You can subscribe to this list here.
| 2002 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
(1) |
Oct
(122) |
Nov
(152) |
Dec
(69) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2003 |
Jan
(6) |
Feb
(25) |
Mar
(73) |
Apr
(82) |
May
(24) |
Jun
(25) |
Jul
(10) |
Aug
(11) |
Sep
(10) |
Oct
(54) |
Nov
(203) |
Dec
(182) |
| 2004 |
Jan
(307) |
Feb
(305) |
Mar
(430) |
Apr
(312) |
May
(187) |
Jun
(342) |
Jul
(487) |
Aug
(637) |
Sep
(336) |
Oct
(373) |
Nov
(441) |
Dec
(210) |
| 2005 |
Jan
(385) |
Feb
(480) |
Mar
(636) |
Apr
(544) |
May
(679) |
Jun
(625) |
Jul
(810) |
Aug
(838) |
Sep
(634) |
Oct
(521) |
Nov
(965) |
Dec
(543) |
| 2006 |
Jan
(494) |
Feb
(431) |
Mar
(546) |
Apr
(411) |
May
(406) |
Jun
(322) |
Jul
(256) |
Aug
(401) |
Sep
(345) |
Oct
(542) |
Nov
(308) |
Dec
(481) |
| 2007 |
Jan
(427) |
Feb
(326) |
Mar
(367) |
Apr
(255) |
May
(244) |
Jun
(204) |
Jul
(223) |
Aug
(231) |
Sep
(354) |
Oct
(374) |
Nov
(497) |
Dec
(362) |
| 2008 |
Jan
(322) |
Feb
(482) |
Mar
(658) |
Apr
(422) |
May
(476) |
Jun
(396) |
Jul
(455) |
Aug
(267) |
Sep
(280) |
Oct
(253) |
Nov
(232) |
Dec
(304) |
| 2009 |
Jan
(486) |
Feb
(470) |
Mar
(458) |
Apr
(423) |
May
(696) |
Jun
(461) |
Jul
(551) |
Aug
(575) |
Sep
(134) |
Oct
(110) |
Nov
(157) |
Dec
(102) |
| 2010 |
Jan
(226) |
Feb
(86) |
Mar
(147) |
Apr
(117) |
May
(107) |
Jun
(203) |
Jul
(193) |
Aug
(238) |
Sep
(300) |
Oct
(246) |
Nov
(23) |
Dec
(75) |
| 2011 |
Jan
(133) |
Feb
(195) |
Mar
(315) |
Apr
(200) |
May
(267) |
Jun
(293) |
Jul
(353) |
Aug
(237) |
Sep
(278) |
Oct
(611) |
Nov
(274) |
Dec
(260) |
| 2012 |
Jan
(303) |
Feb
(391) |
Mar
(417) |
Apr
(441) |
May
(488) |
Jun
(655) |
Jul
(590) |
Aug
(610) |
Sep
(526) |
Oct
(478) |
Nov
(359) |
Dec
(372) |
| 2013 |
Jan
(467) |
Feb
(226) |
Mar
(391) |
Apr
(281) |
May
(299) |
Jun
(252) |
Jul
(311) |
Aug
(352) |
Sep
(481) |
Oct
(571) |
Nov
(222) |
Dec
(231) |
| 2014 |
Jan
(185) |
Feb
(329) |
Mar
(245) |
Apr
(238) |
May
(281) |
Jun
(399) |
Jul
(382) |
Aug
(500) |
Sep
(579) |
Oct
(435) |
Nov
(487) |
Dec
(256) |
| 2015 |
Jan
(338) |
Feb
(357) |
Mar
(330) |
Apr
(294) |
May
(191) |
Jun
(108) |
Jul
(142) |
Aug
(261) |
Sep
(190) |
Oct
(54) |
Nov
(83) |
Dec
(22) |
| 2016 |
Jan
(49) |
Feb
(89) |
Mar
(33) |
Apr
(50) |
May
(27) |
Jun
(34) |
Jul
(53) |
Aug
(53) |
Sep
(98) |
Oct
(206) |
Nov
(93) |
Dec
(53) |
| 2017 |
Jan
(65) |
Feb
(82) |
Mar
(102) |
Apr
(86) |
May
(187) |
Jun
(67) |
Jul
(23) |
Aug
(93) |
Sep
(65) |
Oct
(45) |
Nov
(35) |
Dec
(17) |
| 2018 |
Jan
(26) |
Feb
(35) |
Mar
(38) |
Apr
(32) |
May
(8) |
Jun
(43) |
Jul
(27) |
Aug
(30) |
Sep
(43) |
Oct
(42) |
Nov
(38) |
Dec
(67) |
| 2019 |
Jan
(32) |
Feb
(37) |
Mar
(53) |
Apr
(64) |
May
(49) |
Jun
(18) |
Jul
(14) |
Aug
(53) |
Sep
(25) |
Oct
(30) |
Nov
(49) |
Dec
(31) |
| 2020 |
Jan
(87) |
Feb
(45) |
Mar
(37) |
Apr
(51) |
May
(99) |
Jun
(36) |
Jul
(11) |
Aug
(14) |
Sep
(20) |
Oct
(24) |
Nov
(40) |
Dec
(23) |
| 2021 |
Jan
(14) |
Feb
(53) |
Mar
(85) |
Apr
(15) |
May
(19) |
Jun
(3) |
Jul
(14) |
Aug
(1) |
Sep
(57) |
Oct
(73) |
Nov
(56) |
Dec
(22) |
| 2022 |
Jan
(3) |
Feb
(22) |
Mar
(6) |
Apr
(55) |
May
(46) |
Jun
(39) |
Jul
(15) |
Aug
(9) |
Sep
(11) |
Oct
(34) |
Nov
(20) |
Dec
(36) |
| 2023 |
Jan
(79) |
Feb
(41) |
Mar
(99) |
Apr
(169) |
May
(48) |
Jun
(16) |
Jul
(16) |
Aug
(57) |
Sep
(19) |
Oct
|
Nov
|
Dec
|
| S | M | T | W | T | F | S |
|---|---|---|---|---|---|---|
|
1
(8) |
2
(8) |
3
(15) |
4
(14) |
5
(12) |
6
(40) |
7
(9) |
|
8
(5) |
9
(12) |
10
(9) |
11
(13) |
12
(7) |
13
(7) |
14
(19) |
|
15
(18) |
16
(13) |
17
(16) |
18
(8) |
19
(16) |
20
(16) |
21
(12) |
|
22
(21) |
23
(39) |
24
(27) |
25
(33) |
26
(41) |
27
(17) |
28
(15) |
|
From: <sv...@va...> - 2009-02-26 23:16:05
|
