Commit Graph

952 Commits

Author SHA1 Message Date
Toomas Soome
21da9f14f6 loader: zfs.c is missing malloc checks, fix it
malloc() can return NULL, we need to check the return value.
2019-11-03 14:36:16 +00:00
Toomas Soome
0c0a882c7a loader: we do not support booting from pool with log device
If pool has log device, stop there and tell about it.
2019-11-03 13:25:47 +00:00
Toomas Soome
f4ed004573 loader: should check malloc in zfs_dev_open
malloc can return NULL.
2019-11-03 13:03:47 +00:00
Toomas Soome
abca0bd501 loader: calculate physical vdev psize from asize
Since physical device asize is calculated from psize and the asize is stored
in pool label, we can use asize to set the value of psize, which is used to
calculate the location of the pool labels.

MFC after:	1 week
2019-11-03 11:09:06 +00:00
Toomas Soome
bcbb1e60a1 loader: userboot/test should accept more than one disk
allow to specify multiple -d options, test -d disk1 -d disk2 ..
2019-11-03 09:14:29 +00:00
Toomas Soome
e499793e76 Remove duplicate lz4 implementations
Port illumos change: https://www.illumos.org/issues/11667

Move lz4.c out of zfs tree to opensolaris/common/lz4, adjust it to be
usable from kernel/stand/userland builds, so we can use just one single
source. Add lz4.h to declare lz4_compress() and lz4_decompress().

MFC after:	1 month
Differential Revision:	https://reviews.freebsd.org/D22037
2019-11-02 12:28:04 +00:00
Toomas Soome
0060947db1 loader: libi386/comconsole.c cstyle cleanup
Only cstyle, no functional changes.
2019-11-02 10:53:23 +00:00
Toomas Soome
b9f745fdb1 loader: fall back to term_emu on efi console with serial backend
In case of efi console having serial backend (video + serial or only serial),
we need to stick with old emulator till we can draw console.

Eventually we would need to get console terminal emulator to be removed
from serial console because the serial link already has the terminal.

However, we need to implement comconsole on all efi platforms first, then
we need the ability to draw console, so we do not have to use SimpleTextOutput
protocol (which will write both on video and serial in case of multiplexed
ComOut).

Differential Revision:	https://reviews.freebsd.org/D22161
2019-11-02 09:50:36 +00:00
Kyle Evans
bac5966e2d lualoader: rewrite try_include using lfs + dofile
Actual modules get require()'d in, rather than try_include(). All instances
of try_include should be provided with proper hooks/API in the rest of
loader to do the work they need to do, since we can't rely on them to exist.
Convert this now to lfs + dofile since we won't really be treating them as
modules.

lfs is required because dofile will properly throw an error if the file
doesn't exist, which is not in the spirit of 'optionally included'.

Getting out of the pcall game allows us to provide a loader.exit() style
call that backs out to the common bits of loader (autoboot sequence unless
disabled with a loader.setenv("autoboot_delay", "NO")). The most ideal way
identified so far to implement loader.exit() is to throw a special
abort-style error that indicates to the caller in interp_lua that we've not
actually errored out, just continue execution. Otherwise, we have to hack in
logic to bubble up and return from loader.lua without continuing further,
which gets kind of ugly depending on the context in which we're aborting.

A compat shim is provided temporarily in case the executing loader doesn't
yet have loader.lua_path, which was just added in r354246.
2019-11-02 04:01:39 +00:00
Kyle Evans
63172bf68b liblua: add loader.lua_path
As described previously, loader.lua_path is absolute path where scripts are
installed. A future commit will use this to build paths for dofile in
try_include, rather than the current pcall/require setup that makes it more
difficult to coordinate loader aborts from local.lua -- we do not need the
flexibility of require(), and local.lua is in-fact not a 'module-like' file
as we will not be referencing anything from it.
2019-11-02 03:41:30 +00:00
Kyle Evans
ee74c23624 stand: consolidate knowledge of lua path
Multiple places coordinate to 'know' where lua scripts are installed. Knock
this down to being formally defined (and overridable) in exactly one spot,
defs.mk, and spread the knowledge to loaders and liblua alike. A future
commit will expose this to lua as loader.lua_path, so it can build absolute
paths to lua scripts as needed.

