| To: | Daniel Jacobowitz <dan@debian.org> |
|---|---|
| Subject: | Re: [Patch 6/6] Ptrace support for HARDWARE_WATCHPOINTS. |
| From: | "Maciej W. Rozycki" <macro@linux-mips.org> |
| Date: | Tue, 22 Apr 2008 19:26:54 +0100 (BST) |
| Cc: | David Daney <ddaney@avtrex.com>, linux-mips@linux-mips.org, linux-kernel@vger.kernel.org |
| In-reply-to: | <20080422162343.GA14790@caradoc.them.org> |
| Original-recipient: | rfc822;linux-mips@linux-mips.org |
| References: | <480D2151.2020701@avtrex.com> <480D33EB.30808@avtrex.com> <20080422162343.GA14790@caradoc.them.org> |
| Sender: | linux-mips-bounce@linux-mips.org |
On Tue, 22 Apr 2008, Daniel Jacobowitz wrote:
> On Mon, Apr 21, 2008 at 05:40:11PM -0700, David Daney wrote:
> > +struct mips32_watch_regs {
> > + unsigned int num_valid;
> > + unsigned int reg_mask;
> > + unsigned int irw_mask;
> > + unsigned long watchlo[8];
> > + unsigned int watchhi[8];
> > +};
>
> Please do not use long in new ptrace interfaces. Use either
> uint32_t or uint64_t as appropriate so that it doesn't depend
> on how the kernel or debugger was built.
A minor nit from my side too -- since it is a new API it may be wise to
move wider structure members to the beginning not to waste gaps in memory
due to alignment.
Maciej
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: [Patch 6/6] Ptrace support for HARDWARE_WATCHPOINTS., Daniel Jacobowitz |
|---|---|
| Next by Date: | Alchemy: kill unused PCI_IRQ_TABLE_LOOKUP macro, Sergei Shtylyov |
| Previous by Thread: | Re: [Patch 6/6] Ptrace support for HARDWARE_WATCHPOINTS., Daniel Jacobowitz |
| Next by Thread: | [PATCH] rbtx4938: minor cleanup, Atsushi Nemoto |
| Indexes: | [Date] [Thread] [Top] [All Lists] |