From: Adhemerval Zanella Netto <adhemerval.zanella@linaro.org>
To: "H.J. Lu" <hjl.tools@gmail.com>
Cc: libc-alpha@sourceware.org, Joseph Myers <josmyers@redhat.com>,
Florian Weimer <fweimer@redhat.com>
Subject: Re: [PATCH v3 7/7] math: Provide modf128 for static libm on alpha, s390, and sparcv9
Date: Tue, 21 May 2024 13:41:18 -0300 [thread overview]
Message-ID: <15bdbb7d-aa1c-4159-9c47-20c797047281@linaro.org> (raw)
In-Reply-To: <CAMe9rOoaVXjbzVmQ0OsXqRJzksz+FeCOqrcb4id_67gQBcVnng@mail.gmail.com>
On 21/05/24 12:40, H.J. Lu wrote:
> On Mon, May 20, 2024 at 12:07 PM Adhemerval Zanella Netto
> <adhemerval.zanella@linaro.org> wrote:
>>
>>
>>
>> On 20/05/24 13:59, H.J. Lu wrote:
>>> On Tue, Apr 2, 2024 at 7:07 AM Adhemerval Zanella
>>> <adhemerval.zanella@linaro.org> wrote:
>>>>
>>>> Checked with a build for the affected ABIs
>>>> ---
>>>> sysdeps/ieee754/ldbl-64-128/s_modfl.c | 4 ++--
>>>> 1 file changed, 2 insertions(+), 2 deletions(-)
>>>>
>>>> diff --git a/sysdeps/ieee754/ldbl-64-128/s_modfl.c b/sysdeps/ieee754/ldbl-64-128/s_modfl.c
>>>> index 7d7aeae111..ba3d31334a 100644
>>>> --- a/sysdeps/ieee754/ldbl-64-128/s_modfl.c
>>>> +++ b/sysdeps/ieee754/ldbl-64-128/s_modfl.c
>>>> @@ -1,10 +1,10 @@
>>>> #include <math_ldbl_opt.h>
>>>> #include <libm-alias-ldouble.h>
>>>> -#if IS_IN (libc)
>>>> +#if IS_IN (libc) && defined SHARED
>>>> # undef libm_alias_ldouble
>>>> # define libm_alias_ldouble(from, to)
>>>> #endif
>>>> #include <sysdeps/ieee754/ldbl-128/s_modfl.c>
>>>> -#if IS_IN (libc)
>>>> +#if IS_IN (libc) && defined SHARED
>>>> long_double_symbol (libc, __modfl, modfl);
>>>
>>> Doesn't this remove modfl from static library?
>>
>> As for copysignf128 and frexp128, this symbol is also provided by libc.a.
>> Before this patch:
>>
>> alpha-linux-gnu$ readelf -sW libc.a | grep -w modf.*
>> 19: 0000000000000000 572 FUNC WEAK DEFAULT [STD GPLOAD] 1 modfl
>> 17: 0000000000000000 304 FUNC WEAK DEFAULT [STD GPLOAD] 1 modff32x
>> 18: 0000000000000000 304 FUNC WEAK DEFAULT [STD GPLOAD] 1 modff64
>> 19: 0000000000000000 304 FUNC WEAK DEFAULT [STD GPLOAD] 1 modf
>> 17: 0000000000000000 244 FUNC WEAK DEFAULT [STD GPLOAD] 1 modff32
>> 18: 0000000000000000 244 FUNC WEAK DEFAULT [STD GPLOAD] 1 modff
>>
>> After this patch:
>>
>> alpha-linux-gnu$ readelf -sW libc.a | grep -w modf.*
>> 19: 0000000000000000 572 FUNC WEAK DEFAULT [STD GPLOAD] 1 modff64x
>> 20: 0000000000000000 572 FUNC WEAK DEFAULT [STD GPLOAD] 1 modff128
>> 21: 0000000000000000 572 FUNC WEAK DEFAULT [STD GPLOAD] 1 modfl
>> 17: 0000000000000000 304 FUNC WEAK DEFAULT [STD GPLOAD] 1 modff32x
>> 18: 0000000000000000 304 FUNC WEAK DEFAULT [STD GPLOAD] 1 modff64
>> 19: 0000000000000000 304 FUNC WEAK DEFAULT [STD GPLOAD] 1 modf
>> 17: 0000000000000000 244 FUNC WEAK DEFAULT [STD GPLOAD] 1 modff32
>> 18: 0000000000000000 244 FUNC WEAK DEFAULT [STD GPLOAD] 1 modff
>>
>>
>
> I am working on a patch to check missing symbols in static libraries.
