linux/arch/x86/kernel/ptrace.c

1536 lines
38 KiB
C
Raw Normal View History

/* By Ross Biro 1/23/92 */
/*
* Pentium III FXSR, SSE support
* Gareth Hughes <gareth@valinux.com>, May 2000
*/
#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/mm.h>
#include <linux/smp.h>
#include <linux/errno.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 09:04:11 +01:00
#include <linux/slab.h>
#include <linux/ptrace.h>
#include <linux/regset.h>
#include <linux/tracehook.h>
#include <linux/user.h>
#include <linux/elf.h>
#include <linux/security.h>
#include <linux/audit.h>
#include <linux/seccomp.h>
#include <linux/signal.h>
hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events This patch rebase the implementation of the breakpoints API on top of perf events instances. Each breakpoints are now perf events that handle the register scheduling, thread/cpu attachment, etc.. The new layering is now made as follows: ptrace kgdb ftrace perf syscall \ | / / \ | / / / Core breakpoint API / / | / | / Breakpoints perf events | | Breakpoints PMU ---- Debug Register constraints handling (Part of core breakpoint API) | | Hardware debug registers Reasons of this rewrite: - Use the centralized/optimized pmu registers scheduling, implying an easier arch integration - More powerful register handling: perf attributes (pinned/flexible events, exclusive/non-exclusive, tunable period, etc...) Impact: - New perf ABI: the hardware breakpoints counters - Ptrace breakpoints setting remains tricky and still needs some per thread breakpoints references. Todo (in the order): - Support breakpoints perf counter events for perf tools (ie: implement perf_bpcounter_event()) - Support from perf tools Changes in v2: - Follow the perf "event " rename - The ptrace regression have been fixed (ptrace breakpoint perf events weren't released when a task ended) - Drop the struct hw_breakpoint and store generic fields in perf_event_attr. - Separate core and arch specific headers, drop asm-generic/hw_breakpoint.h and create linux/hw_breakpoint.h - Use new generic len/type for breakpoint - Handle off case: when breakpoints api is not supported by an arch Changes in v3: - Fix broken CONFIG_KVM, we need to propagate the breakpoint api changes to kvm when we exit the guest and restore the bp registers to the host. Changes in v4: - Drop the hw_breakpoint_restore() stub as it is only used by KVM - EXPORT_SYMBOL_GPL hw_breakpoint_restore() as KVM can be built as a module - Restore the breakpoints unconditionally on kvm guest exit: TIF_DEBUG_THREAD doesn't anymore cover every cases of running breakpoints and vcpu->arch.switch_db_regs might not always be set when the guest used debug registers. (Waiting for a reliable optimization) Changes in v5: - Split-up the asm-generic/hw-breakpoint.h moving to linux/hw_breakpoint.h into a separate patch - Optimize the breakpoints restoring while switching from kvm guest to host. We only want to restore the state if we have active breakpoints to the host, otherwise we don't care about messed-up address registers. - Add asm/hw_breakpoint.h to Kbuild - Fix bad breakpoint type in trace_selftest.c Changes in v6: - Fix wrong header inclusion in trace.h (triggered a build error with CONFIG_FTRACE_SELFTEST Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jan Kiszka <jan.kiszka@web.de> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Avi Kivity <avi@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org>
2009-09-09 19:22:48 +02:00
#include <linux/perf_event.h>
#include <linux/hw_breakpoint.h>
#include <linux/rcupdate.h>
#include <linux/export.h>
#include <linux/context_tracking.h>
#include <asm/uaccess.h>
#include <asm/pgtable.h>
#include <asm/processor.h>
#include <asm/i387.h>
#include <asm/fpu-internal.h>
#include <asm/debugreg.h>
#include <asm/ldt.h>
#include <asm/desc.h>
#include <asm/prctl.h>
#include <asm/proto.h>
#include <asm/hw_breakpoint.h>
#include <asm/traps.h>
#include "tls.h"
#define CREATE_TRACE_POINTS
#include <trace/events/syscalls.h>
enum x86_regset {
REGSET_GENERAL,
REGSET_FP,
REGSET_XFP,
REGSET_IOPERM64 = REGSET_XFP,
REGSET_XSTATE,
REGSET_TLS,
REGSET_IOPERM32,
};
x86: Add pt_regs register and stack access APIs Add following APIs for accessing registers and stack entries from pt_regs. These APIs are required by kprobes-based event tracer on ftrace. Some other debugging tools might be able to use it too. - regs_query_register_offset(const char *name) Query the offset of "name" register. - regs_query_register_name(unsigned int offset) Query the name of register by its offset. - regs_get_register(struct pt_regs *regs, unsigned int offset) Get the value of a register by its offset. - regs_within_kernel_stack(struct pt_regs *regs, unsigned long addr) Check the address is in the kernel stack. - regs_get_kernel_stack_nth(struct pt_regs *reg, unsigned int nth) Get Nth entry of the kernel stack. (N >= 0) - regs_get_argument_nth(struct pt_regs *reg, unsigned int nth) Get Nth argument at function call. (N >= 0) Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com> Cc: linux-arch@vger.kernel.org Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Avi Kivity <avi@redhat.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: Frank Ch. Eigler <fche@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jason Baron <jbaron@redhat.com> Cc: Jim Keniston <jkenisto@us.ibm.com> Cc: K.Prasad <prasad@linux.vnet.ibm.com> Cc: Lai Jiangshan <laijs@cn.fujitsu.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Przemysław Pawełczyk <przemyslaw@pawelczyk.it> Cc: Roland McGrath <roland@redhat.com> Cc: Sam Ravnborg <sam@ravnborg.org> Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Tom Zanussi <tzanussi@gmail.com> Cc: Vegard Nossum <vegard.nossum@gmail.com> LKML-Reference: <20090813203444.31965.26374.stgit@localhost.localdomain> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
2009-08-13 22:34:44 +02:00
struct pt_regs_offset {
const char *name;
int offset;
};
#define REG_OFFSET_NAME(r) {.name = #r, .offset = offsetof(struct pt_regs, r)}
#define REG_OFFSET_END {.name = NULL, .offset = 0}
static const struct pt_regs_offset regoffset_table[] = {
#ifdef CONFIG_X86_64
REG_OFFSET_NAME(r15),
REG_OFFSET_NAME(r14),
REG_OFFSET_NAME(r13),
REG_OFFSET_NAME(r12),
REG_OFFSET_NAME(r11),
REG_OFFSET_NAME(r10),
REG_OFFSET_NAME(r9),
REG_OFFSET_NAME(r8),
#endif
REG_OFFSET_NAME(bx),
REG_OFFSET_NAME(cx),
REG_OFFSET_NAME(dx),
REG_OFFSET_NAME(si),
REG_OFFSET_NAME(di),
REG_OFFSET_NAME(bp),
REG_OFFSET_NAME(ax),
#ifdef CONFIG_X86_32
REG_OFFSET_NAME(ds),
REG_OFFSET_NAME(es),
REG_OFFSET_NAME(fs),
REG_OFFSET_NAME(gs),
#endif
REG_OFFSET_NAME(orig_ax),
REG_OFFSET_NAME(ip),
REG_OFFSET_NAME(cs),
REG_OFFSET_NAME(flags),
REG_OFFSET_NAME(sp),
REG_OFFSET_NAME(ss),
REG_OFFSET_END,
};
/**
* regs_query_register_offset() - query register offset from its name
* @name: the name of a register
*
* regs_query_register_offset() returns the offset of a register in struct
* pt_regs from its name. If the name is invalid, this returns -EINVAL;
*/
int regs_query_register_offset(const char *name)
{
const struct pt_regs_offset *roff;
for (roff = regoffset_table; roff->name != NULL; roff++)
if (!strcmp(roff->name, name))
return roff->offset;
return -EINVAL;
}
/**
* regs_query_register_name() - query register name from its offset
* @offset: the offset of a register in struct pt_regs.
*
* regs_query_register_name() returns the name of a register from its
* offset in struct pt_regs. If the @offset is invalid, this returns NULL;
*/
const char *regs_query_register_name(unsigned int offset)
{
const struct pt_regs_offset *roff;
for (roff = regoffset_table; roff->name != NULL; roff++)
if (roff->offset == offset)
return roff->name;
return NULL;
}
static const int arg_offs_table[] = {
#ifdef CONFIG_X86_32
[0] = offsetof(struct pt_regs, ax),
[1] = offsetof(struct pt_regs, dx),
[2] = offsetof(struct pt_regs, cx)
#else /* CONFIG_X86_64 */
[0] = offsetof(struct pt_regs, di),
[1] = offsetof(struct pt_regs, si),
[2] = offsetof(struct pt_regs, dx),
[3] = offsetof(struct pt_regs, cx),
[4] = offsetof(struct pt_regs, r8),
[5] = offsetof(struct pt_regs, r9)
#endif
};
/*
* does not yet catch signals sent when the child dies.
* in exit.c or in signal.c.
*/
/*
* Determines which flags the user has access to [1 = access, 0 = no access].
*/
#define FLAG_MASK_32 ((unsigned long) \
(X86_EFLAGS_CF | X86_EFLAGS_PF | \
X86_EFLAGS_AF | X86_EFLAGS_ZF | \
X86_EFLAGS_SF | X86_EFLAGS_TF | \
X86_EFLAGS_DF | X86_EFLAGS_OF | \
X86_EFLAGS_RF | X86_EFLAGS_AC))
/*
* Determines whether a value may be installed in a segment register.
*/
static inline bool invalid_selector(u16 value)
{
return unlikely(value != 0 && (value & SEGMENT_RPL_MASK) != USER_RPL);
}
#ifdef CONFIG_X86_32
#define FLAG_MASK FLAG_MASK_32
x86-32: Fix invalid stack address while in softirq In 32 bit the stack address provided by kernel_stack_pointer() may point to an invalid range causing NULL pointer access or page faults while in NMI (see trace below). This happens if called in softirq context and if the stack is empty. The address at &regs->sp is then out of range. Fixing this by checking if regs and &regs->sp are in the same stack context. Otherwise return the previous stack pointer stored in struct thread_info. If that address is invalid too, return address of regs. BUG: unable to handle kernel NULL pointer dereference at 0000000a IP: [<c1004237>] print_context_stack+0x6e/0x8d *pde = 00000000 Oops: 0000 [#1] SMP Modules linked in: Pid: 4434, comm: perl Not tainted 3.6.0-rc3-oprofile-i386-standard-g4411a05 #4 Hewlett-Packard HP xw9400 Workstation/0A1Ch EIP: 0060:[<c1004237>] EFLAGS: 00010093 CPU: 0 EIP is at print_context_stack+0x6e/0x8d EAX: ffffe000 EBX: 0000000a ECX: f4435f94 EDX: 0000000a ESI: f4435f94 EDI: f4435f94 EBP: f5409ec0 ESP: f5409ea0 DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068 CR0: 8005003b CR2: 0000000a CR3: 34ac9000 CR4: 000007d0 DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000 DR6: ffff0ff0 DR7: 00000400 Process perl (pid: 4434, ti=f5408000 task=f5637850 task.ti=f4434000) Stack: 000003e8 ffffe000 00001ffc f4e39b00 00000000 0000000a f4435f94 c155198c f5409ef0 c1003723 c155198c f5409f04 00000000 f5409edc 00000000 00000000 f5409ee8 f4435f94 f5409fc4 00000001 f5409f1c c12dce1c 00000000 c155198c Call Trace: [<c1003723>] dump_trace+0x7b/0xa1 [<c12dce1c>] x86_backtrace+0x40/0x88 [<c12db712>] ? oprofile_add_sample+0x56/0x84 [<c12db731>] oprofile_add_sample+0x75/0x84 [<c12ddb5b>] op_amd_check_ctrs+0x46/0x260 [<c12dd40d>] profile_exceptions_notify+0x23/0x4c [<c1395034>] nmi_handle+0x31/0x4a [<c1029dc5>] ? ftrace_define_fields_irq_handler_entry+0x45/0x45 [<c13950ed>] do_nmi+0xa0/0x2ff [<c1029dc5>] ? ftrace_define_fields_irq_handler_entry+0x45/0x45 [<c13949e5>] nmi_stack_correct+0x28/0x2d [<c1029dc5>] ? ftrace_define_fields_irq_handler_entry+0x45/0x45 [<c1003603>] ? do_softirq+0x4b/0x7f <IRQ> [<c102a06f>] irq_exit+0x35/0x5b [<c1018f56>] smp_apic_timer_interrupt+0x6c/0x7a [<c1394746>] apic_timer_interrupt+0x2a/0x30 Code: 89 fe eb 08 31 c9 8b 45 0c ff 55 ec 83 c3 04 83 7d 10 00 74 0c 3b 5d 10 73 26 3b 5d e4 73 0c eb 1f 3b 5d f0 76 1a 3b 5d e8 73 15 <8b> 13 89 d0 89 55 e0 e8 ad 42 03 00 85 c0 8b 55 e0 75 a6 eb cc EIP: [<c1004237>] print_context_stack+0x6e/0x8d SS:ESP 0068:f5409ea0 CR2: 000000000000000a ---[ end trace 62afee3481b00012 ]--- Kernel panic - not syncing: Fatal exception in interrupt V2: * add comments to kernel_stack_pointer() * always return a valid stack address by falling back to the address of regs Reported-by: Yang Wei <wei.yang@windriver.com> Cc: <stable@vger.kernel.org> Signed-off-by: Robert Richter <robert.richter@amd.com> Link: http://lkml.kernel.org/r/20120912135059.GZ8285@erda.amd.com Signed-off-by: H. Peter Anvin <hpa@linux.intel.com> Cc: Jun Zhang <jun.zhang@intel.com>
2012-09-03 20:54:48 +02:00
/*
* X86_32 CPUs don't save ss and esp if the CPU is already in kernel mode
* when it traps. The previous stack will be directly underneath the saved
* registers, and 'sp/ss' won't even have been saved. Thus the '&regs->sp'.
