Skip to contentSkip to conversation

syscalls/copy_file_range02.c: Compatible with new and old kernels

PATCH New Jinhui huang 6 participants 14 messages May 31, 2019 – Jul 4, 2019

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 syscalls/copy_file_range02.c: Compatible with new and old kernels

On new kernels, copy_file_range() returned EISDIR when copyed contents to directory, but on old kernels, it returned EBADF, we should accept EBADF to be compatible with new and old kernels.

The patch as follows:
commit 11cbfb10775a ("vfs: deny copy_file_range() for non regular files")

---
 .../syscalls/copy_file_range/copy_file_range02.c   | 33 +++++++++++++++-------
 1 file changed, 23 insertions(+), 10 deletions(-)
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +23 -10
index 07c0207..9e6356c 100644
@@ -54,19 +54,32 @@ static void verify_copy_file_range(unsigned int n)
 	TEST(sys_copy_file_range(fd_src, 0, *tc->copy_to_fd,
 				0, CONTSIZE, tc->flags));
 
-	if (TST_RET == -1) {
-		if (tc->exp_err == TST_ERR) {
+	if (TST_RET != -1) {
+		tst_res(TFAIL,
+			"copy_file_range returned wrong value: %ld", TST_RET);
+		return;
+	}
+
+	if (tc->exp_err == TST_ERR) {
+		tst_res(TPASS | TTERRNO,
+			"copy_file_range failed as expected");
+	} else {
+		/* copy_file_range() returned EISDIR when copyed contents to
+		 * directory on new kernels, but on old kernels, it returned
+		 * EBADF.
+		 *
+		 * the patch as follws:
+		 * commit 11cbfb10775a ("vfs: deny copy_file_range() for non regular files")
+		 */
+		if (tc->exp_err == EISDIR && TST_ERR == EBADF) {
 			tst_res(TPASS | TTERRNO,
-					"copy_file_range failed as expected");
-		} else {
-			tst_res(TFAIL | TTERRNO,
-				"copy_file_range failed unexpectedly; expected %s, but got",
-				tst_strerrno(tc->exp_err));
+				"copy_file_range failed as expected");
 			return;
 		}
-	} else {
-		tst_res(TFAIL,
-			"copy_file_range returned wrong value: %ld", TST_RET);
+
+		tst_res(TFAIL | TTERRNO,
+			"copy_file_range failed unexpectedly; expected %s, but got",
+			tst_strerrno(tc->exp_err));
 	}
 }
 
Signature · 1 line
1.8.3.1
#1 lore raw

PATCH 1/1 syscalls/copy_file_range02.c: Compatible with new and old kernels

Quoted 7 lines · Jinhui huang

On new kernels, copy_file_range() returned EISDIR when copyed contents to directory, but on old kernels, it returned EBADF, we should accept EBADF to be compatible with new and old kernels.

The patch as follows:
commit 11cbfb10775a ("vfs: deny copy_file_range() for non regular files")

Patch makes sense.

Quoted 72 lines
---
 .../syscalls/copy_file_range/copy_file_range02.c   | 33
+++++++++++++++-------
 1 file changed, 23 insertions(+), 10 deletions(-)
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +11 -2
index 07c0207..9e6356c 100644
@@ -54,19 +54,32 @@ static void verify_copy_file_range(unsigned int n)
        TEST(sys_copy_file_range(fd_src, 0, *tc->copy_to_fd,
                                0, CONTSIZE, tc->flags));
 
-       if (TST_RET == -1) {
-               if (tc->exp_err == TST_ERR) {
+       if (TST_RET != -1) {
+               tst_res(TFAIL,
+                       "copy_file_range returned wrong value: %ld",
 TST_RET);
+               return;
+       }
+
+       if (tc->exp_err == TST_ERR) {
+               tst_res(TPASS | TTERRNO,
+                       "copy_file_range failed as expected");
+       } else {
+               /* copy_file_range() returned EISDIR when copyed contents

to
+ * directory on new kernels, but on old kernels, it
returned

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +4 -0
+                * EBADF.
+                *
+                * the patch as follws:
+                * commit 11cbfb10775a ("vfs: deny copy_file_range() for

non regular files")

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +2 -1
+                */
+               if (tc->exp_err == EISDIR && TST_ERR == EBADF) {
                        tst_res(TPASS | TTERRNO,
-                                       "copy_file_range failed as

expected");

