public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Oleg Nesterov <oleg@redhat.com>
To: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jacob Shin <jacob.shin@amd.com>, Ingo Molnar <mingo@redhat.com>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Arnaldo Carvalho de Melo <acme@ghostprotocols.net>,
	"H. Peter Anvin" <hpa@zytor.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	x86@kernel.org, Stephane Eranian <eranian@google.com>,
	Jiri Olsa <jolsa@redhat.com>,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH V2 1/4] perf: Add hardware breakpoint address mask
Date: Thu, 25 Apr 2013 17:57:20 +0200	[thread overview]
Message-ID: <20130425155720.GA30844@redhat.com> (raw)
In-Reply-To: <20130425151035.GA26760@redhat.com>

On 04/25, Oleg Nesterov wrote:
>
> On 04/25, Frederic Weisbecker wrote:
> >
> > 2013/4/23 Jacob Shin <jacob.shin@amd.com>:
> > > @@ -286,7 +286,10 @@ struct perf_event_attr {
> > >                 __u64           config1; /* extension of config */
> > >         };
> > >         union {
> > > -               __u64           bp_len;
> > > +               struct {
> > > +                       __u32           bp_len;
> > > +                       __u32           bp_addr_mask;
> > > +               };
> >
> > Do we need len and mask to work at the same time? I can't think of a
> > situation when len and mask mix up together in a useful way to define
> > a range.
>
> And it would be nice (I think) if we could simply turn bp_len into
> bp_mask. It is already the mask actually, bp_addr should be aligned.
>
> But I do not see how we can do this, so I guess we need another field.
>
> Well. Another option is to extend bp_len. Fortunately HW_BREAKPOINT_LEN_*
> match the length, so we can simply allow any 2^n length and amd.c can
> translate it into the mask.
>
> Of course, this doesn't allow to use, say, mask=0xF0. But perhaps this
> is not really useful?
>
> I dunno. I leave this to you and Jacob ;)

IOW, I meant something like the incomplete patch below. But let me
repeat I am not sure this is the good idea.

Hmm. And note the change arch_check_bp_in_kernelspace(). Whatever we
do it should be updated if we have a mask...

And this reminds me that arch_check_bp_in_kernelspace() is buggy but
my trivial fix was ignored ;) see http://marc.info/?t=135248593000001

Oleg.

--- x/arch/x86/include/asm/hw_breakpoint.h
+++ x/arch/x86/include/asm/hw_breakpoint.h
@@ -12,6 +12,7 @@
  */
 struct arch_hw_breakpoint {
 	unsigned long	address;
+	unsigned long	mask;
 	u8		len;
 	u8		type;
 };
--- x/arch/x86/kernel/hw_breakpoint.c
+++ x/arch/x86/kernel/hw_breakpoint.c
@@ -128,6 +128,8 @@ int arch_install_hw_breakpoint(struct pe
 	*dr7 |= encode_dr7(i, info->len, info->type);
 
 	set_debugreg(*dr7, 7);
+	if (info->mask)
+		set_dr_addr_mask(...);
 
 	return 0;
 }
@@ -165,29 +167,6 @@ void arch_uninstall_hw_breakpoint(struct
 	set_debugreg(*dr7, 7);
 }
 
-static int get_hbp_len(u8 hbp_len)
-{
-	unsigned int len_in_bytes = 0;
-
-	switch (hbp_len) {
-	case X86_BREAKPOINT_LEN_1:
-		len_in_bytes = 1;
-		break;
-	case X86_BREAKPOINT_LEN_2:
-		len_in_bytes = 2;
-		break;
-	case X86_BREAKPOINT_LEN_4:
-		len_in_bytes = 4;
-		break;
-#ifdef CONFIG_X86_64
-	case X86_BREAKPOINT_LEN_8:
-		len_in_bytes = 8;
-		break;
-#endif
-	}
-	return len_in_bytes;
-}
-
 /*
  * Check for virtual address in kernel space.
  */
@@ -198,7 +177,7 @@ int arch_check_bp_in_kernelspace(struct 
 	struct arch_hw_breakpoint *info = counter_arch_bp(bp);
 
 	va = info->address;
-	len = get_hbp_len(info->len);
+	len = bp->attr.bp_len;
 
 	return (va >= TASK_SIZE) && ((va + len - 1) >= TASK_SIZE);
 }
@@ -278,7 +257,8 @@ static int arch_build_bp_info(struct per
 		return -EINVAL;
 	}
 