*
* Now, if the stack is empty, '&regs->sp' is out of range. In this
* case we try to take the previous stack. To always return a non-null
* stack pointer we fall back to regs as stack if no previous stack
* exists.
*
* This is valid only for kernel mode traps.
*/
unsigned long kernel_stack_pointer(struct pt_regs *regs)
{
unsigned long context = (unsigned long)regs & ~(THREAD_SIZE - 1);
unsigned long sp = (unsigned long)&regs->sp;
struct thread_info *tinfo;
if (context == (sp & ~(THREAD_SIZE - 1)))
return sp;
tinfo = (struct thread_info *)context;
if (tinfo->previous_esp)
return tinfo->previous_esp;
return (unsigned long)regs;
}
EXPORT_SYMBOL_GPL(kernel_stack_pointer);
x86-32: Fix invalid stack address while in softirq In 32 bit the stack address provided by kernel_stack_pointer() may point to an invalid range causing NULL pointer access or page faults while in NMI (see trace below). This happens if called in softirq context and if the stack is empty. The address at &regs->sp is then out of range. Fixing this by checking if regs and &regs->sp are in the same stack context. Otherwise return the previous stack pointer stored in struct thread_info. If that address is invalid too, return address of regs. BUG: unable to handle kernel NULL pointer dereference at 0000000a IP: [<c1004237>] print_context_stack+0x6e/0x8d *pde = 00000000 Oops: 0000 [#1] SMP Modules linked in: Pid: 4434, comm: perl Not tainted 3.6.0-rc3-oprofile-i386-standard-g4411a05 #4 Hewlett-Packard HP xw9400 Workstation/0A1Ch EIP: 0060:[<c1004237>] EFLAGS: 00010093 CPU: 0 EIP is at print_context_stack+0x6e/0x8d EAX: ffffe000 EBX: 0000000a ECX: f4435f94 EDX: 0000000a ESI: f4435f94 EDI: f4435f94 EBP: f5409ec0 ESP: f5409ea0 DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068 CR0: 8005003b CR2: 0000000a CR3: 34ac9000 CR4: 000007d0 DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000 DR6: ffff0ff0 DR7: 00000400 Process perl (pid: 4434, ti=f5408000 task=f5637850 task.ti=f4434000) Stack: 000003e8 ffffe000 00001ffc f4e39b00 00000000 0000000a f4435f94 c155198c f5409ef0 c1003723 c155198c f5409f04 00000000 f5409edc 00000000 00000000 f5409ee8 f4435f94 f5409fc4 00000001 f5409f1c c12dce1c 00000000 c155198c Call Trace: [<c1003723>] dump_trace+0x7b/0xa1 [<c12dce1c>] x86_backtrace+0x40/0x88 [<c12db712>] ? oprofile_add_sample+0x56/0x84 [<c12db731>] oprofile_add_sample+0x75/0x84 [<c12ddb5b>] op_amd_check_ctrs+0x46/0x260 [<c12dd40d>] profile_exceptions_notify+0x23/0x4c [<c1395034>] nmi_handle+0x31/0x4a [<c1029dc5>] ? ftrace_define_fields_irq_handler_entry+0x45/0x45 [<c13950ed>] do_nmi+0xa0/0x2ff [<c1029dc5>] ? ftrace_define_fields_irq_handler_entry+0x45/0x45 [<c13949e5>] nmi_stack_correct+0x28/0x2d [<c1029dc5>] ? ftrace_define_fields_irq_handler_entry+0x45/0x45 [<c1003603>] ? do_softirq+0x4b/0x7f <IRQ> [<c102a06f>] irq_exit+0x35/0x5b [<c1018f56>] smp_apic_timer_interrupt+0x6c/0x7a [<c1394746>] apic_timer_interrupt+0x2a/0x30 Code: 89 fe eb 08 31 c9 8b 45 0c ff 55 ec 83 c3 04 83 7d 10 00 74 0c 3b 5d 10 73 26 3b 5d e4 73 0c eb 1f 3b 5d f0 76 1a 3b 5d e8 73 15 <8b> 13 89 d0 89 55 e0 e8 ad 42 03 00 85 c0 8b 55 e0 75 a6 eb cc EIP: [<c1004237>] print_context_stack+0x6e/0x8d SS:ESP 0068:f5409ea0 CR2: 000000000000000a ---[ end trace 62afee3481b00012 ]--- Kernel panic - not syncing: Fatal exception in interrupt V2: * add comments to kernel_stack_pointer() * always return a valid stack address by falling back to the address of regs Reported-by: Yang Wei <wei.yang@windriver.com> Cc: <stable@vger.kernel.org> Signed-off-by: Robert Richter <robert.richter@amd.com> Link: http://lkml.kernel.org/r/20120912135059.GZ8285@erda.amd.com Signed-off-by: H. Peter Anvin <hpa@linux.intel.com> Cc: Jun Zhang <jun.zhang@intel.com>
2012-09-03 20:54:48 +02:00
static unsigned long *pt_regs_access(struct pt_regs *regs, unsigned long regno)
{
BUILD_BUG_ON(offsetof(struct pt_regs, bx) != 0);
return &regs->bx + (regno >> 2);
}
static u16 get_segment_reg(struct task_struct *task, unsigned long offset)
{
/*
* Returning the value truncates it to 16 bits.
*/
unsigned int retval;
if (offset != offsetof(struct user_regs_struct, gs))
retval = *pt_regs_access(task_pt_regs(task), offset);
else {
if (task == current)
retval = get_user_gs(task_pt_regs(task));
else
retval = task_user_gs(task);
}
return retval;
}
static int set_segment_reg(struct task_struct *task,
unsigned long offset, u16 value)
{
/*
* The value argument was already truncated to 16 bits.
*/
if (invalid_selector(value))
return -EIO;
/*
* For %cs and %ss we cannot permit a null selector.
* We can permit a bogus selector as long as it has USER_RPL.
* Null selectors are fine for other segment registers, but
* we will never get back to user mode with invalid %cs or %ss
* and will take the trap in iret instead. Much code relies
* on user_mode() to distinguish a user trap frame (which can
* safely use invalid selectors) from a kernel trap frame.
*/
switch (offset) {
case offsetof(struct user_regs_struct, cs):
case offsetof(struct user_regs_struct, ss):
if (unlikely(value == 0))
return -EIO;
default:
*pt_regs_access(task_pt_regs(task), offset) = value;
break;
case offsetof(struct user_regs_struct, gs):
if (task == current)
set_user_gs(task_pt_regs(task), value);
else
task_user_gs(task) = value;
}
return 0;
}
#else /* CONFIG_X86_64 */
#define FLAG_MASK (FLAG_MASK_32 | X86_EFLAGS_NT)
static unsigned long *pt_regs_access(struct pt_regs *regs, unsigned long offset)
{
BUILD_BUG_ON(offsetof(struct pt_regs, r15) != 0);
return &regs->r15 + (offset / sizeof(regs->r15));
}
static u16 get_segment_reg(struct task_struct *task, unsigned long offset)
{
/*
* Returning the value truncates it to 16 bits.
*/
unsigned int seg;
switch (offset) {
case offsetof(struct user_regs_struct, fs):
if (task == current) {
/* Older gas can't assemble movq %?s,%r?? */
asm("movl %%fs,%0" : "=r" (seg));
return seg;
}
return task->thread.fsindex;
case offsetof(struct user_regs_struct, gs):
if (task == current) {
asm("movl %%gs,%0" : "=r" (seg));
return seg;
}
return task->thread.gsindex;
case offsetof(struct user_regs_struct, ds):
if (task == current) {
asm("movl %%ds,%0" : "=r" (seg));
return seg;
}
return task->thread.ds;
case offsetof(struct user_regs_struct, es):
if (task == current) {
asm("movl %%es,%0" : "=r" (seg));
return seg;
}
return task->thread.es;
case offsetof(struct user_regs_struct, cs):
case offsetof(struct user_regs_struct, ss):
break;
}
return *pt_regs_access(task_pt_regs(task), offset);
}
static int set_segment_reg(struct task_struct *task,
unsigned long offset, u16 value)
{
/*
* The value argument was already truncated to 16 bits.
*/
if (invalid_selector(value))
return -EIO;
switch (offset) {
case offsetof(struct user_regs_struct,fs):
/*
* If this is setting fs as for normal 64-bit use but
* setting fs_base has implicitly changed it, leave it.
*/
if ((value == FS_TLS_SEL && task->thread.fsindex == 0 &&
task->thread.fs != 0) ||
(value == 0 && task->thread.fsindex == FS_TLS_SEL &&
task->thread.fs == 0))
break;
task->thread.fsindex = value;
if (task == current)
loadsegment(fs, task->thread.fsindex);
break;
case offsetof(struct user_regs_struct,gs):
/*
* If this is setting gs as for normal 64-bit use but
* setting gs_base has implicitly changed it, leave it.
*/
if ((value == GS_TLS_SEL && task->thread.gsindex == 0 &&
task->thread.gs != 0) ||
(value == 0 && task->thread.gsindex == GS_TLS_SEL &&
task->thread.gs == 0))
break;
task->thread.gsindex = value;
if (task == current)
load_gs_index(task->thread.gsindex);
break;
case offsetof(struct user_regs_struct,ds):
task->thread.ds = value;
if (task == current)
loadsegment(ds, task->thread.ds);
break;
case offsetof(struct user_regs_struct,es):
task->thread.es = value;
if (task == current)
loadsegment(es, task->thread.es);
break;
/*
* Can't actually change these in 64-bit mode.
*/
case offsetof(struct user_regs_struct,cs):
if (unlikely(value == 0))
return -EIO;
#ifdef CONFIG_IA32_EMULATION
if (test_tsk_thread_flag(task, TIF_IA32))
task_pt_regs(task)->cs = value;
#endif
break;
case offsetof(struct user_regs_struct,ss):
if (unlikely(value == 0))
return -EIO;
#ifdef CONFIG_IA32_EMULATION
if (test_tsk_thread_flag(task, TIF_IA32))
task_pt_regs(task)->ss = value;
#endif
break;
}
return 0;
}
#endif /* CONFIG_X86_32 */
static unsigned long get_flags(struct task_struct *task)
{
unsigned long retval = task_pt_regs(task)->flags;
/*
* If the debugger set TF, hide it from the readout.
*/
if (test_tsk_thread_flag(task, TIF_FORCED_TF))
retval &= ~X86_EFLAGS_TF;
return retval;
}
static int set_flags(struct task_struct *task, unsigned long value)
{
struct pt_regs *regs = task_pt_regs(task);
/*
* If the user value contains TF, mark that
* it was not "us" (the debugger) that set it.
* If not, make sure it stays set if we had.
*/
if (value & X86_EFLAGS_TF)
clear_tsk_thread_flag(task, TIF_FORCED_TF);
else if (test_tsk_thread_flag(task, TIF_FORCED_TF))
value |= X86_EFLAGS_TF;
regs->flags = (regs->flags & ~FLAG_MASK) | (value & FLAG_MASK);
return 0;
}
static int putreg(struct task_struct *child,
unsigned long offset, unsigned long value)
{
switch (offset) {
case offsetof(struct user_regs_struct, cs):
case offsetof(struct user_regs_struct, ds):
case offsetof(struct user_regs_struct, es):
case offsetof(struct user_regs_struct, fs):
case offsetof(struct user_regs_struct, gs):
case offsetof(struct user_regs_struct, ss):
return set_segment_reg(child, offset, value);
case offsetof(struct user_regs_struct, flags):
return set_flags(child, value);
#ifdef CONFIG_X86_64
case offsetof(struct user_regs_struct,fs_base):
if (value >= TASK_SIZE_OF(child))
return -EIO;
/*
* When changing the segment base, use do_arch_prctl
* to set either thread.fs or thread.fsindex and the
* corresponding GDT slot.
*/
if (child->thread.fs != value)
return do_arch_prctl(child, ARCH_SET_FS, value);
return 0;
case offsetof(struct user_regs_struct,gs_base):
/*
* Exactly the same here as the %fs handling above.
*/
if (value >= TASK_SIZE_OF(child))
return -EIO;
if (child->thread.gs != value)
return do_arch_prctl(child, ARCH_SET_GS, value);
return 0;
#endif
}
*pt_regs_access(task_pt_regs(child), offset) = value;
return 0;
}
static unsigned long getreg(struct task_struct *task, unsigned long offset)
{
switch (offset) {
case offsetof(struct user_regs_struct, cs):
case offsetof(struct user_regs_struct, ds):
case offsetof(struct user_regs_struct, es):
case offsetof(struct user_regs_struct, fs):
case offsetof(struct user_regs_struct, gs):
case offsetof(struct user_regs_struct, ss):
return get_segment_reg(task, offset);
case offsetof(struct user_regs_struct, flags):
return get_flags(task);
#ifdef CONFIG_X86_64
case offsetof(struct user_regs_struct, fs_base): {
/*
* do_arch_prctl may have used a GDT slot instead of
* the MSR. To userland, it appears the same either
* way, except the %fs segment selector might not be 0.
*/
unsigned int seg = task->thread.fsindex;
if (task->thread.fs != 0)
return task->thread.fs;
if (task == current)
asm("movl %%fs,%0" : "=r" (seg));
if (seg != FS_TLS_SEL)
return 0;
return get_desc_base(&task->thread.tls_array[FS_TLS]);
}
case offsetof(struct user_regs_struct, gs_base): {
/*
* Exactly the same here as the %fs handling above.
