From: Alexander Graf <agraf@suse.de>
To: qemu-ppc@nongnu.org
Cc: peter.maydell@linaro.org, qemu-devel@nongnu.org,
"Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Subject: [Qemu-devel] [PULL 02/38] target-ppc: Use right page size with hash table lookup
Date: Sun, 8 Mar 2015 09:44:21 +0100 [thread overview]
Message-ID: <1425804297-53727-3-git-send-email-agraf@suse.de> (raw)
In-Reply-To: <1425804297-53727-1-git-send-email-agraf@suse.de>
From: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
We look at two sizes specified in ISA (4K, 64K). If not found matching,
we consider it 16MB.
Without this patch we would fail to lookup address above 16MB range.
Below 16MB happened to work before because the kernel have a liner
mapping and we always looked up hash for 0xc000000000000000. The
actual real address was computed by using the 16MB offset
with the real address found with the above hash.
Without Fix:
(gdb) x/16x 0xc000000001000000
0xc000000001000000 <list_entries+453208>: Cannot access memory at address 0xc000000001000000
(gdb)
With Fix:
(gdb) x/16x 0xc000000001000000
0xc000000001000000 <list_entries+453208>: 0x00000000 0x00000000 0x00000000 0x00000000
0xc000000001000010 <list_entries+453224>: 0x00000000 0x00000000 0x00000000 0x00000000
0xc000000001000020 <list_entries+453240>: 0x00000000 0x00000000 0x00000000 0x00000000
0xc000000001000030 <list_entries+453256>: 0x00000000 0x00000000 0x00000000 0x00000000
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Alexander Graf <agraf@suse.de>
---
target-ppc/cpu.h | 1 +
target-ppc/mmu-hash64.c | 37 ++++++++++++++++++++++++++-----------
target-ppc/mmu-hash64.h | 3 +++
3 files changed, 30 insertions(+), 11 deletions(-)
diff --git a/target-ppc/cpu.h b/target-ppc/cpu.h
index aae33a9..b706b9f 100644
--- a/target-ppc/cpu.h
+++ b/target-ppc/cpu.h
@@ -45,6 +45,7 @@
# define TARGET_VIRT_ADDR_SPACE_BITS 64
#endif
+#define TARGET_PAGE_BITS_64K 16
#define TARGET_PAGE_BITS_16M 24
#else /* defined (TARGET_PPC64) */
diff --git a/target-ppc/mmu-hash64.c b/target-ppc/mmu-hash64.c
index b0278c9..971751f 100644
--- a/target-ppc/mmu-hash64.c
+++ b/target-ppc/mmu-hash64.c
@@ -388,6 +388,24 @@ static hwaddr ppc_hash64_pteg_search(CPUPPCState *env, hwaddr hash,
return -1;
}
+static uint64_t ppc_hash64_page_shift(ppc_slb_t *slb)
+{
+ uint64_t epnshift;
+
+ /* Page size according to the SLB, which we use to generate the
+ * EPN for hash table lookup.. When we implement more recent MMU
+ * extensions this might be different from the actual page size
+ * encoded in the PTE */
+ if ((slb->vsid & SLB_VSID_LLP_MASK) == SLB_VSID_4K) {
+ epnshift = TARGET_PAGE_BITS;
+ } else if ((slb->vsid & SLB_VSID_LLP_MASK) == SLB_VSID_64K) {
+ epnshift = TARGET_PAGE_BITS_64K;
+ } else {
+ epnshift = TARGET_PAGE_BITS_16M;
+ }
+ return epnshift;
+}
+
static hwaddr ppc_hash64_htab_lookup(CPUPPCState *env,
ppc_slb_t *slb, target_ulong eaddr,
ppc_hash_pte64_t *pte)
@@ -396,12 +414,7 @@ static hwaddr ppc_hash64_htab_lookup(CPUPPCState *env,
hwaddr hash;
uint64_t vsid, epnshift, epnmask, epn, ptem;
- /* Page size according to the SLB, which we use to generate the
- * EPN for hash table lookup.. When we implement more recent MMU
- * extensions this might be different from the actual page size
- * encoded in the PTE */
- epnshift = (slb->vsid & SLB_VSID_L)
- ? TARGET_PAGE_BITS_16M : TARGET_PAGE_BITS;
+ epnshift = ppc_hash64_page_shift(slb);
epnmask = ~((1ULL << epnshift) - 1);
if (slb->vsid & SLB_VSID_B) {
@@ -448,12 +461,14 @@ static hwaddr ppc_hash64_htab_lookup(CPUPPCState *env,
static hwaddr ppc_hash64_pte_raddr(ppc_slb_t *slb, ppc_hash_pte64_t pte,
target_ulong eaddr)
{
+ hwaddr mask;
+ int target_page_bits;
hwaddr rpn = pte.pte1 & HPTE64_R_RPN;
- /* FIXME: Add support for SLLP extended page sizes */
- int target_page_bits = (slb->vsid & SLB_VSID_L)
- ? TARGET_PAGE_BITS_16M : TARGET_PAGE_BITS;
- hwaddr mask = (1ULL << target_page_bits) - 1;
-
+ /*
+ * We support 4K, 64K and 16M now
+ */
+ target_page_bits = ppc_hash64_page_shift(slb);
+ mask = (1ULL << target_page_bits) - 1;
return (rpn & ~mask) | (eaddr & mask);
}
diff --git a/target-ppc/mmu-hash64.h b/target-ppc/mmu-hash64.h
index 49e385d..291750f 100644
--- a/target-ppc/mmu-hash64.h
+++ b/target-ppc/mmu-hash64.h
@@ -37,6 +37,9 @@ void ppc_hash64_store_hpte(CPUPPCState *env, target_ulong index,
#define SLB_VSID_C 0x0000000000000080ULL /* class */
#define SLB_VSID_LP 0x0000000000000030ULL
#define SLB_VSID_ATTR 0x0000000000000FFFULL
+#define SLB_VSID_LLP_MASK (SLB_VSID_L | SLB_VSID_LP)
+#define SLB_VSID_4K 0x0000000000000000ULL
+#define SLB_VSID_64K 0x0000000000000110ULL
/*
* Hash page table definitions
--
1.8.1.4
