Suspend-to-full-encrypted-disk is broken (by KARL?)

Probably something in the bootloader?

Yes, exactly. The bootloader handles the selection of /bsd.booted

Please create a transcript of your default boot. You should see a
'%' character next to your boot disk. If not, update your bootblocks.

I've heard a few reports of failure, but people haven't been exact about
tracking down where the failures comes from.

CVSROOT: /cvs
Module name: src
Changes by: ***@cvs.openbsd.org 2017/06/21 19:26:28

Modified files:
sys/arch/i386/stand/libsa: diskprobe.c
sys/arch/amd64/stand/libsa: diskprobe.c

Log message:
Only print the magic '&' letter on a disk if the hibernate partition is
valid. This is an undocumented little debugging diagnostic, much like
the '*'.

Natasha Kerensikova

2017-07-18 19:01:55 UTC

Hello,

Post by Theo de Raadt
Please create a transcript of your default boot.

Here it is, handcopied but hopefully without typo. I mistyped the
passphrase on the first attempt, because muscle memory was impaired by
the camera, but I'm showing the text as it appears without tampering. If
you need a transcript with the correct passphrase on the first try, I
can try to record another one.

Using drive 0, partition 3.
Loading......
probing: pc0 mem[630K 511M 510M 2471M 486M a20=on]
disk: hd0+ sr0*

OpenBSD/amd64 BOOT 3.33

Passphrase:
incorrect passphrase or keydisk
open(sr0a:/etc/boot.conf): Operation not permitted
boot>
Passphrase:
booting sr0a:/bsd: 7782584+2319368+264992+0+675840 [72+751104+512939]=0bbec78
entry point at 0x1000158
( using 1264760 bytes of bsd ELF symbol table )
Copyright (c) 1982, 1986, 1989, 1991, 1993
The Regents of the University of California. All rights reserved.
Copyright (c) 1995-2017 OpenBSD. All rights reserved. https://www.OpenBSD.org

OpenBSD 6.1-current (GENERIC.MP) #94: Mon Jul 10 18:20:35 MDT 2017
***@amd64.openbsd.org:/usr/src/sys/arch/amd64/compile/GENERIC.MP

and I believe the rest is accounted for in the dmesg.

Post by Theo de Raadt
You should see a
'%' character next to your boot disk. If not, update your bootblocks.

I updated the whole system by downloading the snapshot bsd.rd,
checking the signature, booting on it, and selecting (U)pgrade. I have
been told it does update the bootblocks.

I ran `installboot -v sd1` anyway, and got exactly the same behavior
with the same transcript as above.

Sorry for having made you yell,
Natasha

Sebastien Marie

2017-07-18 17:01:19 UTC

Hi,

(First sorry for the thread break, I have some problems in receiving
mails actually.)

I dig a bit inside the code of bootloader, and I think it is possible
that the bootloader is unable to see hibernate signature when FDE is in
use.

Taking boot(bootdev) as entrypoint (stand/boot/boot.c +60):

it calls machdep() (arch/amd64/stand/libsa/machdep.c +48)
and call each probe routines.

In the list of probe routines, there is diskprobe()
(arch/amd64/stand/libsa/diskprobe.c +246), which call:

- hardprobe() (line 122),
- and only after sr_probe() (line 270)

It is hardprobe() that will call check_hibernate():

- allocating dip struct
- running check_hibernate() to set BDI_HIBVALID if HIBERNATE_MAGIC is
found.

To read the value, it uses dip->strategy() function.

arch/amd64/stand/libsa/diskprobe.c
491 error = dip->strategy(dip, F_READ, (daddr32_t)sec, sizeof hib, &hib, NULL);
492 if (error == 0 && hib.magic == HIBERNATE_MAGIC) {
493 /* Hibernate present */
494 dip->bios_info.flags |= BDI_HIBVALID;
495 printf("&");
496 }

the actual code is biosstrategy() (arch/amd64/stand/libsa/biosdev.c +722)
which will call sr_strategy() if dip->sr_vol is not NULL.

arch/amd64/stand/libsa/biosdev.c
721 /* Intercept strategy for softraid volumes. */
722 if (dip->sr_vol)
723 return sr_strategy(dip->sr_vol, rw, blk, size, buf, rsize);

If I correctly understood the code, dip->sr_vol is initialized on first
biosopen() call. But at this point, we didn't call it. Additionally,
sr_probe() isn't called either.