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +0 -3
-               } else {
-                       tst_res(TFAIL | TTERRNO,
-                               "copy_file_range failed unexpectedly;

expected %s, but got",

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +1 -4
-                               tst_strerrno(tc->exp_err));
+                               "copy_file_range failed as expected");
                        return;
                }
-       } else {
-               tst_res(TFAIL,
-                       "copy_file_range returned wrong value: %ld",

TST_RET);

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +3 -0
+
+               tst_res(TFAIL | TTERRNO,
+                       "copy_file_range failed unexpectedly; expected %s,

but got",

+                       tst_strerrno(tc->exp_err));
        }

}

--
1.8.3.1

--
Mailing list info: https://lists.linux.it/listinfo/ltp

Signature · 5 lines
Regards,
Li Wang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linux.it/pipermail/ltp/attachments/20190531/9c287c61/attachment.html>
#2 lore raw

PATCH 1/1 syscalls/copy_file_range02.c: Compatible with new and old kernels

In reply to Jinhui huang
Quoted 6 lines · Jinhui huang

On new kernels, copy_file_range() returned EISDIR when copyed contents to directory, but on old kernels, it returned EBADF, we should accept EBADF to be compatible with new and old kernels.

The patch as follows:
commit 11cbfb10775a ("vfs: deny copy_file_range() for non regular files")

Hi,

From description of commit, I wonder if we can add more tests for some non regular files(e.g. block, pipe)?
I just want to increase coverage and fix all similar issues as you did. :-)

Best Regards,
Xiao Yang

Quoted 52 lines
---
 .../syscalls/copy_file_range/copy_file_range02.c   | 33 +++++++++++++++-------
 1 file changed, 23 insertions(+), 10 deletions(-)
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +23 -10
index 07c0207..9e6356c 100644
@@ -54,19 +54,32 @@ static void verify_copy_file_range(unsigned int n)
 	TEST(sys_copy_file_range(fd_src, 0, *tc->copy_to_fd,
 				0, CONTSIZE, tc->flags));
 
-	if (TST_RET == -1) {
-		if (tc->exp_err == TST_ERR) {
+	if (TST_RET != -1) {
+		tst_res(TFAIL,
+			"copy_file_range returned wrong value: %ld", TST_RET);
+		return;
+	}
+
+	if (tc->exp_err == TST_ERR) {
+		tst_res(TPASS | TTERRNO,
+			"copy_file_range failed as expected");
+	} else {
+		/* copy_file_range() returned EISDIR when copyed contents to
+		 * directory on new kernels, but on old kernels, it returned
+		 * EBADF.
+		 *
+		 * the patch as follws:
+		 * commit 11cbfb10775a ("vfs: deny copy_file_range() for non regular files")
+		 */
+		if (tc->exp_err == EISDIR && TST_ERR == EBADF) {
 			tst_res(TPASS | TTERRNO,
-					"copy_file_range failed as expected");
-		} else {
-			tst_res(TFAIL | TTERRNO,
-				"copy_file_range failed unexpectedly; expected %s, but got",
-				tst_strerrno(tc->exp_err));
+				"copy_file_range failed as expected");
 			return;
 		}
-	} else {
-		tst_res(TFAIL,
-			"copy_file_range returned wrong value: %ld", TST_RET);
+
+		tst_res(TFAIL | TTERRNO,
+			"copy_file_range failed unexpectedly; expected %s, but got",
+			tst_strerrno(tc->exp_err));
 	}
 }
 
#3 lore raw

PATCH 1/1 syscalls/copy_file_range02.c: Compatible with new and old kernels

Quoted 5 lines · Xiao Yang
Quoted 4 lines · Jinhui huang

On new kernels, copy_file_range() returned EISDIR when copyed contents to directory, but on old kernels, it returned EBADF, we should accept EBADF to be compatible with new and old kernels.

Quoted 7 lines
Quoted 2 lines

The patch as follows:
commit 11cbfb10775a ("vfs: deny copy_file_range() for non regular files")

Hi,

From description of commit, I wonder if we can add more tests for some non regular files(e.g. block, pipe)?

I have no objection on this. But, is there really make sense to test some more non regular files which not being mentioned by Linux Manual Page?

The copy_file_range02 test errors are all extract from manual page, I commented that in Christian's first patch version. And I don't think it's necessary to test undefined behavior in syscall using, because how do we know what error return is the expected?

