Linux kernel mirror (for testing)
git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel
os
linux
Merge tag 'x86-urgent-2025-12-21' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Ingo Molnar:
- Fix FPU core dumps on certain CPU models
- Fix htmldocs build warning
- Export TLB tracing event name via header
- Remove unused constant from <linux/mm_types.h>
- Fix comments
- Fix whitespace noise in documentation
- Fix variadic structure's definition to un-confuse UBSAN
- Fix posted MSI interrupts irq_retrigger() bug
- Fix asm build failure with older GCC builds
* tag 'x86-urgent-2025-12-21' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/bug: Fix old GCC compile fails
x86/msi: Make irq_retrigger() functional for posted MSI
x86/platform/uv: Fix UBSAN array-index-out-of-bounds
mm: Remove tlb_flush_reason::NR_TLB_FLUSH_REASONS from <linux/mm_types.h>
x86/mm/tlb/trace: Export the TLB_REMOTE_WRONG_CPU enum in <trace/events/tlb.h>
x86/sgx: Remove unmatched quote in __sgx_encl_extend function comment
x86/boot/Documentation: Fix whitespace noise in boot.rst
x86/fpu: Fix FPU state core dump truncation on CPUs with no extended xfeatures
x86/boot/Documentation: Fix htmldocs build warning due to malformed table in boot.rst
+138
-108
+96
-96
Documentation/arch/x86/boot.rst
+96
-96
Documentation/arch/x86/boot.rst
···
95
95
The traditional memory map for the kernel loader, used for Image or
96
96
zImage kernels, typically looks like::
97
97
98
-
| |
98
+
| |
99
99
0A0000 +------------------------+
100
-
| Reserved for BIOS | Do not use. Reserved for BIOS EBDA.
100
+
| Reserved for BIOS | Do not use. Reserved for BIOS EBDA.
101
101
09A000 +------------------------+
102
-
| Command line |
103
-
| Stack/heap | For use by the kernel real-mode code.
102
+
| Command line |
103
+
| Stack/heap | For use by the kernel real-mode code.
104
104
098000 +------------------------+
105
-
| Kernel setup | The kernel real-mode code.
105
+
| Kernel setup | The kernel real-mode code.
106
106
090200 +------------------------+
107
-
| Kernel boot sector | The kernel legacy boot sector.
107
+
| Kernel boot sector | The kernel legacy boot sector.
108
108
090000 +------------------------+
109
-
| Protected-mode kernel | The bulk of the kernel image.
109
+
| Protected-mode kernel | The bulk of the kernel image.
110
110
010000 +------------------------+
111
-
| Boot loader | <- Boot sector entry point 0000:7C00
111
+
| Boot loader | <- Boot sector entry point 0000:7C00
112
112
001000 +------------------------+
113
-
| Reserved for MBR/BIOS |
113
+
| Reserved for MBR/BIOS |
114
114
000800 +------------------------+
115
-
| Typically used by MBR |
115
+
| Typically used by MBR |
116
116
000600 +------------------------+
117
-
| BIOS use only |
117
+
| BIOS use only |
118
118
000000 +------------------------+
119
119
120
120
When using bzImage, the protected-mode kernel was relocated to
···
142
142
For a modern bzImage kernel with boot protocol version >= 2.02, a
143
143
memory layout like the following is suggested::
144
144
145
-
~ ~
146
-
| Protected-mode kernel |
145
+
~ ~
146
+
| Protected-mode kernel |
147
147
100000 +------------------------+
148
-
| I/O memory hole |
148
+
| I/O memory hole |
149
149
0A0000 +------------------------+
150
-
| Reserved for BIOS | Leave as much as possible unused
151
-
~ ~
152
-
| Command line | (Can also be below the X+10000 mark)
150
+
| Reserved for BIOS | Leave as much as possible unused
151
+
~ ~
152
+
| Command line | (Can also be below the X+10000 mark)
153
153
X+10000 +------------------------+
154
-
| Stack/heap | For use by the kernel real-mode code.
154
+
| Stack/heap | For use by the kernel real-mode code.
155
155
X+08000 +------------------------+
156
-
| Kernel setup | The kernel real-mode code.
