freebsd-dev/sys/ia64/ia32/ia32_sysvec.c

385 lines
10 KiB
C
Raw Normal View History

/*-
* Copyright (c) 2002 Doug Rabson
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* $FreeBSD$
*/
#define __ELF_WORD_SIZE 32
#include <sys/param.h>
#include <sys/exec.h>
#include <sys/fcntl.h>
#include <sys/imgact.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/malloc.h>
#include <sys/mutex.h>
#include <sys/mman.h>
#include <sys/namei.h>
#include <sys/pioctl.h>
#include <sys/proc.h>
#include <sys/procfs.h>
#include <sys/resourcevar.h>
#include <sys/systm.h>
#include <sys/signalvar.h>
#include <sys/stat.h>
#include <sys/sx.h>
#include <sys/syscall.h>
#include <sys/sysctl.h>
#include <sys/sysent.h>
#include <sys/vnode.h>
#include <sys/imgact_elf.h>
Revamp of the syscall path, exception and context handling. The prime objectives are: o Implement a syscall path based on the epc inststruction (see sys/ia64/ia64/syscall.s). o Revisit the places were we need to save and restore registers and define those contexts in terms of the register sets (see sys/ia64/include/_regset.h). Secundairy objectives: o Remove the requirement to use contigmalloc for kernel stacks. o Better handling of the high FP registers for SMP systems. o Switch to the new cpu_switch() and cpu_throw() semantics. o Add a good unwinder to reconstruct contexts for the rare cases we need to (see sys/contrib/ia64/libuwx) Many files are affected by this change. Functionally it boils down to: o The EPC syscall doesn't preserve registers it does not need to preserve and places the arguments differently on the stack. This affects libc and truss. o The address of the kernel page directory (kptdir) had to be unstaticized for use by the nested TLB fault handler. The name has been changed to ia64_kptdir to avoid conflicts. The renaming affects libkvm. o The trapframe only contains the special registers and the scratch registers. For syscalls using the EPC syscall path no scratch registers are saved. This affects all places where the trapframe is accessed. Most notably the unaligned access handler, the signal delivery code and the debugger. o Context switching only partly saves the special registers and the preserved registers. This affects cpu_switch() and triggered the move to the new semantics, which additionally affects cpu_throw(). o The high FP registers are either in the PCB or on some CPU. context switching for them is done lazily. This affects trap(). o The mcontext has room for all registers, but not all of them have to be defined in all cases. This mostly affects signal delivery code now. The *context syscalls are as of yet still unimplemented. Many details went into the removal of the requirement to use contigmalloc for kernel stacks. The details are mostly CPU specific and limited to exception_save() and exception_restore(). The few places where we create, destroy or switch stacks were mostly simplified by not having to construct physical addresses and additionally saving the virtual addresses for later use. Besides more efficient context saving and restoring, which of course yields a noticable speedup, this also fixes the dreaded SMP bootup problem as a side-effect. The details of which are still not fully understood. This change includes all the necessary backward compatibility code to have it handle older userland binaries that use the break instruction for syscalls. Support for break-based syscalls has been pessimized in favor of a clean implementation. Due to the overall better performance of the kernel, this will still be notived as an improvement if it's noticed at all. Approved by: re@ (jhb)
2003-05-16 21:26:42 +00:00
#include <machine/frame.h>
#include <machine/md_var.h>
#include <machine/pcb.h>
#include <vm/vm.h>
#include <vm/vm_kern.h>
#include <vm/vm_param.h>
#include <vm/pmap.h>
#include <vm/vm_map.h>
#include <vm/vm_object.h>
#include <vm/vm_extern.h>
#include <ia64/ia32/ia32_util.h>
#include <i386/include/psl.h>
#include <i386/include/segments.h>
#include <i386/include/specialreg.h>
static register_t *ia32_copyout_strings(struct image_params *imgp);
static void ia32_setregs(struct thread *td, u_long entry, u_long stack,
u_long ps_strings);
extern struct sysent ia32_sysent[];
static char ia32_sigcode[] = {
0xff, 0x54, 0x24, 0x10, /* call *SIGF_HANDLER(%esp) */
0x8d, 0x44, 0x24, 0x14, /* lea SIGF_UC(%esp),%eax */
0x50, /* pushl %eax */
0xf7, 0x40, 0x54, 0x00, 0x00, 0x02, 0x02, /* testl $PSL_VM,UC_EFLAGS(%eax) */
0x75, 0x03, /* jne 9f */
0x8e, 0x68, 0x14, /* movl UC_GS(%eax),%gs */
0xb8, 0x57, 0x01, 0x00, 0x00, /* 9: movl $SYS_sigreturn,%eax */
0x50, /* pushl %eax */
0xcd, 0x80, /* int $0x80 */
0xeb, 0xfe, /* 0: jmp 0b */
Revamp of the syscall path, exception and context handling. The prime objectives are: o Implement a syscall path based on the epc inststruction (see sys/ia64/ia64/syscall.s). o Revisit the places were we need to save and restore registers and define those contexts in terms of the register sets (see sys/ia64/include/_regset.h). Secundairy objectives: o Remove the requirement to use contigmalloc for kernel stacks. o Better handling of the high FP registers for SMP systems. o Switch to the new cpu_switch() and cpu_throw() semantics. o Add a good unwinder to reconstruct contexts for the rare cases we need to (see sys/contrib/ia64/libuwx) Many files are affected by this change. Functionally it boils down to: o The EPC syscall doesn't preserve registers it does not need to preserve and places the arguments differently on the stack. This affects libc and truss. o The address of the kernel page directory (kptdir) had to be unstaticized for use by the nested TLB fault handler. The name has been changed to ia64_kptdir to avoid conflicts. The renaming affects libkvm. o The trapframe only contains the special registers and the scratch registers. For syscalls using the EPC syscall path no scratch registers are saved. This affects all places where the trapframe is accessed. Most notably the unaligned access