From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 48) id B02323858C3A; Thu, 29 Feb 2024 22:46:45 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org B02323858C3A DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org; s=default; t=1709246805; bh=HpSHylqjo410gkAWxhFCQAowoMj5s7Tx4+MhGvUyoEE=; h=From:To:Subject:Date:In-Reply-To:References:From; b=BVu9vdSUB0pN2MuwBr728ZVhAmKMU/Y9rE61Yu5c89FO4fglG5ON/ZqUCaM4YX5s8 Vv3PqD/j40HglM8ZGVsgUYOwHU8oduGYPGgMJJgeadAMDv+8M2PtYrd1xawP0LwVFa HO2R6sjFQWQHI8Agex6fMWTOqCEzQsictfuiAcmE= From: "ewlu at rivosinc dot com" To: gcc-bugs@gcc.gnu.org Subject: [Bug target/114175] [13/14] RISC-V: Execution test failures on gcc.dg/c23-stdarg-6.c Date: Thu, 29 Feb 2024 22:46:44 +0000 X-Bugzilla-Reason: CC X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: gcc X-Bugzilla-Component: target X-Bugzilla-Version: 14.0 X-Bugzilla-Keywords: X-Bugzilla-Severity: normal X-Bugzilla-Who: ewlu at rivosinc dot com X-Bugzilla-Status: NEW X-Bugzilla-Resolution: X-Bugzilla-Priority: P3 X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org X-Bugzilla-Target-Milestone: --- X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 List-Id: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D114175 --- Comment #16 from Edwin Lu --- (In reply to palmer from comment #15) > It's a little easier to see from the float version of the code. >=20 > $ cat gcc/testsuite/gcc.dg/c23-stdarg-6.c=20 > /* Test C23 variadic functions with no named parameters, or last named > parameter with a declaration not allowed in C17. Execution tests. */ > /* { dg-do run } */ > /* { dg-options "-std=3Dc23 -pedantic-errors" } */ >=20 > #include > #include >=20 > extern void abort (void); > extern void exit (int); > struct s { char c[1000]; }; >=20 > struct s > f (...) > { > va_list ap; > va_start (ap); > int r =3D va_arg (ap, double); > va_end (ap); > struct s ret =3D {}; > ret.c[0] =3D r; > ret.c[999] =3D 42; > return ret; > } >=20 > int > main () > { > struct s x =3D f (1.0); > fprintf(stderr, "%d\n", x.c[0]); > if (x.c[0] !=3D 1) > abort (); > exit (0); > } > $ riscv64-unknown-linux-gnu-gcc gcc/testsuite/gcc.dg/c23-stdarg-6.c -o te= st > -std=3Dc2x -static -O3 > $ riscv64-unknown-linux-gnu-objdump -d test > ... > 0000000000010412
: > ... > 1042e: 850a mv a0,sp > ... > 10438: 112000ef jal 1054a > ... > 000000000001054a : > 1054a: f20507d3 fmv.d.x fa5,a0 >=20 > The psABI says >=20 > A callee with variadic arguments is responsible for copying the conte= nts > of registers used to pass variadic arguments to the vararg save area, > which must be contiguous with arguments passed on the stack. >=20 > which I'm taking to mean the "1.0" is meant to be passed in a register. = It > also says >=20 > Values are returned in the same manner as a first named argument of t= he > same type would be passed. If such an argument would have been passed= by > reference, the caller allocates memory for the return value, and pass= es > the address as an implicit first parameter. >=20 The psABI also says this in the paragraph before In the base integer calling convention, variadic arguments are passed= =20 in the same manner as named arguments, with one exception. Variadic=20 arguments with 2=C3=97XLEN-bit alignment and size at most 2=C3=97XLEN= bits are passed in an aligned register pair (i.e., the first register in the p= air=20 is even-numbered), or on the stack by value if none is available. Aft= er a variadic argument has been passed on the stack, all future arguments = will also be passed on the stack (i.e. the last argument register may be l= eft=20 unused due to the aligned register pair rule). > So I think we're screwing up both ends of this one: the caller is passing > the return struct in a0 (losing the first arg), which the callee is > obtaining the first argument from a0 (losing the return struct). >=20 > That all very much seems like a backend bug to me. So if I understand correctly, there may also be a problem where it's trying= to create that named first argument but also trying to pass it as a variadic argument.=