Hello
This patch serie fixes some minor problem found in the agp subsystem
There are no change since v1 (posted two years ago)
This is simply a repost for trying to get an answer (gentle ping 6 month
ago got no answer also).
Regards
Corentin Labbe (5):
agp: remove unused variable size in agp_generic_create_gatt_table
agp: move AGPGART_MINOR to include/linux/miscdevice.h
agp: remove unused variable num_segments
agp: Add bridge parameter documentation
ia64: agp: Replace empty define with do while
arch/ia64/include/asm/agp.h | 4 ++--
drivers/char/agp/frontend.c | 3 +--
drivers/char/agp/generic.c | 12 +++++-------
include/linux/agpgart.h | 2 --
include/linux/miscdevice.h | 1 +
5 files changed, 9 insertions(+), 13 deletions(-)
--
2.23.0
This patch add documentation about the bridge parameter in several
function.
This will fix the following build warning:
drivers/char/agp/generic.c:220: warning: No description found for parameter 'bridge'
drivers/char/agp/generic.c:364: warning: No description found for parameter 'bridge'
drivers/char/agp/generic.c:1283: warning: No description found for parameter 'bridge'
Signed-off-by: Corentin Labbe <[email protected]>
---
drivers/char/agp/generic.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/char/agp/generic.c b/drivers/char/agp/generic.c
index 4d2eb0800b2a..ab154a75acf0 100644
--- a/drivers/char/agp/generic.c
+++ b/drivers/char/agp/generic.c
@@ -207,6 +207,7 @@ EXPORT_SYMBOL(agp_free_memory);
/**
* agp_allocate_memory - allocate a group of pages of a certain type.
*
+ * @bridge: an agp_bridge_data struct allocated for the AGP host bridge.
* @page_count: size_t argument of the number of pages
* @type: u32 argument of the type of memory to be allocated.
*
@@ -355,6 +356,7 @@ EXPORT_SYMBOL_GPL(agp_num_entries);
/**
* agp_copy_info - copy bridge state information
*
+ * @bridge: an agp_bridge_data struct allocated for the AGP host bridge.
* @info: agp_kern_info pointer. The caller should insure that this pointer is valid.
*
* This function copies information about the agp bridge device and the state of
@@ -1277,6 +1279,7 @@ EXPORT_SYMBOL(agp_generic_destroy_page);
/**
* agp_enable - initialise the agp point-to-point connection.
*
+ * @bridge: an agp_bridge_data struct allocated for the AGP host bridge.
* @mode: agp mode register value to configure with.
*/
void agp_enable(struct agp_bridge_data *bridge, u32 mode)
--
2.23.0
This patch fix the following build warning:
warning: variable 'num_segments' set but not used [-Wunused-but-set-variable]
Signed-off-by: Corentin Labbe <[email protected]>
---
drivers/char/agp/frontend.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/char/agp/frontend.c b/drivers/char/agp/frontend.c
index f6955888e676..47098648502d 100644
--- a/drivers/char/agp/frontend.c
+++ b/drivers/char/agp/frontend.c
@@ -102,14 +102,13 @@ agp_segment_priv *agp_find_seg_in_client(const struct agp_client *client,
int size, pgprot_t page_prot)
{
struct agp_segment_priv *seg;
- int num_segments, i;
+ int i;
off_t pg_start;
size_t pg_count;
pg_start = offset / 4096;
pg_count = size / 4096;
seg = *(client->segments);
- num_segments = client->num_segments;
for (i = 0; i < client->num_segments; i++) {
if ((seg[i].pg_start == pg_start) &&
--
2.23.0
This patch move the define for AGPGART_MINOR to include/linux/miscdevice.h.
It is better that all minor number definitions are in the same place.
