Xen devel - Nov 2006 - [PATCH] 0/3] Update xentrace to be compatable with PPC.

Signed-off-by: Tony Breeds <tony@bakeyournoodle.com>
---

 xen/common/trace.c         |    6 +++---
 xen/include/public/trace.h |    2 +-
 xen/include/xen/trace.h    |   14 +++++++-------
 3 files changed, 11 insertions(+), 11 deletions(-)

Index: xenppc-unstable.hg.working/xen/common/trace.c
==================================================================---
xenppc-unstable.hg.working.orig/xen/common/trace.c
+++ xenppc-unstable.hg.working/xen/common/trace.c
@@ -46,7 +46,7 @@ static int nr_recs;
 static int t_buf_highwater;
 
 /* Number of records lost due to per-CPU trace buffer being full. */
-static DEFINE_PER_CPU(unsigned long, lost_records);
+static DEFINE_PER_CPU(uint32_t, lost_records);
 
 /* a flag recording whether initialization has been done */
 /* or more properly, if the tbuf subsystem is enabled right now */
@@ -233,8 +233,8 @@ int tb_control(xen_sysctl_tbuf_op_t *tbc
  * failure, otherwise 0.  Failure occurs only if the trace buffers are not yet
  * initialised.
  */
-void trace(u32 event, unsigned long d1, unsigned long d2,
-           unsigned long d3, unsigned long d4, unsigned long d5)
+void trace(uint32_t event, uint32_t d1, uint32_t d2, uint32_t d3, uint32_t d4,
+                           uint32_t d5)
 {
     struct t_buf *buf;
     struct t_rec *rec;
Index: xenppc-unstable.hg.working/xen/include/public/trace.h
==================================================================---
xenppc-unstable.hg.working.orig/xen/include/public/trace.h
+++ xenppc-unstable.hg.working/xen/include/public/trace.h
@@ -76,7 +76,7 @@
 struct t_rec {
     uint64_t cycles;          /* cycle counter timestamp */
     uint32_t event;           /* event ID                */
-    unsigned long data[5];    /* event data items        */
+    uint32_t data[5];         /* event data items        */
 };
 
 /*
Index: xenppc-unstable.hg.working/xen/include/xen/trace.h
==================================================================---
xenppc-unstable.hg.working.orig/xen/include/xen/trace.h
+++ xenppc-unstable.hg.working/xen/include/xen/trace.h
@@ -33,19 +33,19 @@ void init_trace_bufs(void);
 /* used to retrieve the physical address of the trace buffers */
 int tb_control(struct xen_sysctl_tbuf_op *tbc);
 
-void trace(u32 event, unsigned long d1, unsigned long d2,
-           unsigned long d3, unsigned long d4, unsigned long d5);
+void trace(uint32_t event, uint32_t d1, uint32_t d2, uint32_t d3, uint32_t d4,
+                           uint32_t d5);
 
 /* Avoids troubling the caller with casting their arguments to a trace macro */
 #define trace_do_casts(e,d1,d2,d3,d4,d5) \
     do {                                 \
         if ( unlikely(tb_init_done) )    \
             trace(e,                     \
-                 (unsigned long)d1,      \
-                 (unsigned long)d2,      \
-                 (unsigned long)d3,      \
-                 (unsigned long)d4,      \
-                 (unsigned long)d5);     \
+                 (uint32_t)d1,           \
+                 (uint32_t)d2,           \
+                 (uint32_t)d3,           \
+                 (uint32_t)d4,           \
+                 (uint32_t)d5);          \
     } while ( 0 )
 
 /* Convenience macros for calling the trace function. */


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Hello All,
	This Patch series updates xentrace such that it will compile and work
correctly (assuming underlying hypervisor support) on all architectures.
Also allowing trace files captured on one architecture to be portable to other
systems with no changes to tools.

Essentially the patches do:
 1. Use unit32_t instead of unsigned long.
 2. Make the xentrace tool (xentrace copies the xentrace buffers from xen
    memory to disk) write data in network byte order.
 3. Various cleanups.
    This patch is optional.

