qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Cc: r.bolshakov@yadro.com, j@getutm.app
Subject: [PATCH v2 26/29] tcg: Round the tb_size default from qemu_get_host_physmem
Date: Sun, 14 Mar 2021 15:27:21 -0600	[thread overview]
Message-ID: <20210314212724.1917075-27-richard.henderson@linaro.org> (raw)
In-Reply-To: <20210314212724.1917075-1-richard.henderson@linaro.org>

If qemu_get_host_physmem returns an odd number of pages,
then physmem / 8 will not be a multiple of the page size.

The following was observed on a gitlab runner:

ERROR qtest-arm/boot-serial-test - Bail out!
ERROR:../util/osdep.c:80:qemu_mprotect__osdep: \
  assertion failed: (!(size & ~qemu_real_host_page_mask))

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
 tcg/region.c | 47 +++++++++++++++++++++--------------------------
 1 file changed, 21 insertions(+), 26 deletions(-)

diff --git a/tcg/region.c b/tcg/region.c
index 27a7e35c8e..4dc1237ff4 100644
--- a/tcg/region.c
+++ b/tcg/region.c
@@ -469,26 +469,6 @@ static size_t tcg_n_regions(size_t tb_size, unsigned max_cpus)
   (DEFAULT_CODE_GEN_BUFFER_SIZE_1 < MAX_CODE_GEN_BUFFER_SIZE \
    ? DEFAULT_CODE_GEN_BUFFER_SIZE_1 : MAX_CODE_GEN_BUFFER_SIZE)
 
-static size_t size_code_gen_buffer(size_t tb_size)
-{
-    /* Size the buffer.  */
-    if (tb_size == 0) {
-        size_t phys_mem = qemu_get_host_physmem();
-        if (phys_mem == 0) {
-            tb_size = DEFAULT_CODE_GEN_BUFFER_SIZE;
-        } else {
-            tb_size = MIN(DEFAULT_CODE_GEN_BUFFER_SIZE, phys_mem / 8);
-        }
-    }
-    if (tb_size < MIN_CODE_GEN_BUFFER_SIZE) {
-        tb_size = MIN_CODE_GEN_BUFFER_SIZE;
-    }
-    if (tb_size > MAX_CODE_GEN_BUFFER_SIZE) {
-        tb_size = MAX_CODE_GEN_BUFFER_SIZE;
-    }
-    return tb_size;
-}
-
 #ifdef __mips__
 /* In order to use J and JAL within the code_gen_buffer, we require
    that the buffer not cross a 256MB boundary.  */
@@ -836,13 +816,29 @@ static int alloc_code_gen_buffer(size_t size, int splitwx, Error **errp)
  */
 void tcg_region_init(size_t tb_size, int splitwx, unsigned max_cpus)
 {
-    size_t page_size;
+    const size_t page_size = qemu_real_host_page_size;
     size_t region_size;
     size_t i;
     int have_prot;
 
-    have_prot = alloc_code_gen_buffer(size_code_gen_buffer(tb_size),
-                                      splitwx, &error_fatal);
+    /* Size the buffer.  */
+    if (tb_size == 0) {
+        size_t phys_mem = qemu_get_host_physmem();
+        if (phys_mem == 0) {
+            tb_size = DEFAULT_CODE_GEN_BUFFER_SIZE;
+        } else {
+            tb_size = QEMU_ALIGN_DOWN(phys_mem / 8, page_size);
+            tb_size = MIN(DEFAULT_CODE_GEN_BUFFER_SIZE, tb_size);
+        }
+    }
+    if (tb_size < MIN_CODE_GEN_BUFFER_SIZE) {
+        tb_size = MIN_CODE_GEN_BUFFER_SIZE;
+    }
+    if (tb_size > MAX_CODE_GEN_BUFFER_SIZE) {
+        tb_size = MAX_CODE_GEN_BUFFER_SIZE;
+    }
+
+    have_prot = alloc_code_gen_buffer(tb_size, splitwx, &error_fatal);
     assert(have_prot >= 0);
 
     /* Request large pages for the buffer and the splitwx.  */
@@ -857,9 +853,8 @@ void tcg_region_init(size_t tb_size, int splitwx, unsigned max_cpus)
      * As a result of this we might end up with a few extra pages at the end of
      * the buffer; we will assign those to the last region.
      */
-    region.n = tcg_n_regions(region.total_size, max_cpus);
-    page_size = qemu_real_host_page_size;
-    region_size = region.total_size / region.n;
+    region.n = tcg_n_regions(tb_size, max_cpus);
+    region_size = tb_size / region.n;
     region_size = QEMU_ALIGN_DOWN(region_size, page_size);
 
