@@ -335,5 +335,18 @@
usually displayed using a so-called "POST card" ISA/PCI/PCI-E
device) on the serial console.
+config DEBUG_MUXED
+ bool "Multiplex debug output from each core"
+ default n
+ help
+ Transmit core number with each nibble of coreboot debug output,
+ causing the debug output to be unintelligible at naked eye but
+ filtrable piping it to demux_debug to separate output from each
+ core instead of watching the garbage produced by simultaneous
+ transmission from each core. It will duplicate the size of debug
+ output, halve the speed and require demuxing, but it felt less
+ invasive that some form of mutex or blocking. Tested for serial,
+ but should also work for usb or network console
+
endmenu
@@ -30,6 +30,27 @@
#define SPECIAL 32 /* 0x */
#define LARGE 64 /* use 'ABCDEF' instead of 'abcdef' */
+unsigned char calc_id_buffer(void) {
+ return 0xf0; /* do we have some function to tell the current
+ * core/node, implemented for each cpu ?
+ */
+}
+
+static void mux_byte(void (*tx_byte)(unsigned char byte), unsigned char idbuffer, unsigned char byte) {
+#ifdef CONFIG_DEBUG_MUXED
+
+ if (!(idbuffer & DISABLE_MUXED)) {
+ tx_byte((idbuffer & 0xf0) | (byte >> 4));
+ tx_byte((idbuffer & 0xf0) | (byte & 0x0f));
+ };
+ if (idbuffer & ENABLE_UNMUXED) {
+ tx_byte(byte);
+ };
+#else
+ tx_byte(byte);
+#endif
+}
+
static int number(void (*tx_byte)(unsigned char byte),
unsigned long long num, int base, int size, int precision, int type)
{
@@ -37,6 +58,7 @@
const char *digits="0123456789abcdefghijklmnopqrstuvwxyz";
int i;
int count = 0;
+ unsigned char idbuffer = calc_id_buffer();
if (type & LARGE)
digits = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ";
@@ -75,26 +97,26 @@
size -= precision;
if (!(type&(ZEROPAD+LEFT)))
while(size-->0)
- tx_byte(' '), count++;
+ mux_byte(tx_byte,idbuffer,' '), count++;
if (sign)
- tx_byte(sign), count++;
+ mux_byte(tx_byte,idbuffer,sign), count++;
if (type & SPECIAL) {
if (base==8)
- tx_byte('0'), count++;
+ mux_byte(tx_byte,idbuffer,'0'), count++;
else if (base==16) {
- tx_byte('0'), count++;
- tx_byte(digits[33]), count++;
+ mux_byte(tx_byte,idbuffer,'0'), count++;
+ mux_byte(tx_byte,idbuffer,digits[33]), count++;
}
}
if (!(type & LEFT))
while (size-- > 0)
- tx_byte(c), count++;
+ mux_byte(tx_byte,idbuffer,c), count++;
while (i < precision--)
- tx_byte('0'), count++;
+ mux_byte(tx_byte,idbuffer,'0'), count++;
while (i-- > 0)
- tx_byte(tmp[i]), count++;
+ mux_byte(tx_byte,idbuffer,tmp[i]), count++;
while (size-- > 0)
- tx_byte(' '), count++;
+ mux_byte(tx_byte,idbuffer,' '), count++;
return count;
}
@@ -105,6 +127,7 @@
unsigned long long num;
int i, base;
const char *s;
+ unsigned char idbuffer = calc_id_buffer();
int flags; /* flags to number() */
@@ -117,7 +140,7 @@