handler, the signal delivery code and the debugger. o Context switching only partly saves the special registers and the preserved registers. This affects cpu_switch() and triggered the move to the new semantics, which additionally affects cpu_throw(). o The high FP registers are either in the PCB or on some CPU. context switching for them is done lazily. This affects trap(). o The mcontext has room for all registers, but not all of them have to be defined in all cases. This mostly affects signal delivery code now. The *context syscalls are as of yet still unimplemented. Many details went into the removal of the requirement to use contigmalloc for kernel stacks. The details are mostly CPU specific and limited to exception_save() and exception_restore(). The few places where we create, destroy or switch stacks were mostly simplified by not having to construct physical addresses and additionally saving the virtual addresses for later use. Besides more efficient context saving and restoring, which of course yields a noticable speedup, this also fixes the dreaded SMP bootup problem as a side-effect. The details of which are still not fully understood. This change includes all the necessary backward compatibility code to have it handle older userland binaries that use the break instruction for syscalls. Support for break-based syscalls has been pessimized in favor of a clean implementation. Due to the overall better performance of the kernel, this will still be notived as an improvement if it's noticed at all. Approved by: re@ (jhb)
2003-05-16 21:26:42 +00:00
0
};
Revamp of the syscall path, exception and context handling. The prime objectives are: o Implement a syscall path based on the epc inststruction (see sys/ia64/ia64/syscall.s). o Revisit the places were we need to save and restore registers and define those contexts in terms of the register sets (see sys/ia64/include/_regset.h). Secundairy objectives: o Remove the requirement to use contigmalloc for kernel stacks. o Better handling of the high FP registers for SMP systems. o Switch to the new cpu_switch() and cpu_throw() semantics. o Add a good unwinder to reconstruct contexts for the rare cases we need to (see sys/contrib/ia64/libuwx) Many files are affected by this change. Functionally it boils down to: o The EPC syscall doesn't preserve registers it does not need to preserve and places the arguments differently on the stack. This affects libc and truss. o The address of the kernel page directory (kptdir) had to be unstaticized for use by the nested TLB fault handler. The name has been changed to ia64_kptdir to avoid conflicts. The renaming affects libkvm. o The trapframe only contains the special registers and the scratch registers. For syscalls using the EPC syscall path no scratch registers are saved. This affects all places where the trapframe is accessed. Most notably the unaligned access handler, the signal delivery code and the debugger. o Context switching only partly saves the special registers and the preserved registers. This affects cpu_switch() and triggered the move to the new semantics, which additionally affects cpu_throw(). o The high FP registers are either in the PCB or on some CPU. context switching for them is done lazily. This affects trap(). o The mcontext has room for all registers, but not all of them have to be defined in all cases. This mostly affects signal delivery code now. The *context syscalls are as of yet still unimplemented. Many details went into the removal of the requirement to use contigmalloc for kernel stacks. The details are mostly CPU specific and limited to exception_save() and exception_restore(). The few places where we create, destroy or switch stacks were mostly simplified by not having to construct physical addresses and additionally saving the virtual addresses for later use. Besides more efficient context saving and restoring, which of course yields a noticable speedup, this also fixes the dreaded SMP bootup problem as a side-effect. The details of which are still not fully understood. This change includes all the necessary backward compatibility code to have it handle older userland binaries that use the break instruction for syscalls. Support for break-based syscalls has been pessimized in favor of a clean implementation. Due to the overall better performance of the kernel, this will still be notived as an improvement if it's noticed at all. Approved by: re@ (jhb)
2003-05-16 21:26:42 +00:00
static int ia32_szsigcode = sizeof(ia32_sigcode);
struct sysentvec ia32_freebsd_sysvec = {
SYS_MAXSYSCALL,
ia32_sysent,
0,
0,
NULL,
0,
NULL,
NULL,
elf32_freebsd_fixup,
sendsig,
ia32_sigcode,
&ia32_szsigcode,
NULL,
"FreeBSD ELF",
elf32_coredump,
NULL,
Revamp of the syscall path, exception and context handling. The prime objectives are: o Implement a syscall path based on the epc inststruction (see sys/ia64/ia64/syscall.s). o Revisit the places were we need to save and restore registers and define those contexts in terms of the register sets (see sys/ia64/include/_regset.h). Secundairy objectives: o Remove the requirement to use contigmalloc for kernel stacks. o Better handling of the high FP registers for SMP systems. o Switch to the new cpu_switch() and cpu_throw() semantics. o Add a good unwinder to reconstruct contexts for the rare cases we need to (see sys/contrib/ia64/libuwx) Many files are affected by this change. Functionally it boils down to: o The EPC syscall doesn't preserve registers it does not need to preserve and places the arguments differently on the stack. This affects libc and truss. o The address of the kernel page directory (kptdir) had to be unstaticized for use by the nested TLB fault handler. The name has been changed to ia64_kptdir to avoid conflicts. The renaming affects libkvm. o The trapframe only contains the special registers and the scratch registers. For syscalls using the EPC syscall path no scratch registers are saved. This affects all places where the