Also add a KINST_TRAMP_FILL_SIZE constant to correct the step in
KINST_TRAMP_INIT.

Add a KINST_TRAMP_FILL constant and use this instead of KINST_PATCHVAL. In
the RISC-V port we want to use the compressed patch value, instead of the
regular one, so KINST_PATCHVAL doesn't work in this case.

Add /* NOTREACHED */ at the end of dtrace_getreg().

Bump version to 1.13.

Add FP.

Forgot riscv_regs.d

Thanks for working on this!

Get rid of amd64-specific calculation comment KINST_TRAMPS_PER_CHUNK.

Provide full diff.

Merge and abandon D39490.

Add RISC-V exception handlers.

Compile instr_size.c into libdtrace and dtrace, move definition below KERNEL ifdef.

Provide full diff.

Use dtrace_instr_size() and get rid of machine-dependent code (depends on
D39489).

Surround amd64-specific code in an ifdef.

Exclude cpu_switch().

Surround amd64 exception handlers in an ifdef.

Fix accidental dangling brace.

Implement dt_sugar_elf_verify_debuglink() to make sure the debug files we're
reading are up to date, and if they're not, skip them. Also stop leaking
elf_info structs in dt_sugar_do_kinst_inline().

Abandoning this one since we no longer search for push %rbp, instead we
exclude just the exception handlers (see D39229).

No longer search for push %rbp, just exclude the exception handlers.
This way we can trace safe functions that do not push %rbp, such as
cpu_switch().

In D38892#893746, @christos wrote:

Since D38825 doesn't need this patch anymore, is this needed?

Since D38825 doesn't need this patch anymore, is this needed?

Add tests. kinst_test_inline() needs modification so that it can be splitted
into DW_AT_ranges.

Do not exclude inline copies that have DW_AT_subprogram instead of
DW_AT_inlined_subroutine.

More bug fixes. I always keep finding them after I submit the patch.

Fix flag-check on line 779

Fix per-module parsing. Create an additional FBT probe if a non-inline definition is
found along an inline definition in the same module.

Fix bugs, surround amd64-specific code in an #ifdef, merge flags into
dtsp_flags, handle dt_sugar_elf_init() and dwarf_elf_init() errors, use
longjmp() for dt_alloc() failures.

Parse all loaded modules and their debug files instead of just kernel.
Mention new bug in summary.

Apparently, this patch doesn't really work and, in fact, allowing a predicate
with no acts is probably impossible in D without introducing a significant
amount of complexity. awk can do that just fine because awk's predicates cannot
contain unescaped "/", since it searches for regexes. D on the other hand
allows normal computation inside the predicate, meaning one can do division
inside it:

Put it in its own function.

I see those warnings when building libdtrace without the patch.

I don't think there are any pending changes for this PR, it can get through.

Mention leaf function with rbp omitted.

Get rid of push %rbp check (depends on D39166).

Improve example in man page.

Do not always go back one instruction.

Handle multiple dn_pdescs properly, move handling away from
dt_sugar_visit_all().

Got rid of deep copying, appending to the main clause is simpler.

Handle leaf functions with their push %rbp ommitted.

remove if/else comment in dt_cc as dt_sugar now also performs transformations
on kinst probes.

remove call to dwarf_next_types_section()

update dtrace_kinst.4 man page to mention inline tracing

fix dtrace_disx86() fail in dt_sugar_kinst_find_caller_func()

use dtp->bootfile in dt_sugar_do_kinst_inline() instead of hardcoding the path