for (count=0; *fmt ; ++fmt) {
if (*fmt != '%') {
- tx_byte(*fmt), count++;
+ mux_byte(tx_byte,idbuffer,*fmt), count++;
continue;
}
@@ -180,10 +203,10 @@
case 'c':
if (!(flags & LEFT))
while (--field_width > 0)
- tx_byte(' '), count++;
- tx_byte((unsigned char) va_arg(args, int)), count++;
+ mux_byte(tx_byte,idbuffer,' '), count++;
+ mux_byte(tx_byte,idbuffer,(unsigned char) va_arg(args, int)), count++;
while (--field_width > 0)
- tx_byte(' '), count++;
+ mux_byte(tx_byte,idbuffer,' '), count++;
continue;
case 's':
@@ -195,11 +218,11 @@
if (!(flags & LEFT))
while (len < field_width--)
- tx_byte(' '), count++;
+ mux_byte(tx_byte,idbuffer,' '), count++;
for (i = 0; i < len; ++i)
- tx_byte(*s++), count++;
+ mux_byte(tx_byte,idbuffer,*s++), count++;
while (len < field_width--)
- tx_byte(' '), count++;
+ mux_byte(tx_byte,idbuffer,' '), count++;
continue;
case 'p':
@@ -227,7 +250,7 @@
continue;
case '%':
- tx_byte('%'), count++;
+ mux_byte(tx_byte,idbuffer,'%'), count++;
continue;
/* integer number formats - set up the flags and "break" */
@@ -248,9 +271,9 @@
break;
default:
- tx_byte('%'), count++;
+ mux_byte(tx_byte,idbuffer,'%'), count++;
if (*fmt)
- tx_byte(*fmt), count++;
+ mux_byte(tx_byte,idbuffer,*fmt), count++;
else
--fmt;
continue;
@@ -23,6 +23,18 @@
#ifdef __PRE_RAM__
#include <cpu/amd/model_10xxx_msr.h>
#endif
+#include <console/vtxprintf.h>
+
+// buffer id for console/vtxprintf
+unsigned char calc_id_buffer(void) {
+ // lower 4 bits should be 0, DISABLE_MUXED, or ENABLE_UNMUXED.
+ // I only use 4 bits, so it may not be useful for systems with many cores
+ // for systems with less cores, I prefer using th higher values to avoid
+ // muxed bytes being interpreted as ASCII control characters or something
+ // so I negate (core_number mod 16)
+ return ((~( (cpuid_ebx(1) >> 24) & 0xf))<<4);
+}
+
//called by bus_cpu_scan too
u32 read_nb_cfg_54(void)
@@ -34,6 +34,21 @@
#include <stdarg.h>
#endif
+#define DISABLE_MUXED 1
+#define ENABLE_UNMUXED 2
+
+ /* lower 4 bits should be 0, DISABLE_MUXED (1) and/or ENABLE_UNMUXED (2),
+ * so :
+ * 0 : output muxed
+ * 1 : no output
+ * 2 : output unmuxed
+ * 3 : output muxed (2 bytes) followed by output unmuxed (1 byte) (any use ?)
+ * higher 4 bits should be the bitwise negation of the buffer id
+ * to use in muxed output
+ */
+unsigned char __attribute__((weak)) calc_id_buffer(void) ;
+
+
int vtxprintf(void (*tx_byte)(unsigned char byte), const char *fmt, va_list args);
#endif
@@ -0,0 +1,593 @@
+#!/usr/bin/perl -w
+# demux_debug.pl -- demultiplex coreboot output with CONFIG_DEBUG_MUXED
+# Copyright (C) 2011 Xavi Drudis Ferran <xdrudis@tinet.cat>
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License along
+# with this program; if not, write to the Free Software Foundation, Inc.,
