All of lore.kernel.org
 help / color / mirror / Atom feed
From: Will Deacon <will@kernel.org>
To: linux-arm-kernel@lists.infradead.org
Cc: Marc Zyngier <maz@kernel.org>, Will Deacon <will@kernel.org>
Subject: [PATCH] arm64: spectre-v2: Favour CPU-specific mitigation at EL2
Date: Wed, 21 Oct 2020 11:52:12 +0100	[thread overview]
Message-ID: <20201021105212.17634-1-will@kernel.org> (raw)

Spectre-v2 can be mitigated on Falkor CPUs either by calling into
firmware or by issuing a magic, CPU-specific sequence of branches.
Although the latter is faster, the size of the code sequence means that
it cannot be used in the EL2 vectors, and so there is a need for both
mitigations to co-exist in order to achieve optimal performance.

Change the mitigation selection logic for Spectre-v2 so that the
CPU-specific mitigation is used only when the firmware mitigation is
also available, rather than when a firmware mitigation is unavailable.

Cc: Marc Zyngier <maz@kernel.org>
Signed-off-by: Will Deacon <will@kernel.org>
---
 arch/arm64/kernel/proton-pack.c | 36 ++++++++++++++++-----------------
 1 file changed, 17 insertions(+), 19 deletions(-)

diff --git a/arch/arm64/kernel/proton-pack.c b/arch/arm64/kernel/proton-pack.c
index 68b710f1b43f..5029ef14fb27 100644
--- a/arch/arm64/kernel/proton-pack.c
+++ b/arch/arm64/kernel/proton-pack.c
@@ -67,7 +67,8 @@ ssize_t cpu_show_spectre_v1(struct device *dev, struct device_attribute *attr,
  * - Mitigated in hardware and advertised by ID_AA64PFR0_EL1.CSV2.
  * - Mitigated in hardware and listed in our "safe list".
  * - Mitigated in software by firmware.
- * - Mitigated in software by a CPU-specific dance in the kernel.
+ * - Mitigated in software by a CPU-specific dance in the kernel and a
+ *   firmware call at EL2.
  * - Vulnerable.
  *
  * It's not unlikely for different CPUs in a big.LITTLE system to fall into
@@ -259,6 +260,16 @@ static void qcom_link_stack_sanitisation(void)
 		     : "=&r" (tmp));
 }
 
+static bp_hardening_cb_t spectre_v2_get_sw_mitigation_cb(void)
+{
+	u32 midr = read_cpuid_id();
+	if (((midr & MIDR_CPU_MODEL_MASK) != MIDR_QCOM_FALKOR) &&
+	    ((midr & MIDR_CPU_MODEL_MASK) != MIDR_QCOM_FALKOR_V1))
+		return NULL;
+
+	return qcom_link_stack_sanitisation;
+}
+
 static enum mitigation_state spectre_v2_enable_fw_mitigation(void)
 {
 	bp_hardening_cb_t cb;
@@ -284,26 +295,15 @@ static enum mitigation_state spectre_v2_enable_fw_mitigation(void)
 		return SPECTRE_VULNERABLE;
 	}
 
+	/*
+	 * Prefer a CPU-specific workaround if it exists. Note that we
+	 * still rely on firmware for the mitigation at EL2.
+	 */
+	cb = spectre_v2_get_sw_mitigation_cb() ?: cb;
 	install_bp_hardening_cb(cb);
 	return SPECTRE_MITIGATED;
 }
 
-static enum mitigation_state spectre_v2_enable_sw_mitigation(void)
-{
-	u32 midr;
-
-	if (spectre_v2_mitigations_off())
-		return SPECTRE_VULNERABLE;
-
-	midr = read_cpuid_id();
-	if (((midr & MIDR_CPU_MODEL_MASK) != MIDR_QCOM_FALKOR) &&
-	    ((midr & MIDR_CPU_MODEL_MASK) != MIDR_QCOM_FALKOR_V1))
-		return SPECTRE_VULNERABLE;
-
-	install_bp_hardening_cb(qcom_link_stack_sanitisation);
-	return SPECTRE_MITIGATED;
-}
-
 void spectre_v2_enable_mitigation(const struct arm64_cpu_capabilities *__unused)
 {
 	enum mitigation_state state;
@@ -313,8 +313,6 @@ void spectre_v2_enable_mitigation(const struct arm64_cpu_capabilities *__unused)
 	state = spectre_v2_get_cpu_hw_mitigation_state();
 	if (state == SPECTRE_VULNERABLE)
 		state = spectre_v2_enable_fw_mitigation();
-	if (state == SPECTRE_VULNERABLE)
-		state = spectre_v2_enable_sw_mitigation();
 
 	update_mitigation_state(&spectre_v2_state, state);
 }
-- 
2.29.0.rc1.297.gfa9743e501-goog


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

             reply	other threads:[~2020-10-21 10:53 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-21 10:52 Will Deacon [this message]
2020-10-21 14:38 ` [PATCH] arm64: spectre-v2: Favour CPU-specific mitigation at EL2 Marc Zyngier

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=20201021105212.17634-1-will@kernel.org \
    --to=will@kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=maz@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.