*/
unsigned int seg = task->thread.gsindex;
if (task->thread.gs != 0)
return task->thread.gs;
if (task == current)
asm("movl %%gs,%0" : "=r" (seg));
if (seg != GS_TLS_SEL)
return 0;
return get_desc_base(&task->thread.tls_array[GS_TLS]);
}
#endif
}
return *pt_regs_access(task_pt_regs(task), offset);
}
static int genregs_get(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
void *kbuf, void __user *ubuf)
{
if (kbuf) {
unsigned long *k = kbuf;
while (count >= sizeof(*k)) {
*k++ = getreg(target, pos);
count -= sizeof(*k);
pos += sizeof(*k);
}
} else {
unsigned long __user *u = ubuf;
while (count >= sizeof(*u)) {
if (__put_user(getreg(target, pos), u++))
return -EFAULT;
count -= sizeof(*u);
pos += sizeof(*u);
}
}
return 0;
}
static int genregs_set(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
const void *kbuf, const void __user *ubuf)
{
int ret = 0;
if (kbuf) {
const unsigned long *k = kbuf;
while (count >= sizeof(*k) && !ret) {
ret = putreg(target, pos, *k++);
count -= sizeof(*k);
pos += sizeof(*k);
}
} else {
const unsigned long __user *u = ubuf;
while (count >= sizeof(*u) && !ret) {
unsigned long word;
ret = __get_user(word, u++);
if (ret)
break;
ret = putreg(target, pos, word);
count -= sizeof(*u);
pos += sizeof(*u);
}
}
return ret;
}
static void ptrace_triggered(struct perf_event *bp,
struct perf_sample_data *data,
struct pt_regs *regs)
{
int i;
hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events This patch rebase the implementation of the breakpoints API on top of perf events instances. Each breakpoints are now perf events that handle the register scheduling, thread/cpu attachment, etc.. The new layering is now made as follows: ptrace kgdb ftrace perf syscall \ | / / \ | / / / Core breakpoint API / / | / | / Breakpoints perf events | | Breakpoints PMU ---- Debug Register constraints handling (Part of core breakpoint API) | | Hardware debug registers Reasons of this rewrite: - Use the centralized/optimized pmu registers scheduling, implying an easier arch integration - More powerful register handling: perf attributes (pinned/flexible events, exclusive/non-exclusive, tunable period, etc...) Impact: - New perf ABI: the hardware breakpoints counters - Ptrace breakpoints setting remains tricky and still needs some per thread breakpoints references. Todo (in the order): - Support breakpoints perf counter events for perf tools (ie: implement perf_bpcounter_event()) - Support from perf tools Changes in v2: - Follow the perf "event " rename - The ptrace regression have been fixed (ptrace breakpoint perf events weren't released when a task ended) - Drop the struct hw_breakpoint and store generic fields in perf_event_attr. - Separate core and arch specific headers, drop asm-generic/hw_breakpoint.h and create linux/hw_breakpoint.h - Use new generic len/type for breakpoint - Handle off case: when breakpoints api is not supported by an arch Changes in v3: - Fix broken CONFIG_KVM, we need to propagate the breakpoint api changes to kvm when we exit the guest and restore the bp registers to the host. Changes in v4: - Drop the hw_breakpoint_restore() stub as it is only used by KVM - EXPORT_SYMBOL_GPL hw_breakpoint_restore() as KVM can be built as a module - Restore the breakpoints unconditionally on kvm guest exit: TIF_DEBUG_THREAD doesn't anymore cover every cases of running breakpoints and vcpu->arch.switch_db_regs might not always be set when the guest used debug registers. (Waiting for a reliable optimization) Changes in v5: - Split-up the asm-generic/hw-breakpoint.h moving to linux/hw_breakpoint.h into a separate patch - Optimize the breakpoints restoring while switching from kvm guest to host. We only want to restore the state if we have active breakpoints to the host, otherwise we don't care about messed-up address registers. - Add asm/hw_breakpoint.h to Kbuild - Fix bad breakpoint type in trace_selftest.c Changes in v6: - Fix wrong header inclusion in trace.h (triggered a build error with CONFIG_FTRACE_SELFTEST Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jan Kiszka <jan.kiszka@web.de> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Avi Kivity <avi@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org>
2009-09-09 19:22:48 +02:00
struct thread_struct *thread = &(current->thread);
/*
* Store in the virtual DR6 register the fact that the breakpoint
* was hit so the thread's debugger will see it.
*/
hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events This patch rebase the implementation of the breakpoints API on top of perf events instances. Each breakpoints are now perf events that handle the register scheduling, thread/cpu attachment, etc.. The new layering is now made as follows: ptrace kgdb ftrace perf syscall \ | / / \ | / / / Core breakpoint API / / | / | / Breakpoints perf events | | Breakpoints PMU ---- Debug Register constraints handling (Part of core breakpoint API) | | Hardware debug registers Reasons of this rewrite: - Use the centralized/optimized pmu registers scheduling, implying an easier arch integration - More powerful register handling: perf attributes (pinned/flexible events, exclusive/non-exclusive, tunable period, etc...) Impact: - New perf ABI: the hardware breakpoints counters - Ptrace breakpoints setting remains tricky and still needs some per thread breakpoints references. Todo (in the order): - Support breakpoints perf counter events for perf tools (ie: implement perf_bpcounter_event()) - Support from perf tools Changes in v2: - Follow the perf "event " rename - The ptrace regression have been fixed (ptrace breakpoint perf events weren't released when a task ended) - Drop the struct hw_breakpoint and store generic fields in perf_event_attr. - Separate core and arch specific headers, drop asm-generic/hw_breakpoint.h and create linux/hw_breakpoint.h - Use new generic len/type for breakpoint - Handle off case: when breakpoints api is not supported by an arch Changes in v3: - Fix broken CONFIG_KVM, we need to propagate the breakpoint api changes to kvm when we exit the guest and restore the bp registers to the host. Changes in v4: - Drop the hw_breakpoint_restore() stub as it is only used by KVM - EXPORT_SYMBOL_GPL hw_breakpoint_restore() as KVM can be built as a module - Restore the breakpoints unconditionally on kvm guest exit: TIF_DEBUG_THREAD doesn't anymore cover every cases of running breakpoints and vcpu->arch.switch_db_regs might not always be set when the guest used debug registers. (Waiting for a reliable optimization) Changes in v5: - Split-up the asm-generic/hw-breakpoint.h moving to linux/hw_breakpoint.h into a separate patch - Optimize the breakpoints restoring while switching from kvm guest to host. We only want to restore the state if we have active breakpoints to the host, otherwise we don't care about messed-up address registers. - Add asm/hw_breakpoint.h to Kbuild - Fix bad breakpoint type in trace_selftest.c Changes in v6: - Fix wrong header inclusion in trace.h (triggered a build error with CONFIG_FTRACE_SELFTEST Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jan Kiszka <jan.kiszka@web.de> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Avi Kivity <avi@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org>
2009-09-09 19:22:48 +02:00
for (i = 0; i < HBP_NUM; i++) {
if (thread->ptrace_bps[i] == bp)
break;
hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events This patch rebase the implementation of the breakpoints API on top of perf events instances. Each breakpoints are now perf events that handle the register scheduling, thread/cpu attachment, etc.. The new layering is now made as follows: ptrace kgdb ftrace perf syscall \ | / / \ | / / / Core breakpoint API / / | / | / Breakpoints perf events | | Breakpoints PMU ---- Debug Register constraints handling (Part of core breakpoint API) | | Hardware debug registers Reasons of this rewrite: - Use the centralized/optimized pmu registers scheduling, implying an easier arch integration - More powerful register handling: perf attributes (pinned/flexible events, exclusive/non-exclusive, tunable period, etc...) Impact: - New perf ABI: the hardware breakpoints counters - Ptrace breakpoints setting remains tricky and still needs some per thread breakpoints references. Todo (in the order): - Support breakpoints perf counter events for perf tools (ie: implement perf_bpcounter_event()) - Support from perf tools Changes in v2: - Follow the perf "event " rename - The ptrace regression have been fixed (ptrace breakpoint perf events weren't released when a task ended) - Drop the struct hw_breakpoint and store generic fields in perf_event_attr. - Separate core and arch specific headers, drop asm-generic/hw_breakpoint.h and create linux/hw_breakpoint.h - Use new generic len/type for breakpoint - Handle off case: when breakpoints api is not supported by an arch Changes in v3: - Fix broken CONFIG_KVM, we need to propagate the breakpoint api changes to kvm when we exit the guest and restore the bp registers to the host. Changes in v4: - Drop the hw_breakpoint_restore() stub as it is only used by KVM - EXPORT_SYMBOL_GPL hw_breakpoint_restore() as KVM can be built as a module - Restore the breakpoints unconditionally on kvm guest exit: TIF_DEBUG_THREAD doesn't anymore cover every cases of running breakpoints and vcpu->arch.switch_db_regs might not always be set when the guest used debug registers. (Waiting for a reliable optimization) Changes in v5: - Split-up the asm-generic/hw-breakpoint.h moving to linux/hw_breakpoint.h into a separate patch - Optimize the breakpoints restoring while switching from kvm guest to host. We only want to restore the state if we have active breakpoints to the host, otherwise we don't care about messed-up address registers. - Add asm/hw_breakpoint.h to Kbuild - Fix bad breakpoint type in trace_selftest.c Changes in v6: - Fix wrong header inclusion in trace.h (triggered a build error with CONFIG_FTRACE_SELFTEST Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jan Kiszka <jan.kiszka@web.de> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Avi Kivity <avi@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org>
2009-09-09 19:22:48 +02:00
}
thread->debugreg6 |= (DR_TRAP0 << i);
}
/*
hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events This patch rebase the implementation of the breakpoints API on top of perf events instances. Each breakpoints are now perf events that handle the register scheduling, thread/cpu attachment, etc.. The new layering is now made as follows: ptrace kgdb ftrace perf syscall \ | / / \ | / / / Core breakpoint API / / | / | / Breakpoints perf events | | Breakpoints PMU ---- Debug Register constraints handling (Part of core breakpoint API) | | Hardware debug registers Reasons of this rewrite: - Use the centralized/optimized pmu registers scheduling, implying an easier arch integration - More powerful register handling: perf attributes (pinned/flexible events, exclusive/non-exclusive, tunable period, etc...) Impact: - New perf ABI: the hardware breakpoints counters - Ptrace breakpoints setting remains tricky and still needs some per thread breakpoints references. Todo (in the order): - Support breakpoints perf counter events for perf tools (ie: implement perf_bpcounter_event()) - Support from perf tools Changes in v2: - Follow the perf "event " rename - The ptrace regression have been fixed (ptrace breakpoint perf events weren't released when a task ended) - Drop the struct hw_breakpoint and store generic fields in perf_event_attr. - Separate core and arch specific headers, drop asm-generic/hw_breakpoint.h and create linux/hw_breakpoint.h - Use new generic len/type for breakpoint - Handle off case: when breakpoints api is not supported by an arch Changes in v3: - Fix broken CONFIG_KVM, we need to propagate the breakpoint api changes to kvm when we exit the guest and restore the bp registers to the host. Changes in v4: - Drop the hw_breakpoint_restore() stub as it is only used by KVM - EXPORT_SYMBOL_GPL hw_breakpoint_restore() as KVM can be built as a module - Restore the breakpoints unconditionally on kvm guest exit: TIF_DEBUG_THREAD doesn't anymore cover every cases of running breakpoints and vcpu->arch.switch_db_regs might not always be set when the guest used debug registers. (Waiting for a reliable optimization) Changes in v5: - Split-up the asm-generic/hw-breakpoint.h moving to linux/hw_breakpoint.h into a separate patch - Optimize the breakpoints restoring while switching from kvm guest to host. We only want to restore the state if we have active breakpoints to the host, otherwise we don't care about messed-up address registers. - Add asm/hw_breakpoint.h to Kbuild - Fix bad breakpoint type in trace_selftest.c Changes in v6: - Fix wrong header inclusion in trace.h (triggered a build error with CONFIG_FTRACE_SELFTEST Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jan Kiszka <jan.kiszka@web.de> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Avi Kivity <avi@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org>
2009-09-09 19:22:48 +02:00
* Walk through every ptrace breakpoints for this thread and
* build the dr7 value on top of their attributes.
