El 12/3/22 a les 12:30, Samuel Thibault ha escrit:
> Is netdde using your modified libpciaccess?
That was it, netdde was using the original libpciaccess.
I'm attaching a patch to fix the problem.
Hello,
Joan Lledó, le sam. 12 mars 2022 12:22:39 +0100, a ecrit:
> El 23/1/22 a les 18:44, Samuel Thibault ha escrit:
> > With a normal hurd system, i.e. no bootstrap pci-arbiter, and using
> > netdde, which is thus supposed to eventually use the translator sitting
> > on /servers/bus/pci
>
> I s
El 23/1/22 a les 18:44, Samuel Thibault ha escrit:
With a normal hurd system, i.e. no bootstrap pci-arbiter, and using
netdde, which is thus supposed to eventually use the translator sitting
on /servers/bus/pci
I still don't know how to reproduce this, if I run a normal Hurd with an
e1000 net
Joan Lledó, le dim. 23 janv. 2022 18:35:58 +0100, a ecrit:
> El 9/1/22 a les 1:02, Samuel Thibault ha escrit:
> > Err, this is breaking everything when pci-arbiter is not running as a
> > bootstrap translator. In that case e.g. netdde succeeds getting the
> > privileged port, and thus tries to open
Hi,
El 9/1/22 a les 1:02, Samuel Thibault ha escrit:
Err, this is breaking everything when pci-arbiter is not running as a
bootstrap translator. In that case e.g. netdde succeeds getting the
privileged port, and thus tries to open pci, but fails, and thus the
whole thing aborts. What problem the
Samuel Thibault, le dim. 09 janv. 2022 13:14:21 +0100, a ecrit:
> Joan Lledó, le dim. 09 janv. 2022 12:16:21 +0100, a ecrit:
> >
> > El 9/1/22 a les 1:15, Samuel Thibault ha escrit:
> > > > Err, this is breaking everything when pci-arbiter is not running as a
> > > > bootstrap translator. In that
Joan Lledó, le dim. 09 janv. 2022 12:16:21 +0100, a ecrit:
>
> El 9/1/22 a les 1:15, Samuel Thibault ha escrit:
> > > Err, this is breaking everything when pci-arbiter is not running as a
> > > bootstrap translator. In that case e.g. netdde succeeds getting the
> > > privileged port, and thus trie
El 9/1/22 a les 1:15, Samuel Thibault ha escrit:
Err, this is breaking everything when pci-arbiter is not running as a
bootstrap translator. In that case e.g. netdde succeeds getting the
privileged port, and thus tries to open pci, but fails, and thus the
whole thing aborts. What problem the ol
Samuel Thibault, le dim. 09 janv. 2022 01:02:57 +0100, a ecrit:
> Joan Lledó via Bug reports for the GNU Hurd, le mer. 05 janv. 2022 13:08:01
> +0100, a ecrit:
> > @@ -523,18 +637,18 @@ pci_system_hurd_create(void)
> >
> > pci_sys->num_devices = 0;
> >
> > -if ((err = get_privileged_p
Joan Lledó via Bug reports for the GNU Hurd, le mer. 05 janv. 2022 13:08:01
+0100, a ecrit:
> @@ -523,18 +637,18 @@ pci_system_hurd_create(void)
>
> pci_sys->num_devices = 0;
>
> -if ((err = get_privileged_ports (NULL, &device_master)) ||
> (device_master == MACH_PORT_NULL)) {
> -
Applied, thanks!!
Samuel
Joan Lledó via Bug reports for the GNU Hurd, le mer. 05 janv. 2022 13:08:01
+0100, a ecrit:
> From: Joan Lledó
>
> * src/hurd_pci.c:
> * Implement device memory mapping functions
> * pci_device_hurd_map_range
> * pci_device_hurd_unmap_
Hi,
> This is not using map.base as set by pci_device_hurd_map_legacy?
You're right, I checked it out and found that libpciaccess has a set of old and
deprecated memory mapping functions to map/unmap entire BARs (regions). Later,
they added new functions to allow manny mappings in the same regi
Joan Lledó, le jeu. 30 déc. 2021 12:19:37 +0100, a ecrit:
> +static int
> +pci_device_hurd_map_range(struct pci_device *dev,
[...]
> +err = vm_map (mach_task_self (), (vm_address_t *)&map->memory, map->size,
> + 0, 1,
> + pager, /* a memory object proxy contain
Hi,
I attached a patch with the changes
Joan Lledó, le mar. 28 déc. 2021 17:48:20 +0100, a ecrit:
> El 28/12/21 a les 13:55, Samuel Thibault ha escrit:
> >
> > ? The interface explicitly says they can be different.
> >
>
> Sorry, I was confused. I meant the function pci_device_hurd_map_range at
> libpciaccess. If io_map returns differ
Hi,
El 28/12/21 a les 13:55, Samuel Thibault ha escrit:
? The interface explicitly says they can be different.
Samuel
Sorry, I was confused. I meant the function pci_device_hurd_map_range at
libpciaccess. If io_map returns different ports for robj and wobj, we
can only assign one of them
Joan Lledó, le mar. 28 déc. 2021 13:26:12 +0100, a ecrit:
> El 19/12/21 a les 16:40, Samuel Thibault ha escrit:
> > “
> > For objects where read data and write data are the same, these objects
> > will be equal, otherwise they will be disjoint.
> > ”
> >
>
> In that case, which one should we retu
Hi,
El 19/12/21 a les 16:40, Samuel Thibault ha escrit:
“
For objects where read data and write data are the same, these objects
will be equal, otherwise they will be disjoint.
”
In that case, which one should we return as pager? Probably the right
thing is to return an error, since a caller
Joan Lledó, le dim. 19 déc. 2021 12:14:18 +0100, a ecrit:
> El 12/12/21 a les 16:19, Samuel Thibault ha escrit:
> >
> > Don't we need
> >
> > else { pager = wobj; } ?
> >
> > Otherwise pager would be undefined.
>
> In the case when the two previous conditions are false, what about robj? can
> w
El 12/12/21 a les 16:19, Samuel Thibault ha escrit:
Don't we need
else { pager = wobj; } ?
Otherwise pager would be undefined.
In the case when the two previous conditions are false, what about robj?
can we be sure it's null and not leaked?
When could the two previous conditions be fal
Hello,
Joan Lledó, le dim. 12 déc. 2021 13:52:14 +0100, a ecrit:
> +err = io_map (file, &robj, &wobj);
> +mach_port_deallocate (mach_task_self(), file);
> +if (err)
> +return err;
> +
> +switch (prot & (VM_PROT_READ|VM_PROT_WRITE)) {
> +case VM_PROT_READ:
> +pag
21 matches
Mail list logo
