[#6204] all test fail in gdb.asm/asm-source.exp through jtag

Document created by Aaronwu Employee on Oct 17, 2013
Version 1Show Document
  • View in full screen mode

[#6204] all test fail in gdb.asm/asm-source.exp through jtag

Submitted By: Vivi Li

Open Date

2010-08-25 06:37:37     Close Date

2011-10-16 23:22:07

Priority:

Medium High     Assignee:

Mingquan Pan

Board:

N/A     Silicon Revision:

Resolution:

Fixed     Fixed In Release:

N/A

Processor:

BF537     

Host Operating System:

toolchain rev.:

trunk head 4.3     kernel rev.:

State:

Closed     Found In Release:

N/A

Is this bug repeatable?:

N/A     

Summary: all test fail in gdb.asm/asm-source.exp through jtag

Details:

 

Since 2010_Jun_24 with svn version 4650, all test fail in gdb.asm/asm-source.exp on bf537-stamp through jtag.

For bf561-ezkit, this test never pass.

 

Bellow is the log:

--

(gdb) set remotetimeout 1600

(gdb) load /home/test/work/cruise/temp/regtest_build4.3/gdb_build/testsuite/gdb. ^Masm/asm-source

Loading section .text, size 0x5c lma 0x0

Loading section .data, size 0x8 lma 0x105c

Start address 0x0, load size 100

Transfer rate: 32000 bits/sec, 50 bytes/write.

(gdb) set remotetimeout 150^M

(gdb) continue

Continuing.

FAIL: gdb.asm/asm-source.exp: running to main in runto (timeout)

f

FAIL: gdb.asm/asm-source.exp: f at main (timeout)

n

FAIL: gdb.asm/asm-source.exp: next over macro (timeout)

s

FAIL: gdb.asm/asm-source.exp: step into foo2 (timeout)

info target

FAIL: gdb.asm/asm-source.exp: info target (timeout)

...

--

 

On 2010_Jun_23 with svn version 4646, some of tests in asm-source.exp can pass on bf537-stamp. Bellow is the list:

--

Running binutils-2.17/gdb/testsuite/gdb.asm/asm-source.exp ...

FAIL: gdb.asm/asm-source.exp: running to main in runto

FAIL: gdb.asm/asm-source.exp: f at main

FAIL: gdb.asm/asm-source.exp: next over macro

FAIL: gdb.asm/asm-source.exp: step into foo2

PASS: gdb.asm/asm-source.exp: info target

PASS: gdb.asm/asm-source.exp: info symbol

PASS: gdb.asm/asm-source.exp: list

PASS: gdb.asm/asm-source.exp: search

FAIL: gdb.asm/asm-source.exp: f in foo2

FAIL: gdb.asm/asm-source.exp: n in foo2

FAIL: gdb.asm/asm-source.exp: bt ALL in foo2

FAIL: gdb.asm/asm-source.exp: bt 2 in foo2

PASS: gdb.asm/asm-source.exp: s 2

PASS: gdb.asm/asm-source.exp: n 2

FAIL: gdb.asm/asm-source.exp: bt 3 in foo3

PASS: gdb.asm/asm-source.exp: info source asmsrc1.s

FAIL: gdb.asm/asm-source.exp: finish from foo3

FAIL: gdb.asm/asm-source.exp: info source asmsrc2.s

PASS: gdb.asm/asm-source.exp: info sources

FAIL: gdb.asm/asm-source.exp: info line

FAIL: gdb.asm/asm-source.exp: next over foo3

FAIL: gdb.asm/asm-source.exp: return from foo2 (GDB internal error)

FAIL: gdb.asm/asm-source.exp: look at global variable

PASS: gdb.asm/asm-source.exp: x/i &globalvar

PASS: gdb.asm/asm-source.exp: disassem &globalvar &globalvar+1

FAIL: gdb.asm/asm-source.exp: look at static variable

PASS: gdb.asm/asm-source.exp: x/i &staticvar

PASS: gdb.asm/asm-source.exp: disassem &staticvar &staticvar+1

PASS: gdb.asm/asm-source.exp: look at static function

--

 

Follow-ups

 

--- Steve Kilbane                                            2010-08-26 08:14:38

Changes between 4646 and 4650 were part of the ice100-b splurge - something

going on there?

 

--- Chad Wentworth                                           2010-08-26 10:15:31

Do we know if the ice100-b was used or the gnICE/gnICE+?  Do you have a dif file

between 4646 and 4650?  Also I noticed there is a timeout on all of these.  I

know that the ice-100b using max wait states which slows it down so could it be

the the timeout needs to be increased.  I have never run these tests so I don't

know much about them.