157
-
| Kernel boot sector | The kernel legacy boot sector.
156
+
| Kernel setup | The kernel real-mode code.
157
+
| Kernel boot sector | The kernel legacy boot sector.
158
158
X +------------------------+
159
-
| Boot loader | <- Boot sector entry point 0000:7C00
159
+
| Boot loader | <- Boot sector entry point 0000:7C00
160
160
001000 +------------------------+
161
-
| Reserved for MBR/BIOS |
161
+
| Reserved for MBR/BIOS |
162
162
000800 +------------------------+
163
-
| Typically used by MBR |
163
+
| Typically used by MBR |
164
164
000600 +------------------------+
165
-
| BIOS use only |
165
+
| BIOS use only |
166
166
000000 +------------------------+
167
167
168
168
... where the address X is as low as the design of the boot loader permits.
···
433
433
434
434
Assigned boot loader IDs:
435
435
436
-
== =======================================
436
+
==== =======================================
437
437
0x0 LILO
438
438
(0x00 reserved for pre-2.00 bootloader)
439
439
0x1 Loadlin
···
456
456
<http://sebastian-plotz.blogspot.de>
457
457
0x12 OVMF UEFI virtualization stack
458
458
0x13 barebox
459
-
== =======================================
459
+
==== =======================================
460
460
461
461
Please contact <hpa@zytor.com> if you need a bootloader ID value assigned.
462
462
···
809
809
as follow::
810
810
811
811
struct setup_data {
812
-
__u64 next;
813
-
__u32 type;
814
-
__u32 len;
815
-
__u8 data[];
812
+
__u64 next;
813
+
__u32 type;
814
+
__u32 len;
815
+
__u8 data[];
816
816
}
817
-
817
+
818
818
Where, the next is a 64-bit physical pointer to the next node of
819
819
linked list, the next field of the last node is 0; the type is used
820
820
to identify the contents of data; the len is the length of data
···
835
835
protocol 2.15::
836
836
837
837
struct setup_indirect {
838
-
__u32 type;
839
-
__u32 reserved; /* Reserved, must be set to zero. */
840
-
__u64 len;
841
-
__u64 addr;
838
+
__u32 type;
839
+
__u32 reserved; /* Reserved, must be set to zero. */
840
+
__u64 len;
841
+
__u64 addr;
842
842
};
843
843
844
844
The type member is a SETUP_INDIRECT | SETUP_* type. However, it cannot be
···
850
850
In this case setup_data and setup_indirect will look like this::
851
851
852
852
struct setup_data {
853
-
.next = 0, /* or <addr_of_next_setup_data_struct> */
854
-
.type = SETUP_INDIRECT,
855
-
.len = sizeof(setup_indirect),
856
-
.data[sizeof(setup_indirect)] = (struct setup_indirect) {
857
-
.type = SETUP_INDIRECT | SETUP_E820_EXT,
858
-
.reserved = 0,
859
-
.len = <len_of_SETUP_E820_EXT_data>,
860
-
.addr = <addr_of_SETUP_E820_EXT_data>,
861
-
},
853
+
.next = 0, /* or <addr_of_next_setup_data_struct> */
854
+
.type = SETUP_INDIRECT,
855
+
.len = sizeof(setup_indirect),
856
+
.data[sizeof(setup_indirect)] = (struct setup_indirect) {
857
+
.type = SETUP_INDIRECT | SETUP_E820_EXT,
858
+
.reserved = 0,
859
+
.len = <len_of_SETUP_E820_EXT_data>,
860
+
.addr = <addr_of_SETUP_E820_EXT_data>,
861
+
},
862
862
}
863
863
864
864
.. note::
···
897
897
The kernel runtime start address is determined by the following algorithm::
898
898
899
899
if (relocatable_kernel) {
900
-
if (load_address < pref_address)
901
-
load_address = pref_address;
902
-
runtime_start = align_up(load_address, kernel_alignment);
900
+
if (load_address < pref_address)
901
+
load_address = pref_address;
