* [PATCH v3 0/4] hw/cxl: Line length reduction and related
@ 2023-09-19 9:34 Jonathan Cameron via
2023-09-19 9:34 ` [PATCH v3 1/4] hw/cxl: Use a switch to explicitly check size in caps_reg_read() Jonathan Cameron via
` (3 more replies)
0 siblings, 4 replies; 10+ messages in thread
From: Jonathan Cameron via @ 2023-09-19 9:34 UTC (permalink / raw)
To: Michael Tokarev, qemu-devel, Michael Tsirkin, Fan Ni,
Philippe Mathieu-Daudé
Cc: linuxarm
v3:
- Fix an odd spacing change that Fan Ni noticed in review.
- Picked up tags.
Suggested-by: Michael S. Tsirkin <mst@redhat.com>
Michael observed that the CXL code regularly went above the 80 character
recommendation and in many cases this was not necessary for readability.
This series is focused on tidying this up for the existing code so that
we can maintain the preferred formatting going forwards.
Based on: [PATCH 0/4] hw/cxl: Minor CXL emulation fixes and cleanup
Based on: [PATCH v2 0/3] hw/cxl: Add dummy ACPI QTG DSM
Based-on: [PATCH V2] hw/pci-bridge/cxl-upstream: Add serial number extended capability support
Based on: Message ID: 20230904132806.6094-1-Jonathan.Cameron@huawei.com
Based on: Message ID: 20230904161847.18468-1-Jonathan.Cameron@huawei.com
Based on: Message ID: 20230913133615.29876-1-Jonathan.Cameron@huawei.com
Jonathan Cameron (4):
hw/cxl: Use a switch to explicitly check size in caps_reg_read()
hw/cxl: Use switch statements for read and write of cachemem registers
hw/cxl: CXLDVSECPortExtensions renamed to CXLDVSECPortExt
hw/cxl: Line length reductions
include/hw/cxl/cxl_component.h | 3 +-
include/hw/cxl/cxl_device.h | 5 +-
include/hw/cxl/cxl_events.h | 3 +-
include/hw/cxl/cxl_pci.h | 6 +-
hw/cxl/cxl-cdat.c | 3 +-
hw/cxl/cxl-component-utils.c | 128 ++++++++++++++++++++-------------
hw/cxl/cxl-device-utils.c | 11 +--
hw/cxl/cxl-events.c | 9 ++-
hw/cxl/cxl-mailbox-utils.c | 21 ++++--
hw/mem/cxl_type3.c | 31 ++++----
hw/mem/cxl_type3_stubs.c | 5 +-
hw/pci-bridge/cxl_downstream.c | 2 +-
hw/pci-bridge/cxl_root_port.c | 2 +-
hw/pci-bridge/cxl_upstream.c | 2 +-
14 files changed, 143 insertions(+), 88 deletions(-)
--
2.39.2
^ permalink raw reply [flat|nested] 10+ messages in thread
* [PATCH v3 1/4] hw/cxl: Use a switch to explicitly check size in caps_reg_read()
2023-09-19 9:34 [PATCH v3 0/4] hw/cxl: Line length reduction and related Jonathan Cameron via
@ 2023-09-19 9:34 ` Jonathan Cameron via
2023-09-19 9:34 ` [PATCH v3 2/4] hw/cxl: Use switch statements for read and write of cachemem registers Jonathan Cameron via
` (2 subsequent siblings)
3 siblings, 0 replies; 10+ messages in thread
From: Jonathan Cameron via @ 2023-09-19 9:34 UTC (permalink / raw)
To: Michael Tokarev, qemu-devel, Michael Tsirkin, Fan Ni,
Philippe Mathieu-Daudé
Cc: linuxarm
Bring this read function inline with the others that do
check for unexpected size values.
Also reduces line lengths to sub 80 chars.