I just want to increase coverage and fix all similar issues as you did. :-)

Quoted 70 lines

Best Regards,
Xiao Yang

Quoted 3 lines
---
 .../syscalls/copy_file_range/copy_file_range02.c   | 33

+++++++++++++++-------

Quoted 3 lines
 1 file changed, 23 insertions(+), 10 deletions(-)

diff --git

a/testcases/kernel/syscalls/copy_file_range/copy_file_range02.c b/testcases/kernel/syscalls/copy_file_range/copy_file_range02.c

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c static void verify_copy_file_range(unsigned int n) · line 59

Quoted 12 lines

index 07c0207..9e6356c 100644

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +3 -2
@@ -54,19 +54,32 @@ static void verify_copy_file_range(unsigned int n)
      TEST(sys_copy_file_range(fd_src, 0, *tc->copy_to_fd,
                              0, CONTSIZE, tc->flags));
 
-     if (TST_RET == -1) {
-             if (tc->exp_err == TST_ERR) {
+     if (TST_RET != -1) {
+             tst_res(TFAIL,
+                     "copy_file_range returned wrong value: %ld",

TST_RET);

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c static void verify_copy_file_range(unsigned int n)

Quoted 8 lines
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +8 -0
+             return;
+     }
+
+     if (tc->exp_err == TST_ERR) {
+             tst_res(TPASS | TTERRNO,
+                     "copy_file_range failed as expected");
+     } else {
+             /* copy_file_range() returned EISDIR when copyed contents

to

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c static void verify_copy_file_range(unsigned int n) · line 68

Quoted 1 line
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +1 -0
+              * directory on new kernels, but on old kernels, it

returned

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c static void verify_copy_file_range(unsigned int n)

Quoted 4 lines
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +4 -0
+              * EBADF.
+              *
+              * the patch as follws:
+              * commit 11cbfb10775a ("vfs: deny copy_file_range() for

non regular files")

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c static void verify_copy_file_range(unsigned int n)

Quoted 4 lines
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +2 -1
+              */
+             if (tc->exp_err == EISDIR && TST_ERR == EBADF) {
                      tst_res(TPASS | TTERRNO,
-                                     "copy_file_range failed as

expected");

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c static void verify_copy_file_range(unsigned int n)

Quoted 3 lines
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +0 -3
-             } else {
-                     tst_res(TFAIL | TTERRNO,
-                             "copy_file_range failed unexpectedly;

expected %s, but got",

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c static void verify_copy_file_range(unsigned int n)

Quoted 7 lines
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +1 -4
-                             tst_strerrno(tc->exp_err));
+                             "copy_file_range failed as expected");
                      return;
              }
-     } else {
-             tst_res(TFAIL,
-                     "copy_file_range returned wrong value: %ld",

TST_RET);

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c static void verify_copy_file_range(unsigned int n)

Quoted 3 lines
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +3 -0
+
+             tst_res(TFAIL | TTERRNO,
+                     "copy_file_range failed unexpectedly; expected %s,

but got",

Quoted 4 lines
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +1 -0
+                     tst_strerrno(tc->exp_err));
      }
 }
 
Signature · 5 lines
Regards,
Li Wang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linux.it/pipermail/ltp/attachments/20190531/e62514d2/attachment.html>
#4 lore raw

PATCH 1/1 syscalls/copy_file_range02.c: Compatible with new and old kernels

Hi!

Quoted 7 lines

I have no objection on this. But, is there really make sense to test some more non regular files which not being mentioned by Linux Manual Page?

The copy_file_range02 test errors are all extract from manual page, I commented that in Christian's first patch version. And I don't think it's necessary to test undefined behavior in syscall using, because how do we know what error return is the expected?

That's not undefined that's undocummented at best. The kernel code for vfs_copy_file_range does:

        if (S_ISDIR(inode_in->i_mode) || S_ISDIR(inode_out->i_mode))
                return -EISDIR;
        if (!S_ISREG(inode_in->i_mode) || !S_ISREG(inode_out->i_mode))
                return -EINVAL;

Which means that directories are treated as special here and all other file descriptors that are not regular files are supposed to fail with EINVAL.

So as far as I can tell it makes sense to pass a pipe fd for example and check for EINVAL. And we should do that both for in_fd and out_fd as well.

Signature · 2 lines
Cyril Hrubis
[email protected]
#5 lore raw

PATCH 1/1 syscalls/copy_file_range02.c: Compatible with new and old kernels

Quoted 12 lines · Cyril Hrubis

Hi!

Quoted 7 lines

I have no objection on this. But, is there really make sense to test some more non regular files which not being mentioned by Linux Manual Page?

The copy_file_range02 test errors are all extract from manual page, I commented that in Christian's first patch version. And I don't think it's necessary to test undefined behavior in syscall using, because how do we know what error return is the expected?

