From: "Ján Stanček" <jan.stancek@gmail.com>
To: gdb-patches@sourceware.org
Subject: patch: fix stack unwind through uClibc syscall() on mips
Date: Sat, 27 Mar 2010 17:55:00 -0000 [thread overview]
Message-ID: <737ad3551003271055o91a78i3f5ff305b927e441@mail.gmail.com> (raw)
uClibc syscall() is macro which modifies stack before syscall
instruction, gdb is only looking at function prologue and misses the
stack modification made in syscall(). Because of this unwind doesn't
work. Attached is a patch, which is looking at actual $pc and $pc-4,
and in case of syscall it modifies $sp, so mip32_scan_prologue finds
correct values.
Description of bug is also available here:
http://www.listware.net/201003/gnu-gdb/26893.html
2010-03-27 Jan Stancek <jan.stancek@gmail.com>
* mips-tdep.c: fix stack unwind through uClibc syscall() on mips
--- cut ---
--- mips-tdep.c.orig 2010-03-27 17:04:57.000000000 +0100
+++ mips-tdep.c 2010-03-27 18:47:46.000000000 +0100
@@ -1895,6 +1895,44 @@ reset_saved_regs (struct gdbarch *gdbarc
}
}
+/*
+ * fix the $sp by looking around actual $pc
+ * Currently this handles only uClibc syscalls,
+ * which adjust $sp before syscall itsels
+ */
+int mips32_get_sp_adjustment(struct frame_info *this_frame, CORE_ADDR start_pc)
+{
+ CORE_ADDR pc = get_frame_address_in_block (this_frame);
+ struct gdbarch *gdbarch = get_frame_arch (this_frame);
+ unsigned long inst, high_word, low_word, ret;
+ int reg;
+
+ inst = (unsigned long) mips_fetch_instruction (gdbarch, pc);
+
+ ret = 0;
+ high_word = (inst >> 16) & 0xffff;
+ low_word = inst & 0xffff;
+ reg = high_word & 0x1f;
+
+ if (high_word == 0x27bd /* addiu $sp,$sp,-i */
+ || high_word == 0x23bd /* addi $sp,$sp,-i */
+ || high_word == 0x67bd) /* daddiu $sp,$sp,-i */
+ {
+ if ( reg == MIPS_SP_REGNUM
+ && (low_word & 0x8000) == 0 /* positive stack adjustment */
+ && (pc-4) > start_pc )
+ {
+ pc = pc - 4;
+ inst = (unsigned long) mips_fetch_instruction (gdbarch, pc);
+ if (inst==0x0000000c) /* syscall */
+ {
+ ret = low_word;
+ }
+ }
+ }
+ return ret;
+}
+
/* Analyze the function prologue from START_PC to LIMIT_PC. Builds
the associated FRAME_CACHE if not null.
Return the address of the first instruction past the prologue. */
@@ -1920,9 +1958,12 @@ mips32_scan_prologue (struct gdbarch *gd
/* Can be called when there's no process, and hence when there's no
THIS_FRAME. */
if (this_frame != NULL)
- sp = get_frame_register_signed (this_frame,
- gdbarch_num_regs (gdbarch)
- + MIPS_SP_REGNUM);
+ {
+ sp = get_frame_register_signed (this_frame,
+ gdbarch_num_regs (gdbarch)
+ + MIPS_SP_REGNUM);
+ sp += mips32_get_sp_adjustment(this_frame, start_pc);
+ }
else
sp = 0;
next reply other threads:[~2010-03-27 17:55 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-03-27 17:55 Ján Stanček [this message]
2010-04-05 15:45 ` Joel Brobecker
2010-04-06 18:55 ` Ján Stanček
2010-04-07 17:11 ` Joel Brobecker
2010-04-05 15:51 ` Daniel Jacobowitz
2010-04-06 20:03 ` Ján Stanček
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=737ad3551003271055o91a78i3f5ff305b927e441@mail.gmail.com \
--to=jan.stancek@gmail.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).