This is the mail archive of the gdb-prs@sourceware.org mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug sim/19447] New: mips sim fails some tests when using 64-bit address space


https://sourceware.org/bugzilla/show_bug.cgi?id=19447

            Bug ID: 19447
           Summary: mips sim fails some tests when using 64-bit address
                    space
           Product: gdb
           Version: unknown
            Status: NEW
          Severity: normal
          Priority: P2
         Component: sim
          Assignee: unassigned at sourceware dot org
          Reporter: vapier at gentoo dot org
                CC: vapier at gentoo dot org
  Target Milestone: ---
            Target: mips-sde-elf

before commit this commit:
https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=26f8bf63bf36f9062a5cc1afacf71462a

all addresses were implicitly truncated on mips to 32-bits via the
AddressTranslation function:
  /* For a simple (flat) memory model, we simply pass virtual
     addressess through (mostly) unchanged. */
  vAddr &= 0xFFFFFFFF;

when i cleaned that up, a bunch of tests started failing because they were sign
extending 32-bit addresses into 64-bit.  most of them were fixed by:
https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=b36d953bced0a4fecdde1823abac70ed7038ee95

but a few remain:
$ ./configure --target=mips-sde-elf
$ make all-sim
$ make check-sim
...
FAIL: mips64 fpu64-ps.s (execution)
FAIL: mips64 mdmx-ob.s (execution)
FAIL: mips32r2 mips32-dsp.s (execution)
FAIL: mips64r2 mips32-dsp.s (execution)
FAIL: mips32r2 mips32-dsp.s (execution)
FAIL: mips64r2 mips32-dsp.s (execution)

with the first one being:
ldc1
paddr: 0xffffffff80010b18 vaddr: 0xffffffff80010b18 mask: 0x7 base:
0xffffffff80010b18 off: 0
mips-core: 8 byte read to unmapped address 0xffffffff80010b18 at 0x80010070
program stopped with signal 10 (User defined signal 1).

these look like they are bugs either in the tests themselves (loading &
extending 32-bit values into 64-bit registers), or they're bugs in the mips
decode logic (sign extending 32-bit registers).  someone more familiar with the
mips logic should take a look.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]