summaryrefslogtreecommitdiff
path: root/arch/s390
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2024-05-13 08:33:52 -0700
committerLinus Torvalds <torvalds@linux-foundation.org>2024-05-13 08:33:52 -0700
commitd65e1a0f305ba3e7aabf6261a37bb871790d9f93 (patch)
tree4033039381e4a35eb76158e5b5130f05ae204139 /arch/s390
parenta38297e3fb012ddfa7ce0321a7e5a8daeb1872b6 (diff)
parent1812dc9c334f98227c65bc9c475f16fb6840a94b (diff)
downloadlinux-d65e1a0f305ba3e7aabf6261a37bb871790d9f93.tar.gz
linux-d65e1a0f305ba3e7aabf6261a37bb871790d9f93.tar.bz2
linux-d65e1a0f305ba3e7aabf6261a37bb871790d9f93.zip
Merge tag 's390-6.10-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
Pull s390 updates from Alexander Gordeev: - Store AP Query Configuration Information in a static buffer - Rework the AP initialization and add missing cleanups to the error path - Swap IRQ and AP bus/device registration to avoid race conditions - Export prot_virt_guest symbol - Introduce AP configuration changes notifier interface to facilitate modularization of the AP bus - Add CONFIG_AP kernel configuration option to allow modularization of the AP bus - Rework CONFIG_ZCRYPT_DEBUG kernel configuration option description and dependency and rename it to CONFIG_AP_DEBUG - Convert sprintf() and snprintf() to sysfs_emit() in CIO code - Adjust indentation of RELOCS command build step - Make crypto performance counters upward compatible - Convert make_page_secure() and gmap_make_secure() to use folio - Rework channel-utilization-block (CUB) handling in preparation of introducing additional CUBs - Use attribute groups to simplify registration, removal and extension of measurement-related channel-path sysfs attributes - Add a per-channel-path binary "ext_measurement" sysfs attribute that provides access to extended channel-path measurement data - Export measurement data for all channel-measurement-groups (CMG), not only for a specific ones. This enables support of new CMG data formats in userspace without the need for kernel changes - Add a per-channel-path sysfs attribute "speed_bps" that provides the operating speed in bits per second or 0 if the operating speed is not available - The CIO tracepoint subchannel-type field "st" is incorrectly set to the value of subchannel-enabled SCHIB "ena" field. Fix that - Do not forcefully limit vmemmap starting address to MAX_PHYSMEM_BITS - Consider the maximum physical address available to a DCSS segment (512GB) when memory layout is set up - Simplify the virtual memory layout setup by reducing the size of identity mapping vs vmemmap overlap - Swap vmalloc and Lowcore/Real Memory Copy areas in virtual memory. This will allow to place the kernel image next to kernel modules - Move everyting KASLR related from <asm/setup.h> to <asm/page.h> - Put virtual memory layout information into a structure to improve code generation - Currently __kaslr_offset is the kernel offset in both physical and virtual memory spaces. Uncouple these offsets to allow uncoupling of the addresses spaces - Currently the identity mapping base address is implicit and is always set to zero. Make it explicit by putting into __identity_base persistent boot variable and use it in proper context - Introduce .amode31 section start and end macros AMODE31_START and AMODE31_END - Introduce OS_INFO entries that do not reference any data in memory, but rather provide only values - Store virtual memory layout in OS_INFO. It is read out by makedumpfile, crash and other tools - Store virtual memory layout in VMCORE_INFO. It is read out by crash and other tools when /proc/kcore device is used - Create additional PT_LOAD ELF program header that covers kernel image only, so that vmcore tools could locate kernel text and data when virtual and physical memory spaces are uncoupled - Uncouple physical and virtual address spaces - Map kernel