From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 122336 invoked by alias); 3 Jan 2017 21:31:34 -0000 Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org Received: (qmail 122316 invoked by uid 89); 3 Jan 2017 21:31:33 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-5.1 required=5.0 tests=BAYES_00,RP_MATCHES_RCVD,SPF_HELO_PASS autolearn=ham version=3.3.2 spammy=resumed, inferior_ptid, target_ops, from_tty X-HELO: mx1.redhat.com Received: from mx1.redhat.com (HELO mx1.redhat.com) (209.132.183.28) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with ESMTP; Tue, 03 Jan 2017 21:31:22 +0000 Received: from int-mx10.intmail.prod.int.phx2.redhat.com (int-mx10.intmail.prod.int.phx2.redhat.com [10.5.11.23]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id B83704D687; Tue, 3 Jan 2017 21:31:22 +0000 (UTC) Received: from localhost (unused-10-15-17-193.yyz.redhat.com [10.15.17.193]) by int-mx10.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id v03LVLk2029034 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=NO); Tue, 3 Jan 2017 16:31:22 -0500 From: Sergio Durigan Junior To: Luis Machado Cc: GDB Patches , Subject: Re: [PATCH 4/6] Share parts of gdb/gdbthread.h with gdbserver References: <1482464361-4068-1-git-send-email-sergiodj@redhat.com> <1482464361-4068-5-git-send-email-sergiodj@redhat.com> X-URL: http://blog.sergiodj.net Date: Tue, 03 Jan 2017 21:31:00 -0000 In-Reply-To: (Luis Machado's message of "Mon, 26 Dec 2016 15:34:43 -0600") Message-ID: <87d1g3q193.fsf@redhat.com> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/25.1 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain X-IsSubscribed: yes X-SW-Source: 2017-01/txt/msg00032.txt.bz2 Thanks for the review. Comments below. On Monday, December 26 2016, Luis Machado wrote: > On 12/22/2016 09:39 PM, Sergio Durigan Junior wrote: >> Again, it was necessary to share a few functions declared on >> gdb/gdbthread.h with gdbserver, because they are needed by >> fork_inferior. I decided to implement them on >> gdb/gdbserver/inferiors.c because that's where the thread functions >> are also implemented on gdbserver. As a side note, due to the way >> gdbserver marks a thread as executing, a new argument needed to be >> added on set_executing. This argument is a 'struct target_waitstatus >> *', and is not necessary on GDB, so it was marked with >> ATTRIBUTE_UNUSED accordingly. >> >> gdb/ChangeLog: >> 2016-12-22 Sergio Durigan Junior >> >> * Makefile.in (HFILES_NO_SRCDIR): Add "common/common-gdbthread.h". >> * common/common-gdbthread.h: New file, with parts from >> "gdb/gdbthread.h". >> * fork-child.c (fork_inferior): Update call of "set_executing". >> * gdbthread.h: Include "common-gdbthread.h". >> (init_thread_list): Moved to "common/common-gdbthread.h". >> (add_thread_silent): Likewise. >> (switch_to_thread): Likewise. >> (set_executing): Likewise. Added an extra argument, "struct >> target_waitstatus *". >> * infrun.c (handle_inferior_event_1): Update call of >> "set_executing". >> * linux-nat.c (attach_proc_task_lwp_callback): Likewise. >> (linux_handle_extended_wait): Likewise. >> * record-btrace.c (get_thread_current_frame): Likewise. >> * record-full.c (record_full_wait_1): Likewise. >> * remote.c (remote_add_thread): Likewise. >> (process_initial_stop_replies): Likewise. >> * solib-spu.c (spu_skip_standalone_loader): Likewise. >> * target.c (target_resume): Likewise. >> * thread.c (set_executing): Update function declaration to add >> third argument. Mark it as unused. >> >> gdb/gdbserver/ChangeLog: >> 2016-12-22 Sergio Durigan Junior >> >> * inferiors.c (init_thread_list): New function. >> (switch_to_thread): Likewise. >> (set_executing): Likewise. >> (add_thread_silent): Likewise. >> --- >> gdb/Makefile.in | 1 + >> gdb/common/common-gdbthread.h | 50 +++++++++++++++++++++++++++++++++++++++++++ >> gdb/fork-child.c | 2 +- >> gdb/gdbserver/inferiors.c | 42 ++++++++++++++++++++++++++++++++++++ >> gdb/gdbthread.h | 20 +---------------- >> gdb/infrun.c | 2 +- >> gdb/linux-nat.c | 4 ++-- >> gdb/record-btrace.c | 6 +++--- >> gdb/record-full.c | 4 ++-- >> gdb/remote.c | 4 ++-- >> gdb/solib-spu.c | 2 +- >> gdb/target.c | 2 +- >> gdb/thread.c | 5 ++++- >> 13 files changed, 111 insertions(+), 33 deletions(-) >> create mode 100644 gdb/common/common-gdbthread.h >> >> diff --git a/gdb/Makefile.in b/gdb/Makefile.in >> index 8372b4a..ca13a80 100644 >> --- a/gdb/Makefile.in >> +++ b/gdb/Makefile.in >> @@ -1467,6 +1467,7 @@ HFILES_NO_SRCDIR = \ >> common/common-debug.h \ >> common/common-defs.h \ >> common/common-exceptions.h \ >> + common/common-gdbthread.h \ >> common/common-regcache.h \ >> common/common-types.h \ >> common/common-utils.h \ >> diff --git a/gdb/common/common-gdbthread.h b/gdb/common/common-gdbthread.h >> new file mode 100644 >> index 0000000..3560892 >> --- /dev/null >> +++ b/gdb/common/common-gdbthread.h >> @@ -0,0 +1,50 @@ >> +/* Multi-process/thread control defs for GDB, the GNU debugger. >> + Copyright (C) 1987-2016 Free Software Foundation, Inc. >> + Contributed by Lynx Real-Time Systems, Inc. Los Gatos, CA. >> + >> + >> + This file is part of GDB. >> + >> + This program is free software; you can redistribute it and/or modify >> + it under the terms of the GNU General Public License as published by >> + the Free Software Foundation; either version 3 of the License, or >> + (at your option) any later version. >> + >> + This program is distributed in the hope that it will be useful, >> + but WITHOUT ANY WARRANTY; without even the implied warranty of >> + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the >> + GNU General Public License for more details. >> + >> + You should have received a copy of the GNU General Public License >> + along with this program. If not, see . */ >> + >> +#ifndef COMMON_THREAD_H >> +#define COMMON_THREAD_H >> + >> +struct target_waitstatus; >> + >> +/* Create an empty thread list, or empty the existing one. */ >> + > > Spurious newline, multiple cases of this throughout. Yeah, now you're right: https://sourceware.org/gdb/wiki/Internals%20GDB-C-Coding-Standards#Empty_line_between_subprogram_description_and_the_subprogram_implementation "Note that this only applies to the case where the comment is placed besides the subprogram implementation (typically in a .c file). In the case of the documentation being placed next to the subprogram declaration, then the comment should be placed immediately before the declaration." I'll fix these. Thanks. >> +extern void init_thread_list (void); >> + >> +/* Switch from one thread to another. */ >> + >> +extern void switch_to_thread (ptid_t ptid); >> + >> +/* Marks thread PTID as executing, or not. If PTID is minus_one_ptid, >> + marks all threads. >> + >> + Note that this is different from the running state. See the >> + description of state and executing fields of struct >> + thread_info. */ >> + >> +extern void set_executing (ptid_t ptid, int executing, >> + struct target_waitstatus *ws); >> + >> +/* Add a thread to the thread list and return the pointer to the new >> + thread. Caller may use this pointer to initialize the private >> + thread data. */ >> + >> +extern struct thread_info *add_thread_silent (ptid_t ptid); >> + >> +#endif /* ! COMMON_THREAD_H */ >> diff --git a/gdb/fork-child.c b/gdb/fork-child.c >> index 15f8249..38fca60 100644 >> --- a/gdb/fork-child.c >> +++ b/gdb/fork-child.c >> @@ -545,7 +545,7 @@ startup_inferior (int ntraps) >> } >> >> /* Mark all threads non-executing. */ >> - set_executing (resume_ptid, 0); >> + set_executing (resume_ptid, 0, NULL); >> } >> >> /* Implement the "unset exec-wrapper" command. */ >> diff --git a/gdb/gdbserver/inferiors.c b/gdb/gdbserver/inferiors.c >> index 574a7ba..6b981d0 100644 >> --- a/gdb/gdbserver/inferiors.c >> +++ b/gdb/gdbserver/inferiors.c >> @@ -468,3 +468,45 @@ make_cleanup_restore_current_thread (void) >> { >> return make_cleanup (do_restore_current_thread_cleanup, current_thread); >> } >> + >> +/* See common/common-gdbthread.h. */ >> + >> +void >> +init_thread_list (void) >> +{ >> + /* To be implemented. */ >> +} >> + >> +/* See common/common-gdbthread.h. */ >> + >> +void >> +switch_to_thread (ptid_t ptid) >> +{ >> + if (!ptid_equal (ptid, minus_one_ptid)) >> + current_thread = find_thread_ptid (ptid); >> +} >> + >> +/* See common/common-gdbthread.h. */ >> + >> +void >> +set_executing (ptid_t ptid ATTRIBUTE_UNUSED, int executing ATTRIBUTE_UNUSED, >> + struct target_waitstatus *ws) >> +{ >> + gdb_assert (current_thread != NULL); >> + current_thread->last_resume_kind = resume_stop; >> + current_thread->last_status = *ws; > > Is there a chance ws will be NULL and cause a crash here? Should we assert? Yeah, I was also thinking about that, but currently there is no chance of it being NULL. The only place that calls this function for gdbserver is fork_inferior, and if the function is called without a current_thread then something really wrong happened. >> +} >> + >> +/* See common/common-gdbthread.h. */ >> + >> +struct thread_info * >> +add_thread_silent (ptid_t ptid) >> +{ >> + pid_t pid = ptid_get_pid (ptid); >> + >> + /* Check if there is a process already. */ >> + if (find_process_pid (pid) == NULL) >> + add_process (pid, 0); >> + >> + return add_thread (ptid_build (pid, pid, 0), NULL); >> +} >> diff --git a/gdb/gdbthread.h b/gdb/gdbthread.h >> index 8f37fbb..d6fdc42 100644 >> --- a/gdb/gdbthread.h >> +++ b/gdb/gdbthread.h >> @@ -31,6 +31,7 @@ struct symtab; >> #include "common/vec.h" >> #include "target/waitstatus.h" >> #include "cli/cli-utils.h" >> +#include "common-gdbthread.h" >> >> /* Frontend view of the thread state. Possible extensions: stepping, >> finishing, until(ling),... */ >> @@ -344,19 +345,12 @@ struct thread_info >> struct thread_info *step_over_next; >> }; >> >> -/* Create an empty thread list, or empty the existing one. */ >> -extern void init_thread_list (void); >> - >> /* Add a thread to the thread list, print a message >> that a new thread is found, and return the pointer to >> the new thread. Caller my use this pointer to >> initialize the private thread data. */ >> extern struct thread_info *add_thread (ptid_t ptid); >> >> -/* Same as add_thread, but does not print a message >> - about new thread. */ >> -extern struct thread_info *add_thread_silent (ptid_t ptid); >> - >> /* Same as add_thread, and sets the private info. */ >> extern struct thread_info *add_thread_with_info (ptid_t ptid, >> struct private_thread_info *); >> @@ -469,10 +463,6 @@ extern struct thread_info *iterate_over_threads (thread_callback_func, void *); >> >> extern