From: "Luis R. Rodriguez" <mcgrof@suse.com>
To: Dan Williams <dan.j.williams@intel.com>
Cc: tglx@linutronix.de, mingo@kernel.org, hpa@zytor.com,
linux-arch@vger.kernel.org, toshi.kani@hp.com,
Arnd Bergmann <arnd@arndb.de>,
linux-nvdimm@lists.01.org, linux-kernel@vger.kernel.org,
rmk+kernel@arm.linux.org.uk, hch@lst.de,
linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v3 05/24] arch: introduce memremap()
Date: Thu, 30 Jul 2015 23:02:04 +0200 [thread overview]
Message-ID: <20150730210204.GQ30479@wotan.suse.de> (raw)
In-Reply-To: <20150730165407.33962.79603.stgit@dwillia2-desk3.amr.corp.intel.com>
On Thu, Jul 30, 2015 at 12:54:07PM -0400, Dan Williams wrote:
> diff --git a/include/linux/io.h b/include/linux/io.h
> index fb5a99800e77..3fcf6256c088 100644
> --- a/include/linux/io.h
> +++ b/include/linux/io.h
> @@ -121,4 +121,13 @@ static inline int arch_phys_wc_index(int handle)
> #endif
> #endif
>
> +enum {
> + /* See memremap() kernel-doc for usage description... */
> + MEMREMAP_WB = 1 << 0,
> + MEMREMAP_WT = 1 << 1,
> +};
Same feedback for enum nameing and also kdoc style.
> diff --git a/kernel/memremap.c b/kernel/memremap.c
> new file mode 100644
> index 000000000000..27637f42f30d
> --- /dev/null
> +++ b/kernel/memremap.c
<-- ... -->
> +/**
> + * memremap() - remap an iomem_resource as cacheable memory
> + * @offset: iomem resource start address
> + * @size: size of remap
> + * @flags: either MEMREMAP_WB or MEMREMAP_WT
> + *
> + * memremap() is "ioremap" for cases where it is known that the resource
> + * being mapped does not have i/o side effects and the __iomem
> + * annotation is not applicable.
> + *
> + * MEMREMAP_WB - matches the default mapping for "System RAM" on
> + * the architecture. This is usually a read-allocate write-back cache.
> + * Morever, if MEMREMAP_WB is specified and the requested remap region is RAM
> + * memremap() will bypass establishing a new mapping and instead return
> + * a pointer into the direct map.
> + *
> + * MEMREMAP_WT - establish a mapping whereby writes either bypass the
> + * cache or are written through to memory and never exist in a
> + * cache-dirty state with respect to program visibility. Attempts to
> + * map "System RAM" with this mapping type will fail.
Then you can extrend all this on kdoc on the enum.
> + */
> +void *memremap(resource_size_t offset, size_t size, unsigned long flags)
> +{
> + int is_ram = region_intersects(offset, size, "System RAM");
This could be the enum region_intersect_type, then if the region enum is
extended you'd get a compiler error if one type was not handled.
Luis
WARNING: multiple messages have this Message-ID (diff)
From: "Luis R. Rodriguez" <mcgrof@suse.com>
To: Dan Williams <dan.j.williams@intel.com>
Cc: tglx@linutronix.de, mingo@kernel.org, hpa@zytor.com,
linux-arch@vger.kernel.org, toshi.kani@hp.com,
Arnd Bergmann <arnd@arndb.de>,
linux-nvdimm@ml01.01.org, linux-kernel@vger.kernel.org,
rmk+kernel@arm.linux.org.uk, hch@lst.de,
linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v3 05/24] arch: introduce memremap()
Date: Thu, 30 Jul 2015 23:02:04 +0200 [thread overview]
Message-ID: <20150730210204.GQ30479@wotan.suse.de> (raw)
In-Reply-To: <20150730165407.33962.79603.stgit@dwillia2-desk3.amr.corp.intel.com>
On Thu, Jul 30, 2015 at 12:54:07PM -0400, Dan Williams wrote:
> diff --git a/include/linux/io.h b/include/linux/io.h
> index fb5a99800e77..3fcf6256c088 100644
> --- a/include/linux/io.h
> +++ b/include/linux/io.h
> @@ -121,4 +121,13 @@ static inline int arch_phys_wc_index(int handle)
> #endif
> #endif
>
> +enum {
> + /* See memremap() kernel-doc for usage description... */
> + MEMREMAP_WB = 1 << 0,
> + MEMREMAP_WT = 1 << 1,
> +};
Same feedback for enum nameing and also kdoc style.