So dip->strategy() will read the encrypted data and not decrypted one,
so it is unable to found HIBERNATE_MAGIC (as it is encrypted).

Thanks.

--
Sebastien Marie

Theo de Raadt

2017-07-18 17:10:00 UTC

Possibly.

cc'ing yasuoka who helped me with this method of disk probing.

But anyways, beyond a code review we need someone to SHOW THEIR BOOT LOADER
OUTPUT, so that we can see the hibernate marker letter is missing.

Post by Sebastien Marie
(First sorry for the thread break, I have some problems in receiving
mails actually.)
I dig a bit inside the code of bootloader, and I think it is possible
that the bootloader is unable to see hibernate signature when FDE is in
use.
it calls machdep() (arch/amd64/stand/libsa/machdep.c +48)
and call each probe routines.
In the list of probe routines, there is diskprobe()
- hardprobe() (line 122),
- and only after sr_probe() (line 270)
- allocating dip struct
- running check_hibernate() to set BDI_HIBVALID if HIBERNATE_MAGIC is
found.
To read the value, it uses dip->strategy() function.
arch/amd64/stand/libsa/diskprobe.c
491 error = dip->strategy(dip, F_READ, (daddr32_t)sec, sizeof hib, &hib, NULL);
492 if (error == 0 && hib.magic == HIBERNATE_MAGIC) {
493 /* Hibernate present */
494 dip->bios_info.flags |= BDI_HIBVALID;
495 printf("&");
496 }
the actual code is biosstrategy() (arch/amd64/stand/libsa/biosdev.c +722)
which will call sr_strategy() if dip->sr_vol is not NULL.
arch/amd64/stand/libsa/biosdev.c
721 /* Intercept strategy for softraid volumes. */
722 if (dip->sr_vol)
723 return sr_strategy(dip->sr_vol, rw, blk, size, buf, rsize);
If I correctly understood the code, dip->sr_vol is initialized on first
biosopen() call. But at this point, we didn't call it. Additionally,
sr_probe() isn't called either.
So dip->strategy() will read the encrypted data and not decrypted one,
so it is unable to found HIBERNATE_MAGIC (as it is encrypted).
Thanks.
--
Sebastien Marie

Theo Buehler

2017-07-18 18:40:53 UTC

Post by Theo de Raadt
Possibly.
cc'ing yasuoka who helped me with this method of disk probing.
But anyways, beyond a code review we need someone to SHOW THEIR BOOT LOADER
OUTPUT, so that we can see the hibernate marker letter is missing.

Using FDE from MBR on my T420, I can confirm Natasha's findings.

Automatic kernel selection to unhibernate fails, but manually selecting
/bsd.booted works fine.

I see the exact same thing on reboot and on boot into unhibernate:
(manually transcribed, no & sign)

Using drive 0, partition 3.
Loading......
probing: pc0 mem[630K 511M 510M 2471M 486M a20=on]
disk: hd0+ sr0*

OpenBSD/amd64 BOOT 3.33

Passphrase: _

Sebastien Marie

2017-07-18 19:15:54 UTC

Using FDE from MBR on my T420, I can confirm Natasha's findings.
Automatic kernel selection to unhibernate fails, but manually selecting
/bsd.booted works fine.
(manually transcribed, no & sign)
Using drive 0, partition 3.
Loading......
probing: pc0 mem[630K 511M 510M 2471M 486M a20=on]
disk: hd0+ sr0*

OpenBSD/amd64 BOOT 3.33

Passphrase: _

To compare, here my output on suspended boot for a laptop (i386)
*without* FDE:

Using drive 0, partition 3.
Loading......
probing: pc0 pci mem[636K 2037M a20=on]
disk: hd0+&

Post by Theo de Raadt
OpenBSD/i386 BOOT 3.31

unhibernate detected: switching to /bsd.booted
boot>
booting hd0a:/bsd.booted: 8122376+....
entry point at 0x2000d4
...

