cxl fixes for 6.2-rc6

- Fix a crash regression due to module load order of cxl_pmem.ko
 
 - Fix wrong register offset read in CXL AER handling path
 -----BEGIN PGP SIGNATURE-----
 
 iHUEABYKAB0WIQSbo+XnGs+rwLz9XGXfioYZHlFsZwUCY9W0XwAKCRDfioYZHlFs
 Z9y9AQDQXnlfFuVrSZ27lM53U+ZKQJWXV8SnPlcvbWbnVJb5owEA9J5p5ZCT0mHq
 sRFVgJoAlU73p39lcJoX8/ZP6OI6sgU=
 =bzei
 -----END PGP SIGNATURE-----

Merge tag 'cxl-fixes-for-6.2-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/cxl/cxl

Pull cxl fixes from Dan Williams:
 "A couple of fixes for bugs introduced during the merge window. One is
  a regression, the other was a bug in the CXL AER handler:

   - Fix a crash regression due to module load order of cxl_pmem.ko

   - Fix wrong register offset read in CXL AER handling path"

* tag 'cxl-fixes-for-6.2-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/cxl/cxl:
  cxl/pmem: Fix nvdimm unregistration when cxl_pmem driver is absent
  cxl: fix cxl_report_and_clear() RAS UE addr mis-assignment

drivers/cxl/acpi.c

@@ -736,4 +736,3 @@ module_exit(cxl_acpi_exit);
 MODULE_LICENSE("GPL v2");
 MODULE_IMPORT_NS(CXL);
 MODULE_IMPORT_NS(ACPI);
-MODULE_SOFTDEP("pre: cxl_pmem");

drivers/cxl/core/pmem.c

@@ -227,34 +227,16 @@ static struct cxl_nvdimm *cxl_nvdimm_alloc(struct cxl_nvdimm_bridge *cxl_nvb,
 	return cxl_nvd;
 }
 
-static void cxl_nvd_unregister(void *_cxl_nvd)
-{
-	struct cxl_nvdimm *cxl_nvd = _cxl_nvd;
-	struct cxl_memdev *cxlmd = cxl_nvd->cxlmd;
-	struct cxl_nvdimm_bridge *cxl_nvb = cxlmd->cxl_nvb;
-
-	/*
-	 * Either the bridge is in ->remove() context under the device_lock(),
-	 * or cxlmd_release_nvdimm() is cancelling the bridge's release action
-	 * for @cxl_nvd and doing it itself (while manually holding the bridge
-	 * lock).
-	 */
-	device_lock_assert(&cxl_nvb->dev);
-	cxl_nvd->cxlmd = NULL;
-	cxlmd->cxl_nvd = NULL;
-	device_unregister(&cxl_nvd->dev);
-}
-
 static void cxlmd_release_nvdimm(void *_cxlmd)
 {
 	struct cxl_memdev *cxlmd = _cxlmd;
+	struct cxl_nvdimm *cxl_nvd = cxlmd->cxl_nvd;
 	struct cxl_nvdimm_bridge *cxl_nvb = cxlmd->cxl_nvb;
 
-	device_lock(&cxl_nvb->dev);
+	cxl_nvd->cxlmd = NULL;
-	if (cxlmd->cxl_nvd)
+	cxlmd->cxl_nvd = NULL;
-		devm_release_action(&cxl_nvb->dev, cxl_nvd_unregister,
+	cxlmd->cxl_nvb = NULL;
-				    cxlmd->cxl_nvd);
+	device_unregister(&cxl_nvd->dev);
-	device_unlock(&cxl_nvb->dev);
 	put_device(&cxl_nvb->dev);
 }
 
@@ -293,22 +275,6 @@ int devm_cxl_add_nvdimm(struct cxl_memdev *cxlmd)
 
 	dev_dbg(&cxlmd->dev, "register %s\n", dev_name(dev));
 
-	/*
-	 * The two actions below arrange for @cxl_nvd to be deleted when either
-	 * the top-level PMEM bridge goes down, or the endpoint device goes
-	 * through ->remove().
-	 */
-	device_lock(&cxl_nvb->dev);
-	if (cxl_nvb->dev.driver)
-		rc = devm_add_action_or_reset(&cxl_nvb->dev, cxl_nvd_unregister,
-					      cxl_nvd);
-	else
-		rc = -ENXIO;
-	device_unlock(&cxl_nvb->dev);
-
-	if (rc)
-		goto err_alloc;
-
 	/* @cxlmd carries a reference on @cxl_nvb until cxlmd_release_nvdimm */
 	return devm_add_action_or_reset(&cxlmd->dev, cxlmd_release_nvdimm, cxlmd);
 

drivers/cxl/pci.c

@@ -554,8 +554,11 @@ static bool cxl_report_and_clear(struct cxl_dev_state *cxlds)
 
 	/* If multiple errors, log header points to first error from ctrl reg */
 	if (hweight32(status) > 1) {
-		addr = cxlds->regs.ras + CXL_RAS_CAP_CONTROL_OFFSET;
+		void __iomem *rcc_addr =
-		fe = BIT(FIELD_GET(CXL_RAS_CAP_CONTROL_FE_MASK, readl(addr)));
+			cxlds->regs.ras + CXL_RAS_CAP_CONTROL_OFFSET;
+
+		fe = BIT(FIELD_GET(CXL_RAS_CAP_CONTROL_FE_MASK,
+				   readl(rcc_addr)));
 	} else {
 		fe = status;
 	}

drivers/cxl/pmem.c

@@ -225,11 +225,35 @@ static int cxl_pmem_ctl(struct nvdimm_bus_descriptor *nd_desc,
 	return cxl_pmem_nvdimm_ctl(nvdimm, cmd, buf, buf_len);
 }
 
+static int detach_nvdimm(struct device *dev, void *data)
+{
+	struct cxl_nvdimm *cxl_nvd;
+	bool release = false;
+
+	if (!is_cxl_nvdimm(dev))
+		return 0;
+
+	device_lock(dev);
+	if (!dev->driver)
+		goto out;
+
+	cxl_nvd = to_cxl_nvdimm(dev);
+	if (cxl_nvd->cxlmd && cxl_nvd->cxlmd->cxl_nvb == data)
+		release = true;
+out:
+	device_unlock(dev);
+	if (release)
+		device_release_driver(dev);
+	return 0;
+}
+
 static void unregister_nvdimm_bus(void *_cxl_nvb)
 {
 	struct cxl_nvdimm_bridge *cxl_nvb = _cxl_nvb;
 	struct nvdimm_bus *nvdimm_bus = cxl_nvb->nvdimm_bus;
 
+	bus_for_each_dev(&cxl_bus_type, NULL, cxl_nvb, detach_nvdimm);
+
 	cxl_nvb->nvdimm_bus = NULL;
 	nvdimm_bus_unregister(nvdimm_bus);
 }