Previous problems mentioned with 2x rumpdisk partitions all fixed.
Booted off rumpdisk / and mounted second partition in userspace:
root@zamhurd:~# showtrans /dev/wd0
/hurd/storeio -T typed device:@/dev/rumpdisk:/dev/wd0
root@zamhurd:~# fdisk -l /dev/wd0
Disk /dev/wd0: 298.9 GiB, 320072933376 by
Hi Samuel,
I need to figure out the next step of rumpdisk, using the arbiter instead of
faking the arbiter.
> youpi: pci-arbiter could be exposed as a device name in the master device
> port and the userland pci-arbiter running on /server/bus/pci can try to open
> that
> youpi: just like netdd
On 14/11/20 8:17 pm, Samuel Thibault wrote:
> Hello,
>
> Damien Zammit, le sam. 14 nov. 2020 14:37:37 +1100, a ecrit:
>> Previous problems mentioned with 2x rumpdisk partitions all fixed.
>
> Congrats!
>
> What was the issue?
The logic around default behaviour
Hi,
On 14/11/20 9:09 pm, Samuel Thibault wrote:
> This looks odd to me. The "@/dev/master:" part should rather be
> dropped way before, by the code that opens /dev/master to access the
> master device port before calling device_open() on it: the translators
> shouldn't have to grok which way they
Hi,
On 15/11/20 11:33 am, Damien Zammit wrote:
> I will test again without it.
It works in both modes with current master!
Damien
On 15/11/20 8:56 pm, Samuel Thibault wrote:
> I have uploaded it in Debian a minute ago. We'll be able to start to use
> this in production :)
Unfortunately, it still does not work with the pci-arbiter.
I had a go at fixing that today but it's not quite working.
I am getting "mapping time: Grat
Hi,
On 16/11/20 8:16 am, Samuel Thibault wrote:
> wd1 at atabus7 drive 0
Do you have two controllers or one? This looks like IDE (?)
> wd1:
> wd1: drive supports 16-sector PIO transfers, LBA48 addressing
> wd1: 20480 MB, 41610 cyl, 16 head, 63 sec, 512 bytes/sect x 41943040 sectors
> wd1: drive
Hi,
On 16/11/20 9:24 am, Samuel Thibault wrote:
> opening /dev/wd1d
> could not open
> ext2fs: device:/dev/wd1: No such device or address
FYI, I am using the following grub entry for booting off a rumpdisk:
menuentry "rumpdisk!" {
set root=(hd0,msdos2)
multiboot /boot/gnumach-1.8
recall having issues with the driver without patching QEMU.
Perhaps this will help?
Damien
>From 2c672e1902757dcd202c58382dcfc0727217aafa Mon Sep 17 00:00:00 2001
From: Damien Zammit
Date: Wed, 24 Jun 2020 20:14:48 +1000
Subject: [PATCH] No built-in AHCI on q35 && sync cache on IDE
---
Hi,
On 16/11/20 11:04 am, Samuel Thibault wrote:
> I tried wd0, it does work (but my setup is using wd1, so doesn't boot :) )
Hmm, I have the same problem on my end trying to boot off wd1:
ahcisata0 port 1: device present, speed: 1.5Gb/s
ahcisata0 port 0: device present, speed: 1.5Gb/s
wd0 at at
On 16/11/20 12:40 pm, Samuel Thibault wrote:
> It booted fine :D
\o/ YAY!
Hi,
How do I expose the hurdish pci subsystem that has no underlying node to attach
to for a netfs
in pci-arbiter during bootstrap?
I have almost completed the loop with rumpdisk and the arbiter, but I cannot
start pci-arbiter
as it has no underlying node to attach to, and the specific implemen
Hi,
On 16/11/20 9:02 pm, Samuel Thibault wrote:
> ? Like rumpdisk does?
start pci-arbiter: Hurd bootstrap pci pci-arbiter: Must be started as a
translator
Copyright (c) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005,
2006, 2007, 2008, 2009, 2010, 2011, 2012, 2013, 2014, 2015, 201
Somehow I was able to boot / via rumpdisk and then the arbiter
still worked afterwards, so networking via netdde started working.