So here the & is present, and an explicit line is echoed.

--
Sebastien Marie

Theo de Raadt

2017-07-18 20:56:45 UTC

Post by Natasha Kerensikova

Post by Theo de Raadt
Please create a transcript of your default boot.

Here it is, handcopied but hopefully without typo. I mistyped the
passphrase on the first attempt, because muscle memory was impaired by
the camera, but I'm showing the text as it appears without tampering. If
you need a transcript with the correct passphrase on the first try, I
can try to record another one.
Using drive 0, partition 3.
Loading......
probing: pc0 mem[630K 511M 510M 2471M 486M a20=on]
disk: hd0+ sr0*

OpenBSD/amd64 BOOT 3.33

incorrect passphrase or keydisk
open(sr0a:/etc/boot.conf): Operation not permitted
boot>
booting sr0a:/bsd: 7782584+2319368+264992+0+675840 [72+751104+512939]=0bbec78

So it looks like the problem has been identified, and hibernate signature
analysis will need to be deferred to a slightly later phase.

That is pretty difficult to do right.

Theo de Raadt

2017-07-18 21:48:48 UTC

Post by Sebastien Marie

OpenBSD/amd64 BOOT 3.33

Passphrase: _

To compare, here my output on suspended boot for a laptop (i386)
Using drive 0, partition 3.
Loading......
probing: pc0 pci mem[636K 2037M a20=on]
disk: hd0+&

Post by Theo de Raadt
OpenBSD/i386 BOOT 3.31

unhibernate detected: switching to /bsd.booted
boot>
booting hd0a:/bsd.booted: 8122376+....
entry point at 0x2000d4
...
So here the & is present, and an explicit line is echoed.

I'll explain what is going on here, and what needs to probably be done.

The changes introduced into the bootloader are largely correct. They
make the right decisions. They just don't make them at the right
points in time. The IO operations should be moved further forward, to
after the key operations.

Havinge done it this way, yasuoka and I were able to create a workable
model, but the operations need to be moved forward.

Someone want to dive in?

YASUOKA Masahiko

2017-07-20 12:45:55 UTC

This post might be inappropriate. Click to display it.

Theo Buehler

2017-07-20 18:44:53 UTC

Post by YASUOKA Masahiko
On Tue, 18 Jul 2017 15:48:48 -0600

Post by Sebastien Marie

OpenBSD/amd64 BOOT 3.33

Passphrase: _

To compare, here my output on suspended boot for a laptop (i386)
Using drive 0, partition 3.
Loading......
probing: pc0 pci mem[636K 2037M a20=on]
disk: hd0+&

Post by Theo de Raadt
OpenBSD/i386 BOOT 3.31

unhibernate detected: switching to /bsd.booted
boot>
booting hd0a:/bsd.booted: 8122376+....
entry point at 0x2000d4
...
So here the & is present, and an explicit line is echoed.

I'll explain what is going on here, and what needs to probably be done.
The changes introduced into the bootloader are largely correct. They
make the right decisions. They just don't make them at the right
points in time. The IO operations should be moved further forward, to
after the key operations.
Havinge done it this way, yasuoka and I were able to create a workable
model, but the operations need to be moved forward.
Someone want to dive in?

The diff is to support selecting bsd.booted for FED.
disk: hd0 sr0*

Post by Sebastien Marie
OpenBSD/amd64 BOOTX64 3.33

Passphrase: <--- Enter the passphrase
unhibernate detected: switching to /bsd.booted
boot>
booting sr0a:/bsd.booted: xxxx...
we can't add "&" sign on the "disk" line. Since the bootloader gets
the passpharase after that line.
The boot codes other than BOOTX64.EFI are not tested yet.

I tried to pxeboot into bsd.rd to install a snapshot with your diff
(with FDE on MBR) and this leads to a hang after password entry.
Therefore I didn't dare to install the bootloader, as I need this
machine for work.

