public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
To: GCC Patches <gcc-patches@gcc.gnu.org>,
	Ramana Radhakrishnan	<Ramana.Radhakrishnan@arm.com>,
	Richard Earnshaw <Richard.Earnshaw@arm.com>,
	James Greenhalgh <James.Greenhalgh@arm.com>,
	Richard Sandiford	<Richard.Sandiford@arm.com>
Cc: nd <nd@arm.com>
Subject: Re: [PATCH][AArch64] Fix symbol offset limit
Date: Mon, 02 Sep 2019 12:12:00 -0000	[thread overview]
Message-ID: <VI1PR0801MB21272E3761FDA739729BCE4D83BE0@VI1PR0801MB2127.eurprd08.prod.outlook.com> (raw)
In-Reply-To: <VI1PR0801MB21275C4C00B04183555DA96483A80@VI1PR0801MB2127.eurprd08.prod.outlook.com>

     
 ping
     
   
  In aarch64_classify_symbol symbols are allowed full-range offsets on relocations. 
   This means the offset can use all of the +/-4GB offset, leaving no offset available
   for the symbol itself.  This results in relocation overflow and link-time errors
   for simple expressions like &global_char + 0xffffff00.
   
   To avoid this, limit the offset to +/-1MB so that the symbol needs to be within a
   3.9GB offset from its references.  For the tiny code model use a 64KB offset, allowing
   most of the 1MB range for code/data between the symbol and its references.
   
   Bootstrapped on AArch64, passes regress, OK for commit?
   
   ChangeLog:
   2018-11-09  Wilco Dijkstra  <wdijkstr@arm.com>
   
       gcc/
           * config/aarch64/aarch64.c (aarch64_classify_symbol):
           Apply reasonable limit to symbol offsets.
   
       testsuite/
           * gcc.target/aarch64/symbol-range.c (foo): Set new limit.
           * gcc.target/aarch64/symbol-range-tiny.c (foo): Likewise.
   
   --
   
   diff --git a/gcc/config/aarch64/aarch64.c b/gcc/config/aarch64/aarch64.c
   index 83453d03095018eddd1801e71ef3836849267444..0023cb37bbae5afe9387840c1bb6b43586d4fac2 100644
   --- a/gcc/config/aarch64/aarch64.c
   +++ b/gcc/config/aarch64/aarch64.c
   @@ -13047,26 +13047,26 @@ aarch64_classify_symbol (rtx x, HOST_WIDE_INT offset)
                 the offset does not cause overflow of the final address.  But
                 we have no way of knowing the address of symbol at compile time
                 so we can't accurately say if the distance between the PC and
   -            symbol + offset is outside the addressible range of +/-1M in the
   -            TINY code model.  So we rely on images not being greater than
   -            1M and cap the offset at 1M and anything beyond 1M will have to
   -            be loaded using an alternative mechanism.  Furthermore if the
   -            symbol is a weak reference to something that isn't known to
   -            resolve to a symbol in this module, then force to memory.  */
   +            symbol + offset is outside the addressible range of +/-1MB in the
   +            TINY code model.  So we limit the maximum offset to +/-64KB and
   +            assume the offset to the symbol is not larger than +/-(1MB - 64KB).
   +            Furthermore force to memory if the symbol is a weak reference to
   +            something that doesn't resolve to a symbol in this module.  */
              if ((SYMBOL_REF_WEAK (x)
                   && !aarch64_symbol_binds_local_p (x))
   -             || !IN_RANGE (offset, -1048575, 1048575))
   +             || !IN_RANGE (offset, -0x10000, 0x10000))
                return SYMBOL_FORCE_TO_MEM;
   +
              return SYMBOL_TINY_ABSOLUTE;
    
            case AARCH64_CMODEL_SMALL:
              /* Same reasoning as the tiny code model, but the offset cap here is
   -            4G.  */
   +            1MB, allowing +/-3.9GB for the offset to the symbol.  */
              if ((SYMBOL_REF_WEAK (x)
                   && !aarch64_symbol_binds_local_p (x))
   -             || !IN_RANGE (offset, HOST_WIDE_INT_C (-4294967263),
   -                           HOST_WIDE_INT_C (4294967264)))
   +             || !IN_RANGE (offset, -0x100000, 0x100000))
                return SYMBOL_FORCE_TO_MEM;
   +
              return SYMBOL_SMALL_ABSOLUTE;
    
