headers
Garrett Cooper | 1 Dec 2009 01:00
Picon
Gravatar

Re: [PATCH] syscalls: fix some failure on arch X86_64

On Sat, Nov 28, 2009 at 5:22 PM, Wei Yongjun <yjwei@...> wrote:
>
>
> Garrett Cooper wrote:
>> On Fri, Nov 27, 2009 at 2:07 AM, liubo <liubo2009@...> wrote:
>>
>>> Hi, Garrett
>>>
>>> On 11/27/2009 04:50 PM, Garrett Cooper wrote:
>>>
>>> On Thu, Nov 26, 2009 at 10:49 PM, Garrett Cooper
<yanegomi@...> wrote:
>>>
>>>
>>> On Thu, Nov 26, 2009 at 9:33 PM, liubo <liubo2009@...> wrote:
>>>
>>>
>>> Hi, Garrett
>>> On 11/26/2009 07:11 PM, Garrett Cooper wrote:
>>>
>>> On Mon, Nov 16, 2009 at 12:53 AM, liubo <liubo2009@...> wrote:
>>>
>>>
>>> Hi, Subrata,
>>> On 11/16/2009 04:13 PM, Subrata Modak wrote:
>>>
>>> Liubo,
>>>
>>> Are you sending an updated patch post this discussion ?
>>>
(Continue reading)

Permalink | Reply | 1 comment |
headers
Garrett Cooper | 1 Dec 2009 05:24
Picon
Gravatar

Re: [PATCH] clock_settime03: improve message about fail

On Wed, Nov 25, 2009 at 10:08 AM, Mike Frysinger <vapier@...> wrote:
> thanks, look good
> -mike

Just committed with other various cleanup items -- thanks!
-Garrett

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing. 
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
Permalink | Reply | 0 comments |
headers
Mitani | 1 Dec 2009 06:18
Picon
Favicon

build failure with ltp-2009-12-01 cvs

Hi,

I tried to "make" with "ltp-2009-12-01.tar.gz" version.
I tried it in systems as follows:
 (1) kernel 2.6.18-164.el5 - architecture x86 
 (2) kernel 2.6.18-164.el5 - architecture x86_64

But both failed with following message:

------------
[...]
make[3]: Entering directory
`/home/LTP/ltp-2009-12-01/testcases/kernel/containers'
make -C libclone -f
"/home/LTP/ltp-2009-12-01/testcases/kernel/containers/libclone/Makefil
e" all
make[4]: Entering directory
`/home/LTP/ltp-2009-12-01/testcases/kernel/containers/libclone
'
gcc -g -O2 -g -O2 -fno-strict-aliasing -pipe -Wall
-I/home/LTP/ltp-2009-12-01/testcases/k
ernel/include -I../../../../include -I../../../../include  -c -o libclone.o
libclone.c
In file included from libclone.h:31,
                 from libclone.c:17:
/home/LTP/ltp-2009-12-01/testcases/kernel/include/linux_syscall_numbers.h:24
: error: weak
declaration of 'syscall_cleanup_stub' must be public
make[4]: *** [libclone.o] Error 1
make[4]: Leaving directory
(Continue reading)

Permalink | Reply | 3 comments |
headers
Garrett Cooper | 1 Dec 2009 11:11
Picon
Gravatar

Re: [PATCH] tst_res: save errno at the begin, otherwise it may be rewritten

On Wed, Nov 25, 2009 at 10:09 AM, Mike Frysinger <vapier@...> wrote:
> On Wednesday 25 November 2009 08:36:35 Andrew Vagin wrote:
>> it may do fprinf for example.
>
> it seems a little paranoid, but i dont see any harm in the change

Committed -- thanks!
-Garrett

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing. 
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
Permalink | Reply | 0 comments |
headers
Garrett Cooper | 1 Dec 2009 11:18
Picon
Gravatar