More precisely, it looks like check_hibernate() doesn't return on a
normally rebooted machine:

probing: pc0 pxe![2.1] mem[630K 511M 518M 2471M 486M a20=on]
disk: hd0+ sr0*
net: mac aa:bb:cc:dd:ee:ff, ip 192.168.10.11, server 192.168.10.10

Post by YASUOKA Masahiko

Post by Theo de Raadt
OpenBSD/amd64 PXEBOOT 3.28

Passphrase:
start check_hibernate()
_

and the machine hangs here.

I used the following simple modification of your diff:

Index: libsa/biosdev.c
===================================================================
RCS file: /var/cvs/src/sys/arch/amd64/stand/libsa/biosdev.c,v
retrieving revision 1.30
diff -u -p -r1.30 biosdev.c
--- libsa/biosdev.c 18 Sep 2016 15:13:10 -0000 1.30
+++ libsa/biosdev.c 20 Jul 2017 17:21:05 -0000
@@ -548,6 +548,9 @@ biosopen(struct open_file *f, ...)
if (sr_getdisklabel(bv, &dip->disklabel))
return ERDLAB;
dip->bios_info.flags &= ~BDI_BADLABEL;
+ printf("start check_hibernate()\n");
+ check_hibernate(dip);
+ printf("stop check_hibernate()\n");
}

bv->sbv_part = part + 'a';

Theo Buehler

2017-07-20 21:26:57 UTC

Post by YASUOKA Masahiko
On Tue, 18 Jul 2017 15:48:48 -0600

Post by Sebastien Marie

OpenBSD/amd64 BOOT 3.33

Passphrase: _

To compare, here my output on suspended boot for a laptop (i386)
Using drive 0, partition 3.
Loading......
probing: pc0 pci mem[636K 2037M a20=on]
disk: hd0+&

Post by Theo de Raadt
OpenBSD/i386 BOOT 3.31

unhibernate detected: switching to /bsd.booted
boot>
booting hd0a:/bsd.booted: 8122376+....
entry point at 0x2000d4
...
So here the & is present, and an explicit line is echoed.

I'll explain what is going on here, and what needs to probably be done.
The changes introduced into the bootloader are largely correct. They
make the right decisions. They just don't make them at the right
points in time. The IO operations should be moved further forward, to
after the key operations.
Havinge done it this way, yasuoka and I were able to create a workable
model, but the operations need to be moved forward.
Someone want to dive in?

The diff is to support selecting bsd.booted for FED.
disk: hd0 sr0*

Post by Sebastien Marie
OpenBSD/amd64 BOOTX64 3.33

I tried to pxeboot into bsd.rd to install a snapshot with your diff
(with FDE on MBR) and this leads to a hang after password entry.
Therefore I didn't dare to install the bootloader, as I need this
machine for work.
More precisely, it looks like check_hibernate() doesn't return on a
probing: pc0 pxe![2.1] mem[630K 511M 518M 2471M 486M a20=on]
disk: hd0+ sr0*
net: mac aa:bb:cc:dd:ee:ff, ip 192.168.10.11, server 192.168.10.10

Post by YASUOKA Masahiko

Post by Theo de Raadt
OpenBSD/amd64 PXEBOOT 3.28

start check_hibernate()
_
and the machine hangs here.

Unfortunately, regular boot is also broken the same way as PXE boot with
this diff. I made a fresh FDE on MBR install to a memory stick. After
the install I rebooted from the memory stick, boot decrypts the volume
but hangs in check_hibernate().

(with the same diff as in my previous mail):

Using drive 0, partition 3
Loading......
probing: pc0 mem[630K 511M 518M 2471M 486M a20=on]
disk hd0+ hd1+ sr0* sr1*

Post by YASUOKA Masahiko
OpenBSD/amd64 BOOT 3.33

Passphrase:
start check_hibernate()

Post by Theo Buehler
Index: libsa/biosdev.c
===================================================================
RCS file: /var/cvs/src/sys/arch/amd64/stand/libsa/biosdev.c,v
retrieving revision 1.30
diff -u -p -r1.30 biosdev.c
--- libsa/biosdev.c 18 Sep 2016 15:13:10 -0000 1.30
+++ libsa/biosdev.c 20 Jul 2017 17:21:05 -0000
@@ -548,6 +548,9 @@ biosopen(struct open_file *f, ...)
if (sr_getdisklabel(bv, &dip->disklabel))
return ERDLAB;
dip->bios_info.flags &= ~BDI_BADLABEL;
+ printf("start check_hibernate()\n");
+ check_hibernate(dip);
+ printf("stop check_hibernate()\n");
}
bv->sbv_part = part + 'a';

YASUOKA Masahiko

2017-07-20 23:41:04 UTC

Thank you for the test. Let me update the diff.

