All of lore.kernel.org
 help / color / mirror / Atom feed
From: Warner Losh <imp@bsdimp.com>
To: qemu-devel@nongnu.org
Cc: Kyle Evans <kevans@FreeBSD.org>,
	richard.henderson@linaro.org, f4bug@amsat.org,
	Warner Losh <imp@bsdimp.com>
Subject: [PATCH v3 8/9] bsd-user/mmap.c: Implement MAP_EXCL, required by jemalloc in head
Date: Fri,  8 Oct 2021 15:23:43 -0600	[thread overview]
Message-ID: <20211008212344.95537-9-imp@bsdimp.com> (raw)
In-Reply-To: <20211008212344.95537-1-imp@bsdimp.com>

From: Kyle Evans <kevans@FreeBSD.org>

jemalloc requires a working MAP_EXCL. Ensure that no page is double
mapped when specified. In addition, use guest_range_valid_untagged to
test for valid ranges of pages rather than an incomplete inlined version
of the test that might be wrong.

Signed-off-by: Kyle Evans <kevans@FreeBSD.org>
Signed-off-by: Warner Losh <imp@bsdimp.com>
---
 bsd-user/mmap.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/bsd-user/mmap.c b/bsd-user/mmap.c
index 4ecd949a10..066d9c10ff 100644
--- a/bsd-user/mmap.c
+++ b/bsd-user/mmap.c
@@ -403,7 +403,7 @@ abi_ulong mmap_find_vma(abi_ulong start, abi_ulong size)
 abi_long target_mmap(abi_ulong start, abi_ulong len, int prot,
                      int flags, int fd, off_t offset)
 {
-    abi_ulong ret, end, real_start, real_end, retaddr, host_offset, host_len;
+    abi_ulong addr, ret, end, real_start, real_end, retaddr, host_offset, host_len;
 
     mmap_lock();
     if (qemu_loglevel_mask(CPU_LOG_PAGE)) {
@@ -574,12 +574,10 @@ abi_long target_mmap(abi_ulong start, abi_ulong len, int prot,
          * It can fail only on 64-bit host with 32-bit target.
          * On any other target/host host mmap() handles this error correctly.
          */
-#if TARGET_ABI_BITS == 32 && HOST_LONG_BITS == 64
-        if ((unsigned long)start + len - 1 > (abi_ulong) -1) {
+        if (!guest_range_valid_untagged(start, len)) {
             errno = EINVAL;
             goto fail;
         }
-#endif
 
         /*
          * worst case: we cannot map the file because the offset is not
@@ -614,6 +612,12 @@ abi_long target_mmap(abi_ulong start, abi_ulong len, int prot,
             goto the_end;
         }
 
+        /* Reject the mapping if any page within the range is mapped */
+        if ((flags & MAP_EXCL) && page_check_range(start, len, 0) < 0) {
+            errno = EINVAL;
+            goto fail;
+        }
+
         /* handle the start of the mapping */
         if (start > real_start) {
             if (real_end == real_start + qemu_host_page_size) {
-- 
2.32.0



  parent reply	other threads:[~2021-10-08 21:29 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-08 21:23 [PATCH v3 0/9] bsd-user mmap fixes Warner Losh
2021-10-08 21:23 ` [PATCH v3 1/9] bsd-user/mmap.c: Always zero MAP_ANONYMOUS memory in mmap_frag() Warner Losh
2021-10-14 15:05   ` Kyle Evans
2021-10-08 21:23 ` [PATCH v3 2/9] bsd-user/mmap.c: check pread's return value to fix warnings with _FORTIFY_SOURCE Warner Losh
2021-10-14 15:06   ` Kyle Evans
2021-10-08 21:23 ` [PATCH v3 3/9] bsd-user/mmap.c: MAP_ symbols are defined, so no need for ifdefs Warner Losh
2021-10-14 15:06   ` Kyle Evans
2021-10-08 21:23 ` [PATCH v3 4/9] bsd-user/mmap.c: mmap return ENOMEM on overflow Warner Losh
2021-10-14 15:13   ` Kyle Evans
2021-10-08 21:23 ` [PATCH v3 5/9] bsd-user/mmap.c: mmap prefer MAP_ANON for BSD Warner Losh
2021-10-14 15:06   ` Kyle Evans
2021-10-08 21:23 ` [PATCH v3 6/9] bsd-user/mmap.c: Convert to qemu_log logging for mmap debugging Warner Losh
2021-10-09 16:01   ` Richard Henderson
2021-10-11 18:52   ` Philippe Mathieu-Daudé
2021-10-18  3:44   ` Kyle Evans
2021-10-08 21:23 ` [PATCH v3 7/9] bsd-user/mmap.c: Don't mmap fd == -1 independently from MAP_ANON flag Warner Losh
2021-10-09 16:03   ` Richard Henderson
2021-10-18  3:43   ` Kyle Evans
2021-10-08 21:23 ` Warner Losh [this message]
2021-10-09 16:04   ` [PATCH v3 8/9] bsd-user/mmap.c: Implement MAP_EXCL, required by jemalloc in head Richard Henderson
2021-10-08 21:23 ` [PATCH v3 9/9] bsd-user/mmap.c: assert that target_mprotect cannot fail Warner Losh
2021-10-14 15:13   ` Kyle Evans

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=20211008212344.95537-9-imp@bsdimp.com \
    --to=imp@bsdimp.com \
    --cc=f4bug@amsat.org \
    --cc=kevans@FreeBSD.org \
    --cc=qemu-devel@nongnu.org \
    --cc=richard.henderson@linaro.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.