[PATCH] parisc: agp/parisc-agp: allow binding of user memory to the AGP GART

[PATCH] parisc: agp/parisc-agp: allow binding of user memory to the AGP GART

[gnidorah]

[-- Attachment #1: Type: text/plain, Size: 705 bytes --]

The patch allows binding of user memory to the AGP GART 
on systems with HP Quicksilver AGP bus. This resolves 'bind memory failed' 
error seen in dmesg:

[   29.365973] [TTM] AGP Bind memory failed.
…
[   29.367030] [drm] Forcing AGP to PCI mode 

The system doesn't more fail to bind the memory, and hence not falling
back to the PCI mode (if other failures aren't detected).

This is just a simple write down from the following patches:
agp/amd-k7: Allow binding user memory to the AGP GART [1]
agp/hp-agp: Allow binding user memory to the AGP GART [2]

Both fixes are in upstream.

[1] https://lkml.org/lkml/2010/10/15/469
[2] http://marc.info/?l=linux-ia64&m=131412890819311&w=2


[-- Attachment #2: patch-parisc_agp_c --]
[-- Type: application/octet-stream, Size: 811 bytes --]

diff --git a/drivers/char/agp/parisc-agp.c b/drivers/char/agp/parisc-agp.c
index 94821ab..9576fad 100644
--- a/drivers/char/agp/parisc-agp.c
+++ b/drivers/char/agp/parisc-agp.c
@@ -129,7 +129,8 @@ parisc_agp_insert_memory(struct agp_memory *mem, off_t pg_start, int type)
 	off_t j, io_pg_start;
 	int io_pg_count;
 
-	if (type != 0 || mem->type != 0) {
+	if (type != mem->type ||
+		agp_bridge->driver->agp_type_to_mask_type(agp_bridge, type)) {
 		return -EINVAL;
 	}
 
@@ -175,7 +176,8 @@ parisc_agp_remove_memory(struct agp_memory *mem, off_t pg_start, int type)
 	struct _parisc_agp_info *info = &parisc_agp_info;
 	int i, io_pg_start, io_pg_count;
 
-	if (type != 0 || mem->type != 0) {
+	if (type != mem->type ||
+		agp_bridge->driver->agp_type_to_mask_type(agp_bridge, type)) {
 		return -EINVAL;
 	}
 

Re: [PATCH] parisc: agp/parisc-agp: allow binding of user memory to the AGP GART

[Helge Deller]

Hi Alex,

On 07/10/2013 08:17 AM, gnidorah@p0n4ik.tk wrote:
> The patch allows binding of user memory to the AGP GART=20
> on systems with HP Quicksilver AGP bus. This resolves 'bind memory fa=
iled'=20
> error seen in dmesg:
>=20
> [   29.365973] [TTM] AGP Bind memory failed.
> =85
> [   29.367030] [drm] Forcing AGP to PCI mode=20
>=20
> The system doesn't more fail to bind the memory, and hence not fallin=
g
> back to the PCI mode (if other failures aren't detected).
>=20
> This is just a simple write down from the following patches:
> agp/amd-k7: Allow binding user memory to the AGP GART [1]
> agp/hp-agp: Allow binding user memory to the AGP GART [2]
>=20
> Both fixes are in upstream.
>=20
> [1] https://lkml.org/lkml/2010/10/15/469
> [2] http://marc.info/?l=3Dlinux-ia64&m=3D131412890819311&w=3D2

Thanks for the patch!!!

Could you please attach patches inline (as text) next time?
Our patchwork tool (https://patchwork.kernel.org/project/linux-parisc/l=
ist/) missed to pick up your patch too.

=46urthermore, I just noticed that vger.kernel.org stopped forwarding m=
e mails to the mailing list
during the last few days. So, I would have missed your patch if I would=
n't have noticed the failing forwarding.

Thanks,
Helge

PS: I attached it here again for patchwork....


diff --git a/drivers/char/agp/parisc-agp.c b/drivers/char/agp/parisc-ag=
p.c
index 94821ab..9576fad 100644
--- a/drivers/char/agp/parisc-agp.c
+++ b/drivers/char/agp/parisc-agp.c
@@ -129,7 +129,8 @@ parisc_agp_insert_memory(struct agp_memory *mem, of=
f_t pg_start, int type)
 	off_t j, io_pg_start;
 	int io_pg_count;
=20
-	if (type !=3D 0 || mem->type !=3D 0) {
+	if (type !=3D mem->type ||
+		agp_bridge->driver->agp_type_to_mask_type(agp_bridge, type)) {
 		return -EINVAL;
 	}
=20
@@ -175,7 +176,8 @@ parisc_agp_remove_memory(struct agp_memory *mem, of=
f_t pg_start, int type)
 	struct _parisc_agp_info *info =3D &parisc_agp_info;
 	int i, io_pg_start, io_pg_count;
=20
-	if (type !=3D 0 || mem->type !=3D 0) {
+	if (type !=3D mem->type ||
+		agp_bridge->driver->agp_type_to_mask_type(agp_bridge, type)) {
 		return -EINVAL;
 	}
=20
--
To unsubscribe from this list: send the line "unsubscribe linux-parisc"=
 in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH] parisc: agp/parisc-agp: allow binding of user memory to the AGP GART