*
*/
hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events This patch rebase the implementation of the breakpoints API on top of perf events instances. Each breakpoints are now perf events that handle the register scheduling, thread/cpu attachment, etc.. The new layering is now made as follows: ptrace kgdb ftrace perf syscall \ | / / \ | / / / Core breakpoint API / / | / | / Breakpoints perf events | | Breakpoints PMU ---- Debug Register constraints handling (Part of core breakpoint API) | | Hardware debug registers Reasons of this rewrite: - Use the centralized/optimized pmu registers scheduling, implying an easier arch integration - More powerful register handling: perf attributes (pinned/flexible events, exclusive/non-exclusive, tunable period, etc...) Impact: - New perf ABI: the hardware breakpoints counters - Ptrace breakpoints setting remains tricky and still needs some per thread breakpoints references. Todo (in the order): - Support breakpoints perf counter events for perf tools (ie: implement perf_bpcounter_event()) - Support from perf tools Changes in v2: - Follow the perf "event " rename - The ptrace regression have been fixed (ptrace breakpoint perf events weren't released when a task ended) - Drop the struct hw_breakpoint and store generic fields in perf_event_attr. - Separate core and arch specific headers, drop asm-generic/hw_breakpoint.h and create linux/hw_breakpoint.h - Use new generic len/type for breakpoint - Handle off case: when breakpoints api is not supported by an arch Changes in v3: - Fix broken CONFIG_KVM, we need to propagate the breakpoint api changes to kvm when we exit the guest and restore the bp registers to the host. Changes in v4: - Drop the hw_breakpoint_restore() stub as it is only used by KVM - EXPORT_SYMBOL_GPL hw_breakpoint_restore() as KVM can be built as a module - Restore the breakpoints unconditionally on kvm guest exit: TIF_DEBUG_THREAD doesn't anymore cover every cases of running breakpoints and vcpu->arch.switch_db_regs might not always be set when the guest used debug registers. (Waiting for a reliable optimization) Changes in v5: - Split-up the asm-generic/hw-breakpoint.h moving to linux/hw_breakpoint.h into a separate patch - Optimize the breakpoints restoring while switching from kvm guest to host. We only want to restore the state if we have active breakpoints to the host, otherwise we don't care about messed-up address registers. - Add asm/hw_breakpoint.h to Kbuild - Fix bad breakpoint type in trace_selftest.c Changes in v6: - Fix wrong header inclusion in trace.h (triggered a build error with CONFIG_FTRACE_SELFTEST Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jan Kiszka <jan.kiszka@web.de> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Avi Kivity <avi@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org>
2009-09-09 19:22:48 +02:00
static unsigned long ptrace_get_dr7(struct perf_event *bp[])
{
hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events This patch rebase the implementation of the breakpoints API on top of perf events instances. Each breakpoints are now perf events that handle the register scheduling, thread/cpu attachment, etc.. The new layering is now made as follows: ptrace kgdb ftrace perf syscall \ | / / \ | / / / Core breakpoint API / / | / | / Breakpoints perf events | | Breakpoints PMU ---- Debug Register constraints handling (Part of core breakpoint API) | | Hardware debug registers Reasons of this rewrite: - Use the centralized/optimized pmu registers scheduling, implying an easier arch integration - More powerful register handling: perf attributes (pinned/flexible events, exclusive/non-exclusive, tunable period, etc...) Impact: - New perf ABI: the hardware breakpoints counters - Ptrace breakpoints setting remains tricky and still needs some per thread breakpoints references. Todo (in the order): - Support breakpoints perf counter events for perf tools (ie: implement perf_bpcounter_event()) - Support from perf tools Changes in v2: - Follow the perf "event " rename - The ptrace regression have been fixed (ptrace breakpoint perf events weren't released when a task ended) - Drop the struct hw_breakpoint and store generic fields in perf_event_attr. - Separate core and arch specific headers, drop asm-generic/hw_breakpoint.h and create linux/hw_breakpoint.h - Use new generic len/type for breakpoint - Handle off case: when breakpoints api is not supported by an arch Changes in v3: - Fix broken CONFIG_KVM, we need to propagate the breakpoint api changes to kvm when we exit the guest and restore the bp registers to the host. Changes in v4: - Drop the hw_breakpoint_restore() stub as it is only used by KVM - EXPORT_SYMBOL_GPL hw_breakpoint_restore() as KVM can be built as a module - Restore the breakpoints unconditionally on kvm guest exit: TIF_DEBUG_THREAD doesn't anymore cover every cases of running breakpoints and vcpu->arch.switch_db_regs might not always be set when the guest used debug registers. (Waiting for a reliable optimization) Changes in v5: - Split-up the asm-generic/hw-breakpoint.h moving to linux/hw_breakpoint.h into a separate patch - Optimize the breakpoints restoring while switching from kvm guest to host. We only want to restore the state if we have active breakpoints to the host, otherwise we don't care about messed-up address registers. - Add asm/hw_breakpoint.h to Kbuild - Fix bad breakpoint type in trace_selftest.c Changes in v6: - Fix wrong header inclusion in trace.h (triggered a build error with CONFIG_FTRACE_SELFTEST Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jan Kiszka <jan.kiszka@web.de> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Avi Kivity <avi@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org>
2009-09-09 19:22:48 +02:00
int i;
int dr7 = 0;
struct arch_hw_breakpoint *info;
for (i = 0; i < HBP_NUM; i++) {
if (bp[i] && !bp[i]->attr.disabled) {
info = counter_arch_bp(bp[i]);
dr7 |= encode_dr7(i, info->len, info->type);
}
}
hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events This patch rebase the implementation of the breakpoints API on top of perf events instances. Each breakpoints are now perf events that handle the register scheduling, thread/cpu attachment, etc.. The new layering is now made as follows: ptrace kgdb ftrace perf syscall \ | / / \ | / / / Core breakpoint API / / | / | / Breakpoints perf events | | Breakpoints PMU ---- Debug Register constraints handling (Part of core breakpoint API) | | Hardware debug registers Reasons of this rewrite: - Use the centralized/optimized pmu registers scheduling, implying an easier arch integration - More powerful register handling: perf attributes (pinned/flexible events, exclusive/non-exclusive, tunable period, etc...) Impact: - New perf ABI: the hardware breakpoints counters - Ptrace breakpoints setting remains tricky and still needs some per thread breakpoints references. Todo (in the order): - Support breakpoints perf counter events for perf tools (ie: implement perf_bpcounter_event()) - Support from perf tools Changes in v2: - Follow the perf "event " rename - The ptrace regression have been fixed (ptrace breakpoint perf events weren't released when a task ended) - Drop the struct hw_breakpoint and store generic fields in perf_event_attr. - Separate core and arch specific headers, drop asm-generic/hw_breakpoint.h and create linux/hw_breakpoint.h - Use new generic len/type for breakpoint - Handle off case: when breakpoints api is not supported by an arch Changes in v3: - Fix broken CONFIG_KVM, we need to propagate the breakpoint api changes to kvm when we exit the guest and restore the bp registers to the host. Changes in v4: - Drop the hw_breakpoint_restore() stub as it is only used by KVM - EXPORT_SYMBOL_GPL hw_breakpoint_restore() as KVM can be built as a module - Restore the breakpoints unconditionally on kvm guest exit: TIF_DEBUG_THREAD doesn't anymore cover every cases of running breakpoints and vcpu->arch.switch_db_regs might not always be set when the guest used debug registers. (Waiting for a reliable optimization) Changes in v5: - Split-up the asm-generic/hw-breakpoint.h moving to linux/hw_breakpoint.h into a separate patch - Optimize the breakpoints restoring while switching from kvm guest to host. We only want to restore the state if we have active breakpoints to the host, otherwise we don't care about messed-up address registers. - Add asm/hw_breakpoint.h to Kbuild - Fix bad breakpoint type in trace_selftest.c Changes in v6: - Fix wrong header inclusion in trace.h (triggered a build error with CONFIG_FTRACE_SELFTEST Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jan Kiszka <jan.kiszka@web.de> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Avi Kivity <avi@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org>
2009-09-09 19:22:48 +02:00
return dr7;
}
static int ptrace_fill_bp_fields(struct perf_event_attr *attr,
int len, int type, bool disabled)
{
int err, bp_len, bp_type;
err = arch_bp_generic_fields(len, type, &bp_len, &bp_type);
if (!err) {
attr->bp_len = bp_len;
attr->bp_type = bp_type;
attr->disabled = disabled;
}
return err;
}
static struct perf_event *
ptrace_register_breakpoint(struct task_struct *tsk, int len, int type,
unsigned long addr, bool disabled)
{
struct perf_event_attr attr;
int err;
ptrace_breakpoint_init(&attr);
attr.bp_addr = addr;
err = ptrace_fill_bp_fields(&attr, len, type, disabled);
if (err)
return ERR_PTR(err);
return register_user_hw_breakpoint(&attr, ptrace_triggered,
NULL, tsk);
}
static int ptrace_modify_breakpoint(struct perf_event *bp, int len, int type,
int disabled)
{
struct perf_event_attr attr = bp->attr;
int err;
err = ptrace_fill_bp_fields(&attr, len, type, disabled);
if (err)
return err;
return modify_user_hw_breakpoint(bp, &attr);
}
/*
* Handle ptrace writes to debug register 7.
*/
static int ptrace_write_dr7(struct task_struct *tsk, unsigned long data)
{
ptrace/x86: dont delay "disable" till second pass in ptrace_write_dr7() ptrace_write_dr7() skips ptrace_modify_breakpoint(disabled => true) unless second_pass, this buys nothing but complicates the code and means that we always do the main loop twice even if "disabled" was never true. The comment says: Don't unregister the breakpoints right-away, unless all register_user_hw_breakpoint() requests have succeeded. Firstly, we do not do register_user_hw_breakpoint(), it was removed by commit 24f1e32c60c4 ("hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events"). We are going to restore register_user_hw_breakpoint() (see the next patch) but this doesn't matter: after commit 44234adcdce3 ("hw-breakpoints: Modify breakpoints without unregistering them") perf_event_disable() can not hurt, hw_breakpoint_del() does not free the slot. Remove the "second_pass" check from the main loop and simplify the code. Since we have to check "bp != NULL" anyway, the patch also removes the same check in ptrace_modify_breakpoint() and moves the comment into ptrace_write_dr7(). With this patch the second pass is only needed to restore the saved old_dr7. This should never fail, so the patch adds WARN_ON() to catch the potential problems as Frederic suggested. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jan Kratochvil <jan.kratochvil@redhat.com> Cc: Michael Neuling <mikey@neuling.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Paul Mundt <lethal@linux-sh.org> Cc: Will Deacon <will.deacon@arm.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Russell King <linux@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-07-09 01:00:58 +02:00
struct thread_struct *thread = &tsk->thread;
hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events This patch rebase the implementation of the breakpoints API on top of perf events instances. Each breakpoints are now perf events that handle the register scheduling, thread/cpu attachment, etc.. The new layering is now made as follows: ptrace kgdb ftrace perf syscall \ | / / \ | / / / Core breakpoint API / / | / | / Breakpoints perf events | | Breakpoints PMU ---- Debug Register constraints handling (Part of core breakpoint API) | | Hardware debug registers Reasons of this rewrite: - Use the centralized/optimized pmu registers scheduling, implying an easier arch integration - More powerful register handling: perf attributes (pinned/flexible events, exclusive/non-exclusive, tunable period, etc...) Impact: - New perf ABI: the hardware breakpoints counters - Ptrace breakpoints setting remains tricky and still needs some per thread breakpoints references. Todo (in the order): - Support breakpoints perf counter events for perf tools (ie: implement perf_bpcounter_event()) - Support from perf tools Changes in v2: - Follow the perf "event " rename - The ptrace regression have been fixed (ptrace breakpoint perf events weren't released when a task ended) - Drop the struct hw_breakpoint and store generic fields in perf_event_attr. - Separate core and arch specific headers, drop asm-generic/hw_breakpoint.h and create linux/hw_breakpoint.h - Use new generic len/type for breakpoint - Handle off case: when breakpoints api is not supported by an arch Changes in v3: - Fix broken CONFIG_KVM, we need to propagate the breakpoint api changes to kvm when we exit the guest and restore the bp registers to the host. Changes in v4: - Drop the hw_breakpoint_restore() stub as it is only used by KVM - EXPORT_SYMBOL_GPL hw_breakpoint_restore() as KVM can be built as a module - Restore the breakpoints unconditionally on kvm guest exit: TIF_DEBUG_THREAD doesn't anymore cover every cases of running breakpoints and vcpu->arch.switch_db_regs might not always be set when the guest used debug registers. (Waiting for a reliable optimization) Changes in v5: - Split-up the asm-generic/hw-breakpoint.h moving to linux/hw_breakpoint.h into a separate patch - Optimize the breakpoints restoring while switching from kvm guest to host. We only want to restore the state if we have active breakpoints to the host, otherwise we don't care about messed-up address registers. - Add asm/hw_breakpoint.h to Kbuild - Fix bad breakpoint type in trace_selftest.c Changes in v6: - Fix wrong header inclusion in trace.h (triggered a build error with CONFIG_FTRACE_SELFTEST Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jan Kiszka <jan.kiszka@web.de> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Avi Kivity <avi@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org>
2009-09-09 19:22:48 +02:00
unsigned long old_dr7;
ptrace/x86: dont delay "disable" till second pass in ptrace_write_dr7() ptrace_write_dr7() skips ptrace_modify_breakpoint(disabled => true) unless second_pass, this buys nothing but complicates the code and means that we always do the main loop twice even if "disabled" was never true. The comment says: Don't unregister the breakpoints right-away, unless all register_user_hw_breakpoint() requests have succeeded. Firstly, we do not do register_user_hw_breakpoint(), it was removed by commit 24f1e32c60c4 ("hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events"). We are going to restore register_user_hw_breakpoint() (see the next patch) but this doesn't matter: after commit 44234adcdce3 ("hw-breakpoints: Modify breakpoints without unregistering them") perf_event_disable() can not hurt, hw_breakpoint_del() does not free the slot. Remove the "second_pass" check from the main loop and simplify the code. Since we have to check "bp != NULL" anyway, the patch also removes the same check in ptrace_modify_breakpoint() and moves the comment into ptrace_write_dr7(). With this patch the second pass is only needed to restore the saved old_dr7. This should never fail, so the patch adds WARN_ON() to catch the potential problems as Frederic suggested. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jan Kratochvil <jan.kratochvil@redhat.com> Cc: Michael Neuling <mikey@neuling.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Paul Mundt <lethal@linux-sh.org> Cc: Will Deacon <will.deacon@arm.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Russell King <linux@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-07-09 01:00:58 +02:00
