nfs: pass explicit offset/count to trace events

nfs_folio_length is unsafe to use without having the folio locked and a
check for a NULL ->f_mapping that protects against truncations and can
lead to kernel crashes.  E.g. when running xfstests generic/065 with
all nfs trace points enabled.

Follow the model of the XFS trace points and pass in an explіcit offset
and length.  This has the additional benefit that these values can
be more accurate as some of the users touch partial folio ranges.

Fixes: eb5654b3b89d ("NFS: Enable tracing of nfs_invalidate_folio() and nfs_launder_folio()")
Reported-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>

diff --git a/fs/nfs/file.c b/fs/nfs/file.c
index 0e2f87120cb840a53d4d8ae856c5ea64fa040612..9aa2ab218c0ac5ce870b97077f301354caf2fe75 100644 (file)
--- a/fs/nfs/file.c
+++ b/fs/nfs/file.c
@@ -436,7 +436,7 @@ static void nfs_invalidate_folio(struct folio *folio, size_t offset,
        /* Cancel any unstarted writes on this page */
        nfs_wb_folio_cancel(inode, folio);
        folio_wait_private_2(folio); /* [DEPRECATED] */
-       trace_nfs_invalidate_folio(inode, folio);
+       trace_nfs_invalidate_folio(inode, folio_pos(folio) + offset, length);
 }
 
 /*
@@ -504,7 +504,8 @@ static int nfs_launder_folio(struct folio *folio)
 
        folio_wait_private_2(folio); /* [DEPRECATED] */
        ret = nfs_wb_folio(inode, folio);
-       trace_nfs_launder_folio_done(inode, folio, ret);
+       trace_nfs_launder_folio_done(inode, folio_pos(folio),
+                       folio_size(folio), ret);
        return ret;
 }
 

diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h
index 1e710654af1173ae7e7f223cd20794725dde93a6..352fdaed4075418762ca5c176f0f2e4e15f3e523 100644 (file)
--- a/fs/nfs/nfstrace.h
+++ b/fs/nfs/nfstrace.h
@@ -939,10 +939,11 @@ TRACE_EVENT(nfs_sillyrename_unlink,
 DECLARE_EVENT_CLASS(nfs_folio_event,
                TP_PROTO(
                        const struct inode *inode,
-                       struct folio *folio
+                       loff_t offset,
+                       size_t count
                ),
 
-               TP_ARGS(inode, folio),
+               TP_ARGS(inode, offset, count),
 
                TP_STRUCT__entry(
                        __field(dev_t, dev)
@@ -950,7 +951,7 @@ DECLARE_EVENT_CLASS(nfs_folio_event,
                        __field(u64, fileid)
                        __field(u64, version)
                        __field(loff_t, offset)
-                       __field(u32, count)
+                       __field(size_t, count)
                ),
 
                TP_fast_assign(
@@ -960,13 +961,13 @@ DECLARE_EVENT_CLASS(nfs_folio_event,
                        __entry->fileid = nfsi->fileid;
                        __entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
                        __entry->version = inode_peek_iversion_raw(inode);
-                       __entry->offset = folio_file_pos(folio);
-                       __entry->count = nfs_folio_length(folio);
+                       __entry->offset = offset,
+                       __entry->count = count;
                ),
 
                TP_printk(
                        "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu "
-                       "offset=%lld count=%u",
+                       "offset=%lld count=%zu",
                        MAJOR(__entry->dev), MINOR(__entry->dev),
                        (unsigned long long)__entry->fileid,
                        __entry->fhandle, __entry->version,
@@ -978,18 +979,20 @@ DECLARE_EVENT_CLASS(nfs_folio_event,
        DEFINE_EVENT(nfs_folio_event, name, \
                        TP_PROTO( \
                                const struct inode *inode, \
-                               struct folio *folio \
+                               loff_t offset, \
+                               size_t count \
                        ), \
-                       TP_ARGS(inode, folio))
+                       TP_ARGS(inode, offset, count))
 
 DECLARE_EVENT_CLASS(nfs_folio_event_done,
                TP_PROTO(
                        const struct inode *inode,
-                       struct folio *folio,
+                       loff_t offset,
+                       size_t count,
                        int ret
                ),
 
-               TP_ARGS(inode, folio, ret),
+               TP_ARGS(inode, offset, count, ret),
 
