|To:||Himanshu Chauhan <firstname.lastname@example.org>|
|Subject:||Re: [PATCH] MIPS: KProbes support v0.1|
|From:||David Daney <email@example.com>|
|Date:||Thu, 10 Jun 2010 17:12:54 -0700|
|Cc:||firstname.lastname@example.org, email@example.com, firstname.lastname@example.org|
|References:||<email@example.com> <firstname.lastname@example.org> <4C0D4B82.email@example.com> <20100608175118.GA2262@hschauhan-desktop>|
|User-agent:||Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:18.104.22.168) Gecko/20100430 Fedora/3.0.4-2.fc12 Thunderbird/3.0.4|
On 06/08/2010 10:51 AM, Himanshu Chauhan wrote:
Hi David, Thanks for taking a look.
+ +#define BREAKPOINT_INSTRUCTION 0x0000000d + +/* + * We do not have hardware single-stepping on MIPS. + * So we implement software single-stepping with breakpoint + * trap 'break 5'. + */ +#define BREAKPOINT_INSTRUCTION_2 0x0000014dThe BREAK codes are defined in asm/break.h This should be added there instead. Why do you use codes (0 and 5) that are already kind of reserved for user space debuggers?As said ealier, this patch was based on some very older patch of 2.6.16 from Sony Corp, I didn't make much changes like this. But anyways, I wan't aware of this either. What would be the best code then?
How about allocating them after BRK_MEMU? Say 515 and 516 or something like that.
|<Prev in Thread]||Current Thread||[Next in Thread>|
|Previous by Date:||Re: [PATCH 7/9] Removing dead CONFIG_PMCTWILED, Shane McDonald|
|Next by Date:||[no subject], Jabir M|
|Previous by Thread:||Re: [PATCH] MIPS: KProbes support v0.1, Himanshu Chauhan|
|Next by Thread:||[PATCH] MIPS: Add NVRAM support for bcm47xx devices, Waldemar Brodkorb|
|Indexes:||[Date] [Thread] [Top] [All Lists]|