902
+
runtime_start = align_up(load_address, kernel_alignment);
903
903
} else {
904
-
runtime_start = pref_address;
904
+
runtime_start = pref_address;
905
905
}
906
906
907
907
Hence the necessary memory window location and size can be estimated by
···
975
975
be prefixed with header/magic and its size, e.g.::
976
976
977
977
kernel_info:
978
-
.ascii "LToP" /* Header, Linux top (structure). */
979
-
.long kernel_info_var_len_data - kernel_info
980
-
.long kernel_info_end - kernel_info
981
-
.long 0x01234567 /* Some fixed size data for the bootloaders. */
978
+
.ascii "LToP" /* Header, Linux top (structure). */
979
+
.long kernel_info_var_len_data - kernel_info
980
+
.long kernel_info_end - kernel_info
981
+
.long 0x01234567 /* Some fixed size data for the bootloaders. */
982
982
kernel_info_var_len_data:
983
983
example_struct: /* Some variable size data for the bootloaders. */
984
-
.ascii "0123" /* Header/Magic. */
985
-
.long example_struct_end - example_struct
986
-
.ascii "Struct"
987
-
.long 0x89012345
984
+
.ascii "0123" /* Header/Magic. */
985
+
.long example_struct_end - example_struct
986
+
.ascii "Struct"
987
+
.long 0x89012345
988
988
example_struct_end:
989
989
example_strings: /* Some variable size data for the bootloaders. */
990
-
.ascii "ABCD" /* Header/Magic. */
991
-
.long example_strings_end - example_strings
992
-
.asciz "String_0"
993
-
.asciz "String_1"
990
+
.ascii "ABCD" /* Header/Magic. */
991
+
.long example_strings_end - example_strings
992
+
.asciz "String_0"
993
+
.asciz "String_1"
994
994
example_strings_end:
995
995
kernel_info_end:
996
996
···
1132
1132
unsigned long base_ptr; /* base address for real-mode segment */
1133
1133
1134
1134
if (setup_sects == 0)
1135
-
setup_sects = 4;
1135
+
setup_sects = 4;
1136
1136
1137
1137
if (protocol >= 0x0200) {
1138
-
type_of_loader = <type code>;
1139
-
if (loading_initrd) {
1140
-
ramdisk_image = <initrd_address>;
1141
-
ramdisk_size = <initrd_size>;
1142
-
}
1138
+
type_of_loader = <type code>;
1139
+
if (loading_initrd) {
1140
+
ramdisk_image = <initrd_address>;
1141
+
ramdisk_size = <initrd_size>;
1142
+
}
1143
1143
1144
-
if (protocol >= 0x0202 && loadflags & 0x01)
1145
-
heap_end = 0xe000;
1146
-
else
1147
-
heap_end = 0x9800;
1144
+
if (protocol >= 0x0202 && loadflags & 0x01)
1145
+
heap_end = 0xe000;
1146
+
else
1147
+
heap_end = 0x9800;
1148
1148
1149
-
if (protocol >= 0x0201) {
1150
-
heap_end_ptr = heap_end - 0x200;
1151
-
loadflags |= 0x80; /* CAN_USE_HEAP */
1152
-
}
1149
+
if (protocol >= 0x0201) {
1150
+
heap_end_ptr = heap_end - 0x200;
1151
+
loadflags |= 0x80; /* CAN_USE_HEAP */
1152
+
}
1153
1153
1154
-
if (protocol >= 0x0202) {
1155
-
cmd_line_ptr = base_ptr + heap_end;
1156
-
strcpy(cmd_line_ptr, cmdline);
1157
-
} else {
1158
-
cmd_line_magic = 0xA33F;
1159
-
cmd_line_offset = heap_end;
1160
-
setup_move_size = heap_end + strlen(cmdline) + 1;
1161
-
strcpy(base_ptr + cmd_line_offset, cmdline);
1162
-
}
1154
+
if (protocol >= 0x0202) {
1155
+
cmd_line_ptr = base_ptr + heap_end;
1156
+
strcpy(cmd_line_ptr, cmdline);
1157
+
} else {
1158
+
cmd_line_magic = 0xA33F;
1159
+
cmd_line_offset = heap_end;
1160
+
setup_move_size = heap_end + strlen(cmdline) + 1;
1161
+
strcpy(base_ptr + cmd_line_offset, cmdline);
1162
+
}
1163
1163
} else {
1164
-
/* Very old kernel */
1164
+
/* Very old kernel */
1165