trapframe is accessed. Most notably the unaligned access handler, the signal delivery code and the debugger. o Context switching only partly saves the special registers and the preserved registers. This affects cpu_switch() and triggered the move to the new semantics, which additionally affects cpu_throw(). o The high FP registers are either in the PCB or on some CPU. context switching for them is done lazily. This affects trap(). o The mcontext has room for all registers, but not all of them have to be defined in all cases. This mostly affects signal delivery code now. The *context syscalls are as of yet still unimplemented. Many details went into the removal of the requirement to use contigmalloc for kernel stacks. The details are mostly CPU specific and limited to exception_save() and exception_restore(). The few places where we create, destroy or switch stacks were mostly simplified by not having to construct physical addresses and additionally saving the virtual addresses for later use. Besides more efficient context saving and restoring, which of course yields a noticable speedup, this also fixes the dreaded SMP bootup problem as a side-effect. The details of which are still not fully understood. This change includes all the necessary backward compatibility code to have it handle older userland binaries that use the break instruction for syscalls. Support for break-based syscalls has been pessimized in favor of a clean implementation. Due to the overall better performance of the kernel, this will still be notived as an improvement if it's noticed at all. Approved by: re@ (jhb)
2003-05-16 21:26:42 +00:00
IA32_MINSIGSTKSZ,
IA32_PAGE_SIZE,
0,
IA32_USRSTACK,
IA32_USRSTACK,
IA32_PS_STRINGS,
VM_PROT_ALL,
ia32_copyout_strings,
ia32_setregs
};
static Elf32_Brandinfo ia32_brand_info = {
ELFOSABI_FREEBSD,
EM_386,
"FreeBSD",
"/compat/ia32",
"/usr/libexec/ld-elf.so.1",
&ia32_freebsd_sysvec
};
SYSINIT(ia32, SI_SUB_EXEC, SI_ORDER_ANY,
(sysinit_cfunc_t) elf32_insert_brand_entry,
&ia32_brand_info);
static register_t *
ia32_copyout_strings(struct image_params *imgp)
{
int argc, envc;
u_int32_t *vectp;
char *stringp, *destp;
u_int32_t *stack_base;
struct ia32_ps_strings *arginfo;
int szsigcode;
/*
* Calculate string base and vector table pointers.
* Also deal with signal trampoline code for this exec type.
*/
arginfo = (struct ia32_ps_strings *)IA32_PS_STRINGS;
szsigcode = *(imgp->proc->p_sysent->sv_szsigcode);
Revamp of the syscall path, exception and context handling. The prime objectives are: o Implement a syscall path based on the epc inststruction (see sys/ia64/ia64/syscall.s). o Revisit the places were we need to save and restore registers and define those contexts in terms of the register sets (see sys/ia64/include/_regset.h). Secundairy objectives: o Remove the requirement to use contigmalloc for kernel stacks. o Better handling of the high FP registers for SMP systems. o Switch to the new cpu_switch() and cpu_throw() semantics. o Add a good unwinder to reconstruct contexts for the rare cases we need to (see sys/contrib/ia64/libuwx) Many files are affected by this change. Functionally it boils down to: o The EPC syscall doesn't preserve registers it does not need to preserve and places the arguments differently on the stack. This affects libc and truss. o The address of the kernel page directory (kptdir) had to be unstaticized for use by the nested TLB fault handler. The name has been changed to ia64_kptdir to avoid conflicts. The renaming affects libkvm. o The trapframe only contains the special registers and the scratch registers. For syscalls using the EPC syscall path no scratch registers are saved. This affects all places where the trapframe is accessed. Most notably the unaligned access handler, the signal delivery code and the debugger. o Context switching only partly saves the special registers and the preserved registers. This affects cpu_switch() and triggered the move to the new semantics, which additionally affects cpu_throw(). o The high FP registers are either in the PCB or on some CPU. context switching for them is done lazily. This affects trap(). o The mcontext has room for all registers, but not all of them have to be defined in all cases. This mostly affects signal delivery code now. The *context syscalls are as of yet still unimplemented. Many details went into the removal of the requirement to use contigmalloc for kernel stacks. The details are mostly CPU specific and limited to exception_save() and exception_restore(). The few places where we create, destroy or switch stacks were mostly simplified by not having to construct physical addresses and additionally saving the virtual addresses for later use. Besides more efficient context saving and restoring, which of course yields a noticable speedup, this also fixes the dreaded SMP bootup problem as a side-effect. The details of which are still not fully understood. This change includes all the necessary backward compatibility code to have it handle older userland binaries that use the break instruction for syscalls. Support for break-based syscalls has been pessimized in favor of a clean implementation. Due to the overall better performance of the kernel, this will still be notived as an improvement if it's noticed at all. Approved by: re@ (jhb)
2003-05-16 21:26:42 +00:00
destp = (caddr_t)arginfo - szsigcode - IA32_USRSPACE -
roundup((ARG_MAX - imgp->stringspace), sizeof(char *));
/*
* install sigcode
*/
if (szsigcode)
copyout(imgp->proc->p_sysent->sv_sigcode,
((caddr_t)arginfo - szsigcode), szsigcode);
/*
* If we have a valid auxargs ptr, prepare some room
* on the stack.
*/
if (imgp->auxargs) {
/*
* 'AT_COUNT*2' is size for the ELF Auxargs data. This is for
* lower compatibility.
*/
imgp->auxarg_size = (imgp->auxarg_size) ? imgp->auxarg_size
: (AT_COUNT * 2);
/*
* The '+ 2' is for the null pointers at the end of each of
* the arg and env vector sets,and imgp->auxarg_size is room
* for argument of Runtime loader.