bool second_pass = false;
int i, rc, ret = 0;
data &= ~DR_CONTROL_RESERVED;
hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events This patch rebase the implementation of the breakpoints API on top of perf events instances. Each breakpoints are now perf events that handle the register scheduling, thread/cpu attachment, etc.. The new layering is now made as follows: ptrace kgdb ftrace perf syscall \ | / / \ | / / / Core breakpoint API / / | / | / Breakpoints perf events | | Breakpoints PMU ---- Debug Register constraints handling (Part of core breakpoint API) | | Hardware debug registers Reasons of this rewrite: - Use the centralized/optimized pmu registers scheduling, implying an easier arch integration - More powerful register handling: perf attributes (pinned/flexible events, exclusive/non-exclusive, tunable period, etc...) Impact: - New perf ABI: the hardware breakpoints counters - Ptrace breakpoints setting remains tricky and still needs some per thread breakpoints references. Todo (in the order): - Support breakpoints perf counter events for perf tools (ie: implement perf_bpcounter_event()) - Support from perf tools Changes in v2: - Follow the perf "event " rename - The ptrace regression have been fixed (ptrace breakpoint perf events weren't released when a task ended) - Drop the struct hw_breakpoint and store generic fields in perf_event_attr. - Separate core and arch specific headers, drop asm-generic/hw_breakpoint.h and create linux/hw_breakpoint.h - Use new generic len/type for breakpoint - Handle off case: when breakpoints api is not supported by an arch Changes in v3: - Fix broken CONFIG_KVM, we need to propagate the breakpoint api changes to kvm when we exit the guest and restore the bp registers to the host. Changes in v4: - Drop the hw_breakpoint_restore() stub as it is only used by KVM - EXPORT_SYMBOL_GPL hw_breakpoint_restore() as KVM can be built as a module - Restore the breakpoints unconditionally on kvm guest exit: TIF_DEBUG_THREAD doesn't anymore cover every cases of running breakpoints and vcpu->arch.switch_db_regs might not always be set when the guest used debug registers. (Waiting for a reliable optimization) Changes in v5: - Split-up the asm-generic/hw-breakpoint.h moving to linux/hw_breakpoint.h into a separate patch - Optimize the breakpoints restoring while switching from kvm guest to host. We only want to restore the state if we have active breakpoints to the host, otherwise we don't care about messed-up address registers. - Add asm/hw_breakpoint.h to Kbuild - Fix bad breakpoint type in trace_selftest.c Changes in v6: - Fix wrong header inclusion in trace.h (triggered a build error with CONFIG_FTRACE_SELFTEST Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jan Kiszka <jan.kiszka@web.de> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Avi Kivity <avi@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org>
2009-09-09 19:22:48 +02:00
old_dr7 = ptrace_get_dr7(thread->ptrace_bps);
ptrace/x86: dont delay "disable" till second pass in ptrace_write_dr7() ptrace_write_dr7() skips ptrace_modify_breakpoint(disabled => true) unless second_pass, this buys nothing but complicates the code and means that we always do the main loop twice even if "disabled" was never true. The comment says: Don't unregister the breakpoints right-away, unless all register_user_hw_breakpoint() requests have succeeded. Firstly, we do not do register_user_hw_breakpoint(), it was removed by commit 24f1e32c60c4 ("hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events"). We are going to restore register_user_hw_breakpoint() (see the next patch) but this doesn't matter: after commit 44234adcdce3 ("hw-breakpoints: Modify breakpoints without unregistering them") perf_event_disable() can not hurt, hw_breakpoint_del() does not free the slot. Remove the "second_pass" check from the main loop and simplify the code. Since we have to check "bp != NULL" anyway, the patch also removes the same check in ptrace_modify_breakpoint() and moves the comment into ptrace_write_dr7(). With this patch the second pass is only needed to restore the saved old_dr7. This should never fail, so the patch adds WARN_ON() to catch the potential problems as Frederic suggested. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jan Kratochvil <jan.kratochvil@redhat.com> Cc: Michael Neuling <mikey@neuling.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Paul Mundt <lethal@linux-sh.org> Cc: Will Deacon <will.deacon@arm.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Russell King <linux@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-07-09 01:00:58 +02:00
restore:
ptrace/x86: dont delay "disable" till second pass in ptrace_write_dr7() ptrace_write_dr7() skips ptrace_modify_breakpoint(disabled => true) unless second_pass, this buys nothing but complicates the code and means that we always do the main loop twice even if "disabled" was never true. The comment says: Don't unregister the breakpoints right-away, unless all register_user_hw_breakpoint() requests have succeeded. Firstly, we do not do register_user_hw_breakpoint(), it was removed by commit 24f1e32c60c4 ("hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events"). We are going to restore register_user_hw_breakpoint() (see the next patch) but this doesn't matter: after commit 44234adcdce3 ("hw-breakpoints: Modify breakpoints without unregistering them") perf_event_disable() can not hurt, hw_breakpoint_del() does not free the slot. Remove the "second_pass" check from the main loop and simplify the code. Since we have to check "bp != NULL" anyway, the patch also removes the same check in ptrace_modify_breakpoint() and moves the comment into ptrace_write_dr7(). With this patch the second pass is only needed to restore the saved old_dr7. This should never fail, so the patch adds WARN_ON() to catch the potential problems as Frederic suggested. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jan Kratochvil <jan.kratochvil@redhat.com> Cc: Michael Neuling <mikey@neuling.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Paul Mundt <lethal@linux-sh.org> Cc: Will Deacon <will.deacon@arm.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Russell King <linux@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-07-09 01:00:58 +02:00
rc = 0;
for (i = 0; i < HBP_NUM; i++) {
unsigned len, type;
bool disabled = !decode_dr7(data, i, &len, &type);
struct perf_event *bp = thread->ptrace_bps[i];
ptrace/x86: dont delay "disable" till second pass in ptrace_write_dr7() ptrace_write_dr7() skips ptrace_modify_breakpoint(disabled => true) unless second_pass, this buys nothing but complicates the code and means that we always do the main loop twice even if "disabled" was never true. The comment says: Don't unregister the breakpoints right-away, unless all register_user_hw_breakpoint() requests have succeeded. Firstly, we do not do register_user_hw_breakpoint(), it was removed by commit 24f1e32c60c4 ("hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events"). We are going to restore register_user_hw_breakpoint() (see the next patch) but this doesn't matter: after commit 44234adcdce3 ("hw-breakpoints: Modify breakpoints without unregistering them") perf_event_disable() can not hurt, hw_breakpoint_del() does not free the slot. Remove the "second_pass" check from the main loop and simplify the code. Since we have to check "bp != NULL" anyway, the patch also removes the same check in ptrace_modify_breakpoint() and moves the comment into ptrace_write_dr7(). With this patch the second pass is only needed to restore the saved old_dr7. This should never fail, so the patch adds WARN_ON() to catch the potential problems as Frederic suggested. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jan Kratochvil <jan.kratochvil@redhat.com> Cc: Michael Neuling <mikey@neuling.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Paul Mundt <lethal@linux-sh.org> Cc: Will Deacon <will.deacon@arm.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Russell King <linux@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-07-09 01:00:58 +02:00
if (!bp) {
if (disabled)
continue;
bp = ptrace_register_breakpoint(tsk,
len, type, 0, disabled);
if (IS_ERR(bp)) {
rc = PTR_ERR(bp);
break;
}
thread->ptrace_bps[i] = bp;
continue;
}
rc = ptrace_modify_breakpoint(bp, len, type, disabled);
if (rc)
hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events This patch rebase the implementation of the breakpoints API on top of perf events instances. Each breakpoints are now perf events that handle the register scheduling, thread/cpu attachment, etc.. The new layering is now made as follows: ptrace kgdb ftrace perf syscall \ | / / \ | / / / Core breakpoint API / / | / | / Breakpoints perf events | | Breakpoints PMU ---- Debug Register constraints handling (Part of core breakpoint API) | | Hardware debug registers Reasons of this rewrite: - Use the centralized/optimized pmu registers scheduling, implying an easier arch integration - More powerful register handling: perf attributes (pinned/flexible events, exclusive/non-exclusive, tunable period, etc...) Impact: - New perf ABI: the hardware breakpoints counters - Ptrace breakpoints setting remains tricky and still needs some per thread breakpoints references. Todo (in the order): - Support breakpoints perf counter events for perf tools (ie: implement perf_bpcounter_event()) - Support from perf tools Changes in v2: - Follow the perf "event " rename - The ptrace regression have been fixed (ptrace breakpoint perf events weren't released when a task ended) - Drop the struct hw_breakpoint and store generic fields in perf_event_attr. - Separate core and arch specific headers, drop asm-generic/hw_breakpoint.h and create linux/hw_breakpoint.h - Use new generic len/type for breakpoint - Handle off case: when breakpoints api is not supported by an arch Changes in v3: - Fix broken CONFIG_KVM, we need to propagate the breakpoint api changes to kvm when we exit the guest and restore the bp registers to the host. Changes in v4: - Drop the hw_breakpoint_restore() stub as it is only used by KVM - EXPORT_SYMBOL_GPL hw_breakpoint_restore() as KVM can be built as a module - Restore the breakpoints unconditionally on kvm guest exit: TIF_DEBUG_THREAD doesn't anymore cover every cases of running breakpoints and vcpu->arch.switch_db_regs might not always be set when the guest used debug registers. (Waiting for a reliable optimization) Changes in v5: - Split-up the asm-generic/hw-breakpoint.h moving to linux/hw_breakpoint.h into a separate patch - Optimize the breakpoints restoring while switching from kvm guest to host. We only want to restore the state if we have active breakpoints to the host, otherwise we don't care about messed-up address registers. - Add asm/hw_breakpoint.h to Kbuild - Fix bad breakpoint type in trace_selftest.c Changes in v6: - Fix wrong header inclusion in trace.h (triggered a build error with CONFIG_FTRACE_SELFTEST Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jan Kiszka <jan.kiszka@web.de> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Avi Kivity <avi@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org>
2009-09-09 19:22:48 +02:00
break;
}
ptrace/x86: dont delay "disable" till second pass in ptrace_write_dr7() ptrace_write_dr7() skips ptrace_modify_breakpoint(disabled => true) unless second_pass, this buys nothing but complicates the code and means that we always do the main loop twice even if "disabled" was never true. The comment says: Don't unregister the breakpoints right-away, unless all register_user_hw_breakpoint() requests have succeeded. Firstly, we do not do register_user_hw_breakpoint(), it was removed by commit 24f1e32c60c4 ("hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events"). We are going to restore register_user_hw_breakpoint() (see the next patch) but this doesn't matter: after commit 44234adcdce3 ("hw-breakpoints: Modify breakpoints without unregistering them") perf_event_disable() can not hurt, hw_breakpoint_del() does not free the slot. Remove the "second_pass" check from the main loop and simplify the code. Since we have to check "bp != NULL" anyway, the patch also removes the same check in ptrace_modify_breakpoint() and moves the comment into ptrace_write_dr7(). With this patch the second pass is only needed to restore the saved old_dr7. This should never fail, so the patch adds WARN_ON() to catch the potential problems as Frederic suggested. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jan Kratochvil <jan.kratochvil@redhat.com> Cc: Michael Neuling <mikey@neuling.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Paul Mundt <lethal@linux-sh.org> Cc: Will Deacon <will.deacon@arm.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Russell King <linux@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-07-09 01:00:58 +02:00
/* Restore if the first pass failed, second_pass shouldn't fail. */
if (rc && !WARN_ON(second_pass)) {
ret = rc;
data = old_dr7;
second_pass = true;
goto restore;
}
ptrace/x86: dont delay "disable" till second pass in ptrace_write_dr7() ptrace_write_dr7() skips ptrace_modify_breakpoint(disabled => true) unless second_pass, this buys nothing but complicates the code and means that we always do the main loop twice even if "disabled" was never true. The comment says: Don't unregister the breakpoints right-away, unless all register_user_hw_breakpoint() requests have succeeded. Firstly, we do not do register_user_hw_breakpoint(), it was removed by commit 24f1e32c60c4 ("hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events"). We are going to restore register_user_hw_breakpoint() (see the next patch) but this doesn't matter: after commit 44234adcdce3 ("hw-breakpoints: Modify breakpoints without unregistering them") perf_event_disable() can not hurt, hw_breakpoint_del() does not free the slot. Remove the "second_pass" check from the main loop and simplify the code. Since we have to check "bp != NULL" anyway, the patch also removes the same check in ptrace_modify_breakpoint() and moves the comment into ptrace_write_dr7(). With this patch the second pass is only needed to restore the saved old_dr7. This should never fail, so the patch adds WARN_ON() to catch the potential problems as Frederic suggested. Signed-off-by: Oleg Nesterov <oleg@redhat.com> Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jan Kratochvil <jan.kratochvil@redhat.com> Cc: Michael Neuling <mikey@neuling.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Paul Mundt <lethal@linux-sh.org> Cc: Will Deacon <will.deacon@arm.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Russell King <linux@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-07-09 01:00:58 +02:00
return ret;
}
/*
* Handle PTRACE_PEEKUSR calls for the debug register area.
