From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 29002 invoked by alias); 14 Nov 2008 15:42:34 -0000 Received: (qmail 28962 invoked by uid 22791); 14 Nov 2008 15:42:33 -0000 X-Spam-Check-By: sourceware.org Received: from ti-out-0910.google.com (HELO ti-out-0910.google.com) (209.85.142.190) by sourceware.org (qpsmtpd/0.31) with ESMTP; Fri, 14 Nov 2008 15:41:57 +0000 Received: by ti-out-0910.google.com with SMTP id d10so1138005tib.12 for ; Fri, 14 Nov 2008 07:40:22 -0800 (PST) Received: by 10.110.47.17 with SMTP id u17mr1265231tiu.43.1226677222613; Fri, 14 Nov 2008 07:40:22 -0800 (PST) Received: by 10.110.103.3 with HTTP; Fri, 14 Nov 2008 07:40:22 -0800 (PST) Message-ID: Date: Fri, 14 Nov 2008 17:29:00 -0000 From: teawater To: "Eli Zaretskii" Subject: Re: [RFA] Process record and replay, 3/10 Cc: gdb-patches@sourceware.org In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: X-IsSubscribed: yes 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 X-SW-Source: 2008-11/txt/msg00343.txt.bz2 Thanks Eli. On Fri, Nov 14, 2008 at 19:58, Eli Zaretskii wrote: >> Date: Tue, 11 Nov 2008 10:59:21 +0800 >> From: teawater >> Cc: gdb-patches@sourceware.org >> >> >> + q = yquery (_("The record instruction number (record-insn-number) is equal to record-insn-number-max. Do you want to close record/replay stop when record/replay buffer becomes full(record-stop-at-limit) then auto delete first record_t?")); >> > >> > There's something wrong with this query. First, why both "close" and >> > "stop"? Also, what is "record_t"? a typo? >> >> What about change it to "Do you want to auto delete first execute log >> entry when record/replay buffer becomes full(record-stop-at-limit)?" > > I suggest > > "Do you want to auto delete previous execute log entries > when record/replay buffer becomes full (record-stop-at-limit)?" I will use it. > >> >> + error (_("Process record: record stop the program.")); >> > >> > Do you mean >> > >> > Process record: program recording stopped. >> > >> > ? >> >> What about "Process record stop inferior." > > I'm not sure. Can you explain what exactly does this message try to > say? I will then suggest a better wording. OK. When P record want make inferior stop. It will use this code. > >> >> + if (ret < 0) >> >> + error (_("Process record record message error.")); >> > >> > Do you mean something like "Process record error."? That is, does >> > this happen when some error is encountered inside >> > gdbarch_process_record? >> > >> >> What about "Process record execute log error"? > > "Error executing process log record"? What about "Process record target record execute log failed"? > >> >> + if (target_read_memory >> >> + (record_list->u.mem.addr, mem, record_list->u.mem.len)) >> >> + { >> >> + error (_("Process record: read memory addr = 0x%s len = %d error."), >> > >> > Here also, the error message should be more clearly phrased. >> >> Could you tell me how to output this message clear? > > "Error reading memory at addr = 0x%s, len = %d OK. I will use it. > > By the way, is the address really a string (%s) here? Yes, cause there is paddr_nz (addr). > >> >> + if (record_arch_list_add_reg (i)) >> >> + { >> >> + record_list_release (record_arch_list_tail); >> >> + error (_("Process record: record message error.")); >> > >> > Same here. (There are more like this one.) >> >> Change it to "record execute log"? > > Yes, that's better. OK. I will use it. > >> >> + nquery (_ >> >> + ("Becuse GDB is in replay mode, changing the value of a register will destroy the record from this point forward. Change all register?")); >> > >> > "all registers", in plural. >> > >> > Also, I'm not sure I would understand what you mean by ``destroy the >> > record''? Are you saying that process recording will effectively stop >> > working from this point onward? >> >> What about change it to "destory the execute log"? > > I think you mean "make the execute log unusable from this point onward". OK. I will use it. > >> >> +static void >> >> +set_record_insn_max_num (char *args, int from_tty, struct cmd_list_element *c) >> >> +{ >> >> + if (record_insn_num > record_insn_max_num && record_insn_max_num) >> >> + { >> >> + printf_unfiltered (_("Process record: record instructions number is bigger than record instructions max number. Auto delete the first ones.\n")); >> > >> > And here. Also, we need a question mark at the end of the second >> > sentence. >> >> I think maybe we need let user know when delete some log entries. > > Yes, but I didn't mean to remove the question, only the "Process > record:" prefix. I see. I will do it. > >> >> + add_com ("delrecord", class_obscure, cmd_record_delete, >> >> + _("When process record target running in replay mode, delete the next running messages and begin to record the running message at current address.")); >> > >> > This doc string should be made shorter, to fit a single terminal line, >> > and the first line should not include comma characters, because >> > otherwise some help commands will not display anything beyond the >> > first comma. >> > >> >> What about this: >> _("When process record target running in replay mode,\n\ >> delete the next running messages and begin to record\n\ >> the execute log at current address.") > > The first line must be a complete sentence, because some help commands > (e.g., `apropos') show only the first line. > > I suggest > > _("Delete the rest of execution log and start recording it anew.\n\") > OK. I will use it. Hui