Author: njn Date: 2009-02-26 23:15:53 +0000 (Thu, 26 Feb 2009) New Revision: 9287 Log: Fix vgprintf, broken by recent filter changes. Modified: branches/DARWIN/none/tests/vgprintf.stdout.exp Modified: branches/DARWIN/none/tests/vgprintf.stdout.exp =================================================================== --- branches/DARWIN/none/tests/vgprintf.stdout.exp 2009-02-26 22:38:42 UTC (rev 9286) +++ branches/DARWIN/none/tests/vgprintf.stdout.exp 2009-02-26 23:15:53 UTC (rev 9287) @@ -1 +1 @@ -. +... |
|
From: <sv...@va...> - 2009-02-26 22:38:49
|
Author: njn
Date: 2009-02-26 22:38:42 +0000 (Thu, 26 Feb 2009)
New Revision: 9286
Log:
Merge r9285 (match-overrun fixes) from the DARWIN branch.
Modified:
trunk/memcheck/tests/match-overrun.c
trunk/memcheck/tests/match-overrun.stderr.exp
trunk/memcheck/tests/match-overrun.supp
trunk/memcheck/tests/match-overrun.vgtest
Modified: trunk/memcheck/tests/match-overrun.c
===================================================================
--- trunk/memcheck/tests/match-overrun.c 2009-02-26 22:36:14 UTC (rev 9285)
+++ trunk/memcheck/tests/match-overrun.c 2009-02-26 22:38:42 UTC (rev 9286)
@@ -1,9 +1,12 @@
#include <stdio.h>
#include <stdlib.h>
+#include "memcheck/memcheck.h"
static int a1234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789(void)
{
- return *(int *)0;
+ int x;
+ VALGRIND_MAKE_MEM_NOACCESS(&x, sizeof(x));
+ return *(int *)&x;
}
int main(int argc, char **argv)
Modified: trunk/memcheck/tests/match-overrun.stderr.exp
===================================================================
--- trunk/memcheck/tests/match-overrun.stderr.exp 2009-02-26 22:36:14 UTC (rev 9285)
+++ trunk/memcheck/tests/match-overrun.stderr.exp 2009-02-26 22:38:42 UTC (rev 9286)
@@ -1,16 +0,0 @@
-
-
-Process terminating with default action of signal 11 (SIGSEGV)
- Access not within mapped region at address 0x........
- at 0x........: a1234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789 (match-overrun.c:6)
- by 0x........: main (match-overrun.c:11)
- If you believe this happened as a result of a stack overflow in your
- program's main thread (unlikely but possible), you can try to increase
- the size of the main thread stack using the --main-stacksize= flag.
- The main thread stack size used in this run was ....
-
-ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)
-malloc/free: in use at exit: ... bytes in ... blocks.
-malloc/free: ... allocs, ... frees, ... bytes allocated.
-For a detailed leak analysis, rerun with: --leak-check=yes
-For counts of detected errors, rerun with: -v
Modified: trunk/memcheck/tests/match-overrun.supp
===================================================================
--- trunk/memcheck/tests/match-overrun.supp 2009-02-26 22:36:14 UTC (rev 9285)
+++ trunk/memcheck/tests/match-overrun.supp 2009-02-26 22:38:42 UTC (rev 9286)
@@ -2,4 +2,5 @@
test
Memcheck:Addr4
fun:a123456789*
+ fun:main
}
Modified: trunk/memcheck/tests/match-overrun.vgtest
===================================================================
--- trunk/memcheck/tests/match-overrun.vgtest 2009-02-26 22:36:14 UTC (rev 9285)
+++ trunk/memcheck/tests/match-overrun.vgtest 2009-02-26 22:38:42 UTC (rev 9286)
@@ -1,3 +1,3 @@
prog: match-overrun
-vgopts: --suppressions=match-overrun.supp
+vgopts: -q --suppressions=match-overrun.supp
stderr_filter: filter_allocs
|
|
From: <sv...@va...> - 2009-02-26 22:36:20
|
Author: njn
Date: 2009-02-26 22:36:14 +0000 (Thu, 26 Feb 2009)
New Revision: 9285
Log:
In memcheck/tests/match-overrun.c, don't assume anything about the memory
layout at address 0x0.
Also, it really looks like this test is meant to be testing wildcard
matching in suppressions. So I rewrote the suppression so it actually
works.
The test now succeeds on Darwin.
