From: "duntan" <dun.tan@intel.com>
To: devel@edk2.groups.io
Cc: Eric Dong <eric.dong@intel.com>, Ray Ni <ray.ni@intel.com>,
Rahul Kumar <rahul1.kumar@intel.com>,
Gerd Hoffmann <kraxel@redhat.com>
Subject: [Patch V2 03/14] UefiCpuPkg/CpuPageTableLib: Fix the non-1:1 mapping issue
Date: Wed, 8 Mar 2023 18:07:47 +0800 [thread overview]
Message-ID: <20230308100758.669-4-dun.tan@intel.com> (raw)
In-Reply-To: <20230308100758.669-1-dun.tan@intel.com>
In previous code logic, when splitting a leaf parent entry to
smaller granularity child page table, if the parent entry
Attribute&Mask(without PageTableBaseAddress field) is equal to the
input attribute&mask(without PageTableBaseAddress field), the split
process won't happen. This may lead to failure in non-1:1 mapping.
For example, there is a page table in which [0, 1G] is mapped(Lv4[0]
,Lv3[0,0], a non-leaf level4 entry and a leaf level3 entry). And we
want to remap [0, 2M] linear address range to [1G, 1G + 2M] with the
same attibute. The expected behaviour should be: split Lv3[0,0]
entry into 512 level2 entries and remap the first level2 entry to
cover [0, 2M]. But the split won't happen in previous code since
PageTableBaseAddress of input Attribute is not checked.
So, when checking if a leaf parent entry needs to be splitted, we
should also check if PageTableBaseAddress calculated by parent entry
is equal to the value caculated by input attribute.
Signed-off-by: Dun Tan <dun.tan@intel.com>
Cc: Eric Dong <eric.dong@intel.com>
Cc: Ray Ni <ray.ni@intel.com>
Cc: Rahul Kumar <rahul1.kumar@intel.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
---
UefiCpuPkg/Library/CpuPageTableLib/CpuPageTableMap.c | 27 +++++++++++++++++----------
1 file changed, 17 insertions(+), 10 deletions(-)
diff --git a/UefiCpuPkg/Library/CpuPageTableLib/CpuPageTableMap.c b/UefiCpuPkg/Library/CpuPageTableLib/CpuPageTableMap.c
index 4c9d70fa0a..ee27238edb 100644
--- a/UefiCpuPkg/Library/CpuPageTableLib/CpuPageTableMap.c
+++ b/UefiCpuPkg/Library/CpuPageTableLib/CpuPageTableMap.c
@@ -258,6 +258,7 @@ PageTableLibMapInLevel (
UINTN BitStart;
UINTN Index;
IA32_PAGING_ENTRY *PagingEntry;
+ UINTN PagingEntryIndex;
IA32_PAGING_ENTRY *CurrentPagingEntry;
UINT64 RegionLength;
UINT64 SubLength;
@@ -288,6 +289,13 @@ PageTableLibMapInLevel (
LocalParentAttribute.Uint64 = ParentAttribute->Uint64;
ParentAttribute = &LocalParentAttribute;
+ //
+ // RegionLength: 256T (1 << 48) 512G (1 << 39), 1G (1 << 30), 2M (1 << 21) or 4K (1 << 12).
+ //
+ BitStart = 12 + (Level - 1) * 9;
+ PagingEntryIndex = (UINTN)BitFieldRead64 (LinearAddress + Offset, BitStart, BitStart + 9 - 1);
+ RegionLength = REGION_LENGTH (Level);
+
//
// ParentPagingEntry ONLY is deferenced for checking Present and MustBeOne bits
// when Modify is FALSE.
@@ -325,8 +333,11 @@ PageTableLibMapInLevel (
// the actual attributes of grand-parents when determing the memory type.
//
PleBAttribute.Uint64 = PageTableLibGetPleBMapAttribute (&ParentPagingEntry->PleB, ParentAttribute);
- if ((IA32_MAP_ATTRIBUTE_ATTRIBUTES (&PleBAttribute) & IA32_MAP_ATTRIBUTE_ATTRIBUTES (Mask))
- == (IA32_MAP_ATTRIBUTE_ATTRIBUTES (Attribute) & IA32_MAP_ATTRIBUTE_ATTRIBUTES (Mask)))
+ if ((((IA32_MAP_ATTRIBUTE_ATTRIBUTES (&PleBAttribute) & IA32_MAP_ATTRIBUTE_ATTRIBUTES (Mask))
+ == (IA32_MAP_ATTRIBUTE_ATTRIBUTES (Attribute) & IA32_MAP_ATTRIBUTE_ATTRIBUTES (Mask)))) &&
+ ( (Mask->Bits.PageTableBaseAddress == 0)
+ || ((IA32_MAP_ATTRIBUTE_PAGE_TABLE_BASE_ADDRESS (&PleBAttribute) + PagingEntryIndex * RegionLength)
+ == (IA32_MAP_ATTRIBUTE_PAGE_TABLE_BASE_ADDRESS (Attribute) + Offset))))
{
//
// This function is called when the memory length is less than the region length of the parent level.
