Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932354Ab0DPPZq (ORCPT ); Fri, 16 Apr 2010 11:25:46 -0400 Received: from mx1.redhat.com ([209.132.183.28]:62001 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932321Ab0DPPZ2 (ORCPT ); Fri, 16 Apr 2010 11:25:28 -0400 Date: Fri, 16 Apr 2010 11:25:02 -0400 From: Jason Baron To: linux-kernel@vger.kernel.org Cc: mingo@elte.hu, mathieu.desnoyers@polymtl.ca, hpa@zytor.com, tglx@linutronix.de, rostedt@goodmis.org, andi@firstfloor.org, roland@redhat.com, rth@redhat.com, mhiramat@redhat.com, fweisbec@gmail.com, avi@redhat.com, davem@davemloft.net, vgoyal@redhat.com Message-Id: In-Reply-To: References: Subject: [PATCH 11/11] jump label: add docs Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 5393 Lines: 162 Add jump label docs as: Documentation/jump-label.txt Signed-off-by: Jason Baron --- Documentation/jump-label.txt | 140 ++++++++++++++++++++++++++++++++++++++++++ 1 files changed, 140 insertions(+), 0 deletions(-) create mode 100644 Documentation/jump-label.txt diff --git a/Documentation/jump-label.txt b/Documentation/jump-label.txt new file mode 100644 index 0000000..4614821 --- /dev/null +++ b/Documentation/jump-label.txt @@ -0,0 +1,140 @@ + Jump Label + ---------- + +By: Jason Baron + + +1) motivation + + +Currently, tracepoints are implemented using a conditional. The conditional +check requires checking a global variable for each tracepoint. Although, +the overhead of this check is small, it increases under memory pressure. As we +increase the number of tracepoints in the kernel this may become more of an +issue. In addition, tracepoints are often dormant (disabled), and provide no +direct kernel functionality. Thus, it is highly desirable to reduce their +impact as much as possible. Although tracepoints are the original motivation +for this work, other kernel code paths should be able to make use of the jump +label optimization. + + +2) jump label description/usage + + +gcc (v4.5) adds a new 'asm goto' statement that allows branching to a label. +http://gcc.gnu.org/ml/gcc-patches/2009-07/msg01556.html + +Thus, this patch set introduces an architecture specific 'JUMP_LABEL()' macro as +follows (x86): + +# define JUMP_LABEL_INITIAL_NOP ".byte 0xe9 \n\t .long 0\n\t" + +# define JUMP_LABEL(tag, label, cond) \ + do { \ + extern const char __jlstrtab_##tag[]; \ + asm goto("1:" \ + JUMP_LABEL_INITIAL_NOP + ".pushsection __jump_table, \"a\" \n\t"\ + _ASM_PTR "1b, %l[" #label "], %c0 \n\t" \ + ".popsection \n\t" \ + : : "i" (__jlstrtab_##tag) : : label);\ + } while (0) + + +For architectures that have not yet introduced jump label support its simply: + +#define JUMP_LABEL(tag, label, cond) \ + if (unlikely(cond)) \ + goto label; + +which then can be used as: + + .... + JUMP_LABEL(trace_name, trace_label, jump_enabled); + printk("not doing tracing\n"); + return; +trace_label: + printk("doing tracing: %d\n", file); + .... + +Thus, when tracing is disabled, we simply have a no-op followed by a jump around +the dormant (disabled) tracing code. The 'JUMP_LABEL()' macro, produces a +'jump_table' which has the following format: + +[instruction address] [jump target] [tracepoint name] + +Thus, to enable a tracepoint, we simply patch the 'instruction address' with +a jump to the 'jump target'. + +The call to enable a jump label is: enable_jump_label(trace_name); to disable: +disable_jump_label(trace_name); + + +3) Jump label analysis (x86) + + +I've tested the performance of using 'get_cycles()' calls around the +tracepoint call sites. For an Intel Core 2 Quad cpu (in cycles, averages): + + idle after tbench run + ---- ---------------- +old code 32 88 +new code 2 4 + + +The performance improvement can be reproduced reliably on both Intel and AMD +hardware. + +In terms of code analysis the current code for the disabled case is a 'cmpl' +followed by a 'je' around the tracepoint code. so: + +cmpl - 83 3d 0e 77 87 00 00 - 7 bytes +je - 74 3e - 2 bytes + +total of 9 instruction bytes. + +The new code is a 'nopl' followed by a 'jmp'. Thus: + +nopl - 0f 1f 44 00 00 - 5 bytes +jmp - eb 3e - 2 bytes + +total of 7 instruction bytes. + +So, the new code also accounts for 2 less bytes in the instruction cache per tracepoint. + + +4) architecture interface + + +There are a few functions and macros which arches must implement in order to +take advantage of this optimization. As previously mentioned, if there is no +architecture support we simply fall back to a traditional, load, test, and +jump sequence. + +* add "HAVE_ARCH_JUMP_LABEL" to arch//Kconfig to indicate support + +* #define JUMP_LABEL_NOP_SIZE, arch/x86/include/asm/jump_label.h + +* #define "JUMP_LABEL(tag, label, cond)", arch/x86/include/asm/jump_label.h + +* add: void arch_jump_label_transform(struct jump_entry *entry, enum jump_label_type type) + and + const u8 *arch_get_jump_label_nop(void) + + see: arch/x86/kernel/jump_label.c + +* finally add a definition for "struct jump_entry". This must be done in a + separate .h file, b/c the modpost.c code uses this definition to sort the + the jump label tabel in the vmlinux, so that it does not have to be sorted at + runtime. see: arch/x86/include/asm/jump_entry.h + + +5) Acknowledgments + + +Thanks to Roland McGrath and Richard Henderson for coming up with the initial +'asm goto' and jump label design. + +Thanks to Mathieu Desnoyers for calling attention to this issue, outlining the +requirements of a solution, and implementing a solution in the form of the +"Immediate Values" work. -- 1.7.0.1 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/