> diff --git a/kernel/memremap.c b/kernel/memremap.c
> new file mode 100644
> index 000000000000..27637f42f30d
> --- /dev/null
> +++ b/kernel/memremap.c
<-- ... -->
> +/**
> + * memremap() - remap an iomem_resource as cacheable memory
> + * @offset: iomem resource start address
> + * @size: size of remap
> + * @flags: either MEMREMAP_WB or MEMREMAP_WT
> + *
> + * memremap() is "ioremap" for cases where it is known that the resource
> + * being mapped does not have i/o side effects and the __iomem
> + * annotation is not applicable.
> + *
> + * MEMREMAP_WB - matches the default mapping for "System RAM" on
> + * the architecture. This is usually a read-allocate write-back cache.
> + * Morever, if MEMREMAP_WB is specified and the requested remap region is RAM
> + * memremap() will bypass establishing a new mapping and instead return
> + * a pointer into the direct map.
> + *
> + * MEMREMAP_WT - establish a mapping whereby writes either bypass the
> + * cache or are written through to memory and never exist in a
> + * cache-dirty state with respect to program visibility. Attempts to
> + * map "System RAM" with this mapping type will fail.
Then you can extrend all this on kdoc on the enum.
> + */
> +void *memremap(resource_size_t offset, size_t size, unsigned long flags)
> +{
> + int is_ram = region_intersects(offset, size, "System RAM");
This could be the enum region_intersect_type, then if the region enum is
extended you'd get a compiler error if one type was not handled.
Luis
WARNING: multiple messages have this Message-ID (diff)
From: mcgrof@suse.com (Luis R. Rodriguez)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v3 05/24] arch: introduce memremap()
Date: Thu, 30 Jul 2015 23:02:04 +0200 [thread overview]
Message-ID: <20150730210204.GQ30479@wotan.suse.de> (raw)
In-Reply-To: <20150730165407.33962.79603.stgit@dwillia2-desk3.amr.corp.intel.com>
On Thu, Jul 30, 2015 at 12:54:07PM -0400, Dan Williams wrote:
> diff --git a/include/linux/io.h b/include/linux/io.h
> index fb5a99800e77..3fcf6256c088 100644
> --- a/include/linux/io.h
> +++ b/include/linux/io.h
> @@ -121,4 +121,13 @@ static inline int arch_phys_wc_index(int handle)
> #endif
> #endif
>
> +enum {
> + /* See memremap() kernel-doc for usage description... */
> + MEMREMAP_WB = 1 << 0,
> + MEMREMAP_WT = 1 << 1,
> +};
Same feedback for enum nameing and also kdoc style.
> diff --git a/kernel/memremap.c b/kernel/memremap.c
> new file mode 100644
> index 000000000000..27637f42f30d
> --- /dev/null
> +++ b/kernel/memremap.c
<-- ... -->
> +/**
> + * memremap() - remap an iomem_resource as cacheable memory
> + * @offset: iomem resource start address
> + * @size: size of remap
> + * @flags: either MEMREMAP_WB or MEMREMAP_WT
> + *
> + * memremap() is "ioremap" for cases where it is known that the resource
> + * being mapped does not have i/o side effects and the __iomem
> + * annotation is not applicable.
> + *
> + * MEMREMAP_WB - matches the default mapping for "System RAM" on
> + * the architecture. This is usually a read-allocate write-back cache.
> + * Morever, if MEMREMAP_WB is specified and the requested remap region is RAM
> + * memremap() will bypass establishing a new mapping and instead return
> + * a pointer into the direct map.
> + *
> + * MEMREMAP_WT - establish a mapping whereby writes either bypass the
> + * cache or are written through to memory and never exist in a
> + * cache-dirty state with respect to program visibility. Attempts to
> + * map "System RAM" with this mapping type will fail.
Then you can extrend all this on kdoc on the enum.
> + */
> +void *memremap(resource_size_t offset, size_t size, unsigned long flags)
> +{
> + int is_ram = region_intersects(offset, size, "System RAM");
This could be the enum region_intersect_type, then if the region enum is
extended you'd get a compiler error if one type was not handled.