This is the first time I've had a rumpdisk / with network access!
Alas, I cannot seem to make it work via the arbiter though.
My latest attempt looks like the arbiter s
On 24/11/20 5:16 am, Samuel Thibault wrote:
> Hello,
>
> Had you tested support for cd-rom? With qemu I am getting a media sense
> error with status 0x50.
I think that was the reason for the qemu patch for syncing IDE cache iirc.
Damien
Hi,
On 24/1/21 3:32 am, Samuel Thibault wrote:
> Yes, but the real answer is that we should fix vim.
I noticed that vim is building with -O2, shouldn't it also have these flags for
hurd-i386?
-mno-sse -mno-mmx -mno-sse2 -mno-3dnow
I will try rebuilding the package with these flags and see if i
Hi,
On 24/1/21 11:28 am, Samuel Thibault wrote:
> Why so? We do support SSE*.
>
> (glibc 2.33 will even use them automatically for memcpy etc. thanks to
> ifunc support recently getting enabled)
OK, I ran the failing test in GDB:
initmaster(int f UNUSED)
{
#ifndef VMS
# ifdef POSIX
tcflush(
Hi,
On 24/1/21 11:41 am, Damien Zammit wrote:
> OK, I ran the failing test in GDB:
Forgot to paste the command I used:
vim/src/vim-gtk3/testdir $ VIMRUNTIME=../../runtime gdb --args ../vim -f -u
unix.vim -U NONE --noplugin --not-a-term -S runtest.vim test_arglist.vim --cmd
'au Sw
On 24/1/21 12:18 pm, Samuel Thibault wrote:
> I'm starting to think that perhaps we should as well just comment the
> #define TIOCFLUSH from the headers, since there are so many bogus
> callers of the interface.
I think we still need to pass -DPOSIX to vim CFLAGS for hurd,
otherwise it will not ca
Hi,
On 23/1/21 12:03 pm, Paul Dufresne wrote:
> I have seen problems with daily images from:
> https://people.debian.org/~sthibault/hurd-i386/installer/cdimage/
>
> Was problems with vim, emacs, gdb... in previous days... I think but unsure
> Idon't seems to see them now.
> Had tried previous mo
Hi Samuel,
On 24/1/21 12:32 pm, Samuel Thibault wrote:
> Damien Zammit, le dim. 24 janv. 2021 12:26:35 +1100, a ecrit:
>> I think we still need to pass -DPOSIX to vim CFLAGS for hurd,
>> otherwise it will not call any flush function if you remove the #define.
>
> Ok, but A
On 25/1/21 11:40 pm, Samuel Thibault wrote:
> Paul Dufresne, le lun. 25 janv. 2021 07:21:33 -0500, a ecrit:
>> Today I tried to install [1]https://people.debian.org/~sthibault/hurd-i386/
>> installer/cdimage/daily/debian-sid-hurd-i386-NETINST-1.iso in QEMU.
>>
>> I got:
>> gdb depends on libboost-r
Hi,
On 25/1/21 11:40 pm, Samuel Thibault wrote:
> Ok, I fixed the missing dep in the cd script, tomorrow's build should be
> fine in that concern.
I burnt the cd to physical media and tried to install natively on X220 thinkpad.
We have an issue because the SSD is not detected by gnumach, but sti
On 26/1/21 4:00 pm, Damien Zammit wrote:
> I burnt the cd to physical media and tried to install natively on X220
> thinkpad.
I also tried the same image on a physical X230 thinkpad, this time the disk was
detected by gnumach,
but it failed to detect the CD ROM so the installer cou
This option disables the built-in disk drivers completely.
Passing " nodisk" to gnumach command line will allow the user
to boot the installer with a ramdisk, and rumpdisk should
initialise AHCI controller in userspace.
This is a workaround for the installer for some modern hardware that
the gnuma
On 31/1/21 7:54 am, Samuel Thibault wrote:
> I uploaded to unreleased a version with that fixed, but apparently the
> build fails a bit later:
>
> [...]
