[PATCH] fbsd-nat: Pacify gcc with no functional changes
Enze Li
enze.li@hotmail.com
Mon Sep 11 14:20:35 GMT 2023
On Sat, Sep 09 2023 at 09:49:40 AM -0700, John Baldwin wrote:
> On 9/9/23 6:50 AM, Enze Li wrote:
>> I see these errors on FreeBSD/aarch64 when using gcc 12 without passing
>> --disable-werror.
>> =====================================================================
>> CXX fbsd-nat.o
>> fbsd-nat.c: In member function 'void fbsd_nat_target::resume_one_process(ptid_t, int, gdb_signal)':
>> fbsd-nat.c:1208:11: error: unused variable 'request' [-Werror=unused-variable]
>> 1208 | int request;
>> | ^~~~~~~
>> fbsd-nat.c: In member function 'virtual ptid_t
>> fbsd_nat_target::wait(ptid_t, target_waitstatus*,
>> target_wait_flags)':
>> fbsd-nat.c:1726:22: error: declaration of 'inf' shadows a previous
>> local [-Werror=shadow=compatible-local]
>> 1726 | for (inferior *inf : all_non_exited_inferiors (this))
>> | ^~~
>> fbsd-nat.c:1697:17: note: shadowed declaration is here
>> 1697 | inferior *inf = find_inferior_ptid (this, wptid);
>> | ^~~
>> fbsd-nat.c: In member function 'virtual void
>> fbsd_nat_target::detach(inferior*, int)':
>> fbsd-nat.c:2044:18: error: variable 'wptid' set but not used
>> [-Werror=unused-but-set-variable]
>> 2044 | ptid_t wptid = wait_1 (ptid, &ws, 0);
>> | ^~~~~
>> cc1plus: all warnings being treated as errors
>> =====================================================================
>> This patch includes the following non-functional changes,
>> 1. Remove unused variable "request".
>> 2. Rename inf to inf_p to avoid shadowed declaration warnings.
>> 3. Set wptid to null_ptid when USE_SIGTRAP_SIGINFO is not defined.
Hi John,
Thanks for your review.
>
> I'm surprised you see this third warning. USE_SIGTRAP_SIGINFO should
> be on for all modern versions of FreeBSD. In particular, fbsd-nat.h
> enables it on 11.3 and later. What version of FreeBSD are you
> building on?
I just realized that the third description is incorrect and should be:
"3. Set wptid to null_ptid when USE_SIGTRAP_SIGINFO is defined."
BTW, I'm using FreeBSD/aarch64 13.2-RELEASE.
>
> From fbsd-nat.h:
>
> /* FreeBSD kernels 11.3 and later report valid si_code values for
> SIGTRAP on all architectures. Older FreeBSD kernels that supported
> TRAP_BRKPT did not report valid values for MIPS and sparc64. Even
> older kernels without TRAP_BRKPT support did not report valid
> values on any architecture. */
> #if (__FreeBSD_kernel_version >= 1102502) || (__FreeBSD_version >= 1102502)
> # define USE_SIGTRAP_SIGINFO
>
>
>> Tested on FreeBSD/aarch64 by rebuilding.
>> ---
>> gdb/fbsd-nat.c | 11 ++++++-----
>> 1 file changed, 6 insertions(+), 5 deletions(-)
>> diff --git a/gdb/fbsd-nat.c b/gdb/fbsd-nat.c
>> index 0ee3bccb5ef..af416557504 100644
>> --- a/gdb/fbsd-nat.c
>> +++ b/gdb/fbsd-nat.c
>> @@ -1205,8 +1205,6 @@ fbsd_nat_target::resume_one_process (ptid_t ptid, int step,
>> for (thread_info *tp : inf->non_exited_threads ())
>> {
>> - int request;
>> -
>> /* If ptid is a specific LWP, suspend all other LWPs in the
>> process, otherwise resume all LWPs in the process.. */
>> if (!ptid.lwp_p() || tp->ptid.lwp () == ptid.lwp ())
>
> Thanks. This a leftover from my recent patches, not sure why clang
> didn't notice.
>
>> @@ -1694,9 +1692,9 @@ fbsd_nat_target::wait (ptid_t ptid, struct target_waitstatus *ourstatus,
>> || ourstatus->kind () == TARGET_WAITKIND_NO_RESUMED)
>> break;
>> - inferior *inf = find_inferior_ptid (this, wptid);
>> - gdb_assert (inf != nullptr);
>> - fbsd_inferior *fbsd_inf = get_fbsd_inferior (inf);
>> + inferior *inf_p = find_inferior_ptid (this, wptid);
>> + gdb_assert (inf_p != nullptr);
>> + fbsd_inferior *fbsd_inf = get_fbsd_inferior (inf_p);
>> gdb_assert (fbsd_inf != nullptr);
>> gdb_assert (fbsd_inf->resumed_lwps != null_ptid);
>> gdb_assert (fbsd_inf->running_lwps > 0);
>
> I might call this variable `winf` to match the `wptid` vs `ptid` pattern.
>
Fixed.
>> @@ -2094,6 +2092,9 @@ fbsd_nat_target::detach (inferior *inf, int from_tty)
>> }
>> }
>> }
>> +#else
>> + /* pacify gcc */
>> + wptid = null_ptid;
>
> Probably the better way to pacify this is to use a (void) cast to mark it
> as used, e.g. `(void)null_ptid;`
>
Okey, fixed. I'll send v2 to the list soon.
Best Regards,
Enze
>> #endif
>> sig = 0;
>> break;
>> base-commit: 38a984fa440c7686c741b7804eae06a528849aa7
More information about the Gdb-patches
mailing list