     /* A region must have at least 2 pages; one code, one guard */
-- 
2.25.1



  parent reply	other threads:[~2021-03-14 21:55 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-03-14 21:26 [PATCH v2 00/29] tcg: Workaround macOS 11.2 mprotect bug Richard Henderson
2021-03-14 21:26 ` [PATCH v2 01/29] meson: Split out tcg/meson.build Richard Henderson
2021-03-15 23:09   ` Roman Bolshakov
2021-03-14 21:26 ` [PATCH v2 02/29] meson: Split out fpu/meson.build Richard Henderson
2021-03-15 23:10   ` Roman Bolshakov
2021-03-14 21:26 ` [PATCH v2 03/29] tcg: Re-order tcg_region_init vs tcg_prologue_init Richard Henderson
2021-03-15 23:37   ` Roman Bolshakov
2021-03-16 14:57     ` Richard Henderson
2021-03-14 21:26 ` [PATCH v2 04/29] tcg: Remove error return from tcg_region_initial_alloc__locked Richard Henderson
2021-03-14 21:27 ` [PATCH v2 05/29] tcg: Split out tcg_region_initial_alloc Richard Henderson
2021-03-14 21:27 ` [PATCH v2 06/29] tcg: Split out tcg_region_prologue_set Richard Henderson
2021-03-14 21:27 ` [PATCH v2 07/29] tcg: Split out region.c Richard Henderson
2021-03-14 21:27 ` [PATCH v2 08/29] accel/tcg: Inline cpu_gen_init Richard Henderson
2021-03-14 21:27 ` [PATCH v2 09/29] accel/tcg: Move alloc_code_gen_buffer to tcg/region.c Richard Henderson
2021-03-14 21:27 ` [PATCH v2 10/29] accel/tcg: Rename tcg_init to tcg_init_machine Richard Henderson
2021-03-14 21:27 ` [PATCH v2 11/29] tcg: Create tcg_init Richard Henderson
2021-03-14 21:27 ` [PATCH v2 12/29] accel/tcg: Merge tcg_exec_init into tcg_init_machine Richard Henderson
2021-03-14 21:27 ` [PATCH v2 13/29] accel/tcg: Pass down max_cpus to tcg_init Richard Henderson
2021-03-14 21:27 ` [PATCH v2 14/29] tcg: Introduce tcg_max_ctxs Richard Henderson
2021-03-14 21:27 ` [PATCH v2 15/29] tcg: Move MAX_CODE_GEN_BUFFER_SIZE to tcg-target.h Richard Henderson
2021-03-14 21:27 ` [PATCH v2 16/29] tcg: Replace region.end with region.total_size Richard Henderson
2021-03-14 21:27 ` [PATCH v2 17/29] tcg: Rename region.start to region.after_prologue Richard Henderson
2021-03-14 21:27 ` [PATCH v2 18/29] tcg: Tidy tcg_n_regions Richard Henderson
2021-03-14 21:27 ` [PATCH v2 19/29] tcg: Tidy split_cross_256mb Richard Henderson
2021-03-14 21:27 ` [PATCH v2 20/29] tcg: Move in_code_gen_buffer and tests to region.c Richard Henderson
2021-03-14 21:27 ` [PATCH v2 21/29] tcg: Allocate code_gen_buffer into struct tcg_region_state Richard Henderson
2021-03-14 21:27 ` [PATCH v2 22/29] tcg: Return the map protection from alloc_code_gen_buffer Richard Henderson
2021-03-14 22:04   ` Philippe Mathieu-Daudé
2021-03-14 21:27 ` [PATCH v2 23/29] tcg: Sink qemu_madvise call to common code Richard Henderson
2021-03-14 21:27 ` [PATCH v2 24/29] tcg: Do not set guard pages in the rx buffer Richard Henderson
2021-03-14 21:27 ` [PATCH v2 25/29] util/osdep: Add qemu_mprotect_rw Richard Henderson
2021-03-14 21:27 ` Richard Henderson [this message]
2021-03-14 21:27 ` [PATCH v2 27/29] tcg: Merge buffer protection and guard page protection Richard Henderson
2021-03-14 21:27 ` [PATCH v2 28/29] tcg: When allocating for !splitwx, begin with PROT_NONE Richard Henderson
2021-03-14 21:27 ` [PATCH v2 29/29] tcg: Move tcg_init_ctx and tcg_ctx from accel/tcg/ Richard Henderson
2021-03-14 22:00   ` Philippe Mathieu-Daudé
2021-03-14 22:12 ` [PATCH v2 00/29] tcg: Workaround macOS 11.2 mprotect bug no-reply
2021-03-15 23:08 ` Roman Bolshakov

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=20210314212724.1917075-27-richard.henderson@linaro.org \
    --to=richard.henderson@linaro.org \
    --cc=j@getutm.app \
    --cc=qemu-devel@nongnu.org \
    --cc=r.bolshakov@yadro.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 NNTP newsgroup(s).