at fixed location when KASLR mode is disabled. The location is defined by CONFIG_KERNEL_IMAGE_BASE kernel configuration value. - Rework deployment of kernel image for both compressed and uncompressed variants as defined by CONFIG_KERNEL_UNCOMPRESSED kernel configuration value - Move .vmlinux.relocs section in front of the compressed kernel. The interim section rescue step is avoided as result - Correct modules thunk offset calculation when branch target is more than 2GB away - Kernel modules contain their own set of expoline thunks. Now that the kernel modules area is less than 4GB away from kernel expoline thunks, make modules use kernel expolines. Also make EXPOLINE_EXTERN the default if the compiler supports it - userfaultfd can insert shared zeropages into processes running VMs, but that is not allowed for s390. Fallback to allocating a fresh zeroed anonymous folio and insert that instead - Re-enable shared zeropages for non-PV and non-skeys KVM guests - Rename hex2bitmap() to ap_hex2bitmap() and export it for external use - Add ap_config sysfs attribute to provide the means for setting or displaying adapters, domains and control domains assigned to a vfio-ap mediated device in a single operation - Make vfio_ap_mdev_link_queue() ignore duplicate link requests - Add write support to ap_config sysfs attribute to allow atomic update a vfio-ap mediated device state - Document ap_config sysfs attribute - Function os_info_old_init() is expected to be called only from a regular kdump kernel. Enable it to be called from a stand-alone dump kernel - Address gcc -Warray-bounds warning and fix array size in struct os_info - s390 does not support SMBIOS, so drop unneeded CONFIG_DMI checks - Use unwinder instead of __builtin_return_address() with ftrace to prevent returning of undefined values - Sections .hash and .gnu.hash are only created when CONFIG_PIE_BUILD kernel is enabled. Drop these for the case CONFIG_PIE_BUILD is disabled - Compile kernel with -fPIC and link with -no-pie to allow kpatch feature always succeed and drop the whole CONFIG_PIE_BUILD option-enabled code - Add missing virt_to_phys() converter for VSIE facility and crypto control blocks * tag 's390-6.10-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (54 commits) Revert "s390: Relocate vmlinux ELF data to virtual address space" KVM: s390: vsie: Use virt_to_phys for crypto control block s390: Relocate vmlinux ELF data to virtual address space s390: Compile kernel with -fPIC and link with -no-pie s390: vmlinux.lds.S: Drop .hash and .gnu.hash for !CONFIG_PIE_BUILD s390/ftrace: Use unwinder instead of __builtin_return_address() s390/pci: Drop unneeded reference to CONFIG_DMI s390/os_info: Fix array size in struct os_info s390/os_info: Initialize old os_info in standalone dump kernel docs: Update s390 vfio-ap doc for ap_config sysfs attribute s390/vfio-ap: Add write support to sysfs attr ap_config s390/vfio-ap: Ignore duplicate link requests in vfio_ap_mdev_link_queue s390/vfio-ap: Add sysfs attr, ap_config, to export mdev state s390/ap: Externalize AP bus specific bitmap reading function s390/mm: Re-enable the shared zeropage for !PV and !skeys KVM guests mm/userfaultfd: Do not place zeropages when zeropages are disallowed s390/expoline: Make modules use kernel expolines s390/nospec: Correct modules thunk offset calculation s390/boot: Do not rescue .vmlinux.relocs section s390/boot: Rework deployment of the kernel image ...
Diffstat (limited to 'arch/s390')
-rw-r--r--arch/s390/Kconfig65
-rw-r--r--arch/s390/Makefile15
-rw-r--r--arch/s390/boot/Makefile9
-rw-r--r--arch/s390/boot/boot.h14
-rw-r--r--arch/s390/boot/decompressor.c15
-rw-r--r--arch/s390/boot/decompressor.h8
-rw-r--r--arch/s390/boot/kaslr.c2
-rw-r--r--arch/s390/boot/pgm_check_info.c4
-rw-r--r--arch/s390/boot/startup.c263
-rw-r--r--arch/s390/boot/vmem.c108
