s390/mm,kvm: improve detection of KVM guest faults

The identification of guest fault currently relies on the PF_VCPU flag.
This is set in guest_entry_irqoff and cleared in guest_exit_irqoff.
Both functions are called by __vcpu_run, the PF_VCPU flag is set for
quite a lot of kernel code outside of the guest execution.

Replace the PF_VCPU scheme with the PIF_GUEST_FAULT in the pt_regs and
make the program check handler code in entry.S set the bit only for
exception that occurred between the .Lsie_gmap and .Lsie_done labels.

Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>

diff --git a/arch/s390/include/asm/ptrace.h b/arch/s390/include/asm/ptrace.h
index 2f84e77f1f1b9844133663842b220f842c529e80..a3788dafc0e1f2272abd0ba9c455b775e81f627a 100644 (file)
--- a/arch/s390/include/asm/ptrace.h
+++ b/arch/s390/include/asm/ptrace.h
@@ -13,10 +13,12 @@
 #define PIF_SYSCALL            0       /* inside a system call */
 #define PIF_PER_TRAP           1       /* deliver sigtrap on return to user */
 #define PIF_SYSCALL_RESTART    2       /* restart the current system call */
+#define PIF_GUEST_FAULT                3       /* indicates program check in sie64a */
 
 #define _PIF_SYSCALL           _BITUL(PIF_SYSCALL)
 #define _PIF_PER_TRAP          _BITUL(PIF_PER_TRAP)
 #define _PIF_SYSCALL_RESTART   _BITUL(PIF_SYSCALL_RESTART)
+#define _PIF_GUEST_FAULT       _BITUL(PIF_GUEST_FAULT)
 
 #ifndef __ASSEMBLY__
 

diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index f498d201f98d09e527c26f815b048a8f48494499..ee53ac7b1ab8c9b11a083a28e4ed80ba41b99871 100644 (file)
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -518,6 +518,7 @@ ENTRY(pgm_check_handler)
        stmg    %r8,%r15,__LC_SAVE_AREA_SYNC
        lg      %r10,__LC_LAST_BREAK
        lg      %r12,__LC_CURRENT
+       lghi    %r11,0
        larl    %r13,cleanup_critical
        lmg     %r8,%r9,__LC_PGM_OLD_PSW
        tmhh    %r8,0x0001              # test problem state bit
@@ -532,6 +533,7 @@ ENTRY(pgm_check_handler)
        ni      __SIE_PROG0C+3(%r14),0xfe       # no longer in SIE
        lctlg   %c1,%c1,__LC_USER_ASCE          # load primary asce
        larl    %r9,sie_exit                    # skip forward to sie_exit
+       lghi    %r11,_PIF_GUEST_FAULT
 #endif
 0:     tmhh    %r8,0x4000              # PER bit set in old PSW ?
        jnz     1f                      # -> enabled, can't be a double fault
@@ -549,13 +551,14 @@ ENTRY(pgm_check_handler)
        jz      3f
        mvc     __THREAD_trap_tdb(256,%r14),0(%r13)
 3:     stg     %r10,__THREAD_last_break(%r14)
-4:     la      %r11,STACK_FRAME_OVERHEAD(%r15)
+4:     lgr     %r13,%r11
+       la      %r11,STACK_FRAME_OVERHEAD(%r15)
        stmg    %r0,%r7,__PT_R0(%r11)
        mvc     __PT_R8(64,%r11),__LC_SAVE_AREA_SYNC
        stmg    %r8,%r9,__PT_PSW(%r11)
        mvc     __PT_INT_CODE(4,%r11),__LC_PGM_ILC
        mvc     __PT_INT_PARM_LONG(8,%r11),__LC_TRANS_EXC_CODE
-       xc      __PT_FLAGS(8,%r11),__PT_FLAGS(%r11)
+       stg     %r13,__PT_FLAGS(%r11)
        stg     %r10,__PT_ARGS(%r11)
        tm      __LC_PGM_ILC+3,0x80     # check for per exception
        jz      5f

diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c
index 242b78c0a9ec84ba7a956d197cdd887bf86c9ebd..be974b3eb7e45a1ae4fdbe8522eef5eb41f2626b 100644 (file)
--- a/arch/s390/mm/fault.c
+++ b/arch/s390/mm/fault.c
@@ -117,7 +117,7 @@ static inline int user_space_fault(struct pt_regs *regs)
                return 1;
        if (trans_exc_code == 2) /* secondary space -> set_fs */
                return current->thread.mm_segment.ar4;
-       if (current->flags & PF_VCPU)
+       if (test_pt_regs_flag(regs, PIF_GUEST_FAULT))
                return 1;
        return 0;
 }
@@ -209,7 +209,7 @@ static void dump_fault_info(struct pt_regs *regs)
                pr_cont("kernel ");
        }
 #ifdef CONFIG_PGSTE
-       else if ((current->flags & PF_VCPU) && S390_lowcore.gmap) {
+       else if (test_pt_regs_flag(regs, PIF_GUEST_FAULT)) {
                struct gmap *gmap = (struct gmap *)S390_lowcore.gmap;
                asce = gmap->asce;
                pr_cont("gmap ");
@@ -438,7 +438,7 @@ static inline int do_exception(struct pt_regs *regs, int access)
        down_read(&mm->mmap_sem);
 
 #ifdef CONFIG_PGSTE
-       gmap = (current->flags & PF_VCPU) ?
+       gmap = test_pt_regs_flag(regs, PIF_GUEST_FAULT) ?
                (struct gmap *) S390_lowcore.gmap : NULL;
        if (gmap) {
                current->thread.gmap_addr = address;