These patches have been tested on ppc64 and x86_32, and datafiles transported
between said hosts.


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

- use err/errx/warn, instead of PERROR, perror and fprintf
 - Match () place ment consistent in this files
 - Use consistent tabs and whitespacing.

Signed-off-by: Tony Breeds <tony@bakeyournoodle.com>
---

 tools/xentrace/xentrace.c      |  137 ++++++++++++++---------------------------
 tools/xentrace/xentrace_format |   51 +++++++--------
 2 files changed, 73 insertions(+), 115 deletions(-)

Index: xenppc-unstable.hg.working/tools/xentrace/xentrace.c
==================================================================---
xenppc-unstable.hg.working.orig/tools/xentrace/xentrace.c
+++ xenppc-unstable.hg.working/tools/xentrace/xentrace.c
@@ -22,6 +22,7 @@
 #include <signal.h>
 #include <inttypes.h>
 #include <string.h>
+#include <err.h>
 
 #include <xen/xen.h>
 #include <xen/trace.h>
@@ -42,16 +43,6 @@
 
 #define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
 
-#define PERROR(_m, _a...)                                       \
-do {                                                            \
-    int __saved_errno = errno;                                  \
-    fprintf(stderr, "ERROR: " _m " (%d = %s)\n" , ## _a ,  
\
-            __saved_errno, strerror(__saved_errno));            \
-    errno = __saved_errno;                                      \
-} while (0)
-
-extern FILE *stderr;
-
 /***** Compile time configuration of defaults ********************************/
 
 /* when we''ve got more records than this waiting, we log it to the
output */
@@ -88,7 +79,7 @@ void close_handler(int signal)
 struct timespec millis_to_timespec(unsigned long millis)
 {
     struct timespec spec;
-    
+
     spec.tv_sec = millis / 1000;
     spec.tv_nsec = (millis % 1000) * 1000;
 
@@ -128,10 +119,7 @@ void write_rec(uint32_t cpu, struct t_re
     }
 
     if ( written != 8 )
-    {
-        PERROR("Failed to write trace record");
-        exit(EXIT_FAILURE);
-    }
+        err(EXIT_FAILURE, "Failed to write trace record");
 }
 
 static void get_tbufs(unsigned long *mfn, unsigned long *size)
@@ -139,21 +127,16 @@ static void get_tbufs(unsigned long *mfn
     int xc_handle = xc_interface_open();
     int ret;
 
-    if ( xc_handle < 0 ) 
-    {
-        exit(EXIT_FAILURE);
-    }
+    if ( xc_handle < 0 )
+        errx(EXIT_FAILURE, "Unable to open the xc interface");
 
-    if(!opts.tbuf_size)
-      opts.tbuf_size = DEFAULT_TBUF_SIZE;
+    if ( !opts.tbuf_size )
+        opts.tbuf_size = DEFAULT_TBUF_SIZE;
 
     ret = xc_tbuf_enable(xc_handle, opts.tbuf_size, mfn, size);
 
     if ( ret != 0 )
-    {
-        perror("Couldn''t enable trace buffers");
-        exit(1);
-    }
+        err(EXIT_FAILURE, "Couldn''t enable trace buffers");
 
     xc_interface_close(xc_handle);
 }
@@ -174,10 +157,8 @@ struct t_buf *map_tbufs(unsigned long tb
 
     xc_handle = xc_interface_open();
 
-    if ( xc_handle < 0 ) 
-    {
-        exit(EXIT_FAILURE);
-    }
+    if ( xc_handle < 0 )
+        errx(EXIT_FAILURE, "Unable to open the xc interface");
 
     /* On PPC (At least) the DOMID arg is ignored in dom0 */
     tbufs_mapped = xc_map_foreign_range(xc_handle, DOMID_XEN,
@@ -186,18 +167,15 @@ struct t_buf *map_tbufs(unsigned long tb
 
     xc_interface_close(xc_handle);
 
