Not sure, but this may want to go into 4.11? Is the bug in that as well? Peter Edwards wrote: >The following reply was made to PR kern/66248; it has been noted by GNATS. > >From: Peter Edwards <peadar_at_freebsd.org> >To: freebsd-gnats-submit_at_FreeBSD.org, bremner_at_unb.ca >Cc: julian_at_elischer.org, imp_at_bsdimp.org >Subject: Re: kern/66248: [panic] bootloader is confused by booting from USB > flash, trashes main drives boot block >Date: Wed, 05 Jan 2005 20:53:02 +0000 > > This is a multi-part message in MIME format. > --------------070200040205060401080901 > Content-Type: text/plain; charset=us-ascii; format=flowed > Content-Transfer-Encoding: 7bit > > The following patch, applied before the one above, shrinks boot0 enough > to squeeze the previous patch in without undeffing the BEL stuff. > > FWIW: The patch to fix the bug looks good to me. > I someone gives a go-ahead, I'll commit a combined patch: I certainly > don't want to make a commit to boot0 and have half the FreeBSD community > left without a bootable machine, unless I can pass the book :-) > > > --------------070200040205060401080901 > Content-Type: text/plain; > name="boot0.smaller.patch" > Content-Transfer-Encoding: 7bit > Content-Disposition: inline; > filename="boot0.smaller.patch" > > Index: boot0.S > =================================================================== > RCS file: /usr/cvs/FreeBSD-CVS/src/sys/boot/i386/boot0/boot0.S,v > retrieving revision 1.12 > diff -u -w -r1.12 boot0.S > --- boot0.S 24 Nov 2004 15:39:04 -0000 1.12 > +++ boot0.S 5 Jan 2005 20:44:32 -0000 > _at__at_ -29,7 +29,7 _at__at_ > .set PRT_OFF,0x1be # Partition table > > .set TBL0SZ,0x3 # Table 0 size > - .set TBL1SZ,0xb # Table 1 size > + .set TBL1SZ,0xc # Table 1 size > > .set MAGIC,0xaa55 # Magic: bootable > .set B0MAGIC,0xbb66 # Identification > _at__at_ -148,13 +148,8 _at__at_ > movb $TBL1SZ,%cl # Number of entries > repne # Known > scasb # type? > - jne main.4 # No > -/* > - * If it matches get the matching element in the next array. If it doesn't, > - * we are already pointing at its first element which points to a "?". > - */ > - addw $TBL1SZ,%di # Adjust > -main.4: movb (%di),%cl # Partition > + addw $TBL1SZ - 1, %di # Adjust > + movb (%di),%cl # Partition > addw %cx,%di # description > callw putx # Display it > main.5: incw %dx # Next item > _at__at_ -314,8 +309,7 _at__at_ > cmpw $MAGIC,0x1fe(%bx) # Bootable? > jne main.10 # No > pushw %si # Save > - movw $crlf,%si # Leave some > - callw puts # space > + callw putn # Leave some space > popw %si # Restore > jmp *%bx # Invoke bootstrap > > _at__at_ -408,14 +402,14 _at__at_ > .byte 0x0, 0x5, 0xf > /* > * These values indicate bootable types we know the names of. > + * The value of the last entry is unimportant: it indicates "everything else" > */ > .byte 0x1, 0x4, 0x6, 0xb, 0xc, 0xe, 0x83 > - .byte 0x9f, 0xa5, 0xa6, 0xa9 > + .byte 0x9f, 0xa5, 0xa6, 0xa9, 0xff > /* > * These are offsets that match the known names above and point to the strings > * that will be printed. > */ > - .byte os_misc-. # Unknown > .byte os_dos-. # DOS > .byte os_dos-. # DOS > .byte os_dos-. # DOS > _at__at_ -427,6 +421,7 _at__at_ > .byte os_freebsd-. # FreeBSD > .byte os_bsd-. # OpenBSD > .byte os_bsd-. # NetBSD > + .byte os_misc-. # Unknown > /* > * And here are the strings themselves. 0x80 or'd into a byte indicates > * the end of the string. (not so great for Russians but...) > > --------------070200040205060401080901-- >_______________________________________________ >freebsd-usb_at_freebsd.org mailing list >http://lists.freebsd.org/mailman/listinfo/freebsd-usb >To unsubscribe, send any mail to "freebsd-usb-unsubscribe_at_freebsd.org" > >Received on Thu Jan 06 2005 - 00:32:45 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:25 UTC