Re: binutils 2.8.1.0.26 is released.

[Robert HYATT <hyatt at cis dot uab dot edu>]

I downloaded the new bin utils today from the below post (I notice
the post says .26, but I found .27 there, no .26).  I tried to build
a kernel after installing them, but it failed, complaining about
invalid x86 instructions.  I re-installed the previous binutils and
this went away.

It appears that .27 is broken...

I have also worked my way up the kernel tree and am now running
2.1.89 on my SMP (ALR quad-6) box.  2.1.90 and 2.1.91 won't boot.
I get the infamous hang after decompressing the kernel I described
last week (and which a few others are also getting.)

Is there anything "serious" in 2.1.89 that I should look out for?

and if anyone is interested in tracking this down, I am available
to help with my SMP machine.  I'm a long-time C programmer so that's
not an issue.  I'll be in my office tomorrow around 9am CST if any-
one would like to try and figure out what is broken.  Note that there
is *no* output from the kernel after "now booting the kernel" which
leads me to believe that it gets broken by either the decompression
or by getting moved around in memory...


Robert Hyatt                    Computer and Information Sciences
hyatt@cis.uab.edu               University of Alabama at Birmingham
(205) 934-2213                  115A Campbell Hall, UAB Station 
(205) 934-5473 FAX              Birmingham, AL 35294-1170

On Fri, 27 Mar 1998, H.J. Lu wrote:

> Hi,
> 
> I fixed a few x86 assembler bugs which may output inefficient opcodes
> in certain cases. I have verified it by recompiling egcs, binutils,
> glibc 2.1 and Linux kernel 2.1.90. Let me know if you find any problems.
> 
> Thanks.
> 
> 
> -- 
> H.J. Lu (hjl@gnu.org)
> ----
> This is the beta release of binutils 2.8.1.0.26 for Linux, which is
> based on the binutils 980327 snapshot plus Linux/ARM support.
> 
> There was some problem on Linux/alpha. It turned out it was a bug
> in the alpha dynamic linker in glibc 2.0.6. The current glibc 2.1
> and glibc 2.0.7 should be ok.
> 
> This release should work on linux/x86. There are some ELF bug fixes.
> But I don't know if it fixes all the bugs reported on linux/sparc,
> Linux/MIPS and linux/PPC. I'd like to hear reports on them.
> 
> Please report any bugs related to binutils 2.8.1.0.26 to hjl@gnu.org.
> 
> Changes from binutils 2.8.1.0.25:
> 
> 1. Update from the binutils gas-980327.
> 2. There are some major x86 assembler changes with operand size suffix.
>    Some x86 assembler performence bugs are fixed. The Linux kernel and
>    applications may run faster when assembled the new x86 assembler on
>    x86. Please double check them on x86. gcc/egcs should be fine. But
>    the Linux kernel does have some invalid x86 assembly code. In
>    arch/i386/lib/locks.S, there was
> 
> 	btl     %dl, SYMBOL_NAME(smp_invalidate_needed)
> 
>    It is not valid. The register should be either 16-bit or 32-bit.
>    8-bit is not allowed. If you have any x86 assembler problem, please
>    let me know.
> 
> Changes from binutils 2.8.1.0.24:
> 
> 1. Update from the binutils gas-980325.
> 2. Change the x86 assembler error from as_bad to as_warn.
> 3. Fix the address for x86's lgdt/lidt.
> 4. Add 'S' to ar to disable symbol-table generation. Maybe used by
>    glibc build to speed up archive update.
> 
> Changes from binutils 2.8.1.0.23:
> 
> 1. Update from the binutils gas-980320.
> 2. Fix an x86 assembler bug.
> 3. Fix a Sparc/ELF bfd bug.
> 
> IMPORTANT:
> The .plt format that the Alpha was using was not thread safe.  So I 
> changed it.  The new format is *not* binary compatible with the old,
> thus you must use glibc 2.0.4 to get an ld.so that can understand it.
> 
> Note that the new ld.so knows how to deal with the old .plt format, so
> old libraries will still work, but they should be relinked eventually
> for performance reasons.
> 
> The file list:
> 
> 1. binutils-2.8.1.0.26.tar.gz. Source code.
> 2. binutils-2.8.1.0.25-2.8.1.0.26.diff.gz. Patch against the previous
>    beta source code.
> 3. binutils-2.8.1.0.26.bin.tar.gz. Precompiled Linux/x86 binaries for
>    libc 5.
> 4. binutils-2.8.1.0.26-glibc.bin.tar.gz. Precompiled Linux/x86 binaries
>    for libc 6 (glibc 2.0.5c or above).
> 
> The primary ftp sites for the compiler/C library are tsx-11.mit.edu
> under pub/linux/packages/GCC and sunsite.unc.edu under pub/Linux/GCC.
> The beta directory is in private/tofu under the GCC directory.
>  
> Most of my stuff can also be found at
> 
> 
>    ftp://ftp.yggdrasil.com/private/hjl
> 
> To install the binary package, please follow the procedure very closely.
> Please backup/save all the files you are instructed to delete and you
> should do
>  