That's not undefined that's undocummented at best. The kernel code for vfs_copy_file_range does:

Quoted 8 lines
        if (S_ISDIR(inode_in->i_mode) || S_ISDIR(inode_out->i_mode))
                return -EISDIR;
        if (!S_ISREG(inode_in->i_mode) || !S_ISREG(inode_out->i_mode))
                return -EINVAL;

Which means that directories are treated as special here and all other file descriptors that are not regular files are supposed to fail with EINVAL.

Yeah, Indeed true.

Quoted 4 lines

So as far as I can tell it makes sense to pass a pipe fd for example and check
for EINVAL. And we should do that both for in_fd and out_fd as well.

Sounds reasonable. Thanks for the explanation.

Signature · 5 lines
Regards,
Li Wang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linux.it/pipermail/ltp/attachments/20190531/9c601a43/attachment.html>
#6 lore raw

PATCH 1/1 syscalls/copy_file_range02.c: Compatible with new and old kernels

In reply to Li Wang
Quoted 12 lines · Li Wang
Quoted 8 lines · Xiao Yang
Quoted 6 lines · Jinhui huang

On new kernels, copy_file_range() returned EISDIR when copyed contents to directory, but on old kernels, it returned EBADF, we should accept EBADF to be compatible with new and old kernels.

The patch as follows:
commit 11cbfb10775a ("vfs: deny copy_file_range() for non regular files")

Quoted 8 lines
Quoted 4 lines

Hi,

From description of commit, I wonder if we can add more tests for some non regular files(e.g. block, pipe)?

I have no objection on this. But, is there really make sense to test some more non regular files which not being mentioned by Linux Manual Page?

FYI, more changes to copy_file_range checks are in the works:
https://lore.kernel.org/linux-fsdevel/[email protected]/T/#me34d4363449118bd3b2ec8421d282a77e9a7d2d1

Quoted 69 lines

The copy_file_range02 test errors are all extract from manual page, I commented that in Christian's first patch version. And I don't think it's necessary to test undefined behavior in syscall using, because how do we know what error return is the expected?

Quoted 48 lines

I just want to increase coverage and fix all similar issues as you did. :-)

Best Regards,
Xiao Yang

Quoted 44 lines
---
 .../syscalls/copy_file_range/copy_file_range02.c   | 33 +++++++++++++++-------
 1 file changed, 23 insertions(+), 10 deletions(-)
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +19 -9
index 07c0207..9e6356c 100644
@@ -54,19 +54,32 @@ static void verify_copy_file_range(unsigned int n)
      TEST(sys_copy_file_range(fd_src, 0, *tc->copy_to_fd,
                              0, CONTSIZE, tc->flags));
 
-     if (TST_RET == -1) {
-             if (tc->exp_err == TST_ERR) {
+     if (TST_RET != -1) {
+             tst_res(TFAIL,
+                     "copy_file_range returned wrong value: %ld", TST_RET);
+             return;
+     }
+
+     if (tc->exp_err == TST_ERR) {
+             tst_res(TPASS | TTERRNO,
+                     "copy_file_range failed as expected");
+     } else {
+             /* copy_file_range() returned EISDIR when copyed contents to
+              * directory on new kernels, but on old kernels, it returned
+              * EBADF.
+              *
+              * the patch as follws:
+              * commit 11cbfb10775a ("vfs: deny copy_file_range() for non regular files")
+              */
+             if (tc->exp_err == EISDIR && TST_ERR == EBADF) {
                      tst_res(TPASS | TTERRNO,
-                                     "copy_file_range failed as expected");
-             } else {
-                     tst_res(TFAIL | TTERRNO,
-                             "copy_file_range failed unexpectedly; expected %s, but got",
-                             tst_strerrno(tc->exp_err));
+                             "copy_file_range failed as expected");
                      return;
              }
-     } else {
-             tst_res(TFAIL,
Quoted 11 lines

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c static void verify_copy_file_range(unsigned int n) · line 85

Quoted 8 lines
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +4 -1
-                     "copy_file_range returned wrong value: %ld", TST_RET);
+
+             tst_res(TFAIL | TTERRNO,
+                     "copy_file_range failed unexpectedly; expected %s, but got",
+                     tst_strerrno(tc->exp_err));
      }
 }
 

--
Regards,
Li Wang

--
Mailing list info: https://lists.linux.it/listinfo/ltp

#7 lore raw

PATCH 1/1 syscalls/copy_file_range02.c: Compatible with new and old kernels

Quoted 13 lines · Amir Goldstein
Quoted 12 lines · Li Wang
Quoted 9 lines · Xiao Yang
Quoted 6 lines · Jinhui huang

On new kernels, copy_file_range() returned EISDIR when copyed contents to directory, but on old kernels, it returned EBADF, we should accept EBADF to be compatible with new and old kernels.

