public inbox for devel@edk2.groups.io
 help / color / mirror / Atom feed
From: "Andrew Fish" <afish@apple.com>
To: edk2-devel-groups-io <devel@edk2.groups.io>, tigerliu@zhaoxin.com
Cc: Mike Kinney <michael.d.kinney@intel.com>,
	"Gao, Liming" <liming.gao@intel.com>
Subject: Re: [edk2-devel] question about MSFT compiler/link option
Date: Wed, 05 Aug 2020 20:46:29 -0700	[thread overview]
Message-ID: <1CFCC7E5-6BB3-467D-885E-E74CF0FCA5A0@apple.com> (raw)
In-Reply-To: <edd1928492d046ec8efda5d7df2c9f71@zhaoxin.com>

[-- Attachment #1: Type: text/plain, Size: 6073 bytes --]



> On Aug 5, 2020, at 8:31 PM, Tiger Liu(BJ-RD) <tigerliu@zhaoxin.com> wrote:
> 
> Hi, Michael:
> I wrote this sentence in C file:
> pTmp = (VOID*)& PdataStructTable;
> 

Tiger,

Working around optimization is complicated. You likely added a statement that got optimized away as the value of pTmp could be ignored and your program could still get the same result. 

You would need to do some meaningful work with pTmp or make pTmp a volatile variable to force the compiler to NOT optimize it away. The better the optimizer, the more layers of abstraction it can see and optimize away. 