Modified:
branches/DARWIN/memcheck/tests/match-overrun.c
branches/DARWIN/memcheck/tests/match-overrun.stderr.exp
branches/DARWIN/memcheck/tests/match-overrun.supp
branches/DARWIN/memcheck/tests/match-overrun.vgtest
Modified: branches/DARWIN/memcheck/tests/match-overrun.c
===================================================================
--- branches/DARWIN/memcheck/tests/match-overrun.c 2009-02-26 22:26:45 UTC (rev 9284)
+++ branches/DARWIN/memcheck/tests/match-overrun.c 2009-02-26 22:36:14 UTC (rev 9285)
@@ -1,9 +1,12 @@
#include <stdio.h>
#include <stdlib.h>
+#include "memcheck/memcheck.h"
static int a1234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789(void)
{
- return *(int *)0;
+ int x;
+ VALGRIND_MAKE_MEM_NOACCESS(&x, sizeof(x));
+ return *(int *)&x;
}
int main(int argc, char **argv)
Modified: branches/DARWIN/memcheck/tests/match-overrun.stderr.exp
===================================================================
--- branches/DARWIN/memcheck/tests/match-overrun.stderr.exp 2009-02-26 22:26:45 UTC (rev 9284)
+++ branches/DARWIN/memcheck/tests/match-overrun.stderr.exp 2009-02-26 22:36:14 UTC (rev 9285)
@@ -1,16 +0,0 @@
-
-
-Process terminating with default action of signal 11 (SIGSEGV)
- Access not within mapped region at address 0x........
- at 0x........: a1234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789 (match-overrun.c:6)
- by 0x........: main (match-overrun.c:11)
- If you believe this happened as a result of a stack overflow in your
- program's main thread (unlikely but possible), you can try to increase
- the size of the main thread stack using the --main-stacksize= flag.
- The main thread stack size used in this run was ....
-
-ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)
-malloc/free: in use at exit: ... bytes in ... blocks.
-malloc/free: ... allocs, ... frees, ... bytes allocated.
-For a detailed leak analysis, rerun with: --leak-check=yes
-For counts of detected errors, rerun with: -v
Modified: branches/DARWIN/memcheck/tests/match-overrun.supp
===================================================================
--- branches/DARWIN/memcheck/tests/match-overrun.supp 2009-02-26 22:26:45 UTC (rev 9284)
+++ branches/DARWIN/memcheck/tests/match-overrun.supp 2009-02-26 22:36:14 UTC (rev 9285)
@@ -2,4 +2,5 @@
test
Memcheck:Addr4
fun:a123456789*
+ fun:main
}
Modified: branches/DARWIN/memcheck/tests/match-overrun.vgtest
===================================================================
--- branches/DARWIN/memcheck/tests/match-overrun.vgtest 2009-02-26 22:26:45 UTC (rev 9284)
+++ branches/DARWIN/memcheck/tests/match-overrun.vgtest 2009-02-26 22:36:14 UTC (rev 9285)
@@ -1,3 +1,3 @@
prog: match-overrun
-vgopts: --suppressions=match-overrun.supp
+vgopts: -q --suppressions=match-overrun.supp
stderr_filter: filter_allocs
|
|
From: Josef W. <Jos...@gm...> - 2009-02-26 22:32:41
|
On Thursday 26 February 2009, Julian Seward wrote: > > Hi, > > I intend to package and ship 3.4.1 over the course of the coming > weekend. If there are any critical issues in 3_4_BRANCH that need > to be addressed before release, please let me know asap. I do not > envisage any such problems as I believe 3.4.1 to be pretty stable. > > The list of changes for 3.4.1 is in the NEWS file in > branches/VALGRIND_3_4_BRANCH. It's not much, primarily fixes to > debuginfo reading. You could mention that cachegrind and callgrind now can cope with non-power-of-two cache sizes, as long as the number of sets is power-of-two. Which makes it work with 24k Atom L1D and a few Core-2-Duo with 3/6MB L2 cache. Josef |
|
From: <sv...@va...> - 2009-02-26 22:26:51
|
Author: njn
Date: 2009-02-26 22:26:45 +0000 (Thu, 26 Feb 2009)
New Revision: 9284
Log:
Merge r9283 (lackey testing fixes) from the DARWIN branch.
Modified:
trunk/lackey/lk_main.c
trunk/lackey/tests/filter_stderr
trunk/lackey/tests/true.stderr.exp
trunk/tests/filter_numbers
Modified: trunk/lackey/lk_main.c
===================================================================
--- trunk/lackey/lk_main.c 2009-02-26 22:25:10 UTC (rev 9283)
+++ trunk/lackey/lk_main.c 2009-02-26 22:26:45 UTC (rev 9284)
@@ -886,12 +886,12 @@
VG_(message)(Vg_UserMsg, "");
VG_(message)(Vg_UserMsg, "Ratios:");
tl_assert(n_SBs_entered); // Paranoia time.
- VG_(message)(Vg_UserMsg, " guest instrs : SB entered = %3llu : 10",
+ VG_(message)(Vg_UserMsg, " guest instrs : SB entered = %'llu : 10",
10 * n_guest_instrs / n_SBs_entered);
- VG_(message)(Vg_UserMsg, " IRStmts : SB entered = %3llu : 10",
+ VG_(message)(Vg_UserMsg, " IRStmts : SB entered = %'llu : 10",
10 * n_IRStmts / n_SBs_entered);
tl_assert(n_guest_instrs); // Paranoia time.
- VG_(message)(Vg_UserMsg, " IRStmts : guest instr = %3llu : 10",
+ VG_(message)(Vg_UserMsg, " IRStmts : guest instr = %'llu : 10",
10 * n_IRStmts / n_guest_instrs);
}
Modified: trunk/lackey/tests/filter_stderr
===================================================================
--- trunk/lackey/tests/filter_stderr 2009-02-26 22:25:10 UTC (rev 9283)
+++ trunk/lackey/tests/filter_stderr 2009-02-26 22:26:45 UTC (rev 9284)
@@ -7,28 +7,5 @@
# Remove "Lackey, ..." line and the following copyright line.
sed "/^Lackey, an example Valgrind tool./ , /./ d" |
-# Output looks like this...
