Buy commercial curl support from WolfSSL. We help you work
out your issues, debug your libcurl applications, use the API, port to new
platforms, add new features and more. With a team lead by the curl founder
himself.
Re: RELEASE: curl 7.75.0
- Contemporary messages sorted: [ by date ] [ by thread ] [ by subject ] [ by author ] [ by messages with attachments ]
From: Dennis Clarke via curl-library <curl-library_at_cool.haxx.se>
Date: Sat, 27 Mar 2021 17:05:34 +0000
On 3/27/21 11:50 AM, Daniel Stenberg wrote:
> On Sat, 27 Mar 2021, Dennis Clarke via curl-library wrote:
>
>> TESTFAIL: These test cases failed: 1188
>
> We've updated test 1188 since to make it more reliable in more
> surroundings. I think it is likely to have fixed the issue for you as well.
>
I tend to stay with the production releases however I think that I had
nightly builds going somewhere. I will look around as I am fairly sure
that there were nightlies running on a cron job in some server in a
corner somewhere. I know that sounds strange but with enough systems
running in a datacenter we sometimes lose track. I am certain that I had
curl building nightly somewhere. Now I need to go look.
>> Also, I did see strange optimization flags inserted into the
>> Makefile(s) and I was able to catch that and edit all the tree
>> Makefile(s) to remove a strange -xO2 which I did not want.
>
> That's done by the CURL_CHECK_COMPILER_SUNPRO_C macro we have in
> m4/curl-compilers.m4
>
> Are those flags wrong?
>
Only a little.
alpha $ echo $CFLAGS
-m64 -xarch=sparc -xO0 -g -Xa -errfmt=error -erroff=%none -errshort=full
-xstrconst -xildoff -xmemalign=8s -xnolibmil -xcode=pic32 -xregs=no%appl
-xlibmieee -mc -ftrap=%none -xbuiltin=%none -xunroll=1 -Qy
-xdebugformat=dwarf
Well on some software from the open source world I push C99 as the spec
to enforce.
The above is much more relaxed. If that is the word. OpenSSL is not C99
and has never been anything past C89. That is fine as it needs to be
used everywhere. Same with curl. So I usually start with a relaxed set
of CFLAGS that allow for C89 and then try again with C99 later. Please
note the -xO0 there. No optimization at all. None. Also the -g for a
debug build.
Strangely the resultant Makefiles all had -xO2 in them. That was odd.
Anyways I manually edited all Makefiles to get the CFLAGS back to the
above and everything seems fine. Except for test 1188 which is odd.
Date: Sat, 27 Mar 2021 17:05:34 +0000
On 3/27/21 11:50 AM, Daniel Stenberg wrote:
> On Sat, 27 Mar 2021, Dennis Clarke via curl-library wrote:
>
>> TESTFAIL: These test cases failed: 1188
>
> We've updated test 1188 since to make it more reliable in more
> surroundings. I think it is likely to have fixed the issue for you as well.
>
I tend to stay with the production releases however I think that I had
nightly builds going somewhere. I will look around as I am fairly sure
that there were nightlies running on a cron job in some server in a
corner somewhere. I know that sounds strange but with enough systems
running in a datacenter we sometimes lose track. I am certain that I had
curl building nightly somewhere. Now I need to go look.
>> Also, I did see strange optimization flags inserted into the
>> Makefile(s) and I was able to catch that and edit all the tree
>> Makefile(s) to remove a strange -xO2 which I did not want.
>
> That's done by the CURL_CHECK_COMPILER_SUNPRO_C macro we have in
> m4/curl-compilers.m4
>
> Are those flags wrong?
>
Only a little.
alpha $ echo $CFLAGS
-m64 -xarch=sparc -xO0 -g -Xa -errfmt=error -erroff=%none -errshort=full
-xstrconst -xildoff -xmemalign=8s -xnolibmil -xcode=pic32 -xregs=no%appl
-xlibmieee -mc -ftrap=%none -xbuiltin=%none -xunroll=1 -Qy
-xdebugformat=dwarf
Well on some software from the open source world I push C99 as the spec
to enforce.
The above is much more relaxed. If that is the word. OpenSSL is not C99
and has never been anything past C89. That is fine as it needs to be
used everywhere. Same with curl. So I usually start with a relaxed set
of CFLAGS that allow for C89 and then try again with C99 later. Please
note the -xO0 there. No optimization at all. None. Also the -g for a
debug build.
Strangely the resultant Makefiles all had -xO2 in them. That was odd.
Anyways I manually edited all Makefiles to get the CFLAGS back to the
above and everything seems fine. Except for test 1188 which is odd.
-- Dennis Clarke RISC-V/SPARC/PPC/ARM/CISC UNIX and Linux spoken GreyBeard and suspenders optional ------------------------------------------------------------------- Unsubscribe: https://cool.haxx.se/list/listinfo/curl-library Etiquette: https://curl.se/mail/etiquette.htmlReceived on 2021-03-27