* Re: [PATCH] sn2 module (& other misc.) fixes
2003-08-04 21:14 [PATCH] sn2 module (& other misc.) fixes Jesse Barnes
@ 2003-08-04 21:54 ` Christoph Hellwig
2003-08-04 22:01 ` Jesse Barnes
` (6 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: Christoph Hellwig @ 2003-08-04 21:54 UTC (permalink / raw)
To: linux-ia64
On Mon, Aug 04, 2003 at 02:14:04PM -0700, Jesse Barnes wrote:
> nasid_t master_baseio_nasid = INVALID_NASID; /* This is the master base I/O nasid */
> +EXPORT_SYMBOL(master_nasid);
why is this exported?
> int num_base_io_scsi_ctlr = 2; /* used by syssgi */
> vertex_hdl_t base_io_scsi_ctlr_vhdl[NUM_BASE_IO_SCSI_CTLR];
> +EXPORT_SYMBOL(base_io_scsi_ctlr_vhdl);
Why do you want this export. Or rather why does this tasteless stuff
exist at all? (And the comment looks wrong, too..)
> partid_t sn_partid = -1;
> +EXPORT_SYMBOL(sn_partid);
> +
> char sn_system_serial_number_string[128];
> u64 sn_partition_serial_number;
> +EXPORT_SYMBOL(sn_system_serial_number_string);
> +EXPORT_SYMBOL(sn_partition_serial_number);
>
> short physical_node_map[MAX_PHYSNODE_ID];
> +EXPORT_SYMBOL(physical_node_map);
Again, what is all this for?
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [PATCH] sn2 module (& other misc.) fixes
2003-08-04 21:14 [PATCH] sn2 module (& other misc.) fixes Jesse Barnes
2003-08-04 21:54 ` Christoph Hellwig
@ 2003-08-04 22:01 ` Jesse Barnes
2003-08-04 22:04 ` Christoph Hellwig
` (5 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: Jesse Barnes @ 2003-08-04 22:01 UTC (permalink / raw)
To: linux-ia64
On Mon, Aug 04, 2003 at 10:54:23PM +0100, Christoph Hellwig wrote:
> On Mon, Aug 04, 2003 at 02:14:04PM -0700, Jesse Barnes wrote:
> > nasid_t master_baseio_nasid = INVALID_NASID; /* This is the master base I/O nasid */
> > +EXPORT_SYMBOL(master_nasid);
>
> why is this exported?
Because no one's ripped it out yet.
> > int num_base_io_scsi_ctlr = 2; /* used by syssgi */
> > vertex_hdl_t base_io_scsi_ctlr_vhdl[NUM_BASE_IO_SCSI_CTLR];
> > +EXPORT_SYMBOL(base_io_scsi_ctlr_vhdl);
>
> Why do you want this export. Or rather why does this tasteless stuff
> exist at all? (And the comment looks wrong, too..)
Same here.
> > partid_t sn_partid = -1;
> > +EXPORT_SYMBOL(sn_partid);
> > +
> > char sn_system_serial_number_string[128];
> > u64 sn_partition_serial_number;
> > +EXPORT_SYMBOL(sn_system_serial_number_string);
> > +EXPORT_SYMBOL(sn_partition_serial_number);
Modules.
> > short physical_node_map[MAX_PHYSNODE_ID];
> > +EXPORT_SYMBOL(physical_node_map);
>
> Again, what is all this for?
Modules.
Jesse
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [PATCH] sn2 module (& other misc.) fixes
2003-08-04 21:14 [PATCH] sn2 module (& other misc.) fixes Jesse Barnes
2003-08-04 21:54 ` Christoph Hellwig
2003-08-04 22:01 ` Jesse Barnes
@ 2003-08-04 22:04 ` Christoph Hellwig
2003-08-04 23:31 ` David Mosberger
` (4 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: Christoph Hellwig @ 2003-08-04 22:04 UTC (permalink / raw)
To: linux-ia64
On Mon, Aug 04, 2003 at 03:01:39PM -0700, Jesse Barnes wrote:
> On Mon, Aug 04, 2003 at 10:54:23PM +0100, Christoph Hellwig wrote:
> > On Mon, Aug 04, 2003 at 02:14:04PM -0700, Jesse Barnes wrote:
> > > nasid_t master_baseio_nasid = INVALID_NASID; /* This is the master base I/O nasid */
> > > +EXPORT_SYMBOL(master_nasid);
> >
> > why is this exported?
>
> Because no one's ripped it out yet.
