public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
* [Bug target/39258] No ABI warnings on __m128i when SSE is disabled [not found] <bug-39258-4@http.gcc.gnu.org/bugzilla/> @ 2021-08-09 6:28 ` pinskia at gcc dot gnu.org 0 siblings, 0 replies; 4+ messages in thread From: pinskia at gcc dot gnu.org @ 2021-08-09 6:28 UTC (permalink / raw) To: gcc-bugs https://gcc.gnu.org/bugzilla/show_bug.cgi?id=39258 --- Comment #4 from Andrew Pinski <pinskia at gcc dot gnu.org> --- union sse2 does not pass via a register for me on x86_64 ^ permalink raw reply [flat|nested] 4+ messages in thread
* [Bug target/39258] New: No ABI warnings on __m128i when SSE is disabled @ 2009-02-21 17:28 hjl dot tools at gmail dot com 2009-02-22 19:21 ` [Bug target/39258] " hjl dot tools at gmail dot com ` (2 more replies) 0 siblings, 3 replies; 4+ messages in thread From: hjl dot tools at gmail dot com @ 2009-02-21 17:28 UTC (permalink / raw) To: gcc-bugs [-- Warning: decoded text below may be mangled, UTF-8 assumed --] [-- Attachment #1: Type: text/plain, Size: 1118 bytes --] bash-3.2$ cat /tmp/u1.i typedef long long __m128i __attribute__ ((__vector_size__ (16), __may_alias__)); union sse2 { float i; __m128i x; }; void foo2 (union sse2); void bar2 (union sse2 x) { foo2 (x); } bash-3.2$ gcc -S -O2 /tmp/u1.i -mno-sse bash-3.2$ gcc -S -O2 /tmp/u1.i -mno-sse -m32 bash-3.2$ cat /tmp/u2.i typedef long long __m128i __attribute__ ((__vector_size__ (16), __may_alias__)); void foo2 (__m128i); void bar2 (__m128i y, __m128i x) { foo2 (x); } bash-3.2$ gcc -S -O2 /tmp/u2.i -mno-sse I am expecting a warning here, like bash-3.2$ gcc -S -O2 /tmp/u2.i -mno-sse -m32 /tmp/u2.i: In function bar2: /tmp/u2.i:7: warning: SSE vector argument without SSE enabled changes the ABI bash-3.2$ -- Summary: No ABI warnings on __m128i when SSE is disabled Product: gcc Version: 4.4.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: target AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: hjl dot tools at gmail dot com http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39258 ^ permalink raw reply [flat|nested] 4+ messages in thread
* [Bug target/39258] No ABI warnings on __m128i when SSE is disabled 2009-02-21 17:28 [Bug target/39258] New: " hjl dot tools at gmail dot com @ 2009-02-22 19:21 ` hjl dot tools at gmail dot com 2009-02-22 22:04 ` hjl dot tools at gmail dot com 2009-04-16 18:33 ` pinskia at gcc dot gnu dot org 2 siblings, 0 replies; 4+ messages in thread From: hjl dot tools at gmail dot com @ 2009-02-22 19:21 UTC (permalink / raw) To: gcc-bugs [-- Warning: decoded text below may be mangled, UTF-8 assumed --] [-- Attachment #1: Type: text/plain, Size: 1015 bytes --] ------- Comment #1 from hjl dot tools at gmail dot com 2009-02-22 19:21 ------- The ABI warnings are inconsistent: bash-3.2$ cat u3.i typedef long long __m128i __attribute__ ((__vector_size__ (16), __may_alias__)); __m128i bar2 (void) { __m128i x = (__m128i) { 0, 0 }; return x; } bash-3.2$ gcc -S -mno-sse u3.i u3.i: In function bar2: u3.i:5: error: SSE register return with SSE disabled bash-3.2$ gcc -S -mno-sse u3.i -m32 bash-3.2$ construct_container has /* We allowed the user to turn off SSE for kernel mode. Don't crash if some less clueful developer tries to use floating-point anyway. */ But we issue an error. -- hjl dot tools at gmail dot com changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |ubizjak at gmail dot com Target Milestone|--- |4.5.0 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39258 ^ permalink raw reply [flat|nested] 4+ messages in thread
* [Bug target/39258] No ABI warnings on __m128i when SSE is disabled 2009-02-21 17:28 [Bug target/39258] New: " hjl dot tools at gmail dot com 2009-02-22 19:21 ` [Bug target/39258] " hjl dot tools at gmail dot com @ 2009-02-22 22:04 ` hjl dot tools at gmail dot com 2009-04-16 18:33 ` pinskia at gcc dot gnu dot org 2 siblings, 0 replies; 4+ messages in thread From: hjl dot tools at gmail dot com @ 2009-02-22 22:04 UTC (permalink / raw) To: gcc-bugs ------- Comment #2 from hjl dot tools at gmail dot com 2009-02-22 22:04 ------- On 32bit, -mno-avx/-mno-sse/-mno-mmx changes ABI for 1. Vector returns. 2. Vector parameters without varargs. On 64bit, 1. -mno-avx/-mno-sse changes ABI for a. Float/Vector and aggregate with float/vector returns. b. Float/Vector and aggregate with float/vector parameters. 2. -mno-i387 changes ABI for a. long double and aggregates with long double returns. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39258 ^ permalink raw reply [flat|nested] 4+ messages in thread
* [Bug target/39258] No ABI warnings on __m128i when SSE is disabled 2009-02-21 17:28 [Bug target/39258] New: " hjl dot tools at gmail dot com 2009-02-22 19:21 ` [Bug target/39258] " hjl dot tools at gmail dot com 2009-02-22 22:04 ` hjl dot tools at gmail dot com @ 2009-04-16 18:33 ` pinskia at gcc dot gnu dot org 2 siblings, 0 replies; 4+ messages in thread From: pinskia at gcc dot gnu dot org @ 2009-04-16 18:33 UTC (permalink / raw) To: gcc-bugs ------- Comment #3 from pinskia at gcc dot gnu dot org 2009-04-16 18:33 ------- Stop setting the target milestone unless it is a regression. -- pinskia at gcc dot gnu dot org changed: What |Removed |Added ---------------------------------------------------------------------------- Target Milestone|4.5.0 |--- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39258 ^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2021-08-09 6:28 UTC | newest] Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- [not found] <bug-39258-4@http.gcc.gnu.org/bugzilla/> 2021-08-09 6:28 ` [Bug target/39258] No ABI warnings on __m128i when SSE is disabled pinskia at gcc dot gnu.org 2009-02-21 17:28 [Bug target/39258] New: " hjl dot tools at gmail dot com 2009-02-22 19:21 ` [Bug target/39258] " hjl dot tools at gmail dot com 2009-02-22 22:04 ` hjl dot tools at gmail dot com 2009-04-16 18:33 ` pinskia at gcc dot gnu dot org
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).