 

--- Chad Wentworth                                           2010-08-26 11:34:26

I got the diff files for what changed between these revs and they look to be all

from the ice-100b branch and not the trunk.  We should not be running anything

against the ice-100b branch at this point.  Please let me know if that is the

case and then we should try to rerun using the trunk.

 

--- Mike Frysinger                                           2010-08-26 15:44:20

all of our tests run against just trunk.  especially because the ice100b branch

only contained urjtag/gdbproxy, so it wouldnt really be possible.

 

that means there were no changes to trunk between svn rev 4646 and 4650 except

for the gdbproxy test changes by Vivi in rev 4647.

 

perhaps the difference in test results is that commit 4647 changed which dir

was being looked at:

--- a/buildscript/compare_toolchain_results

+++ b/buildscript/compare_toolchain_results

@@ -24,7 +24,7 @@ if [  $# -ge 3 ] ; then

fi

 

if [ ! $COMPARE ] ; then

-  COMPARE=~/work/cruise/checkouts/toolchain/gcc-4.1/contrib/compare_tests

+  COMPARE=~/work/cruise/checkouts/toolchain/gcc-4.3/contrib/compare_tests

fi

chmod 755 $COMPARE

 

 

--- Vivi Li                                                  2010-08-31 04:00:44

I don't think compare_tests is related to the failure. It's only a script to

compare test results.

 

--- Mike Frysinger                                           2010-08-31 09:12:23

right ... i was thinking of the log tree, not the compare script.  at any rate,

please recheck that these two versions actually test different.  the svn history

shows that your change is the only thing that occurred in svn toolchain trunk

between those two revisions quoted.

 

--- Vivi Li                                                  2010-09-03 04:56:27

I compared the toolchain build lof between 4646 and 4650, the only difference is

in the urjtag/po.

 

--

110850c110850

< checking code revision... from svn info: 4646

---

> checking code revision... from svn info: 4650

 

 

111605,111628d111604

< make urjtag.pot-update

< make[3]: Entering directory `/home/test/work/cruise/temp/urjtag_build/po'

< sed -e '/^#/d'

/home/test/work/cruise/checkouts/toolchain/urjtag/po/remove-potcdate.sin >

t-remove-potcdate.sed

< mv t-remove-potcdate.sed remove-potcdate.sed

< /usr/bin/xgettext --default-domain=urjtag

--directory=/home/test/work/cruise/checkouts/toolchain/urjtag \

<         --add-comments=TRANSLATORS: --keyword=_ --keyword=N_ \

<       

--files-from=/home/test/work/cruise/checkouts/toolchain/urjtag/po/POTFILES.in \

<         --copyright-holder='ETC s.r.o.' \

<         --msgid-bugs-address='marcel@telka.sk'

< test ! -f urjtag.po || { \

<         if test -f

/home/test/work/cruise/checkouts/toolchain/urjtag/po/urjtag.pot; then \

<           sed -f remove-potcdate.sed <

/home/test/work/cruise/checkouts/toolchain/urjtag/po/urjtag.pot > urjtag.1po

&& \

<           sed -f remove-potcdate.sed < urjtag.po > urjtag.2po

&& \

<           if cmp urjtag.1po urjtag.2po >/dev/null 2>&1; then \

<             rm -f urjtag.1po urjtag.2po urjtag.po; \

<           else \

<             rm -f urjtag.1po urjtag.2po

/home/test/work/cruise/checkouts/toolchain/urjtag/po/urjtag.pot && \

<             mv urjtag.po

/home/test/work/cruise/checkouts/toolchain/urjtag/po/urjtag.pot; \

<           fi; \

<         else \

<           mv urjtag.po

/home/test/work/cruise/checkouts/toolchain/urjtag/po/urjtag.pot; \

<         fi; \

<       }

< make[3]: Leaving directory `/home/test/work/cruise/temp/urjtag_build/po'

111631,111636c111607,111609

< cd /home/test/work/cruise/checkouts/toolchain/urjtag/po &&

/usr/bin/msgmerge --update fr.po urjtag.pot

< cd /home/test/work/cruise/checkouts/toolchain/urjtag/po && rm -f

fr.gmo && /usr/bin/msgfmt -c --statistics -o fr.gmo fr.po

< cd /home/test/work/cruise/checkouts/toolchain/urjtag/po &&

/usr/bin/msgmerge --update rw.po urjtag.pot

< cd /home/test/work/cruise/checkouts/toolchain/urjtag/po && rm -f

rw.gmo && /usr/bin/msgfmt -c --statistics -o rw.gmo rw.po

< cd /home/test/work/cruise/checkouts/toolchain/urjtag/po &&

/usr/bin/msgmerge --update sk.po urjtag.pot

< cd /home/test/work/cruise/checkouts/toolchain/urjtag/po && rm -f

sk.gmo && /usr/bin/msgfmt -c --statistics -o sk.gmo sk.po

---

> make[3]: `/home/test/work/cruise/checkouts/toolchain/urjtag/po/fr.gmo' is

up to date.