-rw-r--r--arch/s390/boot/vmlinux.lds.S28
-rw-r--r--arch/s390/include/asm/ap.h30
-rw-r--r--arch/s390/include/asm/asm-prototypes.h1
-rw-r--r--arch/s390/include/asm/chsc.h15
-rw-r--r--arch/s390/include/asm/extmem.h7
-rw-r--r--arch/s390/include/asm/ftrace.h8
-rw-r--r--arch/s390/include/asm/gmap.h2
-rw-r--r--arch/s390/include/asm/mmu.h5
-rw-r--r--arch/s390/include/asm/mmu_context.h1
-rw-r--r--arch/s390/include/asm/nospec-branch.h20
-rw-r--r--arch/s390/include/asm/nospec-insn.h13
-rw-r--r--arch/s390/include/asm/os_info.h29
-rw-r--r--arch/s390/include/asm/page.h50
-rw-r--r--arch/s390/include/asm/pgtable.h22
-rw-r--r--arch/s390/include/asm/physmem_info.h4
-rw-r--r--arch/s390/include/asm/setup.h14
-rw-r--r--arch/s390/kernel/Makefile2
-rw-r--r--arch/s390/kernel/crash_dump.c41
-rw-r--r--arch/s390/kernel/ipl.c6
-rw-r--r--arch/s390/kernel/nospec-branch.c4
-rw-r--r--arch/s390/kernel/os_info.c29
-rw-r--r--arch/s390/kernel/perf_cpum_cf.c2
-rw-r--r--arch/s390/kernel/perf_cpum_cf_events.c11
-rw-r--r--arch/s390/kernel/setup.c6
-rw-r--r--arch/s390/kernel/stacktrace.c19
-rw-r--r--arch/s390/kernel/uv.c51
-rw-r--r--arch/s390/kernel/vmcore_info.c2
-rw-r--r--arch/s390/kernel/vmlinux.lds.S38
-rw-r--r--arch/s390/kvm/kvm-s390.c4
-rw-r--r--arch/s390/kvm/vsie.c5
-rw-r--r--arch/s390/lib/Makefile2
-rw-r--r--arch/s390/lib/expoline.S (renamed from arch/s390/lib/expoline/expoline.S)0
-rw-r--r--arch/s390/lib/expoline/Makefile3
-rw-r--r--arch/s390/mm/gmap.c165
-rw-r--r--arch/s390/mm/vmem.c5
-rw-r--r--arch/s390/pci/pci_sysfs.c4
-rw-r--r--arch/s390/tools/relocs.c2
47 files changed, 698 insertions, 455 deletions
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index 8f01ada6845e..7e7fe89c9b25 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -17,6 +17,9 @@ config ARCH_HAS_ILOG2_U32
config ARCH_HAS_ILOG2_U64
def_bool n
+config ARCH_PROC_KCORE_TEXT
+ def_bool y
+
config GENERIC_HWEIGHT
def_bool y
@@ -552,7 +555,7 @@ config EXPOLINE
If unsure, say N.
config EXPOLINE_EXTERN
- def_bool n
+ def_bool y if EXPOLINE
depends on EXPOLINE
depends on CC_IS_GCC && GCC_VERSION >= 110200
depends on $(success,$(srctree)/arch/s390/tools/gcc-thunk-extern.sh $(CC))
@@ -590,18 +593,6 @@ config RELOCATABLE
Note: this option exists only for documentation purposes, please do
not remove it.
-config PIE_BUILD
- def_bool CC_IS_CLANG && !$(cc-option,-munaligned-symbols)
- help
- If the compiler is unable to generate code that can manage unaligned
- symbols, the kernel is linked as a position-independent executable
- (PIE) and includes dynamic relocations that are processed early
- during bootup.
-
- For kpatch functionality, it is recommended to build the kernel
- without the PIE_BUILD option. PIE_BUILD is only enabled when the
- compiler lacks proper support for handling unaligned symbols.
-
config RANDOMIZE_BASE
bool "Randomize the address of the kernel image (KASLR)"
default y
@@ -611,6 +602,25 @@ config RANDOMIZE_BASE
as a security feature that deters exploit attempts relying on
knowledge of the location of kernel internals.
+config KERNEL_IMAGE_BASE
+ hex "Kernel image base address"
+ range 0x100000 0x1FFFFFE0000000 if !KASAN
+ range 0x100000 0x1BFFFFE0000000 if KASAN
+ default 0x3FFE0000000 if !KASAN
+ default 0x7FFFE0000000 if KASAN
+ help
+ This is the address at which the kernel image is loaded in case
+ Kernel Address Space Layout Randomization (KASLR) is disabled.
+
+ In case the Protected virtualization guest support is enabled the
+ Ultravisor imposes a virtual address limit. If the value of this
+ option leads to the kernel image exceeding the Ultravisor limit,
+ this option is ignored and the image is loaded below the limit.
+
+ If the value of this option leads to the kernel image overlapping
+ the virtual memory where other data structures are located, this
+ option is ignored and the image is loaded above the structures.