> ../vim -f -u unix.vim -U NONE --noplugin --not-a-term -S runtest.vim
> test_help.vim --cmd 'au SwapExists * let v:swapchoice = "e"' > /dev/nu
Hi all,
[PATCH 1/3] libnetfs: Fix mapping of time fallback to mach device
This could be merged as is.
[PATCH 2/3] Bootstrappable pci-arbiter
This is a work in progress, I would like some comments on how I can
make the arbiter start up without a / filesystem when it needs to
expose the pci inte
---
libnetfs/init-init.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/libnetfs/init-init.c b/libnetfs/init-init.c
index 984a5e71..b63e7ffa 100644
--- a/libnetfs/init-init.c
+++ b/libnetfs/init-init.c
@@ -39,6 +39,8 @@ netfs_init ()
{
error_t err;
err = maptime_map (0, 0, &netfs_mti
---
rumpdisk/main.c | 5 -
1 file changed, 5 deletions(-)
diff --git a/rumpdisk/main.c b/rumpdisk/main.c
index 150dc996..b35cbfa7 100644
--- a/rumpdisk/main.c
+++ b/rumpdisk/main.c
@@ -31,11 +31,6 @@
#include
#include
-/* TODO: Add api to pciaccess to allow selecting backend.
- * For no
---
pci-arbiter/Makefile | 2 +-
pci-arbiter/main.c| 72 ---
pci-arbiter/options.c | 19 +++
pci-arbiter/options.h | 8 +
pci-arbiter/pcifs.h | 3 ++
pci-arbiter/startup-ops.c | 2 ++
6 files changed, 100 insertions(+),
Hi,
Here is the tail of my start up log with some extra debug mach_prints:
I think the "Must be started as a translator" is coming from the netfs start up
when it tries to create a node for the pci filesystem.
start pci-arbiter: PCI start
PCI machdev start
Hurd bootstrap pci PCI machdev done
On 3/2/21 8:53 am, Samuel Thibault wrote:
> Damien Zammit, le mar. 02 févr. 2021 18:29:42 +1100, a ecrit:
>> This is a work in progress, I would like some comments on how I can
>> make the arbiter start up without a / filesystem when it needs to
>> expose the pci interface
Hi,
I think we have another chicken/egg problem:
netfs needs an underlying node to expose pcifs on, so that the disk driver can
start up and expose the real underlying node.
It seems like diskfs_startup_diskfs opens the real disk and finds an actual
root fs node,
but we can't do that in netfs u
I'm getting logs like the following:
PCI netfs done
PCI maptime done
PCI pci_system_init done
machdev_trivfs_server
pci-arbiter: Translator startup failure: fsys_startup: (ipc/send) invalid destin
ation portCopyright (c) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 20
05,
2006, 2
More logs after making the underlying node for netfs not required:
It hangs somewhere on netfs_startup() using a bootstrap port i passed from
libmachdev.
Maybe I am passing the wrong kind of port?
start pci-arbiter: PCI start
PCI machdev start
Hurd bootstrap pci PCI machdev done
PCI netfs done
PC
Hi,
I'm getting very close to working pci + rump.
In pci-arbiter, what port rights do I return in device_open?
I am currently doing this:
static io_return_t
device_open (mach_port_t reply_port, mach_msg_type_name_t reply_port_type,
dev_mode_t mode, char *name, device_t * devp,
On 9/2/21 5:33 am, Samuel Thibault wrote:
> Damien Zammit, le lun. 08 févr. 2021 20:33:16 +1100, a ecrit:
>> I'm getting very close to working pci + rump.
>
> Did you manage to find answers to your previous questions?
> (I have to admit I didn't have time to dive int
I'm getting some issue with libpciaccess, I changed it locally to detect
the pci port from device_open and it seems to be opening the right port, but
there must be a bug with it.
I must be pretty close now, because the arbiter opens the x86 method and rump
opens the hurd method of pciaccess:
sta
Ok so the problem is that its returning ENODIR on the file_name_lookup_under
call.
How do I get the root filesystem of pci arbiter to exist so it can look up
paths on the hurd pci arbiter
before rumpdisk initiates the disk driver?
