Re: install/56132: build.sh target install-image fails in update-mode

[Izumi Tsutsui <tsutsui%ceres.dti.ne.jp@localhost>]

The following reply was made to PR install/56132; it has been noted by GNATS.

From: Izumi Tsutsui <tsutsui%ceres.dti.ne.jp@localhost>
To: gnats-bugs%netbsd.org@localhost
Cc: tsutsui%ceres.dti.ne.jp@localhost
Subject: Re: install/56132: build.sh target install-image fails in update-mode
Date: Tue, 1 Jun 2021 01:22:11 +0900

 I wrote:
 
 > Date: Fri, 30 Apr 2021 01:20:14 +0900
 > 
 >  I wrote:
 >  
 >  > ${WORKFSTAB} needs to depend on ${WORKMBR} image file to get
 >  > a UUID created by ${TOOL_GPT} with ${GPT_TIMESTAMP}?  Umm.
 >  > (note GPT/UEFI part was not implemented by me)
 >  > 
 >  > I wonder if we can calculate the UUID from ${GPT_TIMESTAMP}
 >  > without an actual image, but anyway we need some dummy targets
 >  > that checks and removes ${WORKMBR} if its size is different
 >  > from ${IMAGESECTORS}?
 >  
 >  There are two options:
 >  
 >  1) always build ${WORKMBR} by .PHONY
 
  [snip]
 
 >  2) check a size of ${WORKMBR} before building targets
 >  
 >  Not sure if .BEGIN is correct here..
 >  
 >  ---
 >  Index: distrib/common/bootimage/Makefile.bootimage
 >  ===================================================================
 >  RCS file: /cvsroot/src/distrib/common/bootimage/Makefile.bootimage,v
 >  retrieving revision 1.26
 >  diff -u -p -d -r1.26 Makefile.bootimage
 >  --- distrib/common/bootimage/Makefile.bootimage	5 Aug 2020 01:35:18 -0000	1.26
 >  +++ distrib/common/bootimage/Makefile.bootimage	29 Apr 2021 16:09:58 -0000
 >  @@ -294,6 +294,15 @@ WORKFATDIR?=	work.fatdir
 >   
 >   .if ${USE_GPT} != "no"
 >   ${WORKFSTAB}: ${WORKMBR}
 >  +
 >  +.BEGIN:
 >  +	@if [ -f ${WORKMBR} ]; then			 		\
 >  +		st_size=`${TOOL_STAT} -f %z ${WORKMBR}`;		\
 >  +		if [ "$$st_size" -ne "$$((${IMAGESECTORS} * 512))" ]; then \
 >  +			echo Removing stale ${WORKMBR} ...;		\
 >  +			rm -f ${WORKMBR};				\
 >  +		fi							\
 >  +	fi
 >   .endif
 >   
 >   ${WORKFSTAB}:
 >  ---
 >  
 >  > Needs make(1) guru..
 >  
 >  Anyway we need proper reviews..
 
 If there is no further comment, I'll commit the above 2) soon.
 
 Thanks,
 ---
 Izumi Tsutsui