The patch as follows:
commit 11cbfb10775a ("vfs: deny copy_file_range() for non regular files")

Hi,

Quoted 7 lines
Quoted 5 lines
Quoted 2 lines

From description of commit, I wonder if we can add more tests for some
non regular files(e.g. block, pipe)?

I have no objection on this. But, is there really make sense to test some more non regular files which not being mentioned by Linux Manual Page?

FYI, more changes to copy_file_range checks are in the works:
https://lore.kernel.org/linux-fsdevel/[email protected]/T/#me34d4363449118bd3b2ec8421d282a77e9a7d2d1

Hi Amir

     Meet again.  We can increase ltp copy_file_range02 coverage include( swapfile ->ETXTBUSY,  immutable file->EPERM) as same as xfstests generic/553[4].

Also the two other checks(overlaping and offset wrap). I am glad to do it.

PS: Why we don't have test for overlaping and offset wrap check on xfstests? Or, I miss it?

Thanks
Yang Xu

Quoted 67 lines
Quoted 67 lines

The copy_file_range02 test errors are all extract from manual page, I commented that in Christian's first patch version. And I don't think it's necessary to test undefined behavior in syscall using, because how do we know what error return is the expected?

Quoted 48 lines

I just want to increase coverage and fix all similar issues as you did. :-)

Best Regards,
Xiao Yang

Quoted 44 lines
---
  .../syscalls/copy_file_range/copy_file_range02.c   | 33 +++++++++++++++-------
  1 file changed, 23 insertions(+), 10 deletions(-)
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +19 -9
index 07c0207..9e6356c 100644
@@ -54,19 +54,32 @@ static void verify_copy_file_range(unsigned int n)
       TEST(sys_copy_file_range(fd_src, 0, *tc->copy_to_fd,
                               0, CONTSIZE, tc->flags));
 
-     if (TST_RET == -1) {
-             if (tc->exp_err == TST_ERR) {
+     if (TST_RET != -1) {
+             tst_res(TFAIL,
+                     "copy_file_range returned wrong value: %ld", TST_RET);
+             return;
+     }
+
+     if (tc->exp_err == TST_ERR) {
+             tst_res(TPASS | TTERRNO,
+                     "copy_file_range failed as expected");
+     } else {
+             /* copy_file_range() returned EISDIR when copyed contents to
+              * directory on new kernels, but on old kernels, it returned
+              * EBADF.
+              *
+              * the patch as follws:
+              * commit 11cbfb10775a ("vfs: deny copy_file_range() for non regular files")
+              */
+             if (tc->exp_err == EISDIR&&  TST_ERR == EBADF) {
                       tst_res(TPASS | TTERRNO,
-                                     "copy_file_range failed as expected");
-             } else {
-                     tst_res(TFAIL | TTERRNO,
-                             "copy_file_range failed unexpectedly; expected %s, but got",
-                             tst_strerrno(tc->exp_err));
+                             "copy_file_range failed as expected");
                       return;
               }
-     } else {
-             tst_res(TFAIL,
Quoted 10 lines

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c static void verify_copy_file_range(unsigned int n) · line 85

Quoted 8 lines
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +4 -1
-                     "copy_file_range returned wrong value: %ld", TST_RET);
+
+             tst_res(TFAIL | TTERRNO,
+                     "copy_file_range failed unexpectedly; expected %s, but got",
+                     tst_strerrno(tc->exp_err));
       }
  }
 

--
Regards,
Li Wang

--
Mailing list info: https://lists.linux.it/listinfo/ltp

#8 lore raw

PATCH 1/1 syscalls/copy_file_range02.c: Compatible with new and old kernels

Quoted 21 lines · Yang Xu
Quoted 18 lines · Amir Goldstein
Quoted 17 lines · Li Wang
Quoted 11 lines · Xiao Yang
Quoted 6 lines · Jinhui huang

On new kernels, copy_file_range() returned EISDIR when copyed contents to directory, but on old kernels, it returned EBADF, we should accept EBADF to be compatible with new and old kernels.

