Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 98915 - app-emulation/bochs-2.2.1 compiles with some small patches
Summary: app-emulation/bochs-2.2.1 compiles with some small patches
Status: RESOLVED WONTFIX
Alias: None
Product: Gentoo/Alt
Classification: Unclassified
Component: Mac OSX (show other bugs)
Hardware: PPC OS X
: High normal (vote)
Assignee: Gentoo for Mac OS X
URL:
Whiteboard:
Keywords:
Depends on: 72229 98914
Blocks:
  Show dependency tree
 
Reported: 2005-07-13 12:21 UTC by Fabian Groffen
Modified: 2008-03-01 22:13 UTC (History)
1 user (show)

See Also:
Package list:
Runtime testing required: ---


Attachments
patches applied to the sources (placed in ${FILESDIR}/bochs-2.2.1-osx.patch) (bochs-2.2.1-osx.patch,805 bytes, patch)
2005-07-13 12:23 UTC, Fabian Groffen
Details | Diff
changes made to the ebuild (bochs-2.2.1.ebuild.patch,1.43 KB, patch)
2005-07-13 12:26 UTC, Fabian Groffen
Details | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Fabian Groffen gentoo-dev 2005-07-13 12:21:40 UTC
when patching the sources a bit, and patching the ebuild to configure the right
way for OSX, bochs compiles quite cleanly on OSX.

Reproducible: Always
Steps to Reproduce:
1.
2.
3.
Comment 1 Fabian Groffen gentoo-dev 2005-07-13 12:23:32 UTC
Created attachment 63335 [details, diff]
patches applied to the sources (placed in ${FILESDIR}/bochs-2.2.1-osx.patch)

iodev/cdrom.cc: on Tiger (10.4) dev/disk.h seems to
have been moved to sys/disk.h. On Panther (10.3)
dev/disk.h and sys/disk.h are both availlable. From
dev/disk.h for 10.3:
#warning <dev/disk.h> is obsolete, please use
<sys/disk.h> instead
sys/disk.h defines some extra stuff and structs.

iodev/eth_tuntap.cc: Apple OSX Tiger's GCC4 is picky on types and does not
allow arithmetic operations on void* types, thus:
Comment 2 Fabian Groffen gentoo-dev 2005-07-13 12:26:08 UTC
Created attachment 63336 [details, diff]
changes made to the ebuild

make sure patch #63335 is run for OSX users and configure with the appropriate
sound system, ie. osx on OSX and linux on the rest (other options are win and
freebsd but I don't know it portage supports the last one, otherwise an elif
should be added)
Comment 3 Fabian Groffen gentoo-dev 2005-07-13 13:39:47 UTC
I've just booted GNU-Hurd in my bochs system on OSX, so it seems to work correctly.

Maybe the bios files from the source should be installed somewhere.  I couldn't
find any other option than extract the sources and grab them to get running.
Comment 4 Hasan Khalil (RETIRED) gentoo-dev 2005-07-13 14:22:40 UTC
Adding Diego for input the FreeBSD stuff, cause I know he's good about taking care of stuff.

Diego: If you're not interested, just remove yourself from the CC. Just thought this might interest you.
Comment 5 Diego Elio Pettenò (RETIRED) gentoo-dev 2005-07-13 14:36:04 UTC
Thanks Hasan, although i have moved the cc to bsd@g.o to have it in the right 
query. 
 
Let me look at this, seems like the sys/disk.h change is going to be needed 
for us, too. 
BTW better if the patch is going to be applied unconditionally, simpler for 
everyone, let me see anyway :) 
 
Comment 6 Fabian Groffen gentoo-dev 2005-07-14 00:03:19 UTC
for OSX to get the picture:
dependencies for bochs are as follows, and hence the bugs on them (opensp and
openjade)
[ebuild  N    ] app-text/opensp-1.5-r1  +nls 0 kB [1]
[ebuild  N    ] app-emulation/bochs-2.2.1  -debugger (-gtk2) +readline -sdl
-wxwindows 5,166 kB [1]
[ebuild  N    ] app-text/sgml-common-0.6.3-r4  0 kB
[ebuild  N    ] app-text/openjade-1.3.2-r1  0 kB [1] 

for FreeBSD:
the sys/disk.h is in a "#if defined (__APPLE_CC__)" block, so it won't have
effect on freebsd.  However, there are many defines in there, also for freebsd I
guess, so that probably needs an additional patch.

for all interested:
I also reported my patch as a 'patch' on bochs' sf page:
http://sourceforge.net/tracker/index.php?func=detail&aid=1237763&group_id=12580&atid=312580
Comment 7 Diego Elio Pettenò (RETIRED) gentoo-dev 2005-07-14 01:55:44 UTC
Well if both the changes are applied in a __APPLE__ (or similar) block, go for 
an unconditional patch, simpler for us to add one later. 
Comment 8 Fabian Groffen gentoo-dev 2008-03-01 22:13:10 UTC
lost interest, pretty much slow too.