+# 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+#
+use Getopt::Long;
+use Pod::Usage;
+use List::Util qw[min];
+use IO::Handle;
+
+my $num_buffers = 4;
+my $width_buffer = 2000;
+my @out_buffers=[];
+my @in_buffers=[];
+#my $format='%*3$s|';
+my $format='%4$1d:%1s'."\n";
+my %formats=();
+my %files=();
+my $rtrim=1;
+my $ltrim=0;
+my $lf_does_cr=1;
+my $omit_regex = qr/^\s*$/;
+#my $omit_regex = undef;
+my $line =0;
+my %handles = ();
+
+sub parse_args {
+ use Getopt::Long;
+ GetOptions( "buffers=i" => \$num_buffers,
+ "width=i" => \$width_buffer,
+ "format=s" => \$format,
+ "rtrim!" => \$rtrim,
+ "ltrim!" => \$ltrim,
+ "lf_does_cr!" => \$lf_does_cr,
+ "empty_skipped|omit_regex:s" => \$omit_regex,
+ "help" => \$help,
+ "nthformat=s%" => \%formats,
+ "outputs=s%" => \%files,
+ "chat" => sub {
+ $format='core%4$1d:%1s'."\n";
+ $width_buffer = 2000;
+ $rtrim=1;
+ $ltrim=0;
+ $omit_regex = qr/^\s*$/;
+ } ,
+ "multicolumn" => sub {
+ $format='%*3$s|';
+ $width_buffer = 19;
+ $formats{"-1"}='%*3$s'."\n";
+ $rtrim=0;
+ $ltrim=0;
+ $omit_regex = undef;
+ } ,
+ "classic" => sub {
+ $format='%s';
+ $width_buffer = 0;
+ $rtrim=0;
+ $ltrim=0;
+ $omit_regex = '^$(!:^)';
+ }
+ );
+ if (defined($omit_regex)) {
+ if ($omit_regex eq "") {
+ $omit_regex = qr/^\s*$/;
+ } else {
+ if ($omit_regex eq "none") {
+ $omit_regex=undef;
+ } else {
+ $omit_regex = qr/$omit_regex/s;
+ }
+ }
+ }
+ if ($help) {
+ pod2usage(1);
+ }
+
+}
+
+sub init {
+ for $k (keys( %formats)) {
+ for $r (split(/,/,$k)) {
+ if ( @l = split /-/,$r) {
+ for $i ($l[0]..$l[$#l]) {
+ $formats{$i}=$formats{$k};
+ }
+ }
+ }
+ }
+ for $i (0..($num_buffers-1)) {
+ $out_buffers[$i] = { BUFFER=> (" " x $width_buffer) , POS => 0 , FORMAT => ($formats{$i}||$formats{$i-$num_buffers}|| $format), FILEHANDLE => \*STDOUT, RTRIM => $rtrim , LTRIM => $ltrim, OMITREGEX => $omit_regex};
+ $in_buffers[$i] = { EXPECT => 'HIGH' , BYTE => 0, LINES=> 0, COUNT => 0};
+ }
+ $/=\1; #input line = 1 byte
+ my @time = localtime();
+ my $time = localtime();
+ my $year2 = $time[5] % 100;
+ for $k (keys(%files)) {
+ for $r (split(/,/,$k)) {
+ if (@l = split /-/,$r) {
+ for $i ($l[0]..$l[$#l]) {
+ $fileexpr = sprintf($files{$k},
+ $width_buffer,
+ $i,
+ @time,
+ $year2,
+ $time);
+ if ($fileexpr !~ /^(\+?>|\|)/){
+ $fileexpr = '>'. $fileexpr;
+ }
+
+ if (!exists $handles{$fileexpr}) {
+ open( $handles{$fileexpr} ,$fileexpr) or die "can't open $fileexpr\n";
+ }
+
+ $out_buffers[$i<0?$num_buffers-$i:$i]->{'FILEHANDLE'}=$handles{$fileexpr};
+ }
+ }
+ }
+ }
+}
+
+sub calc_id_buffer {
+ my ($in) = @_;
+ return ((~($in >> 4)) & 0xf);
+}
+
+sub input {
+ my ($in_char) = @_;
+ my $in = ord($in_char);
+ $idbuffer = calc_id_buffer($in);
+ if (($idbuffer < 0) || ($idbuffer >= $num_buffers)) {
