On Fri, Jul 24, 2015 at 09:52:01PM -0700, Andy Lutomirski wrote:
I see your wide terminal and raise you a complete rewrite of that
function. Sigh, why did I assume the old code was the right way to do
it?
That's a mostly wrong assumption, as experience proves.
Hah¸ we both missed it. This
On Wed, Jul 22, 2015 at 12:23:46PM -0700, Andy Lutomirski wrote:
modify_ldt has questionable locking and does not synchronize
threads. Improve it: redesign the locking and synchronize all
threads' LDTs using an IPI on all modifications.
This will dramatically slow down modify_ldt in
On Wed, Jul 22, 2015 at 12:23:46PM -0700, Andy Lutomirski wrote:
modify_ldt has questionable locking and does not synchronize
threads. Improve it: redesign the locking and synchronize all
threads' LDTs using an IPI on all modifications.
This will dramatically slow down modify_ldt in
On 07/22/2015 06:20 PM, Boris Ostrovsky wrote:
On 07/22/2015 03:23 PM, Andy Lutomirski wrote:
+error = -ENOMEM;
+new_ldt = alloc_ldt_struct(newsize);
+if (!new_ldt)
goto out_unlock;
-}
-fill_ldt(ldt, ldt_info);
-if (oldmode)
-ldt.avl = 0;
+if
On Fri, Jul 24, 2015 at 8:29 AM, Borislav Petkov b...@alien8.de wrote:
Let it stick out:
if (!cpumask_equal(mm_cpumask(current_mm),
cpumask_of(smp_processor_id(
smp_call_function(flush_ldt, current_mm, 1);
I see your wide terminal and raise you a complete rewrite
On Fri, Jul 24, 2015 at 9:13 PM, Boris Ostrovsky
boris.ostrov...@oracle.com wrote:
On 07/22/2015 06:20 PM, Boris Ostrovsky wrote:
On 07/22/2015 03:23 PM, Andy Lutomirski wrote:
+error = -ENOMEM;
+new_ldt = alloc_ldt_struct(newsize);
+if (!new_ldt)
goto out_unlock;
modify_ldt has questionable locking and does not synchronize
threads. Improve it: redesign the locking and synchronize all
threads' LDTs using an IPI on all modifications.
This will dramatically slow down modify_ldt in multithreaded
programs, but there shouldn't be any multithreaded programs
On 07/22/2015 03:23 PM, Andy Lutomirski wrote:
modify_ldt has questionable locking and does not synchronize
threads. Improve it: redesign the locking and synchronize all
threads' LDTs using an IPI on all modifications.
This will dramatically slow down modify_ldt in multithreaded
programs, but