-#
-# Counted 53 calls to _dl_runtime_resolve()
-#
-# Executed:
-# BBs: 47131
-# x86 instrs: 193330
-# UInstrs: 523996
-#
-# Jccs:
-# total: 36368
-# % taken: 58%
-#
-# Ratios:
-# x86 instrs : BB = 41 : 10
-# UInstrs : BB = 111 : 10
-# UInstrs : x86_instr = 27 : 10
-#
-# Exit code: 0
-#
-# ...so chop all lines between first and last (inclusive)
-sed "/^Counted [0-9]\+ calls to _dl_runtime_resolve()$/ , \
- /UInstrs : x86_instrs = [0-9]\+ : [0-9]\+/ \
- d"
-
+# Filter all the numbers.
+../../tests/filter_numbers
Modified: trunk/lackey/tests/true.stderr.exp
===================================================================
--- trunk/lackey/tests/true.stderr.exp 2009-02-26 22:25:10 UTC (rev 9283)
+++ trunk/lackey/tests/true.stderr.exp 2009-02-26 22:26:45 UTC (rev 9284)
@@ -1,2 +1,20 @@
+Counted ... calls to _dl_runtime_resolve()
+
+Jccs:
+ total: ...
+ taken: ... ( ...%)
+
+Executed:
+ SBs entered: ...
+ SBs completed: ...
+ guest instrs: ...
+ IRStmts: ...
+
+Ratios:
+ guest instrs : SB entered = ... : ...
+ IRStmts : SB entered = ... : ...
+ IRStmts : guest instr = ... : ...
+
+Exit code: ...
Modified: trunk/tests/filter_numbers
===================================================================
--- trunk/tests/filter_numbers 2009-02-26 22:25:10 UTC (rev 9283)
+++ trunk/tests/filter_numbers 2009-02-26 22:26:45 UTC (rev 9284)
@@ -1,3 +1,3 @@
#!/bin/sh
-sed 's/[0-9]*/./'
+perl -p -e 's/[0-9,]+/.../g'
|
|
From: xinran w. <xin...@gm...> - 2009-02-26 22:25:54
|
Hi,
I'm writting a new Valgrind tool. I have a question: How can I get the
value of a temporary, eg. the value of t1 ?
For exmaple, how can I know the temporary value with index temp1 in
the below function "wrtmp_rdtmp_handler".
// wrtmp_rdtmp_handler is the function to instrument Ist_WrTmp
instruction with Iex_RdTmp data
static VG_REGPARM(2) void wrtmp_rdtmp_handler(IRTemp temp1, IRTemp temp2)
{
// The value of temp1 is just a index. My question is how can I
know the temporary value with index temp1.
// eg. if value of temp1 is 1, what is the value of t1 ?
}
~~~~~~~~~~~~~~~~~~~~~
// code snippet to instrument Ist_WrTmp instruction with Iex_RdTmp data
IRStmt* st = sbIn->stmts[i];
switch (st->tag) {
case Ist_WrTmp:
data = st->Ist.WrTmp.data;
type = typeOfIRExpr(sbOut->tyenv, data);
tl_assert(type != Ity_INVALID);
switch (data->tag) {
case Iex_RdTmp:
argv = mkIRExprVec_2( mkIRExpr_HWord(
data->Iex.RdTmp.tmp ),
mkIRExpr_HWord(st->Ist.WrTmp.tmp) );
di = unsafeIRDirty_0_N( /*regparms*/2,
"wrtmp_rdtmp_handler",
VG_(fnptr_to_fnentry)( wrtmp_rdtmp_handler ),
argv );
addStmtToIRSB( sbOut, IRStmt_Dirty(di) );
break;
}
}
~~~~~~~~~~~~~~~~~~~~~~~
Any suggest will be greatly appreciated.
Thank you very much.
-Xinran
|
|
From: <sv...@va...> - 2009-02-26 22:25:19
|
Author: njn
Date: 2009-02-26 22:25:10 +0000 (Thu, 26 Feb 2009)
New Revision: 9283
Log:
Fix filter_numbers, and do in perl so it works more reliably.
Change Lackey's filtering so it doesn't rely on non-universal sed features.
Tweak Lackey's output a tiny bit (changing spacing and comma-insertion on 3
number outputs) to make it more consistent and easier to filter for testing.
lackey/tests/true now passes on Darwin.
Modified:
branches/DARWIN/lackey/lk_main.c
branches/DARWIN/lackey/tests/filter_stderr
branches/DARWIN/lackey/tests/true.stderr.exp
branches/DARWIN/tests/filter_numbers
Modified: branches/DARWIN/lackey/lk_main.c
===================================================================
--- branches/DARWIN/lackey/lk_main.c 2009-02-26 22:14:20 UTC (rev 9282)
+++ branches/DARWIN/lackey/lk_main.c 2009-02-26 22:25:10 UTC (rev 9283)
@@ -886,12 +886,12 @@
VG_(message)(Vg_UserMsg, "");
VG_(message)(Vg_UserMsg, "Ratios:");
tl_assert(n_SBs_entered); // Paranoia time.
- VG_(message)(Vg_UserMsg, " guest instrs : SB entered = %3llu : 10",
+ VG_(message)(Vg_UserMsg, " guest instrs : SB entered = %'llu : 10",
10 * n_guest_instrs / n_SBs_entered);
- VG_(message)(Vg_UserMsg, " IRStmts : SB entered = %3llu : 10",
+ VG_(message)(Vg_UserMsg, " IRStmts : SB entered = %'llu : 10",
10 * n_IRStmts / n_SBs_entered);
tl_assert(n_guest_instrs); // Paranoia time.