Thanks,

Andrew Fish

> tried to reference the table being generated to prevent the optimizer from removing the data structure from the executable.
> 
> But it is still being optimized away.
> 
> I used MSTF VS2015.
> 
> Thanks
> -----邮件原件-----
> 发件人: devel@edk2.groups.io <mailto:devel@edk2.groups.io> <devel@edk2.groups.io <mailto:devel@edk2.groups.io>> 代表 Michael D Kinney
> 发送时间: 2020年8月5日 23:18
> 收件人: devel@edk2.groups.io <mailto:devel@edk2.groups.io>; Tiger Liu(BJ-RD) <TigerLiu@zhaoxin.com <mailto:TigerLiu@zhaoxin.com>>; Gao, Liming <liming.gao@intel.com <mailto:liming.gao@intel.com>>; Kinney, Michael D <michael.d.kinney@intel.com <mailto:michael.d.kinney@intel.com>>
> 主题: Re: [edk2-devel] question about MSFT compiler/link option
> 
> Hi,
> 
> I think you are seeing the impact of compiler optimizations.
> 
> If a global variable is never referenced after optimizing the code, then the global variable will be optimized away.  This type of optimization is important for size constrained environment like firmware.
> 
> By using the /Gw- flag, you are disabling the ability of the compiler to optimize away unreferenced global variables.
> 
> If you update the code in your example to guarantee the global variable is always accessed for all types of builds, then I suspect you will always see the global in the map.
> 
> Thanks,
> 
> Mike
> 
>> -----Original Message-----
>> From: devel@edk2.groups.io <devel@edk2.groups.io> On Behalf Of Tiger
>> Liu(BJ-RD)
>> Sent: Tuesday, August 4, 2020 11:34 PM
>> To: devel@edk2.groups.io; Gao, Liming
>> <liming.gao@intel.com>
>> Subject: Re: [edk2-devel] question about MSFT compiler/link option
>> 
>> Hi, Liming:
>> Thanks for your reply!
>> 
>> I checked map file, not find this global data array.
>> 
>> I found if added this build option in this PEIM's inf, then it seems
>> ok.
>> [BuildOptions]
>> MSFT:*_*_IA32_CC_FLAGS   = /Gw-
>> 
>> Thanks
>> 
>> -----邮件原件-----
>> 发件人: devel@edk2.groups.io <devel@edk2.groups.io> 代表
>> Liming Gao
>> 发送时间: 2020年8月4日 23:21
>> 收件人: devel@edk2.groups.io; Tiger Liu(BJ-RD) <TigerLiu@zhaoxin.com>
>> 主题: Re: [edk2-devel] question about MSFT compiler/link option
>> 
>> If this global variable is used by code, it will be linked into the
>> final EFI image. You can check whether the generate map file includes
>> it.
>> 
>>> -----Original Message-----
>>> From: devel@edk2.groups.io <devel@edk2.groups.io> On
>> Behalf Of Tiger
>>> Liu(BJ-RD)
>>> Sent: Tuesday, August 4, 2020 5:34 PM
>>> To: devel@edk2.groups.io
>>> Subject: [edk2-devel] question about MSFT
>> compiler/link option
>>> 
>>> Hi, Experts:
>>> I have a question about MSFT compiler/link option.
>>> 
>>> I wrote a PEIM, which had a large global data array
>> definition in a C file.
>>> Such as:
>>> PRI_DATA_STRUCT  PdataStructTable[] = {
>>> {01,           0x75,    0xF0,     0x0,     0x0,
>> 0x0 },
>>> {02,           0x77,    0xF0,     0x0,     0x0,
>> 0x0 },
>>> {03,           0x79,    0xF0,     0x0,     0x0,
>> 0x0 },
>>> {04,           0x7B,    0xF0,     0x0,     0x0,
>> 0x0 },
>>> {05,           0x7D,    0xF0,     0x0,     0x0,
>> 0x0 },
>>> {06,           0x7F,    0xF0,     0x0,     0x0,
>> 0x0 },
>>> ......
>>> };
>>> 
>>> This global data array is very large, about 20KB.
>>> 
>>> But I found the PEIM's compiled size is about 5KB, it
>> seems not include this global data array in the efi binary statically.
>>> Even the release version PEIM's size is larger than
>> debug version.
>>> 
>>> I found when compiling release version, the link
>> option has :
>>> /MERGE:.rdata=.data It seems release binary included
>> the global data array in the final efi image.
>>> 
>>> So, my question is:
>>> Will the global data array be in linked into the final
>> efi binary image when compiling debug version?
>>> 
>>> Thanks
>>> 
>>> 
>>> 保密声明:
>>> 本邮件含有保密或专有信息,仅供指定收件人使用。严禁对本邮件或其内
>> 容做任何未经授权的查阅、使用、复制或转发。
>>> CONFIDENTIAL NOTE:
>>> This email contains confidential or legally privileged
>> information and
>>> is for the sole use of its intended recipient. Any
>> unauthorized review, use, copying or forwarding of this email or the
>> content of this email is strictly prohibited.
>>> 
>>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 保密声明:
>> 本邮件含有保密或专有信息,仅供指定收件人使用。严禁对本邮件或其内容
>> 做任何未经授权的查阅、使用、复制或转发。
>> CONFIDENTIAL NOTE:
>> This email contains confidential or legally privileged information and
>> is for the sole use of its intended recipient. Any unauthorized
>> review, use, copying or forwarding of this email or the content of
>> this email is strictly prohibited.
>> 
>> 
> 
> 
> 
> 
> 
> 
> 保密声明:
> 本邮件含有保密或专有信息,仅供指定收件人使用。严禁对本邮件或其内容做任何未经授权的查阅、使用、复制或转发。
> CONFIDENTIAL NOTE:
> This email contains confidential or legally privileged information and is for the sole use of its intended recipient. Any unauthorized review, use, copying or forwarding of this email or the content of this email is strictly prohibited.
> 
> 


[-- Attachment #2: Type: text/html, Size: 37773 bytes --]

  reply	other threads:[~2020-08-06  3:46 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-08-06  3:31 [edk2-devel] question about MSFT compiler/link option Tiger Liu(BJ-RD)
2020-08-06  3:46 ` Andrew Fish [this message]
  -- strict thread matches above, loose matches on Subject: below --
2020-08-10  3:25 Tiger Liu(BJ-RD)
2020-08-05  6:33 Tiger Liu(BJ-RD)
2020-08-05 15:17 ` Michael D Kinney
2020-08-04  9:34 Tiger Liu(BJ-RD)
2020-08-04 15:20 ` Liming Gao

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=1CFCC7E5-6BB3-467D-885E-E74CF0FCA5A0@apple.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