linux/arch/s390/kernel/entry.S
Linus Torvalds 1ec6d09789 s390 updates for 6.12 merge window
- Optimize ftrace and kprobes code patching and avoid stop machine for
   kprobes if sequential instruction fetching facility is available
 
 - Add hiperdispatch feature to dynamically adjust CPU capacity in
   vertical polarization to improve scheduling efficiency and overall
   performance. Also add infrastructure for handling warning track
   interrupts (WTI), allowing for graceful CPU preemption
 
 - Rework crypto code pkey module and split it into separate, independent
   modules for sysfs, PCKMO, CCA, and EP11, allowing modules to load only
   when the relevant hardware is available
 
 - Add hardware acceleration for HMAC modes and the full AES-XTS cipher,
   utilizing message-security assist extensions (MSA) 10 and 11. It
   introduces new shash implementations for HMAC-SHA224/256/384/512 and
   registers the hardware-accelerated AES-XTS cipher as the preferred
   option. Also add clear key token support
 
 - Add MSA 10 and 11 processor activity instrumentation counters to perf
   and update PAI Extension 1 NNPA counters
 
 - Cleanup cpu sampling facility code and rework debug/WARN_ON_ONCE
   statements
 
 - Add support for SHA3 performance enhancements introduced with MSA 12
 
 - Add support for the query authentication information feature of
   MSA 13 and introduce the KDSA CPACF instruction. Provide query and query
   authentication information in sysfs, enabling tools like cpacfinfo to
   present this data in a human-readable form
 
 - Update kernel disassembler instructions
 
 - Always enable EXPOLINE_EXTERN if supported by the compiler to ensure
   kpatch compatibility
 
 - Add missing warning handling and relocated lowcore support to the
   early program check handler
 
 - Optimize ftrace_return_address() and avoid calling unwinder
 
 - Make modules use kernel ftrace trampolines
 
 - Strip relocs from the final vmlinux ELF file to make it roughly 2
   times smaller
 
 - Dump register contents and call trace for early crashes to the console
 
 - Generate ptdump address marker array dynamically
 
 - Fix rcu_sched stalls that might occur when adding or removing large
   amounts of pages at once to or from the CMM balloon
 
 - Fix deadlock caused by recursive lock of the AP bus scan mutex
 
 - Unify sync and async register save areas in entry code
 
 - Cleanup debug prints in crypto code
 
 - Various cleanup and sanitizing patches for the decompressor
 
 - Various small ftrace cleanups
 -----BEGIN PGP SIGNATURE-----
 
 iQEzBAABCAAdFiEE3QHqV+H2a8xAv27vjYWKoQLXFBgFAmbsZawACgkQjYWKoQLX
 FBg+Ogf+NiKPfvI14NcTwnOHB6qz8ApPdGfN9bNVtQxtK3epeAvtj0cMonAuKpRg
 xckTRRd8y0guhCT7Q2+WitSgA5eYDn+u9/Ux5YuKUdUdXolQ0D64BJNtVeEFkmJj
 s+Lesb8cVI9T2VBZOpuF9lJigfsDALBkFroqN4MDudDeahS+qy33bAc0OoqYNXHo
 S6OwPK1/tEG9O/oTN2V4mN+aP0B3/dl7Msezb0gfAXQJA+WUAyMNK0RHvoG9uzaa
 BWAyWWYABj6woGZEAQAzXcbzkQiRPixTqZVe6e4YndXhIlEnB/Z2AQFdTpT9V7En
 eOmmve3QuJa0hkF9q4H/anvOMPntTg==
 =Xagq
 -----END PGP SIGNATURE-----

Merge tag 's390-6.12-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux

Pull s390 updates from Vasily Gorbik:

 - Optimize ftrace and kprobes code patching and avoid stop machine for
   kprobes if sequential instruction fetching facility is available

 - Add hiperdispatch feature to dynamically adjust CPU capacity in
   vertical polarization to improve scheduling efficiency and overall
   performance. Also add infrastructure for handling warning track
   interrupts (WTI), allowing for graceful CPU preemption

 - Rework crypto code pkey module and split it into separate,
   independent modules for sysfs, PCKMO, CCA, and EP11, allowing modules
   to load only when the relevant hardware is available

 - Add hardware acceleration for HMAC modes and the full AES-XTS cipher,
   utilizing message-security assist extensions (MSA) 10 and 11. It
   introduces new shash implementations for HMAC-SHA224/256/384/512 and
   registers the hardware-accelerated AES-XTS cipher as the preferred
   option. Also add clear key token support

 - Add MSA 10 and 11 processor activity instrumentation counters to perf
   and update PAI Extension 1 NNPA counters

