct z_erofs_bvec *bvec)
-{
-	struct z_erofs_pcluster *const pcl = fe->pcl;
-
-	while (fe->icur > 0) {
-		if (!cmpxchg(&pcl->compressed_bvecs[--fe->icur].page,
-			     NULL, bvec->page)) {
-			pcl->compressed_bvecs[fe->icur] = *bvec;
-			return true;
-		}
-	}
-	return false;
-}
-
 /* callers must be with pcluster lock held */
 static int z_erofs_attach_page(struct z_erofs_decompress_frontend *fe,
 			       struct z_erofs_bvec *bvec, bool exclusive)
 {
+	struct z_erofs_pcluster *pcl = fe->pcl;
 	int ret;
 
 	if (exclusive) {
 		/* give priority for inplaceio to use file pages first */
-		if (z_erofs_try_inplace_io(fe, bvec))
+		spin_lock(&pcl->obj.lockref.lock);
+		while (fe->icur > 0) {
+			if (pcl->compressed_bvecs[--fe->icur].page)
+				continue;
+			pcl->compressed_bvecs[fe->icur] = *bvec;
+			spin_unlock(&pcl->obj.lockref.lock);
 			return 0;
+		}
+		spin_unlock(&pcl->obj.lockref.lock);
+
 		/* otherwise, check if it can be used as a bvpage */
 		if (fe->mode >= Z_EROFS_PCLUSTER_FOLLOWED &&
 		    !fe->candidate_bvpage)
@@ -964,7 +963,7 @@ static int z_erofs_read_fragment(struct super_block *sb, struct page *page,
 }
 
 static int z_erofs_do_read_page(struct z_erofs_decompress_frontend *fe,
-				struct page *page)
+				struct page *page, bool ra)
 {
 	struct inode *const inode = fe->inode;
 	struct erofs_map_blocks *const map = &fe->map;
@@ -1014,6 +1013,7 @@ repeat:
 		err = z_erofs_pcluster_begin(fe);
 		if (err)
 			goto out;
+		fe->pcl->besteffort |= !ra;
 	}
 
 	/*
@@ -1280,6 +1280,9 @@ static int z_erofs_decompress_pcluster(struct z_erofs_decompress_backend *be,
 					.inplace_io = overlapped,
 					.partial_decoding = pcl->partial,
 					.fillgaps = pcl->multibases,
+					.gfp = pcl->besteffort ?
+						GFP_KERNEL | __GFP_NOFAIL :
+						GFP_NOWAIT | __GFP_NORETRY
 				 }, be->pagepool);
 
 	/* must handle all compressed pages before actual file pages */
@@ -1322,6 +1325,7 @@ static int z_erofs_decompress_pcluster(struct z_erofs_decompress_backend *be,
 	pcl->length = 0;
 	pcl->partial = true;
 	pcl->multibases = false;
+	pcl->besteffort = false;
 	pcl->bvset.nextpage = NULL;
 	pcl->vcnt = 0;
 
@@ -1423,23 +1427,26 @@ static void z_erofs_fill_bio_vec(struct bio_vec *bvec,
 {
 	gfp_t gfp = mapping_gfp_mask(mc);
 	bool tocache = false;
-	struct z_erofs_bvec *zbv = pcl->compressed_bvecs + nr;
+	struct z_erofs_bvec zbv;
 	struct address_space *mapping;
-	struct page *page, *oldpage;
+	struct page *page;
 	int justfound, bs = i_blocksize(f->inode);
 
 	/* Except for inplace pages, the entire page can be used for I/Os */
 	bvec->bv_offset = 0;
 	bvec->bv_len = PAGE_SIZE;
 repeat:
-	oldpage = READ_ONCE(zbv->page);
-	if (!oldpage)
+	spin_lock(&pcl->obj.lockref.lock);
+	zbv = pcl->compressed_bvecs[nr];
+	page = zbv.page;
+	justfound = (unsigned long)page & 1UL;
+	page = (struct page *)((unsigned long)page & ~1UL);
+	pcl->compressed_bvecs[nr].page = page;
+	spin_unlock(&pcl->obj.lockref.lock);
+	if (!page)
 		goto out_allocpage;
 