+
endmenu
menu "Memory setup"
@@ -724,6 +734,33 @@ config EADM_SCH
To compile this driver as a module, choose M here: the
module will be called eadm_sch.
+config AP
+ def_tristate y
+ prompt "Support for Adjunct Processors (ap)"
+ help
+ This driver allows usage to Adjunct Processor (AP) devices via
+ the ap bus, cards and queues. Supported Adjunct Processors are
+ the CryptoExpress Cards (CEX).
+
+ To compile this driver as a module, choose M here: the
+ module will be called ap.
+
+ If unsure, say Y (default).
+
+config AP_DEBUG
+ def_bool n
+ prompt "Enable debug features for Adjunct Processor (ap) devices"
+ depends on AP
+ help
+ Say 'Y' here to enable some additional debug features for Adjunct
+ Processor (ap) devices.
+
+ There will be some more sysfs attributes displayed for ap queues.
+
+ Do not enable on production level kernel build.
+
+ If unsure, say N.
+
config VFIO_CCW
def_tristate n
prompt "Support for VFIO-CCW subchannels"
@@ -740,7 +777,7 @@ config VFIO_AP
prompt "VFIO support for AP devices"
depends on KVM
depends on VFIO
- depends on ZCRYPT
+ depends on AP
select VFIO_MDEV
help
This driver grants access to Adjunct Processor (AP) devices
diff --git a/arch/s390/Makefile b/arch/s390/Makefile
index 2dbb2d2f22f9..f2b21c7a70ef 100644
--- a/arch/s390/Makefile
+++ b/arch/s390/Makefile
@@ -14,14 +14,9 @@ KBUILD_AFLAGS_MODULE += -fPIC
KBUILD_CFLAGS_MODULE += -fPIC
KBUILD_AFLAGS += -m64
KBUILD_CFLAGS += -m64
-ifdef CONFIG_PIE_BUILD
-KBUILD_CFLAGS += -fPIE
-LDFLAGS_vmlinux := -pie -z notext
-else
-KBUILD_CFLAGS += $(call cc-option,-munaligned-symbols,)
-LDFLAGS_vmlinux := --emit-relocs --discard-none
+KBUILD_CFLAGS += -fPIC
+LDFLAGS_vmlinux := -no-pie --emit-relocs --discard-none
extra_tools := relocs
-endif
aflags_dwarf := -Wa,-gdwarf-2
KBUILD_AFLAGS_DECOMPRESSOR := $(CLANG_FLAGS) -m64 -D__ASSEMBLY__
ifndef CONFIG_AS_IS_LLVM
@@ -88,7 +83,6 @@ endif
ifdef CONFIG_EXPOLINE
ifdef CONFIG_EXPOLINE_EXTERN
- KBUILD_LDFLAGS_MODULE += arch/s390/lib/expoline/expoline.o
CC_FLAGS_EXPOLINE := -mindirect-branch=thunk-extern
CC_FLAGS_EXPOLINE += -mfunction-return=thunk-extern
else
@@ -167,11 +161,6 @@ vdso_prepare: prepare0
vdso-install-y += arch/s390/kernel/vdso64/vdso64.so.dbg
vdso-install-$(CONFIG_COMPAT) += arch/s390/kernel/vdso32/vdso32.so.dbg
-ifdef CONFIG_EXPOLINE_EXTERN
-modules_prepare: expoline_prepare
-expoline_prepare: scripts
- $(Q)$(MAKE) $(build)=arch/s390/lib/expoline arch/s390/lib/expoline/expoline.o
-endif
endif
# Don't use tabs in echo arguments
diff --git a/arch/s390/boot/Makefile b/arch/s390/boot/Makefile
index 294f08a8811a..070c9b2e905f 100644
--- a/arch/s390/boot/Makefile
+++ b/arch/s390/boot/Makefile
@@ -37,8 +37,7 @@ CFLAGS_sclp_early_core.o += -I$(srctree)/drivers/s390/char
obj-y := head.o als.o startup.o physmem_info.o ipl_parm.o ipl_report.o vmem.o
obj-y += string.o ebcdic.o sclp_early_core.o mem.o ipl_vmparm.o cmdline.o
-obj-y += version.o pgm_check_info.o ctype.o ipl_data.o
-obj-y += $(if $(CONFIG_PIE_BUILD),machine_kexec_reloc.o,relocs.o)
+obj-y += version.o pgm_check_info.o ctype.o ipl_data.o relocs.o
obj-$(findstring y, $(CONFIG_PROTECTED_VIRTUALIZATION_GUEST) $(CONFIG_PGSTE)) += uv.o
obj-$(CONFIG_RANDOMIZE_BASE) += kaslr.o
obj-y += $(if $(CONFIG_KERNEL_UNCOMPRESSED),,decompressor.o) info.o
