Re: [patch 00/28] PNP: convert fixed tables to lists, v4

Re: [patch 00/28] PNP: convert fixed tables to lists, v4

[Len Brown]

v4 has replaced v3 in the acpi-test tree.

thanks,
-Len

On Fri, 27 Jun 2008, Bjorn Helgaas wrote:

> This patch series contains all my PNP patches that are not yet in Linus'
> tree, including both the "convert resource table to dynamic list" series
> and the "convert resource options to unified dynamic list" series.
> 
> This fixes all checkpatch warnings except a "do not add new typedefs"
> warning.  I kept the pnp_irq_mask_t typedef because I don't see a
> better way to declare bitmaps in various places and pass around pointers
> to them.
> 
> Len, I told you I was going to send incremental patches, but this is
> a replacement instead.
> 
> Changes since v3:
> 
>     - Move IRQ test from pnpacpi_parse_irq_option() to
>       pnpacpi_parse_ext_irq_option() (fixes "comparison is always true"
>       warning noticed by David Howells)
>     - Use pci_get_legacy_ide_irq(), not ATA_PRIMARY_IRQ()/ATA_SECONDARY_IRQ()
>       (thanks to Bartlomiej Zolnierkiewicz)
>     - Use snprintf() to make printks atomic
>     - Fix checkpatch warnings (line length, "//" comments)
> 
> Here are the significant changes since the last posting:
> 
>     - When replacing pnp_resource_table, the PNPACPI resource encoders
>       have to explicitly handle disabled resources; previously any
>       unused slots in the table were implicitly disabled.  This should
>       fix Jiri Slaby's oops on suspend/resume (though he hasn't tested
>       this version).
>       
>       This change is in the patch titled "PNP: replace pnp_resource_table
>       with dynamically allocated resources".
>       
>     - Added a new patch to keep disabled IRQ and DMA resources when parsing
>       current config.  This fixes a bug in the current (2.6.25) PNPACPI
>       code: consider a device with (mem, irq0, irq1, io), where irq0 is
>       disabled.  If we drop irq0 when parsing the _CRS, we will mistakenly
>       put irq1 in the irq0 slot when we encode resources for an _SRS call.
> 
>     - Added a new patch to avoid interrupts used by an IDE controller in
>       compatibility mode.  https://bugzilla.novell.com/show_bug.cgi?id=375836
> 
>     - Added a new PNPACPI patch to support HP vendor-specific descriptors.
>       This has long been supported by arch/ia64/kernel/acpi-ext.c, which
>       provides an hp_acpi_csr_space() interface, but I think it's better
>       and more generic to support it in PNPACPI.
> 
> Changes since v2 of "convert resource options to unified dynamic list":
> 
>     - fixed bisection problem in quirk_ad1815_mpu_resources()
>     - fixed checkpatch warning in pnp_show_options()
>       (pnp-convert-resource-options-to-single-linked-list-checkpatch-fixes.patch)
>     - replaced pnp_independent_option() with literal 0
>     - fixed coding style in pnp/manager.c
>     - added EXPORT_SYMBOL(pnp_possible_config)
>       (pnp-add-pnp_possible_config-can-a-device-could-be-configured-this-way-fix.patch)
> 
> Bjorn
> -- 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

Re: [patch 11/28] PNP: whitespace/coding style fixes

[Joe Perches]

On Fri, 2008-06-27 at 16:57 -0600, Bjorn Helgaas wrote:
> Index: work10/drivers/pnp/interface.c
> ===================================================================
> --- work10.orig/drivers/pnp/interface.c	2008-05-09 14:42:25.000000000 -0600
> +++ work10/drivers/pnp/interface.c	2008-05-09 14:45:57.000000000 -0600
> @@ -216,12 +216,12 @@ static ssize_t pnp_show_options(struct d
>  				struct device_attribute *attr, char *buf)
>  {
>  	struct pnp_dev *dev = to_pnp_dev(dmdev);
> +	pnp_info_buffer_t *buffer;
>  	struct pnp_option *independent = dev->independent;
>  	struct pnp_option *dependent = dev->dependent;
>  	int ret, dep = 1;
>  
> -	pnp_info_buffer_t *buffer = (pnp_info_buffer_t *)
> -	    pnp_alloc(sizeof(pnp_info_buffer_t));
> +	buffer = pnp_alloc(sizeof(pnp_info_buffer_t));
>  	if (!buffer)
>  		return -ENOMEM;

Wouldn't it be nicer without the unnecessary cast?

	pnp_info_buffer_t *buffer = pnp_alloc(sizeof(*buffer));

Re: [patch 11/28] PNP: whitespace/coding style fixes

[Rene Herman]

On 28-06-08 21:56, Joe Perches wrote:
> On Fri, 2008-06-27 at 16:57 -0600, Bjorn Helgaas wrote:
>> Index: work10/drivers/pnp/interface.c
>> ===================================================================
>> --- work10.orig/drivers/pnp/interface.c	2008-05-09 14:42:25.000000000 -0600
>> +++ work10/drivers/pnp/interface.c	2008-05-09 14:45:57.000000000 -0600
>> @@ -216,12 +216,12 @@ static ssize_t pnp_show_options(struct d
>>  				struct device_attribute *attr, char *buf)
>>  {
>>  	struct pnp_dev *dev = to_pnp_dev(dmdev);
>> +	pnp_info_buffer_t *buffer;
>>  	struct pnp_option *independent = dev->independent;
>>  	struct pnp_option *dependent = dev->dependent;
>>  	int ret, dep = 1;
>>  
>> -	pnp_info_buffer_t *buffer = (pnp_info_buffer_t *)
>> -	    pnp_alloc(sizeof(pnp_info_buffer_t));
>> +	buffer = pnp_alloc(sizeof(pnp_info_buffer_t));
>>  	if (!buffer)
>>  		return -ENOMEM;
> 
> Wouldn't it be nicer without the unnecessary cast?
> 
> 	pnp_info_buffer_t *buffer = pnp_alloc(sizeof(*buffer));

He deleted the cast (the sizeof(type) versus sizeof(var) thing is mostly 
personal preference).

