public inbox for gcc-prs@sourceware.org help / color / mirror / Atom feed
From: trauscher@loytec.com To: gcc-gnats@gcc.gnu.org Subject: target/3218: ARM/Thumb: Illegal byte-oriented stack instructions Date: Mon, 18 Jun 2001 01:16:00 -0000 [thread overview] Message-ID: <200106180810.f5I8Ahe13239@lexx.office.loytec.com> (raw) >Number: 3218 >Category: target >Synopsis: ARM/Thumb: Illegal byte-oriented stack instructions >Confidential: no >Severity: serious >Priority: medium >Responsible: unassigned >State: open >Class: wrong-code >Submitter-Id: net >Arrival-Date: Mon Jun 18 01:16:00 PDT 2001 >Closed-Date: >Last-Modified: >Originator: Thomas Rauscher >Release: 3.0 20010528 (prerelease) >Organization: LoyTec electronics >Environment: System: Linux lexx 2.2.17 #1 Thu Oct 12 11:50:54 CEST 2000 i686 unknown Architecture: i686 host: i686-pc-linux-gnu build: i686-pc-linux-gnu target: arm-rtems-elf configured with: ./configure --prefix=/prj/rtems/test --target=arm-rtems-elf --enable-languages=c --with-newlib >Description: Under some cirumstances (after some byte-oriented stack access), the compiler generates sp-relative strb instructions, which are not supported by the Thumb instruction set. When the code is less complex than the example below, the sp register gets copied into an accessible low register before accessing the stack, so the problem only occurs in more complex functions. Switches: -mthumb -mbig-endian -mthumb-interwork -O3 >How-To-Repeat: Preprocessor output: # 16 "strb.c" typedef struct { char x1, x2, x3, x4, x5, *x6; } s1; typedef struct { char x, y[8]; } s2; extern int *a1; extern s2 *a2; void f1(void *p1, void *p2); void f2(int n, char *p, int *out) { s1 in; in.x1 = p[0]; in.x2 = p[1]; in.x3 = p[2]; in.x4 = p[3]; in.x5 = p[4]; in.x6 = p; if(!in.x1) { f1(&a2[n].x, in.x1); f1(&a2[n].x, &a1[n]); if (n) *out = 0; if (in.x1) *out = 0; } } >Fix: Probably the thumb_movqi_insn should not contain the ldrb/strb/ldrh/strh instructions in conjunction with the high registers for thumb mode. >Release-Note: >Audit-Trail: >Unformatted:
next reply other threads:[~2001-06-18 1:16 UTC|newest] Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top 2001-06-18 1:16 trauscher [this message] 2002-01-22 3:12 pb 2002-01-22 3:26 Momchil Velikov 2002-09-30 5:08 rearnsha
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=200106180810.f5I8Ahe13239@lexx.office.loytec.com \ --to=trauscher@loytec.com \ --cc=gcc-gnats@gcc.gnu.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: linkBe 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).