[Helge Deller]

Hi Alex,

On 07/10/2013 09:14 PM, Helge Deller wrote:
> On 07/10/2013 08:17 AM, gnidorah@p0n4ik.tk wrote:
>> The patch allows binding of user memory to the AGP GART=20
>> on systems with HP Quicksilver AGP bus. This resolves 'bind memory f=
ailed'=20
>> error seen in dmesg:
>>
>> [   29.365973] [TTM] AGP Bind memory failed.
>> =85
>> [   29.367030] [drm] Forcing AGP to PCI mode=20
>>
>> The system doesn't more fail to bind the memory, and hence not falli=
ng
>> back to the PCI mode (if other failures aren't detected).
>>
>> This is just a simple write down from the following patches:
>> agp/amd-k7: Allow binding user memory to the AGP GART [1]
>> agp/hp-agp: Allow binding user memory to the AGP GART [2]
>>
>> Both fixes are in upstream.
>>
>> [1] https://lkml.org/lkml/2010/10/15/469
>> [2] http://marc.info/?l=3Dlinux-ia64&m=3D131412890819311&w=3D2


Please send an Signed-off-by: line for your patch too...

Helge




>=20
> Thanks for the patch!!!
>=20
> Could you please attach patches inline (as text) next time?
> Our patchwork tool (https://patchwork.kernel.org/project/linux-parisc=
/list/) missed to pick up your patch too.
>=20
> Furthermore, I just noticed that vger.kernel.org stopped forwarding m=
e mails to the mailing list
> during the last few days. So, I would have missed your patch if I wou=
ldn't have noticed the failing forwarding.
>=20
> Thanks,
> Helge
>=20
> PS: I attached it here again for patchwork....
>=20
>=20
> diff --git a/drivers/char/agp/parisc-agp.c b/drivers/char/agp/parisc-=
agp.c
> index 94821ab..9576fad 100644
> --- a/drivers/char/agp/parisc-agp.c
> +++ b/drivers/char/agp/parisc-agp.c
> @@ -129,7 +129,8 @@ parisc_agp_insert_memory(struct agp_memory *mem, =
off_t pg_start, int type)
>  	off_t j, io_pg_start;
>  	int io_pg_count;
> =20
> -	if (type !=3D 0 || mem->type !=3D 0) {
> +	if (type !=3D mem->type ||
> +		agp_bridge->driver->agp_type_to_mask_type(agp_bridge, type)) {
>  		return -EINVAL;
>  	}
> =20
> @@ -175,7 +176,8 @@ parisc_agp_remove_memory(struct agp_memory *mem, =
off_t pg_start, int type)
>  	struct _parisc_agp_info *info =3D &parisc_agp_info;
>  	int i, io_pg_start, io_pg_count;
> =20
> -	if (type !=3D 0 || mem->type !=3D 0) {
> +	if (type !=3D mem->type ||
> +		agp_bridge->driver->agp_type_to_mask_type(agp_bridge, type)) {
>  		return -EINVAL;
>  	}
> =20
>=20

--
To unsubscribe from this list: send the line "unsubscribe linux-parisc"=
 in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH] parisc: agp/parisc-agp: allow binding of user memory to the AGP GART

[Alex Ivanov]

10.07.2013, 23:14, "Helge Deller" <deller@gmx.de>:

> =9AHi Alex,
>
> =9AThanks for the patch!!!
>
> =9ACould you please attach patches inline (as text) next time?
> =9AOur patchwork tool (https://patchwork.kernel.org/project/linux-par=
isc/list/) missed to pick up your patch too.
>
> =9AFurthermore, I just noticed that vger.kernel.org stopped forwardin=
g me mails to the mailing list
> =9Aduring the last few days. So, I would have missed your patch if I =
wouldn't have noticed the failing forwarding.
>
> =9AThanks,
> =9AHelge

Helge,

> Could you please attach patches inline (as text) next time?

My mistake, i'll do next time.

> Please send an Signed-off-by: line for your patch too...

Signed-off-by: Alex Ivanov <gnidorah@p0n4ik.tk>

Thank you!
--
To unsubscribe from this list: send the line "unsubscribe linux-parisc"=
 in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html