From: Alexander Graf <agraf@suse.de>
To: qemu-devel qemu-devel <qemu-devel@nongnu.org>
Cc: Blue Swirl <blauwirbel@gmail.com>,
qemu-ppc Mailing List <qemu-ppc@nongnu.org>,
Aurelien Jarno <aurelien@aurel32.net>
Subject: [Qemu-devel] [PATCH 23/72] PPC: mpc8544ds: Span initial TLB entry over as much RAM as we need
Date: Sun, 24 Jun 2012 01:06:47 +0200 [thread overview]
Message-ID: <1340492856-21126-24-git-send-email-agraf@suse.de> (raw)
In-Reply-To: <1340492856-21126-1-git-send-email-agraf@suse.de>
The initial TLB entry is supposed to help us run the guest -kernel payload.
This means the guest needs to be able to access its own memory, the initrd
memory and the device tree.
So far we only statically reserved a TLB entry from [0;256M[. This patch
fixes it to span from [0;dt_end[, allowing the guest payload to access
everything initially.
Reported-by: Stuart Yoder <stuart.yoder@freescale.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
---
hw/ppce500_mpc8544ds.c | 41 +++++++++++++++++++++++++++--------------
1 files changed, 27 insertions(+), 14 deletions(-)
diff --git a/hw/ppce500_mpc8544ds.c b/hw/ppce500_mpc8544ds.c
index 3eb8a23..b1a0b8c 100644
--- a/hw/ppce500_mpc8544ds.c
+++ b/hw/ppce500_mpc8544ds.c
@@ -31,6 +31,7 @@
#include "elf.h"
#include "sysbus.h"
#include "exec-memory.h"
+#include "host-utils.h"
#define BINARY_DEVICE_TREE_FILE "mpc8544ds.dtb"
#define UIMAGE_LOAD_BASE 0
@@ -55,6 +56,7 @@
struct boot_info
{
uint32_t dt_base;
+ uint32_t dt_size;
uint32_t entry;
};
@@ -164,7 +166,11 @@ static int mpc8544_load_device_tree(CPUPPCState *env,
}
ret = rom_add_blob_fixed(BINARY_DEVICE_TREE_FILE, fdt, fdt_size, addr);
+ if (ret < 0) {
+ goto out;
+ }
g_free(fdt);
+ ret = fdt_size;
out:
#endif
@@ -172,23 +178,27 @@ out:
return ret;
}
-/* Create -kernel TLB entries for BookE, linearly spanning 256MB. */
+/* Create -kernel TLB entries for BookE. */
static inline target_phys_addr_t booke206_page_size_to_tlb(uint64_t size)
{
- return ffs(size >> 10) - 1;
+ return 63 - clz64(size >> 10);
}
-static void mmubooke_create_initial_mapping(CPUPPCState *env,
- target_ulong va,
- target_phys_addr_t pa)
+static void mmubooke_create_initial_mapping(CPUPPCState *env)
{
+ struct boot_info *bi = env->load_info;
ppcmas_tlb_t *tlb = booke206_get_tlbm(env, 1, 0, 0);
- target_phys_addr_t size;
-
- size = (booke206_page_size_to_tlb(256 * 1024 * 1024) << MAS1_TSIZE_SHIFT);
+ target_phys_addr_t size, dt_end;
+ int ps;
+
+ /* Our initial TLB entry needs to cover everything from 0 to
+ the device tree top */
+ dt_end = bi->dt_base + bi->dt_size;
+ ps = booke206_page_size_to_tlb(dt_end) + 1;
+ size = (ps << MAS1_TSIZE_SHIFT);
tlb->mas1 = MAS1_VALID | size;
- tlb->mas2 = va & TARGET_PAGE_MASK;
- tlb->mas7_3 = pa & TARGET_PAGE_MASK;
+ tlb->mas2 = 0;
+ tlb->mas7_3 = 0;
tlb->mas7_3 |= MAS3_UR | MAS3_UW | MAS3_UX | MAS3_SR | MAS3_SW | MAS3_SX;
env->tlb_dirty = true;
@@ -220,7 +230,7 @@ static void mpc8544ds_cpu_reset(void *opaque)
env->gpr[1] = (16<<20) - 8;
env->gpr[3] = bi->dt_base;
env->nip = bi->entry;
- mmubooke_create_initial_mapping(env, 0, 0);
+ mmubooke_create_initial_mapping(env);
}
static void mpc8544ds_init(ram_addr_t ram_size,
@@ -379,13 +389,15 @@ static void mpc8544ds_init(ram_addr_t ram_size,
/* If we're loading a kernel directly, we must load the device tree too. */
if (kernel_filename) {
struct boot_info *boot_info;
+ int dt_size;
#ifndef CONFIG_FDT
cpu_abort(env, "Compiled without FDT support - can't load kernel\n");
#endif
- dt_base = (kernel_size + DTC_LOAD_PAD) & ~DTC_PAD_MASK;
- if (mpc8544_load_device_tree(env, dt_base, ram_size,
- initrd_base, initrd_size, kernel_cmdline) < 0) {
+ dt_base = (loadaddr + kernel_size + DTC_LOAD_PAD) & ~DTC_PAD_MASK;
+ dt_size = mpc8544_load_device_tree(env, dt_base, ram_size, initrd_base,
+ initrd_size, kernel_cmdline);
+ if (dt_size < 0) {
fprintf(stderr, "couldn't load device tree\n");
exit(1);
}
@@ -393,6 +405,7 @@ static void mpc8544ds_init(ram_addr_t ram_size,
boot_info = env->load_info;
boot_info->entry = entry;
boot_info->dt_base = dt_base;
+ boot_info->dt_size = dt_size;
}
if (kvm_enabled()) {
--
1.6.0.2
