From: Frederic Weisbecker <fweisbec@gmail.com>
To: LKML <linux-kernel@vger.kernel.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>,
"Paul E . McKenney" <paulmck@linux.vnet.ibm.com>
Subject: [PATCH 5/5] x86: Support reverse execution
Date: Wed, 1 Apr 2015 01:20:36 +0200 [thread overview]
Message-ID: <1427844036-1325-6-git-send-email-fweisbec@gmail.com> (raw)
In-Reply-To: <1427844036-1325-1-git-send-email-fweisbec@gmail.com>
Support reverse execution for negative number of CPUs. We might be able
to implement that deeper with the function tracer.
No-Yet-Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
---
arch/x86/kernel/cpu/negative.c | 28 ++++++++++++++++++++++++++++
arch/x86/kernel/head64.c | 4 ++++
2 files changed, 32 insertions(+)
create mode 100644 arch/x86/kernel/cpu/negative.c
diff --git a/arch/x86/kernel/cpu/negative.c b/arch/x86/kernel/cpu/negative.c
new file mode 100644
index 0000000..3aff473
--- /dev/null
+++ b/arch/x86/kernel/cpu/negative.c
@@ -0,0 +1,28 @@
+/* Execute function in reverse for negative CPUs */
+
+#define _ASM_RET 0xc3
+#define FUNC_MAX_SIZE 4096
+static DEFINE_PER_CPU(char, reverse_func_buf[FUNC_MAX_SIZE]);
+
+void execute_reverse_function(void *func)
+{
+ char *opcode = (char *)func;
+ int i = 0, j = FUNC_MAX_SIZE - 1;
+ void (*reverse_func)(void);
+
+ /*
+ * Reverse copy the function.
+ * This assumes that every opcode is sizeof(char),
+ * there might be a few situations where that won't work
+ * but we'll fix them as soon as they get reported by users.
+ */
+ while (1) {
+ reverse_func_buf[j--] = opcode[i];
+ if (opcode[i++] == _ASM_RET)
+ break;
+ }
+
+ reverse_func = &reverse_func_buf[FUNC_MAX_SIZE - 1];
+ reverse_func();
+}
+
diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index c4f8d46..9ed2c88 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -197,5 +197,9 @@ void __init x86_64_start_reservations(char *real_mode_data)
reserve_ebda_region();
+#if CONFIG_NR_CPUS >= 0
start_kernel();
+#else
+ execute_reverse_function(start_kernel);
+#endif
}
--
2.1.4
next prev parent reply other threads:[~2015-03-31 23:21 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-03-31 23:20 [PATCH 0/5] Support negative number of CPUs Frederic Weisbecker
2015-03-31 23:20 ` [PATCH 1/5] cpu: Infrastructure for negative cpu handling Frederic Weisbecker
2015-03-31 23:20 ` [PATCH 2/5] smp: IPI handling for negative CPU Frederic Weisbecker
2015-03-31 23:20 ` [PATCH 3/5] cpumask: Basic negative number of CPUs handling Frederic Weisbecker
2015-03-31 23:20 ` [PATCH 4/5] init: Support negative CPUs boot and halt code Frederic Weisbecker
2015-03-31 23:20 ` Frederic Weisbecker [this message]
2015-04-01 7:35 ` [PATCH 0/5] Support negative number of CPUs Paul E. McKenney
2015-04-01 9:37 ` Borislav Petkov
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=1427844036-1325-6-git-send-email-fweisbec@gmail.com \
--to=fweisbec@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=paulmck@linux.vnet.ibm.com \
/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 NNTP newsgroup(s).