From: Joel Brobecker <brobecker@adacore.com>
To: Pedro Alves <palves@redhat.com>
Cc: gdb-patches@sourceware.org
Subject: Re: [PATCH 3/7] procfs.c: Don't install a deprecated_xfer_memory method
Date: Mon, 24 Feb 2014 17:53:00 -0000 [thread overview]
Message-ID: <20140224175346.GB4154@adacore.com> (raw)
In-Reply-To: <53062F86.6020409@redhat.com>
[-- Attachment #1: Type: text/plain, Size: 2007 bytes --]
Hi Pedro,
> >>> 2014-02-19 Pedro Alves <palves@redhat.com>
> >>>
> >>> * procfs.c (procfs_target): Don't install procfs_xfer_memory as
> >>> deprecated_xfer_memory hook.
> >>> (procfs_xfer_partial): Call procfs_xfer_memory instead
> >>> of the deprecated_xfer_memory target hook.
> >>> (procfs_xfer_memory): Adjust interface as a to_xfer_partial
> >>> helper.
> >>
> >> FYI: I tried to test it, but the sparc-solaris debugger appears to be
> >> badly broken at the moment, and I won't have much time left this week
> >> to investigate the failures, so I'll schedule that for next week.
It turns out that your patch fixes the breakage as a side-effect.
Since it's bad breakage, I pushed the patch immediately, after having
updated the revision log:
This removes yet another instance of a deprecated_xfer_memory user,
and fixes a nasty regression as a side-effect:
(gdb) start
Temporary breakpoint 1 at 0x19070: file simple_main.adb, line 4.
Starting program: /[...]/simple_main
Warning:
Cannot insert breakpoint 1.
Cannot access memory at address 0x19070
Cannot insert breakpoint -3.
Temporarily disabling shared library breakpoints:
breakpoint #-3
The regression was introduced by the to_xfer_partial transition
to return a status enum. procfs_xfer_partial was updated but
not the case where object is TARGET_OBJECT_MEMORY. As result,
procfs_xfer_partial was returning the length xfered rather than
the status, and the xfered buffer was left uninitialized.
gdb/
2014-02-19 Pedro Alves <palves@redhat.com>
* procfs.c (procfs_target): Don't install procfs_xfer_memory as
deprecated_xfer_memory hook.
(procfs_xfer_partial): Call procfs_xfer_memory instead
of the deprecated_xfer_memory target hook.
(procfs_xfer_memory): Adjust interface as a to_xfer_partial
helper.
Tested on sparc-solaris, and pushed.
I still see one regression, which I will now investigate separately.
Thanks for the patch!
--
Joel
[-- Attachment #2: 0001-procfs.c-Don-t-install-a-deprecated_xfer_memory-meth.patch --]
[-- Type: text/x-diff, Size: 6131 bytes --]
From e96027e0d9b0a600aabd2c1c7cfdde6ce6d076e8 Mon Sep 17 00:00:00 2001
From: Pedro Alves <palves@redhat.com>
Date: Wed, 19 Feb 2014 20:29:31 +0000
Subject: [PATCH] procfs.c: Don't install a deprecated_xfer_memory method
This removes yet another instance of a deprecated_xfer_memory user,
and fixes a nasty regression as a side-effect:
(gdb) start
Temporary breakpoint 1 at 0x19070: file simple_main.adb, line 4.
Starting program: /[...]/simple_main
Warning:
Cannot insert breakpoint 1.
Cannot access memory at address 0x19070
Cannot insert breakpoint -3.
Temporarily disabling shared library breakpoints:
breakpoint #-3
The regression was introduced by the to_xfer_partial transition
to return a status enum. procfs_xfer_partial was updated but
not the case where object is TARGET_OBJECT_MEMORY. As result,
procfs_xfer_partial was returning the length xfered rather than
the status, and the xfered buffer was left uninitialized.
gdb/
2014-02-19 Pedro Alves <palves@redhat.com>
* procfs.c (procfs_target): Don't install procfs_xfer_memory as
deprecated_xfer_memory hook.