- VG_(message)(Vg_UserMsg, " IRStmts : guest instr = %3llu : 10",
+ VG_(message)(Vg_UserMsg, " IRStmts : guest instr = %'llu : 10",
10 * n_IRStmts / n_guest_instrs);
}
Modified: branches/DARWIN/lackey/tests/filter_stderr
===================================================================
--- branches/DARWIN/lackey/tests/filter_stderr 2009-02-26 22:14:20 UTC (rev 9282)
+++ branches/DARWIN/lackey/tests/filter_stderr 2009-02-26 22:25:10 UTC (rev 9283)
@@ -7,28 +7,5 @@
# Remove "Lackey, ..." line and the following copyright line.
sed "/^Lackey, an example Valgrind tool./ , /./ d" |
-# Output looks like this...
-#
-# Counted 53 calls to _dl_runtime_resolve()
-#
-# Executed:
-# BBs: 47131
-# x86 instrs: 193330
-# UInstrs: 523996
-#
-# Jccs:
-# total: 36368
-# % taken: 58%
-#
-# Ratios:
-# x86 instrs : BB = 41 : 10
-# UInstrs : BB = 111 : 10
-# UInstrs : x86_instr = 27 : 10
-#
-# Exit code: 0
-#
-# ...so chop all lines between first and last (inclusive)
-sed "/^Counted [0-9]\+ calls to _dl_runtime_resolve()$/ , \
- /UInstrs : x86_instrs = [0-9]\+ : [0-9]\+/ \
- d"
-
+# Filter all the numbers.
+../../tests/filter_numbers
Modified: branches/DARWIN/lackey/tests/true.stderr.exp
===================================================================
--- branches/DARWIN/lackey/tests/true.stderr.exp 2009-02-26 22:14:20 UTC (rev 9282)
+++ branches/DARWIN/lackey/tests/true.stderr.exp 2009-02-26 22:25:10 UTC (rev 9283)
@@ -1,2 +1,20 @@
+Counted ... calls to _dl_runtime_resolve()
+
+Jccs:
+ total: ...
+ taken: ... ( ...%)
+
+Executed:
+ SBs entered: ...
+ SBs completed: ...
+ guest instrs: ...
+ IRStmts: ...
+
+Ratios:
+ guest instrs : SB entered = ... : ...
+ IRStmts : SB entered = ... : ...
+ IRStmts : guest instr = ... : ...
+
+Exit code: ...
Modified: branches/DARWIN/tests/filter_numbers
===================================================================
--- branches/DARWIN/tests/filter_numbers 2009-02-26 22:14:20 UTC (rev 9282)
+++ branches/DARWIN/tests/filter_numbers 2009-02-26 22:25:10 UTC (rev 9283)
@@ -1,3 +1,3 @@
#!/bin/sh
-sed 's/[0-9]*/./'
+perl -p -e 's/[0-9,]+/.../g'
|
|
From: Josef W. <Jos...@gm...> - 2009-02-26 22:14:55
|
On Thursday 26 February 2009, Nicholas Nethercote wrote: > Hi, > > Back when Valgrind tools were built as shared objects, we had this > notion of the binary interface version, so you could tell if a tool > would work with a particular version of the Valgrind core. > > Now that tools are statically linked with the core, this isn't > necessary. (IMHO this is excellent.) Turns out there was some > left-over cruft from this. The patch below gets rid of it. Any > objections? The main reason I'm asking is to make sure that Josef is > happy with removing the version-specific argv-handling code. Yeah, no problem from my side. Just get rid of it, thanks. An externally distributed tool still would need to get an version number of the tool interface used in the static VG libraries, at build time. Of course, it just can look at the Valgrind release version (I assume with pkgconfig?). Still, would it be better to have an explicit tool interface version given in valgrind.pc? Josef |
|
From: <sv...@va...> - 2009-02-26 22:14:28
|
Author: njn
Date: 2009-02-26 22:14:20 +0000 (Thu, 26 Feb 2009)
New Revision: 9282
Log:
Merge r9281 (fix memalign2 on AIX) from the trunk; the test now passes on
Darwin too.
Modified:
branches/DARWIN/memcheck/tests/memalign2.c
Modified: branches/DARWIN/memcheck/tests/memalign2.c
===================================================================
--- branches/DARWIN/memcheck/tests/memalign2.c 2009-02-26 22:09:53 UTC (rev 9281)
+++ branches/DARWIN/memcheck/tests/memalign2.c 2009-02-26 22:14:20 UTC (rev 9282)
@@ -24,10 +24,10 @@
int main ( void )
{
# if defined(_AIX)
- printf("AIX 5.2 knows about neither memalign() nor posix_memalign().\n");
+ // AIX 5.2 has neither memalign() nor posix_memalign(); do nothing.
# elif defined(__APPLE__)
- printf("MacOS X knows about neither memalign() nor posix_memalign().\n");
+ // Likewise for Mac OS X.
# else
// Nb: assuming VG_MIN_MALLOC_SZB is 8!
|
|
From: <sv...@va...> - 2009-02-26 22:10:05
|
Author: njn
Date: 2009-02-26 22:09:53 +0000 (Thu, 26 Feb 2009)
New Revision: 9281
Log:
Make this pass on AIX (untested).
Modified:
trunk/memcheck/tests/memalign2.c
Modified: trunk/memcheck/tests/memalign2.c
===================================================================
--- trunk/memcheck/tests/memalign2.c 2009-02-26 21:26:34 UTC (rev 9280)
+++ trunk/memcheck/tests/memalign2.c 2009-02-26 22:09:53 UTC (rev 9281)
@@ -20,7 +20,7 @@
int main ( void )
{
# if defined(_AIX)
- printf("AIX 5.2 knows about neither memalign() nor posix_memalign().\n");
+ // AIX 5.2 has neither memalign() nor posix_memalign(); do nothing.