int thread_count (void); >> >> -/* Switch from one thread to another. Also sets the STOP_PC >> - global. */ >> -extern void switch_to_thread (ptid_t ptid); >> - >> /* Switch from one thread to another. Does not read registers and >> sets STOP_PC to -1. */ >> extern void switch_to_thread_no_regs (struct thread_info *thread); >> @@ -518,14 +508,6 @@ extern int is_exited (ptid_t ptid); >> /* In the frontend's perpective, is this thread stopped? */ >> extern int is_stopped (ptid_t ptid); >> >> -/* Marks thread PTID as executing, or not. If PTID is minus_one_ptid, >> - marks all threads. >> - >> - Note that this is different from the running state. See the >> - description of state and executing fields of struct >> - thread_info. */ >> -extern void set_executing (ptid_t ptid, int executing); >> - >> /* Reports if thread PTID is executing. */ >> extern int is_executing (ptid_t ptid); >> >> diff --git a/gdb/infrun.c b/gdb/infrun.c >> index bf0632e..f5cc965 100644 >> --- a/gdb/infrun.c >> +++ b/gdb/infrun.c >> @@ -4947,7 +4947,7 @@ handle_inferior_event_1 (struct execution_control_state *ecs) >> else >> mark_ptid = ecs->ptid; >> >> - set_executing (mark_ptid, 0); >> + set_executing (mark_ptid, 0, NULL); >> >> /* Likewise the resumed flag. */ >> set_resumed (mark_ptid, 0); >> diff --git a/gdb/linux-nat.c b/gdb/linux-nat.c >> index cbf94ed..9db620c 100644 >> --- a/gdb/linux-nat.c >> +++ b/gdb/linux-nat.c >> @@ -1190,7 +1190,7 @@ attach_proc_task_lwp_callback (ptid_t ptid) >> raw clone). */ >> add_thread (lp->ptid); >> set_running (lp->ptid, 1); >> - set_executing (lp->ptid, 1); >> + set_executing (lp->ptid, 1, NULL); >> } >> >> return 1; >> @@ -2083,7 +2083,7 @@ linux_handle_extended_wait (struct lwp_info *lp, int status) >> and the user/frontend, this new thread is running until >> it next reports a stop. */ >> set_running (new_lp->ptid, 1); >> - set_executing (new_lp->ptid, 1); >> + set_executing (new_lp->ptid, 1, NULL); >> >> if (WSTOPSIG (status) != SIGSTOP) >> { >> diff --git a/gdb/record-btrace.c b/gdb/record-btrace.c >> index 7c0e39f..80cf274 100644 >> --- a/gdb/record-btrace.c >> +++ b/gdb/record-btrace.c >> @@ -1928,7 +1928,7 @@ get_thread_current_frame (struct thread_info *tp) >> move the thread. Since we need to recompute the stack, we temporarily >> set EXECUTING to flase. */ >> executing = is_executing (inferior_ptid); >> - set_executing (inferior_ptid, 0); >> + set_executing (inferior_ptid, 0, NULL); >> >> frame = NULL; >> TRY >> @@ -1938,7 +1938,7 @@ get_thread_current_frame (struct thread_info *tp) >> CATCH (except, RETURN_MASK_ALL) >> { >> /* Restore the previous execution state. */ >> - set_executing (inferior_ptid, executing); >> + set_executing (inferior_ptid, executing, NULL); >> >> /* Restore the previous inferior_ptid. */ >> inferior_ptid = old_inferior_ptid; >> @@ -1948,7 +1948,7 @@ get_thread_current_frame (struct thread_info *tp) >> END_CATCH >> >> /* Restore the previous execution state. */ >> - set_executing (inferior_ptid, executing); >> + set_executing (inferior_ptid, executing, NULL); >> >> /* Restore the previous inferior_ptid. */ >> inferior_ptid = old_inferior_ptid; >> diff --git a/gdb/record-full.c b/gdb/record-full.c >> index 5608e70..3ad9a08 100644 >> --- a/gdb/record-full.c >> +++ b/gdb/record-full.c >> @@ -1156,12 +1156,12 @@ record_full_wait_1 (struct target_ops *ops, >> { >> /* Try to insert the software single step breakpoint. >> If insert success, set step to 0. */ >> - set_executing (inferior_ptid, 0); >> + set_executing (inferior_ptid, 0, NULL); >> reinit_frame_cache (); >> >> step = !insert_single_step_breakpoints (gdbarch); >> >> - set_executing (inferior_ptid, 1); >> + set_executing (inferior_ptid, 1, NULL); >> } >> >> if (record_debug) >> diff --git a/gdb/remote.c b/gdb/remote.c >> index ef6c54e..0b5f837 100644 >> --- a/gdb/remote.c >> +++ b/gdb/remote.c >> @@ -1847,7 +1847,7 @@ remote_add_thread (ptid_t ptid, int running, int executing) >> thread = add_thread (ptid); >> >> get_private_info_thread (thread)->vcont_resumed = executing; >> - set_executing (ptid, executing); >> + set_executing (ptid, executing, NULL); >> set_running (ptid, running); >> } >> >> @@ -3949,7 +3949,7 @@ process_initial_stop_replies (int from_tty) >> || ws.value.sig != GDB_SIGNAL_0) >> thread->suspend.waitstatus_pending_p = 1; >> >> - set_executing (event_ptid, 0); >> + set_executing (event_ptid, 0, NULL); >> set_running (event_ptid, 0); >> thread->priv->vcont_resumed = 0; >> } >> diff --git a/gdb/solib-spu.c b/gdb/solib-spu.c >> index fa2977e..f55e0a0 100644 >> --- a/gdb/solib-spu.c >> +++ b/gdb/solib-spu.c >> @@ -88,7 +88,7 @@ spu_skip_standalone_loader (void) >> >> target_resume (inferior_ptid, 1, GDB_SIGNAL_0); >> target_wait (minus_one_ptid, &ws, 0); >> - set_executing (minus_one_ptid, 0); >> + set_executing (minus_one_ptid, 0, NULL); >> >> inferior_thread ()->control.in_infcall = 0; >> } >> diff --git a/gdb/target.c b/gdb/target.c >> index 246d292..e7c68bf 100644 >> --- a/gdb/target.c >> +++ b/gdb/target.c >> @@ -2325,7 +2325,7 @@ target_resume (ptid_t ptid, int step, enum gdb_signal signal) >> registers_changed_ptid (ptid); >> /* We only set the internal executing state here. The user/frontend >> running state is set at a higher level. */ >> - set_executing (ptid, 1); >> + set_executing (ptid, 1, NULL); >> clear_inline_frame_state (ptid); >> } >> >> diff --git a/gdb/thread.c b/gdb/thread.c >> index e5d6c5f..1114aac 100644 >> --- a/gdb/thread.c >> +++ b/gdb/thread.c >> @@ -1011,8 +1011,11 @@ is_executing (ptid_t ptid) >> return tp->executing; >> } >> >> +/* See common/common-gdbthread.h. */ >> + >> void >> -set_executing (ptid_t ptid, int executing) >> +set_executing (ptid_t ptid, int executing, >> + struct target_waitstatus *ws ATTRIBUTE_UNUSED) > > I'm thinking about this change. Isn't there a better way to do this > without adding gdb/gdbserver-specific parameters and then passing > dummy values for the unused parameters depending on who's calling? > > Maybe making gdbserver accept the same parameters as gdb and then do > some extra processing in gdbserver to get at the waitstatus data if it > still needs that? Or maybe unifying the function and making gdb and > gdbserver operate the same way in this regard. Hm. Re-reading the code now, I've realized there is a way to make things work. startup_inferior already sets the global "last_status" (from gdbserver/server.c) when it's executing. Therefore, it is possible to access this variable (maybe through a getter on server.c), and there is no need to pass it to set_executing. Of course, that'd mean relying on a global value to set current_thread->last_status, but that doesn't really change what is already being done today. With this modification, there is no need to extend set_executing and make it receive a target_waitstatus. However, for gdbserver, both the PTID and the EXECUTING arguments will still be useless. Thanks, -- Sergio GPG key ID: 237A 54B1 0287 28BF 00EF 31F4 D0EB 7628 65FC 5E36 Please send encrypted e-mail if possible http://sergiodj.net/