bug-hurd
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Issues with remap when using a nested pci arbiter


From: Samuel Thibault
Subject: Re: Issues with remap when using a nested pci arbiter
Date: Wed, 11 Dec 2019 02:03:23 +0100
User-agent: NeoMutt/20170609 (1.8.3)

Joan Lledó via Bug reports for the GNU Hurd, le mar. 10 déc. 2019 17:42:58 
+0100, a ecrit:
> Sorry, I said I observed the remap problem in the rpc trace, but I
> observed it in the client's behavior. When running lspci with remap, I
> can see it returns the devices configured by the main arbiter, not the
> nested one,

Ok.

There are a few things to note: in trans/remap.c, 

  /* FIXME: should match just prefix of filename too */

So indeed, when servers/bus/pci/ gets opened, it's not getting remapped.
Feel free to fix the FIXME :)

Also, in pciutils' hurd_detect, 

  lstat (_SERVERS_BUS_PCI, &st);
  ...
  return S_ISDIR (st.st_mode) && ((st.st_mode & S_ITRANS) == S_IROOT);

so the remapped node has to be a translated directory, and can't be a
symlink. I would say that pciutils should be made to use stat instead of
lstat, symlinks are useful.

Samuel



reply via email to

[Prev in Thread] Current Thread [Next in Thread]