Signed-off-by: Corentin Labbe <[email protected]>
---
include/linux/agpgart.h | 2 --
include/linux/miscdevice.h | 1 +
2 files changed, 1 insertion(+), 2 deletions(-)
diff --git a/include/linux/agpgart.h b/include/linux/agpgart.h
index c6b61ca97053..21b34a96cfd8 100644
--- a/include/linux/agpgart.h
+++ b/include/linux/agpgart.h
@@ -30,8 +30,6 @@
#include <linux/agp_backend.h>
#include <uapi/linux/agpgart.h>
-#define AGPGART_MINOR 175
-
struct agp_info {
struct agp_version version; /* version of the driver */
u32 bridge_id; /* bridge vendor/device */
diff --git a/include/linux/miscdevice.h b/include/linux/miscdevice.h
index b06b75776a32..becde6981a95 100644
--- a/include/linux/miscdevice.h
+++ b/include/linux/miscdevice.h
@@ -33,6 +33,7 @@
#define SGI_MMTIMER 153
#define STORE_QUEUE_MINOR 155 /* unused */
#define I2O_MINOR 166
+#define AGPGART_MINOR 175
#define HWRNG_MINOR 183
#define MICROCODE_MINOR 184
#define IRNET_MINOR 187
--
2.23.0
It's dangerous to use empty code define.
Furthermore it lead to the following warning:
drivers/char/agp/generic.c: In function « agp_generic_destroy_page »:
drivers/char/agp/generic.c:1266:28: attention : suggest braces around empty body in an « if » statement [-Wempty-body]
So let's replace emptyness by "do {} while(0)"
Signed-off-by: Corentin Labbe <[email protected]>
---
arch/ia64/include/asm/agp.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/ia64/include/asm/agp.h b/arch/ia64/include/asm/agp.h
index 2b451c4496da..0261507dc264 100644
--- a/arch/ia64/include/asm/agp.h
+++ b/arch/ia64/include/asm/agp.h
@@ -14,8 +14,8 @@
* in coherent mode, which lets us map the AGP memory as normal (write-back) memory
* (unlike x86, where it gets mapped "write-coalescing").
*/
-#define map_page_into_agp(page) /* nothing */
-#define unmap_page_from_agp(page) /* nothing */
+#define map_page_into_agp(page) do { } while (0)
+#define unmap_page_from_agp(page) do { } while (0)
#define flush_agp_cache() mb()
/* GATT allocation. Returns/accepts GATT kernel virtual address. */
--
2.23.0
This patch fix the following warning:
drivers/char/agp/generic.c:853:6: attention : variable ‘size’ set but not used [-Wunused-but-set-variable]
by removing the unused variable size in agp_generic_create_gatt_table
Signed-off-by: Corentin Labbe <[email protected]>
---
drivers/char/agp/generic.c | 9 ++-------
1 file changed, 2 insertions(+), 7 deletions(-)
diff --git a/drivers/char/agp/generic.c b/drivers/char/agp/generic.c
index df1edb5ec0ad..4d2eb0800b2a 100644
--- a/drivers/char/agp/generic.c
+++ b/drivers/char/agp/generic.c
@@ -850,7 +850,6 @@ int agp_generic_create_gatt_table(struct agp_bridge_data *bridge)
{
char *table;
char *table_end;
- int size;
int page_order;
int num_entries;
int i;
@@ -864,25 +863,22 @@ int agp_generic_create_gatt_table(struct agp_bridge_data *bridge)
table = NULL;
i = bridge->aperture_size_idx;
temp = bridge->current_size;
- size = page_order = num_entries = 0;
+ page_order = num_entries = 0;
if (bridge->driver->size_type != FIXED_APER_SIZE) {
do {
switch (bridge->driver->size_type) {
case U8_APER_SIZE:
- size = A_SIZE_8(temp)->size;
page_order =
A_SIZE_8(temp)->page_order;
num_entries =
A_SIZE_8(temp)->num_entries;
break;
case U16_APER_SIZE:
- size = A_SIZE_16(temp)->size;
page_order = A_SIZE_16(temp)->page_order;
num_entries = A_SIZE_16(temp)->num_entries;
break;
case U32_APER_SIZE:
- size = A_SIZE_32(temp)->size;
page_order = A_SIZE_32(temp)->page_order;
num_entries = A_SIZE_32(temp)->num_entries;
break;
@@ -890,7 +886,7 @@ int agp_generic_create_gatt_table(struct agp_bridge_data *bridge)
case FIXED_APER_SIZE:
case LVL2_APER_SIZE:
default:
- size = page_order = num_entries = 0;
+ page_order = num_entries = 0;
break;
}
@@ -920,7 +916,6 @@ int agp_generic_create_gatt_table(struct agp_bridge_data *bridge)
}
} while (!table && (i < bridge->driver->num_aperture_sizes));
} else {
- size = ((struct aper_size_info_fixed *) temp)->size;
page_order = ((struct aper_size_info_fixed *) temp)->page_order;
num_entries = ((struct aper_size_info_fixed *) temp)->num_entries;
table = alloc_gatt_pages(page_order);
--
2.23.0
On Thu, Nov 21, 2019 at 9:15 AM Corentin Labbe <[email protected]> wrote:
>
> Hello
>
> This patch serie fixes some minor problem found in the agp subsystem
> There are no change since v1 (posted two years ago)
> This is simply a repost for trying to get an answer (gentle ping 6 month
> ago got no answer also).