next prev parent reply other threads:[~2012-06-23 23:08 UTC|newest]
Thread overview: 89+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-23 23:06 [Qemu-devel] [PULL 00/72] ppc patch queue 2012-06-24 Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 01/72] ppc: Fix coding style in op_helper.c Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 02/72] ppc: Split exception helpers Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 03/72] ppc: Avoid AREG0 for " Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 04/72] ppc: Fix coding style in helper.c Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 05/72] ppc: Move exception helpers from helper.c to excp_helper.c Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 06/72] ppc: Split FPU and SPE ops Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 07/72] ppc: Avoid AREG0 for FPU and SPE helpers Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 08/72] ppc: Split integer and vector ops Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 09/72] ppc: Avoid AREG0 for integer and vector helpers Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 10/72] ppc: Split MMU etc. helpers from op_helper.c Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 11/72] ppc: Avoid AREG0 for MMU etc. helpers Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 12/72] ppc: Avoid a warning with the next patch Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 13/72] ppc: Move MMU helpers from helper.c to mmu_helper.c Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 14/72] ppc: Cleanup MMU merge Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 15/72] ppc: Split off timebase helpers Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 16/72] ppc: Avoid AREG0 for " Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 17/72] ppc: Split off misc helpers Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 18/72] ppc: Avoid AREG0 for " Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 19/72] ppc: Move misc helpers from helper.c to misc_helper.c Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 20/72] ppc: Move load and store helpers, switch to AREG0 free mode Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 21/72] ppc: Add missing break Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 22/72] ppc: Make hbrev table const Alexander Graf
2012-06-23 23:06 ` Alexander Graf [this message]
2012-06-23 23:06 ` [Qemu-devel] [PATCH 24/72] Avoid segfault in cpu_dump_state Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 25/72] booke_206_tlbwe: Discard invalid bits in MAS2 Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 26/72] ppc64: Rudimentary Support for extra page sizes on server CPUs Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 27/72] pseries: Correctly create ibm, segment-page-sizes property Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 28/72] spapr_vscsi: Error handling fixes Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 29/72] spapr: Add "memop" hypercall Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 30/72] raw-posix: Fix build without is_allocated support Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 31/72] dt: allow add_subnode to create root subnodes Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 32/72] dt: add helpers for multi-cell adds Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 33/72] dt: add helper for phandle references Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 34/72] dt: temporarily disable subtree creation failure check Alexander Graf