The patch as follows:
commit 11cbfb10775a ("vfs: deny copy_file_range() for non regular files")

Hi,

From description of commit, I wonder if we can add more tests for some
non regular files(e.g. block, pipe)?

I have no objection on this. But, is there really make sense to test some more non regular files which not being mentioned by Linux Manual Page?

Quoted 7 lines
Quoted 2 lines

FYI, more changes to copy_file_range checks are in the works:
https://lore.kernel.org/linux-fsdevel/[email protected]/T/#me34d4363449118bd3b2ec8421d282a77e9a7d2d1

Hi Amir

     Meet again.  We can increase ltp copy_file_range02 coverage include( swapfile ->ETXTBUSY,  immutable file->EPERM) as same as xfstests generic/553[4].

Also the two other checks(overlaping and offset wrap). I am glad to do it.

That would be great!

Quoted 2 lines

PS: Why we don't have test for overlaping and offset wrap check on xfstests? Or, I miss it?

The bounds check test was posted to xfstests:
https://marc.info/?l=linux-xfs&m=155947929219690&w=2

But because the test requires a new feature from xfs_io:
https://marc.info/?l=linux-xfs&m=156152984109774&w=2

I recommended that xfstests maintainer will hold off merging the test, before the
required change is at least in xfsprogs for-next branch.

Thanks,
Amir.

#9 lore raw

PATCH 1/1 syscalls/copy_file_range02.c: Compatible with new and old kernels

Quoted 19 lines · Amir Goldstein
Quoted 18 lines · Yang Xu
Quoted 16 lines · Amir Goldstein
Quoted 15 lines · Li Wang
Quoted 11 lines · Xiao Yang
Quoted 6 lines · Jinhui huang

On new kernels, copy_file_range() returned EISDIR when copyed contents to directory, but on old kernels, it returned EBADF, we should accept EBADF to be compatible with new and old kernels.

The patch as follows:
commit 11cbfb10775a ("vfs: deny copy_file_range() for non regular files")

Hi,

From description of commit, I wonder if we can add more tests for some
non regular files(e.g. block, pipe)?

I have no objection on this. But, is there really make sense to test some more non regular files which not being mentioned by Linux Manual Page?

Quoted 7 lines
Quoted 6 lines
Quoted 2 lines

FYI, more changes to copy_file_range checks are in the works:
https://lore.kernel.org/linux-fsdevel/[email protected]/T/#me34d4363449118bd3b2ec8421d282a77e9a7d2d1

Hi Amir

      Meet again.  We can increase ltp copy_file_range02 coverage include( swapfile ->ETXTBUSY,  immutable file->EPERM) as same as xfstests generic/553[4].

Also the two other checks(overlaping and offset wrap). I am glad to do it.

That would be great!

Hi Amir

I have tried it. swapfile and immutable file has no problem, but when I try to reproduce EINVAL(same file overlaping) without xfs_io, I got EFAULT error. It look likes we must depend on the new xfs_io feature patch. Right?

ps: If it must xfs_io command, I think we can not check this situation because we should only check by copy_file_range syscall. what do you think about it?

another question:
I have seen copy_file_range manpage, it says fd_out data can be overwriting, but I got EFAULT when I use the following code.

open(src_path, O_RDWR|O_CREAT, 0644) = fd_copy open(copy_path, O_RDWR|O_CREAT, 0644) = fd_src
SAFE_WRITE(1, fd_src, CONTENT, CONTSIZE);
SAFE_WRITE(1, fd_copy, CONTENT, CONTSIZE); copy_file_range(fd_src,0, fd_copy, CONTSIZE/4, CONTSIZE/2, 0) = -1 EFAULT

fs/read_write.c copy_file_range function copy_from_user reports this error. If off_out or off_in isn't equal to 0, the error occurs.

---------------------
ret= -EFAULT;
....

if (off_out) {
                copy_from_user(&pos_out, off_out, sizeof(loff_t));

                         goto out;
              }

----------------------

Is it a bug or I miss something?

Thanks
Yang Xu

Quoted 17 lines
Quoted 2 lines

PS: Why we don't have test for overlaping and offset wrap check on xfstests? Or, I miss it?

The bounds check test was posted to xfstests:
https://marc.info/?l=linux-xfs&m=155947929219690&w=2

But because the test requires a new feature from xfs_io:
https://marc.info/?l=linux-xfs&m=156152984109774&w=2

I recommended that xfstests maintainer will hold off merging the test, before the
required change is at least in xfsprogs for-next branch.

Thanks,
Amir.

