RE: Yes I have dental and medical lists
Spesial Med Package 395.00 Complete Physician Database 788,114 in total * 17,482 emails Coverage in many different areas of medicine such as Endocrinology, Pathology, Urology, Neurology, Plastic Surgery, Psychiatry, Cardiology and much more Also included are the 6 bonus listed below: 1=7145 Hospitals*23,000 Administrators with Ph/fax and postal address. 2=Licensed Clinic/Dentist 597,959 Total Records*6,494 Emails*6,000 Fax Numbers. 3=US Pharmaceutical Company 47,000 employees in high-ranking positions Executive's Directory Names and email addresses of 4=Chiropractors*108,421 with phone, 3,414 emails and 6,553 fax numbers. 5=Acupuncture Database 23,988 records 1,826 emails 6=Alternative Medicine Clinics and Practitioners Database There are 1,141,602 total records with 29,320 emails and 38.935 fax numbers. For a datacard reply here war...@bestbizlists.info contact rem...@bestbizlists.info for record exclusion.
[PATCH] utrace: don't set -ops = utrace_detached_ops lockless
On 12/04, Roland McGrath wrote:
I think the problem is clear now.
Ok. We should probably move this discussion to utrace-devel.
Yes, I didn't notice we discuss this offlist...
I forgot that there is another issue (iirc a bit discussed too).
finish_callback_report() sets -ops = utrace_detached_ops lockless!
You'll have to remind me why this is a problem.
Re: [PATCH 85] ptrace_attach_task: rely on utrace_barrier(), don't
check -ops
https://www.redhat.com/archives/utrace-devel/2009-October/msg00180.html
We already discussed this, but forgot to finish.
Do you agree with the patch?
--
[PATCH] utrace: don't set -ops = utrace_detached_ops lockless
finish_callback_report() changes -ops lockless. Imho this is not
right in general, the state of !EXIT_DEAD tracee must be stable
under utrace-lock.
And this can confuse ptrace_reuse_engine()-utrace_barrier() logic.
utrace_barrier() can race with reporting loop and return 0 while
engine was already detached or in the middle of detach.
See also https://www.redhat.com/archives/utrace-devel/2009-October/msg00180.html
Signed-off-by: Oleg Nesterov o...@redhat.com
---
kernel/utrace.c |8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
--- UTRACE-PTRACE/kernel/utrace.c~DONT_CHANGE_OPS_LOCKLESS 2009-11-24
17:20:33.0 +0100
+++ UTRACE-PTRACE/kernel/utrace.c 2009-12-04 17:10:37.0 +0100
@@ -1390,11 +1390,15 @@ static inline void finish_callback_repor
struct utrace_engine *engine,
enum utrace_resume_action action)
{
+ if (action == UTRACE_DETACH) {
+ spin_lock(utrace-lock);
+ engine-ops = utrace_detached_ops;
+ spin_unlock(utrace-lock);
+ }
/*
* If utrace_control() was used, treat that like UTRACE_DETACH here.
*/
- if (action == UTRACE_DETACH || engine-ops == utrace_detached_ops) {
- engine-ops = utrace_detached_ops;
+ if (engine-ops == utrace_detached_ops) {
report-detaches = true;
return;
}
Re: [PATCH] utrace: don't set -ops = utrace_detached_ops lockless
Forgot to mention, I did a lot of testing on ppc machine and the
patch helps, finally I was able to reproduce the problem.
But I failed to install the kernel on Cai's machine, perhaps he
could test the patch too ;)
fighting with rhts machines is very tiresome and time consuming...
I'll send the promised patch which kills supress_sigtrap() tomorrow,
it needs testing/checking.
On 12/04, Oleg Nesterov wrote:
On 12/04, Roland McGrath wrote:
I think the problem is clear now.
Ok. We should probably move this discussion to utrace-devel.
Yes, I didn't notice we discuss this offlist...
I forgot that there is another issue (iirc a bit discussed too).
finish_callback_report() sets -ops = utrace_detached_ops lockless!
You'll have to remind me why this is a problem.
Re: [PATCH 85] ptrace_attach_task: rely on utrace_barrier(), don't
check -ops
https://www.redhat.com/archives/utrace-devel/2009-October/msg00180.html
We already discussed this, but forgot to finish.
Do you agree with the patch?
--
[PATCH] utrace: don't set -ops = utrace_detached_ops lockless
finish_callback_report() changes -ops lockless. Imho this is not
right in general, the state of !EXIT_DEAD tracee must be stable
under utrace-lock.
And this can confuse ptrace_reuse_engine()-utrace_barrier() logic.
utrace_barrier() can race with reporting loop and return 0 while
engine was already detached or in the middle of detach.
See also
https://www.redhat.com/archives/utrace-devel/2009-October/msg00180.html
Signed-off-by: Oleg Nesterov o...@redhat.com
---
kernel/utrace.c |8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
--- UTRACE-PTRACE/kernel/utrace.c~DONT_CHANGE_OPS_LOCKLESS2009-11-24
17:20:33.0 +0100
+++ UTRACE-PTRACE/kernel/utrace.c 2009-12-04 17:10:37.0 +0100
@@ -1390,11 +1390,15 @@ static inline void finish_callback_repor
struct utrace_engine *engine,
enum utrace_resume_action action)
{
+ if (action == UTRACE_DETACH) {
+ spin_lock(utrace-lock);
+ engine-ops = utrace_detached_ops;
+ spin_unlock(utrace-lock);
+ }
/*
* If utrace_control() was used, treat that like UTRACE_DETACH here.
*/
- if (action == UTRACE_DETACH || engine-ops == utrace_detached_ops) {
- engine-ops = utrace_detached_ops;
+ if (engine-ops == utrace_detached_ops) {
report-detaches = true;
return;
}
Re: [PATCH] utrace: don't set -ops = utrace_detached_ops lockless
I forgot that there is another issue (iirc a bit discussed too). finish_callback_report() sets -ops = utrace_detached_ops lockless! You'll have to remind me why this is a problem. Re: [PATCH 85] ptrace_attach_task: rely on utrace_barrier(), don't check -ops https://www.redhat.com/archives/utrace-devel/2009-October/msg00180.html We already discussed this, but forgot to finish. Ah, yes. I had that message still sitting in my folder to think about again and reply. Do you agree with the patch? I think so, yes. It could use some more comments about the importance of the lock. I added a comment and merged it in. Thanks, Roland
