Re: [PATCH] Adding a partfs translator.

Hello,

On Mon, 25 May 2026 22:27:53 Samuel Thibault wrote:

 > This way of managing the multiple disks seems to me it would be
 > confusing for users, and doing more than it should be. We of course
 > don't want to require users to set a partfs translator by hand with
 > the enumeration of the disks to be handled, we want that to be
 > automatic. And if we make it automatic, just numbering the entries
 > inside the partfs directory will prevent users from knowing what these
 > devices actually are: usb sticks, sata disks, etc. I believe we do want
 > to keep the underlying driver device name such as wd0. But then it's not
 > really useful to have just one partfs that would contain all disks, we
 > can as well just have one partfs per disk.

Why not create a separate directory with the partfs translator for each type
of media? The hd/ directory for disks, the usb/ directory for usb, and 
so on?

 > These are replicating a lot of the storeio logics, which we'd rather
 > avoid, to avoid having yet more code to maintain. We can just migrate
 > storeio into using netfs rather than trivfs, and make its root behave
 > like before, except it additionally implements get_dirents, and the
 > nodes thusly exposed can behave like a parted-based storeio. Essentially
 > that's a convergence of your current partfs and the existing storeio,
 > throwing away some parts of each.

I think that's the right approach. The only question is about the storeio
translator launch parameters: should keep the part:N:file:/root/... 
mechanism
or open partitions automatically?

 > And notably that does get sense: if the disk is not partitioned, it's
 > not a "directory" :)

I hadn't thought about that. Do you mean the disk has a GPT partitioning
system, but no partitions?

Also, while reviewing the code for the netfs_get_dirents function, I found a
bug that was giving nodes an incorrect dirent_type instead of defining a
dirent_type for each node. Interestingly, this didn't show up when using the
stat and file commands.

Thanks,

-- 
Mikhail Karpov

From c69a792d6b0716ebb8e959938598ea050ce3e90d Mon Sep 17 00:00:00 2001
From: Mikhail Karpov <[email protected]>
Date: Tue, 26 May 2026 17:34:41 +0700
Subject: [PATCH 2/2] fix netfs_get_dirents

---
 partfs/netfs.c | 23 ++++++++++++++---------
 1 file changed, 14 insertions(+), 9 deletions(-)

diff --git a/partfs/netfs.c b/partfs/netfs.c
index 66a30129..ffe9a15b 100644
--- a/partfs/netfs.c
+++ b/partfs/netfs.c
@@ -375,7 +375,11 @@ netfs_get_dirents (struct iouser *cred, struct node *dir, int entry,
   *data = mmap (0, size, PROT_READ|PROT_WRITE, MAP_ANON, 0, 0);
 
   if ((void *) *data == (void *) -1)
-    return errno;
+    {
+      debug ("(void *) *data == (void *) -1\n");
+      debug ("netfs_get_dirents end with errno: %d\n", errno);
+      return errno;
+    }
 
   *datacnt = size;
   *amt = count;
@@ -396,18 +400,19 @@ netfs_get_dirents (struct iouser *cred, struct node *dir, int entry,
     }
 
   debug ("Fill in the real directory entries\n");
-  int dirent_type;
-  if (dir->nn->entries_size > 0)
-    dirent_type = DT_DIR;
-  else
-    if (dir->nn->store->block_size == 1)
-      dirent_type = DT_CHR;
-    else
-      dirent_type = DT_BLK;
 
+  int dirent_type;
   for (size_t i = 0; i < dir->nn->entries_size; ++i)
     {
       current_node = dir->nn->entries[i];
+      if (current_node->nn->entries_size > 0)
+        dirent_type = DT_DIR;
+      else
+        if (current_node->nn->store->block_size == 1)
+          dirent_type = DT_CHR;
+        else
+          dirent_type = DT_BLK;
+
       add_dir_entry (&ptr_data, current_node->nn->name,
                      current_node->nn_stat.st_ino, dirent_type, &count,
                      nentries, &size);
-- 
2.43.0