From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: <linuxppc-dev@ozlabs.org>
Cc: Paul Mackerras <paulus@samba.org>, Christoph Hellwig <hch@lst.de>,
cbe-oss-dev@ozlabs.org
Subject: [PATCH 1/21] unmap_vm_area becomes unmap_kernel_range for the public
Date: Mon, 04 Jun 2007 15:15:35 +1000 [thread overview]
Message-ID: <20070604051539.C5819DDE33@ozlabs.org> (raw)
In-Reply-To: <1180934134.603289.870346178920.qpush@grosgo>
This patch makes unmap_vm_area static and a wrapper around a new
exported unmap_kernel_range that takes an explicit range instead
of a vm_area struct.
This makes it more versatile for code that wants to play with kernel
page tables outside of the standard vmalloc area.
(One example is some rework of the PowerPC PCI IO space mapping
code that depends on that patch and removes some code duplication
and horrible abuse of forged struct vm_struct).
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
This patch is already in -mm
Documentation/cachetlb.txt | 2 +-
arch/powerpc/mm/imalloc.c | 3 ++-
arch/powerpc/mm/pgtable_64.c | 1 -
include/linux/vmalloc.h | 3 ++-
mm/vmalloc.c | 13 +++++++++----
5 files changed, 14 insertions(+), 8 deletions(-)
Index: linux-cell/Documentation/cachetlb.txt
===================================================================
--- linux-cell.orig/Documentation/cachetlb.txt 2007-05-24 14:00:32.000000000 +1000
+++ linux-cell/Documentation/cachetlb.txt 2007-05-24 14:01:31.000000000 +1000
@@ -253,7 +253,7 @@ Here are the routines, one by one:
The first of these two routines is invoked after map_vm_area()
has installed the page table entries. The second is invoked
- before unmap_vm_area() deletes the page table entries.
+ before unmap_kernel_range() deletes the page table entries.
There exists another whole class of cpu cache issues which currently
require a whole different set of interfaces to handle properly.
Index: linux-cell/arch/powerpc/mm/imalloc.c
===================================================================
--- linux-cell.orig/arch/powerpc/mm/imalloc.c 2007-05-24 14:00:32.000000000 +1000
+++ linux-cell/arch/powerpc/mm/imalloc.c 2007-05-24 14:01:31.000000000 +1000
@@ -301,7 +301,8 @@ void im_free(void * addr)
for (p = &imlist ; (tmp = *p) ; p = &tmp->next) {
if (tmp->addr == addr) {
*p = tmp->next;
- unmap_vm_area(tmp);
+ unmap_kernel_range((unsigned long)tmp->addr,
+ tmp->size);
kfree(tmp);
mutex_unlock(&imlist_mutex);
return;
Index: linux-cell/arch/powerpc/mm/pgtable_64.c
===================================================================
--- linux-cell.orig/arch/powerpc/mm/pgtable_64.c 2007-05-24 14:00:32.000000000 +1000
+++ linux-cell/arch/powerpc/mm/pgtable_64.c 2007-05-24 14:01:31.000000000 +1000
@@ -240,7 +240,6 @@ int __ioremap_explicit(phys_addr_t pa, u
/*
* Unmap an IO region and remove it from imalloc'd list.
* Access to IO memory should be serialized by driver.
- * This code is modeled after vmalloc code - unmap_vm_area()
*
* XXX what about calls before mem_init_done (ie python_countermeasures())
*/
Index: linux-cell/include/linux/vmalloc.h
===================================================================
--- linux-cell.orig/include/linux/vmalloc.h 2007-05-24 14:00:32.000000000 +1000
+++ linux-cell/include/linux/vmalloc.h 2007-05-24 14:01:31.000000000 +1000
@@ -65,9 +65,10 @@ extern struct vm_struct *get_vm_area_nod
unsigned long flags, int node,
gfp_t gfp_mask);
extern struct vm_struct *remove_vm_area(void *addr);
+
extern int map_vm_area(struct vm_struct *area, pgprot_t prot,
struct page ***pages);
-extern void unmap_vm_area(struct vm_struct *area);
+extern void unmap_kernel_range(unsigned long addr, unsigned long size);
/*
* Internals. Dont't use..