# else
// Nb: assuming VG_MIN_MALLOC_SZB is 8!
|
|
From: Paul M. <pa...@sa...> - 2009-02-26 21:45:09
|
Julian Seward writes:
> VKI_PAGE_SIZE isn't a #defined constant on ppc{32,64}-linux -- it's a
> variable which is set at startup (of V). This is because on ppc64-linux
> there was a movement at one time to also support page sizes of 64k,
> although I think that has now been abandoned.
Not abandoned at all - 64k pages are fully supported upstream, and
RHEL5 uses 64k pages on ppc64.
Paul.
|
|
From: <sv...@va...> - 2009-02-26 21:26:40
|
Author: njn
Date: 2009-02-26 21:26:34 +0000 (Thu, 26 Feb 2009)
New Revision: 9280
Log:
Merge r9279 (diff -u cleanup) from the trunk.
Modified:
branches/DARWIN/configure.in
Modified: branches/DARWIN/configure.in
===================================================================
--- branches/DARWIN/configure.in 2009-02-26 21:25:50 UTC (rev 9279)
+++ branches/DARWIN/configure.in 2009-02-26 21:26:34 UTC (rev 9280)
@@ -122,14 +122,16 @@
# Comparing two identical files results in 0, unless -u isn't supported (as
# it's not on AIX).
-touch tmp-xxx;
-if diff -u tmp-xxx tmp-xxx ; then
+tmpfile="tmp-xxx-yyy-zzz"
+touch $tmpfile;
+if diff -u $tmpfile $tmpfile ; then
AC_MSG_RESULT([yes])
DIFF="diff -u"
else
AC_MSG_RESULT([no])
DIFF="diff"
fi
+rm $tmpfile
# We don't want gcc < 3.0
|
|
From: Greg P. <gp...@ap...> - 2009-02-26 21:26:00
|
On Feb 26, 2009, at 1:15 PM, Nicholas Nethercote wrote: > On Fri, Feb 27, 2009 at 6:41 AM, Greg Parker <gp...@ap...> > wrote: >> >> That's correct, Darwin uses its own home-grown assembler. I don't >> know of >> any GAS support for Mach-O, and -mgas is not supported. > > Greg, do you have any opinion if it's worth trying to use a common > subset of the two dialects, or should we just shove all the current > failing asm tests into x86-linux/ directories? The instructions themselves are easy enough. Symbols and most other compiler directives need to be handled differently, and there are some ABI differences like stack alignment. There may be a way to macro-ize the most common constructs (like a basic main() with only ordinary instructions and no external code or data references), but I wouldn't try too hard to make it work everywhere. -- Greg Parker gp...@ap... Runtime Wrangler |
|
From: <sv...@va...> - 2009-02-26 21:25:56
|
Author: njn
Date: 2009-02-26 21:25:50 +0000 (Thu, 26 Feb 2009)
New Revision: 9279
Log:
Clean up the diff -u test a little.
Modified:
trunk/configure.in
Modified: trunk/configure.in
===================================================================
--- trunk/configure.in 2009-02-26 21:21:38 UTC (rev 9278)
+++ trunk/configure.in 2009-02-26 21:25:50 UTC (rev 9279)
@@ -121,14 +121,16 @@
# Comparing two identical files results in 0, unless -u isn't supported (as
# it's not on AIX).
-touch tmp-xxx;
-if diff -u tmp-xxx tmp-xxx ; then
+tmpfile="tmp-xxx-yyy-zzz"
+touch $tmpfile;
+if diff -u $tmpfile $tmpfile ; then
AC_MSG_RESULT([yes])
DIFF="diff -u"
else
AC_MSG_RESULT([no])
DIFF="diff"
fi
+rm $tmpfile
# We don't want gcc < 3.0
|
|
From: <sv...@va...> - 2009-02-26 21:21:50
|
Author: njn
Date: 2009-02-26 21:21:38 +0000 (Thu, 26 Feb 2009)
New Revision: 9278
Log:
Merge r9275 (port faultstatus to ppc) from the trunk.
Modified:
branches/DARWIN/none/tests/faultstatus.c
Modified: branches/DARWIN/none/tests/faultstatus.c
===================================================================
--- branches/DARWIN/none/tests/faultstatus.c 2009-02-26 21:19:25 UTC (rev 9277)
+++ branches/DARWIN/none/tests/faultstatus.c 2009-02-26 21:21:38 UTC (rev 9278)
@@ -9,6 +9,19 @@
#include <sys/mman.h>
#include <unistd.h>
+/*
+ * Division by zero triggers a SIGFPE on x86 and x86_64,
+ * but not on the PowerPC architecture.
+ */
+#if defined(__powerpc__)
+#define DIVISION_BY_ZERO_TRIGGERS_FPE 0
+#define DIVISION_BY_ZERO_SI_CODE SI_TKILL
+#else
+#define DIVISION_BY_ZERO_TRIGGERS_FPE 1
+#define DIVISION_BY_ZERO_SI_CODE FPE_INTDIV
+#endif
+
+
struct test {
void (*test)(void);
int sig;
@@ -93,6 +106,9 @@
volatile int v = 44/zero();
(void)v;
+#if DIVISION_BY_ZERO_TRIGGERS_FPE == 0
+ raise(SIGFPE);
+#endif
}
int main()
@@ -125,7 +141,7 @@
T(1, SIGSEGV, SEGV_MAPERR, BADADDR),
T(2, SIGSEGV, SEGV_ACCERR, mapping),
T(3, SIGBUS, BUS_ADRERR, &mapping[FILESIZE+10]),
- T(4, SIGFPE, FPE_INTDIV, 0),
+ T(4, SIGFPE, DIVISION_BY_ZERO_SI_CODE, 0),
#undef T
};
|
|
From: <sv...@va...> - 2009-02-26 21:19:36
|
Author: njn
Date: 2009-02-26 21:19:25 +0000 (Thu, 26 Feb 2009)
New Revision: 9277
Log:
Merge r9276 (unit_libcbase fix) from trunk.