2012-06-23 23:06 ` [Qemu-devel] [PATCH 35/72] dt: add helper for phandle enumeration Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 36/72] dt: add helper for empty dt creation Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 37/72] dt: add helper for phandle allocation Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 38/72] dt: add helper for 64bit cell adds Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 39/72] PPC: e500: require libfdt Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 40/72] PPC: e500: dt: create memory node dynamically Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 41/72] PPC: e500: dt: create /cpus " Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 42/72] PPC: e500: dt: create /hypervisor " Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 43/72] PPC: e500: dt: create / " Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 44/72] PPC: e500: dt: create /chosen " Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 45/72] PPC: e500: dt: create /soc8544 " Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 46/72] PPC: e500: dt: create serial nodes dynamically Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 47/72] PPC: e500: dt: create mpic node dynamically Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 48/72] PPC: e500: dt: create global-utils " Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 49/72] PPC: e500: dt: create pci " Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 50/72] PPC: e500: dt: start with empty device tree Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 51/72] dt: Add -machine dumpdtb option to dump the current dtb Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 52/72] PPC: e500: dt: use 64bit cell helper Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 53/72] PPC: e500: dt: use target_phys_addr_t for ramsize Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 54/72] PPC: e500: enable manual loading of dtb blob Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 55/72] Revert "dt: temporarily disable subtree creation failure check" Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 56/72] PPC: e500: Use new MPIC dt format Alexander Graf
2012-08-08 21:16 ` [Qemu-devel] [Qemu-ppc] " Alexander Graf
2012-08-08 22:40 ` Scott Wood
2012-08-09 20:48 ` Alexander Graf
2012-08-09 20:50 ` Scott Wood
2012-08-09 20:52 ` Alexander Graf
2012-08-09 20:58 ` Scott Wood
2012-08-09 21:01 ` Alexander Graf
2012-08-09 21:11 ` Scott Wood
2012-08-09 21:19 ` Alexander Graf
2012-08-09 21:28 ` Scott Wood
2012-08-09 21:36 ` Alexander Graf
2012-08-09 21:45 ` Scott Wood
2012-08-09 21:48 ` Alexander Graf
2012-08-08 22:40 ` [Qemu-devel] " Peter Maydell
2012-08-08 22:43 ` Peter Maydell
2012-06-23 23:07 ` [Qemu-devel] [PATCH 57/72] PPC: e500: Use new SOC " Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 58/72] PPC: e500: Define addresses as always 64bit Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 59/72] PPC: e500: Extend address/size of / to 64bit Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 60/72] dt: Add global option to set phandle start offset Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 61/72] PPC: e500: Refactor serial dt generation Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 62/72] dt: make setprop argument static Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 63/72] PPC: e500: allow users to set the /compatible property via -machine Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 64/72] uImage: increase the gzip load size Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 65/72] PPC: Add some booke SPR defines Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 66/72] PPC: Add support for MSR_CM Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 67/72] PPC: BookE: Implement EPR SPR Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 68/72] PPC: BookE: Make ivpr selectable by CPU type Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 69/72] PPC: Add e5500 CPU target Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 70/72] PPC: Extract SPR dump generation into its own function Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 71/72] PPC: BookE: Support 32 and 64 bit wide MAS2 Alexander Graf
2012-06-23 23:07 ` [Qemu-devel] [PATCH 72/72] PPC: BookE206: Bump MAS2 to 64bit Alexander Graf
2012-06-24 12:27 ` [Qemu-devel] [PULL 00/72] ppc patch queue 2012-06-24 Blue Swirl
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=1340492856-21126-24-git-send-email-agraf@suse.de \
--to=agraf@suse.de \
--cc=aurelien@aurel32.net \
--cc=blauwirbel@gmail.com \
--cc=qemu-devel@nongnu.org \
--cc=qemu-ppc@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).