Author: kevans
Date: Thu Mar 29 00:55:11 2018
New Revision: 331718
URL: https://svnweb.freebsd.org/changeset/base/331718

Log:
  stand: Add workaround for HP BIOS issues
  
  hrs@ and kuriyama@ have found that on some HP BIOS, a system will fail to
  boot immediately after installation with the claim that it can't work out
  which disk they are booting from.
  
  They tracked it down to a buffer overrun, and found that it could be
  alleviated by doing a dummy read before-hand.
  
  Submitted by: kuriyama
  MFC after:    1 week
  Differential Revision:        https://reviews.freebsd.org/D14341

Modified:
  head/stand/i386/libi386/biosdisk.c

Modified: head/stand/i386/libi386/biosdisk.c
==============================================================================
--- head/stand/i386/libi386/biosdisk.c  Thu Mar 29 00:42:44 2018        
(r331717)
+++ head/stand/i386/libi386/biosdisk.c  Thu Mar 29 00:55:11 2018        
(r331718)
@@ -122,6 +122,8 @@ static int nbdinfo = 0;
 
 #define        BD(dev)         (bdinfo[(dev)->dd.d_unit])
 
+static void bd_io_workaround(struct disk_devdesc *dev);
+
 static int bd_read(struct disk_devdesc *dev, daddr_t dblk, int blks,
     caddr_t dest);
 static int bd_write(struct disk_devdesc *dev, daddr_t dblk, int blks,
@@ -724,6 +726,15 @@ bd_chs_io(struct disk_devdesc *dev, daddr_t dblk, int 
     return (0);
 }
 
+static void
+bd_io_workaround(struct disk_devdesc *dev)
+{
+       uint8_t buf[8 * 1024];
+
+       bd_edd_io(dev, 0xffffffff, 1, (caddr_t)buf, 0);
+}
+
+
 static int
 bd_io(struct disk_devdesc *dev, daddr_t dblk, int blks, caddr_t dest, int 
write)
 {
@@ -736,6 +747,17 @@ bd_io(struct disk_devdesc *dev, daddr_t dblk, int blks
 
     resid = blks;
     p = dest;
+
+    /*
+     * Workaround for a problem with some HP ProLiant BIOS failing to work out
+     * the boot disk after installation. hrs and kuriyama discovered this
+     * problem with an HP ProLiant DL320e Gen 8 with a 3TB HDD, and discovered
+     * that an int13h call seems to cause a buffer overrun in the bios. The
+     * problem is alleviated by doing an extra read before the buggy read. It
+     * is not immediately known whether other models are similarly affected.
+     */
+    if (dblk >= 0x100000000)
+       bd_io_workaround(dev);
 
     /* Decide whether we have to bounce */
     if (VTOP(dest) >> 20 != 0 || (BD(dev).bd_unit < 0x80 &&
_______________________________________________
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"

Reply via email to