+ printf STDERR 'Unknown input byte %1$x = %1$d = %1$b (%2$s) would give buffer id %3$d=%3$x=%3$b'."\n", $in, $in_char, $idbuffer;
+ } else {
+ $nibble = ($in & 0xf);
+ if ($in_buffers[$idbuffer]->{'EXPECT'} eq "HIGH") {
+ $in_buffers[$idbuffer]->{'BYTE'}=$nibble;
+ $in_buffers[$idbuffer]->{'EXPECT'}="LOW";
+ } else {
+ $in_buffers[$idbuffer]->{'BYTE'}=(($in_buffers[$idbuffer]->{'BYTE'} << 4)|($nibble));
+ $in_buffers[$idbuffer]->{'EXPECT'}="HIGH";
+ process_byte($idbuffer, $in_buffers[$idbuffer]->{'BYTE'});
+ $in_buffers[$idbuffer]->{'COUNT'}++;
+ if ($in_buffers[$idbuffer]->{'BYTE'} eq '\n') {
+ $in_buffers[$idbuffer]->{'LINES'}++;
+ }
+ }
+ }
+}
+
+sub process_byte {
+ my ($idbuffer,$byte) = @_;
+ my $char=chr($byte);
+ my $out=$out_buffers[$idbuffer];
+ if (($out->{'POS'} >= $width_buffer )
+ || ($char eq "\n")){
+ my $pos=$out->{'POS'};
+ flush_output();
+ if ($char eq "\n" && !$lf_does_cr) {
+ $out->{'POS'}=$pos;
+ }
+ }
+ if ($char eq "\r") {
+ $out->{'POS'}=0;
+ } elsif ($char eq "\t") {
+ my $chars = min(8,$width_buffer-$out->{'POS'});
+ substr($out->{'BUFFER'},$out->{'POS'},$chars)=" " x $chars;
+ $out->{'POS'} += $chars;
+ } elsif (($char ne "\n")||($width_buffer == 0)) {
+ substr($out->{'BUFFER'},$out->{'POS'},1)=$char;
+ $out->{'POS'} ++;
+ }
+
+}
+
+sub flush_output {
+ for $i (0..$num_buffers-1) {
+ my @time = localtime();
+ my $time = localtime();
+ my $year2 = $time[5] % 100;
+ my $buffer = $out_buffers[$i]->{'BUFFER'};
+
+
+ if ($out_buffers[$i]->{'LTRIM'} ) {
+ $buffer =~ s/^\s*//;
+ }
+
+ if ($out_buffers[$i]->{'RTRIM'} ) {
+ $buffer =~ s/\s*$//;
+ }
+
+ if ( (!defined($out_buffers[$i]->{'OMITREGEX'}))
+ || ($buffer !~ $out_buffers[$i]->{'OMITREGEX'}) ) {
+ printf { $out_buffers[$i]->{'FILEHANDLE'} }
+ $out_buffers[$i]->{'FORMAT'},
+
+ $buffer,
+ $out_buffers[$i]->{'POS'},
+ $width_buffer,
+ $i,
+ $line++,
+ $in_buffers[$i]->{'COUNT'},
+ $in_buffers[$i]->{'LINES'},
+ $in_buffers[$i]->{'EXPECT'},
+ $in_buffers[$i]->{'BYTE'},
+ @time,
+ $year2,
+ $time;
+ }
+ $out_buffers[$i]->{'BUFFER'} = (" " x $width_buffer) ;
+ $out_buffers[$i]->{'POS'} = 0;
+ };
+
+ # print("\n");
+
+ for $fhr (values(%handles)) {
+ $fhr->flush();
+ }
+ STDOUT->flush();
+ STDERR->flush();
+}
+
+sub end() {
+ for $i (0..$num_buffers-1) {
+ if ( $out_buffers[$i]->{'POS'} > 0 ) {
+ flush_output();
+ last;
+ }
+ }
+ for $fhr (values(%handles)) {
+ close $fhr;
+ }
+}
+
+parse_args();
+init();
+while(<>){
+ input($_);
+}
+end();
+
+
+__END__
+=head1 NAME
+
+demux_Debug.pl - demultiplexing coreboot output when CONFIF_DEBUG_MUXED to something readable
+
+=head1 SYNOPSYS
+
+sample [modes] [options]
+
+
+modes:
+ --chat
+ -m | --multicolumn
+ --classic
+
+options:
+
+ -b | --buffers Number of buffers.