Luis
next prev parent reply other threads:[~2015-07-30 21:02 UTC|newest]
Thread overview: 118+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-07-30 16:53 [PATCH v3 00/24] replace ioremap_{cache|wt} with memremap Dan Williams
2015-07-30 16:53 ` Dan Williams
2015-07-30 16:53 ` Dan Williams
2015-07-30 16:53 ` Dan Williams
2015-07-30 16:53 ` Dan Williams
2015-07-30 16:53 ` [PATCH v3 01/24] mm: enhance region_is_ram() to region_intersects() Dan Williams
2015-07-30 16:53 ` Dan Williams
2015-07-30 16:53 ` Dan Williams
2015-07-30 20:42 ` Luis R. Rodriguez
2015-07-30 20:42 ` Luis R. Rodriguez
2015-07-30 20:42 ` Luis R. Rodriguez
2015-07-30 20:44 ` Dan Williams
2015-07-30 20:44 ` Dan Williams
2015-07-30 20:44 ` Dan Williams
2015-07-30 20:54 ` Luis R. Rodriguez
2015-07-30 20:54 ` Luis R. Rodriguez
2015-07-30 20:54 ` Luis R. Rodriguez
2015-07-30 20:59 ` Dan Williams
2015-07-30 20:59 ` Dan Williams
2015-07-30 20:59 ` Dan Williams
2015-07-30 21:10 ` Luis R. Rodriguez
2015-07-30 21:10 ` Luis R. Rodriguez
2015-07-30 21:10 ` Luis R. Rodriguez
2015-07-30 20:58 ` Luis R. Rodriguez
2015-07-30 20:58 ` Luis R. Rodriguez
2015-07-30 20:58 ` Luis R. Rodriguez
2015-07-30 16:53 ` [PATCH v3 02/24] arch, drivers: don't include <asm/io.h> directly, use <linux/io.h> instead Dan Williams
2015-07-30 16:53 ` Dan Williams
2015-07-30 16:53 ` Dan Williams
2015-07-30 16:53 ` [PATCH v3 03/24] cleanup IORESOURCE_CACHEABLE vs ioremap() Dan Williams
2015-07-30 16:53 ` Dan Williams
2015-07-30 16:53 ` Dan Williams
2015-07-30 16:54 ` [PATCH v3 04/24] intel_iommu: fix leaked ioremap mapping Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-08-03 14:26 ` Joerg Roedel
2015-08-03 14:26 ` Joerg Roedel
2015-08-03 14:26 ` Joerg Roedel
2015-07-30 16:54 ` [PATCH v3 05/24] arch: introduce memremap() Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 21:02 ` Luis R. Rodriguez [this message]
2015-07-30 21:02 ` Luis R. Rodriguez
2015-07-30 21:02 ` Luis R. Rodriguez
2015-07-30 21:11 ` Dan Williams
2015-07-30 21:11 ` Dan Williams
2015-07-30 21:11 ` Dan Williams
2015-07-30 16:54 ` [PATCH v3 06/24] arm: switch from ioremap_cache to memremap Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` [PATCH v3 07/24] x86: " Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` [PATCH v3 08/24] gma500: switch from acpi_os_ioremap " Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` [PATCH v3 09/24] i915: " Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` [PATCH v3 10/24] acpi: switch from ioremap_cache " Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` [PATCH v3 11/24] toshiba laptop: replace ioremap_cache with ioremap Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` [PATCH v3 12/24] memconsole: fix __iomem mishandling, switch to memremap Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` [PATCH v3 13/24] visorbus: switch from ioremap_cache " Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-08-06 0:27 ` Greg Kroah-Hartman
2015-08-06 0:27 ` Greg Kroah-Hartman
2015-08-06 0:27 ` Greg Kroah-Hartman
2015-07-30 16:54 ` [PATCH v3 14/24] intel-iommu: " Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:54 ` Dan Williams
2015-07-30 16:55 ` [PATCH v3 15/24] libnvdimm, pmem: push call to ioremap_cache out of line Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` [PATCH v3 16/24] pxa2xx-flash: switch from ioremap_cache to memremap Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` [PATCH v3 17/24] sfi: " Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` [PATCH v3 18/24] fbdev: switch from ioremap_wt " Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` [PATCH v3 19/24] pmem: " Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` [PATCH v3 20/24] arch: kill ioremap_cached() Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` [PATCH v3 21/24] arch: kill ioremap_fullcache() Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` [PATCH v3 22/24] arch: remove ioremap_cache, replace with arch_memremap Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` [PATCH v3 23/24] arch: remove ioremap_wt, optionally " Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` [PATCH v3 24/24] pmem: convert to generic memremap Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-07-30 16:55 ` Dan Williams
2015-08-03 17:21 ` Ross Zwisler
2015-08-03 17:21 ` Ross Zwisler
2015-08-03 17:21 ` Ross Zwisler
2015-08-03 18:01 ` Dan Williams
2015-08-03 18:01 ` Dan Williams
2015-08-03 18:01 ` Dan Williams
2015-08-04 16:53 ` Ross Zwisler
2015-08-04 16:53 ` Ross Zwisler
2015-08-04 16:53 ` Ross Zwisler
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20150730210204.GQ30479@wotan.suse.de \
--to=mcgrof@suse.com \
--cc=arnd@arndb.de \
--cc=dan.j.williams@intel.com \
--cc=hch@lst.de \
--cc=hpa@zytor.com \
--cc=linux-arch@vger.kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-nvdimm@lists.01.org \
--cc=mingo@kernel.org \
--cc=rmk+kernel@arm.linux.org.uk \
--cc=tglx@linutronix.de \
--cc=toshi.kani@hp.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.