qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Cc: deller@gmx.de, laurent@vivier.eu, akihiko.odaki@daynix.com,
	joel@jms.id.au
Subject: [PATCH v7 12/14] linux-user: Use elf_et_dyn_base for ET_DYN with interpreter
Date: Wed,  2 Aug 2023 18:53:00 -0700	[thread overview]
Message-ID: <20230803015302.407219-13-richard.henderson@linaro.org> (raw)
In-Reply-To: <20230803015302.407219-1-richard.henderson@linaro.org>

Follow the lead of the linux kernel in fs/binfmt_elf.c,
in which an ET_DYN executable which uses an interpreter
(usually a PIE executable) is loaded away from where the
interpreter itself will be loaded.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
 linux-user/elfload.c | 43 +++++++++++++++++++++++++++++++++++++++----
 1 file changed, 39 insertions(+), 4 deletions(-)

diff --git a/linux-user/elfload.c b/linux-user/elfload.c
index 2aee2298ec..fef9a0bc8f 100644
--- a/linux-user/elfload.c
+++ b/linux-user/elfload.c
@@ -3021,7 +3021,7 @@ static void load_elf_image(const char *image_name, int image_fd,
     struct elfhdr *ehdr = (struct elfhdr *)bprm_buf;
     struct elf_phdr *phdr;
     abi_ulong load_addr, load_bias, loaddr, hiaddr, error;
-    int i, retval, prot_exec;
+    int i, retval, prot_exec, load_map_flags;
     Error *err = NULL;
 
     /* First of all, some simple consistency checks */
@@ -3106,6 +3106,19 @@ static void load_elf_image(const char *image_name, int image_fd,
         }
     }
 
+    load_addr = loaddr;
+
+    /*
+     * For ET_EXEC, load_addr is required.  Use MAP_FIXED_NOREPLACE instead
+     * of MAP_FIXED on the off-chance that the guest address overlaps a
+     * host address.  There should be no other conflict this early in the
+     * loading process.
+     */
+    if (ehdr->e_type == ET_EXEC) {
+        load_map_flags = MAP_FIXED_NOREPLACE;
+    } else {
+        load_map_flags = 0;
+    }
     if (pinterp_name != NULL) {
         /*
          * This is the main executable.
@@ -3135,11 +3148,34 @@ static void load_elf_image(const char *image_name, int image_fd,
              */
             probe_guest_base(image_name, loaddr, hiaddr);
         } else {
+            abi_ulong align;
+
             /*
              * The binary is dynamic, but we still need to
              * select guest_base.  In this case we pass a size.
              */
             probe_guest_base(image_name, 0, hiaddr - loaddr);
+
+            /*
+             * Avoid collision with the loader by providing a different
+             * default load address.
+             */
+            load_addr = loaddr + elf_et_dyn_base;
+
+            /*
+             * TODO: Better support for mmap alignment is desirable.
+             * Without reserved_va we would prefer any host conflict be
+             * resolved by choosing a different address, therefore we
+             * don't want to use MAP_FIXED.  But without that we cannot
+             * cannot guarantee alignment, only suggest it.
+             */
+            align = pow2ceil(info->alignment);
+            if (align) {
+                load_addr &= -align;
+            }
+            if (reserved_va) {
+                load_map_flags = MAP_FIXED_NOREPLACE;
+            }
         }
     }
 
@@ -3157,10 +3193,9 @@ static void load_elf_image(const char *image_name, int image_fd,
      * In both cases, we will overwrite pages in this range with mappings
      * from the executable.
      */
-    load_addr = target_mmap(loaddr, (size_t)hiaddr - loaddr + 1, PROT_NONE,
+    load_addr = target_mmap(load_addr, (size_t)hiaddr - loaddr + 1, PROT_NONE,
                             MAP_PRIVATE | MAP_ANON | MAP_NORESERVE |
-                            (ehdr->e_type == ET_EXEC ? MAP_FIXED : 0),
-                            -1, 0);
+                            load_map_flags, -1, 0);
     if (load_addr == -1) {
         goto exit_mmap;
     }
-- 
2.34.1



  parent reply	other threads:[~2023-08-03  1:55 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-03  1:52 [PATCH v7 00/14] linux-user: brk fixes Richard Henderson
2023-08-03  1:52 ` [PATCH v7 01/14] linux-user: Unset MAP_FIXED_NOREPLACE for host Richard Henderson
2023-08-03  1:52 ` [PATCH v7 02/14] linux-user: Fix MAP_FIXED_NOREPLACE on old kernels Richard Henderson
2023-08-03  1:52 ` [PATCH v7 03/14] linux-user: Do not call get_errno() in do_brk() Richard Henderson
2023-08-03  1:52 ` [PATCH v7 04/14] linux-user: Use MAP_FIXED_NOREPLACE for do_brk() Richard Henderson
2023-08-03  1:52 ` [PATCH v7 05/14] linux-user: Do nothing if too small brk is specified Richard Henderson
2023-08-03  1:52 ` [PATCH v7 06/14] linux-user: Do not align brk with host page size Richard Henderson
2023-08-03  1:52 ` [PATCH v7 07/14] linux-user: Remove last_brk Richard Henderson
2023-08-03  1:52 ` [PATCH v7 08/14] bsd-user: " Richard Henderson
2023-08-03  1:52 ` [PATCH v7 09/14] linux-user: Adjust task_unmapped_base for reserved_va Richard Henderson
2023-08-03  1:52 ` [PATCH v7 10/14] linux-user: Define TASK_UNMAPPED_BASE in $guest/target_mman.h Richard Henderson
2023-08-03  1:52 ` [PATCH v7 11/14] linux-user: Add ELF_ET_DYN_BASE Richard Henderson
2023-08-03  1:53 ` Richard Henderson [this message]
2023-08-03  1:53 ` [PATCH v7 13/14] linux-user: Adjust initial brk when interpreter is close to executable Richard Henderson
2023-08-03 13:00   ` Helge Deller
2023-08-03  1:53 ` [PATCH v7 14/14] linux-user: Properly set image_info.brk in flatload Richard Henderson
2023-08-03 13:11 ` [PATCH v7 00/14] linux-user: brk fixes Joel Stanley
2023-08-03 13:55   ` Helge Deller
2023-08-03 14:17     ` Joel Stanley
2023-08-03 15:01       ` Helge Deller
2023-08-03 15:11         ` Richard Henderson
2023-08-03 16:09           ` Helge Deller
2023-08-03 15:20         ` Richard Henderson
2023-08-03 16:10           ` Helge Deller

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=20230803015302.407219-13-richard.henderson@linaro.org \
    --to=richard.henderson@linaro.org \
    --cc=akihiko.odaki@daynix.com \
    --cc=deller@gmx.de \
    --cc=joel@jms.id.au \
    --cc=laurent@vivier.eu \
    --cc=qemu-devel@nongnu.org \
    /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 NNTP newsgroup(s).