qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Tristan Gingold <gingold@adacore.com>
To: qemu-devel@nongnu.org
Subject: [Qemu-devel] [PATCH 01/25] Add support for multi-level phys map.
Date: Tue, 24 Mar 2009 16:47:43 +0100	[thread overview]
Message-ID: <1237909687-31711-2-git-send-email-gingold@adacore.com> (raw)
In-Reply-To: <1237909687-31711-1-git-send-email-gingold@adacore.com>

As Alpha physical addresses are 44 bits, l1_phys_map can't be anymore 2 levels.
Use a more generic multi-level approach and explain why we don't need to
extend l1_map.

Signed-off-by: Tristan Gingold <gingold@adacore.com>
---
 exec.c |   50 ++++++++++++++++++++++++++++++++++++++------------
 1 files changed, 38 insertions(+), 12 deletions(-)

diff --git a/exec.c b/exec.c
index e869a20..4330003 100644
--- a/exec.c
+++ b/exec.c
@@ -147,14 +147,17 @@ typedef struct PhysPageDesc {
 } PhysPageDesc;
 
 #define L2_BITS 10
-#if defined(CONFIG_USER_ONLY) && defined(TARGET_VIRT_ADDR_SPACE_BITS)
-/* XXX: this is a temporary hack for alpha target.
- *      In the future, this is to be replaced by a multi-level table
- *      to actually be able to handle the complete 64 bits address space.
- */
-#define L1_BITS (TARGET_VIRT_ADDR_SPACE_BITS - L2_BITS - TARGET_PAGE_BITS)
+
+#if TARGET_PHYS_ADDR_SPACE_BITS > (32 + 32 - L2_BITS - TARGET_PAGE_BITS)
+# define MULTILEVEL_PHYS_MAP
+# define L1_BITS_ ((TARGET_PHYS_ADDR_SPACE_BITS - TARGET_PAGE_BITS) % L2_BITS)
+# if L1_BITS_ < 4 /* avoid ridiculous small l1 */
+#  define L1_BITS (L1_BITS_ + L2_BITS)
+# else
+#  define L1_BITS L1_BITS_
+# endif
 #else
-#define L1_BITS (32 - L2_BITS - TARGET_PAGE_BITS)
+# define L1_BITS (32 - L2_BITS - TARGET_PAGE_BITS)
 #endif
 
 #define L1_SIZE (1 << L1_BITS)
@@ -165,7 +168,9 @@ unsigned long qemu_host_page_bits;
 unsigned long qemu_host_page_size;
 unsigned long qemu_host_page_mask;
 
-/* XXX: for system emulation, it could just be an array */
+/* XXX: for system emulation, it could just be an array.  As this is currently
+   a two level map this limits the size of RAM memory that can contains
+   target code.  In practice this is large enough (>= 4GB) */
 static PageDesc *l1_map[L1_SIZE];
 static PhysPageDesc **l1_phys_map;
 
@@ -339,12 +344,31 @@ static PhysPageDesc *phys_page_find_alloc(target_phys_addr_t index, int alloc)
     void **lp, **p;
     PhysPageDesc *pd;
 
+#ifdef MULTILEVEL_PHYS_MAP
+    int i;
+#define L1_SHIFT (TARGET_PHYS_ADDR_SPACE_BITS - TARGET_PAGE_BITS - L1_BITS)
+
+    /* Level 1.  */
     p = (void **)l1_phys_map;
-#if TARGET_PHYS_ADDR_SPACE_BITS > 32
+    lp = p + ((index >> L1_SHIFT) & (L1_SIZE - 1));
 
-#if TARGET_PHYS_ADDR_SPACE_BITS > (32 + L1_BITS)
-#error unsupported TARGET_PHYS_ADDR_SPACE_BITS
-#endif
+    /* Level 2..n-1 */
+    for (i = (L1_SHIFT / L2_BITS) - 1; i > 0; i--) {
+        p = *lp;
+        if (!p) {
+            /* allocate if not found */
+            if (!alloc)
+                return NULL;
+            p = qemu_vmalloc(sizeof(void *) * L2_SIZE);
+            memset(p, 0, sizeof(void *) * L2_SIZE);
+            *lp = p;
+        }
+        lp = p + ((index >> (i * L2_BITS)) & (L2_SIZE - 1));
+    }
+#else
+    p = (void **)l1_phys_map;
+
+#if TARGET_PHYS_ADDR_SPACE_BITS > 32
     lp = p + ((index >> (L1_BITS + L2_BITS)) & (L1_SIZE - 1));
     p = *lp;
     if (!p) {
@@ -357,6 +381,8 @@ static PhysPageDesc *phys_page_find_alloc(target_phys_addr_t index, int alloc)
     }
 #endif
     lp = p + ((index >> L2_BITS) & (L1_SIZE - 1));
+#endif
+
     pd = *lp;
     if (!pd) {
         int i;
-- 
1.6.2

