This is the mail archive of the gdb-testers@sourceware.cygnus.com mailing list for the GDB project. See the GDB home page for more information.
| Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
|---|---|---|
| Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
In message <m10In2q-00000YC@ocean.lucon.org>you write: > > > There are several problems. A very annoying one: > > > > > > 1. Create 4 watchponts watching 4 different values. > > > 2. Check if they are hardware and they work. > > > 3. Delete all 4 watchponts. > > > 4. Create another 4 watchponts watching 4 different values. > > > 5. Check if they are hardware and they work. > > So, submit a patch for this specific problem and follow it through until > the > > issue is resolved. That way the software actually improves over time. > > > > I don't think it is possible. There are so many problems with the > hardware watchpoints on x86. I wound up almost rewriting most part > of the hardware watchpoints for x86. Only things left are the data > structures and some basic contol functions. The patch I send will > fix many problems. I am enclosing the incomplete ChangeLog entry for > my x86 hardware watchpoint changes. Feel free to contact me if you > want to go further. I'm not involved in reviewing gdb code anymore, so I can't strictly speak for what can and can not be included. However, like your egcs work, you've failed to actually include the patch to fix the problem. If you fail to include the patch, then there's little chance it will get reviewed, must less installed. It may be the case that the patch is mostly OK and would just need minor work, or it may be totally wrong. But since you didn't include the patch nobody actually knows. Again, I recommend you extract the changes and submit them. Just sending the ChangeLog entry is almost useless and does absolutely nothing to improve GNU software. jeff