If a crash occurs, show the loaded UEFI images to facilitate analysis.
This is an example output:
=> bootefi 0x1000000
Found 0 disks
Hello world of bugs!
Invalid Opcode (Undefined Opcode)
EIP: 0010:[<06ceb06e>] EFLAGS: 00010206
Original EIP :[<fec9906e>]
EAX: 00000000 EBX: 06cec000 ECX: 00000fd0 EDX: 00000001
ESI: 06ced18a EDI: 07d0fe10 EBP: 07fe27a0 ESP: 07d0fde0
DS: 0018 ES: 0018 FS: 0020 GS: 0018 SS: 0018
CR0: 00000033 CR2: 00000000 CR3: 00000000 CR4: 00000000
DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000
DR6: ffff0ff0 DR7: 00000400
Stack:
0x07d0fde8 : 0x00000000
0x07d0fde4 : 0x06ced040
--->0x07d0fde0 : 0x07fe27a0
0x07d0fddc : 0x00010206
0x07d0fdd8 : 0x00000010
0x07d0fdd4 : 0x06ceb06e
UEFI image [0x06cea000:0x06cf0fff] pc=0x106e '/bug-i386.efi'
### ERROR ### Please RESET the board ###
With the additional information provided by this patch we know that the
problem occurred 0x106e after the load address of bug-i386.efi.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Bin Meng <bmeng.cn@gmail.com>
We cannot init the CPU fully both than once during a boot. Add a new
function which can be called to figure out the CPU identity, but which
does not change anything. For x86_64, this is empty for now.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
It is unnecessary to use a RAM version GDT for 64-bit U-Boot proper.
In fact we can just use the ROM version directly, which not only
eliminates the risk of being overwritten by application, but also
removes the complexity of patching the cpu_call64().
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
Before jumping to 64-bit U-Boot proper, SPL copies the cpu_call64()
function to a hardcoded address 0x3000000. This can have potential
conflicts with application usage. Switch the destination address
to be allocated from the heap to avoid such risk.
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
At present the 4-level page table base address for 64-bit U-Boot
proper is assigned an address that conflicts with CONFIG_LOADADDR.
Change it to an address within the low memory range instead.
Fixes crashes seen when 'dhcp' on QEMU x86_64 with
"-net nic -net user,tftp=.,bootfile=u-boot".
Reported-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Bin Meng <bmeng.cn@gmail.com>
Tested-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
Reviewed-by: Simon Glass <sjg@chromium.org>
There are still systems running which do not have any LAPIC or even
IOAPIC. Responsible MSRs for those do not exist and the systems are
crashing on trying to setup LAPIC.
This commit makes the APIC stuff able to switch off for those boards
which dont' have an LAPIC / IOAPIC.
Signed-off-by: Hannes Schmelzer <oe5hpm@oevsv.at>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
When U-Boot started using SPDX tags we were among the early adopters and
there weren't a lot of other examples to borrow from. So we picked the
area of the file that usually had a full license text and replaced it
with an appropriate SPDX-License-Identifier: entry. Since then, the
Linux Kernel has adopted SPDX tags and they place it as the very first
line in a file (except where shebangs are used, then it's second line)
and with slightly different comment styles than us.
In part due to community overlap, in part due to better tag visibility
and in part for other minor reasons, switch over to that style.
This commit changes all instances where we have a single declared
license in the tag as both the before and after are identical in tag
contents. There's also a few places where I found we did not have a tag
and have introduced one.
Signed-off-by: Tom Rini <trini@konsulko.com>
Depending upon the compiler used, IRQ entries could vary in sizes. With
GCC 5.x, the code generator will use short jumps for some IRQ entries
but near jumps for others. For example, GCC 5.4.0 generates the
following:
$ objdump -d interrupt.o
<snip>
00000207 <irq_18>:
207: 6a 12 push $0x12
209: eb 85 jmp 190 <irq_common_entry>
0000020b <irq_19>:
20b: 6a 13 push $0x13
20d: eb 81 jmp 190 <irq_common_entry>
0000020f <irq_20>:
20f: 6a 14 push $0x14
211: e9 7a ff ff ff jmp 190 <irq_common_entry>
00000216 <irq_21>:
216: 6a 15 push $0x15
218: e9 73 ff ff ff jmp 190 <irq_common_entry>
This causes a problem in cpu_init_interrupts(), because the IDT setup
assumed same sizes for all IRQ entries. GCC 4.x always generated 32-bit
jumps, so this previously was not a problem.
The fix is to force 32-bit near jumps for all entries within the
inline assembly. This works for GCC 5.x, and 4.x was already using
that form of jumping.
Signed-off-by: Jason Tang <tang@jtang.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
This code is only used in 32-bit mode. Move it so that it does not get
built with 64-bit U-Boot.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
This code is only used in 32-bit mode. Move it so that it does not get
built with 64-bit U-Boot.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Add a rough function to handle jumping from 32-bit SPL to 64-bit U-Boot.
This still needs work to clean it up.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
When SPL and U-Boot proper have different settings for this flag, we need to
use the correct one. Fix this up in the interrupt code.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
Much of the cpu and interrupt code cannot be compiled on 64-bit x86. Move it
into its own directory and build it only in 32-bit mode.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Bin Meng <bmeng.cn@gmail.com>