> make[3]: `/home/test/work/cruise/checkouts/toolchain/urjtag/po/rw.gmo' is

up to date.

> make[3]: `/home/test/work/cruise/checkouts/toolchain/urjtag/po/sk.gmo' is

up to date.

--

 

--- Vivi Li                                                  2010-09-06 22:27:34

I cleaned urjtag/po and rebuild toolchain, that has no influence to this bug.

 

--- Vivi Li                                                  2010-09-06 22:32:39

I revert toolchain to 4646, some cases pass.

I revert toolchain to 4650, all tests in asm-source.exp fail.

 

On the basis 4650, I revert the change of toolchain-regtest to 4646, all tests

in asm-source.exp fail.

Then I delete two added file in svn4650, some cases pass.

 

--- Steve Kilbane                                            2010-09-09 07:55:55

When you say "two added file in svn4650", you mean the two files added

in 4647, right? These two:

 

trunk/buildscript/start_gdbproxy.exp

trunk/buildscript/toolchain_test.gdbproxy

 

And when you say "pass" or "fail", do you mean you've

checked the log to verify that the test is passing (or failing), or do you mean

that you're running compare_toolchain_results, and it's reporting a failure?

 

--- Vivi Li                                                  2010-09-10 00:33:43

Hi Steve,

 

>When you say "two added file in svn4650", you mean the two files

added

>in 4647, right? These two:

>trunk/buildscript/start_gdbproxy.exp

>trunk/buildscript/toolchain_test.gdbproxy

 

Yes, that's what I mean.

 

>And when you say "pass" or "fail", do you mean you've

>checked the log to verify that the test is passing (or failing), or do you

mean

>that you're running compare_toolchain_results, and it's reporting a

failure?

 

I only check gdb.log generated during gdb test and compare_toolchain_results is

not run. "Pass" doesn't mean this case all passed, but better then all

fail; "fail" means all tests run by asm-source.exp are failed.

 

--- David Gibson                                             2010-09-13 09:26:08

What's the status of this bug Vivi?

Are you still looking into it?

 

--- Vivi Li                                                  2010-10-12 02:42:05

2010 branch still fail.

No trunk result at present as we are testing branch.

 

--- Stuart Henderson                                         2010-10-15 07:26:23

i'm not convinced this is failing differently due to any particular change.  i

managed to reproduce both the some-fail and all-fail results with the same

toolchain.  i'm still investigating this.

 

--- Stuart Henderson                                         2010-11-19 04:44:17

yeah, the test wasn't configured for hardware testing at all.  there was no ld

file being passed to the linker, so it was defaulting to address 0x0.  the fact

that it was passing anything seems to have been purely by chance.  force things

in to l1 and everything starts passing:

 

Index: asm-source.exp

===================================================================

--- asm-source.exp      (revision 4646)

+++ asm-source.exp      (working copy)

@@ -53,6 +53,7 @@

     }

     "bfin-*-elf" {

         set asm-arch bfin

+       append link-flags " -Ttext=0xFFA00000"

        set debug-flags "-g"

     }

     "bfin-*-uclinux" {

 

--- Stuart Henderson                                         2010-11-22 09:54:57

fixed on trunk. r5014.

 

--- Mingquan Pan                                             2011-06-07 02:25:58

These asm-source.exp cases now are not run on trunk for both bf537 and bf561

board.

 

Running

/home/test/work/cruise/checkouts/toolchain/binutils-2.17/gdb/testsuite/gdb.asm/asm-source.exp

...

UNTESTED: gdb.asm/asm-source.exp: asm-source.exp

 

--- Stuart Henderson                                         2011-06-22 11:57:04

Test stopped working due to the qemu rejig which added multilib_flags.  the test

won't run if these exist.  it should run fine with our multilib_flags, so i've

disabled this check for bfin.  test runs and passes now.

 

Set to fixed on trunk.

 

--- Mingquan Pan                                             2011-10-13 02:35:18

Seen from the latest jtag testing on bf537 board, these cases passed now.