Rene.

[patch 00/28] PNP: convert fixed tables to lists, v3

[Bjorn Helgaas]

This patch series contains all my PNP patches that are not yet in Linus'
tree, including both the "convert resource table to dynamic list" series
and the "convert resource options to unified dynamic list" series.

This series should replace all the PNP patches in -mm from
    pnp-add-detail-to-debug-resource-dump.patch
    ...
    not-for-merging-pnp-changes-suspend-oops.patch
except for isa-set-24-bit-dma_mask-for-isa-devices.patch, which got
stuck in the middle but isn't really PNP-related.

I'll be on vacation for the rest of the week, so I won't be able to
respond to any issues until Monday.

Here are the significant changes since the last posting:

    - When replacing pnp_resource_table, the PNPACPI resource encoders
      have to explicitly handle disabled resources; previously any
      unused slots in the table were implicitly disabled.  This should
      fix Jiri Slaby's oops on suspend/resume (though he hasn't tested
      this version).
      
      This change is in the patch titled "PNP: replace pnp_resource_table
      with dynamically allocated resources".
      
    - Added a new patch to keep disabled IRQ and DMA resources when parsing
      current config.  This fixes a bug in the current (2.6.25) PNPACPI
      code: consider a device with (mem, irq0, irq1, io), where irq0 is
      disabled.  If we drop irq0 when parsing the _CRS, we will mistakenly
      put irq1 in the irq0 slot when we encode resources for an _SRS call.

    - Added a new patch to avoid interrupts used by an IDE controller in
      compatibility mode.  https://bugzilla.novell.com/show_bug.cgi?id=375836

    - Added a new PNPACPI patch to support HP vendor-specific descriptors.
      This has long been supported by arch/ia64/kernel/acpi-ext.c, which
      provides an hp_acpi_csr_space() interface, but I think it's better
      and more generic to support it in PNPACPI.

Changes since v2 of "convert resource options to unified dynamic list":

    - fixed bisection problem in quirk_ad1815_mpu_resources()
    - fixed checkpatch warning in pnp_show_options()
      (pnp-convert-resource-options-to-single-linked-list-checkpatch-fixes.patch)
    - replaced pnp_independent_option() with literal 0
    - fixed coding style in pnp/manager.c
    - added EXPORT_SYMBOL(pnp_possible_config)
      (pnp-add-pnp_possible_config-can-a-device-could-be-configured-this-way-fix.patch)

Bjorn
-- 

[patch 11/28] PNP: whitespace/coding style fixes

[Bjorn Helgaas]

[-- Attachment #1: pnp-interface-style --]
[-- Type: text/plain, Size: 1897 bytes --]

No functional change; just make a couple declarations
consistent with the rest of the file.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Acked-by: Rene Herman <rene.herman@gmail.com>

Index: work10/drivers/pnp/interface.c
===================================================================
--- work10.orig/drivers/pnp/interface.c	2008-05-09 14:42:25.000000000 -0600
+++ work10/drivers/pnp/interface.c	2008-05-09 14:45:57.000000000 -0600
@@ -216,12 +216,12 @@ static ssize_t pnp_show_options(struct d
 				struct device_attribute *attr, char *buf)
 {
 	struct pnp_dev *dev = to_pnp_dev(dmdev);
+	pnp_info_buffer_t *buffer;
 	struct pnp_option *independent = dev->independent;
 	struct pnp_option *dependent = dev->dependent;
 	int ret, dep = 1;
 
-	pnp_info_buffer_t *buffer = (pnp_info_buffer_t *)
-	    pnp_alloc(sizeof(pnp_info_buffer_t));
+	buffer = pnp_alloc(sizeof(pnp_info_buffer_t));
 	if (!buffer)
 		return -ENOMEM;
 
@@ -248,17 +248,18 @@ static ssize_t pnp_show_current_resource
 					  char *buf)
 {
 	struct pnp_dev *dev = to_pnp_dev(dmdev);
+	pnp_info_buffer_t *buffer;
 	struct pnp_resource *pnp_res;
 	struct resource *res;
 	int ret;
-	pnp_info_buffer_t *buffer;
 
 	if (!dev)
 		return -EINVAL;
 
-	buffer = (pnp_info_buffer_t *) pnp_alloc(sizeof(pnp_info_buffer_t));
+	buffer = pnp_alloc(sizeof(pnp_info_buffer_t));
 	if (!buffer)
 		return -ENOMEM;
+
 	buffer->len = PAGE_SIZE;
 	buffer->buffer = buf;
 	buffer->curr = buffer->buffer;
@@ -295,9 +296,9 @@ static ssize_t pnp_show_current_resource
 	return ret;
 }
 
-static ssize_t
-pnp_set_current_resources(struct device *dmdev, struct device_attribute *attr,
-			  const char *ubuf, size_t count)
+static ssize_t pnp_set_current_resources(struct device *dmdev,
+					 struct device_attribute *attr,
+					 const char *ubuf, size_t count)
 {
 	struct pnp_dev *dev = to_pnp_dev(dmdev);
 	char *buf = (void *)ubuf;

--