-	justfound = (unsigned long)oldpage & 1UL;
-	page = (struct page *)((unsigned long)oldpage & ~1UL);
 	bvec->bv_page = page;
-
 	DBG_BUGON(z_erofs_is_shortlived_page(page));
 	/*
 	 * Handle preallocated cached pages.  We tried to allocate such pages
@@ -1448,7 +1455,6 @@ repeat:
 	 */
 	if (page->private == Z_EROFS_PREALLOCATED_PAGE) {
 		set_page_private(page, 0);
-		WRITE_ONCE(zbv->page, page);
 		tocache = true;
 		goto out_tocache;
 	}
@@ -1459,9 +1465,9 @@ repeat:
 	 * therefore it is impossible for `mapping` to be NULL.
 	 */
 	if (mapping && mapping != mc) {
-		if (zbv->offset < 0)
-			bvec->bv_offset = round_up(-zbv->offset, bs);
-		bvec->bv_len = round_up(zbv->end, bs) - bvec->bv_offset;
+		if (zbv.offset < 0)
+			bvec->bv_offset = round_up(-zbv.offset, bs);
+		bvec->bv_len = round_up(zbv.end, bs) - bvec->bv_offset;
 		return;
 	}
 
@@ -1471,7 +1477,6 @@ repeat:
 
 	/* the cached page is still in managed cache */
 	if (page->mapping == mc) {
-		WRITE_ONCE(zbv->page, page);
 		/*
 		 * The cached page is still available but without a valid
 		 * `->private` pcluster hint.  Let's reconnect them.
@@ -1503,11 +1508,15 @@ repeat:
 	put_page(page);
 out_allocpage:
 	page = erofs_allocpage(&f->pagepool, gfp | __GFP_NOFAIL);
-	if (oldpage != cmpxchg(&zbv->page, oldpage, page)) {
+	spin_lock(&pcl->obj.lockref.lock);
+	if (pcl->compressed_bvecs[nr].page) {
 		erofs_pagepool_add(&f->pagepool, page);
+		spin_unlock(&pcl->obj.lockref.lock);
 		cond_resched();
 		goto repeat;
 	}
+	pcl->compressed_bvecs[nr].page = page;
+	spin_unlock(&pcl->obj.lockref.lock);
 	bvec->bv_page = page;
 out_tocache:
 	if (!tocache || bs != PAGE_SIZE ||
@@ -1685,6 +1694,7 @@ submit_bio_retry:
 
 			if (cur + bvec.bv_len > end)
 				bvec.bv_len = end - cur;
+			DBG_BUGON(bvec.bv_len < sb->s_blocksize);
 			if (!bio_add_page(bio, bvec.bv_page, bvec.bv_len,
 					  bvec.bv_offset))
 				goto submit_bio_retry;
@@ -1785,7 +1795,7 @@ static void z_erofs_pcluster_readmore(struct z_erofs_decompress_frontend *f,
 			if (PageUptodate(page))
 				unlock_page(page);
 			else
-				(void)z_erofs_do_read_page(f, page);
+				(void)z_erofs_do_read_page(f, page, !!rac);
 			put_page(page);
 		}
 
@@ -1806,7 +1816,7 @@ static int z_erofs_read_folio(struct file *file, struct folio *folio)
 	f.headoffset = (erofs_off_t)folio->index << PAGE_SHIFT;
 
 	z_erofs_pcluster_readmore(&f, NULL, true);
-	err = z_erofs_do_read_page(&f, &folio->page);
+	err = z_erofs_do_read_page(&f, &folio->page, false);
 	z_erofs_pcluster_readmore(&f, NULL, false);
 	z_erofs_pcluster_end(&f);
 
@@ -1847,7 +1857,7 @@ static void z_erofs_readahead(struct readahead_control *rac)
 		folio = head;
 		head = folio_get_private(folio);
 
-		err = z_erofs_do_read_page(&f, &folio->page);
+		err = z_erofs_do_read_page(&f, &folio->page, true);
 		if (err && err != -EINTR)
 			erofs_err(inode->i_sb, "readahead error at folio %lu @ nid %llu",
 				  folio->index, EROFS_I(inode)->nid);