From: "Gerd Hoffmann" <kraxel@redhat.com>
To: devel@edk2.groups.io, yuanhao.xie@intel.com
Cc: Guo Dong <guo.dong@intel.com>, Ray Ni <ray.ni@intel.com>,
Sean Rhodes <sean@starlabs.systems>,
James Lu <james.lu@intel.com>, Gua Guo <gua.guo@intel.com>
Subject: Re: [edk2-devel] [PATCH 2/5] UefiCpuPkg: Contiguous memory allocation and code clean-up.
Date: Wed, 8 Feb 2023 12:09:33 +0100 [thread overview]
Message-ID: <20230208110933.bho5lybb6evnqcoh@sirius.home.kraxel.org> (raw)
In-Reply-To: <20230207134939.273-3-yuanhao.xie@intel.com>
> + AllocSize = EFI_PAGES_TO_SIZE (
> + EFI_SIZE_TO_PAGES (
> + CpuMpData->CpuCount * AP_SAFE_STACK_SIZE + ApLoopFuncSize
> + )
> + );
> + Status = gBS->AllocatePages (
> + AllocateMaxAddress,
> + EfiReservedMemoryType,
> + EFI_SIZE_TO_PAGES (AllocSize),
> + &Address
> + );
Hmm? You convert size to pages, pages to size, size to pages again.
Also you don't want stack and code being on the same page, so I guess
the logic you actually need is this:
StackPages = EFI_SIZE_TO_PAGES(CpuMpData->CpuCount * AP_SAFE_STACK_SIZE);
FuncPages = EFI_SIZE_TO_PAGES(ApLoopFuncSize)
gBS->AllocatePages(..., StackPages + FuncPages, ...);
> +//
> +// Union holds the relocate APs loop entries for different cases
> +//
> +typedef union {
> + VOID *Data;
> + ASM_RELOCATE_AP_LOOP_AMD64 Amd64Entry; // 64-bit AMD Processor
> + ASM_RELOCATE_AP_LOOP GenericEntry; // Intel Processor (32-bit or 64-bit), or 32-bit AMD Processor
> +} RELOCATE_AP_LOOP_ENTRY;
I'm sure I've mentioned this before. The special case you have to
handle is not running on a AMD Processor, but AmdSev being active
(i.e. UseSevEsAPMethod == True). Otherwise it should be just standard
Ia32 and X64, there should be no need to check whenever you are running
on a AMD processor.
take care,
Gerd
next prev parent reply other threads:[~2023-02-08 11:09 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-02-07 13:49 [PATCH 0/5] Put APs in 64 bit mode before handoff to OS Yuanhao Xie
2023-02-07 13:49 ` [PATCH 1/5] UefiCpuPkg: Duplicate RelocateApLoop for Amd x64 processors Yuanhao Xie
2023-02-07 13:49 ` [PATCH 2/5] UefiCpuPkg: Contiguous memory allocation and code clean-up Yuanhao Xie
2023-02-07 16:40 ` [edk2-devel] " Marvin Häuser
2023-02-07 16:43 ` Marvin Häuser
2023-02-08 10:36 ` Yuanhao Xie
2023-02-08 11:09 ` Gerd Hoffmann [this message]
2023-02-10 9:12 ` Yuanhao Xie
2023-02-10 11:09 ` Gerd Hoffmann
2023-02-10 12:52 ` Ni, Ray
2023-02-07 13:49 ` [PATCH 3/5] OvmfPkg: Add CpuPageTableLib required by MpInitLib Yuanhao Xie
2023-02-08 10:53 ` Gerd Hoffmann
2023-02-09 16:30 ` [edk2-devel] " Ard Biesheuvel
2023-02-07 13:49 ` [PATCH 4/5] UefiPayloadPkg: " Yuanhao Xie
2023-02-08 5:38 ` Guo, Gua
2023-02-08 17:02 ` Guo Dong
2023-02-07 13:49 ` [PATCH 5/5] UefiCpuPkg: Put APs in 64 bit mode before handoff to OS Yuanhao Xie
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=20230208110933.bho5lybb6evnqcoh@sirius.home.kraxel.org \
--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