Do I need to call fsys_getroot somewhere in the arbiter?
pci0 at
Hi Samuel,
On 20/2/21 10:16 pm, Samuel Thibault wrote:
> It's hard to tell exactly without seeing what exactly you have done
> already. As I wrote before, device_open should return a port to the
> root directory of the pci-arbiter tree. I'm not sure exactly what you
> wrote to achieve that.
I thi
Hi,
On 21/2/21 12:24 pm, Samuel Thibault wrote:
> Ah, possibly. Perhaps you could make libnetfs to accept any access when
> auth/proc are not available
Done this!
> ? once libpciaccess gets a port on pci-arbiter's root, it can use
> file_name_lookup_under to open directories and files in it. lib
On 21/2/21 9:21 pm, Samuel Thibault wrote:
> Mmm. I guess that's missing _hurd_init initialization. It would
> be tricky to separate out the really basic pieces from proc/auth
> initialization, I'd say it will be a lot less tricky to just make
> hurd_pci.c use the dir_readdir RPC directly on the ro
Hi,
On 25/2/21 9:30 am, Samuel Thibault wrote:
> From reading the libnetfs/dir-readdir.c source and
> pci-arbiter/netfs_impl.c source I don't see anything obvious. Probably
> some prints in there would allow to make sure how far it is going
> actually.
I got past a hang, by installing the port re
Hi all,
I got the arbiter to play nice with rumpdisk!
But as soon as you log in it crashes the disk and qemu, I think probably
because there is a mismatched
libpciaccess.so in userspace versus the statically linked ones in the pci/disk
servers?
I'm not sure yet.
--
PCI device_open (mem)
PC
---
libmachdev/trivfs_server.c | 31 ---
1 file changed, 24 insertions(+), 7 deletions(-)
diff --git a/libmachdev/trivfs_server.c b/libmachdev/trivfs_server.c
index 3ab15049..06756f82 100644
--- a/libmachdev/trivfs_server.c
+++ b/libmachdev/trivfs_server.c
@@ -398,7 +3
---
rumpdisk/block-rump.c | 16 +---
rumpdisk/main.c | 12 +++-
2 files changed, 20 insertions(+), 8 deletions(-)
diff --git a/rumpdisk/block-rump.c b/rumpdisk/block-rump.c
index 3ee0773d..584056f7 100644
--- a/rumpdisk/block-rump.c
+++ b/rumpdisk/block-rump.c
@@ -27,6 +
Hi all,
This has been a work in progress for some time.
I am hoping it does not need much reworking.
The only problem I can see is the uid=0 faking the cred by allowing
netfs to install ports that have no cred at all, perhaps we need
a new RPC for obtaining a temporary uid=0 cred during bootstr
---
pci-arbiter/Makefile | 7 +-
pci-arbiter/main.c| 140 --
pci-arbiter/netfs_impl.c | 39 +++
pci-arbiter/options.c | 19 ++
pci-arbiter/options.h | 8 +++
pci-arbiter/pci-ops.c | 6 +-
pci-arbiter/pcifs.c |
---
libnetfs/dir-lookup.c | 30 +-
libnetfs/dir-readdir.c | 3 ---
libnetfs/file-getcontrol.c | 13 ++---
libnetfs/init-startup.c| 5 ++---
libnetfs/make-protid.c | 8 +---
5 files changed, 30 insertions(+), 29 deletions(-)
diff --git a/
This patch for libpciaccess goes in lockstep with
the previous patches I sent for hurd.
---
src/hurd_pci.c | 125 ++---
1 file changed, 98 insertions(+), 27 deletions(-)
diff --git a/src/hurd_pci.c b/src/hurd_pci.c
index ada7af8..ff14285 100644
--- a/
This is a cleaned up version of my previous patch.
It fixes a bug by sorting the device array by B/D/F for all backends.
(It assumes only domain exists and I think that assumption is also made
elsewhere).
---
src/common_init.c | 37
src/hurd_pci.c| 108
This is a revised version of previous patch.
When there is no pci "device" present in the system, the fallback
mechanism to use /servers/bus/pci works.
It also boots from a static arbiter.
---
src/common_init.c | 37 +
src/hurd_pci.c| 138 ++---
This splits the previous patch into two, one for sorting devices,
one for hurd improvements.