-    if ( tbufs_mapped == 0 ) 
-    {
-        PERROR("Failed to mmap trace buffers");
-        exit(EXIT_FAILURE);
-    }
+    if ( tbufs_mapped == 0 )
+        err(EXIT_FAILURE, "Failed to mmap trace buffers");
 
     return tbufs_mapped;
 }
 
 /**
  * set_mask - set the cpu/event mask in HV
- * @mask:           the new mask 
+ * @mask:           the new mask
  * @type:           the new mask type,0-event mask, 1-cpu mask
  *
  */
@@ -206,21 +184,19 @@ void set_mask(uint32_t mask, int type)
     int ret = 0;
     int xc_handle = xc_interface_open(); /* for accessing control interface */
 
-    if (type == 1) {
+    if ( type == 1 ) {
         ret = xc_tbuf_set_cpu_mask(xc_handle, mask);
-        fprintf(stderr, "change cpumask to 0x%x\n", mask);
-    } else if (type == 0) {
+        warnx("change cpumask to 0x%x\n", mask);
+    } else if ( type == 0 ) {
         ret = xc_tbuf_set_evt_mask(xc_handle, mask);
-        fprintf(stderr, "change evtmask to 0x%x\n", mask);
+        warnx("change evtmask to 0x%x\n", mask);
     }
 
     xc_interface_close(xc_handle);
 
     if ( ret != 0 )
-    {
-        PERROR("Failure to get trace buffer pointer from Xen and set the
new mask");
-        exit(EXIT_FAILURE);
-    }
+        err(EXIT_FAILURE, "Failure to get trace buffer pointer from "
+	                  "Xen and set the new mask");
 }
 
 /**
@@ -240,11 +216,8 @@ struct t_buf **init_bufs_ptrs(void *bufs
 
     user_ptrs = (struct t_buf **)calloc(num, sizeof(struct t_buf *));
     if ( user_ptrs == NULL )
-    {
-        PERROR( "Failed to allocate memory for buffer pointers\n");
-        exit(EXIT_FAILURE);
-    }
-    
+        err(EXIT_FAILURE, "Failed to allocate memory for buffer
pointers");
+
     /* initialise pointers to the trace buffers - given the size of a trace
      * buffer and the value of bufs_maped, we can easily calculate these */
     for ( i = 0; i<num; i++ )
@@ -269,13 +242,10 @@ struct t_rec **init_rec_ptrs(struct t_bu
 {
     int i;
     struct t_rec **data;
-    
+
     data = calloc(num, sizeof(struct t_rec *));
     if ( data == NULL )
-    {
-        PERROR("Failed to allocate memory for data pointers\n");
-        exit(EXIT_FAILURE);
-    }
+        err(EXIT_FAILURE, "Failed to allocate memory for data
pointers");
 
     for ( i = 0; i < num; i++ )
         data[i] = (struct t_rec *)(meta[i] + 1);
@@ -291,14 +261,11 @@ uint32_t get_num_cpus(void)
     xc_physinfo_t physinfo;
     int xc_handle = xc_interface_open();
     int ret;
-    
+
     ret = xc_physinfo(xc_handle, &physinfo);
-    
+
     if ( ret != 0 )
-    {
-        PERROR("Failure to get logical CPU count from Xen");
-        exit(EXIT_FAILURE);
-    }
+        errx(EXIT_FAILURE, "Failure to get logical CPU count from
Xen");
 
     xc_interface_close(xc_handle);
 
@@ -377,17 +344,17 @@ int parse_evtmask(char *arg, struct argp
     char *inval;
 