1165
1166
-
heap_end = 0x9800;
1166
+
heap_end = 0x9800;
1167
1167
1168
-
cmd_line_magic = 0xA33F;
1169
-
cmd_line_offset = heap_end;
1168
+
cmd_line_magic = 0xA33F;
1169
+
cmd_line_offset = heap_end;
1170
1170
1171
-
/* A very old kernel MUST have its real-mode code loaded at 0x90000 */
1172
-
if (base_ptr != 0x90000) {
1173
-
/* Copy the real-mode kernel */
1174
-
memcpy(0x90000, base_ptr, (setup_sects + 1) * 512);
1175
-
base_ptr = 0x90000; /* Relocated */
1176
-
}
1171
+
/* A very old kernel MUST have its real-mode code loaded at 0x90000 */
1172
+
if (base_ptr != 0x90000) {
1173
+
/* Copy the real-mode kernel */
1174
+
memcpy(0x90000, base_ptr, (setup_sects + 1) * 512);
1175
+
base_ptr = 0x90000; /* Relocated */
1176
+
}
1177
1177
1178
-
strcpy(0x90000 + cmd_line_offset, cmdline);
1178
+
strcpy(0x90000 + cmd_line_offset, cmdline);
1179
1179
1180
-
/* It is recommended to clear memory up to the 32K mark */
1181
-
memset(0x90000 + (setup_sects + 1) * 512, 0, (64 - (setup_sects + 1)) * 512);
1180
+
/* It is recommended to clear memory up to the 32K mark */
1181
+
memset(0x90000 + (setup_sects + 1) * 512, 0, (64 - (setup_sects + 1)) * 512);
1182
1182
}
1183
1183
1184
1184
+1
-1
arch/x86/include/asm/bug.h
+1
-1
arch/x86/include/asm/bug.h
+7
arch/x86/include/asm/irq_remapping.h
+7
arch/x86/include/asm/irq_remapping.h
+1
-1
arch/x86/include/asm/uv/bios.h
+1
-1
arch/x86/include/asm/uv/bios.h
+1
-1
arch/x86/kernel/cpu/sgx/ioctl.c
+1
-1
arch/x86/kernel/cpu/sgx/ioctl.c
···
242
242
/*
243
243
* If the caller requires measurement of the page as a proof for the content,
244
244
* use EEXTEND to add a measurement for 256 bytes of the page. Repeat this
245
-
* operation until the entire page is measured."
245
+
* operation until the entire page is measured.
246
246
*/
247
247
static int __sgx_encl_extend(struct sgx_encl *encl,
248
248
struct sgx_epc_page *epc_page)
+2
-2
arch/x86/kernel/fpu/xstate.c
+2
-2
arch/x86/kernel/fpu/xstate.c
···
1946
1946
};
1947
1947
1948
1948
if (!dump_emit(cprm, &xc, sizeof(xc)))
1949
-
return 0;
1949
+
return -1;
1950
1950
1951
1951
num_records++;
1952
1952
}
···
1984
1984
return 1;
1985
1985
1986
1986
num_records = dump_xsave_layout_desc(cprm);
1987
-
if (!num_records)
1987
+
if (num_records < 0)
1988
1988
return 1;
1989
1989
1990
1990
/* Total size should be equal to the number of records */
+23
arch/x86/kernel/irq.c
+23
arch/x86/kernel/irq.c
···
397
397
398
398
/* Posted Interrupt Descriptors for coalesced MSIs to be posted */
399
399
DEFINE_PER_CPU_ALIGNED(struct pi_desc, posted_msi_pi_desc);
400
+
static DEFINE_PER_CPU_CACHE_HOT(bool, posted_msi_handler_active);
400
401
401
402
void intel_posted_msi_init(void)
402
403
{
···
413
412
apic_id = this_cpu_read(x86_cpu_to_apicid);
414
413
destination = x2apic_enabled() ? apic_id : apic_id << 8;
415
414
this_cpu_write(posted_msi_pi_desc.ndst, destination);
415
+
}
416
+
417
+
void intel_ack_posted_msi_irq(struct irq_data *irqd)
418
+
{
419
+
irq_move_irq(irqd);
420
+
421
+
/*
422
+
* Handle the rare case that irq_retrigger() raised the actual
423
+
* assigned vector on the target CPU, which means that it was not
424
+
* invoked via the posted MSI handler below. In that case APIC EOI
425
+
* is required as otherwise the ISR entry becomes stale and lower
426
+
* priority interrupts are never going to be delivered after that.