On Thu, 20 Jul 2017 23:26:57 +0200

Post by Theo Buehler
I tried to pxeboot into bsd.rd to install a snapshot with your diff
(with FDE on MBR) and this leads to a hang after password entry.
Therefore I didn't dare to install the bootloader, as I need this
machine for work.
More precisely, it looks like check_hibernate() doesn't return on a
probing: pc0 pxe![2.1] mem[630K 511M 518M 2471M 486M a20=on]
disk: hd0+ sr0*
net: mac aa:bb:cc:dd:ee:ff, ip 192.168.10.11, server 192.168.10.10

Post by Theo de Raadt
OpenBSD/amd64 PXEBOOT 3.28

start check_hibernate()
_
and the machine hangs here.

Unfortunately, regular boot is also broken the same way as PXE boot with
this diff. I made a fresh FDE on MBR install to a memory stick. After
the install I rebooted from the memory stick, boot decrypts the volume
but hangs in check_hibernate().
Using drive 0, partition 3
Loading......
probing: pc0 mem[630K 511M 518M 2471M 486M a20=on]
disk hd0+ hd1+ sr0* sr1*

OpenBSD/amd64 BOOT 3.33

start check_hibernate()

Index: sys/arch/amd64/stand/efiboot/efidev.c
===================================================================
RCS file: /cvs/src/sys/arch/amd64/stand/efiboot/efidev.c,v
retrieving revision 1.26
diff -u -p -r1.26 efidev.c
--- sys/arch/amd64/stand/efiboot/efidev.c 16 May 2017 02:56:23 -0000 1.26
+++ sys/arch/amd64/stand/efiboot/efidev.c 20 Jul 2017 23:37:58 -0000
@@ -625,6 +625,7 @@ efiopen(struct open_file *f, ...)
if (sr_getdisklabel(bv, &dip->disklabel))
return ERDLAB;
dip->bios_info.flags &= ~BDI_BADLABEL;
+ check_hibernate(dip);
}

bv->sbv_part = part + 'a';
Index: sys/arch/amd64/stand/libsa/biosdev.c
===================================================================
RCS file: /cvs/src/sys/arch/amd64/stand/libsa/biosdev.c,v
retrieving revision 1.30
diff -u -p -r1.30 biosdev.c
--- sys/arch/amd64/stand/libsa/biosdev.c 18 Sep 2016 15:13:10 -0000 1.30
+++ sys/arch/amd64/stand/libsa/biosdev.c 20 Jul 2017 23:37:58 -0000
@@ -535,6 +535,7 @@ biosopen(struct open_file *f, ...)
if (bv->sbv_diskinfo == NULL) {
dip = alloc(sizeof(struct diskinfo));
bzero(dip, sizeof(*dip));
+ dip->strategy = biosstrategy;
bv->sbv_diskinfo = dip;
dip->sr_vol = bv;
dip->bios_info.flags |= BDI_BADLABEL;
@@ -548,6 +549,7 @@ biosopen(struct open_file *f, ...)
if (sr_getdisklabel(bv, &dip->disklabel))
return ERDLAB;
dip->bios_info.flags &= ~BDI_BADLABEL;
+ check_hibernate(dip);
}

bv->sbv_part = part + 'a';
Index: sys/arch/amd64/stand/libsa/disk.h
===================================================================
RCS file: /cvs/src/sys/arch/amd64/stand/libsa/disk.h,v
retrieving revision 1.5
diff -u -p -r1.5 disk.h
--- sys/arch/amd64/stand/libsa/disk.h 2 Sep 2015 01:52:26 -0000 1.5
+++ sys/arch/amd64/stand/libsa/disk.h 20 Jul 2017 23:37:58 -0000
@@ -59,4 +59,6 @@ extern struct disklist_lh disklist;

void dump_diskinfo(void);