*/
vectp = (u_int32_t *) (destp - (imgp->argc + imgp->envc + 2 +
imgp->auxarg_size) * sizeof(u_int32_t));
} else
/*
* The '+ 2' is for the null pointers at the end of each of
* the arg and env vector sets
*/
vectp = (u_int32_t *)
(destp - (imgp->argc + imgp->envc + 2) * sizeof(u_int32_t));
/*
* vectp also becomes our initial stack base
*/
Revamp of the syscall path, exception and context handling. The prime objectives are: o Implement a syscall path based on the epc inststruction (see sys/ia64/ia64/syscall.s). o Revisit the places were we need to save and restore registers and define those contexts in terms of the register sets (see sys/ia64/include/_regset.h). Secundairy objectives: o Remove the requirement to use contigmalloc for kernel stacks. o Better handling of the high FP registers for SMP systems. o Switch to the new cpu_switch() and cpu_throw() semantics. o Add a good unwinder to reconstruct contexts for the rare cases we need to (see sys/contrib/ia64/libuwx) Many files are affected by this change. Functionally it boils down to: o The EPC syscall doesn't preserve registers it does not need to preserve and places the arguments differently on the stack. This affects libc and truss. o The address of the kernel page directory (kptdir) had to be unstaticized for use by the nested TLB fault handler. The name has been changed to ia64_kptdir to avoid conflicts. The renaming affects libkvm. o The trapframe only contains the special registers and the scratch registers. For syscalls using the EPC syscall path no scratch registers are saved. This affects all places where the trapframe is accessed. Most notably the unaligned access handler, the signal delivery code and the debugger. o Context switching only partly saves the special registers and the preserved registers. This affects cpu_switch() and triggered the move to the new semantics, which additionally affects cpu_throw(). o The high FP registers are either in the PCB or on some CPU. context switching for them is done lazily. This affects trap(). o The mcontext has room for all registers, but not all of them have to be defined in all cases. This mostly affects signal delivery code now. The *context syscalls are as of yet still unimplemented. Many details went into the removal of the requirement to use contigmalloc for kernel stacks. The details are mostly CPU specific and limited to exception_save() and exception_restore(). The few places where we create, destroy or switch stacks were mostly simplified by not having to construct physical addresses and additionally saving the virtual addresses for later use. Besides more efficient context saving and restoring, which of course yields a noticable speedup, this also fixes the dreaded SMP bootup problem as a side-effect. The details of which are still not fully understood. This change includes all the necessary backward compatibility code to have it handle older userland binaries that use the break instruction for syscalls. Support for break-based syscalls has been pessimized in favor of a clean implementation. Due to the overall better performance of the kernel, this will still be notived as an improvement if it's noticed at all. Approved by: re@ (jhb)
2003-05-16 21:26:42 +00:00
vectp = (void*)((uintptr_t)vectp & ~15);
stack_base = vectp;
stringp = imgp->stringbase;
argc = imgp->argc;
envc = imgp->envc;
/*
* Copy out strings - arguments and environment.
*/
copyout(stringp, destp, ARG_MAX - imgp->stringspace);
/*
* Fill in "ps_strings" struct for ps, w, etc.
*/
suword32(&arginfo->ps_argvstr, (u_int32_t)(intptr_t)vectp);
suword32(&arginfo->ps_nargvstr, argc);
/*
* Fill in argument portion of vector table.
*/
for (; argc > 0; --argc) {
suword32(vectp++, (u_int32_t)(intptr_t)destp);
while (*stringp++ != 0)
destp++;
destp++;
}
/* a null vector table pointer separates the argp's from the envp's */
suword32(vectp++, 0);
suword32(&arginfo->ps_envstr, (u_int32_t)(intptr_t)vectp);
suword32(&arginfo->ps_nenvstr, envc);
/*
* Fill in environment portion of vector table.
*/
for (; envc > 0; --envc) {
suword32(vectp++, (u_int32_t)(intptr_t)destp);
while (*stringp++ != 0)
destp++;
destp++;
}
/* end of vector table is a null pointer */
suword32(vectp, 0);
return ((register_t *)stack_base);
}
static void
ia32_setregs(struct thread *td, u_long entry, u_long stack, u_long ps_strings)
{
Revamp of the syscall path, exception and context handling. The prime objectives are: o Implement a syscall path based on the epc inststruction (see sys/ia64/ia64/syscall.s). o Revisit the places were we need to save and restore registers and define those contexts in terms of the register sets (see sys/ia64/include/_regset.h). Secundairy objectives: o Remove the requirement to use contigmalloc for kernel stacks. o Better handling of the high FP registers for SMP systems. o Switch to the new cpu_switch() and cpu_throw() semantics. o Add a good unwinder to reconstruct contexts for the rare cases we need to (see sys/contrib/ia64/libuwx) Many files are affected by this change. Functionally it boils down to: o The EPC syscall doesn't preserve registers it does not need to preserve and places the arguments differently on the stack. This affects libc and truss. o The address of the kernel page directory (kptdir) had to be unstaticized for use by the nested TLB fault handler. The name has been changed to ia64_kptdir to avoid conflicts. The renaming affects libkvm. o The trapframe only contains the special registers and the scratch registers. For syscalls using the EPC syscall path no scratch registers are saved. This affects all places where the trapframe is accessed. Most notably the unaligned access handler, the signal delivery code and the debugger. o Context switching only partly saves the special registers and the preserved registers. This affects cpu_switch() and triggered the move to the new semantics, which additionally affects cpu_throw(). o The high FP registers are either in the PCB or on some CPU. context switching for them is done lazily. This affects trap(). o The mcontext has room for all registers, but not all of them have to be defined in all cases. This mostly affects signal delivery code now. The *context syscalls are as of yet still unimplemented. Many details went into the removal of the requirement to use contigmalloc for kernel stacks. The details are mostly CPU specific and limited to exception_save() and exception_restore(). The few places where we create, destroy or switch stacks were mostly simplified by not having to construct physical addresses and additionally saving the virtual addresses for later use. Besides more efficient context saving and restoring, which of course yields a noticable speedup, this also fixes the dreaded SMP bootup problem as a side-effect. The details of which are still not fully understood. This change includes all the necessary backward compatibility code to have it handle older userland binaries that use the break instruction for syscalls. Support for break-based syscalls has been pessimized in favor of a clean implementation. Due to the overall better performance of the kernel, this will still be notived as an improvement if it's noticed at all. Approved by: re@ (jhb)