Thanks,
Damien
---
src/common_init.c | 37 +
1 file changed, 37 insertions(+)
diff --git a/src/common_init.c b/src/common_init.c
index 1940cff..14f0c6b 100644
--- a/src/common_init.c
+++ b/src/common_init.c
@@ -40,6 +40,40 @@
_pci_hidden struct pci_system * pci_sys;
+sta
---
src/hurd_pci.c | 135 -
1 file changed, 88 insertions(+), 47 deletions(-)
diff --git a/src/hurd_pci.c b/src/hurd_pci.c
index ada7af8..6b6e299 100644
--- a/src/hurd_pci.c
+++ b/src/hurd_pci.c
@@ -33,9 +33,12 @@
#include
#include
#include
+#
---
src/hurd_pci.c | 122 ++---
1 file changed, 74 insertions(+), 48 deletions(-)
diff --git a/src/hurd_pci.c b/src/hurd_pci.c
index ada7af8..d988a6e 100644
--- a/src/hurd_pci.c
+++ b/src/hurd_pci.c
@@ -33,9 +33,12 @@
#include
#include
#include
+#
Reverted change to .destroy
---
src/hurd_pci.c | 121 ++---
1 file changed, 74 insertions(+), 47 deletions(-)
diff --git a/src/hurd_pci.c b/src/hurd_pci.c
index ada7af8..40c9925 100644
--- a/src/hurd_pci.c
+++ b/src/hurd_pci.c
@@ -33,9 +33,12 @@
#incl
On 7/3/21 11:51 am, Samuel Thibault wrote:
> That looks good, I'd say you can submit to upstream libpciaccess :)
>
> Samuel
I created an issue upstream with a merge request that fixes it:
https://gitlab.freedesktop.org/xorg/lib/libpciaccess/-/issues/14
Damien
---
pci-arbiter/netfs_impl.c | 42 ++--
pci-arbiter/pci-ops.c| 6 +++---
pci-arbiter/pcifs.c | 18 +++--
3 files changed, 42 insertions(+), 24 deletions(-)
diff --git a/pci-arbiter/netfs_impl.c b/pci-arbiter/netfs_impl.c
index b987a0bc..e9
---
pci-arbiter/Makefile | 7 +-
pci-arbiter/main.c| 147 +++---
pci-arbiter/options.c | 19 +
pci-arbiter/options.h | 8 +++
pci-arbiter/pcifs.h | 3 +
pci-arbiter/startup-ops.c | 2 +
6 files changed, 174 insertions(+), 12 d
---
libnetfs/dir-lookup.c | 30 +-
libnetfs/dir-readdir.c | 3 ---
libnetfs/file-getcontrol.c | 13 ++---
libnetfs/make-protid.c | 8 +---
pci-arbiter/netfs_impl.c | 2 ++
5 files changed, 30 insertions(+), 26 deletions(-)
diff --git a/lib
---
rumpdisk/block-rump.c | 16 +---
rumpdisk/main.c | 5 -
2 files changed, 13 insertions(+), 8 deletions(-)
diff --git a/rumpdisk/block-rump.c b/rumpdisk/block-rump.c
index 3ee0773d..584056f7 100644
--- a/rumpdisk/block-rump.c
+++ b/rumpdisk/block-rump.c
@@ -27,6 +27,7 @@
Hi,
[PATCH 1/6] pci-arbiter: Fix brokenness
[PATCH 2/6] pci-arbiter: Introduce machdev into this server
First one or both patches could be applied as is to the tree,
and work with the new libpciaccess.
