Description of problem: There was a request for GDB (Bug 253117) to support hardware watchpoints in RHEL-5.2. But it still does not (reliably) work as it is broken even upstream. No one knows if it is a kernel or hardware problem. Version-Release number of selected component (if applicable): Not found the right kernel/RHEL-5 combo machine in RH. All the Fedora kernels (such as kernel-2.6.25-0.121.rc5.git4.fc9.ppc64). All the upstream kernels. How reproducible: 99% of runs with enough (>16) threads (set by THREADS). Steps to Reproduce: 1. wget -O ppc-dabr-race.c 'http://sources.redhat.com/cgi-bin/cvsweb.cgi/~checkout~/tests/ptrace-tests/tests/ppc-dabr-race.c?cvsroot=systemtap' 2. gcc -o ppc-dabr-race ppc-dabr-race.c -Wall -ggdb2 -m32 -D_GNU_SOURCE -pthread 3. ./ppc-dabr-race ;echo $? Actual results: 1 Tested on: ppc03.englab.brq.redhat.com kernel-2.6.25-0.121.rc5.git4.fc9.ppc64 Dual: cpu : POWER5 (gr) clock : 1502.496000MHz revision : 2.2 (pvr 003a 0202) Expected results: 0 Additional info: Compile with -DDEBUG to see why it fails. It was discussed upstream with IBM, mail threads at: http://ozlabs.org/pipermail/linuxppc-dev/2007-November/thread.html#46951 http://ozlabs.org/pipermail/linuxppc-dev/2008-March/thread.html#52788 -m64 has the same error as -m32. Return code 77 means the kernel/CPU combo does not support hardware watchpoints.
See upstream commit a2ceff5e555e664751bc653a4d9b133efa18c742, seems to have fixed it.
Created attachment 299586 [details] Upstream fix.
This request was evaluated by Red Hat Product Management for inclusion, but this component is not scheduled to be updated in the current Red Hat Enterprise Linux release. If you would like this request to be reviewed for the next minor release, ask your support representative to set the next rhel-x.y flag to "?".
Unfortunately the previous automated notification about the non-inclusion of this request in Red Hat Enterprise Linux 5.3 used the wrong text template. It should have read: this request has been reviewed by Product Management and is not planned for inclusion in the current minor release of Red Hat Enterprise Linux. If you would like this request to be reviewed for the next minor release, ask your support representative to set the next rhel-x.y flag to "?" or raise an exception.
I tested this on the *.128 kernel for RHEL5.3 and the test case passed. Kernel version: [root@ibm-p5-185-01 ~]# uname -a Linux ibm-p5-185-01.rhts.bos.redhat.com 2.6.18-128.el5 #1 SMP Wed Dec 17 11:58:23 EST 2008 ppc64 ppc64 ppc64 GNU/Linux Test case results: [root@ibm-p5-185-01 ~]# ./ppc-dabr-race ;echo $? 0 [root@ibm-p5-185-01 ~]# If there are no objections, I would like to close this as fixed in the latest release. I'll wait until Friday, 2/20 for a response.
*** This bug has been marked as a duplicate of bug 444076 ***
As Bug 444076 is not public, it should (with no guarantees) be found in RHEL-5.3: linux-2.6-ppc64-missed-hw-breakpoints-across-multiple-threads.patch - [ppc64] missed hw breakpoints across multiple threads (Brad Peters ) [444076]