> I opened:
>
> https://sourceware.org/bugzilla/show_bug.cgi?id=31781
>
> scalbnf128 seems also missing in alpha libm.a.
>
It seems so, the static tests I have added only cover for the auto-generated
function from the libm-test-funcs-auto, libm-test-funcs-noauto, and
libm-test-funcs-narrow; and the are some function that are not covered by
these (scalbn for instance).
next prev parent reply other threads:[~2024-05-21 16:41 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-04-02 14:06 [PATCH v3 0/7] Fix some libm static issues Adhemerval Zanella
2024-04-02 14:06 ` [PATCH v3 1/7] math: Add support for auto static math tests Adhemerval Zanella
2024-05-20 16:48 ` H.J. Lu
2024-05-20 17:21 ` Adhemerval Zanella Netto
2024-04-02 14:06 ` [PATCH v3 2/7] math: Fix i386 and m68k fmod/fmodf on static build (BZ 31488) Adhemerval Zanella
2024-05-11 9:59 ` Aurelien Jarno
2024-05-21 12:40 ` H.J. Lu
2024-05-21 12:54 ` Adhemerval Zanella Netto
2024-04-02 14:06 ` [PATCH v3 3/7] math: Fix i386 and m68k exp10 on static build Adhemerval Zanella
2024-05-11 9:59 ` Aurelien Jarno
2024-05-21 12:48 ` H.J. Lu
2024-05-21 12:54 ` Adhemerval Zanella Netto
2024-04-02 14:06 ` [PATCH v3 4/7] math: Fix isnanf128 " Adhemerval Zanella
2024-05-20 16:51 ` H.J. Lu
2024-05-20 18:53 ` Adhemerval Zanella Netto
2024-05-20 21:34 ` H.J. Lu
2024-05-21 12:32 ` Adhemerval Zanella Netto
2024-05-21 12:36 ` H.J. Lu
2024-05-21 12:56 ` Adhemerval Zanella Netto
2024-04-02 14:06 ` [PATCH v3 5/7] math: Provided copysignf128 for static libm on alpha, s390, and sparcv9 Adhemerval Zanella
2024-05-20 16:55 ` H.J. Lu
2024-05-20 19:04 ` Adhemerval Zanella Netto
2024-04-02 14:06 ` [PATCH v3 6/7] math: Provide frexpf128 " Adhemerval Zanella
2024-05-20 16:57 ` H.J. Lu
2024-05-20 19:06 ` Adhemerval Zanella Netto
2024-04-02 14:06 ` [PATCH v3 7/7] math: Provide modf128 " Adhemerval Zanella
2024-05-20 16:59 ` H.J. Lu
2024-05-20 19:07 ` Adhemerval Zanella Netto
2024-05-21 15:40 ` H.J. Lu
2024-05-21 16:41 ` Adhemerval Zanella Netto [this message]
2024-05-20 16:31 ` [PATCH v3 0/7] Fix some libm static issues Adhemerval Zanella Netto
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-all 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=15bdbb7d-aa1c-4159-9c47-20c797047281@linaro.org \
--to=adhemerval.zanella@linaro.org \
--cc=fweimer@redhat.com \
--cc=hjl.tools@gmail.com \
--cc=josmyers@redhat.com \
--cc=libc-alpha@sourceware.org \
/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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).