@@ -49,9 +48,7 @@ targets := bzImage section_cmp.boot.data section_cmp.boot.preserved.data $(obj-y
targets += vmlinux.lds vmlinux vmlinux.bin vmlinux.bin.gz vmlinux.bin.bz2
targets += vmlinux.bin.xz vmlinux.bin.lzma vmlinux.bin.lzo vmlinux.bin.lz4
targets += vmlinux.bin.zst info.bin syms.bin vmlinux.syms $(obj-all)
-ifndef CONFIG_PIE_BUILD
targets += relocs.S
-endif
OBJECTS := $(addprefix $(obj)/,$(obj-y))
OBJECTS_ALL := $(addprefix $(obj)/,$(obj-all))
@@ -110,13 +107,11 @@ OBJCOPYFLAGS_vmlinux.bin := -O binary --remove-section=.comment --remove-section
$(obj)/vmlinux.bin: vmlinux FORCE
$(call if_changed,objcopy)
-ifndef CONFIG_PIE_BUILD
CMD_RELOCS=arch/s390/tools/relocs
-quiet_cmd_relocs = RELOCS $@
+quiet_cmd_relocs = RELOCS $@
cmd_relocs = $(CMD_RELOCS) $< > $@
$(obj)/relocs.S: vmlinux FORCE
$(call if_changed,relocs)
-endif
suffix-$(CONFIG_KERNEL_GZIP) := .gz
suffix-$(CONFIG_KERNEL_BZIP2) := .bz2
diff --git a/arch/s390/boot/boot.h b/arch/s390/boot/boot.h
index 567d60f78bbc..18027fdc92b0 100644
--- a/arch/s390/boot/boot.h
+++ b/arch/s390/boot/boot.h
@@ -17,7 +17,6 @@ struct machine_info {
};
struct vmlinux_info {
- unsigned long default_lma;
unsigned long entry;
unsigned long image_size; /* does not include .bss */
unsigned long bss_size; /* uncompressed image .bss size */
@@ -25,14 +24,8 @@ struct vmlinux_info {
unsigned long bootdata_size;
unsigned long bootdata_preserved_off;
unsigned long bootdata_preserved_size;
-#ifdef CONFIG_PIE_BUILD
- unsigned long dynsym_start;
- unsigned long rela_dyn_start;
- unsigned long rela_dyn_end;
-#else
unsigned long got_start;
unsigned long got_end;
-#endif
unsigned long amode31_size;
unsigned long init_mm_off;
unsigned long swapper_pg_dir_off;
@@ -74,10 +67,11 @@ void sclp_early_setup_buffer(void);
void print_pgm_check_info(void);
unsigned long randomize_within_range(unsigned long size, unsigned long align,
unsigned long min, unsigned long max);
-void setup_vmem(unsigned long asce_limit);
+void setup_vmem(unsigned long kernel_start, unsigned long kernel_end, unsigned long asce_limit);
void __printf(1, 2) decompressor_printk(const char *fmt, ...);
void print_stacktrace(unsigned long sp);
void error(char *m);
+int get_random(unsigned long limit, unsigned long *value);
extern struct machine_info machine;
@@ -98,6 +92,10 @@ extern struct vmlinux_info _vmlinux_info;
#define vmlinux _vmlinux_info
#define __abs_lowcore_pa(x) (((unsigned long)(x) - __abs_lowcore) % sizeof(struct lowcore))
+#define __kernel_va(x) ((void *)((unsigned long)(x) - __kaslr_offset_phys + __kaslr_offset))
+#define __kernel_pa(x) ((unsigned long)(x) - __kaslr_offset + __kaslr_offset_phys)
+#define __identity_va(x) ((void *)((unsigned long)(x) + __identity_base))
+#define __identity_pa(x) ((unsigned long)(x) - __identity_base)
static inline bool intersects(unsigned long addr0, unsigned long size0,
unsigned long addr1, unsigned long size1)
diff --git a/arch/s390/boot/decompressor.c b/arch/s390/boot/decompressor.c
index d762733a0753..f478e8e9cbda 100644
--- a/arch/s390/boot/decompressor.c
+++ b/arch/s390/boot/decompressor.c
@@ -63,24 +63,13 @@ static unsigned long free_mem_end_ptr = (unsigned long) _end + BOOT_HEAP_SIZE;
#include "../../../../lib/decompress_unzstd.c"
#endif
-#define decompress_offset ALIGN((unsigned long)_end + BOOT_HEAP_SIZE, PAGE_SIZE)