Re: build failure with ltp-2009-12-01 cvs

On Mon, Nov 30, 2009 at 9:18 PM, Mitani <mitani@...> wrote:
> Hi,
>
> I tried to "make" with "ltp-2009-12-01.tar.gz" version.
> I tried it in systems as follows:
>  (1) kernel 2.6.18-164.el5 - architecture x86
>  (2) kernel 2.6.18-164.el5 - architecture x86_64
>
> But both failed with following message:
>
> ------------
> [...]
> make[3]: Entering directory
> `/home/LTP/ltp-2009-12-01/testcases/kernel/containers'
> make -C libclone -f
> "/home/LTP/ltp-2009-12-01/testcases/kernel/containers/libclone/Makefil
> e" all
> make[4]: Entering directory
> `/home/LTP/ltp-2009-12-01/testcases/kernel/containers/libclone
> '
> gcc -g -O2 -g -O2 -fno-strict-aliasing -pipe -Wall
> -I/home/LTP/ltp-2009-12-01/testcases/k
> ernel/include -I../../../../include -I../../../../include  -c -o libclone.o
> libclone.c
> In file included from libclone.h:31,
>                 from libclone.c:17:
> /home/LTP/ltp-2009-12-01/testcases/kernel/include/linux_syscall_numbers.h:24
> : error: weak
> declaration of 'syscall_cleanup_stub' must be public
> make[4]: *** [libclone.o] Error 1
(Continue reading)

Permalink | Reply | 2 comments |
headers
Garrett Cooper | 1 Dec 2009 11:22
Picon
Gravatar

Re: Running LTP on Android Emulator

On Sun, Nov 29, 2009 at 10:39 PM, Biswajit Mohanty
<biswajit.mohanty@...> wrote:
> Hi
> Can anyone help me on running the LTP on android amulator.
> If any pointer to any site or document, please share.

    Please, go read a recent copy of the archives; you'll find your
answer there.
Thanks,
-Garrett

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing. 
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
Permalink | Reply | 0 comments |
headers
Jiri Palecek | 1 Dec 2009 23:16
Picon

[PATCH] Fix a wrong (reversed) success condition in echo01

Hello,

this is the first of another series of small ltp fixes. I'm sorry I wasn't able to send them earlier.

This patch reverses the success check in echo01, obviously the checksums being equal is the right outcome.

Regards
    Jiri Palecek

Signed-off-by: Jiri Palecek <jpalecek@...>
---
 testcases/network/tcp_cmds/echo/echo01 |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/testcases/network/tcp_cmds/echo/echo01 b/testcases/network/tcp_cmds/echo/echo01
index eb798ae..2bc4621 100755
--- a/testcases/network/tcp_cmds/echo/echo01
+++ b/testcases/network/tcp_cmds/echo/echo01
 <at>  <at>  -95,7 +95,7  <at>  <at>  do_test()
         while [ $NUM -lt $NUMPROCESSES ]
         do
             csum2=$(sum "$TCtmp/${ECHOFILE}${NUM}" | awk '{print $1}')
-            if [ "$csum1" = "$csum2" ]; then
+            if ! [ "$csum1" = "$csum2" ]; then
                 end_testcase "Checksums differ in loop $TST_COUNT"
             fi
             NUM=$(( $NUM + 1 ))
--

-- 
1.6.4.3
(Continue reading)

Permalink | Reply | 0 comments |
headers
Jiri Palecek | 1 Dec 2009 23:22
Picon

[PATCH] Don't fail the ssh tests if the home directory we are creating already exists (eg. when it was created by adduser)

Hello,

this patch silences the error message "mkdir: directory already exists" in ssh tests, caused by the code
that handles home directory creation in case adduser didn't do it. Unfortunately, the same code failed
when adduser did create the directory.

