64 bit s390 and 2.4.0-test11 (was Memory management bug)

[schwidefsky@de.ibm.com]

[-- Attachment #1: Type: text/plain, Size: 1960 bytes --]




Hi,
good news (at least for us): linux on the 64 bit S/390 (aka zServer)
is now running pretty stable. Our implementation of ptep_get_and_clear
didn't clear the pte if the invalid bit was already set. But a swapped
page has the invalid bit set too and in that case we didn't clear the
pte. That did cause the BUG() in swap_state.c:60.

With this new backend in mind I'd like to suggest two small changes
for the common code.
1) move establish_pte to the architecture dependent folders. Our
implementation looks like this:

static inline void ptep_invalidate_and_flush(pte_t *ptep, unsigned long
addr)
{
        if (!(pte_val(*ptep) & _PAGE_INVALID))
                __asm__ __volatile__ ("ipte %0,%1" : : "a" (ptep), "a"
(addr));
}

static inline void establish_pte(struct vm_area_struct * vma,
                                 unsigned long address,
                                 pte_t *page_table, pte_t entry)
{
        ptep_invalidate_and_flush(page_table, address);
        set_pte(page_table, entry);
}

The question I face at the moment is: where is the right place in
include/asm for establish_pte. I added it to include/asm-generic/pgtable.h
and include/asm-i386/pgtable.h but I now face the problem that the
default implementation has a call to flush_tlb_page and that is defined
in pgalloc.h. I added an #include <asm/pgalloc.h> but I fear that this
could
cause compile errors. For details see establish_pte.diff.

2) add a check for EI_CLASS in the binfmt_elf loader. We will use the
same EM_S390 for 31 bit and 64 bit binaries. The distinction is done
by means of the EI_CLASS byte in the ELF header. See binfmt_elf.diff

blue skies,
   Martin

Linux/390 Design & Development, IBM Deutschland Entwicklung GmbH
Schönaicherstr. 220, D-71032 Böblingen, Telefon: 49 - (0)7031 - 16-2247
E-Mail: schwidefsky@de.ibm.com

(See attached file: establish_pte.diff)(See attached file: binfmt_elf.diff)

[-- Attachment #2: establish_pte.diff --]
[-- Type: application/octet-stream, Size: 2700 bytes --]

diff -u -r --new-file linux-2.4.0-test11/include/asm-generic/pgtable.h linux-2.4.0-test11-establish_pte/include/asm-generic/pgtable.h
--- linux-2.4.0-test11/include/asm-generic/pgtable.h	Fri Oct 20 00:51:16 2000
+++ linux-2.4.0-test11-establish_pte/include/asm-generic/pgtable.h	Wed Dec  6 20:52:00 2000
@@ -1,6 +1,8 @@
 #ifndef _ASM_GENERIC_PGTABLE_H
 #define _ASM_GENERIC_PGTABLE_H

+#include <asm/pgalloc.h>
+
 static inline int ptep_test_and_clear_young(pte_t *ptep)
 {
 	pte_t pte = *ptep;
@@ -40,4 +40,17 @@
 
 #define pte_same(A,B)	(pte_val(A) == pte_val(B))
 
+/*
+ * Establish a new mapping:
+ *  - flush the old one
+ *  - update the page tables
+ *  - inform the TLB about the new one
+ */
+static inline void establish_pte(struct vm_area_struct * vma, unsigned long address, pte_t *page_table, pte_t entry)
+{
+	set_pte(page_table, entry);
+	flush_tlb_page(vma, address);
+	update_mmu_cache(vma, address, entry);
+}
+
 #endif /* _ASM_GENERIC_PGTABLE_H */
diff -u -r --new-file linux-2.4.0-test11/include/asm-i386/pgtable.h linux-2.4.0-test11-establish_pte/include/asm-i386/pgtable.h
--- linux-2.4.0-test11/include/asm-i386/pgtable.h	Sun Nov 19 05:56:59 2000
+++ linux-2.4.0-test11-establish_pte/include/asm-i386/pgtable.h	Wed Dec  6 20:53:43 2000
@@ -287,6 +287,20 @@
 static inline void ptep_mkdirty(pte_t *ptep)			{ set_bit(_PAGE_BIT_RW, ptep); }
 
 /*
+ * Establish a new mapping:
+ *  - flush the old one
+ *  - update the page tables
+ *  - inform the TLB about the new one
+ */
+static inline void establish_pte(struct vm_area_struct * vma, unsigned long address, pte_t *page_table, pte_t entry)
+{
+	set_pte(page_table, entry);
+	if (vma->vm_mm == current->active_mm)
+		__flush_tlb_one(addr);
+	update_mmu_cache(vma, address, entry);
+}
+
+/*
  * Conversion functions: convert a page and protection to a page entry,
  * and a page entry and page directory to the page they refer to.
  */
diff -u -r --new-file linux-2.4.0-test11/mm/memory.c linux-2.4.0-test11-establish_pte/mm/memory.c
--- linux-2.4.0-test11/mm/memory.c	Wed Nov  1 16:45:09 2000
+++ linux-2.4.0-test11-establish_pte/mm/memory.c	Wed Dec  6 20:52:00 2000
@@ -773,19 +773,6 @@
 	return error;
 }
 
-/*
- * Establish a new mapping:
- *  - flush the old one
- *  - update the page tables
- *  - inform the TLB about the new one
- */
-static inline void establish_pte(struct vm_area_struct * vma, unsigned long address, pte_t *page_table, pte_t entry)
-{
-	set_pte(page_table, entry);
-	flush_tlb_page(vma, address);
-	update_mmu_cache(vma, address, entry);
-}
-
 static inline void break_cow(struct vm_area_struct * vma, struct page *	old_page, struct page * new_page, unsigned long address, 
 		pte_t *page_table)
 {

[-- Attachment #3: binfmt_elf.diff --]
[-- Type: application/octet-stream, Size: 724 bytes --]

diff -u -r --new-file linux-2.4.0-test11/fs/binfmt_elf.c linux-2.4.0-test11-establish_pte/fs/binfmt_elf.c
--- linux-2.4.0-test11/fs/binfmt_elf.c	Fri Oct 27 20:04:43 2000
+++ linux-2.4.0-test11-establish_pte/fs/binfmt_elf.c	Wed Dec  6 20:52:00 2000
@@ -247,6 +247,8 @@
 		goto out;
 	if (!elf_check_arch(interp_elf_ex))
 		goto out;
+        if (interp_elf_ex->e_ident[EI_CLASS] != ELF_CLASS)
+                goto out;
 	if (!interpreter->f_op->mmap)
 		goto out;
 
@@ -422,6 +424,8 @@
 	if (elf_ex.e_type != ET_EXEC && elf_ex.e_type != ET_DYN)
 		goto out;
 	if (!elf_check_arch(&elf_ex))
+		goto out;
+	if (elf_ex.e_ident[EI_CLASS] != ELF_CLASS)
 		goto out;
 	if (!bprm->file->f_op||!bprm->file->f_op->mmap)
 		goto out;