From: Ard Biesheuvel <ard.biesheuvel@linaro.org>
To: "Gao, Liming" <liming.gao@intel.com>
Cc: edk2-devel-01 <edk2-devel@lists.01.org>,
"Zhu, Yonghong" <yonghong.zhu@intel.com>
Subject: Re: [Patch] BaseTools PeCoffLib: Fix the issue to get RelocationsStripped from TE image
Date: Mon, 22 Aug 2016 09:30:58 +0200 [thread overview]
Message-ID: <CAKv+Gu_1_CmaBKxC4yLHQqq7i8GF1OSwxYcXdUZDrfnj+t1kFw@mail.gmail.com> (raw)
In-Reply-To: <4A89E2EF3DFEDB4C8BFDE51014F606A1155ED73E@shsmsx102.ccr.corp.intel.com>
On 22 August 2016 at 09:07, Gao, Liming <liming.gao@intel.com> wrote:
> Ard:
> This is done to convert PE/COFF image to TE image. TE image Relocation VirtualAddress will be set to non-zero. You can find the following logic in GenFw.c line 2515. Then, PeCoffLib can restore RELOCS_STRIPPED flag from TE image.
>
> if (((PeHdr->Pe32.FileHeader.Characteristics & EFI_IMAGE_FILE_RELOCS_STRIPPED) == 0) && \
> (TEImageHeader.DataDirectory[EFI_TE_IMAGE_DIRECTORY_ENTRY_BASERELOC].VirtualAddress == 0) && \
> (TEImageHeader.DataDirectory[EFI_TE_IMAGE_DIRECTORY_ENTRY_BASERELOC].Size == 0)) {
> //
> // PeImage can be loaded into memory, but it has no relocation section.
> // Fix TeImage Header to set VA of relocation data directory to not zero, the size is still zero.
> //
> if (Optional32 != NULL) {
> TEImageHeader.DataDirectory[EFI_TE_IMAGE_DIRECTORY_ENTRY_BASERELOC].VirtualAddress = Optional32->SizeOfImage - sizeof (EFI_IMAGE_BASE_RELOCATION);
> } else if (Optional64 != NULL) {
> TEImageHeader.DataDirectory[EFI_TE_IMAGE_DIRECTORY_ENTRY_BASERELOC].VirtualAddress = Optional64->SizeOfImage - sizeof (EFI_IMAGE_BASE_RELOCATION);
> }
> }
>
Apologies. Your patch does work, but I failed to clean BaseTools/
before rebuilding it.
Tested-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Thanks for fixing this,
Ard.
>> -----Original Message-----
>> From: Ard Biesheuvel [mailto:ard.biesheuvel@linaro.org]
>> Sent: Monday, August 22, 2016 3:00 PM
>> To: Gao, Liming <liming.gao@intel.com>
>> Cc: edk2-devel-01 <edk2-devel@lists.01.org>; Zhu, Yonghong
>> <yonghong.zhu@intel.com>
>> Subject: Re: [Patch] BaseTools PeCoffLib: Fix the issue to get
>> RelocationsStripped from TE image
>>
>> On 22 August 2016 at 05:09, Liming Gao <liming.gao@intel.com> wrote:
>> > If PE image has no relocation section, and has not set RELOCS_STRIPPED,
>> > after it is converted to TE image, GenFw will set its relocation section
>> > VirtualAddress to non-zero address, and keep Size value as Zero.
>>
>> I cannot find this in the code. Instead, I see
>>
>> Elf64Convert.c-1031- Dir->Size = mCoffOffset - mRelocOffset;
>> Elf64Convert.c-1032- if (Dir->Size == 0) {
>> Elf64Convert.c:1033: // If no relocations, null out the directory
>> entry and don't add the .reloc section
>> Elf64Convert.c-1034- Dir->VirtualAddress = 0;
>> Elf64Convert.c-1035- NtHdr->Pe32Plus.FileHeader.NumberOfSections--;
>> Elf64Convert.c-1036- } else {
>> Elf64Convert.c-1037- Dir->VirtualAddress = mRelocOffset;
>> Elf64Convert.c:1038: CreateSectionHeader (".reloc", mRelocOffset,
>> mCoffOffset - mRelocOffset,
>> Elf64Convert.c-1039- EFI_IMAGE_SCN_CNT_INITIALIZED_DATA
>> Elf64Convert.c-1040- | EFI_IMAGE_SCN_MEM_DISCARDABLE
>> Elf64Convert.c-1041- | EFI_IMAGE_SCN_MEM_READ);
>> Elf64Convert.c-1042- }
>> Elf64Convert.c-1043-}
>>
>> so the virtual address is set to zero.
>>
>> I also tried applying the patch, and it does not solve the issue for me.
>>
>> Thanks,
>> Ard.
>>
>> > MdePkg
>> > BasePeCoffLib applied this rule to get RelocationsStripped attribute. But,
>> > it is missing in BaseTools BasePeCoffLib.
>> >
>> > Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
>> > Cc: Yonghong Zhu <yonghong.zhu@intel.com>
>> > Contributed-under: TianoCore Contribution Agreement 1.0
>> > Signed-off-by: Liming Gao <liming.gao@intel.com>
>> > ---
>> > BaseTools/Source/C/Common/BasePeCoff.c | 4 ++--
>> > 1 file changed, 2 insertions(+), 2 deletions(-)
>> >
>> > diff --git a/BaseTools/Source/C/Common/BasePeCoff.c
>> b/BaseTools/Source/C/Common/BasePeCoff.c
>> > index 9652557..d0cc1af 100644
>> > --- a/BaseTools/Source/C/Common/BasePeCoff.c
>> > +++ b/BaseTools/Source/C/Common/BasePeCoff.c
>> > @@ -2,7 +2,7 @@
>> >
>> > Functions to get info and load PE/COFF image.
>> >
>> > -Copyright (c) 2004 - 2010, Intel Corporation. All rights reserved.<BR>
>> > +Copyright (c) 2004 - 2016, Intel Corporation. All rights reserved.<BR>
>> > Portions Copyright (c) 2011 - 2013, ARM Ltd. All rights reserved.<BR>
>> > This program and the accompanying materials
>> > are licensed and made available under the terms and conditions of the BSD
>> License
>> > @@ -336,7 +336,7 @@ Returns:
>> > //
>> > if ((!(ImageContext->IsTeImage)) && ((PeHdr-
>> >Pe32.FileHeader.Characteristics & EFI_IMAGE_FILE_RELOCS_STRIPPED) !=
>> 0)) {
>> > ImageContext->RelocationsStripped = TRUE;
>> > - } else if ((ImageContext->IsTeImage) && (TeHdr->DataDirectory[0].Size
>> == 0)) {
>> > + } else if ((ImageContext->IsTeImage) && (TeHdr->DataDirectory[0].Size
>> == 0) && (TeHdr->DataDirectory[0].VirtualAddress == 0)) {
>> > ImageContext->RelocationsStripped = TRUE;
>> > } else {
>> > ImageContext->RelocationsStripped = FALSE;
>> > --
>> > 2.8.0.windows.1
>> >
next prev parent reply other threads:[~2016-08-22 7:30 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-08-22 3:09 [Patch] BaseTools PeCoffLib: Fix the issue to get RelocationsStripped from TE image Liming Gao
2016-08-22 6:59 ` Ard Biesheuvel
2016-08-22 7:07 ` Gao, Liming
2016-08-22 7:30 ` Ard Biesheuvel [this message]
2016-08-22 7:43 ` Zhu, Yonghong
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_1_CmaBKxC4yLHQqq7i8GF1OSwxYcXdUZDrfnj+t1kFw@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