427
+
*
428
+
* If the posted handler invoked the device interrupt handler then
429
+
* the EOI would be premature because it would acknowledge the
430
+
* posted vector.
431
+
*/
432
+
if (unlikely(!__this_cpu_read(posted_msi_handler_active)))
433
+
apic_eoi();
416
434
}
417
435
418
436
static __always_inline bool handle_pending_pir(unsigned long *pir, struct pt_regs *regs)
···
466
446
467
447
pid = this_cpu_ptr(&posted_msi_pi_desc);
468
448
449
+
/* Mark the handler active for intel_ack_posted_msi_irq() */
450
+
__this_cpu_write(posted_msi_handler_active, true);
469
451
inc_irq_stat(posted_msi_notification_count);
470
452
irq_enter();
471
453
···
496
474
497
475
apic_eoi();
498
476
irq_exit();
477
+
__this_cpu_write(posted_msi_handler_active, false);
499
478
set_irq_regs(old_regs);
500
479
}
501
480
#endif /* X86_POSTED_MSI */
+4
-4
drivers/iommu/intel/irq_remapping.c
+4
-4
drivers/iommu/intel/irq_remapping.c
···
1303
1303
* irq_enter();
1304
1304
* handle_edge_irq()
1305
1305
* irq_chip_ack_parent()
1306
-
* irq_move_irq(); // No EOI
1306
+
* intel_ack_posted_msi_irq(); // No EOI
1307
1307
* handle_irq_event()
1308
1308
* driver_handler()
1309
1309
* handle_edge_irq()
1310
1310
* irq_chip_ack_parent()
1311
-
* irq_move_irq(); // No EOI
1311
+
* intel_ack_posted_msi_irq(); // No EOI
1312
1312
* handle_irq_event()
1313
1313
* driver_handler()
1314
1314
* handle_edge_irq()
1315
1315
* irq_chip_ack_parent()
1316
-
* irq_move_irq(); // No EOI
1316
+
* intel_ack_posted_msi_irq(); // No EOI
1317
1317
* handle_irq_event()
1318
1318
* driver_handler()
1319
1319
* apic_eoi()
···
1322
1322
*/
1323
1323
static struct irq_chip intel_ir_chip_post_msi = {
1324
1324
.name = "INTEL-IR-POST",
1325
-
.irq_ack = irq_move_irq,
1325
+
.irq_ack = intel_ack_posted_msi_irq,
1326
1326
.irq_set_affinity = intel_ir_set_affinity,
1327
1327
.irq_compose_msi_msg = intel_ir_compose_msi_msg,
1328
1328
.irq_set_vcpu_affinity = intel_ir_set_vcpu_affinity,
-1
include/linux/mm_types.h
-1
include/linux/mm_types.h
+3
-2
include/trace/events/tlb.h
+3
-2
include/trace/events/tlb.h
···
12
12
EM( TLB_FLUSH_ON_TASK_SWITCH, "flush on task switch" ) \
13
13
EM( TLB_REMOTE_SHOOTDOWN, "remote shootdown" ) \
14
14
EM( TLB_LOCAL_SHOOTDOWN, "local shootdown" ) \
15
-
EM( TLB_LOCAL_MM_SHOOTDOWN, "local mm shootdown" ) \
16
-
EMe( TLB_REMOTE_SEND_IPI, "remote ipi send" )
15
+
EM( TLB_LOCAL_MM_SHOOTDOWN, "local MM shootdown" ) \
16
+
EM( TLB_REMOTE_SEND_IPI, "remote IPI send" ) \
17
+
EMe( TLB_REMOTE_WRONG_CPU, "remote wrong CPU" )
17
18
18
19
/*
19
20
* First define the enums in TLB_FLUSH_REASON to be exported to userspace