-	/* Len */
+	/* info->len == info->mask == 0 */
+
 	switch (bp->attr.bp_len) {
 	case HW_BREAKPOINT_LEN_1:
 		info->len = X86_BREAKPOINT_LEN_1;
@@ -295,7 +275,10 @@ static int arch_build_bp_info(struct per
 		break;
 #endif
 	default:
-		return -EINVAL;
+		if (!is_power_of_2(bp->attr.bp_len))
+			return -EINVAL;
+		info->mask = bp->attr.bp_len - 1;
+		info->len = X86_BREAKPOINT_LEN_1;
 	}
 
 	return 0;
@@ -314,11 +297,15 @@ int arch_validate_hwbkpt_settings(struct
 	if (ret)
 		return ret;
 
-	ret = -EINVAL;
-
 	switch (info->len) {
 	case X86_BREAKPOINT_LEN_1:
 		align = 0;
+		if (info->mask) {
+			align = info->mask;
+			ret = arch_validate_hwbkpt_addr_mask(...);
+			if (ret)
+				return ret;
+		}
 		break;
 	case X86_BREAKPOINT_LEN_2:
 		align = 1;
@@ -332,7 +319,7 @@ int arch_validate_hwbkpt_settings(struct
 		break;
 #endif
 	default:
-		return ret;
+		BUG();
 	}
 
 	/*


  reply	other threads:[~2013-04-25 16:24 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-04-23  7:57 [PATCH V2 0/4] perf: Add support for hardware breakpoint address masks Jacob Shin
2013-04-23  7:57 ` [PATCH V2 1/4] perf: Add hardware breakpoint address mask Jacob Shin
2013-04-23  9:54   ` Will Deacon
2013-04-23 14:34     ` Jacob Shin
2013-04-23 14:40       ` Jacob Shin
2013-04-23 15:02         ` Will Deacon
2013-04-23 15:18           ` Jacob Shin
2013-04-24  9:48             ` Will Deacon
2013-04-24 16:30               ` Jacob Shin
2013-04-25 17:06                 ` Oleg Nesterov
2013-04-25 17:17                   ` H. Peter Anvin
2013-04-25 23:19                     ` Jacob Shin
2013-04-26 16:20                       ` Will Deacon
2013-04-26 16:31                         ` Jacob Shin
2013-04-26 16:47                       ` Oleg Nesterov
2013-04-23 13:18   ` Oleg Nesterov
2013-04-23 14:25     ` Jacob Shin
2013-04-24 23:08       ` Frederic Weisbecker
2013-04-25 15:10         ` Oleg Nesterov
2013-04-25 15:57           ` Oleg Nesterov [this message]
2013-04-25 16:59           ` Jacob Shin
2013-04-25 17:33             ` Oleg Nesterov
2013-04-23  7:57 ` [PATCH V2 2/4] perf/x86/amd: AMD implementation for " Jacob Shin
2013-04-23 13:22   ` Oleg Nesterov
2013-04-23  7:57 ` [PATCH V2 3/4] perf tools: Add hardware breakpoint address mask event parser Jacob Shin
2013-04-23  7:57 ` [PATCH V2 4/4] perf tools: Add hardware breakpoint address mask test cases Jacob Shin

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=20130425155720.GA30844@redhat.com \
    --to=oleg@redhat.com \
    --cc=a.p.zijlstra@chello.nl \
    --cc=acme@ghostprotocols.net \
    --cc=eranian@google.com \
    --cc=fweisbec@gmail.com \
    --cc=hpa@zytor.com \
    --cc=jacob.shin@amd.com \
    --cc=jolsa@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=tglx@linutronix.de \
    --cc=x86@kernel.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