* [PATCH][RFC] Add alloc_pages_exact() and free_pages_exact()
@ 2008-05-13 21:26 Timur Tabi
2008-05-14 14:05 ` Jonathan Corbet
2008-05-14 16:03 ` Randy Dunlap
0 siblings, 2 replies; 3+ messages in thread
From: Timur Tabi @ 2008-05-13 21:26 UTC (permalink / raw)
To: andi, linux-kernel
alloc_pages_exact() is similar to alloc_pages(), except that it allocates
the minimum number of pages to fulfill the request. This is useful if you
want to allocate a very large buffer that is slightly larger than an
even power-of-two number of pages. In that case, alloc_pages() will waste
a lot of memory.
I have a video driver that wants to allocate a 5MB buffer. alloc_pages()
will waste 3MB of physically-contiguous memory. Therefore, I would
like to see alloc_pages_exact() added to 2.6.27.
Please note that I am not a Linux VM expert. I wrote these functions based
on guidance from Andi Kleen. I have no familiarity with NUMA, so I don't know
how to handle that. Any and all suggestions are welcome.
Signed-off-by: Timur Tabi <timur@freescale.com>
---
include/linux/gfp.h | 3 ++
mm/page_alloc.c | 53 +++++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 56 insertions(+), 0 deletions(-)
diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index b414be3..1054801 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -215,6 +215,9 @@ extern struct page *alloc_page_vma(gfp_t gfp_mask,
extern unsigned long __get_free_pages(gfp_t gfp_mask, unsigned int order);
extern unsigned long get_zeroed_page(gfp_t gfp_mask);
+void *alloc_pages_exact(size_t size, gfp_t gfp_mask);
+void free_pages_exact(void *virt, size_t size);
+
#define __get_free_page(gfp_mask) \
__get_free_pages((gfp_mask),0)
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index bdd5c43..2b685eb 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1736,6 +1736,59 @@ void free_pages(unsigned long addr, unsigned int order)
EXPORT_SYMBOL(free_pages);
+/**
+ * Allocate an exact number of pages of physically-contiguous memory.
+ *
+ * This function is similar to alloc_pages(), except that it allocates the
+ * minimum number of pages to satisfy the request. alloc_pages() can only
+ * allocate memory in power-of-two pages.
+ *
+ * This function is also limited by MAX_ORDER.
+ *
+ * Memory allocated by this function must be released by free_pages_exact().
+ */
+void *alloc_pages_exact(size_t size, gfp_t gfp_mask)
+{
+ unsigned int order = get_order(size);
+ unsigned long addr;
+
+ addr = __get_free_pages(gfp_mask, order);
+ if (addr) {
+ unsigned long alloc_end = addr + (PAGE_SIZE << order);
+ unsigned long used = addr + PAGE_ALIGN(size);
+
+ split_page(virt_to_page(addr), order);
+ while (used < alloc_end) {
+ free_page(used);
+ used += PAGE_SIZE;
+ }
+ }
+
+ return (void *)addr;
+}
+EXPORT_SYMBOL(alloc_pages_exact);
+
+/**
+ * Releases memory allocated via alloc_pages_exact()
+ *
+ * Release the memory allocated by a previous call to alloc_pages_exact.
+ *
+ * The first parameter must be the value returned by alloc_pages_exact(),
+ * and the 'size' parameter must be the same as was passed to
+ * alloc_pages_exact().
+ */
+void free_pages_exact(void *virt, size_t size)
+{
+ unsigned long addr = (unsigned long)virt;
+ unsigned long end = addr + PAGE_ALIGN(size);
+
+ while (addr < end) {
+ free_page(addr);
+ addr += PAGE_SIZE;
+ }
+}
+EXPORT_SYMBOL(free_pages_exact);
+
static unsigned int nr_free_zone_pages(int offset)
{
struct zoneref *z;
--
1.5.5
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH][RFC] Add alloc_pages_exact() and free_pages_exact()
2008-05-13 21:26 [PATCH][RFC] Add alloc_pages_exact() and free_pages_exact() Timur Tabi
@ 2008-05-14 14:05 ` Jonathan Corbet
2008-05-14 16:03 ` Randy Dunlap
1 sibling, 0 replies; 3+ messages in thread
From: Jonathan Corbet @ 2008-05-14 14:05 UTC (permalink / raw)
To: Timur Tabi; +Cc: andi, linux-kernel
Timur Tabi writes:
> alloc_pages_exact() is similar to alloc_pages(), except that it allocates
> the minimum number of pages to fulfill the request. This is useful if you
> want to allocate a very large buffer that is slightly larger than an
> even power-of-two number of pages. In that case, alloc_pages() will waste
> a lot of memory.