 

Running binutils-2.17/gdb/testsuite/gdb.asm/asm-source.exp ...

PASS: gdb.asm/asm-source.exp: f at main

PASS: gdb.asm/asm-source.exp: next over macro

PASS: gdb.asm/asm-source.exp: step into foo2

PASS: gdb.asm/asm-source.exp: info target

PASS: gdb.asm/asm-source.exp: info symbol

PASS: gdb.asm/asm-source.exp: list

PASS: gdb.asm/asm-source.exp: search

PASS: gdb.asm/asm-source.exp: f in foo2

PASS: gdb.asm/asm-source.exp: n in foo2

PASS: gdb.asm/asm-source.exp: bt ALL in foo2

PASS: gdb.asm/asm-source.exp: bt 2 in foo2

PASS: gdb.asm/asm-source.exp: s 2

PASS: gdb.asm/asm-source.exp: n 2

PASS: gdb.asm/asm-source.exp: bt 3 in foo3

PASS: gdb.asm/asm-source.exp: info source asmsrc1.s

PASS: gdb.asm/asm-source.exp: finish from foo3

PASS: gdb.asm/asm-source.exp: info source asmsrc2.s

PASS: gdb.asm/asm-source.exp: info sources

PASS: gdb.asm/asm-source.exp: info line

PASS: gdb.asm/asm-source.exp: next over foo3

PASS: gdb.asm/asm-source.exp: return from foo2

PASS: gdb.asm/asm-source.exp: look at global variable

PASS: gdb.asm/asm-source.exp: x/i &globalvar

PASS: gdb.asm/asm-source.exp: disassem &globalvar &globalvar+1

PASS: gdb.asm/asm-source.exp: look at static variable

PASS: gdb.asm/asm-source.exp: x/i &staticvar

PASS: gdb.asm/asm-source.exp: disassem &staticvar &staticvar+1

PASS: gdb.asm/asm-source.exp: look at static function

 

Since jtag test on bf561 is stuck now as filed in bug [6800], still leave this

bug open till it is to be checked on bf561 board lator.

 

--- Mingquan Pan                                             2011-10-16 23:26:54

These cases also pass on bf561-ezkit too.

 

Running binutils-2.17/gdb/testsuite/gdb.asm/asm-source.exp ...

PASS: gdb.asm/asm-source.exp: f at main

PASS: gdb.asm/asm-source.exp: next over macro

PASS: gdb.asm/asm-source.exp: step into foo2

PASS: gdb.asm/asm-source.exp: info target

PASS: gdb.asm/asm-source.exp: info symbol

PASS: gdb.asm/asm-source.exp: list

PASS: gdb.asm/asm-source.exp: search

PASS: gdb.asm/asm-source.exp: f in foo2

PASS: gdb.asm/asm-source.exp: n in foo2

PASS: gdb.asm/asm-source.exp: bt ALL in foo2

PASS: gdb.asm/asm-source.exp: bt 2 in foo2

PASS: gdb.asm/asm-source.exp: s 2

PASS: gdb.asm/asm-source.exp: n 2

PASS: gdb.asm/asm-source.exp: bt 3 in foo3

PASS: gdb.asm/asm-source.exp: info source asmsrc1.s

PASS: gdb.asm/asm-source.exp: finish from foo3

PASS: gdb.asm/asm-source.exp: info source asmsrc2.s

PASS: gdb.asm/asm-source.exp: info sources

PASS: gdb.asm/asm-source.exp: info line

PASS: gdb.asm/asm-source.exp: next over foo3

PASS: gdb.asm/asm-source.exp: return from foo2

PASS: gdb.asm/asm-source.exp: look at global variable

PASS: gdb.asm/asm-source.exp: x/i &globalvar

PASS: gdb.asm/asm-source.exp: disassem &globalvar &globalvar+1

PASS: gdb.asm/asm-source.exp: look at static variable

PASS: gdb.asm/asm-source.exp: x/i &staticvar

PASS: gdb.asm/asm-source.exp: disassem &staticvar &staticvar+1

PASS: gdb.asm/asm-source.exp: look at static function

 

So close.

 

 

 

    Files

    Changes

    Commits

    Dependencies

    Duplicates

    Associations

    Tags

 

File Name     File Type     File Size     Posted By

gdb-2.17.log_Jun_23.tar.gz    application/x-gzip    322249    Vivi Li

gdb-2.17.log_Jun_24.tar.gz    application/x-gzip    321331    Vivi Li

build_log.4646vs4650    application/octet-stream    2620    Vivi Li

Outcomes