#10 lore raw

PATCH 1/1 syscalls/copy_file_range02.c: Compatible with new and old kernels

Quoted 29 lines · Yang Xu
Quoted 19 lines · Amir Goldstein
Quoted 18 lines · Yang Xu
Quoted 16 lines · Amir Goldstein
Quoted 15 lines · Li Wang
Quoted 11 lines · Xiao Yang
Quoted 6 lines · Jinhui huang

On new kernels, copy_file_range() returned EISDIR when copyed contents to directory, but on old kernels, it returned EBADF, we should accept EBADF to be compatible with new and old kernels.

The patch as follows:
commit 11cbfb10775a ("vfs: deny copy_file_range() for non regular files")

Hi,

From description of commit, I wonder if we can add more tests for some
non regular files(e.g. block, pipe)?

I have no objection on this. But, is there really make sense to test some more non regular files which not being mentioned by Linux Manual Page?

Quoted 7 lines
Quoted 6 lines
Quoted 2 lines

FYI, more changes to copy_file_range checks are in the works:
https://lore.kernel.org/linux-fsdevel/[email protected]/T/#me34d4363449118bd3b2ec8421d282a77e9a7d2d1

Hi Amir

      Meet again.  We can increase ltp copy_file_range02 coverage include( swapfile ->ETXTBUSY,  immutable file->EPERM) as same as xfstests generic/553[4].

Also the two other checks(overlaping and offset wrap). I am glad to do it.

That would be great!

Quoted 8 lines

Hi Amir

I have tried it. swapfile and immutable file has no problem, but when I try to reproduce EINVAL(same file overlaping) without xfs_io, I got EFAULT error. It look likes we must depend on the new xfs_io feature patch. Right?

ps: If it must xfs_io command, I think we can not check this situation because we should only check by copy_file_range syscall. what do you think about it?

I don't understand how xfs_io is related.
LTP should use only copy_file_range() syscall.
Do you have patches I can look at?

Quoted 14 lines

another question:
I have seen copy_file_range manpage, it says fd_out data can be overwriting, but I got EFAULT when I use the following code.

open(src_path, O_RDWR|O_CREAT, 0644) = fd_copy open(copy_path, O_RDWR|O_CREAT, 0644) = fd_src
SAFE_WRITE(1, fd_src, CONTENT, CONTSIZE);
SAFE_WRITE(1, fd_copy, CONTENT, CONTSIZE); copy_file_range(fd_src,0, fd_copy, CONTSIZE/4, CONTSIZE/2, 0) = -1 EFAULT

fs/read_write.c copy_file_range function copy_from_user reports this error. If off_out or off_in isn't equal to 0, the error occurs.

---------------------
ret= -EFAULT;
....

Quoted 9 lines
if (off_out) {
                copy_from_user(&pos_out, off_out, sizeof(loff_t));

                         goto out;
              }

----------------------

Is it a bug or I miss something?

You know off_out/off_out are either NULL or pointer to loff_t offset variable?

If you have a sample code you may post it for review.

Thanks,
Amir.

#11 lore raw

PATCH 1/1 syscalls/copy_file_range02.c: Compatible with new and old kernels

In reply to Jinhui huang

Hi!

Quoted 28 lines

The patch as follows:
commit 11cbfb10775a ("vfs: deny copy_file_range() for non regular files")

---
 .../syscalls/copy_file_range/copy_file_range02.c   | 33 +++++++++++++++-------
 1 file changed, 23 insertions(+), 10 deletions(-)
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +10 -2
index 07c0207..9e6356c 100644
@@ -54,19 +54,32 @@ static void verify_copy_file_range(unsigned int n)
 	TEST(sys_copy_file_range(fd_src, 0, *tc->copy_to_fd,
 				0, CONTSIZE, tc->flags));
 
-	if (TST_RET == -1) {
-		if (tc->exp_err == TST_ERR) {
+	if (TST_RET != -1) {
+		tst_res(TFAIL,
+			"copy_file_range returned wrong value: %ld", TST_RET);
+		return;
+	}
+
+	if (tc->exp_err == TST_ERR) {
+		tst_res(TPASS | TTERRNO,
+			"copy_file_range failed as expected");
+	} else {

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c static void verify_copy_file_range(unsigned int n) · line 60

Quoted 10 lines
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +8 -1
+		/* copy_file_range() returned EISDIR when copyed contents to
+		 * directory on new kernels, but on old kernels, it returned
+		 * EBADF.
+		 *
+		 * the patch as follws:
+		 * commit 11cbfb10775a ("vfs: deny copy_file_range() for non regular files")
+		 */
+		if (tc->exp_err == EISDIR && TST_ERR == EBADF) {
 			tst_res(TPASS | TTERRNO,
-					"copy_file_range failed as expected");

If nothing else this should be guarded by kernel version check because if new kernel starts to return EBADFD again, that would be regression.