     /* search filtering class */
-    if (strcmp(arg, "gen") == 0){ 
+    if ( strcmp(arg, "gen") == 0 )
         setup->evt_mask |= TRC_GEN;
-    } else if(strcmp(arg, "sched") == 0){ 
+    else if ( strcmp(arg, "sched") == 0 )
         setup->evt_mask |= TRC_SCHED;
-    } else if(strcmp(arg, "dom0op") == 0){ 
+    else if ( strcmp(arg, "dom0op") == 0 )
         setup->evt_mask |= TRC_DOM0OP;
-    } else if(strcmp(arg, "vmx") == 0){ 
+    else if ( strcmp(arg, "vmx") == 0 )
         setup->evt_mask |= TRC_VMX;
-    } else if(strcmp(arg, "all") == 0){ 
+    else if ( strcmp(arg, "all") == 0 )
         setup->evt_mask |= TRC_ALL;
-    } else {
+    else {
         setup->evt_mask = strtol(arg, &inval, 0);
         if ( inval == arg )
             argp_usage(state);
@@ -430,13 +397,13 @@ error_t cmd_parser(int key, char *arg, s
             argp_usage(state);
     }
     break;
-    
+
     case ''e'': /* set new event mask for filtering*/
     {
         parse_evtmask(arg, state);
     }
     break;
-    
+
     case ''S'': /* set tbuf size (given in pages) */
     {
         char *inval;
@@ -454,7 +421,7 @@ error_t cmd_parser(int key, char *arg, s
             argp_usage(state);
     }
     break;
-        
+
     default:
         return ARGP_ERR_UNKNOWN;
     }
@@ -473,16 +440,16 @@ const struct argp_option cmd_opts[]        "(default
" xstr(NEW_DATA_THRESH) ")." },
 
     { .name = "poll-sleep", .key=''s'',
.arg="p",
-      .doc = 
+      .doc        "Set sleep time, p, in milliseconds between polling the
trace buffer "
       "for new data (default " xstr(POLL_SLEEP_MILLIS) ")."
},
 
     { .name = "cpu-mask", .key=''c'',
.arg="c",
-      .doc = 
+      .doc        "set cpu-mask " },
 
     { .name = "evt-mask", .key=''e'',
.arg="e",
-      .doc = 
+      .doc        "set evt-mask " },
 
     { .name = "trace-buf-size", .key=''S'',
.arg="N",
@@ -504,7 +471,7 @@ const struct argp parser_def      "\v"
     "This tool is used to capture trace buffer data from Xen.  The data is
"
     "output in a binary format, in the following order:\n\n"
-    "  CPU(uint) TSC(uint64_t) EVENT(uint32_t) D1 D2 D3 D4 D5 "
+    "  CPU(uint32_t) TSC(uint64_t) EVENT(uint32_t) D1 D2 D3 D4 D5 "
     "(all uint32_t)\n\n"
     "The output should be parsed using the tool xentrace_format, which can
"
     "produce human-readable output in ASCII format."
@@ -513,8 +480,8 @@ const struct argp parser_def  
 const char *argp_program_version     = "xentrace v1.1";
 const char *argp_program_bug_address =
"<mark.a.williamson@intel.com>";
-        
-    
+
+
 int main(int argc, char **argv)
 {
     int outfd = 1, ret;
@@ -529,31 +496,23 @@ int main(int argc, char **argv)
 
     argp_parse(&parser_def, argc, argv, 0, 0, &opts);
 
-    if (opts.evt_mask != 0) { 
+    if ( opts.evt_mask != 0 )
         set_mask(opts.evt_mask, 0);
-    }
 
-    if (opts.cpu_mask != 0) {
+    if ( opts.cpu_mask != 0 )
         set_mask(opts.cpu_mask, 1);
-    }
 
     if ( opts.outfile )
         outfd = open(opts.outfile, O_WRONLY | O_CREAT | O_LARGEFILE, 0644);
 
-    if(outfd < 0)
-    {
-        perror("Could not open output file");
-        exit(EXIT_FAILURE);
-    }        
+    if ( outfd < 0 )
+        err(EXIT_FAILURE, "Could not open output file");
 
-    if(isatty(outfd))
-    {
-        fprintf(stderr, "Cannot output to a TTY, specify a log
file.\n");
-        exit(EXIT_FAILURE);
-    }
+    if ( isatty(outfd) )
+        errx(EXIT_FAILURE, "Cannot output to a TTY, specify a log
file.\n");
 