2003-05-16 21:26:42 +00:00
struct trapframe *tf = td->td_frame;
vm_offset_t gdt, ldt;
u_int64_t codesel, datasel, ldtsel;
u_int64_t codeseg, dataseg, gdtseg, ldtseg;
struct segment_descriptor desc;
struct vmspace *vmspace = td->td_proc->p_vmspace;
Revamp of the syscall path, exception and context handling. The prime objectives are: o Implement a syscall path based on the epc inststruction (see sys/ia64/ia64/syscall.s). o Revisit the places were we need to save and restore registers and define those contexts in terms of the register sets (see sys/ia64/include/_regset.h). Secundairy objectives: o Remove the requirement to use contigmalloc for kernel stacks. o Better handling of the high FP registers for SMP systems. o Switch to the new cpu_switch() and cpu_throw() semantics. o Add a good unwinder to reconstruct contexts for the rare cases we need to (see sys/contrib/ia64/libuwx) Many files are affected by this change. Functionally it boils down to: o The EPC syscall doesn't preserve registers it does not need to preserve and places the arguments differently on the stack. This affects libc and truss. o The address of the kernel page directory (kptdir) had to be unstaticized for use by the nested TLB fault handler. The name has been changed to ia64_kptdir to avoid conflicts. The renaming affects libkvm. o The trapframe only contains the special registers and the scratch registers. For syscalls using the EPC syscall path no scratch registers are saved. This affects all places where the trapframe is accessed. Most notably the unaligned access handler, the signal delivery code and the debugger. o Context switching only partly saves the special registers and the preserved registers. This affects cpu_switch() and triggered the move to the new semantics, which additionally affects cpu_throw(). o The high FP registers are either in the PCB or on some CPU. context switching for them is done lazily. This affects trap(). o The mcontext has room for all registers, but not all of them have to be defined in all cases. This mostly affects signal delivery code now. The *context syscalls are as of yet still unimplemented. Many details went into the removal of the requirement to use contigmalloc for kernel stacks. The details are mostly CPU specific and limited to exception_save() and exception_restore(). The few places where we create, destroy or switch stacks were mostly simplified by not having to construct physical addresses and additionally saving the virtual addresses for later use. Besides more efficient context saving and restoring, which of course yields a noticable speedup, this also fixes the dreaded SMP bootup problem as a side-effect. The details of which are still not fully understood. This change includes all the necessary backward compatibility code to have it handle older userland binaries that use the break instruction for syscalls. Support for break-based syscalls has been pessimized in favor of a clean implementation. Due to the overall better performance of the kernel, this will still be notived as an improvement if it's noticed at all. Approved by: re@ (jhb)
2003-05-16 21:26:42 +00:00
exec_setregs(td, entry, stack, ps_strings);
/* Non-syscall frames are cleared by exec_setregs() */
if (tf->tf_flags & FRAME_SYSCALL) {
bzero(&tf->tf_scratch, sizeof(tf->tf_scratch));
bzero(&tf->tf_scratch_fp, sizeof(tf->tf_scratch_fp));
} else
tf->tf_special.ndirty = 0;
tf->tf_special.psr |= IA64_PSR_IS;
tf->tf_special.sp = stack;
/* Point the RSE backstore to something harmless. */
tf->tf_special.bspstore = (IA32_PS_STRINGS - ia32_szsigcode -
IA32_USRSPACE + 15) & ~15;
codesel = LSEL(LUCODE_SEL, SEL_UPL);
datasel = LSEL(LUDATA_SEL, SEL_UPL);
ldtsel = GSEL(GLDT_SEL, SEL_UPL);
Revamp of the syscall path, exception and context handling. The prime objectives are: o Implement a syscall path based on the epc inststruction (see sys/ia64/ia64/syscall.s). o Revisit the places were we need to save and restore registers and define those contexts in terms of the register sets (see sys/ia64/include/_regset.h). Secundairy objectives: o Remove the requirement to use contigmalloc for kernel stacks. o Better handling of the high FP registers for SMP systems. o Switch to the new cpu_switch() and cpu_throw() semantics. o Add a good unwinder to reconstruct contexts for the rare cases we need to (see sys/contrib/ia64/libuwx) Many files are affected by this change. Functionally it boils down to: o The EPC syscall doesn't preserve registers it does not need to preserve and places the arguments differently on the stack. This affects libc and truss. o The address of the kernel page directory (kptdir) had to be unstaticized for use by the nested TLB fault handler. The name has been changed to ia64_kptdir to avoid conflicts. The renaming affects libkvm. o The trapframe only contains the special registers and the scratch registers. For syscalls using the EPC syscall path no scratch registers are saved. This affects all places where the trapframe is accessed. Most notably the unaligned access handler, the signal delivery code and the debugger. o Context switching only partly saves the special registers and the preserved registers. This affects cpu_switch() and triggered the move to the new semantics, which additionally affects cpu_throw(). o The high FP registers are either in the PCB or on some CPU. context switching for them is done lazily. This affects trap(). o The mcontext has room for all registers, but not all of them have to be defined in all cases. This mostly affects signal delivery code now. The *context syscalls are as of yet still unimplemented. Many details went into the removal of the requirement to use contigmalloc for kernel stacks. The details are mostly CPU specific and limited to exception_save() and exception_restore(). The few places where we create, destroy or switch stacks were mostly simplified by not having to construct physical addresses and additionally saving the virtual addresses for later use. Besides more efficient context saving and restoring, which of course yields a noticable speedup, this also fixes the dreaded SMP bootup problem as a side-effect. The details of which are still not fully understood. This change includes all the necessary backward compatibility code to have it handle older userland binaries that use the break instruction for syscalls. Support for break-based syscalls has been pessimized in favor of a clean implementation. Due to the overall better performance of the kernel, this will still be notived as an improvement if it's noticed at all. Approved by: re@ (jhb)
2003-05-16 21:26:42 +00:00
/* Setup ia32 segment registers. */
tf->tf_scratch.gr16 = (datasel << 48) | (datasel << 32) |
(datasel << 16) | datasel;
tf->tf_scratch.gr17 = (ldtsel << 32) | (datasel << 16) | codesel;
/*
* Build the GDT and LDT.