> 	gzip -dc binutils-2.8.1.0.26.bin.tar.gz | tar tvvf -
>  
> or
> 	gzip -dc binutils-2.8.1.0.26-glibc.bin.tar.gz | tar tvvf -
>  
> to see what is in there.
>  
> Please do back up before you remove things.
> 
> To install for libc 5, PLEASE DO
>  
> 1. su root
> 2. cd /
> 3. rm -f /usr/bin/encaps /usr/bin/nm
> 4. gzip -dc binutils-2.8.1.0.26.bin.tar.gz | tar xvvf -
> 5. ldconfig
>  
> To install for libc 6 (glibc 2.0.5c or above), PLEASE DO
>  
> 1. su root
> 2. cd /
> 3. rm -f /usr/bin/encaps /usr/bin/nm
> 4. gzip -dc binutils-2.8.1.0.26-glibc.bin.tar.gz | tar xvvf -
> 5. ldconfig
>  
> Now you have the new gas/binutils under /usr/bin and
> /usr/ix86-linuxaout/bin. You have to use
>  
> /usr/ix86-linuxaout/bin/as
>  
> and
>  
> /usr/ix86-linuxaout/bin/ld -m i386linux
>  
> if you want to use a.out as and ld directly.
>  
> I have changed the target names for Linux/x86 and Linux/x86 (a.out)
> to ix86-linux and ix86-linuxaout respectively. The precompiled
> binaries are installed under /usr/ix86-linux and /usr/ix86-linuxaout.
> You should make appropriate symbolic links if you have a different
> name for the Linux/x86 target, like i586-unknown-linux.
> 
> If you have an old linux library in the a.out format and you cannot
> obtain the newer version in the ELF format for whatever reason, you
> can try "objcopy --remove-leading-char" on the a.out library and see
> if it can link with your code in ELF. For Sybase client libraries,
> I did
> 
> # cd sybase/lib
> # for f in *.a
> do
>   objcopy --remove-leading-char $f
> done
> 
> It seems to compile/link fine with libc 5.4.2x. But I don't know
> if it really works or not. The resulting binaries should work
> if the libc ABI used by the old library is unchanged in the new
> libc.
> 
> According to Bruce Milner <Bruce.Milner@genetics.utah.edu>:
> 
> Just a quick note about your example using objcopy for
> sybase. The sybase a.out libraries "are" dependent on a changed
> feature of the libc ABI.
>  
> The ctype macros changed to a byte order independent format
> (little endian). In order to use the sybase libraries, one needs to
> rename the the ctype table variables in the sybase library files
> and link in an old version of the libc's "C-ctype.o ctype-info.o"
> files.
> 
> thain@sunquest.sunquest.com put a ELF'ized version of these
> libraries on 
> 
> ftp://mudshark.sunquest.com/pub/ctlib-linux-elf
> 
> 
> Thanks.
> 
> 
> H.J. Lu
> hjl@gnu.org
> 03/27/98
> 
> rom owner-crafty@alpha.jpunix.com  Sat Mar 28 16:54:13 1998
> Return-Path: owner-crafty@alpha.jpunix.com
> Received: from cis.uab.edu (redmont1.cis.uab.edu [138.26.65.64]) by crafty.cis.uab.edu (8.7.5/8.7.3) with SMTP id QAA00244 for <hyatt@crafty>; Sat, 28 Mar 1998 16:54:13 -0600
> Received: from alpha.jpunix.com by cis.uab.edu (5.x/SMI-SVR4)
> 	id AA00516; Fri, 27 Mar 1998 21:08:35 -0600
> Received: (from majordom@localhost)
> 	by alpha.jpunix.com (8.8.8/8.8.7) id VAA17426
> 	for crafty-outgoing; Fri, 27 Mar 1998 21:05:53 -0600 (CST)
> Message-Id: <01BD59C3.ECBAC5C0@SPAM>
> From: "John P. Beck" <jpbeck@swbell.net>
> To: "'crafty@alpha.jpunix.com'" <crafty@alpha.jpunix.com>
> Subject: RE: [Crafty] Dual Processors
> Date: Fri, 27 Mar 1998 21:04:20 -0600
> Mime-Version: 1.0
> Content-Type: multipart/mixed; boundary="---- =_NextPart_000_01BD59C3.ECBAC5C0"