next prev parent reply other threads:[~2015-03-08 8:45 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-03-08 8:44 [Qemu-devel] [PULL 2.3 00/38] ppc patch queue 2015-03-08 Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 01/38] spapr_vio/spapr_iommu: Move VIO bypass where it belongs Alexander Graf
2015-03-08 8:44 ` Alexander Graf [this message]
2015-03-08 8:44 ` [Qemu-devel] [PULL 03/38] pseries: Limit PCI host bridge "index" value Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 04/38] spapr: Add pseries-2.3 machine Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 05/38] spapr-pci: Enable huge BARs Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 06/38] Generalize QOM publishing of date and time from mc146818rtc.c Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 07/38] Add more VMSTATE_*_TEST variants for integers Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 08/38] pseries: Move sPAPR RTC code into its own file Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 09/38] pseries: Add more parameter validation in RTAS time of day functions Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 10/38] pseries: Add spapr_rtc_read() helper function Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 11/38] pseries: Make RTAS time of day functions respect -rtc options Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 12/38] pseries: Make the PAPR RTC a qdev device Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 13/38] pseries: Move rtc_offset into RTC device's state structure Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 14/38] pseries: Export RTC time via QOM Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 15/38] PPC: Clean up misuse of qdev_init() in kvm-openpic creation Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 16/38] spapr: Clean up misuse of qdev_init() in xics-kvm creation Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 17/38] vga: Expose framebuffer byteorder as a QOM property Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 18/38] pseries: Switch VGA endian on H_SET_MODE Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 19/38] Openpic: check that cpu id is within the number of cpus Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 20/38] display cpu id dump state Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 21/38] macio.c: include parent PCIDevice state in VMStateDescription Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 22/38] adb.c: include ADBDevice parent state in KBDState and MouseState Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 23/38] cuda.c: include adb_poll_timer in VMStateDescription Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 24/38] target-ppc: move sdr1 value change detection logic to helper_store_sdr1() Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 25/38] target-ppc: force update of msr bits in cpu_post_load Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 26/38] openpic: fix segfault on -M mac99 savevm Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 27/38] openpic: fix up loadvm under -M mac99 Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 28/38] openpic: switch IRQQueue queue from inline to bitmap Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 29/38] openpic: convert to vmstate Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 30/38] spapr_vio: Convert to realize() Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 31/38] Revert "default-configs/ppc64: add all components of i82378 SuperIO chip used by prep" Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 32/38] ppc64-softmmu: Remove unsupported FDC from config Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 33/38] ppc64-softmmu: Remove duplicated OPENPIC " Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 34/38] PPC: Remove duplicate OPENPIC defines in default-configs Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 35/38] PPC: Introduce the Virtual Time Base (VTB) SPR register Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 36/38] target-ppc: Add versions to server CPU descriptions Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 37/38] sPAPR: Implement EEH RTAS calls Alexander Graf
2015-03-08 8:44 ` [Qemu-devel] [PULL 38/38] sPAPR: Implement sPAPRPHBClass EEH callbacks Alexander Graf
2015-03-09 9:13 ` [Qemu-devel] [PULL 2.3 00/38] ppc patch queue 2015-03-08 Peter Maydell
2015-03-09 12:30 ` Alexander Graf
2015-03-09 13:16 ` Peter Maydell
2015-03-09 14:02 ` Alexander Graf
2015-03-09 15:14 ` Peter Maydell
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=1425804297-53727-3-git-send-email-agraf@suse.de \
--to=agraf@suse.de \
--cc=aneesh.kumar@linux.vnet.ibm.com \
--cc=peter.maydell@linaro.org \
--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).