Instead of iterating through all instructions to find the last
instruction each time .rela.discard.(un)reachable points beyond the
section, use find_insn to locate the last instruction by looking at
the last bytes of the section instead.
Suggested-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: NSami Tolvanen <samitolvanen@google.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200421220843.188260-3-samitolvanen@google.com

Currently, objtool fails to load the correct section for symbols when
the index is greater than SHN_LORESERVE. Use gelf_getsymshndx instead
of gelf_getsym to handle >64k sections.
Signed-off-by: NSami Tolvanen <samitolvanen@google.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NKees Cook <keescook@chromium.org>
Link: https://lkml.kernel.org/r/20200421220843.188260-2-samitolvanen@google.com

Randy reported a false-positive:

  arch/x86/hyperv/hv_apic.o: warning: objtool: hv_apic_write()+0x25: alternative modifies stack

What happens is that:

	alternative_io("movl %0, %P1", "xchgl %0, %P1", X86_BUG_11AP,
 13d:   89 9d 00 d0 7f ff       mov    %ebx,-0x803000(%rbp)

decodes to an instruction with CFI-ops because it modifies RBP.
However, due to this being a !frame-pointer build, that should not in
fact change the CFI state.

So instead of dis-allowing any CFI-op, verify the op would've actually
changed the CFI state.

Fixes: 7117f16b ("objtool: Fix ORC vs alternatives")
Reported-by: NRandy Dunlap <rdunlap@infradead.org>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Acked-by: NRandy Dunlap <rdunlap@infradead.org>
Tested-by: NRandy Dunlap <rdunlap@infradead.org>

Currently objtool cannot understand retpolines, and thus cannot
generate ORC unwind information for them. This means that we cannot
unwind from the middle of a retpoline.

The recent ANNOTATE_INTRA_FUNCTION_CALL and UNWIND_HINT_RET_OFFSET
support in objtool enables it to understand the basic retpoline
construct. A further problem is that the ORC unwind information is
alternative invariant; IOW. every alternative should have the same
ORC, retpolines obviously violate this. This means we need to
out-of-line them.

Since all GCC generated code already uses out-of-line retpolines, this
should not affect performance much, if anything.

This will enable objtool to generate valid ORC data for the
out-of-line copies, which means we can correctly and reliably unwind
through a retpoline.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200428191700.210835357@infradead.org

In order to change the {JMP,CALL}_NOSPEC macros to call out-of-line
versions of the retpoline magic, we need to remove the '%' from the
argument, such that we can paste it onto symbol names.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200428191700.151623523@infradead.org

Because of how KSYM works, we need one declaration per line. Seeing
how we're going to be doubling the amount of retpoline symbols,
simplify the machinery in order to avoid having to copy/paste even
more.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200428191700.091696925@infradead.org

Change FILL_RETURN_BUFFER so that objtool groks it and can generate
correct ORC unwind information.

 - Since ORC is alternative invariant; that is, all alternatives
   should have the same ORC entries, the __FILL_RETURN_BUFFER body
   can not be part of an alternative.

   Therefore, move it out of the alternative and keep the alternative
   as a sort of jump_label around it.

 - Use the ANNOTATE_INTRA_FUNCTION_CALL annotation to white-list
   these 'funny' call instructions to nowhere.

 - Use UNWIND_HINT_EMPTY to 'fill' the speculation traps, otherwise
   objtool will consider them unreachable.

 - Move the RSP adjustment into the loop, such that the loop has a
   deterministic stack layout.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200428191700.032079304@infradead.org

Change objtool to support intra-function calls. On x86, an intra-function
call is represented in objtool as a push onto the stack (of the return
address), and a jump to the destination address. That way the stack
information is correctly updated and the call flow is still accurate.
Signed-off-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200414103618.12657-4-alexandre.chartre@oracle.com

Quoting Julien:

  "And the other suggestion is my other email was that you don't even
  need to add INSN_EXCEPTION_RETURN. You can keep IRET as
  INSN_CONTEXT_SWITCH by default and x86 decoder lookups the symbol
  conaining an iret. If it's a function symbol, it can just set the type
  to INSN_OTHER so that it caries on to the next instruction after
  having handled the stack_op."
Suggested-by: NJulien Thierry <jthierry@redhat.com>
Signed-off-by: NMiroslav Benes <mbenes@suse.cz>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200428191659.913283807@infradead.org

With the unconditional use of handle_insn_ops(), INSN_STACK has lost
its purpose. Remove it.
Suggested-by: NJulien Thierry <jthierry@redhat.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200428191659.854203028@infradead.org

Now that every instruction has a list of stack_ops; we can trivially
distinquish those instructions that do not have stack_ops, their list
is empty.

This means we can now call handle_insn_ops() unconditionally.
Suggested-by: NJulien Thierry <jthierry@redhat.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200428191659.795115188@infradead.org

Wrap each stack_op in a macro that allocates and adds it to the list.
This simplifies trying to figure out what to do with the pre-allocated
stack_op at the end.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200428191659.736151601@infradead.org

UNWIND_HINT_RET_OFFSET will adjust a modified stack. However if a
callee-saved register was pushed on the stack then the stack frame
will still appear modified. So stop checking registers when
UNWIND_HINT_RET_OFFSET is used.
Signed-off-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200407073142.20659-3-alexandre.chartre@oracle.com

Fix is_fentry_call() so that it works if a call has no destination
set (call_dest). This needs to be done in order to support intra-
function calls.
Signed-off-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200414103618.12657-2-alexandre.chartre@oracle.com

As reported by objtool:

  lib/ubsan.o: warning: objtool: .altinstr_replacement+0x0: alternative modifies stack
  lib/ubsan.o: warning: objtool: .altinstr_replacement+0x7: alternative modifies stack

the smap_{save,restore}() alternatives violate (the newly enforced)
rule on stack invariance. That is, due to there only being a single
ORC table it must be valid to any alternative. These alternatives
violate this with the direct result that unwinds will not be correct
when it hits between the PUSH and POP instructions.

Rewrite the functions to only have a conditional jump.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200429101802.GI13592@hirez.programming.kicks-ass.net

Jann reported that (for instance) entry_64.o:general_protection has
very odd ORC data:

  0000000000000f40 <general_protection>:
  #######sp:sp+8 bp:(und) type:iret end:0
    f40:       90                      nop
  #######sp:(und) bp:(und) type:call end:0
    f41:       90                      nop
    f42:       90                      nop
  #######sp:sp+8 bp:(und) type:iret end:0
    f43:       e8 a8 01 00 00          callq  10f0 <error_entry>
  #######sp:sp+0 bp:(und) type:regs end:0
    f48:       f6 84 24 88 00 00 00    testb  $0x3,0x88(%rsp)
    f4f:       03
    f50:       74 00                   je     f52 <general_protection+0x12>
    f52:       48 89 e7                mov    %rsp,%rdi
    f55:       48 8b 74 24 78          mov    0x78(%rsp),%rsi
    f5a:       48 c7 44 24 78 ff ff    movq   $0xffffffffffffffff,0x78(%rsp)
    f61:       ff ff
    f63:       e8 00 00 00 00          callq  f68 <general_protection+0x28>
    f68:       e9 73 02 00 00          jmpq   11e0 <error_exit>
  #######sp:(und) bp:(und) type:call end:0
    f6d:       0f 1f 00                nopl   (%rax)

Note the entry at 0xf41. Josh found this was the result of commit:

  764eef4b ("objtool: Rewrite alt->skip_orig")

Due to the early return in validate_branch() we no longer set
insn->cfi of the original instruction stream (the NOPs at 0xf41 and
0xf42) and we'll end up with the above weirdness.

In other discussions we realized alternatives should be ORC invariant;
that is, due to there being only a single ORC table, it must be valid
for all alternatives. The easiest way to ensure this is to not allow
any stack modifications in alternatives.

When we enforce this latter observation, we get the property that the
whole alternative must have the same CFI, which we can employ to fix
the former report.

Fixes: 764eef4b ("objtool: Rewrite alt->skip_orig")
Reported-by: NJann Horn <jannh@google.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200428191659.499074346@infradead.org

Assign a unique identifier to every alternative instruction group in
order to be able to tell which instructions belong to what
alternative.

[peterz: extracted from a larger patch]
Signed-off-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>

While jumping from outside an alternative region to the middle of an
alternative region is very likely wrong, jumping from an alternative
region into the same region is valid. It is a common pattern on arm64.

The first pattern is unlikely to happen in practice and checking only
for this adds a lot of complexity.

Just remove the current check.
Suggested-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: NJulien Thierry <jthierry@redhat.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Link: https://lkml.kernel.org/r/20200327152847.15294-6-jthierry@redhat.com

Mostly straightforward constification, except that WARN_FUNC()
needs a writable pointer while we have read-only pointers,
so deflect this to WARN().
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: NIngo Molnar <mingo@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Sami Tolvanen <samitolvanen@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20200422103205.61900-4-mingo@kernel.org

'struct elf *' handling is an open/close paradigm, make sure the naming
matches that:

   elf_open_read()
   elf_write()
   elf_close()
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: NIngo Molnar <mingo@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Sami Tolvanen <samitolvanen@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20200422103205.61900-3-mingo@kernel.org

In preparation to parallelize certain parts of objtool, map out which uses
of various data structures are read-only vs. read-write.

As a first step constify 'struct elf' pointer passing, most of the secondary
uses of it in find_symbol_*() methods are read-only.

Also, while at it, better group the 'struct elf' handling methods in elf.h.
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: NIngo Molnar <mingo@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Sami Tolvanen <samitolvanen@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20200422103205.61900-2-mingo@kernel.org

Consider all of .entry.text as noinstr. This gets us coverage across
the PTI boundary. While we could add everything .noinstr.text into
.entry.text that would bloat the amount of code in the user mapping.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200416115119.525037514@infradead.orgSigned-off-by: NIngo Molnar <mingo@kernel.org>

Make sure to also check STT_NOTYPE symbols for noinstr violations.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200416115119.465335884@infradead.orgSigned-off-by: NIngo Molnar <mingo@kernel.org>

In preparation of further changes, once again break out the loop body.
No functional changes intended.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200416115119.405863817@infradead.orgSigned-off-by: NIngo Molnar <mingo@kernel.org>

validate_functions() iterates all sections their symbols; this is
pointless to do for !text sections as they won't have instructions
anyway.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200416115119.346582716@infradead.orgSigned-off-by: NIngo Molnar <mingo@kernel.org>

Now that objtool is capable of processing vmlinux.o and actually has
something useful to do there, (conditionally) add it to the final link
pass.

This will increase build time by a few seconds.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200416115119.287494491@infradead.orgSigned-off-by: NIngo Molnar <mingo@kernel.org>

In preparation for find_insn_containing(), change insn_hash to use
sec_offset_hash().

This actually reduces runtime; probably because mixing in the section
index reduces the collisions due to text sections all starting their
instructions at offset 0.

Runtime on vmlinux.o from 3.1 to 2.5 seconds.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200416115119.227240432@infradead.orgSigned-off-by: NIngo Molnar <mingo@kernel.org>

When doing kbuild tests to see if the objtool changes affected those I
found that there was a measurable regression:

          pre		  post

  real    1m13.594        1m16.488s
  user    34m58.246s      35m23.947s
  sys     4m0.393s        4m27.312s

Perf showed that for small files the increased hash-table sizes were a
measurable difference. Since we already have -l "vmlinux" to
distinguish between the modes, make it also use a smaller portion of
the hash-tables.

This flips it into a small win:

  real    1m14.143s
  user    34m49.292s
  sys     3m44.746s
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200416115119.167588731@infradead.orgSigned-off-by: NIngo Molnar <mingo@kernel.org>

Validate that any call out of .noinstr.text is in between
instr_begin() and instr_end() annotations.

This annotation is useful to ensure correct behaviour wrt tracing
sensitive code like entry/exit and idle code. When we run code in a
sensitive context we want a guarantee no unknown code is ran.

Since this validation relies on knowing the section of call
destination symbols, we must run it on vmlinux.o instead of on
individual object files.

Add two options:

 -d/--duplicate "duplicate validation for vmlinux"
 -l/--vmlinux "vmlinux.o validation"

Where the latter auto-detects when objname ends with "vmlinux.o" and
the former will force all validations, also those already done on
!vmlinux object files.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200416115119.106268040@infradead.orgSigned-off-by: NIngo Molnar <mingo@kernel.org>

Objtool keeps per instruction CFI state in struct insn_state and will
save/restore this where required. However, insn_state has grown some
!CFI state, and this must not be saved/restored (that would
loose/destroy state).

Fix this by moving the CFI specific parts of insn_state into struct
cfi_state.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200416115119.045821071@infradead.orgSigned-off-by: NIngo Molnar <mingo@kernel.org>

There's going to be a new struct cfi_state, rename this one to make
place.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200416115118.986441913@infradead.orgSigned-off-by: NIngo Molnar <mingo@kernel.org>

The SAVE/RESTORE hints are now unused; remove them.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200416115118.926738768@infradead.orgSigned-off-by: NIngo Molnar <mingo@kernel.org>

'Optimize' ftrace_regs_caller. Instead of comparing against an
immediate, the more natural way to test for zero on x86 is: 'test
%r,%r'.

  48 83 f8 00             cmp    $0x0,%rax
  74 49                   je     226 <ftrace_regs_call+0xa3>

  48 85 c0                test   %rax,%rax
  74 49                   je     225 <ftrace_regs_call+0xa2>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200416115118.867411350@infradead.orgSigned-off-by: NIngo Molnar <mingo@kernel.org>

There's a convenient macro for 'SS+8' called FRAME_SIZE. Use it to
clarify things.

(entry/calling.h calls this SIZEOF_PTREGS but we're using
asm/ptrace-abi.h)
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200416115118.808485515@infradead.orgSigned-off-by: NIngo Molnar <mingo@kernel.org>

The ftrace_regs_caller() trampoline does something 'funny' when there
is a direct-caller present. In that case it stuffs the 'direct-caller'
address on the return stack and then exits the function. This then
results in 'returning' to the direct-caller with the exact registers
we came in with -- an indirect tail-call without using a register.

This however (rightfully) confuses objtool because the function shares
a few instruction in order to have a single exit path, but the stack
layout is different for them, depending through which path we came
there.

This is currently cludged by forcing the stack state to the non-direct
case, but this generates actively wrong (ORC) unwind information for
the direct case, leading to potential broken unwinds.

Fix this issue by fully separating the exit paths. This results in
having to poke a second RET into the trampoline copy, see
ftrace_regs_caller_ret.

This brings us to a second objtool problem, in order for it to
perceive the 'jmp ftrace_epilogue' as a function exit, it needs to be
recognised as a tail call. In order to make that happen,
ftrace_epilogue needs to be the start of an STT_FUNC, so re-arrange
code to make this so.

Finally, a third issue is that objtool requires functions to exit with
the same stack layout they started with, which is obviously violated
in the direct case, employ the new HINT_RET_OFFSET to tell objtool
this is an expected exception.

Together, this results in generating correct ORC unwind information
for the ftrace_regs_caller() function and it's trampoline copies.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200416115118.749606694@infradead.orgSigned-off-by: NIngo Molnar <mingo@kernel.org>

Normally objtool ensures a function keeps the stack layout invariant.
But there is a useful exception, it is possible to stuff the return
stack in order to 'inject' a 'call':

	push $fun
	ret

In this case the invariant mentioned above is violated.

Add an objtool HINT to annotate this and allow a function exit with a
modified stack frame.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200416115118.690601403@infradead.orgSigned-off-by: NIngo Molnar <mingo@kernel.org>

Teach objtool a little more about IRET so that we can avoid using the
SAVE/RESTORE annotation. In particular, make the weird corner case in
insn->restore go away.

The purpose of that corner case is to deal with the fact that
UNWIND_HINT_RESTORE lands on the instruction after IRET, but that
instruction can end up being outside the basic block, consider:

	if (cond)
		sync_core()
	foo();

Then the hint will land on foo(), and we'll encounter the restore
hint without ever having seen the save hint.

By teaching objtool about the arch specific exception frame size, and
assuming that any IRET in an STT_FUNC symbol is an exception frame
sized POP, we can remove the use of save/restore hints for this code.
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200416115118.631224674@infradead.orgSigned-off-by: NIngo Molnar <mingo@kernel.org>

Instruction sets can include more or less complex operations which might
not fit the currently defined set of stack_ops.

Combining more than one stack_op provides more flexibility to describe
the behaviour of an instruction. This also reduces the need to define
new stack_ops specific to a single instruction set.

Allow instruction decoders to generate multiple stack_op per
instruction.
Signed-off-by: NJulien Thierry <jthierry@redhat.com>
Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NMiroslav Benes <mbenes@suse.cz>
Reviewed-by: NAlexandre Chartre <alexandre.chartre@oracle.com>
Acked-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/20200327152847.15294-11-jthierry@redhat.comSigned-off-by: NIngo Molnar <mingo@kernel.org>

If the prefix of section name is not '.rodata', the following
function call can never return 0.

    strcmp(sec->name, C_JUMP_TABLE_SECTION)

So the name comparison is pointless, just remove it.
Signed-off-by: NMuchun Song <songmuchun@bytedance.com>
Signed-off-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: NIngo Molnar <mingo@kernel.org>

Compiling with Clang and CONFIG_KASAN=y was exposing a few warnings:

  call to memset() with UACCESS enabled

Document how to fix these for future travelers.

Link: https://github.com/ClangBuiltLinux/linux/issues/876Suggested-by: NKamalesh Babulal <kamalesh@linux.vnet.ibm.com>
Suggested-by: NMatt Helsley <mhelsley@vmware.com>
Suggested-by: NPeter Zijlstra <peterz@infradead.org>
Suggested-by: NRandy Dunlap <rdunlap@infradead.org>
Signed-off-by: NNick Desaulniers <ndesaulniers@google.com>
Signed-off-by: NJosh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: NIngo Molnar <mingo@kernel.org>