From: Ard Biesheuvel <ard.biesheuvel@linaro.org>
To: Laszlo Ersek <lersek@redhat.com>
Cc: "edk2-devel@lists.01.org" <edk2-devel@ml01.01.org>,
Shannon Zhao <zhaoshenglong@huawei.com>,
Sakar Arora <sakar.arora@nxp.com>
Subject: Re: [PATCH 4/4] ArmVirtPkg/HighMemDxe: move to FDT client protocol
Date: Thu, 15 Sep 2016 15:18:17 +0100 [thread overview]
Message-ID: <CAKv+Gu-NkQ8S4QUO2zotV+OoJaLRZ5XpvdDyJ7eRDk76Wy6Xag@mail.gmail.com> (raw)
In-Reply-To: <6c55da74-6982-d31e-c487-0fb1d6ccda8f@redhat.com>
On 15 September 2016 at 15:15, Laszlo Ersek <lersek@redhat.com> wrote:
> On 09/15/16 15:30, Ard Biesheuvel wrote:
>> Use the FDT client protocol rather than parsing the DT directly using
>> fdtlib. While we're at it, update the code so it deals correctly with
>> memory nodes that describe multiple disjoint regions in their "reg"
>> properties, and make the code work with #address-cells/#size-cells
>> properties of <1> as well as <2>.
>>
>> Contributed-under: TianoCore Contribution Agreement 1.0
>> Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
>> ---
>> ArmVirtPkg/HighMemDxe/HighMemDxe.c | 120 +++++++++-----------
>> ArmVirtPkg/HighMemDxe/HighMemDxe.inf | 16 ++-
>> 2 files changed, 62 insertions(+), 74 deletions(-)
>>
>> diff --git a/ArmVirtPkg/HighMemDxe/HighMemDxe.c b/ArmVirtPkg/HighMemDxe/HighMemDxe.c
>> index 4d56e6236b54..08de3cbb7e9c 100644
>> --- a/ArmVirtPkg/HighMemDxe/HighMemDxe.c
>> +++ b/ArmVirtPkg/HighMemDxe/HighMemDxe.c
>> @@ -1,7 +1,7 @@
>> /** @file
>> * High memory node enumeration DXE driver for ARM Virtual Machines
>> *
>> -* Copyright (c) 2015, Linaro Ltd. All rights reserved.
>> +* Copyright (c) 2015-2016, Linaro Ltd. All rights reserved.
>> *
>> * This program and the accompanying materials are licensed and made available
>> * under the terms and conditions of the BSD License which accompanies this
>> @@ -15,12 +15,12 @@
>> **/
>>
>> #include <Library/BaseLib.h>
>> -#include <Library/UefiDriverEntryPoint.h>
>> #include <Library/DebugLib.h>
>> -#include <Library/PcdLib.h>
>> -#include <Library/HobLib.h>
>> -#include <libfdt.h>
>> #include <Library/DxeServicesTableLib.h>
>> +#include <Library/PcdLib.h>
>> +#include <Library/UefiBootServicesTableLib.h>
>> +
>> +#include <Protocol/FdtClient.h>
>>
>> EFI_STATUS
>> EFIAPI
>> @@ -29,76 +29,66 @@ InitializeHighMemDxe (
>> IN EFI_SYSTEM_TABLE *SystemTable
>> )
>> {
>> - VOID *Hob;
>> - VOID *DeviceTreeBase;
>> - INT32 Node, Prev;
>> - EFI_STATUS Status;
>> - CONST CHAR8 *Type;
>> - INT32 Len;
>> - CONST VOID *RegProp;
>> - UINT64 CurBase;
>> - UINT64 CurSize;
>> -
>> - Hob = GetFirstGuidHob(&gFdtHobGuid);
>> - if (Hob == NULL || GET_GUID_HOB_DATA_SIZE (Hob) != sizeof (UINT64)) {
>> - return EFI_NOT_FOUND;
>> - }
>> - DeviceTreeBase = (VOID *)(UINTN)*(UINT64 *)GET_GUID_HOB_DATA (Hob);
>> -
>> - if (fdt_check_header (DeviceTreeBase) != 0) {
>> - DEBUG ((EFI_D_ERROR, "%a: No DTB found @ 0x%p\n", __FUNCTION__,
>> - DeviceTreeBase));
>> - return EFI_NOT_FOUND;
>> - }
>> + FDT_CLIENT_PROTOCOL *FdtClient;
>> + EFI_STATUS Status, FindNodeStatus;
>> + INT32 Node;
>> + CONST UINT32 *Reg;
>> + UINT32 RegSize;
>> + UINTN AddressCells, SizeCells;
>> + UINT64 CurBase;
>> + UINT64 CurSize;
>>
>> - DEBUG ((EFI_D_INFO, "%a: DTB @ 0x%p\n", __FUNCTION__, DeviceTreeBase));
>> + Status = gBS->LocateProtocol (&gFdtClientProtocolGuid, NULL,
>> + (VOID **)&FdtClient);
>> + ASSERT_EFI_ERROR (Status);
>>
>> //
>> - // Check for memory node and add the memory spaces expect the lowest one
>> + // Check for memory node and add the memory spaces except the lowest one
>> //
>> - for (Prev = 0;; Prev = Node) {
>> - Node = fdt_next_node (DeviceTreeBase, Prev, NULL);
>> - if (Node < 0) {
>> - break;
>> - }
>> + for (FindNodeStatus = FdtClient->FindMemoryNodeReg (FdtClient, &Node,
>> + (CONST VOID **) &Reg, &AddressCells,
>> + &SizeCells, &RegSize);
>> + !EFI_ERROR (FindNodeStatus);
>> + FindNodeStatus = FdtClient->FindNextMemoryNodeReg (FdtClient, Node,
>> + &Node, (CONST VOID **) &Reg, &AddressCells,
>> + &SizeCells, &RegSize)) {
>> +
>
> I think we can remove this empty line. Not too important of course.