[PATCH 3/6] libnetfs: Hacks to allow pci to run with no cred
This is still needed to bypass
---
pci-arbiter/main.c | 16
1 file changed, 12 insertions(+), 4 deletions(-)
diff --git a/pci-arbiter/main.c b/pci-arbiter/main.c
index 4a10b264..ddea5aa7 100644
--- a/pci-arbiter/main.c
+++ b/pci-arbiter/main.c
@@ -62,7 +62,7 @@ pci_device_open (mach_port_t reply_port, mach_msg
---
libmachdev/ds_routines.c | 4 +--
libmachdev/machdev-device_emul.h | 2 +-
libmachdev/machdev.h | 2 +-
libmachdev/trivfs_server.c | 2 +-
pci-arbiter/Makefile | 2 +-
pci-arbiter/main.c | 19 -
pci-arbiter/startup-ops.c
[PATCH 1/2] pci-arbiter: Fix brokenness
This patch does not include mustbedir option because I believe there
is already logic that detects if it is looking up a directory,
by checking the entry->dir != 0.
When I tried adding it in, it did not work correctly.
[PATCH 2/2] pci-arbiter: Introduce ma
* pci-arbiter/netfs_impl.c
(get_dirents): Fix count and size detection
(netfs_attempt_lookup): Strip trailing slashes and leading dot slashes
* pci-arbiter/pci-ops.c
(check_permissions): Use the overridable method instead
---
pci-arbiter/netfs_impl.c | 41
* pci-arbiter/Makefile
Add machdev lib and simplify
* pci-arbiter/main.c
(pci_device_{open,close,shutdown}): New methods
(netfs_server_func): Thread the demuxer loop
(pcifs_startup): Custom startup method
(main): Use machdev for server and detach worker threads
* pci-arbiter/options.{c,h}
Add disk-
On 7/3/21 9:25 pm, Samuel Thibault wrote:
> Damien Zammit, le dim. 07 mars 2021 17:39:04 +1100, a ecrit:
>> [PATCH 3/6] libnetfs: Hacks to allow pci to run with no cred
>>
>> This is still needed to bypass the cred uid=0.
>> I tried to call fsys_getroot, but it failed
Fix count and size
---
acpi/netfs_impl.c | 41 +
1 file changed, 29 insertions(+), 12 deletions(-)
diff --git a/acpi/netfs_impl.c b/acpi/netfs_impl.c
index 84f52c89..d9853a08 100644
--- a/acpi/netfs_impl.c
+++ b/acpi/netfs_impl.c
@@ -58,25 +58,26 @@ get_di
* pci-arbiter/Makefile
Add machdev lib and simplify
* pci-arbiter/main.c
(pci_device_{open,close,shutdown}): New methods
(netfs_server_func): Thread the demuxer loop
(pcifs_startup): Custom startup method
(main): Use machdev for server and detach worker threads
* pci-arbiter/options.{c,h}
Add disk-
---
rumpdisk/block-rump.c | 16 +---
rumpdisk/main.c | 5 -
2 files changed, 13 insertions(+), 8 deletions(-)
diff --git a/rumpdisk/block-rump.c b/rumpdisk/block-rump.c
index 9c6fa10c..a61f7b80 100644
--- a/rumpdisk/block-rump.c
+++ b/rumpdisk/block-rump.c
@@ -27,6 +27,7 @@
Hi all,
This series introduces a working disk driver
with no more ugly netfs hacks.
Cheers,
Damien
---
pci-arbiter/netfs_impl.c | 39 ++-
pci-arbiter/pci-ops.c| 2 +-
2 files changed, 27 insertions(+), 14 deletions(-)
diff --git a/pci-arbiter/netfs_impl.c b/pci-arbiter/netfs_impl.c
index b987a0bc..2748c382 100644
--- a/pci-arbiter/netfs_impl.c
+++ b/pci
---
acpi/netfs_impl.c | 39 ++-
1 file changed, 26 insertions(+), 13 deletions(-)
diff --git a/acpi/netfs_impl.c b/acpi/netfs_impl.c
index 84f52c89..b6cb466d 100644
--- a/acpi/netfs_impl.c
+++ b/acpi/netfs_impl.c
@@ -58,25 +58,22 @@ get_dirents (struct acpifs_d
This set of changes allows the fsys_init flow of RPCs to work better
during bootstrap. It also removes duplicated code for the
startup shutdown notify between machdev and the arbiter.
However, I am still unsure why the lookup of _SERVERS_STARTUP fails
in the arbiter's fsys_init, see the WARNING a
This fixes the WARNING on the previous patch.
The lookup of _SERVERS_STARTUP now works.