                TP_STRUCT__entry(
                        __field(dev_t, dev)
@@ -998,7 +1001,7 @@ DECLARE_EVENT_CLASS(nfs_folio_event_done,
                        __field(u64, fileid)
                        __field(u64, version)
                        __field(loff_t, offset)
-                       __field(u32, count)
+                       __field(size_t, count)
                ),
 
                TP_fast_assign(
@@ -1008,14 +1011,14 @@ DECLARE_EVENT_CLASS(nfs_folio_event_done,
                        __entry->fileid = nfsi->fileid;
                        __entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
                        __entry->version = inode_peek_iversion_raw(inode);
-                       __entry->offset = folio_file_pos(folio);
-                       __entry->count = nfs_folio_length(folio);
+                       __entry->offset = offset,
+                       __entry->count = count,
                        __entry->ret = ret;
                ),
 
                TP_printk(
                        "fileid=%02x:%02x:%llu fhandle=0x%08x version=%llu "
-                       "offset=%lld count=%u ret=%d",
+                       "offset=%lld count=%zu ret=%d",
                        MAJOR(__entry->dev), MINOR(__entry->dev),
                        (unsigned long long)__entry->fileid,
                        __entry->fhandle, __entry->version,
@@ -1027,10 +1030,11 @@ DECLARE_EVENT_CLASS(nfs_folio_event_done,
        DEFINE_EVENT(nfs_folio_event_done, name, \
                        TP_PROTO( \
                                const struct inode *inode, \
-                               struct folio *folio, \
+                               loff_t offset, \
+                               size_t count, \
                                int ret \
                        ), \
-                       TP_ARGS(inode, folio, ret))
+                       TP_ARGS(inode, offset, count, ret))
 
 DEFINE_NFS_FOLIO_EVENT(nfs_aop_readpage);
 DEFINE_NFS_FOLIO_EVENT_DONE(nfs_aop_readpage_done);

diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 036ede4875cab6a15107385d1935528ff7d1b82d..6fee86a2eb420b9e4911a556b3ab8ad26a0f4a73 100644 (file)
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -333,13 +333,15 @@ out:
 int nfs_read_folio(struct file *file, struct folio *folio)
 {
        struct inode *inode = file_inode(file);
+       loff_t pos = folio_pos(folio);
+       size_t len = folio_size(folio);
        struct nfs_pageio_descriptor pgio;
        struct nfs_open_context *ctx;
        int ret;
 
-       trace_nfs_aop_readpage(inode, folio);
+       trace_nfs_aop_readpage(inode, pos, len);
        nfs_inc_stats(inode, NFSIOS_VFSREADPAGE);
-       task_io_account_read(folio_size(folio));
+       task_io_account_read(len);
 
        /*
         * Try to flush any pending writes to the file..
@@ -383,7 +385,7 @@ int nfs_read_folio(struct file *file, struct folio *folio)
 out_put:
        put_nfs_open_context(ctx);
 out:
-       trace_nfs_aop_readpage_done(inode, folio, ret);
+       trace_nfs_aop_readpage_done(inode, pos, len, ret);
        return ret;
 out_unlock:
        folio_unlock(folio);

diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index acf2d942d78fe2dbb0cdce4adc28c57aedcae86d..b297833a4514a182dc0a07a07521a34bd321bd16 100644 (file)
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -2063,17 +2063,17 @@ int nfs_wb_folio_cancel(struct inode *inode, struct folio *folio)
  */
 int nfs_wb_folio(struct inode *inode, struct folio *folio)
 {
-       loff_t range_start = folio_file_pos(folio);
-       loff_t range_end = range_start + (loff_t)folio_size(folio) - 1;
+       loff_t range_start = folio_pos(folio);
+       size_t len = folio_size(folio);
        struct writeback_control wbc = {
                .sync_mode = WB_SYNC_ALL,
                .nr_to_write = 0,
                .range_start = range_start,
-               .range_end = range_end,
+               .range_end = range_start + len - 1,
        };
        int ret;
 
-       trace_nfs_writeback_folio(inode, folio);
+       trace_nfs_writeback_folio(inode, range_start, len);
 
        for (;;) {
                folio_wait_writeback(folio);
@@ -2091,7 +2091,7 @@ int nfs_wb_folio(struct inode *inode, struct folio *folio)
                        goto out_error;
        }
 out_error:
-       trace_nfs_writeback_folio_done(inode, folio, ret);
+       trace_nfs_writeback_folio_done(inode, range_start, len, ret);
        return ret;
 }