*/
static unsigned long ptrace_get_debugreg(struct task_struct *tsk, int n)
{
struct thread_struct *thread = &tsk->thread;
unsigned long val = 0;
hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events This patch rebase the implementation of the breakpoints API on top of perf events instances. Each breakpoints are now perf events that handle the register scheduling, thread/cpu attachment, etc.. The new layering is now made as follows: ptrace kgdb ftrace perf syscall \ | / / \ | / / / Core breakpoint API / / | / | / Breakpoints perf events | | Breakpoints PMU ---- Debug Register constraints handling (Part of core breakpoint API) | | Hardware debug registers Reasons of this rewrite: - Use the centralized/optimized pmu registers scheduling, implying an easier arch integration - More powerful register handling: perf attributes (pinned/flexible events, exclusive/non-exclusive, tunable period, etc...) Impact: - New perf ABI: the hardware breakpoints counters - Ptrace breakpoints setting remains tricky and still needs some per thread breakpoints references. Todo (in the order): - Support breakpoints perf counter events for perf tools (ie: implement perf_bpcounter_event()) - Support from perf tools Changes in v2: - Follow the perf "event " rename - The ptrace regression have been fixed (ptrace breakpoint perf events weren't released when a task ended) - Drop the struct hw_breakpoint and store generic fields in perf_event_attr. - Separate core and arch specific headers, drop asm-generic/hw_breakpoint.h and create linux/hw_breakpoint.h - Use new generic len/type for breakpoint - Handle off case: when breakpoints api is not supported by an arch Changes in v3: - Fix broken CONFIG_KVM, we need to propagate the breakpoint api changes to kvm when we exit the guest and restore the bp registers to the host. Changes in v4: - Drop the hw_breakpoint_restore() stub as it is only used by KVM - EXPORT_SYMBOL_GPL hw_breakpoint_restore() as KVM can be built as a module - Restore the breakpoints unconditionally on kvm guest exit: TIF_DEBUG_THREAD doesn't anymore cover every cases of running breakpoints and vcpu->arch.switch_db_regs might not always be set when the guest used debug registers. (Waiting for a reliable optimization) Changes in v5: - Split-up the asm-generic/hw-breakpoint.h moving to linux/hw_breakpoint.h into a separate patch - Optimize the breakpoints restoring while switching from kvm guest to host. We only want to restore the state if we have active breakpoints to the host, otherwise we don't care about messed-up address registers. - Add asm/hw_breakpoint.h to Kbuild - Fix bad breakpoint type in trace_selftest.c Changes in v6: - Fix wrong header inclusion in trace.h (triggered a build error with CONFIG_FTRACE_SELFTEST Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jan Kiszka <jan.kiszka@web.de> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Avi Kivity <avi@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org>
2009-09-09 19:22:48 +02:00
if (n < HBP_NUM) {
struct perf_event *bp = thread->ptrace_bps[n];
if (bp)
val = bp->hw.info.address;
hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events This patch rebase the implementation of the breakpoints API on top of perf events instances. Each breakpoints are now perf events that handle the register scheduling, thread/cpu attachment, etc.. The new layering is now made as follows: ptrace kgdb ftrace perf syscall \ | / / \ | / / / Core breakpoint API / / | / | / Breakpoints perf events | | Breakpoints PMU ---- Debug Register constraints handling (Part of core breakpoint API) | | Hardware debug registers Reasons of this rewrite: - Use the centralized/optimized pmu registers scheduling, implying an easier arch integration - More powerful register handling: perf attributes (pinned/flexible events, exclusive/non-exclusive, tunable period, etc...) Impact: - New perf ABI: the hardware breakpoints counters - Ptrace breakpoints setting remains tricky and still needs some per thread breakpoints references. Todo (in the order): - Support breakpoints perf counter events for perf tools (ie: implement perf_bpcounter_event()) - Support from perf tools Changes in v2: - Follow the perf "event " rename - The ptrace regression have been fixed (ptrace breakpoint perf events weren't released when a task ended) - Drop the struct hw_breakpoint and store generic fields in perf_event_attr. - Separate core and arch specific headers, drop asm-generic/hw_breakpoint.h and create linux/hw_breakpoint.h - Use new generic len/type for breakpoint - Handle off case: when breakpoints api is not supported by an arch Changes in v3: - Fix broken CONFIG_KVM, we need to propagate the breakpoint api changes to kvm when we exit the guest and restore the bp registers to the host. Changes in v4: - Drop the hw_breakpoint_restore() stub as it is only used by KVM - EXPORT_SYMBOL_GPL hw_breakpoint_restore() as KVM can be built as a module - Restore the breakpoints unconditionally on kvm guest exit: TIF_DEBUG_THREAD doesn't anymore cover every cases of running breakpoints and vcpu->arch.switch_db_regs might not always be set when the guest used debug registers. (Waiting for a reliable optimization) Changes in v5: - Split-up the asm-generic/hw-breakpoint.h moving to linux/hw_breakpoint.h into a separate patch - Optimize the breakpoints restoring while switching from kvm guest to host. We only want to restore the state if we have active breakpoints to the host, otherwise we don't care about messed-up address registers. - Add asm/hw_breakpoint.h to Kbuild - Fix bad breakpoint type in trace_selftest.c Changes in v6: - Fix wrong header inclusion in trace.h (triggered a build error with CONFIG_FTRACE_SELFTEST Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jan Kiszka <jan.kiszka@web.de> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Avi Kivity <avi@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org>
2009-09-09 19:22:48 +02:00
} else if (n == 6) {
val = thread->debugreg6;
} else if (n == 7) {
val = thread->ptrace_dr7;
hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events This patch rebase the implementation of the breakpoints API on top of perf events instances. Each breakpoints are now perf events that handle the register scheduling, thread/cpu attachment, etc.. The new layering is now made as follows: ptrace kgdb ftrace perf syscall \ | / / \ | / / / Core breakpoint API / / | / | / Breakpoints perf events | | Breakpoints PMU ---- Debug Register constraints handling (Part of core breakpoint API) | | Hardware debug registers Reasons of this rewrite: - Use the centralized/optimized pmu registers scheduling, implying an easier arch integration - More powerful register handling: perf attributes (pinned/flexible events, exclusive/non-exclusive, tunable period, etc...) Impact: - New perf ABI: the hardware breakpoints counters - Ptrace breakpoints setting remains tricky and still needs some per thread breakpoints references. Todo (in the order): - Support breakpoints perf counter events for perf tools (ie: implement perf_bpcounter_event()) - Support from perf tools Changes in v2: - Follow the perf "event " rename - The ptrace regression have been fixed (ptrace breakpoint perf events weren't released when a task ended) - Drop the struct hw_breakpoint and store generic fields in perf_event_attr. - Separate core and arch specific headers, drop asm-generic/hw_breakpoint.h and create linux/hw_breakpoint.h - Use new generic len/type for breakpoint - Handle off case: when breakpoints api is not supported by an arch Changes in v3: - Fix broken CONFIG_KVM, we need to propagate the breakpoint api changes to kvm when we exit the guest and restore the bp registers to the host. Changes in v4: - Drop the hw_breakpoint_restore() stub as it is only used by KVM - EXPORT_SYMBOL_GPL hw_breakpoint_restore() as KVM can be built as a module - Restore the breakpoints unconditionally on kvm guest exit: TIF_DEBUG_THREAD doesn't anymore cover every cases of running breakpoints and vcpu->arch.switch_db_regs might not always be set when the guest used debug registers. (Waiting for a reliable optimization) Changes in v5: - Split-up the asm-generic/hw-breakpoint.h moving to linux/hw_breakpoint.h into a separate patch - Optimize the breakpoints restoring while switching from kvm guest to host. We only want to restore the state if we have active breakpoints to the host, otherwise we don't care about messed-up address registers. - Add asm/hw_breakpoint.h to Kbuild - Fix bad breakpoint type in trace_selftest.c Changes in v6: - Fix wrong header inclusion in trace.h (triggered a build error with CONFIG_FTRACE_SELFTEST Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jan Kiszka <jan.kiszka@web.de> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Avi Kivity <avi@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org>
2009-09-09 19:22:48 +02:00
}
return val;
}
hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events This patch rebase the implementation of the breakpoints API on top of perf events instances. Each breakpoints are now perf events that handle the register scheduling, thread/cpu attachment, etc.. The new layering is now made as follows: ptrace kgdb ftrace perf syscall \ | / / \ | / / / Core breakpoint API / / | / | / Breakpoints perf events | | Breakpoints PMU ---- Debug Register constraints handling (Part of core breakpoint API) | | Hardware debug registers Reasons of this rewrite: - Use the centralized/optimized pmu registers scheduling, implying an easier arch integration - More powerful register handling: perf attributes (pinned/flexible events, exclusive/non-exclusive, tunable period, etc...) Impact: - New perf ABI: the hardware breakpoints counters - Ptrace breakpoints setting remains tricky and still needs some per thread breakpoints references. Todo (in the order): - Support breakpoints perf counter events for perf tools (ie: implement perf_bpcounter_event()) - Support from perf tools Changes in v2: - Follow the perf "event " rename - The ptrace regression have been fixed (ptrace breakpoint perf events weren't released when a task ended) - Drop the struct hw_breakpoint and store generic fields in perf_event_attr. - Separate core and arch specific headers, drop asm-generic/hw_breakpoint.h and create linux/hw_breakpoint.h - Use new generic len/type for breakpoint - Handle off case: when breakpoints api is not supported by an arch Changes in v3: - Fix broken CONFIG_KVM, we need to propagate the breakpoint api changes to kvm when we exit the guest and restore the bp registers to the host. Changes in v4: - Drop the hw_breakpoint_restore() stub as it is only used by KVM - EXPORT_SYMBOL_GPL hw_breakpoint_restore() as KVM can be built as a module - Restore the breakpoints unconditionally on kvm guest exit: TIF_DEBUG_THREAD doesn't anymore cover every cases of running breakpoints and vcpu->arch.switch_db_regs might not always be set when the guest used debug registers. (Waiting for a reliable optimization) Changes in v5: - Split-up the asm-generic/hw-breakpoint.h moving to linux/hw_breakpoint.h into a separate patch - Optimize the breakpoints restoring while switching from kvm guest to host. We only want to restore the state if we have active breakpoints to the host, otherwise we don't care about messed-up address registers. - Add asm/hw_breakpoint.h to Kbuild - Fix bad breakpoint type in trace_selftest.c Changes in v6: - Fix wrong header inclusion in trace.h (triggered a build error with CONFIG_FTRACE_SELFTEST Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jan Kiszka <jan.kiszka@web.de> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Avi Kivity <avi@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org>
2009-09-09 19:22:48 +02:00
static int ptrace_set_breakpoint_addr(struct task_struct *tsk, int nr,
unsigned long addr)
{
struct thread_struct *t = &tsk->thread;
struct perf_event *bp = t->ptrace_bps[nr];
int err = 0;
if (!bp) {
/*
* Put stub len and type to create an inactive but correct bp.
*
* CHECKME: the previous code returned -EIO if the addr wasn't
* a valid task virtual addr. The new one will return -EINVAL in
* this case.
* -EINVAL may be what we want for in-kernel breakpoints users,
* but -EIO looks better for ptrace, since we refuse a register
* writing for the user. And anyway this is the previous
* behaviour.
*/
bp = ptrace_register_breakpoint(tsk,
X86_BREAKPOINT_LEN_1, X86_BREAKPOINT_WRITE,
addr, true);
if (IS_ERR(bp))
err = PTR_ERR(bp);
else
t->ptrace_bps[nr] = bp;
hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events This patch rebase the implementation of the breakpoints API on top of perf events instances. Each breakpoints are now perf events that handle the register scheduling, thread/cpu attachment, etc.. The new layering is now made as follows: ptrace kgdb ftrace perf syscall \ | / / \ | / / / Core breakpoint API / / | / | / Breakpoints perf events | | Breakpoints PMU ---- Debug Register constraints handling (Part of core breakpoint API) | | Hardware debug registers Reasons of this rewrite: - Use the centralized/optimized pmu registers scheduling, implying an easier arch integration - More powerful register handling: perf attributes (pinned/flexible events, exclusive/non-exclusive, tunable period, etc...) Impact: - New perf ABI: the hardware breakpoints counters - Ptrace breakpoints setting remains tricky and still needs some per thread breakpoints references. Todo (in the order): - Support breakpoints perf counter events for perf tools (ie: implement perf_bpcounter_event()) - Support from perf tools Changes in v2: - Follow the perf "event " rename - The ptrace regression have been fixed (ptrace breakpoint perf events weren't released when a task ended) - Drop the struct hw_breakpoint and store generic fields in perf_event_attr. - Separate core and arch specific headers, drop asm-generic/hw_breakpoint.h and create linux/hw_breakpoint.h - Use new generic len/type for breakpoint - Handle off case: when breakpoints api is not supported by an arch Changes in v3: - Fix broken CONFIG_KVM, we need to propagate the breakpoint api changes to kvm when we exit the guest and restore the bp registers to the host. Changes in v4: - Drop the hw_breakpoint_restore() stub as it is only used by KVM - EXPORT_SYMBOL_GPL hw_breakpoint_restore() as KVM can be built as a module - Restore the breakpoints unconditionally on kvm guest exit: TIF_DEBUG_THREAD doesn't anymore cover every cases of running breakpoints and vcpu->arch.switch_db_regs might not always be set when the guest used debug registers. (Waiting for a reliable optimization) Changes in v5: - Split-up the asm-generic/hw-breakpoint.h moving to linux/hw_breakpoint.h into a separate patch - Optimize the breakpoints restoring while switching from kvm guest to host. We only want to restore the state if we have active breakpoints to the host, otherwise we don't care about messed-up address registers. - Add asm/hw_breakpoint.h to Kbuild - Fix bad breakpoint type in trace_selftest.c Changes in v6: - Fix wrong header inclusion in trace.h (triggered a build error with CONFIG_FTRACE_SELFTEST Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jan Kiszka <jan.kiszka@web.de> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Avi Kivity <avi@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org>
2009-09-09 19:22:48 +02:00
} else {
struct perf_event_attr attr = bp->attr;
attr.bp_addr = addr;
err = modify_user_hw_breakpoint(bp, &attr);
}
return err;
}
/*
* Handle PTRACE_POKEUSR calls for the debug register area.