Modified:
branches/DARWIN/memcheck/tests/unit_libcbase.c
Modified: branches/DARWIN/memcheck/tests/unit_libcbase.c
===================================================================
--- branches/DARWIN/memcheck/tests/unit_libcbase.c 2009-02-26 21:17:39 UTC (rev 9276)
+++ branches/DARWIN/memcheck/tests/unit_libcbase.c 2009-02-26 21:19:25 UTC (rev 9277)
@@ -10,6 +10,7 @@
#define CHECK(x) \
if (!x) { fprintf(stderr, "failure: %s:%d\n", __FILE__, __LINE__); }
+
void test_VG_STREQ(void)
{
CHECK( ! VG_STREQ(NULL, NULL) ); // Nb: strcmp() considers these equal
@@ -53,6 +54,11 @@
CHECK( VG_STREQN(3, "abcd", "abce"));
}
+// On PPC/Linux VKI_PAGE_SIZE is a variable, not a macro.
+#if defined(VGP_ppc32_linux) || defined(VGP_ppc64_linux)
+unsigned long VKI_PAGE_SIZE = 1UL << 12;
+#endif
+
void test_VG_IS_XYZ_ALIGNED(void)
{
CHECK( VG_IS_2_ALIGNED(0x0) );
|
|
From: <sv...@va...> - 2009-02-26 21:17:50
|
Author: njn
Date: 2009-02-26 21:17:39 +0000 (Thu, 26 Feb 2009)
New Revision: 9276
Log:
Make the VKI_PAGE_SIZE conditional match that in m_vki.c.
Modified:
trunk/memcheck/tests/unit_libcbase.c
Modified: trunk/memcheck/tests/unit_libcbase.c
===================================================================
--- trunk/memcheck/tests/unit_libcbase.c 2009-02-26 17:50:14 UTC (rev 9275)
+++ trunk/memcheck/tests/unit_libcbase.c 2009-02-26 21:17:39 UTC (rev 9276)
@@ -11,11 +11,6 @@
if (!x) { fprintf(stderr, "failure: %s:%d\n", __FILE__, __LINE__); }
-#if ! defined(VKI_PAGE_SIZE)
-unsigned long VKI_PAGE_SIZE = VKI_MAX_PAGE_SIZE;
-#endif
-
-
void test_VG_STREQ(void)
{
CHECK( ! VG_STREQ(NULL, NULL) ); // Nb: strcmp() considers these equal
@@ -59,6 +54,11 @@
CHECK( VG_STREQN(3, "abcd", "abce"));
}
+// On PPC/Linux VKI_PAGE_SIZE is a variable, not a macro.
+#if defined(VGP_ppc32_linux) || defined(VGP_ppc64_linux)
+unsigned long VKI_PAGE_SIZE = 1UL << 12;
+#endif
+
void test_VG_IS_XYZ_ALIGNED(void)
{
CHECK( VG_IS_2_ALIGNED(0x0) );
|
|
From: Nicholas N. <n.n...@gm...> - 2009-02-26 21:15:28
|
On Fri, Feb 27, 2009 at 6:41 AM, Greg Parker <gp...@ap...> wrote: > > That's correct, Darwin uses its own home-grown assembler. I don't know of > any GAS support for Mach-O, and -mgas is not supported. Greg, do you have any opinion if it's worth trying to use a common subset of the two dialects, or should we just shove all the current failing asm tests into x86-linux/ directories? N |
|
From: Greg P. <gp...@ap...> - 2009-02-26 19:41:26
|
On Feb 26, 2009, at 9:54 AM, Bart Van Assche wrote: > On Thu, Feb 26, 2009 at 3:47 AM, Nicholas Nethercote > <n.n...@gm...> wrote: >> On Wed, Feb 25, 2009 at 5:56 PM, Bart Van Assche >> <bar...@gm...> wrote: >>> On Wed, Feb 25, 2009 at 3:50 AM, Nicholas Nethercote >>> <n.n...@gm...> wrote: >>>> x86/Linux assembly code and x86/Darwin assembly code appear to be >>>> different dialects. Some of the tests in none/tests/x86/ and >>>> memcheck/tests/x86/ compile on x86/Darwin, eg. >>>> memcheck/tests/bug152022. But some don't, eg. pushfpopf, because >>>> x86/Darwin assembler doesn't accept the .type and .size options. >>>> >>>> The easiest thing to do is move the ones that don't compile on >>>> Darwin >>>> into an x86-linux/ directory. Anyone have any strong opinons about >>>> this, or other insights? >>> >>> Which standards apply to x86 assembler syntax ? >> >> I'm not sure that there are any. This seems to be the definitive >> documentation on Mac/x86 assembly: >> >> http://developer.apple.com/DOCUMENTATION/DeveloperTools/Reference/Assembler/index.html > > Does this mean that gcc on Darwin does not use gas but another > assembler ? Is the gcc command-line option -mgas supported by gcc on > Darwin ? That's correct, Darwin uses its own home-grown assembler. I don't know of any GAS support for Mach-O, and -mgas is not supported. -- Greg Parker gp...@ap... Runtime Wrangler |
|
From: Bart V. A. <bar...@gm...> - 2009-02-26 18:06:51
|
On Thu, Feb 26, 2009 at 3:47 AM, Nicholas Nethercote <n.n...@gm...> wrote: > On Wed, Feb 25, 2009 at 5:56 PM, Bart Van Assche > <bar...@gm...> wrote: >> On Wed, Feb 25, 2009 at 3:50 AM, Nicholas Nethercote >> <n.n...@gm...> wrote: >>> x86/Linux assembly code and x86/Darwin assembly code appear to be >>> different dialects. Some of the tests in none/tests/x86/ and >>> memcheck/tests/x86/ compile on x86/Darwin, eg. >>> memcheck/tests/bug152022. But some don't, eg. pushfpopf, because >>> x86/Darwin assembler doesn't accept the .type and .size options. >>> >>> The easiest thing to do is move the ones that don't compile on Darwin >>> into an x86-linux/ directory. Anyone have any strong opinons about >>> this, or other insights? >> >> Which standards apply to x86 assembler syntax ? > > I'm not sure that there are any. This seems to be the definitive > documentation on Mac/x86 assembly: > > http://developer.apple.com/DOCUMENTATION/DeveloperTools/Reference/Assembler/index.html Does this mean that gcc on Darwin does not use gas but another assembler ? Is the gcc command-line option -mgas supported by gcc on Darwin ? Bart. |
|
From: <sv...@va...> - 2009-02-26 17:50:19
|
Author: bart
Date: 2009-02-26 17:50:14 +0000 (Thu, 26 Feb 2009)
New Revision: 9275
Log:
Ported this test to ppc.