MFC after:	1 week
2019-11-02 03:37:58 +00:00
Warner Losh
08b86a1a8a We don't support configuring serial PCI cards in EFI. Make this clearer in the
source rather than obfuscaring it behind NO_PCI (nothing else declares that,
so it's not making the ifdefs clearer).
2019-11-01 21:26:43 +00:00
Toomas Soome
98e28b71b2 loader: asprinf does crash arm64 due to missing NULL pointer check
PCHAR macro needs to check if d is NULL.

MFC after:	3 days
2019-11-01 06:54:07 +00:00
Toomas Soome
4efce32a51 loader: rs_alloc() may return NULL
rs_alloc() in zfs reader code may return NULL, so we need to check the return value and error out if needed.
MFC after:	1 week
2019-10-26 18:29:02 +00:00
Simon J. Gerraty
21686d9e64 Guard reference to x86_hypervisor
Obviously we only want x86_hypervisor on x86
Guard reference to x86_hypervisor

Obviously we only want x86_hypervisor on x86
2019-10-24 22:26:06 +00:00
Simon J. Gerraty
e9b148a318 Add support for hypervisor check on x86
Add ficl words for isvirtualized
and move ficl inb and outb words to ficl/x86/sysdep.c
so can be shared by i386 and amd64

Reviewed by:	imp bdrewery
MFC after:	1 week
Sponsored by:	Juniper Networks
Differential Revision:	https://reviews.freebsd.org/D22069
2019-10-24 20:02:48 +00:00
Simon J. Gerraty
78cd72c914 Allow loader.efi to identify non-standard boot setup
PATH_BOOTABLE_TOKEN can be set to a non-standard
path that identifies a device as bootable.

Reviewed by: kevans, bcran
Differential Revision:  https://reviews.freebsd.org/D22062
2019-10-24 19:52:41 +00:00
Toomas Soome
96b2f9c996 userboot/test should use PRIx64 as one would expect from prefix 0x
Test is printing decimal value after prefix 0x.
2019-10-24 07:49:33 +00:00
Kyle Evans
1ce57df231 lualoader: fix setting of loader_color=NO in loader.conf(5)
Previously color.disabled would be calculated at color module load time,
then never touched again. We can detect serial boots beyond just what we're
told by loader.conf(5) so this works out in many cases, but we must
re-evaluate the situation after the config is loaded to make sure we're not
supposed to be forcing it enabled/disabled.

Discovered while trying to test r353872.
2019-10-21 20:17:31 +00:00
Kyle Evans
a6cf4be042 lualoader: don't botch disabling of color
When colors are disabled, color.escape{fg,bg} would return the passed in
color rather than the proper ANSI sequence for the color.
color.escape{fg,bg} would be wrong.

Instead return '', as the associated reset* functions will also return ''.
This should get rid of the funky '2' and '4' in the kernel selector if
you're booting serial.

Reported by:	npn
2019-10-21 20:09:43 +00:00
Toomas Soome
6b74887f03 loader: zfs_fmtdev can crash when pool discovery did fail and we have no spa
When zfs probe did fail and no spa was created, but zfs_fmtdev() is called,
we will crash while dereferencing spa (NULL pointer dereference).

MFC after:	1 week
2019-10-19 08:08:06 +00:00
Simon J. Gerraty
c7e6f9dcf5 Allow loader.efi to identify non-standard boot setup
PATH_BOOTABLE_TOKEN can be set to a non-standard
path that identifies a device as bootable.

Reviewed by: kevans, bcran
Differential Revision:  https://reviews.freebsd.org/D22062
2019-10-17 20:40:06 +00:00
Toomas Soome
abc23d5932 boot1.efi: provide generic exit() and stub getchar()
panic() is expecting us to have exit and getchar, lets provide those.
2019-10-15 08:33:05 +00:00
Justin Hibbits
dbef5f7155 loader/powerpc64: Include generic PVR values in CAS architecture list
Add generic PVR values for PowerISA 2.07 and 3.00.  This allows booting pseries
in QEMU with compatibilty mode enabled.