-
unsigned long mem_safe_offset(void)
{
- /*
- * due to 4MB HEAD_SIZE for bzip2
- * 'decompress_offset + vmlinux.image_size' could be larger than
- * kernel at final position + its .bss, so take the larger of two
- */
- return max(decompress_offset + vmlinux.image_size,
- vmlinux.default_lma + vmlinux.image_size + vmlinux.bss_size);
+ return ALIGN(free_mem_end_ptr, PAGE_SIZE);
}
-void *decompress_kernel(void)
+void deploy_kernel(void *output)
{
- void *output = (void *)decompress_offset;
-
__decompress(_compressed_start, _compressed_end - _compressed_start,
NULL, NULL, output, vmlinux.image_size, NULL, error);
- return output;
}
diff --git a/arch/s390/boot/decompressor.h b/arch/s390/boot/decompressor.h
index 92b81d2ea35d..4f966f06bd65 100644
--- a/arch/s390/boot/decompressor.h
+++ b/arch/s390/boot/decompressor.h
@@ -2,11 +2,9 @@
#ifndef BOOT_COMPRESSED_DECOMPRESSOR_H
#define BOOT_COMPRESSED_DECOMPRESSOR_H
-#ifdef CONFIG_KERNEL_UNCOMPRESSED
-static inline void *decompress_kernel(void) { return NULL; }
-#else
-void *decompress_kernel(void);
-#endif
+#ifndef CONFIG_KERNEL_UNCOMPRESSED
unsigned long mem_safe_offset(void);
+void deploy_kernel(void *output);
+#endif
#endif /* BOOT_COMPRESSED_DECOMPRESSOR_H */
diff --git a/arch/s390/boot/kaslr.c b/arch/s390/boot/kaslr.c
index 90602101e2ae..bd3bf5ef472d 100644
--- a/arch/s390/boot/kaslr.c
+++ b/arch/s390/boot/kaslr.c
@@ -43,7 +43,7 @@ static int check_prng(void)
return PRNG_MODE_TDES;
}
-static int get_random(unsigned long limit, unsigned long *value)
+int get_random(unsigned long limit, unsigned long *value)
{
struct prng_parm prng = {
/* initial parameter block for tdes mode, copied from libica */
diff --git a/arch/s390/boot/pgm_check_info.c b/arch/s390/boot/pgm_check_info.c
index 97244cd7a206..ea96275b0380 100644
--- a/arch/s390/boot/pgm_check_info.c
+++ b/arch/s390/boot/pgm_check_info.c
@@ -153,8 +153,10 @@ void print_pgm_check_info(void)
decompressor_printk("Kernel command line: %s\n", early_command_line);
decompressor_printk("Kernel fault: interruption code %04x ilc:%x\n",
S390_lowcore.pgm_code, S390_lowcore.pgm_ilc >> 1);
- if (kaslr_enabled())
+ if (kaslr_enabled()) {
decompressor_printk("Kernel random base: %lx\n", __kaslr_offset);
+ decompressor_printk("Kernel random base phys: %lx\n", __kaslr_offset_phys);
+ }
decompressor_printk("PSW : %016lx %016lx (%pS)\n",
S390_lowcore.psw_save_area.mask,
S390_lowcore.psw_save_area.addr,
diff --git a/arch/s390/boot/startup.c b/arch/s390/boot/startup.c
index 6cf89314209a..467283b112cd 100644
--- a/arch/s390/boot/startup.c
+++ b/arch/s390/boot/startup.c
@@ -3,6 +3,7 @@
#include <linux/elf.h>
#include <asm/page-states.h>
#include <asm/boot_data.h>
+#include <asm/extmem.h>
#include <asm/sections.h>
#include <asm/maccess.h>
#include <asm/cpu_mf.h>
@@ -18,7 +19,7 @@
#include "boot.h"
#include "uv.h"
-unsigned long __bootdata_preserved(__kaslr_offset);
+struct vm_layout __bootdata_preserved(vm_layout);
unsigned long __bootdata_preserved(__abs_lowcore);
unsigned long __bootdata_preserved(__memcpy_real_area);
pte_t *__bootdata_preserved(memcpy_real_ptep);
@@ -29,7 +30,6 @@ unsigned long __bootdata_preserved(vmemmap_size);
unsigned long __bootdata_preserved(MODULES_VADDR);
unsigned long __bootdata_preserved(MODULES_END);