> Sender: owner-crafty@alpha.jpunix.com
> Precedence: bulk
> Reply-To: crafty@alpha.jpunix.com
> 
> 
> ------ =_NextPart_000_01BD59C3.ECBAC5C0
> Content-Type: text/plain; charset="us-ascii"
> Content-Transfer-Encoding: 7bit
> 
> Also, if the .pgn files an sort files were on cd-rom a machine would only have to make the binary file right?
> 
> Just ideas.
> 
> John
> 
> 
> -----Original Message-----
> From:	Robert HYATT [SMTP:hyatt@cis.uab.edu]
> Sent:	Friday, March 27, 1998 12:41 PM
> To:	John P. Beck
> Cc:	'crafty@alpha.jpunix.com'
> Subject:	Re: [Crafty] Dual Processors
> 
> 
> a dual cpu machine should be better, assuming the software kernel
> is smart enough to handle interrupts on the idle processor rather
> than on the busy one, and assuming that when the busy one does get
> interrupted, the process is scheduled on that processor asap rather
> than switching to another with the resulting cache misses...
> 
> I've now got my new quad-p6 sort of running, using a non-current SMP
> kernel from linux... I hope to have it up normally soon, now...
> 
> 
> 
> Robert Hyatt                    Computer and Information Sciences
> hyatt@cis.uab.edu               University of Alabama at Birmingham
> (205) 934-2213                  115A Campbell Hall, UAB Station 
> (205) 934-5473 FAX              Birmingham, AL 35294-1170
> 
> ------ =_NextPart_000_01BD59C3.ECBAC5C0
> Content-Type: application/ms-tnef
> Content-Transfer-Encoding: base64
> 
> eJ8+IhoDAQaQCAAEAAAAAAABAAEAAQeQBgAIAAAA5AQAAAAAAADoAAEIgAcAGAAAAElQTS5NaWNy
> b3NvZnQgTWFpbC5Ob3RlADEIAQ2ABAACAAAAAgACAAEEkAYAwAEAAAEAAAAQAAAAAwAAMAIAAAAL
> AA8OAAAAAAIB/w8BAAAATQAAAAAAAACBKx+kvqMQGZ1uAN0BD1QCAAAAAGNyYWZ0eUBhbHBoYS5q
> cHVuaXguY29tAFNNVFAAY3JhZnR5QGFscGhhLmpwdW5peC5jb20AAAAAHgACMAEAAAAFAAAAU01U
> UAAAAAAeAAMwAQAAABgAAABjcmFmdHlAYWxwaGEuanB1bml4LmNvbQADABUMAQAAAAMA/g8GAAAA
> HgABMAEAAAAaAAAAJ2NyYWZ0eUBhbHBoYS5qcHVuaXguY29tJwAAAAIBCzABAAAAHQAAAFNNVFA6
> Q1JBRlRZQEFMUEhBLkpQVU5JWC5DT00AAAAAAwAAOQAAAAALAEA6AQAAAB4A9l8BAAAAGAAAAGNy
> YWZ0eUBhbHBoYS5qcHVuaXguY29tAAIB918BAAAATQAAAAAAAACBKx+kvqMQGZ1uAN0BD1QCAAAA
> AGNyYWZ0eUBhbHBoYS5qcHVuaXguY29tAFNNVFAAY3JhZnR5QGFscGhhLmpwdW5peC5jb20AAAAA
> AwD9XwEAAAADAP9fAAAAAAIB9g8BAAAABAAAAAAAAAI5YgEEgAEAHQAAAFJFOiBbQ3JhZnR5XSBE
> dWFsIFByb2Nlc3NvcnMACwoBBYADAA4AAADOBwMAGwAVAAQAFAAFACUBASCAAwAOAAAAzgcDABsA
> FQADACgABQA4AQEJgAEAIQAAAEMxQzFDNEUyRTJDNUQxMTE4OTM3MDBDMEYwMjAzN0MxAAAHAQOQ
> BgB4CwAAIQAAAAsAAgABAAAACwAjAAAAAAADACYAAAAAAAsAKQAAAAAAAwAuAAAAAAADADYAAAAA
> AEAAOQBgReoz9lm9AR4AcAABAAAAHQAAAFJFOiBbQ3JhZnR5XSBEdWFsIFByb2Nlc3NvcnMAAAAA
> AgFxAAEAAAAWAAAAAb1Z9jPn4sTBwsXiEdGJNwDA8CA3wQAAHgAeDAEAAAAFAAAAU01UUAAAAAAe
> AB8MAQAAABIAAABqcGJlY2tAc3diZWxsLm5ldAAAAAMABhCtPjauAwAHEBwDAAAeAAgQAQAAAGUA
> AABBTFNPLElGVEhFUEdORklMRVNBTlNPUlRGSUxFU1dFUkVPTkNELVJPTUFNQUNISU5FV09VTERP
> TkxZSEFWRVRPTUFLRVRIRUJJTkFSWUZJTEVSSUdIVD9KVVNUSURFQVNKT0hOAAAAAAIBCRABAAAA
> VwgAAFMIAAChEwAATFpGdbxtj3oDAAoAcmNwZzEyNXIyDGBjMQMwAQcLYG6RDhAwMzMPFmZlD5JP
> AfcCpANjAgBjaArAc4RldALRcHJxMgAAkioKoW5vElAgMAHQhQHQNg+gMDUwNBQh8wHQFBA0fQdt
> AoMAUAPU+xH/EwtiE+EUUBOyGPQU0IsHExXkNhGOMjM4F1SiIAdtIENFFeQ3Gn+nFEAbrxy1eXIV
> 5DkRjq8aUBYxHv8DgkcJ0WsCg98MASD/DlAiLwNzVAhwI9S7FjEhDTgaYSWfA4JCB0D+dA3gI9Ql
> YRZsG3gHEx0G/xtwKv8etyyVIFUOMBZOIej/LJQjiRphME4lZiyUJucdkb8wTSiXLJQqJgKRCOY7
