Re: [PATCH glibc] Hurd: Fix definition of `PAGE_COPY_THRESHOLD'

Quoting Samuel Thibault (2014-12-10 00:07:28) > Hello, > > Justus Winter, le Tue 09 Dec 2014 18:06:18 +0100, a écrit : > > Previously, `PAGE_THRESHOLD' was defined instead of > > `PAGE_COPY_THRESHOLD', preventing the optimization of memory copy > > operations using `vm_copy'. > > Actually this cu

Re: [PATCH hurd 3/5] proc: implement `proc_make_task_namespace'

Justus Winter, le Thu 13 Nov 2014 13:26:17 +0100, a écrit : > * proc/proc.h (struct proc): Add field `p_task_namespace'. > * proc/mgt.c (S_proc_child): Propagate `p_task_namespace' to child. > (allocate_proc): Initialize `p_task_namespace'. > (namespace_terminate): New function. > (process_has_exit

Re: [PATCH hurd 4/5] hurd: add `proc_make_task_namespace'

Justus Winter, le Thu 13 Nov 2014 13:26:18 +0100, a écrit : > Add a new RPC to the process protocol to create task namespaces. > These can be used by an unprivileged process to claims the > responsibility to manage all tasks in this namespace. Any task > created in this namespace will automaticall

Re: [PATCH hurd 2/5] proc: register for new task notifications

Justus Winter, le Thu 13 Nov 2014 13:26:16 +0100, a écrit : > * proc/Makefile (MIGSTUBS): Add `gnumachServer.o'. > * proc/main.c (message_demuxer): Handle the `task_notify' protocol. > (main): Register for new task notificatinos. > * proc/mgt.c (S_mach_notify_new_task): Add server function. Ack.

Re: [PATCH hurd 1/5] Makeconf: handle the task_notify protocol

Justus Winter, le Thu 13 Nov 2014 13:26:15 +0100, a écrit : > * Makeconf (mach_defs_names): Add `task_notify'. Ack. > --- > Makeconf | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/Makeconf b/Makeconf > index f0d3fe3..5439913 100644 > --- a/Makeconf > +++ b/Makeconf > @@ -577,6 +577,7

Re: Please help me test this glibc patch

Samuel Thibault, le Tue 09 Dec 2014 20:02:59 +0100, a écrit : > My native laptop: > >size iterations gcc_memcpy memcpy vm_copy >40968388608 8.698.6918.83 >81924194304 8.528.529.39

Re: [PATCH glibc] Hurd: Fix definition of `PAGE_COPY_THRESHOLD'

Hello, Justus Winter, le Tue 09 Dec 2014 18:06:18 +0100, a écrit : > Previously, `PAGE_THRESHOLD' was defined instead of > `PAGE_COPY_THRESHOLD', preventing the optimization of memory copy > operations using `vm_copy'. Actually this currently doesn't make any difference since the Makefile rules e

Re: Please help me test this glibc patch

Justus Winter, le Tue 09 Dec 2014 18:06:17 +0100, a écrit : > teythoon@darnassus:~/build/time-memcpy$ bash driver.bash >size iterations gcc_memcpy memcpy vm_copy >4096 16777216 6.096.0715.67 >81928388608 7.026.29

Re: Please help me test this glibc patch

Quoting Samuel Thibault (2014-12-09 19:10:28) > Justus Winter, le Tue 09 Dec 2014 18:06:17 +0100, a écrit : > > teythoon@darnassus:~/build/time-memcpy$ bash driver.bash > > It's probably worth trying it on real hardware too. On darnassus there > is some cost related with the kvm virtualization. I

Re: Please help me test this glibc patch

Justus Winter, le Tue 09 Dec 2014 18:06:17 +0100, a écrit : > teythoon@darnassus:~/build/time-memcpy$ bash driver.bash It's probably worth trying it on real hardware too. On darnassus there is some cost related with the kvm virtualization. Samuel

[PATCH glibc] Hurd: Fix definition of `PAGE_COPY_THRESHOLD'

Previously, `PAGE_THRESHOLD' was defined instead of `PAGE_COPY_THRESHOLD', preventing the optimization of memory copy operations using `vm_copy'. Also, increase the threshold to 4 pages. XXX Tune this. * sysdeps/mach/pagecopy.h (PAGE_THRESHOLD): Rename to `PAGE_COPY_THRESHOLD'. Raise threshold t

Please help me test this glibc patch

Hello, please help me test this glibc patch. It could bring a significant speedup to certain workloads, namely those that copy lots of memory in chunks. I wrote a micro-benchmark for this: teythoon@darnassus:~/build/time-memcpy$ bash driver.bash size iterations gcc_memcpy memcpy