[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20180603011121.GU1392@brightrain.aerifal.cx>
Date: Sat, 2 Jun 2018 21:11:21 -0400
From: Rich Felker <dalias@...c.org>
To: musl@...ts.openwall.com
Subject: Re: [PATCH] Fix TLS layout of TLS variant I when there is a
gap above TP
On Sat, Jun 02, 2018 at 09:15:23AM +0200, Szabolcs Nagy wrote:
> * Rich Felker <dalias@...c.org> [2018-06-01 22:59:11 -0400]:
> > On Sat, Jun 02, 2018 at 01:52:01AM +0200, Szabolcs Nagy wrote:
> > > In TLS variant I the TLS is above TP (or above a fixed offset from TP)
> > > but on some targets there is a reserved gap above TP before TLS starts.
> > >
> > > This matters for the local-exec tls access model when the offsets of
> > > TLS variables from the TP are hard coded by the linker into the
> > > executable, so the libc must compute these offsets the same way as the
> > > linker. The tls offset of the main module has to be
> > >
> > > alignup(GAP_ABOVE_TP, main_tls_align).
> > >
> > > If there is no TLS in the main module then the gap can be ignored
> > > since musl does not use it and the tls access models of shared
> > > libraries are not affected.
> > >
> > > The previous setup only worked if (tls_align & -GAP_ABOVE_TP) == 0
> > > (i.e. TLS did not require large alignment) because the gap was
> > > treated as a fixed offset from TP. Now the TP points at the end
> > > of the pthread struct (which is aligned) and there is a gap above
> > > it (which may also need alignment).
> > >
> > > The fix required changing TP_ADJ and __pthread_self on affected
> > > targets (aarch64, arm and sh) and in the tlsdesc asm the offset to
> > > access the dtv changed too.
> > > ---
> >
> > On first glance it all looks right. I'll read in more detail soon.
> > Thanks!
> >
> > > passed my simple local-exec tests.
> >
> > Did you test all archs or just some? I think we should at least run
> > libc-test (if it sufficiently tests TLS) on the affected archs to make
> > sure there are no regressions.
> >
>
> ran libc-test on various targets via qemu-user, i didnt see any regressions.
I just setup libc-test to run under qemu-system-sh4, and confirmed
failure before the patch, success with the patch applied. I don't have
any other qemu-system environments setup, but success here looks like
a good sign!
Rich
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