(procfs_xfer_partial): Call procfs_xfer_memory instead
of the deprecated_xfer_memory target hook.
(procfs_xfer_memory): Adjust interface as a to_xfer_partial
helper.
---
gdb/ChangeLog | 9 +++++++
gdb/procfs.c | 73 +++++++++++++++++++++------------------------------------
2 files changed, 36 insertions(+), 46 deletions(-)
diff --git a/gdb/ChangeLog b/gdb/ChangeLog
index bc8573c..b642ac9 100644
--- a/gdb/ChangeLog
+++ b/gdb/ChangeLog
@@ -1,3 +1,12 @@
+2014-02-24 Pedro Alves <palves@redhat.com>
+
+ * procfs.c (procfs_target): Don't install procfs_xfer_memory as
+ deprecated_xfer_memory hook.
+ (procfs_xfer_partial): Call procfs_xfer_memory instead
+ of the deprecated_xfer_memory target hook.
+ (procfs_xfer_memory): Adjust interface as a to_xfer_partial
+ helper.
+
2014-02-24 Yuanhui Zhang <asmwarrior@gmail.com>
* windows-nat.c (windows_xfer_shared_libraries): Return
diff --git a/gdb/procfs.c b/gdb/procfs.c
index 822e1e0..8204747 100644
--- a/gdb/procfs.c
+++ b/gdb/procfs.c
@@ -127,9 +127,10 @@ static void procfs_create_inferior (struct target_ops *, char *,
char *, char **, int);
static ptid_t procfs_wait (struct target_ops *,
ptid_t, struct target_waitstatus *, int);
-static int procfs_xfer_memory (CORE_ADDR, gdb_byte *, int, int,
- struct mem_attrib *attrib,
- struct target_ops *);
+static enum target_xfer_status procfs_xfer_memory (gdb_byte *,
+ const gdb_byte *,
+ ULONGEST, ULONGEST,
+ ULONGEST *);
static target_xfer_partial_ftype procfs_xfer_partial;
static int procfs_thread_alive (struct target_ops *ops, ptid_t);
@@ -197,7 +198,6 @@ procfs_target (void)
t->to_fetch_registers = procfs_fetch_registers;
t->to_store_registers = procfs_store_registers;
t->to_xfer_partial = procfs_xfer_partial;
- t->deprecated_xfer_memory = procfs_xfer_memory;
t->to_pass_signals = procfs_pass_signals;
t->to_files_info = procfs_files_info;
t->to_stop = procfs_stop;
@@ -3986,13 +3986,7 @@ procfs_xfer_partial (struct target_ops *ops, enum target_object object,
switch (object)
{
case TARGET_OBJECT_MEMORY:
- if (readbuf)
- return (*ops->deprecated_xfer_memory) (offset, readbuf,
- len, 0/*read*/, NULL, ops);
- if (writebuf)
- return (*ops->deprecated_xfer_memory) (offset, (gdb_byte *) writebuf,
- len, 1/*write*/, NULL, ops);
- return TARGET_XFER_E_IO;
+ return procfs_xfer_memory (readbuf, writebuf, offset, len, xfered_len);
#ifdef NEW_PROC_API
case TARGET_OBJECT_AUXV:
@@ -4009,23 +4003,15 @@ procfs_xfer_partial (struct target_ops *ops, enum target_object object,
}
}
+/* Helper for procfs_xfer_partial that handles memory transfers.
+ Arguments are like target_xfer_partial. */
-/* Transfer LEN bytes between GDB address MYADDR and target address
- MEMADDR. If DOWRITE is non-zero, transfer them to the target,
- otherwise transfer them from the target. TARGET is unused.