     logfile = fdopen(outfd, "w");
-    
+
     /* ensure that if we get a signal, we''ll do cleanup, then exit */
     act.sa_handler = close_handler;
     act.sa_flags = 0;
Index: xenppc-unstable.hg.working/tools/xentrace/xentrace_format
==================================================================---
xenppc-unstable.hg.working.orig/tools/xentrace/xentrace_format
+++ xenppc-unstable.hg.working/tools/xentrace/xentrace_format
@@ -24,7 +24,7 @@ def usage():
           Which correspond to the CPU number, event ID, timestamp counter and
           the 5 data fields from the trace record.  There should be one such
           rule for each type of event.
-          
+
           Depending on your system and the volume of trace buffer data,
           this script may not be able to keep up with the output of xentrace
           if it is piped directly.  In these circumstances you should have
@@ -34,7 +34,7 @@ def usage():
 
 def read_defs(defs_file):
     defs = {}
-    
+
     fd = open(defs_file)
 
     reg = re.compile(''(\S+)\s+(\S.*)'')
@@ -43,14 +43,14 @@ def read_defs(defs_file):
         line = fd.readline()
         if not line:
             break
-	
-	if line[0] == ''#'' or line[0] == ''\n'':
-	    continue
-        
+
+        if line[0] == ''#'' or line[0] ==
''\n'':
+            continue
+
         m = reg.match(line)
 
         if not m: print >> sys.stderr, "Bad format file" ;
sys.exit(1)
-        
+
         defs[str(eval(m.group(1)))] = m.group(2)
 
     return defs
@@ -70,7 +70,7 @@ try:
     opts, arg = getopt.getopt(sys.argv[1:], "c:" )
 
     for opt in opts:
-	if opt[0] == ''-c'' : mhz = int(opt[1])
+        if opt[0] == ''-c'' : mhz = int(opt[1])
 
 except getopt.GetoptError:
     usage()
@@ -96,7 +96,7 @@ i=0
 
 while not interrupted:
     try:
-	i=i+1
+        i=i+1
         line = sys.stdin.read(struct.calcsize(CPUREC))
         if not line:
             break
@@ -108,19 +108,20 @@ while not interrupted:
 
         (tsc, event, d1, d2, d3, d4, d5) = struct.unpack(TRCREC, line)
 
-	#tsc = (tscH<<32) | tscL
+        #tsc = (tscH<<32) | tscL
 
-	#print i, tsc
+        #print i, tsc
 
         if cpu >= len(last_tsc):
             last_tsc += [0] * (cpu - len(last_tsc) + 1)
-	elif tsc < last_tsc[cpu]:
-	    print "TSC stepped backward cpu %d !  %d %d" %
(cpu,tsc,last_tsc[cpu])
+        elif tsc < last_tsc[cpu]:
+            print "TSC stepped backward cpu %d !  %d %d" % (cpu, tsc,
+                                                            last_tsc[cpu])
 
-	last_tsc[cpu] = tsc
+        last_tsc[cpu] = tsc
 
-	if mhz:
-	    tsc = tsc / (mhz*1000000.0)
+        if mhz:
+            tsc = tsc / (mhz*1000000.0)
 
         args = {''cpu''   : cpu,
                 ''tsc''   : tsc,
@@ -131,15 +132,13 @@ while not interrupted:
                 ''4''     : d4,
                 ''5''     : d5    }
 
-	try:
-
-	    if defs.has_key(str(event)): 
-		print defs[str(event)] % args
-	    else: 
-		if defs.has_key(str(0)): print defs[str(0)] % args
-	except TypeError:
-	    print defs[str(event)]
-	    print args
-	    
+        try:
+            if defs.has_key(str(event)):
+                print defs[str(event)] % args
+            else:
+                if defs.has_key(str(0)): print defs[str(0)] % args
+        except TypeError:
+            print defs[str(event)]
+            print args
 
     except IOError, struct.error: sys.exit()


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Use network byte order when writing data to disk, making the data portable to
other architectures.