            case AARCH64_CMODEL_TINY_PIC:
   diff --git a/gcc/testsuite/gcc.target/aarch64/symbol-range-tiny.c b/gcc/testsuite/gcc.target/aarch64/symbol-range-tiny.c
   index d7e46b059e41f2672b3a1da5506fa8944e752e01..d49ff4dbe5786ef6d343d2b90052c09676dd7fe5 100644
   --- a/gcc/testsuite/gcc.target/aarch64/symbol-range-tiny.c
   +++ b/gcc/testsuite/gcc.target/aarch64/symbol-range-tiny.c
   @@ -1,12 +1,12 @@
   -/* { dg-do compile } */
   +/* { dg-do link } */
    /* { dg-options "-O3 -save-temps -mcmodel=tiny" } */
    
   -int fixed_regs[0x00200000];
   +char fixed_regs[0x00200000];
    
    int
   -foo()
   +main ()
    {
   -  return fixed_regs[0x00080000];
   +  return fixed_regs[0x000ff000];
    }
    
    /* { dg-final { scan-assembler-not "adr\tx\[0-9\]+, fixed_regs\\\+" } } */
   diff --git a/gcc/testsuite/gcc.target/aarch64/symbol-range.c b/gcc/testsuite/gcc.target/aarch64/symbol-range.c
   index 6574cf4310430b847e77ea56bf8f20ef312d53e4..75c87c12f08004c153efc5192e5cfab566c089db 100644
   --- a/gcc/testsuite/gcc.target/aarch64/symbol-range.c
   +++ b/gcc/testsuite/gcc.target/aarch64/symbol-range.c
   @@ -1,12 +1,12 @@
   -/* { dg-do compile } */
   +/* { dg-do link } */
    /* { dg-options "-O3 -save-temps -mcmodel=small" } */
    
   -int fixed_regs[0x200000000ULL];
   +char fixed_regs[0x200000000ULL];
    
    int
   -foo()
   +main ()
    {
   -  return fixed_regs[0x100000000ULL];
   +  return fixed_regs[0xfffff000ULL];
    }
    
    /* { dg-final { scan-assembler-not "adrp\tx\[0-9\]+, fixed_regs\\\+" } } */
   
               

  reply	other threads:[~2019-09-02 12:12 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-05-28 17:27 Wilco Dijkstra
2019-07-31 17:00 ` Wilco Dijkstra
2019-08-19 15:57   ` Wilco Dijkstra
2019-09-02 12:12     ` Wilco Dijkstra [this message]
2019-09-09 17:07       ` Wilco Dijkstra
2019-10-10 17:24         ` Wilco Dijkstra
2019-10-10 18:33           ` Richard Sandiford
2019-10-11 18:31             ` Wilco Dijkstra
2019-10-12 10:20               ` Richard Sandiford
2019-10-12 11:23                 ` Richard Sandiford
2019-10-13  8:45                 ` Jeff Law
2019-10-14 15:51                 ` Wilco Dijkstra
2019-10-14 16:34                   ` Richard Sandiford
2019-10-15 18:27                     ` Wilco Dijkstra
2019-10-16  8:50                       ` Richard Sandiford
  -- strict thread matches above, loose matches on Subject: below --
2018-11-09 14:47 Wilco Dijkstra
2016-08-23 14:11 Wilco Dijkstra
2016-08-26 10:43 ` Richard Earnshaw (lists)
2016-08-26 19:07   ` Wilco Dijkstra

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=VI1PR0801MB21272E3761FDA739729BCE4D83BE0@VI1PR0801MB2127.eurprd08.prod.outlook.com \
    --to=wilco.dijkstra@arm.com \
    --cc=James.Greenhalgh@arm.com \
    --cc=Ramana.Radhakrishnan@arm.com \
    --cc=Richard.Earnshaw@arm.com \
    --cc=Richard.Sandiford@arm.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=nd@arm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).