So rip it out :)
> > > char sn_system_serial_number_string[128];
> > > u64 sn_partition_serial_number;
> > > +EXPORT_SYMBOL(sn_system_serial_number_string);
> > > +EXPORT_SYMBOL(sn_partition_serial_number);
>
> Modules.
>
> > > short physical_node_map[MAX_PHYSNODE_ID];
> > > +EXPORT_SYMBOL(physical_node_map);
> >
> > Again, what is all this for?
>
> Modules.
What modules. My limited grep abilities can't find users and I don't
really see much proper use for those either..
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [PATCH] sn2 module (& other misc.) fixes
2003-08-04 21:14 [PATCH] sn2 module (& other misc.) fixes Jesse Barnes
` (2 preceding siblings ...)
2003-08-04 22:04 ` Christoph Hellwig
@ 2003-08-04 23:31 ` David Mosberger
2003-08-05 0:06 ` Jesse Barnes
` (3 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: David Mosberger @ 2003-08-04 23:31 UTC (permalink / raw)
To: linux-ia64
>>>>> On Mon, 4 Aug 2003 14:14:04 -0700, jbarnes@sgi.com (Jesse Barnes) said:
Jesse> This one goes on top of the one I sent you last week. I'm
Jesse> shooting for making Linus' tree boot on sn2 as well...
This one didn't apply for me. I just updated the to-linus-2.5 tree.
Can you rediff against that?
Thanks,
--david
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [PATCH] sn2 module (& other misc.) fixes
2003-08-04 21:14 [PATCH] sn2 module (& other misc.) fixes Jesse Barnes
` (3 preceding siblings ...)
2003-08-04 23:31 ` David Mosberger
@ 2003-08-05 0:06 ` Jesse Barnes
2003-08-05 1:01 ` David Mosberger
` (2 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: Jesse Barnes @ 2003-08-05 0:06 UTC (permalink / raw)
To: linux-ia64
On Mon, Aug 04, 2003 at 04:31:32PM -0700, David Mosberger wrote:
> >>>>> On Mon, 4 Aug 2003 14:14:04 -0700, jbarnes@sgi.com (Jesse Barnes) said:
>
> Jesse> This one goes on top of the one I sent you last week. I'm
> Jesse> shooting for making Linus' tree boot on sn2 as well...
>
> This one didn't apply for me. I just updated the to-linus-2.5 tree.
> Can you rediff against that?
Sure, does this one work for you?
Thanks,
Jesse
diff -Nru a/arch/ia64/sn/io/hwgfs/hcl.c b/arch/ia64/sn/io/hwgfs/hcl.c
--- a/arch/ia64/sn/io/hwgfs/hcl.c Mon Aug 4 17:02:37 2003
+++ b/arch/ia64/sn/io/hwgfs/hcl.c Mon Aug 4 17:02:37 2003
@@ -898,6 +898,5 @@
EXPORT_SYMBOL(hwgraph_info_get_next_LBL);
EXPORT_SYMBOL(hwgraph_info_export_LBL);
EXPORT_SYMBOL(hwgraph_info_unexport_LBL);
-EXPORT_SYMBOL(hwgraph_path_lookup);
EXPORT_SYMBOL(hwgraph_traverse);
EXPORT_SYMBOL(hwgraph_vertex_name_get);
diff -Nru a/arch/ia64/sn/io/machvec/iomv.c b/arch/ia64/sn/io/machvec/iomv.c
--- a/arch/ia64/sn/io/machvec/iomv.c Mon Aug 4 17:02:37 2003
+++ b/arch/ia64/sn/io/machvec/iomv.c Mon Aug 4 17:02:37 2003
@@ -73,3 +73,4 @@
SH_PIO_WRITE_STATUS_0_PENDING_WRITE_COUNT_MASK)
udelay(1);
}
+EXPORT_SYMBOL(sn_mmiob);
diff -Nru a/arch/ia64/sn/io/sn2/shub.c b/arch/ia64/sn/io/sn2/shub.c
--- a/arch/ia64/sn/io/sn2/shub.c Mon Aug 4 17:02:37 2003
+++ b/arch/ia64/sn/io/sn2/shub.c Mon Aug 4 17:02:37 2003
@@ -199,10 +199,9 @@
{
cnodeid_t cnode;
uint64_t longarg;
- vertex_hdl_t d;
int nasid;
- cnode = (cnodeid_t)hwgraph_fastinfo_get(d);
+ cnode = (cnodeid_t)file->f_dentry->d_fsdata;
switch (cmd) {
case SNDRV_SHUB_CONFIGURE:
diff -Nru a/arch/ia64/sn/io/sn2/shuberror.c b/arch/ia64/sn/io/sn2/shuberror.c
--- a/arch/ia64/sn/io/sn2/shuberror.c Mon Aug 4 17:02:37 2003