Regards
    Jiri Palecek

Signed-off-by: Jiri Palecek <jpalecek@...>
---
 testcases/network/tcp_cmds/ssh/ssh01 |    2 +-
 testcases/network/tcp_cmds/ssh/ssh02 |    2 +-
 testcases/network/tcp_cmds/ssh/ssh03 |    2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/testcases/network/tcp_cmds/ssh/ssh01 b/testcases/network/tcp_cmds/ssh/ssh01
index 657449a..718f198 100755
--- a/testcases/network/tcp_cmds/ssh/ssh01
+++ b/testcases/network/tcp_cmds/ssh/ssh01
 <at>  <at>  -57,7 +57,7  <at>  <at>  do_setup()
     # in adduser)
     USER_UID=$(id -u $TEST_USER)
     USER_GID=$(id -g $TEST_USER)
-    if ! mkdir "$TEST_USER_HOMEDIR"; then
+    if ! mkdir -p "$TEST_USER_HOMEDIR"; then
         end_testcase "Failed to create $TEST_USER_HOMEDIR"
     fi
     chown -Rf $USER_UID.$USER_GID "$TEST_USER_HOMEDIR"
diff --git a/testcases/network/tcp_cmds/ssh/ssh02 b/testcases/network/tcp_cmds/ssh/ssh02
(Continue reading)

Permalink | Reply | 0 comments |
headers
Jiri Palecek | 1 Dec 2009 23:32
Picon

[PATCH] Fix a bug in awk code in rsh01

Hello,

this patch fixes the following problem with awk code in rsh01. It caused the test to fail.

 The END label is executed always, even in the event of
 exit. Therefore, the "exit 1" statement was executed even when the
 line matched (as a consequence of "exit 0") and awk always returned
 1 (ie. failure).

Regards
    Jiri Palecek

Signed-off-by: Jiri Palecek <jpalecek@...>
---
 testcases/network/tcp_cmds/rsh/rsh01 |    6 +++---
 1 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/testcases/network/tcp_cmds/rsh/rsh01 b/testcases/network/tcp_cmds/rsh/rsh01
index e87a6a0..0a2b58f 100755
--- a/testcases/network/tcp_cmds/rsh/rsh01
+++ b/testcases/network/tcp_cmds/rsh/rsh01
 <at>  <at>  -72,9 +72,9  <at>  <at>  do_test()
             # gcooper <at> orangebox ~ $ ls -l /etc/hosts         
             # -rw-r--r-- 1 root root 463 Jul  5 09:26 /etc/hosts
             #
-            echo "$OUT" | \
-                awk 'NF == 9 && $NF == "/etc/hosts" { exit 0 } END { exit 1 }' \
-                >& /dev/null
+            echo "$OUT" |
+                awk 'BEGIN { RET=1 } NF == 9 && $NF == "/etc/hosts" { RET=0 } END { exit RET }' \
(Continue reading)

Permalink | Reply | 0 comments |
headers
Jiri Palecek | 1 Dec 2009 23:30
Picon

[PATCH] Fixes of some errors discovered by "set -u"

Hello,

this patch fixes the following problems:

 - read of $TEST_USER before it's set in ftp03

 - remnants of the $COMMAND variable which isn't really there anymore
   in netstat01

 - probably a typo in rsh01 ($COUNT vs. $TST_COUNT)

 - $2 has to be escaped twice in rwho01, to appear in the awk script

 - when no interface is found tcpdump01, $IF is unset. Don't spit out
   an error in that case

Regards
    Jiri Palecek

Signed-off-by: Jiri Palecek <jpalecek@...>
---
 testcases/network/tcp_cmds/ftp/ftp03         |    2 +-
 testcases/network/tcp_cmds/netstat/netstat01 |    2 +-
 testcases/network/tcp_cmds/rsh/rsh01         |    2 +-
 testcases/network/tcp_cmds/rwho/rwho01       |    8 ++++----
 testcases/network/tcp_cmds/tcpdump/tcpdump01 |    2 +-
 5 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/testcases/network/tcp_cmds/ftp/ftp03 b/testcases/network/tcp_cmds/ftp/ftp03
index 24e30e9..e85a985 100755
(Continue reading)

Permalink | Reply | 0 comments |

Gmane