unsigned long __bootdata_preserved(max_mappable);
-unsigned long __bootdata(ident_map_size);
u64 __bootdata_preserved(stfle_fac_list[16]);
u64 __bootdata_preserved(alt_stfle_fac_list[16]);
@@ -109,9 +109,19 @@ static void setup_lpp(void)
}
#ifdef CONFIG_KERNEL_UNCOMPRESSED
-unsigned long mem_safe_offset(void)
+static unsigned long mem_safe_offset(void)
{
- return vmlinux.default_lma + vmlinux.image_size + vmlinux.bss_size;
+ return (unsigned long)_compressed_start;
+}
+
+static void deploy_kernel(void *output)
+{
+ void *uncompressed_start = (void *)_compressed_start;
+
+ if (output == uncompressed_start)
+ return;
+ memmove(output, uncompressed_start, vmlinux.image_size);
+ memset(uncompressed_start, 0, vmlinux.image_size);
}
#endif
@@ -141,70 +151,18 @@ static void copy_bootdata(void)
memcpy((void *)vmlinux.bootdata_preserved_off, __boot_data_preserved_start, vmlinux.bootdata_preserved_size);
}
-#ifdef CONFIG_PIE_BUILD
-static void kaslr_adjust_relocs(unsigned long min_addr, unsigned long max_addr, unsigned long offset)
-{
- Elf64_Rela *rela_start, *rela_end, *rela;
- int r_type, r_sym, rc;
- Elf64_Addr loc, val;
- Elf64_Sym *dynsym;
-
- rela_start = (Elf64_Rela *) vmlinux.rela_dyn_start;
- rela_end = (Elf64_Rela *) vmlinux.rela_dyn_end;
- dynsym = (Elf64_Sym *) vmlinux.dynsym_start;
- for (rela = rela_start; rela < rela_end; rela++) {
- loc = rela->r_offset + offset;
- val = rela->r_addend;
- r_sym = ELF64_R_SYM(rela->r_info);
- if (r_sym) {
- if (dynsym[r_sym].st_shndx != SHN_UNDEF)
- val += dynsym[r_sym].st_value + offset;
- } else {
- /*
- * 0 == undefined symbol table index (STN_UNDEF),
- * used for R_390_RELATIVE, only add KASLR offset
- */
- val += offset;
- }
- r_type = ELF64_R_TYPE(rela->r_info);
- rc = arch_kexec_do_relocs(r_type, (void *) loc, val, 0);
- if (rc)
- error("Unknown relocation type");
- }
-}
-
-static void kaslr_adjust_got(unsigned long offset) {}
-static void rescue_relocs(void) {}
-static void free_relocs(void) {}
-#else
-static int *vmlinux_relocs_64_start;
-static int *vmlinux_relocs_64_end;
-
-static void rescue_relocs(void)
-{
- unsigned long size = __vmlinux_relocs_64_end - __vmlinux_relocs_64_start;
-
- vmlinux_relocs_64_start = (void *)physmem_alloc_top_down(RR_RELOC, size, 0);
- vmlinux_relocs_64_end = (void *)vmlinux_relocs_64_start + size;
- memmove(vmlinux_relocs_64_start, __vmlinux_relocs_64_start, size);
-}
-
-static void free_relocs(void)
-{
- physmem_free(RR_RELOC);
-}
-
-static void kaslr_adjust_relocs(unsigned long min_addr, unsigned long max_addr, unsigned long offset)
+static void kaslr_adjust_relocs(unsigned long min_addr, unsigned long max_addr,
+ unsigned long offset, unsigned long phys_offset)
{
int *reloc;
long loc;
/* Adjust R_390_64 relocations */
- for (reloc = vmlinux_relocs_64_start; reloc < vmlinux_relocs_64_end; reloc++) {
- loc = (long)*reloc + offset;
+ for (reloc = (int *)__vmlinux_relocs_64_start; reloc < (int *)__vmlinux_relocs_64_end; reloc++) {
+ loc = (long)*reloc + phys_offset;
if (loc < min_addr || loc > max_addr)
error("64-bit relocation outside of kernel!\n");
- *(u64 *)loc += offset;
+ *(u64 *)loc += offset - __START_KERNEL;
}
}
@@ -217,9 +175,8 @@ static void kaslr_adjust_got(unsigned long offset)
* reason. Adjust the GOT entries.
