Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 466ae837 authored by Shaohua Li's avatar Shaohua Li Committed by Ingo Molnar
Browse files

reduce tlb/cache flush times of agpgart memory allocation



To reduce tlb/cache flush, makes agp memory allocation do one flush
after all pages in a region are changed to uc.

All agp drivers except agp-sgi uses agp_generic_alloc_page()
for .agp_alloc_page, so the patch should work for them. agp-sgi is only
for ia64, so not a problem too.

Signed-off-by: default avatarShaohua Li <shaohua.li@intel.com>
Cc: airlied@linux.ie
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
parent 1ac2f7d5
Loading
Loading
Loading
Loading
+4 −0
Original line number Diff line number Diff line
@@ -30,6 +30,10 @@
#define _AGP_BACKEND_PRIV_H 1

#include <asm/agp.h>	/* for flush_agp_cache() */
#ifndef map_page_into_agp_noflush
#define map_page_into_agp_noflush(page) map_page_into_agp(page)
#define map_page_into_agp_global_flush()
#endif

#define PFX "agpgart: "

+3 −1
Original line number Diff line number Diff line
@@ -274,6 +274,7 @@ struct agp_memory *agp_allocate_memory(struct agp_bridge_data *bridge,
		new->memory[i] = virt_to_gart(addr);
		new->page_count++;
	}
	map_page_into_agp_global_flush();
	new->bridge = bridge;

	return new;
@@ -1186,7 +1187,8 @@ void *agp_generic_alloc_page(struct agp_bridge_data *bridge)
	if (page == NULL)
		return NULL;

	map_page_into_agp(page);
	/* agp_allocate_memory will do flush */
	map_page_into_agp_noflush(page);

	get_page(page);
	atomic_inc(&agp_bridge->current_memory_agp);
+3 −0
Original line number Diff line number Diff line
@@ -15,6 +15,9 @@
#define map_page_into_agp(page) set_pages_uc(page, 1)
#define unmap_page_from_agp(page) set_pages_wb(page, 1)

#define map_page_into_agp_noflush(page) set_pages_uc_noflush(page, 1)
#define map_page_into_agp_global_flush() set_memory_flush_all()

/*
 * Could use CLFLUSH here if the cpu supports it. But then it would
 * need to be called for each cacheline of the whole page so it may