Index: linux-cell/mm/vmalloc.c
===================================================================
--- linux-cell.orig/mm/vmalloc.c 2007-05-24 14:00:32.000000000 +1000
+++ linux-cell/mm/vmalloc.c 2007-05-24 14:02:03.000000000 +1000
@@ -68,12 +68,12 @@ static inline void vunmap_pud_range(pgd_
} while (pud++, addr = next, addr != end);
}
-void unmap_vm_area(struct vm_struct *area)
+void unmap_kernel_range(unsigned long addr, unsigned long size)
{
pgd_t *pgd;
unsigned long next;
- unsigned long addr = (unsigned long) area->addr;
- unsigned long end = addr + area->size;
+ unsigned long start = addr;
+ unsigned long end = addr + size;
BUG_ON(addr >= end);
pgd = pgd_offset_k(addr);
@@ -84,7 +84,12 @@ void unmap_vm_area(struct vm_struct *are
continue;
vunmap_pud_range(pgd, addr, next);
} while (pgd++, addr = next, addr != end);
- flush_tlb_kernel_range((unsigned long) area->addr, end);
+ flush_tlb_kernel_range(start, end);
+}
+
+static void unmap_vm_area(struct vm_struct *area)
+{
+ unmap_kernel_range((unsigned long)area->addr, area->size);
}
static int vmap_pte_range(pmd_t *pmd, unsigned long addr,
next prev parent reply other threads:[~2007-06-04 5:15 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-06-04 5:15 [PATCH 0/21] This is my pending series for 2.6.23 Benjamin Herrenschmidt
2007-06-04 5:15 ` Benjamin Herrenschmidt [this message]
2007-06-04 5:15 ` [PATCH 2/21] powerpc: Rewrite IO allocation & mapping on powerpc64 Benjamin Herrenschmidt
2007-06-04 5:15 ` [PATCH 3/21] spufs: Add support for SPU single stepping Benjamin Herrenschmidt
2007-06-04 13:06 ` Jeremy Kerr
2007-06-04 5:15 ` [PATCH 4/21] spufs: Add a "capabilities" file to spu contexts Benjamin Herrenschmidt
2007-06-04 13:06 ` Jeremy Kerr
2007-06-04 5:15 ` [PATCH 5/21] powerpc: Disable broken PPC_PTRACE_GETFPREGS on 32 bits Benjamin Herrenschmidt
2007-06-04 5:15 ` [PATCH 6/21] powerpc: ptrace cleanups Benjamin Herrenschmidt
2007-06-04 5:15 ` [PATCH 7/21] powerpc: ptrace updates & new better requests Benjamin Herrenschmidt
2007-06-04 5:15 ` [PATCH 8/21] powerpc: uninline common ptrace bits Benjamin Herrenschmidt
2007-06-04 5:15 ` [PATCH 9/21] powerpc: remove some useless ifdef's in ptrace Benjamin Herrenschmidt
2007-06-04 5:15 ` [PATCH 10/21] powerpc: Allow ptrace write to pt_regs trap and orig_r3 Benjamin Herrenschmidt
2007-06-04 5:15 ` [PATCH 11/21] powerpc: ptrace shouldn't touch FP exec mode Benjamin Herrenschmidt
2007-06-04 5:15 ` [PATCH 12/21] powerpc: ptrace can set DABR on both 32 and 64 bits Benjamin Herrenschmidt
2007-06-04 5:15 ` [PATCH 13/21] powerpc: Always apply DABR changes on context switches Benjamin Herrenschmidt
2007-06-04 5:15 ` [PATCH 14/21] powerpc: Make syscall restart code more common Benjamin Herrenschmidt
2007-06-04 7:06 ` Christoph Hellwig
2007-06-04 5:15 ` [PATCH 15/21] powerpc: consolidate sys_sigaltstack Benjamin Herrenschmidt
2007-06-04 5:15 ` [PATCH 16/21] powerpc: consolidate restore_sigmask Benjamin Herrenschmidt
2007-06-04 5:15 ` [PATCH 17/21] powerpc: consolidate do_signal Benjamin Herrenschmidt
2007-06-04 5:15 ` [PATCH 18/21] powerpc: Remove obsolete freezer bits Benjamin Herrenschmidt
2007-06-04 5:15 ` [PATCH 19/21] powerpc: Merge creation of signal frame Benjamin Herrenschmidt
2007-06-04 7:22 ` PATCH 19/21] powerpc: Merge creation of signal frame (#2) Benjamin Herrenschmidt
2007-06-04 8:01 ` Christoph Hellwig
2007-06-04 10:04 ` Benjamin Herrenschmidt
2007-06-04 5:15 ` [PATCH 20/21] powerpc: remove #ifdef around set_dabr in signal code Benjamin Herrenschmidt
2007-06-04 5:15 ` [PATCH 21/21] powerpc: Less ifdef's in signal.c/signal.h Benjamin Herrenschmidt
2007-06-04 7:00 ` [PATCH 0/21] This is my pending series for 2.6.23 Benjamin Herrenschmidt
2007-06-04 7:21 ` Benjamin Herrenschmidt
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=20070604051539.C5819DDE33@ozlabs.org \
--to=benh@kernel.crashing.org \
--cc=cbe-oss-dev@ozlabs.org \
--cc=hch@lst.de \
--cc=linuxppc-dev@ozlabs.org \
--cc=paulus@samba.org \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).