---
libmachdev/trivfs_server.c | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/libmachdev/trivfs_server.c b/libmachdev/trivfs_server.c
index 8ee6be65..be5abc9d 100644
--- a/libmachdev
Hi,
This patch fixes the bootstrapping,
except for a strange anomaly with:
cat /proc/5/stat /proc/6/stat
My grub config looks like this:
menuentry "pci-arbiter + rumpdisk!" {
set root=(hd0,msdos2)
multiboot /boot/gnumach-1.8-486-dbg.gz root=part:2:device:wd0 noide
modul
* libmachdev/machdev.h
(machdev_trivfs_init): Add controlport output parameter
* libmachdev/trivfs_server.c
Rework the bootstrapping of machdev
* pci-arbiter/Makefile
Remove startup.c
* pci-arbiter/startup.{c,h}
Delete files
* pci-arbiter/main.c
(main): Rework bootstrap, remove notification, add co
* libmachdev/trivfs_server.c
Rework the bootstrapping of machdev
* pci-arbiter/Makefile
Remove startup.c
* pci-arbiter/startup.{c,h}
Delete files
* pci-arbiter/main.c
(main): Rework bootstrap, remove duplicated notification
* rumpdisk/main.c
(main): Don't pretend to be the arbiter anymore
---
lib
---
libmachdev/machdev.h | 2 +-
libmachdev/trivfs_server.c | 9 ++---
pci-arbiter/main.c | 2 +-
rumpdisk/main.c| 2 +-
4 files changed, 9 insertions(+), 6 deletions(-)
diff --git a/libmachdev/machdev.h b/libmachdev/machdev.h
index 8f613b35..9884c70d 100644
--- a/li
---
libmachdev/trivfs_server.c | 17 -
1 file changed, 12 insertions(+), 5 deletions(-)
diff --git a/libmachdev/trivfs_server.c b/libmachdev/trivfs_server.c
index 607166f9..60711348 100644
--- a/libmachdev/trivfs_server.c
+++ b/libmachdev/trivfs_server.c
@@ -79,6 +79,9 @@ struct p
Hi all,
As you know, AlmuHS did a bunch of great work for SMP support so far.
However, I have noticed there are a stack of commits in 0.0.10_2 branch
that are not rebased onto master.
Is it desirable/possible to squash all the smp support into one or two commits
so we can rebase it onto master a
)
diff --git a/i386/i386at/ioapic.c b/i386/i386at/ioapic.c
new file mode 100644
index ..72830653
--- /dev/null
+++ b/i386/i386at/ioapic.c
@@ -0,0 +1,320 @@
+/*
+ * Copyright (c) 2019 Damien Zammit
+ */
+
+#include
+#include
+#include
+#include
+#include
+#include
+#include
+#in
On 21/3/21 11:54 am, Samuel Thibault wrote:
> Damien Zammit, le dim. 21 mars 2021 11:48:29 +1100, a ecrit:
>> {cpu0} ../kern/slab.c:1021: kmem_cache_free_to_slab: Assertion `((unsigned
>> long)
>> buf + cache->buf_size) <= vm_page_trunc((unsigned long)slab->a
---
kern/mach_clock.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/kern/mach_clock.c b/kern/mach_clock.c
index e4d44f57..ed5cdfd3 100644
--- a/kern/mach_clock.c
+++ b/kern/mach_clock.c
@@ -152,7 +152,9 @@ void clock_interrupt(
timer_bump(&thread->user_timer,
---
i386/i386at/acpi_parse_apic.c | 25 +
1 file changed, 9 insertions(+), 16 deletions(-)
diff --git a/i386/i386at/acpi_parse_apic.c b/i386/i386at/acpi_parse_apic.c
index 7855e8a5..9dbc6825 100644
--- a/i386/i386at/acpi_parse_apic.c
+++ b/i386/i386at/acpi_parse_apic.c
@@
This patchset provides a way to disable linux/ drivers and build
gnumach with basic IOAPIC interrupts and LAPIC timer support in SMP
mode.
You can still build gnumach the old way with PIC interrupts and single
core, with or without linux/ drivers.