Signed-off-by: Tony Breeds <tony@bakeyournoodle.com>
---

 tools/xentrace/xentrace.c      |   47 ++++++++++++++++++++++++++++++++++-------
 tools/xentrace/xentrace_format |    7 +++---
 2 files changed, 44 insertions(+), 10 deletions(-)

Index: xenppc-unstable.hg.working/tools/xentrace/xentrace.c
==================================================================---
xenppc-unstable.hg.working.orig/tools/xentrace/xentrace.c
+++ xenppc-unstable.hg.working/tools/xentrace/xentrace.c
@@ -28,6 +28,20 @@
 
 #include <xenctrl.h>
 
+#include <arpa/inet.h> /* hton*(), ntoh*() */
+#include <endian.h>
+
+/* There is no 64-bit htonll, so create one */
+#if __BYTE_ORDER == __LITTLE_ENDIAN
+#define        htonll(x)   ( (((uint64_t)htonl(x)) << 32) + htonl(x
>> 32) )
+#define        ntohll(x)   ( (((uint64_t)ntohl(x)) << 32) + ntohl(x
>> 32) )
+#else
+#define        htonll(x)   (x)
+#define        ntohll(x)   (x)
+#endif
+
+#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
+
 #define PERROR(_m, _a...)                                       \
 do {                                                            \
     int __saved_errno = errno;                                  \
@@ -90,12 +104,30 @@ struct timespec millis_to_timespec(unsig
  * Outputs the trace record to a filestream, prepending the CPU ID of the
  * source trace buffer.
  */
-void write_rec(unsigned int cpu, struct t_rec *rec, FILE *out)
+void write_rec(uint32_t cpu, struct t_rec *rec, FILE *out)
 {
     size_t written = 0;
-    written += fwrite(&cpu, sizeof(cpu), 1, out);
-    written += fwrite(rec, sizeof(*rec), 1, out);
-    if ( written != 2 )
+    int i;
+    /* Place network byte order representation in temp vars, rather than
+     * write back into kernel/xen memory */
+    uint64_t tmp64;
+    uint32_t tmp32;
+
+    tmp32 = htonl(cpu);
+    written += fwrite(&tmp32, sizeof(tmp32), 1, out);
+
+    tmp64 = htonll(rec->cycles);
+    written += fwrite(&tmp64, sizeof(tmp64), 1, out);
+
+    tmp32 = htonl(rec->event);
+    written += fwrite(&tmp32, sizeof(tmp32), 1, out);
+
+    for ( i=0; i<ARRAY_SIZE(rec->data); i++ ) {
+        tmp32 = htonl(rec->data[i]);
+        written += fwrite(&tmp32, sizeof(tmp32), 1, out);
+    }
+
+    if ( written != 8 )
     {
         PERROR("Failed to write trace record");
         exit(EXIT_FAILURE);
@@ -147,6 +179,7 @@ struct t_buf *map_tbufs(unsigned long tb
         exit(EXIT_FAILURE);
     }
 
+    /* On PPC (At least) the DOMID arg is ignored in dom0 */
     tbufs_mapped = xc_map_foreign_range(xc_handle, DOMID_XEN,
                                         size * num, PROT_READ | PROT_WRITE,
                                         tbufs_mfn);
@@ -253,7 +286,7 @@ struct t_rec **init_rec_ptrs(struct t_bu
 /**
  * get_num_cpus - get the number of logical CPUs
  */
-unsigned int get_num_cpus(void)
+uint32_t get_num_cpus(void)
 {
     xc_physinfo_t physinfo;
     int xc_handle = xc_interface_open();
@@ -282,14 +315,14 @@ unsigned int get_num_cpus(void)
  */
 int monitor_tbufs(FILE *logfile)
 {
-    int i;
+    uint32_t i;
 
     void *tbufs_mapped;          /* pointer to where the tbufs are mapped    */
     struct t_buf **meta;         /* pointers to the trace buffer metadata    */
     struct t_rec **data;         /* pointers to the trace buffer data areas
                                   * where they are mapped into user space.   */
     unsigned long tbufs_mfn;     /* mfn of the tbufs                         */
-    unsigned int  num;           /* number of trace buffers / logical CPUS   */
+    uint32_t num;                /* number of trace buffers / logical CPUS   */
     unsigned long size;          /* size of a single trace buffer            */
 
     int size_in_recs;
Index: xenppc-unstable.hg.working/tools/xentrace/xentrace_format
==================================================================---
xenppc-unstable.hg.working.orig/tools/xentrace/xentrace_format
+++ xenppc-unstable.hg.working/tools/xentrace/xentrace_format
@@ -84,10 +84,11 @@ interrupted = 0
 defs = read_defs(arg[0])
 