FWIW, I wished I had something like this when doing the cafe_ccic
driver. I can't really attest to the validity of the core VM stuff, but
I will say that the interface is worth having.
Thanks for doing this!
jon
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH][RFC] Add alloc_pages_exact() and free_pages_exact()
2008-05-13 21:26 [PATCH][RFC] Add alloc_pages_exact() and free_pages_exact() Timur Tabi
2008-05-14 14:05 ` Jonathan Corbet
@ 2008-05-14 16:03 ` Randy Dunlap
1 sibling, 0 replies; 3+ messages in thread
From: Randy Dunlap @ 2008-05-14 16:03 UTC (permalink / raw)
To: Timur Tabi; +Cc: andi, linux-kernel
On Tue, 13 May 2008 16:26:46 -0500 Timur Tabi wrote:
> include/linux/gfp.h | 3 ++
> mm/page_alloc.c | 53 +++++++++++++++++++++++++++++++++++++++++++++++++++
> 2 files changed, 56 insertions(+), 0 deletions(-)
>
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index bdd5c43..2b685eb 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -1736,6 +1736,59 @@ void free_pages(unsigned long addr, unsigned int order)
>
> EXPORT_SYMBOL(free_pages);
Hi Timur,
Please use kernel-doc notation/format for these functions as documented
in Documentation/kernel-doc-nano-HOWTO.txt. Ask me if you have
any questions/problems with it.
> +/**
> + * Allocate an exact number of pages of physically-contiguous memory.
> + *
> + * This function is similar to alloc_pages(), except that it allocates the
> + * minimum number of pages to satisfy the request. alloc_pages() can only
> + * allocate memory in power-of-two pages.
> + *
> + * This function is also limited by MAX_ORDER.
> + *
> + * Memory allocated by this function must be released by free_pages_exact().
> + */
> +void *alloc_pages_exact(size_t size, gfp_t gfp_mask)
> +{
> + unsigned int order = get_order(size);
> + unsigned long addr;
> +
> + addr = __get_free_pages(gfp_mask, order);
> + if (addr) {
> + unsigned long alloc_end = addr + (PAGE_SIZE << order);
> + unsigned long used = addr + PAGE_ALIGN(size);
> +
> + split_page(virt_to_page(addr), order);
> + while (used < alloc_end) {
> + free_page(used);
> + used += PAGE_SIZE;
> + }
> + }
> +
> + return (void *)addr;
> +}
> +EXPORT_SYMBOL(alloc_pages_exact);
> +
> +/**
> + * Releases memory allocated via alloc_pages_exact()
> + *
> + * Release the memory allocated by a previous call to alloc_pages_exact.
> + *
> + * The first parameter must be the value returned by alloc_pages_exact(),
> + * and the 'size' parameter must be the same as was passed to
> + * alloc_pages_exact().
> + */
> +void free_pages_exact(void *virt, size_t size)
> +{
> + unsigned long addr = (unsigned long)virt;
> + unsigned long end = addr + PAGE_ALIGN(size);
> +
> + while (addr < end) {
> + free_page(addr);
> + addr += PAGE_SIZE;
> + }
> +}
> +EXPORT_SYMBOL(free_pages_exact);
---
~Randy
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2008-05-14 16:04 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2008-05-13 21:26 [PATCH][RFC] Add alloc_pages_exact() and free_pages_exact() Timur Tabi
2008-05-14 14:05 ` Jonathan Corbet
2008-05-14 16:03 ` Randy Dunlap
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.