*/
static int ptrace_set_debugreg(struct task_struct *tsk, int n,
unsigned long val)
{
struct thread_struct *thread = &tsk->thread;
/* There are no DR4 or DR5 registers */
int rc = -EIO;
if (n < HBP_NUM) {
hw-breakpoints: Rewrite the hw-breakpoints layer on top of perf events This patch rebase the implementation of the breakpoints API on top of perf events instances. Each breakpoints are now perf events that handle the register scheduling, thread/cpu attachment, etc.. The new layering is now made as follows: ptrace kgdb ftrace perf syscall \ | / / \ | / / / Core breakpoint API / / | / | / Breakpoints perf events | | Breakpoints PMU ---- Debug Register constraints handling (Part of core breakpoint API) | | Hardware debug registers Reasons of this rewrite: - Use the centralized/optimized pmu registers scheduling, implying an easier arch integration - More powerful register handling: perf attributes (pinned/flexible events, exclusive/non-exclusive, tunable period, etc...) Impact: - New perf ABI: the hardware breakpoints counters - Ptrace breakpoints setting remains tricky and still needs some per thread breakpoints references. Todo (in the order): - Support breakpoints perf counter events for perf tools (ie: implement perf_bpcounter_event()) - Support from perf tools Changes in v2: - Follow the perf "event " rename - The ptrace regression have been fixed (ptrace breakpoint perf events weren't released when a task ended) - Drop the struct hw_breakpoint and store generic fields in perf_event_attr. - Separate core and arch specific headers, drop asm-generic/hw_breakpoint.h and create linux/hw_breakpoint.h - Use new generic len/type for breakpoint - Handle off case: when breakpoints api is not supported by an arch Changes in v3: - Fix broken CONFIG_KVM, we need to propagate the breakpoint api changes to kvm when we exit the guest and restore the bp registers to the host. Changes in v4: - Drop the hw_breakpoint_restore() stub as it is only used by KVM - EXPORT_SYMBOL_GPL hw_breakpoint_restore() as KVM can be built as a module - Restore the breakpoints unconditionally on kvm guest exit: TIF_DEBUG_THREAD doesn't anymore cover every cases of running breakpoints and vcpu->arch.switch_db_regs might not always be set when the guest used debug registers. (Waiting for a reliable optimization) Changes in v5: - Split-up the asm-generic/hw-breakpoint.h moving to linux/hw_breakpoint.h into a separate patch - Optimize the breakpoints restoring while switching from kvm guest to host. We only want to restore the state if we have active breakpoints to the host, otherwise we don't care about messed-up address registers. - Add asm/hw_breakpoint.h to Kbuild - Fix bad breakpoint type in trace_selftest.c Changes in v6: - Fix wrong header inclusion in trace.h (triggered a build error with CONFIG_FTRACE_SELFTEST Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Jan Kiszka <jan.kiszka@web.de> Cc: Jiri Slaby <jirislaby@gmail.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Cc: Avi Kivity <avi@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Masami Hiramatsu <mhiramat@redhat.com> Cc: Paul Mundt <lethal@linux-sh.org>
2009-09-09 19:22:48 +02:00
rc = ptrace_set_breakpoint_addr(tsk, n, val);
} else if (n == 6) {
thread->debugreg6 = val;
rc = 0;
} else if (n == 7) {
rc = ptrace_write_dr7(tsk, val);
if (!rc)
thread->ptrace_dr7 = val;
}
return rc;
}
/*
* These access the current or another (stopped) task's io permission
* bitmap for debugging or core dump.
*/
static int ioperm_active(struct task_struct *target,
const struct user_regset *regset)
{
return target->thread.io_bitmap_max / regset->size;
}
static int ioperm_get(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
void *kbuf, void __user *ubuf)
{
if (!target->thread.io_bitmap_ptr)
return -ENXIO;
return user_regset_copyout(&pos, &count, &kbuf, &ubuf,
target->thread.io_bitmap_ptr,
0, IO_BITMAP_BYTES);
}
/*
* Called by kernel/ptrace.c when detaching..
*
* Make sure the single step bit is not set.
*/
void ptrace_disable(struct task_struct *child)
{
user_disable_single_step(child);
#ifdef TIF_SYSCALL_EMU
clear_tsk_thread_flag(child, TIF_SYSCALL_EMU);
#endif
}
#if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION
static const struct user_regset_view user_x86_32_view; /* Initialized below. */
#endif
long arch_ptrace(struct task_struct *child, long request,
unsigned long addr, unsigned long data)
{
int ret;
unsigned long __user *datap = (unsigned long __user *)data;
switch (request) {
/* read the word at location addr in the USER area. */
case PTRACE_PEEKUSR: {
unsigned long tmp;
ret = -EIO;
if ((addr & (sizeof(data) - 1)) || addr >= sizeof(struct user))
break;
tmp = 0; /* Default return condition */
if (addr < sizeof(struct user_regs_struct))
tmp = getreg(child, addr);
else if (addr >= offsetof(struct user, u_debugreg[0]) &&
addr <= offsetof(struct user, u_debugreg[7])) {
addr -= offsetof(struct user, u_debugreg[0]);
tmp = ptrace_get_debugreg(child, addr / sizeof(data));
}
ret = put_user(tmp, datap);
break;
}
case PTRACE_POKEUSR: /* write the word at location addr in the USER area */
ret = -EIO;
if ((addr & (sizeof(data) - 1)) || addr >= sizeof(struct user))
break;
if (addr < sizeof(struct user_regs_struct))
ret = putreg(child, addr, data);
else if (addr >= offsetof(struct user, u_debugreg[0]) &&
addr <= offsetof(struct user, u_debugreg[7])) {
addr -= offsetof(struct user, u_debugreg[0]);
ret = ptrace_set_debugreg(child,
addr / sizeof(data), data);
}
break;
case PTRACE_GETREGS: /* Get all gp regs from the child. */
return copy_regset_to_user(child,
task_user_regset_view(current),
REGSET_GENERAL,
0, sizeof(struct user_regs_struct),
datap);
case PTRACE_SETREGS: /* Set all gp regs in the child. */
return copy_regset_from_user(child,
task_user_regset_view(current),
REGSET_GENERAL,
0, sizeof(struct user_regs_struct),
datap);
case PTRACE_GETFPREGS: /* Get the child FPU state. */
return copy_regset_to_user(child,
task_user_regset_view(current),
REGSET_FP,
0, sizeof(struct user_i387_struct),
datap);
case PTRACE_SETFPREGS: /* Set the child FPU state. */
return copy_regset_from_user(child,
task_user_regset_view(current),
REGSET_FP,
0, sizeof(struct user_i387_struct),
datap);
#ifdef CONFIG_X86_32
case PTRACE_GETFPXREGS: /* Get the child extended FPU state. */
return copy_regset_to_user(child, &user_x86_32_view,
REGSET_XFP,
0, sizeof(struct user_fxsr_struct),
datap) ? -EIO : 0;
case PTRACE_SETFPXREGS: /* Set the child extended FPU state. */
return copy_regset_from_user(child, &user_x86_32_view,
REGSET_XFP,
0, sizeof(struct user_fxsr_struct),
datap) ? -EIO : 0;
#endif
#if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION
case PTRACE_GET_THREAD_AREA:
if ((int) addr < 0)
return -EIO;
ret = do_get_thread_area(child, addr,
(struct user_desc __user *)data);
break;
case PTRACE_SET_THREAD_AREA:
if ((int) addr < 0)
return -EIO;
ret = do_set_thread_area(child, addr,
(struct user_desc __user *)data, 0);
break;
#endif
#ifdef CONFIG_X86_64
/* normal 64bit interface to access TLS data.
Works just like arch_prctl, except that the arguments
are reversed. */
case PTRACE_ARCH_PRCTL:
ret = do_arch_prctl(child, data, addr);
break;
#endif
default:
ret = ptrace_request(child, request, addr, data);
break;
}
return ret;
}
#ifdef CONFIG_IA32_EMULATION
#include <linux/compat.h>
#include <linux/syscalls.h>
#include <asm/ia32.h>
#include <asm/user32.h>
#define R32(l,q) \
case offsetof(struct user32, regs.l): \
regs->q = value; break
#define SEG32(rs) \
case offsetof(struct user32, regs.rs): \
return set_segment_reg(child, \
offsetof(struct user_regs_struct, rs), \
value); \
break
static int putreg32(struct task_struct *child, unsigned regno, u32 value)
{
struct pt_regs *regs = task_pt_regs(child);
switch (regno) {
SEG32(cs);
SEG32(ds);
SEG32(es);
SEG32(fs);
SEG32(gs);
SEG32(ss);
R32(ebx, bx);
R32(ecx, cx);
R32(edx, dx);
R32(edi, di);
R32(esi, si);
R32(ebp, bp);
R32(eax, ax);
R32(eip, ip);
R32(esp, sp);
case offsetof(struct user32, regs.orig_eax):
/*
* A 32-bit debugger setting orig_eax means to restore
* the state of the task restarting a 32-bit syscall.
* Make sure we interpret the -ERESTART* codes correctly
* in case the task is not actually still sitting at the
* exit from a 32-bit syscall with TS_COMPAT still set.
*/
regs->orig_ax = value;
if (syscall_get_nr(child, regs) >= 0)
task_thread_info(child)->status |= TS_COMPAT;
break;
case offsetof(struct user32, regs.eflags):
return set_flags(child, value);
case offsetof(struct user32, u_debugreg[0]) ...
offsetof(struct user32, u_debugreg[7]):
regno -= offsetof(struct user32, u_debugreg[0]);
return ptrace_set_debugreg(child, regno / 4, value);
default:
if (regno > sizeof(struct user32) || (regno & 3))
return -EIO;
/*
* Other dummy fields in the virtual user structure
* are ignored
*/
break;
}
return 0;
}
#undef R32
#undef SEG32
#define R32(l,q) \
case offsetof(struct user32, regs.l): \
*val = regs->q; break
#define SEG32(rs) \
case offsetof(struct user32, regs.rs): \
*val = get_segment_reg(child, \
offsetof(struct user_regs_struct, rs)); \
break
static int getreg32(struct task_struct *child, unsigned regno, u32 *val)
{
struct pt_regs *regs = task_pt_regs(child);
switch (regno) {
SEG32(ds);
SEG32(es);
SEG32(fs);
SEG32(gs);
R32(cs, cs);
R32(ss, ss);
R32(ebx, bx);
R32(ecx, cx);
R32(edx, dx);
R32(edi, di);
R32(esi, si);
R32(ebp, bp);
R32(eax, ax);
R32(orig_eax, orig_ax);
R32(eip, ip);
R32(esp, sp);
case offsetof(struct user32, regs.eflags):
*val = get_flags(child);
break;
case offsetof(struct user32, u_debugreg[0]) ...
