| commit | 45507ff3a2c58bc4781688eac1a80979ea972dce | |
| author | Benjamin Herrenschmidt <benh@kernel.crashing.org> | |
| Tue, 4 Jul 2006 04:14:07 +0000 (4 14:14 +1000) | ||
| committer | Paul Mackerras <paulus@samba.org> | |
| Tue, 25 Jul 2006 15:27:27 +0000 (26 01:27 +1000) | ||
| tree | b4e6db21249a59bf03d7e63d1539166421040a75 | treesnapshot (tar.gz zip) |
| parent | 171505dafef56e4c84c48399b36604f4be55cf36 | commitdiff |
[POWERPC] Fix legacy_serial.c error handling on 32 bits
The code in legacy_serial.c wouldn't properly compare OF translation
results against OF_BAD_ADDR as it's using a phys_addr_t which is 32
bits on some 32-bit powerpc platforms. This fixes it by always using
a u64 which is what is returned by the OF parsing routines. It also
makes translation failure harmless for ISA serial ports. If they
can't translate, we can't use the UART early, but we can still let the
8250 driver use it later on by using IO port accessors.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The code in legacy_serial.c wouldn't properly compare OF translation
results against OF_BAD_ADDR as it's using a phys_addr_t which is 32
bits on some 32-bit powerpc platforms. This fixes it by always using
a u64 which is what is returned by the OF parsing routines. It also
makes translation failure harmless for ISA serial ports. If they
can't translate, we can't use the UART early, but we can still let the
8250 driver use it later on by using IO port accessors.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
| arch/powerpc/kernel/legacy_serial.c | diffblobblamehistory |