Re: Increase page fault rate by prezeroing V1 [3/3]: Altix SN2 BTEZeroing
From: Christoph Lameter
Date: Wed Dec 22 2004 - 14:59:53 EST
I have done some additional tests with a 128 cpu SMP machine and it shows
that the bte slows things down during memory benchmarking by about 10-20%
although its causing less load when the system is not under high stress.
So its not always win and I may drop bte support in a future version. Can
we talk off list about this since this is mostly an SGI thing?
On Wed, 22 Dec 2004, Robin Holt wrote:
> We still need to talk. This is a much smaller patch, which I like. The
> problem I see in my 30 second review is you are doing things per-cpu
> when they really need to be done per-node. It is very likely that
> there will be M-Bricks in the system (cranberry2 has one if you want
> to test your code out there or you can take any altix and disable the
> cpus on a C-Brick). With M-Bricks, you will essentially limit
> yourself to one zero operation per controlling node instead of one
> per node.
>
> I think the easy answer is to not have the structure allocated
> within bte_copy(), but rather within bte_start_zero and passed
> in as the notification address.
>
> Give me a call sometime today (Wed. I am in the office from about
> 10:00 CDT until around 4:00 CDT) Maybe we can get this straightened
> out quickly. If you are not calling from the office, email me with
> other arrangements.
>
> Thanks,
> Robin
>
> On Tue, Dec 21, 2004 at 11:57:57AM -0800, Christoph Lameter wrote:
> > o Use the Block Transfer Engine in the Altix SN2 SHub for background zeroing
> >
> > Index: linux-2.6.9/arch/ia64/sn/kernel/bte.c
> > ===================================================================
> > --- linux-2.6.9.orig/arch/ia64/sn/kernel/bte.c 2004-12-17 14:40:10.000000000 -0800
> > +++ linux-2.6.9/arch/ia64/sn/kernel/bte.c 2004-12-21 11:03:49.000000000 -0800
> > @@ -4,6 +4,8 @@
> > * for more details.
> > *
> > * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved.
> > + *
> > + * Support for zeroing pages, Christoph Lameter, SGI, December 2004.
> > */
> >
> > #include <linux/config.h>
> > @@ -20,6 +22,8 @@
> > #include <linux/bootmem.h>
> > #include <linux/string.h>
> > #include <linux/sched.h>
> > +#include <linux/mm.h>
> > +#include <linux/scrub.h>
> >
> > #include <asm/sn/bte.h>
> >
> > @@ -30,7 +34,11 @@
> > /* two interfaces on two btes */
> > #define MAX_INTERFACES_TO_TRY 4
> >
> > -static struct bteinfo_s *bte_if_on_node(nasid_t nasid, int interface)
> > +DEFINE_PER_CPU(u64 *, bte_zero_notify);
> > +
> > +#define bte_zero_notify __get_cpu_var(bte_zero_notify)
> > +
> > +static inline struct bteinfo_s *bte_if_on_node(nasid_t nasid, int interface)
> > {
> > nodepda_t *tmp_nodepda;
> >
> > @@ -132,7 +140,6 @@
> > if (bte == NULL) {
> > continue;
> > }
> > -
> > if (spin_trylock(&bte->spinlock)) {
> > if (!(*bte->most_rcnt_na & BTE_WORD_AVAILABLE) ||
> > (BTE_LNSTAT_LOAD(bte) & BTE_ACTIVE)) {
> > @@ -157,7 +164,7 @@
> > }
> > } while (1);
> >
> > - if (notification == NULL) {
> > + if (notification == NULL || (mode & BTE_NOTIFY_AND_GET_POINTER)) {
> > /* User does not want to be notified. */
> > bte->most_rcnt_na = &bte->notify;
> > } else {
> > @@ -192,6 +199,8 @@
> >
> > itc_end = ia64_get_itc() + (40000000 * local_cpu_data->cyc_per_usec);
> >
> > + if (mode & BTE_NOTIFY_AND_GET_POINTER)
> > + *(u64 volatile **)(notification) = &bte->notify;
> > spin_unlock_irqrestore(&bte->spinlock, irq_flags);
> >
> > if (notification != NULL) {
> > @@ -449,5 +458,31 @@
> > mynodepda->bte_if[i].cleanup_active = 0;
> > mynodepda->bte_if[i].bh_error = 0;
> > }
> > +}
> > +
> > +static int bte_check_bzero(void)
> > +{
> > + return *bte_zero_notify != BTE_WORD_BUSY;
> > +}
> > +
> > +static int bte_start_bzero(void *p, unsigned long len)
> > +{
> > + /* Check limitations.
> > + 1. System must be running (weird things happen during bootup)
> > + 2. Size >64KB. Smaller requests cause too much bte traffic
> > + */
> > + if (len >= BTE_MAX_XFER || len < 60000 || system_state != SYSTEM_RUNNING)
> > + return EINVAL;
> > +
> > + return bte_zero(ia64_tpa(p), len, BTE_NOTIFY_AND_GET_POINTER, &bte_zero_notify);
> > +}
> > +
> > +static struct zero_driver bte_bzero = {
> > + .start = bte_start_bzero,
> > + .check = bte_check_bzero,
> > + .rate = 500000000 /* 500 MB /sec */
> > +};
> >
> > +void sn_bte_bzero_init(void) {
> > + register_zero_driver(&bte_bzero);
> > }
> > Index: linux-2.6.9/arch/ia64/sn/kernel/setup.c
> > ===================================================================
> > --- linux-2.6.9.orig/arch/ia64/sn/kernel/setup.c 2004-12-17 14:40:10.000000000 -0800
> > +++ linux-2.6.9/arch/ia64/sn/kernel/setup.c 2004-12-21 11:02:35.000000000 -0800
> > @@ -243,6 +243,7 @@
> > int pxm;
> > int major = sn_sal_rev_major(), minor = sn_sal_rev_minor();
> > extern void sn_cpu_init(void);
> > + extern void sn_bte_bzero_init(void);
> >
> > /*
> > * If the generic code has enabled vga console support - lets
> > @@ -333,6 +334,7 @@
> > screen_info = sn_screen_info;
> >
> > sn_timer_init();
> > + sn_bte_bzero_init();
> > }
> >
> > /**
> > Index: linux-2.6.9/include/asm-ia64/sn/bte.h
> > ===================================================================
> > --- linux-2.6.9.orig/include/asm-ia64/sn/bte.h 2004-12-17 14:40:16.000000000 -0800
> > +++ linux-2.6.9/include/asm-ia64/sn/bte.h 2004-12-21 11:02:35.000000000 -0800
> > @@ -48,6 +48,8 @@
> > #define BTE_ZERO_FILL (BTE_NOTIFY | IBCT_ZFIL_MODE)
> > /* Use a reserved bit to let the caller specify a wait for any BTE */
> > #define BTE_WACQUIRE (0x4000)
> > +/* Return the pointer to the notification cacheline to the user */
> > +#define BTE_NOTIFY_AND_GET_POINTER (0x8000)
> > /* Use the BTE on the node with the destination memory */
> > #define BTE_USE_DEST (BTE_WACQUIRE << 1)
> > /* Use any available BTE interface on any node for the transfer */
>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/