*/
gdt = IA32_USRSTACK;
Revamp of the syscall path, exception and context handling. The prime objectives are: o Implement a syscall path based on the epc inststruction (see sys/ia64/ia64/syscall.s). o Revisit the places were we need to save and restore registers and define those contexts in terms of the register sets (see sys/ia64/include/_regset.h). Secundairy objectives: o Remove the requirement to use contigmalloc for kernel stacks. o Better handling of the high FP registers for SMP systems. o Switch to the new cpu_switch() and cpu_throw() semantics. o Add a good unwinder to reconstruct contexts for the rare cases we need to (see sys/contrib/ia64/libuwx) Many files are affected by this change. Functionally it boils down to: o The EPC syscall doesn't preserve registers it does not need to preserve and places the arguments differently on the stack. This affects libc and truss. o The address of the kernel page directory (kptdir) had to be unstaticized for use by the nested TLB fault handler. The name has been changed to ia64_kptdir to avoid conflicts. The renaming affects libkvm. o The trapframe only contains the special registers and the scratch registers. For syscalls using the EPC syscall path no scratch registers are saved. This affects all places where the trapframe is accessed. Most notably the unaligned access handler, the signal delivery code and the debugger. o Context switching only partly saves the special registers and the preserved registers. This affects cpu_switch() and triggered the move to the new semantics, which additionally affects cpu_throw(). o The high FP registers are either in the PCB or on some CPU. context switching for them is done lazily. This affects trap(). o The mcontext has room for all registers, but not all of them have to be defined in all cases. This mostly affects signal delivery code now. The *context syscalls are as of yet still unimplemented. Many details went into the removal of the requirement to use contigmalloc for kernel stacks. The details are mostly CPU specific and limited to exception_save() and exception_restore(). The few places where we create, destroy or switch stacks were mostly simplified by not having to construct physical addresses and additionally saving the virtual addresses for later use. Besides more efficient context saving and restoring, which of course yields a noticable speedup, this also fixes the dreaded SMP bootup problem as a side-effect. The details of which are still not fully understood. This change includes all the necessary backward compatibility code to have it handle older userland binaries that use the break instruction for syscalls. Support for break-based syscalls has been pessimized in favor of a clean implementation. Due to the overall better performance of the kernel, this will still be notived as an improvement if it's noticed at all. Approved by: re@ (jhb)
2003-05-16 21:26:42 +00:00
vm_map_find(&vmspace->vm_map, 0, 0, &gdt, IA32_PAGE_SIZE << 1, 0,
VM_PROT_ALL, VM_PROT_ALL, 0);
ldt = gdt + IA32_PAGE_SIZE;
desc.sd_lolimit = 8*NLDT-1;
desc.sd_lobase = ldt & 0xffffff;
desc.sd_type = SDT_SYSLDT;
desc.sd_dpl = SEL_UPL;
desc.sd_p = 1;
desc.sd_hilimit = 0;
desc.sd_def32 = 0;
desc.sd_gran = 0;
desc.sd_hibase = ldt >> 24;
copyout(&desc, (caddr_t) gdt + 8*GLDT_SEL, sizeof(desc));
desc.sd_lolimit = ((IA32_USRSTACK >> 12) - 1) & 0xffff;
desc.sd_lobase = 0;
desc.sd_type = SDT_MEMERA;
desc.sd_dpl = SEL_UPL;
desc.sd_p = 1;
desc.sd_hilimit = ((IA32_USRSTACK >> 12) - 1) >> 16;
desc.sd_def32 = 1;
desc.sd_gran = 1;
desc.sd_hibase = 0;
copyout(&desc, (caddr_t) ldt + 8*LUCODE_SEL, sizeof(desc));
desc.sd_type = SDT_MEMRWA;
copyout(&desc, (caddr_t) ldt + 8*LUDATA_SEL, sizeof(desc));
codeseg = 0 /* base */
+ (((IA32_USRSTACK >> 12) - 1) << 32) /* limit */
+ ((long)SDT_MEMERA << 52)
+ ((long)SEL_UPL << 57)
+ (1L << 59) /* present */
+ (1L << 62) /* 32 bits */
+ (1L << 63); /* page granularity */
dataseg = 0 /* base */
+ (((IA32_USRSTACK >> 12) - 1) << 32) /* limit */
+ ((long)SDT_MEMRWA << 52)
+ ((long)SEL_UPL << 57)
+ (1L << 59) /* present */
+ (1L << 62) /* 32 bits */
+ (1L << 63); /* page granularity */
Revamp of the syscall path, exception and context handling. The prime objectives are: o Implement a syscall path based on the epc inststruction (see sys/ia64/ia64/syscall.s). o Revisit the places were we need to save and restore registers and define those contexts in terms of the register sets (see sys/ia64/include/_regset.h). Secundairy objectives: o Remove the requirement to use contigmalloc for kernel stacks. o Better handling of the high FP registers for SMP systems. o Switch to the new cpu_switch() and cpu_throw() semantics. o Add a good unwinder to reconstruct contexts for the rare cases we need to (see sys/contrib/ia64/libuwx) Many files are affected by this change. Functionally it boils down to: o The EPC syscall doesn't preserve registers it does not need to preserve and places the arguments differently on the stack. This affects libc and truss. o The address of the kernel page directory (kptdir) had to be unstaticized for use by the nested TLB fault handler. The name has been changed to ia64_kptdir to avoid conflicts. The renaming affects libkvm. o The trapframe only contains the special registers and the scratch registers. For syscalls using the EPC syscall path no scratch registers are saved. This affects all places where the trapframe is accessed. Most notably the unaligned access handler, the signal delivery code and the debugger. o Context switching only partly saves the special registers and the preserved registers. This affects cpu_switch() and triggered the move to the new semantics, which additionally affects cpu_throw(). o The high FP registers are either in the PCB or on some