  reply	other threads:[~2009-03-24 15:48 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-03-24 15:47 [Qemu-devel] [PATCH 0/25]: add alpha es40 system emulation (v3) Tristan Gingold
2009-03-24 15:47 ` Tristan Gingold [this message]
2009-03-24 15:47   ` [Qemu-devel] [PATCH 02/25] Fix bug: palcode is at least 6 bits Tristan Gingold
2009-03-24 15:47     ` [Qemu-devel] [PATCH 03/25] Fix bug: do not mask address LSBs for ldwu Tristan Gingold
2009-03-24 15:47       ` [Qemu-devel] [PATCH 04/25] Fix bug: integer conditionnal branch offset is 21 bits wide Tristan Gingold
2009-03-24 15:47         ` [Qemu-devel] [PATCH 05/25] bug fix: avoid nop to override next instruction Tristan Gingold
2009-03-24 15:47           ` [Qemu-devel] [PATCH 06/25] Fix temp free for hw_st Tristan Gingold
2009-03-24 15:47             ` [Qemu-devel] [PATCH 07/25] Increase Alpha physical address size to 44 bits Tristan Gingold
2009-03-24 15:47               ` [Qemu-devel] [PATCH 08/25] Alpha: set target page size to 13 bits Tristan Gingold
2009-03-24 15:47                 ` [Qemu-devel] [PATCH 09/25] Allow 5 mmu indexes Tristan Gingold
2009-03-24 15:47                   ` [Qemu-devel] [PATCH 10/25] Split cpu_mmu_index into cpu_mmu_index_data and cpu_mmu_index_code Tristan Gingold
2009-03-24 15:47                     ` [Qemu-devel] [PATCH 11/25] Add square wave output support Tristan Gingold
2009-03-24 15:47                       ` [Qemu-devel] [PATCH 12/25] Add ali1543 super IO pci device Tristan Gingold
2009-03-24 15:47                         ` [Qemu-devel] [PATCH 13/25] Add 21272 chipset (memory and pci controller for alpha) Tristan Gingold
2009-03-24 15:47                           ` [Qemu-devel] [PATCH 14/25] Add target-alpha/machine.c and hw/es40.c for es40 machine emulation Tristan Gingold
2009-03-24 15:47                             ` [Qemu-devel] [PATCH 15/25] Move softmmu_helper.h from exec.h to op_helper.c on alpha Tristan Gingold
2009-03-24 15:47                               ` [Qemu-devel] [PATCH 16/25] Document which IPR are used by 21264 Tristan Gingold
2009-03-24 15:47                                 ` [Qemu-devel] [PATCH 17/25] tb_flush helper should flush the tb (and not the tlb) Tristan Gingold
2009-03-24 15:48                                   ` [Qemu-devel] [PATCH 18/25] Add instruction name in comments for hw_ld opcode Tristan Gingold
2009-03-24 15:48                                     ` [Qemu-devel] [PATCH 19/25] Remove PALCODE_ declarations (unused) Tristan Gingold
2009-03-24 15:48                                       ` [Qemu-devel] [PATCH 20/25] alpha ld helpers now directly return the value Tristan Gingold
2009-03-24 15:48                                         ` [Qemu-devel] [PATCH 21/25] Add alpha_cpu_list Tristan Gingold
2009-03-24 15:48                                           ` [Qemu-devel] [PATCH 22/25] Alpha: lower parent irq when irq is lowered Tristan Gingold
2009-03-24 15:48                                             ` [Qemu-devel] [PATCH 23/25] Move linux-user pal emulation to linux-user/ Tristan Gingold
2009-03-24 15:48                                               ` [Qemu-devel] [PATCH 24/25] Correctly decode hw_ld/hw_st opcodes for all alpha implementations Tristan Gingold
2009-03-24 15:48                                                 ` [Qemu-devel] [PATCH 25/25] Add full emulation for 21264 Tristan Gingold
2009-03-24 23:00                           ` [Qemu-devel] [PATCH 13/25] Add 21272 chipset (memory and pci controller for alpha) Robert Reif
2009-03-25  7:58                             ` Tristan Gingold
2009-03-25  8:09                             ` Tristan Gingold
2009-03-29  0:37                   ` [Qemu-devel] [PATCH 09/25] Allow 5 mmu indexes Aurelien Jarno
2009-03-24 16:46   ` [Qemu-devel] [PATCH 01/25] Add support for multi-level phys map Paul Brook
2009-03-24 19:42 ` [Qemu-devel] [PATCH 0/25]: add alpha es40 system emulation (v3) Brian Wheeler
2009-03-25  7:37   ` Tristan Gingold
2009-03-25 12:43     ` Brian Wheeler
2009-03-25 12:53       ` Tristan Gingold
2009-03-29  0:14 ` Aurelien Jarno
2009-03-29  0:31   ` Aurelien Jarno

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=1237909687-31711-2-git-send-email-gingold@adacore.com \
    --to=gingold@adacore.com \
    --cc=qemu-devel@nongnu.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).