Bug 19862

Summary:	unresolved symbols in sr_mod
Product:	[Retired] Red Hat Linux	Reporter:	Need Real Name <gravato>
Component:	kernel	Assignee:	Arjan van de Ven <arjanv>
Status:	CLOSED CURRENTRELEASE	QA Contact:	Brock Organ <borgan>
Severity:	medium	Docs Contact:
Priority:	medium
Version:	7.0	CC:	dr, mjt
Target Milestone:	---
Target Release:	---
Hardware:	i686
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2002-12-15 02:56:21 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Need Real Name 2000-10-26 14:29:50 UTC

I compiled kernel-2.2.16-22 which ships with redhat 7.0 with support for
scsi as modules (when possible).
Everything compiled ok but when loading sr_mod.o it complained about an
unresolved symbol (req_finished_io).
It seems that symbols req_new_io and req_finished_io are not exported...
I solved the problem by adding the following lines to the end of
linux/drivers/block/ll_rw_blk.c

#if CONFIG_SCSI == m
EXPORT_SYMBOL(req_new_io);
EXPORT_SYMBOL(req_finished_io);
#endif

Sorry if this bug was already reported but I couldn't find anything related
to it.

Regards,
Bruno Gravato.

Comment 1 Daniel Roesen 2000-10-26 20:16:26 UTC

At least you look for reports... others don't. See Bug #18032 :-]

Comment 2 Rob McMillin 2000-11-20 18:29:46 UTC

Also seeing this bug ... thanks for the patch, Bruno, I'll try it now.

Comment 3 Rob McMillin 2000-11-20 18:40:39 UTC

That patch definitely worked! Thanks... now my IDE CD-RW drive is back in
service. Whew!

Comment 4 Michael Tokarev 2001-02-14 00:48:27 UTC

People, are you kidding?! (that is to RedHat)
I reported this bug already for rawhide, see bug id 16174
http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=16174 .
I wrote about it to sct also, I provided a patch for a patch (this
is in -sard patch, for the beginning of it's "life").  And this issue
is still with us, and errata 2.2.17-14 also contain it!
I'm curious at the end -- why so trivial issue must take so long time
to resolve (and ever amount of that long time still unknown...)??? :)))
Why it's SO hard to solve this?!?!?

BTW, proposed here patch will not work in all cases -- it basic scsi
support built in but e.g. sr_mod itself (scsi_generic) is module
(this is what I tried here recently).

Comment 5 Stephen Tweedie 2001-02-14 12:13:57 UTC

Just export the symbols unconditionally: there's no need to worry about making
it conditional on the SCSI subsystem being modularised.

Comment 6 Michael Tokarev 2001-02-15 18:46:27 UTC

That's almost exactly the same as I proposed to do at my initial
bugreport... :) -- the only difference is that i was not shure
if some additional symbols should also be exported (no them
shouldn't as it seemed to be).

Comment 7 Arjan van de Ven 2001-02-21 17:34:30 UTC

*** Bug 15632 has been marked as a duplicate of this bug. ***