Update some comments

TODO

@@ -173,9 +173,11 @@ o Task/Scheduler (sched/)
                NOTE: glibc behaves this way unless __thread is defined then,
                in that case, it behaves like NuttX (using TLS to save the
                thread local errno).
-  Status:      Closed.  The existing solution is better (although its
-               incompatibilities could show up in porting some code).
-  Priority:    Low
+  Status:      Closed.  The existing solution is better and compatible with
+               thread-aware GLIBC (although its incompatibilities could show
+               up in porting some code).  I will retain this issue for
+               referencei only.
+  Priority:    N/A
 
   Title:       SCALABILITY
   Description: Task control information is retained in simple lists.  This

drivers/sensors/bmp180.c

@@ -318,7 +318,7 @@ static int bmp180_checkid(FAR struct bmp180_dev_s *priv)
   devid = bmp180_getreg8(priv, BMP180_DEVID);
   sninfo("devid: 0x%02x\n", devid);
 
-  if (devid != (uint16_t) DEVID)
+  if (devid != (uint16_t)DEVID)
     {
       /* ID is not Correct */
 

sched/irq/irq_procfs.c

@@ -181,7 +181,23 @@ static int irq_callback(int irq, FAR struct irq_info_s *info,
 #endif
   leave_critical_section(flags);
 
-  /* Don't bother if count == 0 */
+  /* Don't bother if count == 0.
+   *
+   * REVISIT:  There is a logic problem with skipping if the count is zero.
+   * Normally this is a good thing because it makes the output concise.
+   * However, it can be a problem under certain conditions:
+   *
+   * It may take multiple passes through the IRQ table to enumerate the
+   * interrupts if the number of interrupts reported is large or if the size
+   * of the user buffer is small.  If a count is zero it will be skipped on
+   * the first time through but if it becomes non-zero on the second time
+   * through, the output will be corrupted.  Similarly if the count is non-
+   * zero the first time through and zero the second.
+   *
+   * A proper fix would require keep better track of where we left off
+   * between passes.  Current that position is remembered only by the
+   * byte offset into the pseudo-file, f_pos.
+   */
 
   if (copy.count == 0)
     {
@@ -210,7 +226,7 @@ static int irq_callback(int irq, FAR struct irq_info_s *info,
     }
   else
     {
-      uint64_t intcount = (uint64_t)intpart * elapsed / TICK_PER_SEC;
+      uint64_t intcount = ((uint64_t)intpart * elapsed) / TICK_PER_SEC;
       fracpart = (unsigned int)
         (((copy.count - intcount) * TICK_PER_SEC * 1000) / elapsed);
     }
@@ -225,7 +241,6 @@ static int irq_callback(int irq, FAR struct irq_info_s *info,
     {
       count = (unsigned long)copy.count;
     }
-
 #else
 #  error Missing logic
 #endif
@@ -270,10 +285,8 @@ static int irq_open(FAR struct file *filep, FAR const char *relpath,
 
   finfo("Open '%s'\n", relpath);
 
-  /* PROCFS is read-only.  Any attempt to open with any kind of write
-   * access is not permitted.
-   *
-   * REVISIT:  Write-able proc files could be quite useful.
+  /* This PROCFS file is read-only.  Any attempt to open with write access
+   * is not permitted.
    */
 
   if ((oflags & O_WRONLY) != 0 || (oflags & O_RDONLY) == 0)

sched/sched/sched_sporadic.c

@@ -140,8 +140,8 @@ static int sporadic_set_lowpriority(FAR struct tcb_s *tcb)
    *
    * In order to do this we would need to know the highest priority from
    * among all tasks waiting for the all semaphores held by the sporadic
-   * task.  Perhaps that information could be retained by the priority
-   * inheritance logic for use here?
+   * task.  That information could be retained by the priority inheritance
+   * logic of sem_holder.c for use here.
    */
 
   if (tcb->sched_priority > tcb->base_priority)