 - Cleanup cpu sampling facility code and rework debug/WARN_ON_ONCE
   statements

 - Add support for SHA3 performance enhancements introduced with MSA 12

 - Add support for the query authentication information feature of MSA
   13 and introduce the KDSA CPACF instruction. Provide query and query
   authentication information in sysfs, enabling tools like cpacfinfo to
   present this data in a human-readable form

 - Update kernel disassembler instructions

 - Always enable EXPOLINE_EXTERN if supported by the compiler to ensure
   kpatch compatibility

 - Add missing warning handling and relocated lowcore support to the
   early program check handler

 - Optimize ftrace_return_address() and avoid calling unwinder

 - Make modules use kernel ftrace trampolines

 - Strip relocs from the final vmlinux ELF file to make it roughly 2
   times smaller

 - Dump register contents and call trace for early crashes to the
   console

 - Generate ptdump address marker array dynamically

 - Fix rcu_sched stalls that might occur when adding or removing large
   amounts of pages at once to or from the CMM balloon

 - Fix deadlock caused by recursive lock of the AP bus scan mutex

 - Unify sync and async register save areas in entry code

 - Cleanup debug prints in crypto code

 - Various cleanup and sanitizing patches for the decompressor

 - Various small ftrace cleanups

* tag 's390-6.12-1' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (84 commits)
  s390/crypto: Display Query and Query Authentication Information in sysfs
  s390/crypto: Add Support for Query Authentication Information
  s390/crypto: Rework RRE and RRF CPACF inline functions
  s390/crypto: Add KDSA CPACF Instruction
  s390/disassembler: Remove duplicate instruction format RSY_RDRU
  s390/boot: Move boot_printk() code to own file
  s390/boot: Use boot_printk() instead of sclp_early_printk()
  s390/boot: Rename decompressor_printk() to boot_printk()
  s390/boot: Compile all files with the same march flag
  s390: Use MARCH_HAS_*_FEATURES defines
  s390: Provide MARCH_HAS_*_FEATURES defines
  s390/facility: Disable compile time optimization for decompressor code
  s390/boot: Increase minimum architecture to z10
  s390/als: Remove obsolete comment
  s390/sha3: Fix SHA3 selftests failures
  s390/pkey: Add AES xts and HMAC clear key token support
  s390/cpacf: Add MSA 10 and 11 new PCKMO functions
  s390/mm: Add cond_resched() to cmm_alloc/free_pages()
  s390/pai_ext: Update PAI extension 1 counters
  s390/pai_crypto: Add support for MSA 10 and 11 pai counters
  ...
2024-09-21 09:02:54 -07:00

668 lines
19 KiB
ArmAsm

/* SPDX-License-Identifier: GPL-2.0 */
/*
* S390 low-level entry points.
*
* Copyright IBM Corp. 1999, 2012
* Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com),
* Hartmut Penner (hp@de.ibm.com),
* Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com),
*/
#include <linux/export.h>
#include <linux/init.h>
#include <linux/linkage.h>
#include <asm/asm-extable.h>
#include <asm/alternative.h>
#include <asm/processor.h>
#include <asm/cache.h>
#include <asm/dwarf.h>
#include <asm/errno.h>
#include <asm/ptrace.h>
#include <asm/thread_info.h>
#include <asm/asm-offsets.h>
#include <asm/unistd.h>
#include <asm/page.h>
#include <asm/sigp.h>
#include <asm/irq.h>
#include <asm/fpu-insn.h>
#include <asm/setup.h>
#include <asm/nmi.h>
#include <asm/nospec-insn.h>
#include <asm/lowcore.h>
_LPP_OFFSET = __LC_LPP
.macro STBEAR address
ALTERNATIVE "nop", ".insn s,0xb2010000,\address", ALT_FACILITY(193)
.endm
.macro LBEAR address
ALTERNATIVE "nop", ".insn s,0xb2000000,\address", ALT_FACILITY(193)
.endm
.macro LPSWEY address, lpswe
ALTERNATIVE_2 "b \lpswe;nopr", \
".insn siy,0xeb0000000071,\address,0", ALT_FACILITY(193), \
__stringify(.insn siy,0xeb0000000071,LOWCORE_ALT_ADDRESS+\address,0), \
ALT_LOWCORE
.endm
.macro MBEAR reg, lowcore
ALTERNATIVE "brcl 0,0", __stringify(mvc __PT_LAST_BREAK(8,\reg),__LC_LAST_BREAK(\lowcore)),\
ALT_FACILITY(193)
.endm
.macro CHECK_STACK savearea, lowcore
#ifdef CONFIG_CHECK_STACK
tml %r15,THREAD_SIZE - CONFIG_STACK_GUARD
la %r14,\savearea(\lowcore)
jz stack_overflow
#endif
.endm
.macro CHECK_VMAP_STACK savearea, lowcore, oklabel
#ifdef CONFIG_VMAP_STACK
lgr %r14,%r15
nill %r14,0x10000 - THREAD_SIZE
oill %r14,STACK_INIT_OFFSET
clg %r14,__LC_KERNEL_STACK(\lowcore)
je \oklabel
clg %r14,__LC_ASYNC_STACK(\lowcore)
je \oklabel
clg %r14,__LC_MCCK_STACK(\lowcore)
je \oklabel
clg %r14,__LC_NODAT_STACK(\lowcore)
je \oklabel
clg %r14,__LC_RESTART_STACK(\lowcore)
je \oklabel
la %r14,\savearea(\lowcore)
j stack_overflow
#else
j \oklabel
#endif
.endm
/*
* The TSTMSK macro generates a test-under-mask instruction by
* calculating the memory offset for the specified mask value.
* Mask value can be any constant. The macro shifts the mask
* value to calculate the memory offset for the test-under-mask
* instruction.
*/
.macro TSTMSK addr, mask, size=8, bytepos=0
.if (\bytepos < \size) && (\mask >> 8)
.if (\mask & 0xff)
.error "Mask exceeds byte boundary"
.endif
TSTMSK \addr, "(\mask >> 8)", \size, "(\bytepos + 1)"
.exitm
.endif
.ifeq \mask
.error "Mask must not be zero"
.endif
off = \size - \bytepos - 1
tm off+\addr, \mask
.endm
.macro BPOFF
ALTERNATIVE "nop", ".insn rrf,0xb2e80000,0,0,12,0", ALT_SPEC(82)
.endm
.macro BPON
ALTERNATIVE "nop", ".insn rrf,0xb2e80000,0,0,13,0", ALT_SPEC(82)
.endm
.macro BPENTER tif_ptr,tif_mask
ALTERNATIVE "TSTMSK \tif_ptr,\tif_mask; jz .+8; .insn rrf,0xb2e80000,0,0,13,0", \
"j .+12; nop; nop", ALT_SPEC(82)
.endm
.macro BPEXIT tif_ptr,tif_mask
TSTMSK \tif_ptr,\tif_mask
ALTERNATIVE "jz .+8; .insn rrf,0xb2e80000,0,0,12,0", \
"jnz .+8; .insn rrf,0xb2e80000,0,0,13,0", ALT_SPEC(82)
.endm
#if IS_ENABLED(CONFIG_KVM)
.macro SIEEXIT sie_control,lowcore
lg %r9,\sie_control # get control block pointer
ni __SIE_PROG0C+3(%r9),0xfe # no longer in SIE
lctlg %c1,%c1,__LC_KERNEL_ASCE(\lowcore) # load primary asce
lg %r9,__LC_CURRENT(\lowcore)
mvi __TI_sie(%r9),0
larl %r9,sie_exit # skip forward to sie_exit
.endm
#endif
.macro STACKLEAK_ERASE
#ifdef CONFIG_GCC_PLUGIN_STACKLEAK
brasl %r14,stackleak_erase_on_task_stack
#endif
.endm
GEN_BR_THUNK %r14
.section .kprobes.text, "ax"
.Ldummy:
/*
* The following nop exists only in order to avoid that the next
* symbol starts at the beginning of the kprobes text section.
* In that case there would be several symbols at the same address.
* E.g. objdump would take an arbitrary symbol when disassembling
* the code.
* With the added nop in between this cannot happen.
*/
nop 0
/*
* Scheduler resume function, called by __switch_to
* gpr2 = (task_struct *)prev
* gpr3 = (task_struct *)next
* Returns:
* gpr2 = prev
*/
SYM_FUNC_START(__switch_to_asm)
stmg %r6,%r15,__SF_GPRS(%r15) # store gprs of prev task
lghi %r4,__TASK_stack
lghi %r1,__TASK_thread
llill %r5,STACK_INIT_OFFSET
stg %r15,__THREAD_ksp(%r1,%r2) # store kernel stack of prev
lg %r15,0(%r4,%r3) # start of kernel stack of next
agr %r15,%r5 # end of kernel stack of next
GET_LC %r13
stg %r3,__LC_CURRENT(%r13) # store task struct of next
stg %r15,__LC_KERNEL_STACK(%r13) # store end of kernel stack
lg %r15,__THREAD_ksp(%r1,%r3) # load kernel stack of next
aghi %r3,__TASK_pid
mvc __LC_CURRENT_PID(4,%r13),0(%r3) # store pid of next
ALTERNATIVE "nop", "lpp _LPP_OFFSET(%r13)", ALT_FACILITY(40)
lmg %r6,%r15,__SF_GPRS(%r15) # load gprs of next task
BR_EX %r14
SYM_FUNC_END(__switch_to_asm)
#if IS_ENABLED(CONFIG_KVM)
/*
* __sie64a calling convention:
* %r2 pointer to sie control block phys
* %r3 pointer to sie control block virt
* %r4 guest register save area
* %r5 guest asce
*/
SYM_FUNC_START(__sie64a)
stmg %r6,%r14,__SF_GPRS(%r15) # save kernel registers
GET_LC %r13
lg %r14,__LC_CURRENT(%r13)
stg %r2,__SF_SIE_CONTROL_PHYS(%r15) # save sie block physical..
stg %r3,__SF_SIE_CONTROL(%r15) # ...and virtual addresses
stg %r4,__SF_SIE_SAVEAREA(%r15) # save guest register save area
stg %r5,__SF_SIE_GUEST_ASCE(%r15) # save guest asce
xc __SF_SIE_REASON(8,%r15),__SF_SIE_REASON(%r15) # reason code = 0
mvc __SF_SIE_FLAGS(8,%r15),__TI_flags(%r14) # copy thread flags
lmg %r0,%r13,0(%r4) # load guest gprs 0-13
mvi __TI_sie(%r14),1
lctlg %c1,%c1,__SF_SIE_GUEST_ASCE(%r15) # load primary asce
lg %r14,__SF_SIE_CONTROL(%r15) # get control block pointer
oi __SIE_PROG0C+3(%r14),1 # we are going into SIE now
tm __SIE_PROG20+3(%r14),3 # last exit...
jnz .Lsie_skip
lg %r14,__SF_SIE_CONTROL_PHYS(%r15) # get sie block phys addr
BPEXIT __SF_SIE_FLAGS(%r15),_TIF_ISOLATE_BP_GUEST
.Lsie_entry:
sie 0(%r14)
# Let the next instruction be NOP to avoid triggering a machine check
# and handling it in a guest as result of the instruction execution.
nopr 7
.Lsie_leave:
BPOFF
BPENTER __SF_SIE_FLAGS(%r15),_TIF_ISOLATE_BP_GUEST
.Lsie_skip:
lg %r14,__SF_SIE_CONTROL(%r15) # get control block pointer
ni __SIE_PROG0C+3(%r14),0xfe # no longer in SIE
GET_LC %r14
lctlg %c1,%c1,__LC_KERNEL_ASCE(%r14) # load primary asce
lg %r14,__LC_CURRENT(%r14)
mvi __TI_sie(%r14),0
# some program checks are suppressing. C code (e.g. do_protection_exception)
# will rewind the PSW by the ILC, which is often 4 bytes in case of SIE. There
# are some corner cases (e.g. runtime instrumentation) where ILC is unpredictable.
# Other instructions between __sie64a and .Lsie_done should not cause program
# interrupts. So lets use 3 nops as a landing pad for all possible rewinds.
.Lrewind_pad6:
nopr 7
.Lrewind_pad4:
nopr 7
.Lrewind_pad2:
nopr 7
SYM_INNER_LABEL(sie_exit, SYM_L_GLOBAL)
lg %r14,__SF_SIE_SAVEAREA(%r15) # load guest register save area
stmg %r0,%r13,0(%r14) # save guest gprs 0-13
xgr %r0,%r0 # clear guest registers to
xgr %r1,%r1 # prevent speculative use
xgr %r3,%r3
xgr %r4,%r4
xgr %r5,%r5
lmg %r6,%r14,__SF_GPRS(%r15) # restore kernel registers
lg %r2,__SF_SIE_REASON(%r15) # return exit reason code
BR_EX %r14
.Lsie_fault:
lghi %r14,-EFAULT
stg %r14,__SF_SIE_REASON(%r15) # set exit reason code
j sie_exit
EX_TABLE(.Lrewind_pad6,.Lsie_fault)
EX_TABLE(.Lrewind_pad4,.Lsie_fault)
EX_TABLE(.Lrewind_pad2,.Lsie_fault)
EX_TABLE(sie_exit,.Lsie_fault)
SYM_FUNC_END(__sie64a)
EXPORT_SYMBOL(__sie64a)
EXPORT_SYMBOL(sie_exit)
#endif
/*
* SVC interrupt handler routine. System calls are synchronous events and
* are entered with interrupts disabled.
*/
SYM_CODE_START(system_call)
STMG_LC %r8,%r15,__LC_SAVE_AREA
GET_LC %r13
stpt __LC_SYS_ENTER_TIMER(%r13)
BPOFF
lghi %r14,0
.Lsysc_per:
STBEAR __LC_LAST_BREAK(%r13)
lctlg %c1,%c1,__LC_KERNEL_ASCE(%r13)
lg %r15,__LC_KERNEL_STACK(%r13)
xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15)
stmg %r0,%r7,STACK_FRAME_OVERHEAD+__PT_R0(%r15)
# clear user controlled register to prevent speculative use
xgr %r0,%r0
xgr %r1,%r1
xgr %r4,%r4
xgr %r5,%r5
xgr %r6,%r6
xgr %r7,%r7
xgr %r8,%r8
xgr %r9,%r9
xgr %r10,%r10
xgr %r11,%r11
la %r2,STACK_FRAME_OVERHEAD(%r15) # pointer to pt_regs
mvc __PT_R8(64,%r2),__LC_SAVE_AREA(%r13)
MBEAR %r2,%r13
lgr %r3,%r14
brasl %r14,__do_syscall
STACKLEAK_ERASE
lctlg %c1,%c1,__LC_USER_ASCE(%r13)
mvc __LC_RETURN_PSW(16,%r13),STACK_FRAME_OVERHEAD+__PT_PSW(%r15)
BPON
LBEAR STACK_FRAME_OVERHEAD+__PT_LAST_BREAK(%r15)
stpt __LC_EXIT_TIMER(%r13)
lmg %r0,%r15,STACK_FRAME_OVERHEAD+__PT_R0(%r15)
LPSWEY __LC_RETURN_PSW,__LC_RETURN_LPSWE
SYM_CODE_END(system_call)
#
# a new process exits the kernel with ret_from_fork
#
SYM_CODE_START(ret_from_fork)
lgr %r3,%r11
brasl %r14,__ret_from_fork
STACKLEAK_ERASE
GET_LC %r13
lctlg %c1,%c1,__LC_USER_ASCE(%r13)
mvc __LC_RETURN_PSW(16,%r13),STACK_FRAME_OVERHEAD+__PT_PSW(%r15)
BPON
LBEAR STACK_FRAME_OVERHEAD+__PT_LAST_BREAK(%r15)
stpt __LC_EXIT_TIMER(%r13)
lmg %r0,%r15,STACK_FRAME_OVERHEAD+__PT_R0(%r15)
LPSWEY __LC_RETURN_PSW,__LC_RETURN_LPSWE
SYM_CODE_END(ret_from_fork)
/*
* Program check handler routine
*/
SYM_CODE_START(pgm_check_handler)
STMG_LC %r8,%r15,__LC_SAVE_AREA
GET_LC %r13
stpt __LC_SYS_ENTER_TIMER(%r13)
BPOFF
lgr %r10,%r15
lmg %r8,%r9,__LC_PGM_OLD_PSW(%r13)
tmhh %r8,0x0001 # coming from user space?
jno .Lpgm_skip_asce
lctlg %c1,%c1,__LC_KERNEL_ASCE(%r13)
j 3f # -> fault in user space
.Lpgm_skip_asce:
1: tmhh %r8,0x4000 # PER bit set in old PSW ?
jnz 2f # -> enabled, can't be a double fault
tm __LC_PGM_ILC+3(%r13),0x80 # check for per exception
jnz .Lpgm_svcper # -> single stepped svc
2: CHECK_STACK __LC_SAVE_AREA,%r13
aghi %r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE)
# CHECK_VMAP_STACK branches to stack_overflow or 4f
CHECK_VMAP_STACK __LC_SAVE_AREA,%r13,4f
3: lg %r15,__LC_KERNEL_STACK(%r13)
4: la %r11,STACK_FRAME_OVERHEAD(%r15)
xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11)
xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15)
stmg %r0,%r7,__PT_R0(%r11)
mvc __PT_R8(64,%r11),__LC_SAVE_AREA(%r13)
mvc __PT_LAST_BREAK(8,%r11),__LC_PGM_LAST_BREAK(%r13)
stctg %c1,%c1,__PT_CR1(%r11)
#if IS_ENABLED(CONFIG_KVM)
ltg %r12,__LC_GMAP(%r13)
jz 5f
clc __GMAP_ASCE(8,%r12), __PT_CR1(%r11)
jne 5f
BPENTER __SF_SIE_FLAGS(%r10),_TIF_ISOLATE_BP_GUEST
SIEEXIT __SF_SIE_CONTROL(%r10),%r13
#endif
5: stmg %r8,%r9,__PT_PSW(%r11)
# clear user controlled registers to prevent speculative use
xgr %r0,%r0
xgr %r1,%r1
xgr %r3,%r3
xgr %r4,%r4
xgr %r5,%r5
xgr %r6,%r6
xgr %r7,%r7
lgr %r2,%r11
brasl %r14,__do_pgm_check
tmhh %r8,0x0001 # returning to user space?
jno .Lpgm_exit_kernel
STACKLEAK_ERASE
lctlg %c1,%c1,__LC_USER_ASCE(%r13)
BPON
stpt __LC_EXIT_TIMER(%r13)
.Lpgm_exit_kernel:
mvc __LC_RETURN_PSW(16,%r13),STACK_FRAME_OVERHEAD+__PT_PSW(%r15)
LBEAR STACK_FRAME_OVERHEAD+__PT_LAST_BREAK(%r15)
lmg %r0,%r15,STACK_FRAME_OVERHEAD+__PT_R0(%r15)
LPSWEY __LC_RETURN_PSW,__LC_RETURN_LPSWE
#
# single stepped system call
#
.Lpgm_svcper:
mvc __LC_RETURN_PSW(8,%r13),__LC_SVC_NEW_PSW(%r13)
larl %r14,.Lsysc_per
stg %r14,__LC_RETURN_PSW+8(%r13)
lghi %r14,1
LBEAR __LC_PGM_LAST_BREAK(%r13)
LPSWEY __LC_RETURN_PSW,__LC_RETURN_LPSWE # branch to .Lsysc_per
SYM_CODE_END(pgm_check_handler)
/*
* Interrupt handler macro used for external and IO interrupts.
*/
.macro INT_HANDLER name,lc_old_psw,handler
SYM_CODE_START(\name)
STMG_LC %r8,%r15,__LC_SAVE_AREA
GET_LC %r13
stckf __LC_INT_CLOCK(%r13)
stpt __LC_SYS_ENTER_TIMER(%r13)
STBEAR __LC_LAST_BREAK(%r13)
BPOFF
lmg %r8,%r9,\lc_old_psw(%r13)
tmhh %r8,0x0001 # interrupting from user ?
jnz 1f
#if IS_ENABLED(CONFIG_KVM)
lg %r10,__LC_CURRENT(%r13)
tm __TI_sie(%r10),0xff
jz 0f
BPENTER __SF_SIE_FLAGS(%r15),_TIF_ISOLATE_BP_GUEST
SIEEXIT __SF_SIE_CONTROL(%r15),%r13
#endif
0: CHECK_STACK __LC_SAVE_AREA,%r13
aghi %r15,-(STACK_FRAME_OVERHEAD + __PT_SIZE)
j 2f
1: lctlg %c1,%c1,__LC_KERNEL_ASCE(%r13)
lg %r15,__LC_KERNEL_STACK(%r13)
2: xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15)
la %r11,STACK_FRAME_OVERHEAD(%r15)
stmg %r0,%r7,__PT_R0(%r11)
# clear user controlled registers to prevent speculative use
xgr %r0,%r0
xgr %r1,%r1
xgr %r3,%r3
xgr %r4,%r4
xgr %r5,%r5
xgr %r6,%r6
xgr %r7,%r7
xgr %r10,%r10
xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11)
mvc __PT_R8(64,%r11),__LC_SAVE_AREA(%r13)
MBEAR %r11,%r13
stmg %r8,%r9,__PT_PSW(%r11)
lgr %r2,%r11 # pass pointer to pt_regs
brasl %r14,\handler
mvc __LC_RETURN_PSW(16,%r13),__PT_PSW(%r11)
tmhh %r8,0x0001 # returning to user ?
jno 2f
STACKLEAK_ERASE
lctlg %c1,%c1,__LC_USER_ASCE(%r13)
BPON
stpt __LC_EXIT_TIMER(%r13)
2: LBEAR __PT_LAST_BREAK(%r11)
lmg %r0,%r15,__PT_R0(%r11)
LPSWEY __LC_RETURN_PSW,__LC_RETURN_LPSWE
SYM_CODE_END(\name)
.endm
INT_HANDLER ext_int_handler,__LC_EXT_OLD_PSW,do_ext_irq
INT_HANDLER io_int_handler,__LC_IO_OLD_PSW,do_io_irq
/*
* Machine check handler routines
*/
SYM_CODE_START(mcck_int_handler)
BPOFF
GET_LC %r13
lmg %r8,%r9,__LC_MCK_OLD_PSW(%r13)
TSTMSK __LC_MCCK_CODE(%r13),MCCK_CODE_SYSTEM_DAMAGE
jo .Lmcck_panic # yes -> rest of mcck code invalid
TSTMSK __LC_MCCK_CODE(%r13),MCCK_CODE_CR_VALID
jno .Lmcck_panic # control registers invalid -> panic
ptlb
lay %r14,__LC_CPU_TIMER_SAVE_AREA(%r13)
mvc __LC_MCCK_ENTER_TIMER(8,%r13),0(%r14)
TSTMSK __LC_MCCK_CODE(%r13),MCCK_CODE_CPU_TIMER_VALID
jo 3f
la %r14,__LC_SYS_ENTER_TIMER(%r13)
clc 0(8,%r14),__LC_EXIT_TIMER(%r13)
jl 1f
la %r14,__LC_EXIT_TIMER(%r13)
1: clc 0(8,%r14),__LC_LAST_UPDATE_TIMER(%r13)
jl 2f
la %r14,__LC_LAST_UPDATE_TIMER(%r13)
2: spt 0(%r14)
mvc __LC_MCCK_ENTER_TIMER(8,%r13),0(%r14)
3: TSTMSK __LC_MCCK_CODE(%r13),MCCK_CODE_PSW_MWP_VALID
jno .Lmcck_panic
tmhh %r8,0x0001 # interrupting from user ?
jnz .Lmcck_user
TSTMSK __LC_MCCK_CODE(%r13),MCCK_CODE_PSW_IA_VALID
jno .Lmcck_panic
#if IS_ENABLED(CONFIG_KVM)
lg %r10,__LC_CURRENT(%r13)
tm __TI_sie(%r10),0xff
jz .Lmcck_user
# Need to compare the address instead of __TI_SIE flag.
# Otherwise there would be a race between setting the flag
# and entering SIE (or leaving and clearing the flag). This
# would cause machine checks targeted at the guest to be
# handled by the host.
larl %r14,.Lsie_entry
clgrjl %r9,%r14, 4f
larl %r14,.Lsie_leave
clgrjhe %r9,%r14, 4f
lg %r10,__LC_PCPU
oi __PCPU_FLAGS+7(%r10), _CIF_MCCK_GUEST
4: BPENTER __SF_SIE_FLAGS(%r15),_TIF_ISOLATE_BP_GUEST
SIEEXIT __SF_SIE_CONTROL(%r15),%r13
#endif
.Lmcck_user:
lg %r15,__LC_MCCK_STACK(%r13)
la %r11,STACK_FRAME_OVERHEAD(%r15)
stctg %c1,%c1,__PT_CR1(%r11)
lctlg %c1,%c1,__LC_KERNEL_ASCE(%r13)
xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15)
lay %r14,__LC_GPREGS_SAVE_AREA(%r13)
mvc __PT_R0(128,%r11),0(%r14)
# clear user controlled registers to prevent speculative use
xgr %r0,%r0
xgr %r1,%r1
xgr %r3,%r3
xgr %r4,%r4
xgr %r5,%r5
xgr %r6,%r6
xgr %r7,%r7
xgr %r10,%r10
stmg %r8,%r9,__PT_PSW(%r11)
xc __PT_FLAGS(8,%r11),__PT_FLAGS(%r11)
xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15)
lgr %r2,%r11 # pass pointer to pt_regs
brasl %r14,s390_do_machine_check
lctlg %c1,%c1,__PT_CR1(%r11)
lmg %r0,%r10,__PT_R0(%r11)
mvc __LC_RETURN_MCCK_PSW(16,%r13),__PT_PSW(%r11) # move return PSW
tm __LC_RETURN_MCCK_PSW+1(%r13),0x01 # returning to user ?
jno 0f
BPON
stpt __LC_EXIT_TIMER(%r13)
0: ALTERNATIVE "brcl 0,0", __stringify(lay %r12,__LC_LAST_BREAK_SAVE_AREA(%r13)),\
ALT_FACILITY(193)
LBEAR 0(%r12)
lmg %r11,%r15,__PT_R11(%r11)
LPSWEY __LC_RETURN_MCCK_PSW,__LC_RETURN_MCCK_LPSWE
.Lmcck_panic:
/*
* Iterate over all possible CPU addresses in the range 0..0xffff
* and stop each CPU using signal processor. Use compare and swap
* to allow just one CPU-stopper and prevent concurrent CPUs from
* stopping each other while leaving the others running.
*/
lhi %r5,0
lhi %r6,1
larl %r7,stop_lock
cs %r5,%r6,0(%r7) # single CPU-stopper only
jnz 4f
larl %r7,this_cpu
stap 0(%r7) # this CPU address
lh %r4,0(%r7)
nilh %r4,0
lhi %r0,1
sll %r0,16 # CPU counter
lhi %r3,0 # next CPU address
0: cr %r3,%r4
je 2f
1: sigp %r1,%r3,SIGP_STOP # stop next CPU
brc SIGP_CC_BUSY,1b
2: ahi %r3,1
brct %r0,0b
3: sigp %r1,%r4,SIGP_STOP # stop this CPU
brc SIGP_CC_BUSY,3b
4: j 4b
SYM_CODE_END(mcck_int_handler)
SYM_CODE_START(restart_int_handler)
ALTERNATIVE "nop", "lpp _LPP_OFFSET", ALT_FACILITY(40)
stg %r15,__LC_SAVE_AREA_RESTART
TSTMSK __LC_RESTART_FLAGS,RESTART_FLAG_CTLREGS,4
jz 0f
lctlg %c0,%c15,__LC_CREGS_SAVE_AREA
0: larl %r15,daton_psw
lpswe 0(%r15) # turn dat on, keep irqs off
.Ldaton:
GET_LC %r15
lg %r15,__LC_RESTART_STACK(%r15)
xc STACK_FRAME_OVERHEAD(__PT_SIZE,%r15),STACK_FRAME_OVERHEAD(%r15)
stmg %r0,%r14,STACK_FRAME_OVERHEAD+__PT_R0(%r15)
GET_LC %r13
mvc STACK_FRAME_OVERHEAD+__PT_R15(8,%r15),__LC_SAVE_AREA_RESTART(%r13)
mvc STACK_FRAME_OVERHEAD+__PT_PSW(16,%r15),__LC_RST_OLD_PSW(%r13)
xc 0(STACK_FRAME_OVERHEAD,%r15),0(%r15)
lg %r1,__LC_RESTART_FN(%r13) # load fn, parm & source cpu
lg %r2,__LC_RESTART_DATA(%r13)
lgf %r3,__LC_RESTART_SOURCE(%r13)
ltgr %r3,%r3 # test source cpu address
jm 1f # negative -> skip source stop
0: sigp %r4,%r3,SIGP_SENSE # sigp sense to source cpu
brc 10,0b # wait for status stored
1: basr %r14,%r1 # call function
stap __SF_EMPTY(%r15) # store cpu address
llgh %r3,__SF_EMPTY(%r15)
2: sigp %r4,%r3,SIGP_STOP # sigp stop to current cpu
brc 2,2b
3: j 3b
SYM_CODE_END(restart_int_handler)
__INIT
SYM_CODE_START(early_pgm_check_handler)
STMG_LC %r8,%r15,__LC_SAVE_AREA
GET_LC %r13
aghi %r15,-(STACK_FRAME_OVERHEAD+__PT_SIZE)
la %r11,STACK_FRAME_OVERHEAD(%r15)
xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15)
stmg %r0,%r7,__PT_R0(%r11)
mvc __PT_PSW(16,%r11),__LC_PGM_OLD_PSW(%r13)
mvc __PT_R8(64,%r11),__LC_SAVE_AREA(%r13)
lgr %r2,%r11
brasl %r14,__do_early_pgm_check
mvc __LC_RETURN_PSW(16,%r13),STACK_FRAME_OVERHEAD+__PT_PSW(%r15)
lmg %r0,%r15,STACK_FRAME_OVERHEAD+__PT_R0(%r15)
LPSWEY __LC_RETURN_PSW,__LC_RETURN_LPSWE
SYM_CODE_END(early_pgm_check_handler)
__FINIT
.section .kprobes.text, "ax"
#if defined(CONFIG_CHECK_STACK) || defined(CONFIG_VMAP_STACK)
/*
* The synchronous or the asynchronous stack overflowed. We are dead.
* No need to properly save the registers, we are going to panic anyway.
* Setup a pt_regs so that show_trace can provide a good call trace.
*/
SYM_CODE_START(stack_overflow)
GET_LC %r15
lg %r15,__LC_NODAT_STACK(%r15) # change to panic stack
la %r11,STACK_FRAME_OVERHEAD(%r15)
stmg %r0,%r7,__PT_R0(%r11)
stmg %r8,%r9,__PT_PSW(%r11)
mvc __PT_R8(64,%r11),0(%r14)
stg %r10,__PT_ORIG_GPR2(%r11) # store last break to orig_gpr2
xc __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15)
lgr %r2,%r11 # pass pointer to pt_regs
jg kernel_stack_overflow
SYM_CODE_END(stack_overflow)
#endif
.section .data, "aw"
.balign 4
SYM_DATA_LOCAL(stop_lock, .long 0)
SYM_DATA_LOCAL(this_cpu, .short 0)
.balign 8
SYM_DATA_START_LOCAL(daton_psw)
.quad PSW_KERNEL_BITS
.quad .Ldaton
SYM_DATA_END(daton_psw)
.section .rodata, "a"
.balign 8
#define SYSCALL(esame,emu) .quad __s390x_ ## esame
SYM_DATA_START(sys_call_table)
#include "asm/syscall_table.h"
SYM_DATA_END(sys_call_table)
#undef SYSCALL
#ifdef CONFIG_COMPAT
#define SYSCALL(esame,emu) .quad __s390_ ## emu
SYM_DATA_START(sys_call_table_emu)
#include "asm/syscall_table.h"
SYM_DATA_END(sys_call_table_emu)
#undef SYSCALL
#endif