Summary: | ICE with invalid %y operand (inline-asm) | ||
---|---|---|---|
Product: | gcc | Reporter: | Andrew Pinski <pinskia> |
Component: | target | Assignee: | Andrew Pinski <pinskia> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | gcc-bugs |
Priority: | P3 | Keywords: | ice-on-invalid-code, patch |
Version: | 4.3.0 | ||
Target Milestone: | 4.4.0 | ||
URL: | http://gcc.gnu.org/ml/gcc-patches/2008-07/msg01058.html | ||
Host: | Target: | powerpc*-*-* | |
Build: | Known to work: | ||
Known to fail: | Last reconfirmed: | 2007-07-09 08:50:46 |
Description
Andrew Pinski
2007-04-13 23:05:44 UTC
I have a patch for this when I get back in August. Testing this patch right now: Index: testsuite/gcc.target/powerpc/asm-y.c =================================================================== --- testsuite/gcc.target/powerpc/asm-y.c (revision 0) +++ testsuite/gcc.target/powerpc/asm-y.c (revision 0) @@ -0,0 +1,11 @@ +/* { dg-do compile } */ +/* { dg-options "-O1" } */ + +/* Test that %yN does not cause an internal error if used incorrectly. */ + +int f(int *a) +{ + asm ("#%y0" : "=m"(a[2])); /* { dg-error "try using the 'Z' constraint" } */ + asm ("#%y0" : "=m"(a[1])); /* { dg-error "try using the 'Z' constraint" } */ + asm ("#%y0" : "=m"(a[0])); +} Index: config/rs6000/rs6000.c =================================================================== --- config/rs6000/rs6000.c (revision 137784) +++ config/rs6000/rs6000.c (working copy) @@ -12288,9 +12288,13 @@ print_operand (FILE *file, rtx x, int co fprintf (file, "0,%s", reg_names[REGNO (tmp)]); else { - gcc_assert (GET_CODE (tmp) == PLUS - && REG_P (XEXP (tmp, 0)) - && REG_P (XEXP (tmp, 1))); + if (!GET_CODE (tmp) == PLUS + || !REG_P (XEXP (tmp, 0)) + || !REG_P (XEXP (tmp, 1))) + { + output_operand_lossage ("invalid %%y value, try using the 'Z' constraint"); + break; + } if (REGNO (XEXP (tmp, 0)) == 0) fprintf (file, "%s,%s", reg_names[ REGNO (XEXP (tmp, 1)) ], Fixed. Subject: Bug 31568 Author: pinskia Date: Tue Jul 15 12:52:56 2008 New Revision: 137837 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=137837 Log: 2008-07-15 Andrew Pinski <andrew_pinski@playstation.sony.com> PR target/31568 * config/rs6000/rs6000.c (print_operand <case 'y'>): Don't use gcc_assert, instead call output_operand_lossage. 2008-07-15 Andrew Pinski <andrew_pinski@playstation.sony.com> PR target/31568 * gcc.target/powerpc/asm-y.c: New testcase. Added: trunk/gcc/testsuite/gcc.target/powerpc/asm-y.c Modified: trunk/gcc/ChangeLog trunk/gcc/config/rs6000/rs6000.c trunk/gcc/testsuite/ChangeLog |