+++ b/arch/ia64/sn/io/sn2/shuberror.c Mon Aug 4 17:02:37 2003
@@ -52,7 +52,6 @@
hub_error_clear(nasid_t nasid)
{
int i;
- hubreg_t idsr;
/*
* Make sure spurious write response errors are cleared
diff -Nru a/arch/ia64/sn/kernel/sn2/cache.c b/arch/ia64/sn/kernel/sn2/cache.c
--- a/arch/ia64/sn/kernel/sn2/cache.c Mon Aug 4 17:02:37 2003
+++ b/arch/ia64/sn/kernel/sn2/cache.c Mon Aug 4 17:02:37 2003
@@ -8,7 +8,9 @@
*
*/
-#include <asm/pgalloc.h>
+#include <linux/module.h>
+#include <asm/cacheflush.h>
+#include <asm/system.h>
/**
* sn_flush_all_caches - flush a range of address from all caches (incl. L4)
@@ -26,3 +28,4 @@
flush_icache_range(flush_addr, flush_addr+bytes);
mb();
}
+EXPORT_SYMBOL(sn_flush_all_caches);
diff -Nru a/arch/ia64/sn/kernel/sn2/sn2_smp.c b/arch/ia64/sn/kernel/sn2/sn2_smp.c
--- a/arch/ia64/sn/kernel/sn2/sn2_smp.c Mon Aug 4 17:02:37 2003
+++ b/arch/ia64/sn/kernel/sn2/sn2_smp.c Mon Aug 4 17:02:37 2003
@@ -41,6 +41,7 @@
#include <linux/interrupt.h>
#include <linux/irq.h>
#include <linux/mmzone.h>
+#include <linux/module.h>
#include <asm/processor.h>
#include <asm/irq.h>
@@ -214,6 +215,7 @@
}
}
+EXPORT_SYMBOL(sn_send_IPI_phys);
/**
* sn2_send_IPI - send an IPI to a processor
diff -Nru a/arch/ia64/sn/kernel/sn_ksyms.c b/arch/ia64/sn/kernel/sn_ksyms.c
--- a/arch/ia64/sn/kernel/sn_ksyms.c Mon Aug 4 17:02:37 2003
+++ b/arch/ia64/sn/kernel/sn_ksyms.c Mon Aug 4 17:02:37 2003
@@ -13,48 +13,25 @@
#include <linux/config.h>
#include <linux/module.h>
-
+#include <linux/mm.h>
#include <asm/machvec.h>
#include <asm/sn/intr.h>
-
-#include <linux/mm.h>
#include <asm/sn/sgi.h>
-extern vertex_hdl_t base_io_scsi_ctlr_vhdl[];
#include <asm/sn/types.h>
-extern cnodeid_t master_node_get(devfs_handle_t vhdl);
#include <asm/sn/arch.h>
-EXPORT_SYMBOL(base_io_scsi_ctlr_vhdl);
-EXPORT_SYMBOL(master_node_get);
+#include <asm/sn/bte.h>
+#include <asm/sal.h>
+#include <asm/sn/sn_sal.h>
#ifdef CONFIG_IA64_SGI_SN_DEBUG
EXPORT_SYMBOL(__pa_debug);
EXPORT_SYMBOL(__va_debug);
#endif
-/* Support IPIs for loaded modules. */
-EXPORT_SYMBOL(sn_send_IPI_phys);
-
-/* symbols referenced by partitioning modules */
-#include <asm/sn/bte.h>
EXPORT_SYMBOL(bte_copy);
EXPORT_SYMBOL(bte_unaligned_copy);
-#include <asm/sal.h>
EXPORT_SYMBOL(ia64_sal);
-EXPORT_SYMBOL(physical_node_map);
-
-#include <asm/sn/sn_sal.h>
EXPORT_SYMBOL(sal_lock);
-EXPORT_SYMBOL(sn_partid);
EXPORT_SYMBOL(sn_local_partid);
-EXPORT_SYMBOL(sn_system_serial_number_string);
-EXPORT_SYMBOL(sn_partition_serial_number);
-
-EXPORT_SYMBOL(sn_mmiob);
-
-/* added by tduffy 04.08.01 to fix depmod issues */
-#include <linux/mmzone.h>
-extern nasid_t master_nasid;
-EXPORT_SYMBOL(master_nasid);
-EXPORT_SYMBOL(sn_flush_all_caches);
diff -Nru a/include/asm-ia64/sn/xtalk/xtalk.h b/include/asm-ia64/sn/xtalk/xtalk.h
--- a/include/asm-ia64/sn/xtalk/xtalk.h Mon Aug 4 17:02:37 2003
+++ b/include/asm-ia64/sn/xtalk/xtalk.h Mon Aug 4 17:02:37 2003
@@ -19,7 +19,7 @@
/*
* User-level device driver visible types
*/
-typedef char xwidgetnum_t; /* xtalk widget number (0..15) */
+typedef int xwidgetnum_t; /* xtalk widget number (0..15) */
#define XWIDGET_NONE (-1)
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [PATCH] sn2 module (& other misc.) fixes
2003-08-04 21:14 [PATCH] sn2 module (& other misc.) fixes Jesse Barnes
` (4 preceding siblings ...)