Note that it currently assumes there is only one
Defaults to --enable-apic
To build with SMP and APIC support, use for example:
--enable-ncpus=4
Otherwise to build with PIC only and no SMP:
--enable-apic=no
---
i386/Makefrag.am | 13 +-
i386/configfrag.ac | 12 ++
i386/i386/apic.c | 34 ++-
i3
Defaults to build with linux drivers.
To build without linux drivers, use:
--enable-qemu-group=no --enable-net-group=no
--enable-scsi-group=no --enable-wireless-group=no --enable-pcmcia-group=no
---
device/intr.c | 36
linux/configfrag.ac | 38 +
---
i386/i386at/acpi_parse_apic.c | 25 +
1 file changed, 9 insertions(+), 16 deletions(-)
diff --git a/i386/i386at/acpi_parse_apic.c b/i386/i386at/acpi_parse_apic.c
index 7855e8a5..9dbc6825 100644
--- a/i386/i386at/acpi_parse_apic.c
+++ b/i386/i386at/acpi_parse_apic.c
@@
---
kern/mach_clock.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/kern/mach_clock.c b/kern/mach_clock.c
index e4d44f57..ed5cdfd3 100644
--- a/kern/mach_clock.c
+++ b/kern/mach_clock.c
@@ -152,7 +152,9 @@ void clock_interrupt(
timer_bump(&thread->user_timer,
---
linux/dev/arch/i386/kernel/irq.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/linux/dev/arch/i386/kernel/irq.c b/linux/dev/arch/i386/kernel/irq.c
index 522ed0ac..06889e58 100644
--- a/linux/dev/arch/i386/kernel/irq.c
+++ b/linux/dev/arch/i386/kernel/irq.c
@@ -743,7 +743,
On 26/3/21 8:59 pm, Samuel Thibault wrote:
> Damien Zammit, le ven. 26 mars 2021 20:48:45 +1100, a ecrit:
>> As discussed, this should be done via a new pci-arbiter RPC that
>> requests the IRQ id for a particular device.
>
> Mmm, AIUI a driver can already read the PCI
I have split the patches up into logical pieces.
Perhaps all can be merged except the last patch until we fix
the selection of apic via kernel cmdline (?)
It now almost boots when you give it the right IRQ.
I am not sure why it hangs just before "proc" is supposed to appear,
but it probes and find
---
i386/i386at/acpi_parse_apic.c | 32 ++--
1 file changed, 14 insertions(+), 18 deletions(-)
diff --git a/i386/i386at/acpi_parse_apic.c b/i386/i386at/acpi_parse_apic.c
index 7855e8a5..36e2805d 100644
--- a/i386/i386at/acpi_parse_apic.c
+++ b/i386/i386at/acpi_parse_ap
---
i386/i386/apic.c | 35 +++-
i386/i386/apic.h | 75 +--
i386/i386at/acpi_parse_apic.c | 15 ---
i386/i386at/acpi_parse_apic.h | 2 +-
4 files changed, 106 insertions(+), 21 deletions(-)
diff --git a/i386/i386/apic.c b/i
Defaults to build with linux drivers.
To build without linux drivers, use:
--enable-block-group=no --enable-net-group=no
--enable-scsi-group=no --enable-wireless-group=no --enable-pcmcia-group=no
---
device/intr.c | 91 +
linux/configfrag.ac | 32
---
i386/i386/locore.S | 15 +++
i386/i386/pic.h| 4 +++-
i386/i386at/idt.h | 17 +++--
i386/i386at/int_init.c | 15 +--
4 files changed, 42 insertions(+), 9 deletions(-)
diff --git a/i386/i386/locore.S b/i386/i386/locore.S
index bee3630c..8a1
---
i386/i386at/acpi_parse_apic.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/i386/i386at/acpi_parse_apic.c b/i386/i386at/acpi_parse_apic.c
index 36e2805d..af70934d 100644
--- a/i386/i386at/acpi_parse_apic.c
+++ b/i386/i386at/acpi_parse_apic.c
@@ -280,7 +280,7 @@ acpi_get_a
1 - 100 of 684 matches
Mail list logo