> CW/qMDi/ZQ4wNTnqOwE6v/87yTnUO/I6Xz4vPe09bzuf8znvEGAyOEO6RNFEj0WZ/znURcJEL0f/
> R71HP0VvSTR+OQ5QTIRN4UYDTeACgnOodHlsB5BoCeB0AABDE1AD8GRjdGwKsVxhUDhhZGp1T1AF
> EGfuaAVCFjIMAWMJwFBAAzDwc25leBcwB7AFsADAxQJzcwBQc2IyFFBPQNJhE/BcawngcAuQUB+v
> UIMIYFBwC4BlT4B2V0D/AUBRewwwUkQbkFUgBKALgKZnRdFSxmJhFxBkAiCfU4BTJk+wUXBZcSAx
> TxP/DlBUf1WPVp8AUVfcAKBSTv9aX1tmTwQPwFxvXX9ejw5Qb1fPYO9h/1uTMwKCExBjP1RAaYFR
> cFuQKlBXcCBESQEQYXUqQCBQCsBhwQnAYXBoIEYCIVQE6TARaS0PkDgBQFcQbhPrZO9Qg2ILIHIJ
> UHAyFqDZcDJ3NEMhFwBwAdBrUn9Rn2h/aYZtsGxwBRACMC0rbRADYTopEG91kFN1FGJqBZB0dZBE
> YXT8ZTpUBBphbf9vD3AfcS//cjdPoFuDDiFpgViWDlBzT810XlJXYRcBIEhbcQSQ/1QEHZF3b3h/
> eY96m1bve5+XD5CHUAjQYgqwdDhn2n8PVGPwfZ9+pofgf7ALUHneL20gepALEYAlc1QEG5H/gS+C
> P4NPep9yP4lPil+LZH91snVUdokg0I2vUS+HhDmHkX+Sj5iARG9jdQeA/wIwBdBs4DfhlrKWEJZQ
> jzH9AYBudhAAYAnwa4Ca4AIBu1PAfDJlAPCa4E9gcDxgpFx2CJB3awuAZB7A/56CBPAHQBBhAUAO
> AI8CW2J7n+UCEG8FQhchEvJ2oG1HC1F2oB0AOlxcdOBvvWzBbW0QAxAHkKKQTQ3gyQNgc28BgCBP
> ASAN4LWd0FykRkUAwAMQLmlQvnSbsBcQllBTAYVSeAFAvZzhbk+wONCl5GwUYwMg5xLzAIAFkGx2
> X4FksA5w/1PAqHIBkAAgqQKe0ZshAcF/qHEW4A9wAABksAzQAZAg/i438qhoDlCpIipAlqCpn/+q
> r6u/D8BksAWBrV+ub69/dmwewGSwbK0fsd+y5Skfq+wwELC/tZ+y1GIgKP8Ckba/qLMaYLRvuS+6
> P7tP/6jgHZC8kqlvvf+/D6vsG5D/vJ/CH8MvxD+o4CDQwR/Gr3/Hv8jECvkDMJYPlx+YrXskQWyj
> 8CwgBpAgdL9PsKvADgADoIywT4EgA5FXo/AAINKVd6chIAIgIFhjZC0DYdLwIADBaPeFYdPQCGBs
> UwACIIsQW1ByYWvRdG/U4WSA0gNirwuACsDWANKiINAjPwqFvQqFSs/x0dABAGkgLthd+G9obgqF
> zYTOT89fBUD+e9hc23aFUMuC2//dDwjBh1LSU4AS8mJrbWvKkxwgX5tgAxB2gWF9Ld3kEk/QIddB
> AyBNB5BkMPxnZeQT2wvN5oDdjQDfz/+On4+vhR+GL4c/iE/RJOIWN3VTDIIH8WIEkAVASFkAQVRU
> IFtTTVQQUDpoeXaQdEBjKQQALnUBoC4JgHVd/+W/5sSMm+f/6Q/qH5Cf0Kz/DCHiJQZgAjDwpDgC
> 8Ab2AGxhedHAm2ByFtBjQDeb0cAgwDnuEA4gOjQOkPxQTdsG+ql1sfut2sJscDYuKhAFkGv+j1OA
> Q2OV+54nBQBhnHB5QAdAQWzwYS5qcHXLEHjyLgjgbScBn/sxdgX7rdN/sHWQW0MEA11r8PLQ36gg
> orGkgBaQU7Bz81/OIv/fj+Cf+V/RM96fC08MX93a97Mw8yAswWME0NTnT6DVg7/xIBQh8mBZQNHA
> aSBzmwD/W6LSEqPyi3EYQFky67AOpf8wYNMg1PDTUTlQ61BnoNZiHbKBZNfBdQFZQHJ1cP500uDU
> MdISECDXwSjQCVV/1+B2kKcRDqXSENMBGWVi99lA14DUMGUUwZzwFNp2kG/T0E+wG/wSoG/S0eVA
> dP8WphjGOODRwNISGiUXAzCw//MRT4DVsh20GihpIJtwGr//0xEwUKgA1SEVUdaA0wChIP+nEdPQ
> a6AX8dIhONAVACpBv1uxn3Ah8dTgMGHS0C4oUHnYXEkna9EP8R9woSFtN9eAU0Ap0HHy0NRwcDb/
> 0ySkANfgBOBZYtHA2UBbovOzMA/wbi2a8CBgmyLx4L5QDqUWRNKQ1JLrgXUFEPcoUClQW1BvnkAY
> BGvRa6D/LCAj0A/wU8LV8aPw1DDRwN8P8Shf2FzxBvJCIDSPNLCeQ3VwExAUkRzjSW6dUvV2kGnU
> MVPykDlQCWEOpfvyPzS9VcsQf9JroByB0fCf0YDw0KMAszAd4UJpU8ArWXEiwG0OpSiZADUpICA5
> MzQtfXAxM7c0v/3wbbBBArCjAHDxII8xAPFgMPHRwFVBQvtAR+MgNwM8fzU0Nz2wRpxBWD3cO9jR
> wEFMapCYNTI5PWA+4DcwzMBP4pac4eNpDqZ9AEcAAAMAEBAAAAAAAwAREAAAAAADAIAQ/////0AA
> BzDgPkcc9lm9AUAACDDgPkcc9lm9AQsAAIAIIAYAAAAAAMAAAAAAAABGAAAAAAOFAAAAAAAAAwAC
> gAggBgAAAAAAwAAAAAAAAEYAAAAAEIUAAAAAAAADAAWACCAGAAAAAADAAAAAAAAARgAAAABShQAA
> tw0AAB4AJYAIIAYAAAAAAMAAAAAAAABGAAAAAFSFAAABAAAABAAAADguMAADACaACCAGAAAAAADA
> AAAAAAAARgAAAAABhQAAAAAAAAsAL4AIIAYAAAAAAMAAAAAAAABGAAAAAA6FAAAAAAAAAwAwgAgg
> BgAAAAAAwAAAAAAAAEYAAAAAEYUAAAAAAAADADKACCAGAAAAAADAAAAAAAAARgAAAAAYhQAAAAAA
> AB4AQYAIIAYAAAAAAMAAAAAAAABGAAAAADaFAAABAAAAAQAAAAAAAAAeAEKACCAGAAAAAADAAAAA
> AAAARgAAAAA3hQAAAQAAAAEAAAAAAAAAHgBDgAggBgAAAAAAwAAAAAAAAEYAAAAAOIUAAAEAAAAB
> AAAAAAAAAB4APQABAAAABQAAAFJFOiAAAAAAAwANNP03AACU8Q==
> 
> ------ =_NextPart_000_01BD59C3.ECBAC5C0--
>