Submitted by:	Shawn Anastasio <shawn@anastas.io>
2019-10-08 02:36:53 +00:00
Kyle Evans
605261992c Revert r352557: powerpc/loader: Install ubldr without stripping
This was committed due to what was later diagnosed as an msdosfs bug
preventing in-place strip. This bug was fixed in r352564, and we agreed to
keep the workaround in for a bit to allow the driver fix a suitable amount
of propagation time for folks building/installing powerpc/ubldr, seeing as
how we were not in any hurry to revert.
2019-10-07 03:28:11 +00:00
Justin Hibbits
cd51f7df9c loader/powerpc64: Fix HV check for CAS usage
Logic was backwards.  The function returns true if it *is* running as a
hypervisor, whereas we want to only call the CAS utility if we're running as a
guest.

Reported by:	Shawn Anastasio <shawn@anastas.io>
2019-10-07 03:05:32 +00:00
Toomas Soome
03c9cdf71d loader.efi: for text mode, use STM to scroll the whole screen
Since local UEFI console is implemented on top of framebuffer,
we need to avoid redrawing the whole screen ourselves, but let
Simple Text Mode to do the scroll for us.
2019-10-06 18:38:58 +00:00
Kyle Evans
e12ff89136 Further normalize copyright notices
- s/C/c/ where I've been inconsistent about it
- +SPDX tags
- Remove "All rights reserved" where possible

Requested by:	rgrimes (all rights reserved)
2019-09-26 16:19:22 +00:00
Toomas Soome
8074c5cc86 loader: fix indentation in efi_console and vidconsole
Remove extra tab.

Reported by:	yuripv
2019-09-25 07:36:35 +00:00
Toomas Soome
233ab015c0 loader: add teken.fg_color and teken.bg_color variables
Add settable variables to control teken default color attributes.
The supported colors are 0-7 or basic color names:
black, red, green, brown, blue, magenta, cyan, white.

The current implementation does add some duplication which will be addressed
later.
2019-09-25 07:09:25 +00:00
Toomas Soome
6dd078df19 loader_lua: lua color changes should end with reset
The color change should have reset sequence, not switch to white.
2019-09-22 17:39:20 +00:00
Toomas Soome
2de5a21edc loader_4th: menu items need to reset color attribute, not switch to white
Forth menu kernel and BE entries, instead of resetting the color attribute,
are switching to white color.
2019-09-22 16:10:25 +00:00
Kyle Evans
bb7b569d36 loader: Respect loader_color=YES for serial consoles
It's not uncommon these days for the terminals attached to serial consoles
to support ANSI escape sequences. However, we assume escape sequences may
break some serial consoles and default to not using them when boot_serial or
boot_multicons (or if console contains "comconsole" in the forth loader) for
broader compatibility. We also have loader_color which can be explicitly set
to "NO" to disable the use of ANSI escape sequences.

The problem is that loader_color=YES gets ignored when boot_serial=YES or
boot_multicons=YES (or when console contains "comconsole" in the forth
loader).

To fix, the existing default behavior remains unchanged when loader_color is
unset, loader_color=NO explicitly disables the use of ANSI escape sequences
still, and the change is that loader_color=YES can now be used to explicitly
allow ANSI escapes when a serial console is enabled.

Submitted by:	Ryan Moeller <ryan@ixsystems.com>
Reviewed by:	tsoome (forth), kevans (lua)
MFC after:	1 week
Sponsored by:	iXsystems, Inc. (Ryan)
Differential Revision:	https://reviews.freebsd.org/D21732
2019-09-20 19:43:40 +00:00
Justin Hibbits
39404490b7 powerpc/loader: Install ubldr without stripping
Summary:
Install's strip capability, by way of strip(1), doesn't seem to work
correctly on msdosfs, and instead ends up truncating the resulting
binary to 0-length.  As a workaround, don't strip ubldr(8).  This
fixes installworld on Book-E ubldr-based platforms, which prior to this
would need to manually install ubldr separately after installworld, in
order to have a functional ubldr.

The same thing could be done on PowerNV platforms that use msdosfs /boot
volumes, since loader and loader.kboot, etc, all get truncated to 0 on
install.  However, PowerNV does not use loader, instead loading from
petitboot, so it's not really necessary at this time.

Reviewed by:	kevans
Differential Revision: https://reviews.freebsd.org/D21725
2019-09-20 13:35:28 +00:00
Toomas Soome
23883413d4 loader: fix typo in zalloc. 2019-09-20 05:22:34 +00:00
Kyle Evans
8b2195605b mips: ubldr: use truncated load address for mips32
BFD appears to silently truncate 0xffffffff80800000 when it processes the
ldscript for 32-bit mips, but LLD chokes on it as the linker script tries to
place elements above 32-bit range. It's unclear to me if silent truncation
is kosher or not and whether this patch is really what we want to do, but it
is one approach at least.