 # structure of trace record + prepended CPU id (as output by xentrace):
-# CPU(I) TSC(Q) EVENT(L) D1(L) D2(L) D3(L) D4(L) D5(L)
+# CPU(L) TSC(Q) EVENT(L) D1(L) D2(L) D3(L) D4(L) D5(L)
 # read CPU id separately to avoid structure packing problems on 64-bit arch.
-CPUREC = "I"
-TRCREC = "QLLLLLL"
+# Force network byte order.
+CPUREC = "!L"
+TRCREC = "!QLLLLLL"
 
 last_tsc = [0]
 


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

>Hello All,
>	This Patch series updates xentrace such that it will 
>compile and work
>correctly (assuming underlying hypervisor support) on all 
>architectures.
>Also allowing trace files captured on one architecture to be 
>portable to other
>systems with no changes to tools.
>
>Essentially the patches do:
> 1. Use unit32_t instead of unsigned long.
I''d like to keep the trace value unsigned long, and add a parameter to
user level tools the width in raw data file.
-Xin

> 2. Make the xentrace tool (xentrace copies the xentrace 
>buffers from xen
>    memory to disk) write data in network byte order.
> 3. Various cleanups.
>    This patch is optional.
>
>These patches have been tested on ppc64 and x86_32, and 
>datafiles transported
>between said hosts.
>
>
>_______________________________________________
>Xen-devel mailing list
>Xen-devel@lists.xensource.com
>http://lists.xensource.com/xen-devel
>
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

On Wed, Nov 29, 2006 at 12:47:07PM +0800, Li, Xin B wrote:
 
> I''d like to keep the trace value unsigned long, and add a
parameter to
> user level tools the width in raw data file.
For backwards compatability? (so you can read your current files with
new tools?)  or do you need to write 64bit data to disk?

Yours Tony

   linux.conf.au       http://linux.conf.au/ || http://lca2007.linux.org.au/
   Jan 15-20 2007      The Australian Linux Technical Conference!


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

>> I''d like to keep the trace value unsigned long, and add a 
>parameter to
>> user level tools the width in raw data file.
>
>For backwards compatability? (so you can read your current files with
>new tools?)  or do you need to write 64bit data to disk?
>
Write 64bit data to disk.
-Xin

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

On 29/11/06 6:30 am, "Li, Xin B" <xin.b.li@intel.com> wrote:
>> For backwards compatability? (so you can read your current files with
>> new tools?)  or do you need to write 64bit data to disk?
>> 
> 
> Write 64bit data to disk.
Let''s write native-endian long values into the memory buffer, then have
the
tool turn that into network-endian 64-bit values on disc.

 -- Keir



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

On Wed, Nov 29, 2006 at 07:56:17AM +0000, Keir Fraser wrote:
> Let''s write native-endian long values into the memory buffer, then
have the
> tool turn that into network-endian 64-bit values on disc.
Sure.  I''ll respin the patches tomorrow.

Yours Tony

   linux.conf.au       http://linux.conf.au/ || http://lca2007.linux.org.au/
   Jan 15-20 2007      The Australian Linux Technical Conference!


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

On Nov 29, 2006, at 1:30 AM, Li, Xin B wrote:
>>> I''d like to keep the trace value unsigned long, and add a
>> parameter to
>>> user level tools the width in raw data file.
>>
>> For backwards compatability? (so you can read your current files with
>> new tools?)  or do you need to write 64bit data to disk?
>>
>
> Write 64bit data to disk.
Is there a specific trace stanza you require this for that 32bits  
would not work?

-JX



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

On 29/11/06 12:20, "Jimi Xenidis" <jimix@watson.ibm.com> wrote:
>> Write 64bit data to disk.
> 
> Is there a specific trace stanza you require this for that 32bits
> would not work?
I expect guest addresses are logged on some HVM/VMX trace paths. Also
various interesting VMCS fields are longs or 64 bits.

 -- Keir


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel