Re: what is the difference with and without crc extension support
On 3/3/22 14:52, Dongjiu Geng wrote: Thanks for the answer, my source file is complex，it is not convenient attach, because it is company code. I see. So then you can try -fdump-tree-optimized and compare the corresponding dump file. Or you can use -fdump-tree-all -fdump-rtl-all and do the same. According to your experience， what is the reason about it? Dunno. Cheers, Martin Martin Liška 于2022年3月3日周四 21:48写道： On 3/3/22 14:41, Dongjiu Geng via Gcc wrote: Hi, My program does not use CRC instructions，but I find the compiled binary has much difference between using "-march=armv8-a+crc" and using "-march=armv8-a". Even stranger, when I use "-march=armv8-a+crc", I find my compiled binary can not run. but when I change -O2 to -O0 based on "-march=armv8-a+crc", it can run. I do not know what is the reason.. can you answer it? Thanks. Hello. Please report bug here: https://gcc.gnu.org/bugzilla/ and attach a pre-processed source file: https://gcc.gnu.org/bugs/ Martin
Save/Store information, which variable are not initialized
I known, IDE will detect cases, whereas variable is not initialized, but some developers still have problems with variable initialization. I suppose gcc do not support any run-time sanitizer, because I do not found any information. So: gcc will introduce hidden array of variables for each context. One bit stores information about one variable in current scope. So, if it is set - variable was initialized. In other case, variable is not initialized. When new scope is created, gcc will allocate this array and set each element to 0. When program set variable, also bit is set. Each structure will contains this array at top of self. GCC must hack access to this struct in some way. That is big problem to handle this case, because some languages allow to free memory access. Maybe for struct store information about fields state in different part of memory? Idea was inspired by way to detect cases, where return address was written. Programmer should compile program with special switch to enable this mode. GDB could support this feature of gcc and read, which variable was not set.
Re: what is the difference with and without crc extension support
On 03/03/2022 13:41, Dongjiu Geng via Gcc wrote: Hi, My program does not use CRC instructions，but I find the compiled binary has much difference between using "-march=armv8-a+crc" and using "-march=armv8-a". Even stranger, when I use "-march=armv8-a+crc", I find my compiled binary can not run. but when I change -O2 to -O0 based on "-march=armv8-a+crc", it can run. I do not know what is the reason.. can you answer it? Thanks. The most common reason for a program failing to run when the optimizer is turned on is that it contains undefined behaviour. Make sure that you turn on GCC's warning options and fix anything that these report. It's very unlikely that the change of -march=armv8-a to -march=armv8-a+crc is the reason for the differences in compiled code that you are seeing. Something else to try here is to identify an object file that is different between the two options and then rebuild it both ways, but add "-g". You can then use readelf -debug-dump=str to print out the real options used by the compiler (there may be some additional strings reported in the dump, but the options are the bit that is interesting). The only difference should be the -march option you mentioned above. R.
gcc-10-20220304 is now available
Snapshot gcc-10-20220304 is now available on https://gcc.gnu.org/pub/gcc/snapshots/10-20220304/ and on various mirrors, see http://gcc.gnu.org/mirrors.html for details. This snapshot has been generated from the GCC 10 git branch with the following options: git://gcc.gnu.org/git/gcc.git branch releases/gcc-10 revision 69dadb7dc3507a1e6dd59a3a905901dd1653db85 You'll find: gcc-10-20220304.tar.xz Complete GCC SHA256=04baba86b4538c0fc758351b4df8e22b4267be85df737da651db94848e35450b SHA1=2fc5088d05b28d1303e76a11c5433facbd646f5e Diffs from 10-20220225 are available in the diffs/ subdirectory. When a particular snapshot is ready for public consumption the LATEST-10 link is updated and a message is sent to the gcc list. Please do not use a snapshot before it has been announced that way.
Re: Reqister to become gcc mirror site
Hi Ceasar, On Tue, 22 Feb 2022, Ceasar Sun wrote: > We sent the follow message two weeks ago, buy didn't get response yet. > So, send this again and sorry for the inconvenience ~ I am sorry this didn't see an answer earlier. It looks like you are actively mirroring our download site already? > I'm Ceasar Sun , from Free Software Lab, NCHC , Taiwan. > We provide several free software mirror service in Taiwan : free.nchc.org.tw > We would like to become a gcc mirror site and the service would be presented > as : > > Country : Taiwan ,TW > City: Hsinchu > Site: https://free.nchc.org.tw/gcc/ > Contact e-mail : fs...@nchc.org.tw > Contact Name : FSLab Team , NCHC > > So, which upstream gcc site we should use ? It's best if you use gcc.gnu.org directly. Is this what you are now doing? Is there anything you need? (I believe it's probably best to mirror using rsync.) As a next step we then just need to add you to our mirros page at https://gcc.gnu.org/mirrors.html, something like the following: diff --git a/htdocs/mirrors.html b/htdocs/mirrors.html index 75c71b95..8e35bc08 100644 --- a/htdocs/mirrors.html +++ b/htdocs/mirrors.html @@ -38,6 +38,7 @@ mirrors. The following sites mirror the gcc.gnu.org download site The Netherlands, Nijmegen: https://ftp.nluug.nl/languages/gcc/";>ftp.nluug.nl, thanks to Jan Cristiaan van Winkel (j...@atcomputing.nl) Russia, Novosibirsk: http://mirror.linux-ia64.org/gnu/gcc/";>http://mirror.linux-ia64.org/gnu/gcc/, thanks to Daniel Volchixin (dan...@volchixin.co.uk) Slovakia, Bratislava: http://gcc.fyxm.net/";>gcc.fyxm.net, thanks to Jan Teluch (ad...@2600.sk) +Taiwan, Hsinchu: https://free.nchc.org.tw/gcc/";>nchc.org.tw, thanks to FSLab Team, NCHC (fs...@nchc.org.tw) UK: https://mirrorservice.org/sites/sourceware.org/pub/gcc/";>mirrorservice.org, thanks to mir...@mirrorservice.org US, San Francisco: https://bigsearcher.com/mirrors/gcc/";>https://bigsearcher.com/mirrors/gcc/, thanks to i...@bigsearcher.com US, San Jose: http://www.netgull.com/gcc/";>http://www.netgull.com, thanks to ad...@netgull.com What do you think? Are there any changes to this proposed patch you suggest? Thank you, Gerald