CPU. context switching for them is done lazily. This affects trap(). o The mcontext has room for all registers, but not all of them have to be defined in all cases. This mostly affects signal delivery code now. The *context syscalls are as of yet still unimplemented. Many details went into the removal of the requirement to use contigmalloc for kernel stacks. The details are mostly CPU specific and limited to exception_save() and exception_restore(). The few places where we create, destroy or switch stacks were mostly simplified by not having to construct physical addresses and additionally saving the virtual addresses for later use. Besides more efficient context saving and restoring, which of course yields a noticable speedup, this also fixes the dreaded SMP bootup problem as a side-effect. The details of which are still not fully understood. This change includes all the necessary backward compatibility code to have it handle older userland binaries that use the break instruction for syscalls. Support for break-based syscalls has been pessimized in favor of a clean implementation. Due to the overall better performance of the kernel, this will still be notived as an improvement if it's noticed at all. Approved by: re@ (jhb)
2003-05-16 21:26:42 +00:00
tf->tf_scratch.csd = codeseg;
tf->tf_scratch.ssd = dataseg;
tf->tf_scratch.gr24 = dataseg; /* ESD */
tf->tf_scratch.gr27 = dataseg; /* DSD */
tf->tf_scratch.gr28 = dataseg; /* FSD */
tf->tf_scratch.gr29 = dataseg; /* GSD */
gdtseg = gdt /* base */
+ ((8L*NGDT - 1) << 32) /* limit */
+ ((long)SDT_SYSNULL << 52)
+ ((long)SEL_UPL << 57)
+ (1L << 59) /* present */
+ (0L << 62) /* 16 bits */
+ (0L << 63); /* byte granularity */
ldtseg = ldt /* base */
+ ((8L*NLDT - 1) << 32) /* limit */
+ ((long)SDT_SYSLDT << 52)
+ ((long)SEL_UPL << 57)
+ (1L << 59) /* present */
+ (0L << 62) /* 16 bits */
+ (0L << 63); /* byte granularity */
Revamp of the syscall path, exception and context handling. The prime objectives are: o Implement a syscall path based on the epc inststruction (see sys/ia64/ia64/syscall.s). o Revisit the places were we need to save and restore registers and define those contexts in terms of the register sets (see sys/ia64/include/_regset.h). Secundairy objectives: o Remove the requirement to use contigmalloc for kernel stacks. o Better handling of the high FP registers for SMP systems. o Switch to the new cpu_switch() and cpu_throw() semantics. o Add a good unwinder to reconstruct contexts for the rare cases we need to (see sys/contrib/ia64/libuwx) Many files are affected by this change. Functionally it boils down to: o The EPC syscall doesn't preserve registers it does not need to preserve and places the arguments differently on the stack. This affects libc and truss. o The address of the kernel page directory (kptdir) had to be unstaticized for use by the nested TLB fault handler. The name has been changed to ia64_kptdir to avoid conflicts. The renaming affects libkvm. o The trapframe only contains the special registers and the scratch registers. For syscalls using the EPC syscall path no scratch registers are saved. This affects all places where the trapframe is accessed. Most notably the unaligned access handler, the signal delivery code and the debugger. o Context switching only partly saves the special registers and the preserved registers. This affects cpu_switch() and triggered the move to the new semantics, which additionally affects cpu_throw(). o The high FP registers are either in the PCB or on some CPU. context switching for them is done lazily. This affects trap(). o The mcontext has room for all registers, but not all of them have to be defined in all cases. This mostly affects signal delivery code now. The *context syscalls are as of yet still unimplemented. Many details went into the removal of the requirement to use contigmalloc for kernel stacks. The details are mostly CPU specific and limited to exception_save() and exception_restore(). The few places where we create, destroy or switch stacks were mostly simplified by not having to construct physical addresses and additionally saving the virtual addresses for later use. Besides more efficient context saving and restoring, which of course yields a noticable speedup, this also fixes the dreaded SMP bootup problem as a side-effect. The details of which are still not fully understood. This change includes all the necessary backward compatibility code to have it handle older userland binaries that use the break instruction for syscalls. Support for break-based syscalls has been pessimized in favor of a clean implementation. Due to the overall better performance of the kernel, this will still be notived as an improvement if it's noticed at all. Approved by: re@ (jhb)
2003-05-16 21:26:42 +00:00
tf->tf_scratch.gr30 = ldtseg; /* LDTD */
tf->tf_scratch.gr31 = gdtseg; /* GDTD */
/* Set ia32 control registers on this processor. */
ia64_set_cflg(CR0_PE | CR0_PG | ((long)(CR4_XMM | CR4_FXSR) << 32));
ia64_set_eflag(PSL_USER);
/* PS_STRINGS value for BSD/OS binaries. It is 0 for non-BSD/OS. */