2003-08-05 0:06 ` Jesse Barnes
@ 2003-08-05 1:01 ` David Mosberger
2003-08-05 1:15 ` Jesse Barnes
2003-08-05 6:04 ` David Mosberger
7 siblings, 0 replies; 9+ messages in thread
From: David Mosberger @ 2003-08-05 1:01 UTC (permalink / raw)
To: linux-ia64
>>>>> On Mon, 4 Aug 2003 17:06:31 -0700, jbarnes@sgi.com (Jesse Barnes) said:
Jesse> On Mon, Aug 04, 2003 at 04:31:32PM -0700, David Mosberger
Jesse> wrote:
>> >>>>> On Mon, 4 Aug 2003 14:14:04 -0700, jbarnes@sgi.com (Jesse
>> Barnes) said:
Jesse> This one goes on top of the one I sent you last week. I'm
Jesse> shooting for making Linus' tree boot on sn2 as well...
>> This one didn't apply for me. I just updated the to-linus-2.5
>> tree. Can you rediff against that?
Jesse> Sure, does this one work for you?
Yes, this one worked. It's in my tree now.
--david
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [PATCH] sn2 module (& other misc.) fixes
2003-08-04 21:14 [PATCH] sn2 module (& other misc.) fixes Jesse Barnes
` (5 preceding siblings ...)
2003-08-05 1:01 ` David Mosberger
@ 2003-08-05 1:15 ` Jesse Barnes
2003-08-05 6:04 ` David Mosberger
7 siblings, 0 replies; 9+ messages in thread
From: Jesse Barnes @ 2003-08-05 1:15 UTC (permalink / raw)
To: linux-ia64
On Mon, Aug 04, 2003 at 06:01:42PM -0700, David Mosberger wrote:
> >>>>> On Mon, 4 Aug 2003 17:06:31 -0700, jbarnes@sgi.com (Jesse Barnes) said:
>
> Jesse> On Mon, Aug 04, 2003 at 04:31:32PM -0700, David Mosberger
> Jesse> wrote:
> >> >>>>> On Mon, 4 Aug 2003 14:14:04 -0700, jbarnes@sgi.com (Jesse
> >> Barnes) said:
>
> Jesse> This one goes on top of the one I sent you last week. I'm
> Jesse> shooting for making Linus' tree boot on sn2 as well...
> >> This one didn't apply for me. I just updated the to-linus-2.5
> >> tree. Can you rediff against that?
>
> Jesse> Sure, does this one work for you?
>
> Yes, this one worked. It's in my tree now.
Cool! Now I only have to apply a couple of patches (one being the
discontig patch) to get the 2.6 tree booting on sn2.
What needs to be done to fix the assembler? It always spews a bunch of
warnings during the build. The ia64 build would be really quiet if that
was fixed...
Thanks,
Jesse
^ permalink raw reply [flat|nested] 9+ messages in thread* Re: [PATCH] sn2 module (& other misc.) fixes
2003-08-04 21:14 [PATCH] sn2 module (& other misc.) fixes Jesse Barnes
` (6 preceding siblings ...)
2003-08-05 1:15 ` Jesse Barnes
@ 2003-08-05 6:04 ` David Mosberger
7 siblings, 0 replies; 9+ messages in thread
From: David Mosberger @ 2003-08-05 6:04 UTC (permalink / raw)
To: linux-ia64
>>>>> On Mon, 4 Aug 2003 18:15:58 -0700, jbarnes@sgi.com (Jesse Barnes) said:
Jesse> What needs to be done to fix the assembler? It always spews
Jesse> a bunch of warnings during the build. The ia64 build would
Jesse> be really quiet if that was fixed...
We need to add directives to let the assembler know that the code is
indeed safe.
--david
^ permalink raw reply [flat|nested] 9+ messages in thread