offsetof(struct user32, u_debugreg[7]):
regno -= offsetof(struct user32, u_debugreg[0]);
*val = ptrace_get_debugreg(child, regno / 4);
break;
default:
if (regno > sizeof(struct user32) || (regno & 3))
return -EIO;
/*
* Other dummy fields in the virtual user structure
* are ignored
*/
*val = 0;
break;
}
return 0;
}
#undef R32
#undef SEG32
static int genregs32_get(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
void *kbuf, void __user *ubuf)
{
if (kbuf) {
compat_ulong_t *k = kbuf;
while (count >= sizeof(*k)) {
getreg32(target, pos, k++);
count -= sizeof(*k);
pos += sizeof(*k);
}
} else {
compat_ulong_t __user *u = ubuf;
while (count >= sizeof(*u)) {
compat_ulong_t word;
getreg32(target, pos, &word);
if (__put_user(word, u++))
return -EFAULT;
count -= sizeof(*u);
pos += sizeof(*u);
}
}
return 0;
}
static int genregs32_set(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
const void *kbuf, const void __user *ubuf)
{
int ret = 0;
if (kbuf) {
const compat_ulong_t *k = kbuf;
while (count >= sizeof(*k) && !ret) {
ret = putreg32(target, pos, *k++);
count -= sizeof(*k);
pos += sizeof(*k);
}
} else {
const compat_ulong_t __user *u = ubuf;
while (count >= sizeof(*u) && !ret) {
compat_ulong_t word;
ret = __get_user(word, u++);
if (ret)
break;
ret = putreg32(target, pos, word);
count -= sizeof(*u);
pos += sizeof(*u);
}
}
return ret;
}
#ifdef CONFIG_X86_X32_ABI
static long x32_arch_ptrace(struct task_struct *child,
compat_long_t request, compat_ulong_t caddr,
compat_ulong_t cdata)
{
unsigned long addr = caddr;
unsigned long data = cdata;
void __user *datap = compat_ptr(data);
int ret;
switch (request) {
/* Read 32bits at location addr in the USER area. Only allow
to return the lower 32bits of segment and debug registers. */
case PTRACE_PEEKUSR: {
u32 tmp;
ret = -EIO;
if ((addr & (sizeof(data) - 1)) || addr >= sizeof(struct user) ||
addr < offsetof(struct user_regs_struct, cs))
break;
tmp = 0; /* Default return condition */
if (addr < sizeof(struct user_regs_struct))
tmp = getreg(child, addr);
else if (addr >= offsetof(struct user, u_debugreg[0]) &&
addr <= offsetof(struct user, u_debugreg[7])) {
addr -= offsetof(struct user, u_debugreg[0]);
tmp = ptrace_get_debugreg(child, addr / sizeof(data));
}
ret = put_user(tmp, (__u32 __user *)datap);
break;
}
/* Write the word at location addr in the USER area. Only allow
to update segment and debug registers with the upper 32bits
zero-extended. */
case PTRACE_POKEUSR:
ret = -EIO;
if ((addr & (sizeof(data) - 1)) || addr >= sizeof(struct user) ||
addr < offsetof(struct user_regs_struct, cs))
break;
if (addr < sizeof(struct user_regs_struct))
ret = putreg(child, addr, data);
else if (addr >= offsetof(struct user, u_debugreg[0]) &&
addr <= offsetof(struct user, u_debugreg[7])) {
addr -= offsetof(struct user, u_debugreg[0]);
ret = ptrace_set_debugreg(child,
addr / sizeof(data), data);
}
break;
case PTRACE_GETREGS: /* Get all gp regs from the child. */
return copy_regset_to_user(child,
task_user_regset_view(current),
REGSET_GENERAL,
0, sizeof(struct user_regs_struct),
datap);
case PTRACE_SETREGS: /* Set all gp regs in the child. */
return copy_regset_from_user(child,
task_user_regset_view(current),
REGSET_GENERAL,
0, sizeof(struct user_regs_struct),
datap);
case PTRACE_GETFPREGS: /* Get the child FPU state. */
return copy_regset_to_user(child,
task_user_regset_view(current),
REGSET_FP,
0, sizeof(struct user_i387_struct),
datap);
case PTRACE_SETFPREGS: /* Set the child FPU state. */
return copy_regset_from_user(child,
task_user_regset_view(current),
REGSET_FP,
0, sizeof(struct user_i387_struct),
datap);
default:
return compat_ptrace_request(child, request, addr, data);
}
return ret;
}
#endif
long compat_arch_ptrace(struct task_struct *child, compat_long_t request,
compat_ulong_t caddr, compat_ulong_t cdata)
{
unsigned long addr = caddr;
unsigned long data = cdata;
void __user *datap = compat_ptr(data);
int ret;
__u32 val;
#ifdef CONFIG_X86_X32_ABI
if (!is_ia32_task())
return x32_arch_ptrace(child, request, caddr, cdata);
#endif
switch (request) {
case PTRACE_PEEKUSR:
ret = getreg32(child, addr, &val);
if (ret == 0)
ret = put_user(val, (__u32 __user *)datap);
break;
case PTRACE_POKEUSR:
ret = putreg32(child, addr, data);
break;
case PTRACE_GETREGS: /* Get all gp regs from the child. */
return copy_regset_to_user(child, &user_x86_32_view,
REGSET_GENERAL,
0, sizeof(struct user_regs_struct32),
datap);
case PTRACE_SETREGS: /* Set all gp regs in the child. */
return copy_regset_from_user(child, &user_x86_32_view,
REGSET_GENERAL, 0,
sizeof(struct user_regs_struct32),
datap);
case PTRACE_GETFPREGS: /* Get the child FPU state. */
return copy_regset_to_user(child, &user_x86_32_view,
REGSET_FP, 0,
sizeof(struct user_i387_ia32_struct),
datap);
case PTRACE_SETFPREGS: /* Set the child FPU state. */
return copy_regset_from_user(
child, &user_x86_32_view, REGSET_FP,
0, sizeof(struct user_i387_ia32_struct), datap);
case PTRACE_GETFPXREGS: /* Get the child extended FPU state. */
return copy_regset_to_user(child, &user_x86_32_view,
REGSET_XFP, 0,
sizeof(struct user32_fxsr_struct),
datap);
case PTRACE_SETFPXREGS: /* Set the child extended FPU state. */
return copy_regset_from_user(child, &user_x86_32_view,
REGSET_XFP, 0,
sizeof(struct user32_fxsr_struct),
datap);
case PTRACE_GET_THREAD_AREA:
case PTRACE_SET_THREAD_AREA:
return arch_ptrace(child, request, addr, data);
default:
return compat_ptrace_request(child, request, addr, data);
}
return ret;
}
#endif /* CONFIG_IA32_EMULATION */
#ifdef CONFIG_X86_64
static struct user_regset x86_64_regsets[] __read_mostly = {
[REGSET_GENERAL] = {
.core_note_type = NT_PRSTATUS,
.n = sizeof(struct user_regs_struct) / sizeof(long),
.size = sizeof(long), .align = sizeof(long),
.get = genregs_get, .set = genregs_set
},
[REGSET_FP] = {
.core_note_type = NT_PRFPREG,
.n = sizeof(struct user_i387_struct) / sizeof(long),
.size = sizeof(long), .align = sizeof(long),
.active = xfpregs_active, .get = xfpregs_get, .set = xfpregs_set
},
[REGSET_XSTATE] = {
.core_note_type = NT_X86_XSTATE,
.size = sizeof(u64), .align = sizeof(u64),
.active = xstateregs_active, .get = xstateregs_get,
.set = xstateregs_set
},
[REGSET_IOPERM64] = {
.core_note_type = NT_386_IOPERM,
.n = IO_BITMAP_LONGS,
.size = sizeof(long), .align = sizeof(long),
.active = ioperm_active, .get = ioperm_get
},
};
static const struct user_regset_view user_x86_64_view = {
.name = "x86_64", .e_machine = EM_X86_64,
.regsets = x86_64_regsets, .n = ARRAY_SIZE(x86_64_regsets)
};
#else /* CONFIG_X86_32 */
#define user_regs_struct32 user_regs_struct
#define genregs32_get genregs_get
#define genregs32_set genregs_set
#endif /* CONFIG_X86_64 */
#if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION
static struct user_regset x86_32_regsets[] __read_mostly = {
[REGSET_GENERAL] = {
.core_note_type = NT_PRSTATUS,
.n = sizeof(struct user_regs_struct32) / sizeof(u32),
.size = sizeof(u32), .align = sizeof(u32),
.get = genregs32_get, .set = genregs32_set
},
[REGSET_FP] = {
.core_note_type = NT_PRFPREG,
.n = sizeof(struct user_i387_ia32_struct) / sizeof(u32),
.size = sizeof(u32), .align = sizeof(u32),
.active = fpregs_active, .get = fpregs_get, .set = fpregs_set
},
[REGSET_XFP] = {
.core_note_type = NT_PRXFPREG,
.n = sizeof(struct user32_fxsr_struct) / sizeof(u32),
.size = sizeof(u32), .align = sizeof(u32),
.active = xfpregs_active, .get = xfpregs_get, .set = xfpregs_set
},
[REGSET_XSTATE] = {
.core_note_type = NT_X86_XSTATE,
.size = sizeof(u64), .align = sizeof(u64),
.active = xstateregs_active, .get = xstateregs_get,
.set = xstateregs_set
},
[REGSET_TLS] = {
.core_note_type = NT_386_TLS,
.n = GDT_ENTRY_TLS_ENTRIES, .bias = GDT_ENTRY_TLS_MIN,
.size = sizeof(struct user_desc),
.align = sizeof(struct user_desc),
.active = regset_tls_active,
.get = regset_tls_get, .set = regset_tls_set
},
[REGSET_IOPERM32] = {
.core_note_type = NT_386_IOPERM,
.n = IO_BITMAP_BYTES / sizeof(u32),
.size = sizeof(u32), .align = sizeof(u32),
.active = ioperm_active, .get = ioperm_get
},
};
static const struct user_regset_view user_x86_32_view = {
.name = "i386", .e_machine = EM_386,
.regsets = x86_32_regsets, .n = ARRAY_SIZE(x86_32_regsets)
};
#endif
/*
* This represents bytes 464..511 in the memory layout exported through
* the REGSET_XSTATE interface.
*/
u64 xstate_fx_sw_bytes[USER_XSTATE_FX_SW_WORDS];
void update_regset_xstate_info(unsigned int size, u64 xstate_mask)
{
#ifdef CONFIG_X86_64
x86_64_regsets[REGSET_XSTATE].n = size / sizeof(u64);
#endif
#if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION
x86_32_regsets[REGSET_XSTATE].n = size / sizeof(u64);
#endif
xstate_fx_sw_bytes[USER_XSTATE_XCR0_WORD] = xstate_mask;
}
const struct user_regset_view *task_user_regset_view(struct task_struct *task)
{
#ifdef CONFIG_IA32_EMULATION
if (test_tsk_thread_flag(task, TIF_IA32))
#endif
#if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION
return &user_x86_32_view;
#endif
#ifdef CONFIG_X86_64
return &user_x86_64_view;
#endif
}
static void fill_sigtrap_info(struct task_struct *tsk,
struct pt_regs *regs,
int error_code, int si_code,
struct siginfo *info)
{
tsk->thread.trap_nr = X86_TRAP_DB;
tsk->thread.error_code = error_code;
memset(info, 0, sizeof(*info));
info->si_signo = SIGTRAP;
info->si_code = si_code;
info->si_addr = user_mode_vm(regs) ? (void __user *)regs->ip : NULL;
}
void user_single_step_siginfo(struct task_struct *tsk,
struct pt_regs *regs,
struct siginfo *info)
{
fill_sigtrap_info(tsk, regs, 0, TRAP_BRKPT, info);
}
void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs,
int error_code, int si_code)
{
struct siginfo info;
fill_sigtrap_info(tsk, regs, error_code, si_code, &info);
/* Send us the fake SIGTRAP */
force_sig_info(SIGTRAP, &info, tsk);
}
#ifdef CONFIG_X86_32
# define IS_IA32 1
#elif defined CONFIG_IA32_EMULATION
# define IS_IA32 is_compat_task()
#else
# define IS_IA32 0
#endif
/*
* We must return the syscall number to actually look up in the table.
* This can be -1L to skip running any syscall at all.
*/
long syscall_trace_enter(struct pt_regs *regs)
{
long ret = 0;
user_exit();
/*
* If we stepped into a sysenter/syscall insn, it trapped in
* kernel mode; do_debug() cleared TF and set TIF_SINGLESTEP.
* If user-mode had set TF itself, then it's still clear from
* do_debug() and we need to set it again to restore the user
* state. If we entered on the slow path, TF was already set.
*/
if (test_thread_flag(TIF_SINGLESTEP))
regs->flags |= X86_EFLAGS_TF;
/* do the secure computing check first */
if (secure_computing(regs->orig_ax)) {
/* seccomp failures shouldn't expose any additional code. */
ret = -1L;
goto out;
}
if (unlikely(test_thread_flag(TIF_SYSCALL_EMU)))
ret = -1L;
if ((ret || test_thread_flag(TIF_SYSCALL_TRACE)) &&
tracehook_report_syscall_entry(regs))
ret = -1L;
if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
trace_sys_enter(regs, regs->orig_ax);
if (IS_IA32)
audit_syscall_entry(AUDIT_ARCH_I386,
regs->orig_ax,
regs->bx, regs->cx,
regs->dx, regs->si);
#ifdef CONFIG_X86_64
else
audit_syscall_entry(AUDIT_ARCH_X86_64,
regs->orig_ax,
regs->di, regs->si,
regs->dx, regs->r10);
#endif
out:
return ret ?: regs->orig_ax;
}
void syscall_trace_leave(struct pt_regs *regs)
{
bool step;
/*
* We may come here right after calling schedule_user()
* or do_notify_resume(), in which case we can be in RCU
* user mode.
*/
user_exit();
Audit: push audit success and retcode into arch ptrace.h The audit system previously expected arches calling to audit_syscall_exit to supply as arguments if the syscall was a success and what the return code was. Audit also provides a helper AUDITSC_RESULT which was supposed to simplify things by converting from negative retcodes to an audit internal magic value stating success or failure. This helper was wrong and could indicate that a valid pointer returned to userspace was a failed syscall. The fix is to fix the layering foolishness. We now pass audit_syscall_exit a struct pt_reg and it in turns calls back into arch code to collect the return value and to determine if the syscall was a success or failure. We also define a generic is_syscall_success() macro which determines success/failure based on if the value is < -MAX_ERRNO. This works for arches like x86 which do not use a separate mechanism to indicate syscall failure. We make both the is_syscall_success() and regs_return_value() static inlines instead of macros. The reason is because the audit function must take a void* for the regs. (uml calls theirs struct uml_pt_regs instead of just struct pt_regs so audit_syscall_exit can't take a struct pt_regs). Since the audit function takes a void* we need to use static inlines to cast it back to the arch correct structure to dereference it. The other major change is that on some arches, like ia64, MIPS and ppc, we change regs_return_value() to give us the negative value on syscall failure. THE only other user of this macro, kretprobe_example.c, won't notice and it makes the value signed consistently for the audit functions across all archs. In arch/sh/kernel/ptrace_64.c I see that we were using regs[9] in the old audit code as the return value. But the ptrace_64.h code defined the macro regs_return_value() as regs[3]. I have no idea which one is correct, but this patch now uses the regs_return_value() function, so it now uses regs[3]. For powerpc we previously used regs->result but now use the regs_return_value() function which uses regs->gprs[3]. regs->gprs[3] is always positive so the regs_return_value(), much like ia64 makes it negative before calling the audit code when appropriate. Signed-off-by: Eric Paris <eparis@redhat.com> Acked-by: H. Peter Anvin <hpa@zytor.com> [for x86 portion] Acked-by: Tony Luck <tony.luck@intel.com> [for ia64] Acked-by: Richard Weinberger <richard@nod.at> [for uml] Acked-by: David S. Miller <davem@davemloft.net> [for sparc] Acked-by: Ralf Baechle <ralf@linux-mips.org> [for mips] Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> [for ppc]
2012-01-03 20:23:06 +01:00
audit_syscall_exit(regs);
if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
trace_sys_exit(regs, regs->ax);
/*
* If TIF_SYSCALL_EMU is set, we only get here because of
* TIF_SINGLESTEP (i.e. this is PTRACE_SYSEMU_SINGLESTEP).
* We already reported this syscall instruction in
* syscall_trace_enter().
*/
step = unlikely(test_thread_flag(TIF_SINGLESTEP)) &&
!test_thread_flag(TIF_SYSCALL_EMU);
if (step || test_thread_flag(TIF_SYSCALL_TRACE))
tracehook_report_syscall_exit(regs, step);
user_enter();
}