+void check_hibernate(struct diskinfo *);
+
#endif /* _DISK_H */
Index: sys/arch/amd64/stand/libsa/diskprobe.c
===================================================================
RCS file: /cvs/src/sys/arch/amd64/stand/libsa/diskprobe.c,v
retrieving revision 1.19
diff -u -p -r1.19 diskprobe.c
--- sys/arch/amd64/stand/libsa/diskprobe.c 22 Jun 2017 01:26:28 -0000 1.19
+++ sys/arch/amd64/stand/libsa/diskprobe.c 20 Jul 2017 23:37:58 -0000
@@ -55,7 +55,6 @@

/* Local Prototypes */
static int disksum(int);
-static void check_hibernate(struct diskinfo *);

int bootdev_has_hibernate(void); /* export for loadfile() */

@@ -176,6 +175,8 @@ hardprobe(void)
dip->bsddev = dip->bios_info.bsd_dev =
MAKEBOOTDEV(type, 0, 0, bsdunit, RAW_PART);
check_hibernate(dip);
+ if (dip->bios_info.flags & BDI_HIBVALID)
+ printf("&");

/* Add to queue of disks */
TAILQ_INSERT_TAIL(&disklist, dip, list);
@@ -232,6 +233,8 @@ efi_hardprobe(void)
dip->bsddev = dip->bios_info.bsd_dev =
MAKEBOOTDEV(type, 0, 0, bsdunit, RAW_PART);
check_hibernate(dip);
+ if (dip->bios_info.flags & BDI_HIBVALID)
+ printf("&");

/* Add to queue of disks */
TAILQ_INSERT_TAIL(&disklist, dip, list);
@@ -472,7 +475,7 @@ bootdev_has_hibernate(void)
return ((bootdev_dip->bios_info.flags & BDI_HIBVALID)? 1 : 0);
}

-static void
+void
check_hibernate(struct diskinfo *dip)
{
daddr_t sec;
@@ -489,9 +492,6 @@ check_hibernate(struct diskinfo *dip)
(sizeof(union hibernate_info) / DEV_BSIZE);

error = dip->strategy(dip, F_READ, (daddr32_t)sec, sizeof hib, &hib, NULL);
- if (error == 0 && hib.magic == HIBERNATE_MAGIC) {
- /* Hibernate present */
- dip->bios_info.flags |= BDI_HIBVALID;
- printf("&");
- }
+ if (error == 0 && hib.magic == HIBERNATE_MAGIC)
+ dip->bios_info.flags |= BDI_HIBVALID; /* Hibernate present */
}
Index: sys/arch/i386/stand/libsa/biosdev.c
===================================================================
RCS file: /cvs/src/sys/arch/i386/stand/libsa/biosdev.c,v
retrieving revision 1.95
diff -u -p -r1.95 biosdev.c
--- sys/arch/i386/stand/libsa/biosdev.c 18 Sep 2016 15:13:10 -0000 1.95
+++ sys/arch/i386/stand/libsa/biosdev.c 20 Jul 2017 23:37:58 -0000
@@ -536,6 +536,7 @@ biosopen(struct open_file *f, ...)
if (bv->sbv_diskinfo == NULL) {
dip = alloc(sizeof(struct diskinfo));
bzero(dip, sizeof(*dip));
+ dip->strategy = biosstrategy;
bv->sbv_diskinfo = dip;
dip->sr_vol = bv;
dip->bios_info.flags |= BDI_BADLABEL;
@@ -549,6 +550,7 @@ biosopen(struct open_file *f, ...)
if (sr_getdisklabel(bv, &dip->disklabel))
return ERDLAB;
dip->bios_info.flags &= ~BDI_BADLABEL;
+ check_hibernate(dip);
}

bv->sbv_part = part + 'a';
Index: sys/arch/i386/stand/libsa/disk.h
===================================================================
RCS file: /cvs/src/sys/arch/i386/stand/libsa/disk.h,v
retrieving revision 1.6
diff -u -p -r1.6 disk.h
--- sys/arch/i386/stand/libsa/disk.h 2 Sep 2015 04:09:24 -0000 1.6
+++ sys/arch/i386/stand/libsa/disk.h 20 Jul 2017 23:37:58 -0000
@@ -59,4 +59,6 @@ extern struct disklist_lh disklist;

void dump_diskinfo(void);