@@ -353,8 +364,7 @@ PageTableLibMapInLevel (
//
PageTableLibSetPnle (&ParentPagingEntry->Pnle, &NopAttribute, &AllOneMask);
- RegionLength = REGION_LENGTH (Level);
- PagingEntry = (IA32_PAGING_ENTRY *)(UINTN)IA32_PNLE_PAGE_TABLE_BASE_ADDRESS (&ParentPagingEntry->Pnle);
+ PagingEntry = (IA32_PAGING_ENTRY *)(UINTN)IA32_PNLE_PAGE_TABLE_BASE_ADDRESS (&ParentPagingEntry->Pnle);
for (SubOffset = 0, Index = 0; Index < 512; Index++) {
PagingEntry[Index].Uint64 = OneOfPagingEntry.Uint64 + SubOffset;
SubOffset += RegionLength;
@@ -425,14 +435,11 @@ PageTableLibMapInLevel (
}
//
- // RegionLength: 256T (1 << 48) 512G (1 << 39), 1G (1 << 30), 2M (1 << 21) or 4K (1 << 12).
// RegionStart: points to the linear address that's aligned on RegionLength and lower than (LinearAddress + Offset).
//
- BitStart = 12 + (Level - 1) * 9;
- Index = (UINTN)BitFieldRead64 (LinearAddress + Offset, BitStart, BitStart + 9 - 1);
- RegionLength = LShiftU64 (1, BitStart);
- RegionMask = RegionLength - 1;
- RegionStart = (LinearAddress + Offset) & ~RegionMask;
+ Index = PagingEntryIndex;
+ RegionMask = RegionLength - 1;
+ RegionStart = (LinearAddress + Offset) & ~RegionMask;
ParentAttribute->Uint64 = PageTableLibGetPnleMapAttribute (&ParentPagingEntry->Pnle, ParentAttribute);
--
2.31.1.windows.1
next prev parent reply other threads:[~2023-03-08 10:08 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-03-08 10:07 [Patch V2 00/14] Fix issues in CpuPageTableLib duntan
2023-03-08 10:07 ` [Patch V2 01/14] UefiCpuPkg/CpuPageTableLib: Remove unneeded 'if' condition duntan
2023-03-15 1:23 ` Ni, Ray
2023-03-15 1:23 ` Ni, Ray
2023-03-15 1:45 ` duntan
2023-03-08 10:07 ` [Patch V2 02/14] UefiCpuPkg/CpuPageTableLib: Add check for input Length duntan
2023-03-15 1:25 ` Ni, Ray
2023-03-15 1:46 ` duntan
2023-03-08 10:07 ` duntan [this message]
2023-03-15 1:28 ` [Patch V2 03/14] UefiCpuPkg/CpuPageTableLib: Fix the non-1:1 mapping issue Ni, Ray
2023-03-15 1:44 ` duntan
2023-03-08 10:07 ` [Patch V2 04/14] UefiCpuPkg/CpuPageTableLib: Fix issue when splitting leaf entry duntan
2023-03-15 1:51 ` Ni, Ray
2023-03-08 10:07 ` [Patch V2 05/14] UefiCpuPkg/CpuPageTebleLib: Check Mask and Attr in PageTableMap duntan
2023-03-15 5:33 ` Ni, Ray
2023-03-15 9:49 ` duntan
2023-03-08 10:07 ` [Patch V2 06/14] UefiCpuPkg/CpuPageTableLib: Add manual test to check Mask and Attr duntan
2023-03-15 5:36 ` Ni, Ray
2023-03-08 10:07 ` [Patch V2 07/14] UefiCpuPkg/CpuPageTableLib:Modify RandomTest to check Mask/Attr duntan
2023-03-15 5:48 ` Ni, Ray
2023-03-15 9:49 ` duntan
2023-03-08 10:07 ` [Patch V2 08/14] UefiCpuPkg/CpuPageTableLib: Enable non-1:1 mapping in random test duntan
2023-03-15 5:49 ` Ni, Ray
2023-03-08 10:07 ` [Patch V2 09/14] UefiCpuPkg/CpuPageTableLib: Add OUTPUT IsModified parameter duntan
2023-03-15 6:01 ` Ni, Ray
2023-03-15 9:49 ` duntan
2023-03-08 10:07 ` [Patch V2 10/14] UefiCpuPkg/CpuPageTableLib: Modify RandomTest to check IsModified duntan
2023-03-15 6:09 ` Ni, Ray
2023-03-15 9:49 ` duntan
2023-03-08 10:07 ` [Patch V2 11/14] UefiCpuPkg/CpuPageTableLib: Enable PAE paging duntan
2023-03-15 6:24 ` Ni, Ray
2023-03-15 9:49 ` duntan
2023-03-08 10:07 ` [Patch V2 12/14] UefiCpuPkg/CpuPageTableLib: Add RandomTest for " duntan
2023-03-15 6:27 ` Ni, Ray
2023-03-08 10:07 ` [Patch V2 13/14] UefiCpuPkg: Fix IA32 build failure in CpuPageTableLib.inf duntan
2023-03-15 6:35 ` Ni, Ray
2023-03-15 9:49 ` duntan
2023-03-08 10:07 ` [Patch V2 14/14] UefiCpuPkg: Modify UnitTest code since tested API is changed duntan
2023-03-15 6:42 ` Ni, Ray
2023-03-15 9:26 ` Zhiguang Liu
2023-03-15 9:27 ` Ni, Ray
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-list 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=20230308100758.669-4-dun.tan@intel.com \
--to=devel@edk2.groups.io \
/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