So we should check the kernel version in test setup and set a flag that would be checked here as well.

Signature · 2 lines
Cyril Hrubis
[email protected]
#12 lore raw

PATCH 1/1 syscalls/copy_file_range02.c: Compatible with new and old kernels

Quoted 46 lines · Cyril Hrubis

Hi!

Quoted 6 lines

The patch as follows:
commit 11cbfb10775a ("vfs: deny copy_file_range() for non regular files")

---
 .../syscalls/copy_file_range/copy_file_range02.c   | 33

+++++++++++++++-------

Quoted 3 lines
 1 file changed, 23 insertions(+), 10 deletions(-)

diff --git

a/testcases/kernel/syscalls/copy_file_range/copy_file_range02.c b/testcases/kernel/syscalls/copy_file_range/copy_file_range02.c

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c static void verify_copy_file_range(unsigned int n) · line 59

Quoted 12 lines

index 07c0207..9e6356c 100644

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +3 -2
@@ -54,19 +54,32 @@ static void verify_copy_file_range(unsigned int n)
      TEST(sys_copy_file_range(fd_src, 0, *tc->copy_to_fd,
                              0, CONTSIZE, tc->flags));
 
-     if (TST_RET == -1) {
-             if (tc->exp_err == TST_ERR) {
+     if (TST_RET != -1) {
+             tst_res(TFAIL,
+                     "copy_file_range returned wrong value: %ld",

TST_RET);

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c static void verify_copy_file_range(unsigned int n)

Quoted 8 lines
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +8 -0
+             return;
+     }
+
+     if (tc->exp_err == TST_ERR) {
+             tst_res(TPASS | TTERRNO,
+                     "copy_file_range failed as expected");
+     } else {
+             /* copy_file_range() returned EISDIR when copyed contents

to

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c static void verify_copy_file_range(unsigned int n) · line 68

Quoted 1 line
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +1 -0
+              * directory on new kernels, but on old kernels, it

returned

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c static void verify_copy_file_range(unsigned int n)

Quoted 4 lines
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +4 -0
+              * EBADF.
+              *
+              * the patch as follws:
+              * commit 11cbfb10775a ("vfs: deny copy_file_range() for

non regular files")

testcases/kernel/syscalls/copy_file_range/copy_file_range02.c static void verify_copy_file_range(unsigned int n)

Quoted 4 lines
testcases/kernel/syscalls/copy_file_range/copy_file_range02.c +2 -1
+              */
+             if (tc->exp_err == EISDIR && TST_ERR == EBADF) {
                      tst_res(TPASS | TTERRNO,
-                                     "copy_file_range failed as
Quoted 8 lines

expected");

If nothing else this should be guarded by kernel version check because if new kernel starts to return EBADFD again, that would be regression.

So we should check the kernel version in test setup and set a flag that would be checked here as well.

This is a good suggestion. Another point I can come up is, if an LTS Linux distribution backports that commit 11cbfb10775a to their old kernel as regression fix, then this flag will make no sense.

So, to strict we maybe need to regards the EISDIR as the only one legal errno(copying content to dir) when kernel >= 4.10(includes commit 11cbfb10775a).

Quoted 8 lines

--
Cyril Hrubis
[email protected]

--
Mailing list info: https://lists.linux.it/listinfo/ltp

Signature · 5 lines
Regards,
Li Wang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linux.it/pipermail/ltp/attachments/20190531/61205c12/attachment-0001.html>
#13 lore raw

PATCH 1/1 syscalls/copy_file_range02.c: Compatible with new and old kernels

Hi!

Quoted 3 lines

This is a good suggestion. Another point I can come up is, if an LTS Linux distribution backports that commit 11cbfb10775a to their old kernel as regression fix, then this flag will make no sense.

Well that's life, there is no way how to detect if patch was backported or not, so either we make the test pass on both EISDIR and EBADFD on older kernels or we leave it as it is. Neither of the solutions is perfect.

Quoted 3 lines

So, to strict we maybe need to regards the EISDIR as the only one legal errno(copying content to dir) when kernel >= 4.10(includes commit 11cbfb10775a).

Yes, exactly.

Signature · 2 lines
Cyril Hrubis
[email protected]