Revamp of the syscall path, exception and context handling. The prime objectives are: o Implement a syscall path based on the epc inststruction (see sys/ia64/ia64/syscall.s). o Revisit the places were we need to save and restore registers and define those contexts in terms of the register sets (see sys/ia64/include/_regset.h). Secundairy objectives: o Remove the requirement to use contigmalloc for kernel stacks. o Better handling of the high FP registers for SMP systems. o Switch to the new cpu_switch() and cpu_throw() semantics. o Add a good unwinder to reconstruct contexts for the rare cases we need to (see sys/contrib/ia64/libuwx) Many files are affected by this change. Functionally it boils down to: o The EPC syscall doesn't preserve registers it does not need to preserve and places the arguments differently on the stack. This affects libc and truss. o The address of the kernel page directory (kptdir) had to be unstaticized for use by the nested TLB fault handler. The name has been changed to ia64_kptdir to avoid conflicts. The renaming affects libkvm. o The trapframe only contains the special registers and the scratch registers. For syscalls using the EPC syscall path no scratch registers are saved. This affects all places where the trapframe is accessed. Most notably the unaligned access handler, the signal delivery code and the debugger. o Context switching only partly saves the special registers and the preserved registers. This affects cpu_switch() and triggered the move to the new semantics, which additionally affects cpu_throw(). o The high FP registers are either in the PCB or on some CPU. context switching for them is done lazily. This affects trap(). o The mcontext has room for all registers, but not all of them have to be defined in all cases. This mostly affects signal delivery code now. The *context syscalls are as of yet still unimplemented. Many details went into the removal of the requirement to use contigmalloc for kernel stacks. The details are mostly CPU specific and limited to exception_save() and exception_restore(). The few places where we create, destroy or switch stacks were mostly simplified by not having to construct physical addresses and additionally saving the virtual addresses for later use. Besides more efficient context saving and restoring, which of course yields a noticable speedup, this also fixes the dreaded SMP bootup problem as a side-effect. The details of which are still not fully understood. This change includes all the necessary backward compatibility code to have it handle older userland binaries that use the break instruction for syscalls. Support for break-based syscalls has been pessimized in favor of a clean implementation. Due to the overall better performance of the kernel, this will still be notived as an improvement if it's noticed at all. Approved by: re@ (jhb)
2003-05-16 21:26:42 +00:00
tf->tf_scratch.gr11 = IA32_PS_STRINGS;
/*
* XXX - Linux emulator
* Make sure sure edx is 0x0 on entry. Linux binaries depend
* on it.
*/
td->td_retval[1] = 0;
}
Revamp of the syscall path, exception and context handling. The prime objectives are: o Implement a syscall path based on the epc inststruction (see sys/ia64/ia64/syscall.s). o Revisit the places were we need to save and restore registers and define those contexts in terms of the register sets (see sys/ia64/include/_regset.h). Secundairy objectives: o Remove the requirement to use contigmalloc for kernel stacks. o Better handling of the high FP registers for SMP systems. o Switch to the new cpu_switch() and cpu_throw() semantics. o Add a good unwinder to reconstruct contexts for the rare cases we need to (see sys/contrib/ia64/libuwx) Many files are affected by this change. Functionally it boils down to: o The EPC syscall doesn't preserve registers it does not need to preserve and places the arguments differently on the stack. This affects libc and truss. o The address of the kernel page directory (kptdir) had to be unstaticized for use by the nested TLB fault handler. The name has been changed to ia64_kptdir to avoid conflicts. The renaming affects libkvm. o The trapframe only contains the special registers and the scratch registers. For syscalls using the EPC syscall path no scratch registers are saved. This affects all places where the trapframe is accessed. Most notably the unaligned access handler, the signal delivery code and the debugger. o Context switching only partly saves the special registers and the preserved registers. This affects cpu_switch() and triggered the move to the new semantics, which additionally affects cpu_throw(). o The high FP registers are either in the PCB or on some CPU. context switching for them is done lazily. This affects trap(). o The mcontext has room for all registers, but not all of them have to be defined in all cases. This mostly affects signal delivery code now. The *context syscalls are as of yet still unimplemented. Many details went into the removal of the requirement to use contigmalloc for kernel stacks. The details are mostly CPU specific and limited to exception_save() and exception_restore(). The few places where we create, destroy or switch stacks were mostly simplified by not having to construct physical addresses and additionally saving the virtual addresses for later use. Besides more efficient context saving and restoring, which of course yields a noticable speedup, this also fixes the dreaded SMP bootup problem as a side-effect. The details of which are still not fully understood. This change includes all the necessary backward compatibility code to have it handle older userland binaries that use the break instruction for syscalls. Support for break-based syscalls has been pessimized in favor of a clean implementation. Due to the overall better performance of the kernel, this will still be notived as an improvement if it's noticed at all. Approved by: re@ (jhb)
2003-05-16 21:26:42 +00:00
void
ia32_restorectx(struct pcb *pcb)
{
ia64_set_cflg(pcb->pcb_ia32_cflg);
ia64_set_eflag(pcb->pcb_ia32_eflag);
ia64_set_fcr(pcb->pcb_ia32_fcr);
ia64_set_fdr(pcb->pcb_ia32_fdr);
ia64_set_fir(pcb->pcb_ia32_fir);
ia64_set_fsr(pcb->pcb_ia32_fsr);
}
void
ia32_savectx(struct pcb *pcb)
{
pcb->pcb_ia32_cflg = ia64_get_cflg();
pcb->pcb_ia32_eflag = ia64_get_eflag();
pcb->pcb_ia32_fcr = ia64_get_fcr();
pcb->pcb_ia32_fdr = ia64_get_fdr();
pcb->pcb_ia32_fir = ia64_get_fir();
pcb->pcb_ia32_fsr = ia64_get_fsr();
}