>
Sure
>> + ASSERT (AddressCells <= 2);
>> + ASSERT (SizeCells <= 2);
>>
>> - Type = fdt_getprop (DeviceTreeBase, Node, "device_type", &Len);
>> - if (Type && AsciiStrnCmp (Type, "memory", Len) == 0) {
>> - //
>> - // Get the 'reg' property of this node. For now, we will assume
>> - // two 8 byte quantities for base and size, respectively.
>> - //
>> - RegProp = fdt_getprop (DeviceTreeBase, Node, "reg", &Len);
>> - if (RegProp != NULL && Len == (2 * sizeof (UINT64))) {
>> + while (RegSize > 0) {
>>
>
> This one too. (Also just a matter of taste.)
>
If you prefer, I don't care either way.
>> - CurBase = fdt64_to_cpu (((UINT64 *)RegProp)[0]);
>> - CurSize = fdt64_to_cpu (((UINT64 *)RegProp)[1]);
>> + CurBase = SwapBytes32 (*Reg++);
>> + if (AddressCells > 1) {
>> + CurBase |= (CurBase << 32) | SwapBytes32 (*Reg++);
>
> The operator |= seems wrong; it should be just =.
>
> (You could do { CurBase <<= 32; CurBase |= SwapBytes32 (*Reg++); }. The
> above is more compact, but the operator should be fixed.)
>
You are right. I got away with it due to the fact that I tested with
base/size pairs <= 4GB, and so the leading cell == 0 in all cases.
Thanks for spotting that!
>> + }
>> + CurSize = SwapBytes32 (*Reg++);
>> + if (SizeCells > 1) {
>> + CurSize |= (CurSize << 32) | SwapBytes32 (*Reg++);
>
> Same here.
>
> With these fixed:
>
> Reviewed-by: Laszlo Ersek <lersek@redhat.com>
>
OK, thanks.
next prev parent reply other threads:[~2016-09-15 14:18 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-09-15 13:30 [PATCH 0/4] ArmVirtPkg: FdtClientDxe & HighMemDxe updates Ard Biesheuvel
2016-09-15 13:30 ` [PATCH 1/4] ArmVirtPkg/FdtClientDxe: fix check for size of "reg" properties Ard Biesheuvel
2016-09-15 13:38 ` Laszlo Ersek
2016-09-15 13:40 ` Ard Biesheuvel
2016-09-15 13:30 ` [PATCH 2/4] ArmVirtPkg/FdtClientDxe: report address and size cell count directly Ard Biesheuvel
2016-09-15 13:42 ` Laszlo Ersek
2016-09-15 13:30 ` [PATCH 3/4] ArmVirtPkg/FdtClient: add methods to iterate over memory nodes Ard Biesheuvel
2016-09-15 13:57 ` Laszlo Ersek
2016-09-15 14:04 ` Ard Biesheuvel
2016-09-15 13:30 ` [PATCH 4/4] ArmVirtPkg/HighMemDxe: move to FDT client protocol Ard Biesheuvel
2016-09-15 14:15 ` Laszlo Ersek
2016-09-15 14:18 ` Ard Biesheuvel [this message]
2016-09-15 13:34 ` [PATCH 0/4] ArmVirtPkg: FdtClientDxe & HighMemDxe updates Ard Biesheuvel
2016-09-15 14:40 ` Ard Biesheuvel
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=CAKv+Gu-NkQ8S4QUO2zotV+OoJaLRZ5XpvdDyJ7eRDk76Wy6Xag@mail.gmail.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