+ -w | --width Maximum size of each buffer.
+ -f | format perl's sprintf format expression for outputting each buffer.
+ -n | nthformat format for specific buffers.
+ -o | --outputs filenames or perl open expressions for the especified buffers.
+ --rtrim remove all trailing whitespace before writing an output buffer
+ --nortrim ouput all trailing whitespace
+ --ltrim remove all leading whitespace before writing an output buffer
+ --noltrim output all leading whitespace
+ -e | --empty_skipped Don't write anything when an outputbuffer is empty after any trimming
+ --omit_regex "none" or a regex to suppress matching output
+ --help this text
+
+
+=head1 OPTIONS
+
+=over 8
+
+
+=item B<--chat>
+
+output messages from each core (until newline at any core)
+on a line, prefixed with the node/core number, a little ressembling
+an IRC session where each participant presses all participants
+enter key at once (ignores empty lines).
+
+Equivalent to (in bash)
+
+-f $'core%4$1d:%1s\n' -w 2000 --rtrim --noltrim --omit_regex '^\s*$'
+
+
+
+=item B<-m | --multicolumn>
+
+output mesages from each core side by side, roughly
+keeping simultaneus output in the same line.
+Reserves a part of a line for each core even if its
+empty and outputs a new line each time any core outputs
+a new line or fills its reserved space in its column.
+
+Equivalent to (in bash)
+
+-f '%*3$s|' -n -1=$'%*3$s'\n' -w 19 --nortrim --noltrim --omit_regex none
+
+
+
+=item B<--classic>
+
+Try to recover output similar to coreboot without CONFIG_DEBUG_MUXED
+Outputs one character after another as they come mixing output from all cores.
+Might be useful with --outputs
+
+Equivalent to
+
+-f '%s' -w 0 --nortrim --noltrim --omit_regex '^$(!:^)'
+
+
+=item B<-b | --buffers>
+
+Number of buffers. Usually it'd be the number
+of cores. The "protocol" maximum is 16. The
+bitwise negation of the high 4 bits in each byte
+indicate the destination buffer.
+
+
+=item B<-w | --width>
+
+Maximum size of each buffer. When this number of bytes
+have been demuxed without a newline, all buffers are
+flushed to output regardless. Width of each column in
+characters when in multicolumn mode. -w 0 also turns on
+output of newline characters as received. Other values
+eat newlines and cause an output buffer flush instead,
+which can output newline or not depending on formats
+
+
+=item B<-f | format>
+
+perl's sprintf format expression for outputting each buffer.
+See perldoc -f sprintf for details. Arguments are:
+
+=over 3
+
+=item 1
+
+output characters in this buffer (0-width characters)
+
+=item 2
+
+number of putput characters in this buffer (length of 1)
+
+=item 3
+
+width of buffer (-w value)
+
+=item 4
+
+buffer / core / node number (0-15)
+
+=item 5
+
+ouput line number (roughly, number of times output
+buffers have been flushed)
+
+=item 6
+
+count of charaacters seen so far for this buffer.
+
+=item 7
+
+count of lines seen so far in this buffer (newline count)
+
+=item 8
+
+state of this buffer HIGH if it is expecting a new char
+or LOW if it is expecting the final half of a character
+when it has seen the first half already in this buffer
+
+=item 9
+
+last byte (if 8=HIGH) or half byte (when 8=LOW)
+seen so far in this buffer
+
+=item 10
+
+second of start of this buffer flush. See perldoc -f localtime
+
+=item 11
+
+minute of start of this buffer flush. See perldoc -f localtime
+
+=item 12
+
+hour of start of this buffer flush. See perldoc -f localtime
+
+=item 13
+
+day in month of start of this buffer flush. See perldoc -f localtime
+
+=item 14
+
+month of start of this buffer flush. See perldoc -f localtime
+
+=item 15
+
+ year (4 digits) of start of this buffer flush. See perldoc -f localtime
+
+=item 16
+
+day in week of this buffer flush. See perldoc -f localtime
+
+=item 17
+
+day in year of start of this buffer flush. See perldoc -f localtime
+
+=item 18
+
+daylight saving time of start of this buffer flush. See perldoc -f localtime
+
+=item 19
+
+year (2 digits) of start of this buffer flush.