-
- The return value is 0 if an error occurred or no bytes were
- transferred. Otherwise, it will be a positive value which
- indicates the number of bytes transferred between gdb and the
- target. (Note that the interface also makes provisions for
- negative values, but this capability isn't implemented here.) */
-
-static int
-procfs_xfer_memory (CORE_ADDR memaddr, gdb_byte *myaddr, int len, int dowrite,
- struct mem_attrib *attrib, struct target_ops *target)
+static enum target_xfer_status
+procfs_xfer_memory (gdb_byte *readbuf, const gdb_byte *writebuf,
+ ULONGEST memaddr, ULONGEST len, ULONGEST *xfered_len)
{
procinfo *pi;
- int nbytes = 0;
+ int nbytes;
/* Find procinfo for main process. */
pi = find_procinfo_or_die (ptid_get_pid (inferior_ptid), 0);
@@ -4033,31 +4019,26 @@ procfs_xfer_memory (CORE_ADDR memaddr, gdb_byte *myaddr, int len, int dowrite,
open_procinfo_files (pi, FD_AS) == 0)
{
proc_warn (pi, "xfer_memory, open_proc_files", __LINE__);
- return 0;
+ return TARGET_XFER_E_IO;
}
- if (lseek (pi->as_fd, (off_t) memaddr, SEEK_SET) == (off_t) memaddr)
+ if (lseek (pi->as_fd, (off_t) memaddr, SEEK_SET) != (off_t) memaddr)
+ return TARGET_XFER_E_IO;
+
+ if (writebuf != NULL)
{
- if (dowrite)
- {
-#ifdef NEW_PROC_API
- PROCFS_NOTE ("write memory:\n");
-#else
- PROCFS_NOTE ("write memory:\n");
-#endif
- nbytes = write (pi->as_fd, myaddr, len);
- }
- else
- {
- PROCFS_NOTE ("read memory:\n");
- nbytes = read (pi->as_fd, myaddr, len);
- }
- if (nbytes < 0)
- {
- nbytes = 0;
- }
+ PROCFS_NOTE ("write memory:\n");
+ nbytes = write (pi->as_fd, writebuf, len);
+ }
+ else
+ {
+ PROCFS_NOTE ("read memory:\n");
+ nbytes = read (pi->as_fd, readbuf, len);
}
- return nbytes;
+ if (nbytes <= 0)
+ return TARGET_XFER_E_IO;
+ *xfered_len = nbytes;
+ return TARGET_XFER_OK;
}
/* Called by target_resume before making child runnable. Mark cached
--
1.7.9.5
next prev parent reply other threads:[~2014-02-24 17:53 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-02-19 20:29 [PATCH 0/7] eliminate deprecated_xfer_memory Pedro Alves
2014-02-19 20:29 ` [PATCH 1/7] remote-mips.c: Don't install a deprecated_xfer_memory method Pedro Alves
2014-02-19 20:29 ` [PATCH 7/7] bsd-uthread.c: Don't install a to_xfer_partial method Pedro Alves
2014-02-20 13:26 ` Mark Kettenis
2014-02-20 16:42 ` Pedro Alves
2014-02-19 20:29 ` [PATCH 6/7] eliminate target_ops->deprecated_xfer_memory Pedro Alves
2014-02-19 20:29 ` [PATCH 5/7] go32-nat.c: Don't install a deprecated_xfer_memory method Pedro Alves
[not found] ` <00c801cf2e59$acae60f0$060b22d0$@muller@ics-cnrs.unistra.fr>
2014-02-20 16:37 ` Pedro Alves
2014-02-19 20:29 ` [PATCH 2/7] remote-m32r-sdi.c: " Pedro Alves
2014-02-19 20:29 ` [PATCH 3/7] procfs.c: " Pedro Alves
2014-02-20 15:48 ` Joel Brobecker
2014-02-20 15:56 ` Joel Brobecker
2014-02-20 16:38 ` Pedro Alves
2014-02-24 17:53 ` Joel Brobecker [this message]
2014-02-24 18:21 ` Pedro Alves
2014-02-24 18:37 ` Joel Brobecker
2014-02-19 20:29 ` [PATCH 4/7] nto-procfs.c: " Pedro Alves
2014-02-20 18:27 ` [PATCH 0/7] eliminate deprecated_xfer_memory Tom Tromey
2014-02-26 14:50 ` Pedro Alves
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=20140224175346.GB4154@adacore.com \
--to=brobecker@adacore.com \
--cc=gdb-patches@sourceware.org \
--cc=palves@redhat.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