As proc/native is arch related, it should move some functions to arch-relate file. And this patch can help us to separate the architecture code, make code tidy. So that the merge of arm64 code later will not cause chaos.(#118)

the tracee may coredumped after detach: trace/breakpoint trap (core dumped).

Use the name specified by compile unit attribute DW_AT_go_package_name,
introduced in Go 1.13, to map package names to package paths, instead of
trying to deduce it from names of types.
Also use this mapping for resolving global variables and function
expressions.

We used to not do anything say anything at all about them anyway.

This prevents other users on the same machine (e.g. a production server) from
using dlv under the credentials of another user, which poses a security issue.

This allows to execute and debug signal handlers in inferior.

Move the loclist code out of proc and into its own package in
`pkg/dwarf`.

* Makefile: discard stderr of "go list"

In module mode "go" will print messages about downloading modules to
stderr, we shouldn't confuse them for the real command output.

* proc/core: enable PIE tests

PIE tests for core files were never enabled due to a missing flag.Parse
call.

* proc/linux: do not route signals to threads while stopping

While we are trying to stop the process we should not route signals
sent to threads because that will result in threads being resumed.
Also keep better track of which threads are stopped.

This fixes an incompatibility with Go 1.14, which sends a lot of
signals to its threads to implement non-cooperative preemption,
resulting in Delve hanging waiting for an already-stopped thread to
stop.

In principle however this bug has nothing to do with Go 1.14 and could
manifest in any instance of high signal pressure.

* Makefile: discard stderr of "go list"

In module mode "go" will print messages about downloading modules to
stderr, we shouldn't confuse them for the real command output.

Instead of just sending unhandled signals back to the process send them
to the specific thread that received them.
This is important because:

1. debugserver does not appear to support the vCont;CXX packet without
specifying a target thread
2. the non-cooperative preemption change in an upcoming version of Go
(1.15?) will require sending signals to a specific thread.

Fixes #1744

Changes CreateBreakpoint to create a logical breakpoint when multiple
addresses are specified, FindLocation and the api.Location type to
return logical locations and the cli to support logical breakpoints.

Modifies FindFileLocation, FindFunctionLocation and LineToPC as well as
service/debugger to support inlining and introduces the concept of
logical breakpoints.

For inlined functions FindFileLocation, FindFunctionLocation and
LineToPC will now return one PC address for each inlining and one PC
for the concrete implementation of the function (if present).

A proc.Breakpoint will continue to represent a physical breakpoint, at
a single memory location.

Breakpoints returned by service/debugger, however, will represent
logical breakpoints and may be associated with multiple memory
locations and, therefore, multiple proc.Breakpoints.

The necessary logic is introduced in service/debugger so that a change
to a logical breakpoint will be mirrored to all its physical
breakpoints and physical breakpoints are aggregated into a single
logical breakpoint when returned.

Maintain naming convention

program

When evaluating type casts always resolve array types.

Instead of resolving them by looking up the string in debug_info
construct a fake array type so that a type cast to an array type always
works as long as the element type exists.

We already did this for byte arrays, this commit extends this to any
array type. The reason is that we return a fake array type (that
doesn't exist in the target program) for the array of a channel type.

Fixes #1736

Fixes a case of breakpoint confusion on resume caused by having two
breakpoints one byte apart. This bug can cause the target program to
resume execution a single byte inside an instruction and crash either
with SIGILL or a SIGSEGV, or misbehave (depending on how the truncated
instruction is decoded).

native.(*Thread).StepInstruction should call FindBreakpoint using
adjustPC==false because at that point the PC of the thread should
already have been adjusted (and it has been).

This script re-runs a program until it fails or hits a breakpoint, it
is useful for debugging non-deterministic problems when used in
conjunction with rr.

Make the 'list' command succeed for file:line expressions that don't
map to any instruction.
Adds an argument to the FindLocations API call that makes FindLocations
return if the expression can be parsed, even if it doesn't end up
matching any instruction in debug_line.

Fixes #1713

Thanks to everyone who contributed!

Adds a '-r' option to the 'restart' command (and to the Restart API)
that re-records the target when using rr.

Also moves the code to delete the trace directory inside the gdbserial
package.

Updates #1711

Avoid crashing with nil pointer dereference, signal error instead.

Fixes #1720

A compile unit can produce a debug_line program consisting of multiple
sequences according to the DWARF standard. The standard guarantees that
addresses monotonically increment within a single sequence but
different sequences may not follow this rule.

This commit changes dwarf/line (in particular PCToLine and
AllPCsBetween) to support debug_line sections containing units with
multiple sequences.

TestPCToLine needs to be changed so that it picks valid addresses (i.e.
addresses covered by a sequence) as values for basePC, instead of just
rounding.

Fixes #1694

Splits the code that loads function information from debug_info into
multiple functions.
This makes the changes needed to implement logical breakpoints easier
to make.

* Fix starlark map iteration for maps > 64 entries

* Fix TestMapEvaluation

The origin error message is confusing, so add some details.

Fixes #1700

The Disassemble API call should work when there is no selected
goroutine (for example because the program was just started).

Fixes #1703

It was failing silently.

`func (v *Variable) findMethod` should support for searching methods of an
embedded filed.

Fixes #1688

Trust argument order to determine argument frame layout when calling
functions, this allows calling optimized functions and removes the
special cases for runtime.mallocgc.

Fixes #1589

Add options to start a stacktrace from the values saved in the
runtime.g struct as well as a way to disable the stackSwitch logic and
just get a normal stacktrace.

If a closure captures a variable but also defines a variable of the
same name in its root scope the shadowed flag would, sometimes, not be
appropriately applied to the captured variable.

This change:

1. sorts the variable list by depth *and* declaration line, so that
closure captured variables always appear before other root-scope
variables, regardless of the order used by the compiler

2. marks variable with the same name as shadowed even if there is only
one scope at play.

This fixes the problem but as a side effect:

1. programs compiled with Go prior to version 1.9 will have the
shadowed flag applied arbitrarily (previously the shadowed flag was not
applied at all)
2. programs compiled with Go prior to versoin 1.11 will still exhibit
the bug, as they do not have DeclLine information.

Fixes #1672

Bump continuous integration to include Go 1.13, drop 1.10 from
compatiblity file.

This change adds the `make uninstall` command which will uninstall delve.

According to the description of "CIE: length, CIE_id, version, augmentation"
in Page 122 of  http://dwarfstd.org/doc/Dwarf3.pdf ,
`augmentation` should exclude `version`

The fix for #1428 was buggy, partly because I communicated poorly. Sorry
about that.

The size of the TLS segment should be padded such that TLS addresses
are congruent in the file to where they will end up memory, i.e.
(tlsoffset%align) == (vaddr%align). In most cases, vaddr will be aligned
and it won't matter, but if not then simply aligning the end of the
segment is incorrect. This should be right.

(For the record, the current rounding logic is working in bits, but
PtrSize is in bytes, so it wasn't working as originally intended
either.)

* all: Bump to v1.3.0

Add new version to CHANGELOG and update internal version.

Thank you @Ladicle, @qaisjp, @justinclift, @tschundler, @two,
@dpapastamos, @qingyunha, @rayrapetyan, @briandealwis and @msaf1980,
@jeremyfaller, @stmuk, @dr2chase, @pjot726.

* all: Add date to changelog