Reviewed by:	imp, mizhka
Differential Revision:	https://reviews.freebsd.org/D21487
2019-09-18 01:33:17 +00:00
Toomas Soome
68861a62f5 loader: provide u> and xemit words if needed
We have external interpreter (userboot.so) which may be lagging behind
with updates and may be missing u> xemit words.
2019-09-17 18:05:33 +00:00
Toomas Soome
d6b090d1d4 loader: revert r352421
As insisted by kib, malloc(0) is quite legal.
2019-09-17 16:16:46 +00:00
Toomas Soome
840fa0f86d loader.efi: efipart needs to use ioalign
UEFI specification 2.7A, EFI_BLOCK_IO_PROTOCOL, page 566.

The ioalign property does define the alignment of data buffer.

If the alignment is required and our buffer is not aligned, or if
the data buffer is not multiple of Blocksize, we need to use bounce buffer
to perform the block IO. This is much like with BIOS version, except
there the INT13 needs buffer to be located in low memory.

Additionally, we need to handle disk writes properly.
2019-09-17 13:50:25 +00:00
Toomas Soome
11db1a1654 loader: add memalign() to libsa
Implement memalign(size_t alignment, size_t size) to allocate aligned memory.
2019-09-17 13:15:27 +00:00
Toomas Soome
62ea4c11d5 loader: stand.h should define reallocf as Reallocf
Use the same approach as other zalloc functions.
2019-09-17 13:07:02 +00:00
Toomas Soome
e57c0c2afb loader: cstyle cleanup libsa zalloc sources
Clean up libstand zalloc* sources. Note that it is not 100% whitespace cleanup.
I also reduced block in znalloc and zfree as those were obvious simplifications
and did help to save one level of indent.
2019-09-17 11:35:53 +00:00
Toomas Soome
144c4ca039 loader: factor out guard location setup from Realloc
To simplify and make it easier to read, use ga_Bytes field from guard
structure.
2019-09-17 09:47:35 +00:00
Toomas Soome
41e2c61669 loader: Malloc(0) should return NULL.
We really should not allocate anything with size 0.
2019-09-16 20:28:08 +00:00
Toomas Soome
1781ad707e loader_4th: scan_buffer can leave empty string on stack
When the file processing is done, we will have string with lenght 0 in stack and we will attempt to
allocate 0 bytes.
2019-09-16 20:26:53 +00:00
Kyle Evans
af876563d1 lualoader: Add reload-conf loader command
This command will trigger a reload of the configuration from disk. This is
useful if you've changed currdev from recovery media to local disk as much
as I have over the past ~2 hours and are tired of the extra keystrokes.

This is really just a glorified shortcut, but reload-conf is likely easier
to remember for other people and does save some keystrokes when reloading
the configuration. It is also resilient to the underlying config method
changing interface, but this is unlikely to happen.

MFC after:	1 week
2019-09-14 03:38:18 +00:00
Toomas Soome
e7da26b6ca loader: vidconsole should set LINES and COLUMNS
Set LINES and COLUMNS based on terminal dimensions.
2019-09-10 21:53:42 +00:00
Kyle Evans
90a2541772 lualoader: Revert to ASCII menu frame for serial console
The box drawing characters we use aren't necessarily safe with a serial
console; for instance, in the report by npn@, these were causing his xterm
to send back a sequence that lua picked up as input and halted the boot.
This is less than ideal.

Fallback to ASCII frames for console with 'comconsole' in it.  This is a
partial revert r338108 by imp@ -- instead of removing the menu entirely and
disabling color/cursor sequences, just reverting the default frame to ASCII
is enough to not break in this setup.

Reported by:	npn
Triaged and recommended by:	tsoome
2019-09-10 21:30:38 +00:00
Toomas Soome
094b40659f loader: --gc-sections needs sections to work with
--gc-sections is not really useful unless we generate sections with
-ffunction-sections -fdata-sections

While there, i386/loader would win from --gc-sections too.
2019-09-08 19:39:45 +00:00
Toomas Soome
21388f5cf8 loader.efi: UEFI text mode background colors are only using 3 bits
Need to mask background colors to pass allowed value.
2019-09-06 19:30:23 +00:00