Looks all good to me,
Acked-by: Arnd Bergmann <[email protected]>
On Thu, Nov 21, 2019 at 08:14:40AM +0000, Corentin Labbe wrote:
> Hello
>
> This patch serie fixes some minor problem found in the agp subsystem
> There are no change since v1 (posted two years ago)
> This is simply a repost for trying to get an answer (gentle ping 6 month
> ago got no answer also).
>
> Regards
>
Hello
Does the AGP maintainer still maintain it ?
Regards
On Mon, Dec 2, 2019 at 11:33 PM LABBE Corentin <[email protected]> wrote:
>
> On Thu, Nov 21, 2019 at 08:14:40AM +0000, Corentin Labbe wrote:
> > Hello
> >
> > This patch serie fixes some minor problem found in the agp subsystem
> > There are no change since v1 (posted two years ago)
> > This is simply a repost for trying to get an answer (gentle ping 6 month
> > ago got no answer also).
> >
> > Regards
> >
>
> Hello
>
> Does the AGP maintainer still maintain it ?
It's maintained but really loathe to touch it, I've no hw to validate
any changes on so making any changes to it really has to get past my
internal, I care enough about this change to risk applying anything to
AGP.
I'll try and look and apply those patches today.
Dave.
On Tue, Dec 03, 2019 at 06:01:15AM +1000, David Airlie wrote:
> On Mon, Dec 2, 2019 at 11:33 PM LABBE Corentin <[email protected]> wrote:
> >
> > On Thu, Nov 21, 2019 at 08:14:40AM +0000, Corentin Labbe wrote:
> > > Hello
> > >
> > > This patch serie fixes some minor problem found in the agp subsystem
> > > There are no change since v1 (posted two years ago)
> > > This is simply a repost for trying to get an answer (gentle ping 6 month
> > > ago got no answer also).
> > >
> > > Regards
> > >
> >
> > Hello
> >
> > Does the AGP maintainer still maintain it ?
>
> It's maintained but really loathe to touch it, I've no hw to validate
> any changes on so making any changes to it really has to get past my
> internal, I care enough about this change to risk applying anything to
> AGP.
>
> I'll try and look and apply those patches today.
>
Thanks for applying.
Perhaps you need to fix your address in MAINTAINERS.
When you has hardware, What was your tests procedure ?
I can on my freetime add some AGP hw on my kernelCI lab.
Regards
> Thanks for applying.
> Perhaps you need to fix your address in MAINTAINERS.
>
> When you has hardware, What was your tests procedure ?
Just boot and run gears type thing.
> I can on my freetime add some AGP hw on my kernelCI lab.
>
I don't know where to get AGP hw these days that isn't horrible power
consumption wise, I can't really wish for anyone to ever put it into
CI for an area that we change once a year.
I'd rather we just don't touch AGP too much and let it die and remove
support in 5-10 years.
Dave.