linux-kernel - Re: [RFC PATCH 6/7] tools: separate out shared radix-tree components

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date: Thu, 27 Jun 2024 21:39:14 +0100
From: Lorenzo Stoakes <lstoakes@...il.com>
To: "Liam R. Howlett" <Liam.Howlett@...cle.com>,
	Andrew Morton <akpm@...ux-foundation.org>,
	linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org,
	linux-mm@...ck.org, Vlastimil Babka <vbabka@...e.cz>,
	Matthew Wilcox <willy@...radead.org>,
	Alexander Viro <viro@...iv.linux.org.uk>,
	Christian Brauner <brauner@...nel.org>, Jan Kara <jack@...e.cz>,
	Eric Biederman <ebiederm@...ssion.com>, Kees Cook <kees@...nel.org>,
	Suren Baghdasaryan <surenb@...gle.com>
Subject: Re: [RFC PATCH 6/7] tools: separate out shared radix-tree components

On Thu, Jun 27, 2024 at 04:03:21PM -0400, Liam R. Howlett wrote:
> * Lorenzo Stoakes <lstoakes@...il.com> [240627 15:47]:
> > On Thu, Jun 27, 2024 at 01:59:18PM -0400, Liam R. Howlett wrote:
> > > * Lorenzo Stoakes <lstoakes@...il.com> [240627 06:39]:
> > > > The core components contained within the radix-tree tests which provide
> > > > shims for kernel headers and access to the maple tree are useful for
> > > > testing other things, so separate them out and make the radix tree tests
> > > > dependent on the shared components.
> > > >
> > > > This lays the groundwork for us to add VMA tests of the newly introduced
> > > > vma.c file.
> > >
> > > This separation and subsequent patch requires building the
> > > xarray-hsared, radix-tree, idr, find_bit, and bitmap .o files which are
> > > unneeded for the target 'main'.  I'm not a build expert on how to fix
> > > this, but could that be reduced to the minimum set somehow?
> >
> > I'm confused, the existing Makefile specified:
> >
> > CORE_OFILES := xarray.o radix-tree.o idr.o linux.o test.o find_bit.o bitmap.o \
> > 			 slab.o maple.o
> >
> > OFILES = main.o $(CORE_OFILES) regression1.o regression2.o regression3.o \
> > 	 regression4.o tag_check.o multiorder.o idr-test.o iteration_check.o \
> > 	 iteration_check_2.o benchmark.o
> >
> > main:	$(OFILES)
> >
> > Making all of the files you mentioned dependencies of main no? (xarray-shared
> > being a subset of xarray.o which requires it anyway)
>
> After replacing main with vma and dropping the vma.c, I can generate the
> vma executable. I had to generate map-shift.h and bit-lenght.h, then
> execute:
>
> cc -O2 -g -Wall -I../shared -I. -I../../include -I../../../lib -g -Og -Wall -D_LGPL_SOURCE -fsanitize=address -fsanitize=undefined -c -o vma.o vma.c
> cc -c -O2 -g -Wall -I../shared -I. -I../../include -I../../../lib -g -Og -Wall -D_LGPL_SOURCE -fsanitize=address -fsanitize=undefined ../shared/linux.c -o linux.o
> cc -O2 -g -Wall -I../shared -I. -I../../include -I../../../lib -g -Og -Wall -D_LGPL_SOURCE -fsanitize=address -fsanitize=undefined -c -o slab.o ../../lib/slab.c
> cc -c -O2 -g -Wall -I../shared -I. -I../../include -I../../../lib -g -Og -Wall -D_LGPL_SOURCE -fsanitize=address -fsanitize=undefined ../shared/maple-shim.c -o maple-shim.o
> cc -O2 -g -Wall -I../shared -I. -I../../include -I../../../lib -g -Og -Wall -D_LGPL_SOURCE -fsanitize=address -fsanitize=undefined -c -o vma_stub.o vma_stub.c
> cc -fsanitize=address -fsanitize=undefined vma.o linux.o slab.o maple-shim.o vma_stub.o -lpthread -lurcu -o vma
>
> Dropping a number of the pre-built items.  When I looked at the list, it
> seemed too numerous for what we were doing (why do we need idr, and
> radix-tree?)  So I tried removing as many as I could.

I did see link errors at some point when radix-tree wasn't included there,
I'll take another look and see if we can reduce to a smaller subset if
posible.

>
> >
> > I'm not sure this is a huge big deal as these are all rather small :)
>
> Yeah, maybe not worth the effort.
>
> Cheers,
> Liam