From mboxrd@z Thu Jan 1 00:00:00 1970 From: Masami Hiramatsu Subject: Re: [PATCH -tip -v10 5/7] x86: add pt_regs register and stack access APIs Date: Mon, 06 Jul 2009 15:28:02 -0400 Message-ID: <4A525042.5030702@redhat.com> References: <20090701010838.32547.62843.stgit@localhost.localdomain> <20090701010911.32547.1313.stgit@localhost.localdomain> <871votop6a.fsf@basil.nowhere.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <871votop6a.fsf@basil.nowhere.org> List-Unsubscribe: List-Subscribe: List-Post: List-Help: , Sender: systemtap-owner@sourceware.org To: Andi Kleen Cc: Ingo Molnar , Steven Rostedt , lkml , systemtap , kvm , DLE , Christoph Hellwig , Ananth N Mavinakayanahalli , Frederic Weisbecker , Roland McGrath , Srikar Dronamraju , linux-arch@vger.kernel.org List-Id: linux-arch.vger.kernel.org Andi Kleen wrote: > Masami Hiramatsu writes: > >> Add following APIs for accessing registers and stack entries from pt_regs. > > You forgot to state who calls these functions/why are they added? > Who only has strings for registers? Oh, yes. This patch is needed for kprobes based event tracer on ftrace. Some other debugging tools might be able to use it. > I can see the point of having a function for nth argument though, > that's useful. > >> +static inline unsigned long regs_get_argument_nth(struct pt_regs *regs, >> + unsigned n) >> +{ >> + if (n < NR_REGPARMS) { >> + switch (n) { >> + case 0: >> + return regs->ax; >> + case 1: >> + return regs->dx; >> + case 2: >> + return regs->cx; > > > [....] > > That could be done shorter with a offsetof table. > >> + if (n < NR_REGPARMS) { >> + switch (n) { >> + case 0: >> + return regs->di; >> + case 1: >> + return regs->si; >> + case 2: >> + return regs->dx; >> + case 3: >> + return regs->cx; >> + case 4: >> + return regs->r8; >> + case 5: >> + return regs->r9; > > and that too. I'm not so sure about your idea. Would you mean below code? int offs_table[NR_REGPARMS] = { [0] = offsetof(struct pt_regs, di), ... }; if (n < NR_REGPARMS) return *((unsigned long *)regs + offs_table[n]); Thank you, -- Masami Hiramatsu Software Engineer Hitachi Computer Products (America), Inc. Software Solutions Division e-mail: mhiramat@redhat.com From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx2.redhat.com ([66.187.237.31]:43352 "EHLO mx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753062AbZGFTZs (ORCPT ); Mon, 6 Jul 2009 15:25:48 -0400 Message-ID: <4A525042.5030702@redhat.com> Date: Mon, 06 Jul 2009 15:28:02 -0400 From: Masami Hiramatsu MIME-Version: 1.0 Subject: Re: [PATCH -tip -v10 5/7] x86: add pt_regs register and stack access APIs References: <20090701010838.32547.62843.stgit@localhost.localdomain> <20090701010911.32547.1313.stgit@localhost.localdomain> <871votop6a.fsf@basil.nowhere.org> In-Reply-To: <871votop6a.fsf@basil.nowhere.org> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-arch-owner@vger.kernel.org List-ID: To: Andi Kleen Cc: Ingo Molnar , Steven Rostedt , lkml , systemtap , kvm , DLE , Christoph Hellwig , Ananth N Mavinakayanahalli , Frederic Weisbecker , Roland McGrath , Srikar Dronamraju , linux-arch@vger.kernel.org Message-ID: <20090706192802.QVixDKD8ThNE10RSrz26-R7zqJPu1LQAPzVkQ-kLs3c@z> Andi Kleen wrote: > Masami Hiramatsu writes: > >> Add following APIs for accessing registers and stack entries from pt_regs. > > You forgot to state who calls these functions/why are they added? > Who only has strings for registers? Oh, yes. This patch is needed for kprobes based event tracer on ftrace. Some other debugging tools might be able to use it. > I can see the point of having a function for nth argument though, > that's useful. > >> +static inline unsigned long regs_get_argument_nth(struct pt_regs *regs, >> + unsigned n) >> +{ >> + if (n < NR_REGPARMS) { >> + switch (n) { >> + case 0: >> + return regs->ax; >> + case 1: >> + return regs->dx; >> + case 2: >> + return regs->cx; > > > [....] > > That could be done shorter with a offsetof table. > >> + if (n < NR_REGPARMS) { >> + switch (n) { >> + case 0: >> + return regs->di; >> + case 1: >> + return regs->si; >> + case 2: >> + return regs->dx; >> + case 3: >> + return regs->cx; >> + case 4: >> + return regs->r8; >> + case 5: >> + return regs->r9; > > and that too. I'm not so sure about your idea. Would you mean below code? int offs_table[NR_REGPARMS] = { [0] = offsetof(struct pt_regs, di), ... }; if (n < NR_REGPARMS) return *((unsigned long *)regs + offs_table[n]); Thank you, -- Masami Hiramatsu Software Engineer Hitachi Computer Products (America), Inc. Software Solutions Division e-mail: mhiramat@redhat.com