+
+=item 20
+
+date & time locale independent string of start of this buffer flush.
+see man 3 ctime or perldoc -f localtime (in scalar context)
+
+=back
+
+=item B<-n | nthformat>
+
+format for specific buffers. It must be followed by <n>=<format>
+where <n> is a comma separated list of ranges and each range
+if either a number or two numbers separated by dash. numbers must
+be between - (number of buffers) (see -b) and + (number of buffers -1).
+Negative numbers count from the last buffer backwards. <format>
+if a output buffer format (see -f) that will be applied exclusively
+to these buffers. This argument can be repeated as many times as needed
+to use different formats for different buffers.
+
+=item B<-o | --outputs>
+
+filenames or perl open expressions for the especified buffers.
+Must be followed by <n>=<output> where <n> is like for -n and
+<output> is either a filename like "output.log" or an output
+file open expression (like ">>previous.log" or "|grep ERROR").
+In fact <output> is a format for perl sprintf that will give
+the open expression, so it can be made to include core/node number
+or date etc. See perldoc -f sprintf and perldoc -f open
+The arguments passed to sprintf will be:
+
+=over 3
+
+=item 1
+
+width of buffer (-w value)
+
+=item 2
+
+number of buffer
+
+=item 3
+
+second of program invocation (roughly). See perldoc -f localtime
+
+=item 4
+
+minute of program invocation (roughly). See perldoc -f localtime
+
+=item 5
+
+hour of program invocation (roughly). See perldoc -f localtime
+
+
+=item 6
+
+day in month of program invocation (roughly). See perldoc -f localtime
+
+=item 7
+
+month of program invocation (roughly). See perldoc -f localtime
+
+=item 8
+
+year (4 digits) of program invocation (roughly). See perldoc -f localtime
+
+=item 9
+
+day in week of this buffer flush. See perldoc -f localtime
+
+=item 10
+
+day in year of program invocation (roughly). See perldoc -f localtime
+
+=item 11
+
+daylight saving time of program invocation (roughly). See perldoc -f localtime
+
+=item 12
+
+year (2 digits) of program invocation (roughly).
+
+=item 13
+
+date & time locale independent string of program invocation (roughly).
+
+=back
+
+=item B<--rtrim>
+
+remove all trailing whitespace before writing an output buffer
+
+=item B<--nortrim>
+
+output all trailing whitespace
+
+
+=item B<--ltrim>
+
+remove all leading whitespace before writing an output buffer
+
+=item B<--noltrim>
+
+output all leading whitespace
+
+=item B<-e | --empty_skipped>
+
+Don't write anything when an outputbuffer is empty after any trimming
+Equivalent to --omit-regex '^\s*$'
+
+=item B<--omit_regex>
+
+If followed by "none" prints all output. Otherwise it is a perl reglar
+expression that causes ouput to be suprresed when the contents of a buffer
+matches it.
+
+=item B<--lf_does_cr>
+
+When it gets a new line ('\n', ASCII 10, 0xA) it returns the buffer position
+to 0 even if there is no carriage return '\r'.
+
+=item B<--nolf_does_cr>
+
+When it gets a new line ('\n', ASCII 10, 0xA) it leaves the first characters
+of next line empty to keep the current position in the buffer, unless a carriage
+return changes teh current position back to 0
+
+=item B<--help>
+
+this text
+
+
+=back
+
+
+=head1 DESCRIPTION
+
+B<demux_debug.pl> will read standard input and write to standard ouput (or the
+specified files) after separating characters from each core. It expects it's
+inputs to be encoded as coreboot does with COFIG_DEBUG_MUXED, that is,
+each byte is split in two. The 4 high bits give the buffer id (core/node number,
+for instance) but it is bitwise negated to try to avoid problems with
+characters near 0 being interpreted as control characters. the low 4 bits are
+the high 4 bits of the transmitted char (in the first byte) or the 4 low bits
+of the transmitted char (in the second byte).
