public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Tom Tromey <tom@tromey.com>
To: gdb-patches@sourceware.org
Subject: [PATCH 2/5] Capture directory in index_cache_store_context
Date: Sun, 28 Jan 2024 09:28:06 -0700	[thread overview]
Message-ID: <20240128-pr-31262-index-cache-race-v1-2-4fe53c5265e3@tromey.com> (raw)
In-Reply-To: <20240128-pr-31262-index-cache-race-v1-0-4fe53c5265e3@tromey.com>

I noticed that index_cache_store_context captures the 'enabled'
setting, but not the index cache directory.  This patch makes this
change, which avoids a possible race -- with background reading, the
user could possibly change this directory at the exact moment the
writer examines the variable.
---
 gdb/dwarf2/index-cache.c | 9 +++++----
 gdb/dwarf2/index-cache.h | 3 +++
 2 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/gdb/dwarf2/index-cache.c b/gdb/dwarf2/index-cache.c
index 645e36c66f7..a269eb46a36 100644
--- a/gdb/dwarf2/index-cache.c
+++ b/gdb/dwarf2/index-cache.c
@@ -90,7 +90,8 @@ index_cache::disable ()
 
 index_cache_store_context::index_cache_store_context (const index_cache &ic,
 						      dwarf2_per_bfd *per_bfd)
-  :  m_enabled (ic.enabled ())
+  :  m_enabled (ic.enabled ()),
+     m_dir (ic.m_dir)
 {
   if (!m_enabled)
     return;
@@ -124,7 +125,7 @@ index_cache_store_context::index_cache_store_context (const index_cache &ic,
       m_dwz_build_id_str = build_id_to_string (dwz_build_id);
     }
 
-  if (ic.m_dir.empty ())
+  if (m_dir.empty ())
     {
       warning (_("The index cache directory name is empty, skipping store."));
       m_enabled = false;
@@ -134,7 +135,7 @@ index_cache_store_context::index_cache_store_context (const index_cache &ic,
   try
     {
       /* Try to create the containing directory.  */
-      if (!mkdir_recursive (ic.m_dir.c_str ()))
+      if (!mkdir_recursive (m_dir.c_str ()))
 	{
 	  warning (_("index cache: could not make cache directory: %s"),
 		   safe_strerror (errno));
@@ -170,7 +171,7 @@ index_cache::store (dwarf2_per_bfd *per_bfd,
 
       /* Write the index itself to the directory, using the build id as the
 	 filename.  */
-      write_dwarf_index (per_bfd, m_dir.c_str (),
+      write_dwarf_index (per_bfd, ctx.m_dir.c_str (),
 			 ctx.m_build_id_str.c_str (), dwz_build_id_ptr,
 			 dw_index_kind::GDB_INDEX);
     }
diff --git a/gdb/dwarf2/index-cache.h b/gdb/dwarf2/index-cache.h
index 079ed412d2e..f66f72caea1 100644
--- a/gdb/dwarf2/index-cache.h
+++ b/gdb/dwarf2/index-cache.h
@@ -48,6 +48,9 @@ struct index_cache_store_context
   /* Captured value of enabled ().  */
   bool m_enabled;
 
+  /* Captured value of index cache directory.  */
+  std::string m_dir;
+
   /* Captured value of build id.  */
   std::string m_build_id_str;
 

-- 
2.43.0


  parent reply	other threads:[~2024-01-28 16:28 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-01-28 16:28 [PATCH 0/5] Fix race when writing to index cache Tom Tromey
2024-01-28 16:28 ` [PATCH 1/5] Rename members of index_cache_store_context Tom Tromey
2024-01-28 16:28 ` Tom Tromey [this message]
2024-01-28 16:28 ` [PATCH 3/5] Capture the per-BFD object in index_cache_store_context Tom Tromey
2024-01-28 16:28 ` [PATCH 4/5] Move the 'store' method to index_cache_store_context Tom Tromey
2024-01-28 16:28 ` [PATCH 5/5] Avoid race when writing to index cache Tom Tromey
2024-03-09  0:25 ` [PATCH 0/5] Fix " Tom Tromey

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=20240128-pr-31262-index-cache-race-v1-2-4fe53c5265e3@tromey.com \
    --to=tom@tromey.com \
    --cc=gdb-patches@sourceware.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 read-only IMAP folder(s) and NNTP newsgroup(s).