Bugfix/fix uio partial copies

In zfs_write(), the loop continues to the next iteration without
accounting for partial copies occurring in uiomove_iov when 
copy_from_user/__copy_from_user_inatomic return a non-zero status.
This results in "zfs: accessing past end of object..." in the 
kernel log, and the write failing.

Account for partial copies and update uio struct before returning
EFAULT, leave a comment explaining the reason why this is done.

Reviewed-by: Brian Behlendorf <[email protected]>
Reviewed-by: ilbsmart <[email protected]>
Signed-off-by: Fabio Scaccabarozzi <[email protected]>
Closes #8673 
Closes #10148

Author: fsvm88

module/zcommon/zfs_uio.c

@@ -80,22 +80,31 @@ uiomove_iov(void *p, size_t n, enum uio_rw rw, struct uio *uio)
 				if (copy_to_user(iov->iov_base+skip, p, cnt))
 					return (EFAULT);
 			} else {
+				unsigned long b_left = 0;
 				if (uio->uio_fault_disable) {
 					if (!zfs_access_ok(VERIFY_READ,
 					    (iov->iov_base + skip), cnt)) {
 						return (EFAULT);
 					}
 					pagefault_disable();
-					if (__copy_from_user_inatomic(p,
-					    (iov->iov_base + skip), cnt)) {
-						pagefault_enable();
-						return (EFAULT);
-					}
+					b_left =
+					    __copy_from_user_inatomic(p,
+					    (iov->iov_base + skip), cnt);
 					pagefault_enable();
 				} else {
-					if (copy_from_user(p,
-					    (iov->iov_base + skip), cnt))
-						return (EFAULT);
+					b_left =
+					    copy_from_user(p,
+					    (iov->iov_base + skip), cnt);
+				}
+				if (b_left > 0) {
+					unsigned long c_bytes =
+					    cnt - b_left;
+					uio->uio_skip += c_bytes;
+					ASSERT3U(uio->uio_skip, <,
+					    iov->iov_len);
+					uio->uio_resid -= c_bytes;
+					uio->uio_loffset += c_bytes;
+					return (EFAULT);
 				}
 			}
 			break;

module/zfs/zfs_vnops.c

@@ -829,6 +829,15 @@ zfs_write(struct inode *ip, uio_t *uio, int ioflag, cred_t *cr)
 			uio->uio_fault_disable = B_FALSE;
 			if (error == EFAULT) {
 				dmu_tx_commit(tx);
+				/*
+				 * Account for partial writes before
+				 * continuing the loop.
+				 * Update needs to occur before the next
+				 * uio_prefaultpages, or prefaultpages may
+				 * error, and we may break the loop early.
+				 */
+				if (tx_bytes != uio->uio_resid)
+					n -= tx_bytes - uio->uio_resid;
 				if (uio_prefaultpages(MIN(n, max_blksz), uio)) {
 					break;
 				}