+This way concurrent output from different cores can become readable again.
+
+It keeps a buffer for each core (or each buffer id) and outputs all
+buffers to the specified file(s) or standard output when either it
+gets a newline ('\n', ASCII 10, 0xA) from any one of the cores or any
+of the buffers fills.
+
+This allows for different formats of output, like storing the output of each
+core in a different file, or printing in one column per core, or in separated
+lines with a prefix indicating the core "speaking".
+
+It outputs up to 8 white spaces (until end of buffer) instead of each tab ('\t').
+
+=head1 LICENSE
+
+GPL >= 2
Hello. This patch works for me but needs a small function calc_id_buffer for each board/cpu/whatever. I only made one for amd quadcore because it is what I have. My board does not get to ramstage, so it might not work there. It works for my serial console but should work for net or usb if I'm not mistaken. Also my tree is not up to date with svn head. Likely things could be placed elsewhere or done better, so it is mostly an idea. But I'm unlikely to have more time for it and it works for me so far, so I'm sending it so that it does not get lost and in case somebody likes the idea and wants to commit or do it better. Signed-off-by: Xavi Drudis Ferran <xdrudis@tinet.cat> My problem was with having serial output like [...] init node: 00 cores: 03 Start other core - nodeid: 00 cores: 03 start_other_cores() retuPPPrOOOnSSSeTTTd::: P000OxxxST3:330 00 0 x3 7cc cosootrrraeeexxxr:::t e d-- --a---p-- {{a{ p iAAAPPPcIIIiCCdCI:IID DD === 000231 NNNOOODDDEEEIIIDDD === 000000 CCCOOORRREEEIIIDDD === 000312}}} --------- [...] So I changed it because I like it better like this: [...] ���������������������������������������������������������������� [...] which is ugly, slow, etc. but I can pipe it to a little script which turns it back into almost the original ("boooh!!!" :( ) or writes it a little nicer ( :) ) [...] core0:init node: 00 cores: 03 core0:Start other core - nodeid: 00 cores: 03 core0:start_other_cores() returned core1:POST: core2:POST: core3:POST: core0:POST: core1: 0x30 core2: 0x30 core3: 0x30 core0:0 core0:x37 core2: c core3: c core0:started ap apicid: core1:corex: --- { APICID = 01 NODEID = 00 COREID = 01} --- core2:orex: --- { APICID = 02 NODEID = 00 COREID = 02} --- core3:orex: --- { APICID = 03 NODEID = 00 COREID = 03} --- [...] or even ( :) ) [...] init node: 00 cores: 03| | | | | | Start other core - nodei| | | d: 00 cores: 03 | | | start_other_cores() retu|P |P |P rned |OST: |OST: |OST: POST: | 0x30 | 0x30 | 0x30 0 | | | | | | x37 | | c | c started ap apicid: |corex: --- { APICID = 0|orex: --- { APICID = 02|orex: --- { APICID = 03 |1 NODEID = 00 COREID = 0| NODEID = 00 COREID = 0 | NODEID = 00 COREID = 03 |1} --- |2} --- |} --- | | | [...] or other options to if you fancy. I observed characters from each core got mixed but bits from each character not, so the superIO or someone probably did alreday synchronize bytes correctly. So I could split each byte in two, add a 4 bit buffer id and pretend each core has a different serial port and I'm multiplexing up to 16 channels for up to 16 cores into one physical serial port, by sending together the channel/buffer id with the character (4 bits buffer id, 4 bits half character). Then a small perl script demultiplexes each channel into a different buffer and formats output. If someone has more than 16 cores does she really want to see ouput from all at a time? Redefining the weak function calc_id_buffer you can choose to have some of them mix into the same buffer or just filter out the output for some of them. Hope the patch is simple enough, else I can explain more later. I must leave now.