*/
for (entry = (u64 *)vmlinux.got_start; entry < (u64 *)vmlinux.got_end; entry++)
- *entry += offset;
+ *entry += offset - __START_KERNEL;
}
-#endif
/*
* Merge information from several sources into a single ident_map_size value.
@@ -261,9 +218,26 @@ static void setup_ident_map_size(unsigned long max_physmem_end)
#endif
}
-static unsigned long setup_kernel_memory_layout(void)
+#define FIXMAP_SIZE round_up(MEMCPY_REAL_SIZE + ABS_LOWCORE_MAP_SIZE, sizeof(struct lowcore))
+
+static unsigned long get_vmem_size(unsigned long identity_size,
+ unsigned long vmemmap_size,
+ unsigned long vmalloc_size,
+ unsigned long rte_size)
+{
+ unsigned long max_mappable, vsize;
+
+ max_mappable = max(identity_size, MAX_DCSS_ADDR);
+ vsize = round_up(SZ_2G + max_mappable, rte_size) +
+ round_up(vmemmap_size, rte_size) +
+ FIXMAP_SIZE + MODULES_LEN + KASLR_LEN;
+ return size_add(vsize, vmalloc_size);
+}
+
+static unsigned long setup_kernel_memory_layout(unsigned long kernel_size)
{
unsigned long vmemmap_start;
+ unsigned long kernel_start;
unsigned long asce_limit;
unsigned long rte_size;
unsigned long pages;
@@ -275,12 +249,19 @@ static unsigned long setup_kernel_memory_layout(void)
vmemmap_size = SECTION_ALIGN_UP(pages) * sizeof(struct page);
/* choose kernel address space layout: 4 or 3 levels. */
- vsize = round_up(ident_map_size, _REGION3_SIZE) + vmemmap_size +
- MODULES_LEN + MEMCPY_REAL_SIZE + ABS_LOWCORE_MAP_SIZE;
- vsize = size_add(vsize, vmalloc_size);
- if (IS_ENABLED(CONFIG_KASAN) || (vsize > _REGION2_SIZE)) {
+ BUILD_BUG_ON(!IS_ALIGNED(__START_KERNEL, THREAD_SIZE));
+ BUILD_BUG_ON(!IS_ALIGNED(__NO_KASLR_START_KERNEL, THREAD_SIZE));
+ BUILD_BUG_ON(__NO_KASLR_END_KERNEL > _REGION1_SIZE);
+ vsize = get_vmem_size(ident_map_size, vmemmap_size, vmalloc_size, _REGION3_SIZE);
+ if (IS_ENABLED(CONFIG_KASAN) || __NO_KASLR_END_KERNEL > _REGION2_SIZE ||
+ (vsize > _REGION2_SIZE && kaslr_enabled())) {
asce_limit = _REGION1_SIZE;
- rte_size = _REGION2_SIZE;
+ if (__NO_KASLR_END_KERNEL > _REGION2_SIZE) {
+ rte_size = _REGION2_SIZE;
+ vsize = get_vmem_size(ident_map_size, vmemmap_size, vmalloc_size, _REGION2_SIZE);
+ } else {
+ rte_size = _REGION3_SIZE;
+ }
} else {
asce_limit = _REGION2_SIZE;
rte_size = _REGION3_SIZE;
@@ -290,38 +271,67 @@ static unsigned long setup_kernel_memory_lay