+void check_hibernate(struct diskinfo *);
+
#endif /* _DISK_H */
Index: sys/arch/i386/stand/libsa/diskprobe.c
===================================================================
RCS file: /cvs/src/sys/arch/i386/stand/libsa/diskprobe.c,v
retrieving revision 1.43
diff -u -p -r1.43 diskprobe.c
--- sys/arch/i386/stand/libsa/diskprobe.c 22 Jun 2017 01:26:28 -0000 1.43
+++ sys/arch/i386/stand/libsa/diskprobe.c 20 Jul 2017 23:37:58 -0000
@@ -55,7 +55,6 @@

/* Local Prototypes */
static int disksum(int);
-static void check_hibernate(struct diskinfo *);

int bootdev_has_hibernate(void); /* export for loadfile() */

@@ -176,6 +175,8 @@ hardprobe(void)
dip->bsddev = dip->bios_info.bsd_dev =
MAKEBOOTDEV(type, 0, 0, bsdunit, RAW_PART);
check_hibernate(dip);
+ if (dip->bios_info.flags & BDI_HIBVALID)
+ printf("&");

/* Add to queue of disks */
TAILQ_INSERT_TAIL(&disklist, dip, list);
@@ -232,6 +233,8 @@ efi_hardprobe(void)
dip->bsddev = dip->bios_info.bsd_dev =
MAKEBOOTDEV(type, 0, 0, bsdunit, RAW_PART);
check_hibernate(dip);
+ if (dip->bios_info.flags & BDI_HIBVALID)
+ printf("&");

/* Add to queue of disks */
TAILQ_INSERT_TAIL(&disklist, dip, list);
@@ -472,7 +475,7 @@ bootdev_has_hibernate(void)
return ((bootdev_dip->bios_info.flags & BDI_HIBVALID)? 1 : 0);
}

-static void
+void
check_hibernate(struct diskinfo *dip)
{
daddr_t sec;
@@ -489,9 +492,6 @@ check_hibernate(struct diskinfo *dip)
(sizeof(union hibernate_info) / DEV_BSIZE);

error = dip->strategy(dip, F_READ, (daddr32_t)sec, sizeof hib, &hib, NULL);
- if (error == 0 && hib.magic == HIBERNATE_MAGIC) {
- /* Hibernate present */
- dip->bios_info.flags |= BDI_HIBVALID;
- printf("&");
- }
+ if (error == 0 && hib.magic == HIBERNATE_MAGIC)
+ dip->bios_info.flags |= BDI_HIBVALID; /* Hibernate present */
}

YASUOKA Masahiko

2017-07-20 23:44:06 UTC

FYI,

On Fri, 21 Jul 2017 08:41:04 +0900 (JST)

Post by YASUOKA Masahiko
Thank you for the test. Let me update the diff.

The previous diff lacks

Post by YASUOKA Masahiko
Index: sys/arch/i386/stand/libsa/biosdev.c
===================================================================
RCS file: /cvs/src/sys/arch/i386/stand/libsa/biosdev.c,v
retrieving revision 1.95
diff -u -p -r1.95 biosdev.c
--- sys/arch/i386/stand/libsa/biosdev.c 18 Sep 2016 15:13:10 -0000 1.95
+++ sys/arch/i386/stand/libsa/biosdev.c 20 Jul 2017 23:37:58 -0000
@@ -536,6 +536,7 @@ biosopen(struct open_file *f, ...)
if (bv->sbv_diskinfo == NULL) {
dip = alloc(sizeof(struct diskinfo));
bzero(dip, sizeof(*dip));
+ dip->strategy = biosstrategy;
bv->sbv_diskinfo = dip;
dip->sr_vol = bv;
dip->bios_info.flags |= BDI_BADLABEL;

this initialization of dip ->strategy

--yasuoka

Theo Buehler

2017-07-21 02:08:55 UTC