Modified:
trunk/none/tests/faultstatus.c
Modified: trunk/none/tests/faultstatus.c
===================================================================
--- trunk/none/tests/faultstatus.c 2009-02-26 13:21:50 UTC (rev 9274)
+++ trunk/none/tests/faultstatus.c 2009-02-26 17:50:14 UTC (rev 9275)
@@ -9,6 +9,19 @@
#include <sys/mman.h>
#include <unistd.h>
+/*
+ * Division by zero triggers a SIGFPE on x86 and x86_64,
+ * but not on the PowerPC architecture.
+ */
+#if defined(__powerpc__)
+#define DIVISION_BY_ZERO_TRIGGERS_FPE 0
+#define DIVISION_BY_ZERO_SI_CODE SI_TKILL
+#else
+#define DIVISION_BY_ZERO_TRIGGERS_FPE 1
+#define DIVISION_BY_ZERO_SI_CODE FPE_INTDIV
+#endif
+
+
struct test {
void (*test)(void);
int sig;
@@ -93,6 +106,9 @@
volatile int v = 44/zero();
(void)v;
+#if DIVISION_BY_ZERO_TRIGGERS_FPE == 0
+ raise(SIGFPE);
+#endif
}
int main()
@@ -125,7 +141,7 @@
T(1, SIGSEGV, SEGV_MAPERR, BADADDR),
T(2, SIGSEGV, SEGV_ACCERR, mapping),
T(3, SIGBUS, BUS_ADRERR, &mapping[FILESIZE+10]),
- T(4, SIGFPE, FPE_INTDIV, 0),
+ T(4, SIGFPE, DIVISION_BY_ZERO_SI_CODE, 0),
#undef T
};
|
|
From: Julian S. <js...@ac...> - 2009-02-26 17:26:16
|
> unit_libcbase.o:(.toc+0x78): undefined reference to `VKI_PAGE_SIZE'
Not sure exactly what happened here, but ..
VKI_PAGE_SIZE isn't a #defined constant on ppc{32,64}-linux -- it's a
variable which is set at startup (of V). This is because on ppc64-linux
there was a movement at one time to also support page sizes of 64k,
although I think that has now been abandoned.
One easy fix is just to do the test with VKI_MAX_PAGE_SIZE, which
afaict is a constant on all platforms.
J
|
|
From: Julian S. <js...@ac...> - 2009-02-26 17:26:07
|
Hi, I intend to package and ship 3.4.1 over the course of the coming weekend. If there are any critical issues in 3_4_BRANCH that need to be addressed before release, please let me know asap. I do not envisage any such problems as I believe 3.4.1 to be pretty stable. The list of changes for 3.4.1 is in the NEWS file in branches/VALGRIND_3_4_BRANCH. It's not much, primarily fixes to debuginfo reading. J |
|
From: Stefan K. <en...@ho...> - 2009-02-26 15:48:46
|
Bart Van Assche schrieb: > On Tue, Feb 24, 2009 at 8:23 PM, Stefan Kost <en...@ho...> wrote: > >> but then you can anyway do >> make CFLAGS=-Werror >> :) >> > > What will the above command to with the CFLAGS variable that has > already been defined in each Makefile ? > > $ grep '^CFLAGS =' Makefile > CFLAGS = -Wno-long-long -Wno-pointer-sign > -Wdeclaration-after-statement -fno-stack-protector > > You should not override CFLAGS in Makefile.am. You can set AM_CFLAGS. Have a look at http://www.lrde.epita.fr/~adl/autotools.html (be aware its quite long). Stefan > Bart. > |
|
From: <sv...@va...> - 2009-02-26 13:22:06
|
Author: bart
Date: 2009-02-26 13:21:50 +0000 (Thu, 26 Feb 2009)
New Revision: 9274
Log:
Builds again on powerpc.
Modified:
trunk/memcheck/tests/unit_libcbase.c
Modified: trunk/memcheck/tests/unit_libcbase.c
===================================================================
--- trunk/memcheck/tests/unit_libcbase.c 2009-02-26 04:07:21 UTC (rev 9273)
+++ trunk/memcheck/tests/unit_libcbase.c 2009-02-26 13:21:50 UTC (rev 9274)
@@ -10,6 +10,12 @@
#define CHECK(x) \
if (!x) { fprintf(stderr, "failure: %s:%d\n", __FILE__, __LINE__); }
+
+#if ! defined(VKI_PAGE_SIZE)
+unsigned long VKI_PAGE_SIZE = VKI_MAX_PAGE_SIZE;
+#endif
+
+
void test_VG_STREQ(void)
{
CHECK( ! VG_STREQ(NULL, NULL) ); // Nb: strcmp() considers these equal
|