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Fan Ni <fan.ni@samsung.com>
---
hw/cxl/cxl-device-utils.c | 11 +++++++----
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/hw/cxl/cxl-device-utils.c b/hw/cxl/cxl-device-utils.c
index 517f06d869..cd0c45a2ed 100644
--- a/hw/cxl/cxl-device-utils.c
+++ b/hw/cxl/cxl-device-utils.c
@@ -32,10 +32,13 @@ static uint64_t caps_reg_read(void *opaque, hwaddr offset, unsigned size)
{
CXLDeviceState *cxl_dstate = opaque;
- if (size == 4) {
- return cxl_dstate->caps_reg_state32[offset / sizeof(*cxl_dstate->caps_reg_state32)];
- } else {
- return cxl_dstate->caps_reg_state64[offset / sizeof(*cxl_dstate->caps_reg_state64)];
+ switch (size) {
+ case 4:
+ return cxl_dstate->caps_reg_state32[offset / size];
+ case 8:
+ return cxl_dstate->caps_reg_state64[offset / size];
+ default:
+ g_assert_not_reached();
}
}
--
2.39.2
^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH v3 2/4] hw/cxl: Use switch statements for read and write of cachemem registers
2023-09-19 9:34 [PATCH v3 0/4] hw/cxl: Line length reduction and related Jonathan Cameron via
2023-09-19 9:34 ` [PATCH v3 1/4] hw/cxl: Use a switch to explicitly check size in caps_reg_read() Jonathan Cameron via
@ 2023-09-19 9:34 ` Jonathan Cameron via
2023-09-20 5:08 ` Michael Tokarev
2023-10-03 20:42 ` Michael S. Tsirkin
2023-09-19 9:34 ` [PATCH v3 3/4] hw/cxl: CXLDVSECPortExtensions renamed to CXLDVSECPortExt Jonathan Cameron via
2023-09-19 9:34 ` [PATCH v3 4/4] hw/cxl: Line length reductions Jonathan Cameron via
3 siblings, 2 replies; 10+ messages in thread
From: Jonathan Cameron via @ 2023-09-19 9:34 UTC (permalink / raw)
To: Michael Tokarev, qemu-devel, Michael Tsirkin, Fan Ni,
Philippe Mathieu-Daudé
Cc: linuxarm
Establishing that only register accesses of size 4 and 8 can occur
using these functions requires looking at their callers. Make it
easier to see that by using switch statements.
Assertions are used to enforce that the register storage is of the
matching size, allowing fixed values to be used for divisors of
the array indices.
Suggested-by: Michael Tokarev <mjt@tls.msk.ru>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Fan Ni <fan.ni@samsung.com>
---
hw/cxl/cxl-component-utils.c | 65 +++++++++++++++++++++++-------------
1 file changed, 42 insertions(+), 23 deletions(-)
diff --git a/hw/cxl/cxl-component-utils.c b/hw/cxl/cxl-component-utils.c
index f3bbf0fd13..7ef3ef2bd6 100644
--- a/hw/cxl/cxl-component-utils.c
+++ b/hw/cxl/cxl-component-utils.c
@@ -67,16 +67,24 @@ static uint64_t cxl_cache_mem_read_reg(void *opaque, hwaddr offset,
CXLComponentState *cxl_cstate = opaque;
ComponentRegisters *cregs = &cxl_cstate->crb;
- if (size == 8) {
+ switch (size) {
+ case 4:
+ if (cregs->special_ops && cregs->special_ops->read) {
+ return cregs->special_ops->read(cxl_cstate, offset, 4);
+ } else {
+ static_assert(sizeof(*cregs->cache_mem_registers) == 4);
+ return cregs->cache_mem_registers[offset / 4];
+ }
+ case 8:
qemu_log_mask(LOG_UNIMP,
"CXL 8 byte cache mem registers not implemented\n");
return 0;
- }
-
- if (cregs->special_ops && cregs->special_ops->read) {
- return cregs->special_ops->read(cxl_cstate, offset, size);
- } else {
- return cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)];
+ default:
+ /*
+ * In line with specifiction limitaions on access sizes, this
+ * routine is not called with other sizes.
+ */
+ g_assert_not_reached();
}
}
@@ -117,25 +125,36 @@ static void cxl_cache_mem_write_reg(void *opaque, hwaddr offset, uint64_t value,
ComponentRegisters *cregs = &cxl_cstate->crb;
uint32_t mask;
- if (size == 8) {
+ switch (size) {
+ case 4:
+ static_assert(sizeof(*cregs->cache_mem_regs_write_mask) == 4);
+ static_assert(sizeof(*cregs->cache_mem_registers) == 4);
+ mask = cregs->cache_mem_regs_write_mask[offset / 4];
+ value &= mask;
+ /* RO bits should remain constant. Done by reading existing value */
+ value |= ~mask & cregs->cache_mem_registers[offset / 4];
+ if (cregs->special_ops && cregs->special_ops->write) {
+ cregs->special_ops->write(cxl_cstate, offset, value, size);
+ return;
+ }
+
+ if (offset >= A_CXL_HDM_DECODER_CAPABILITY &&
+ offset <= A_CXL_HDM_DECODER3_TARGET_LIST_HI) {
+ dumb_hdm_handler(cxl_cstate, offset, value);
+ } else {
+ cregs->cache_mem_registers[offset / 4] = value;
+ }
+ return;
+ case 8:
qemu_log_mask(LOG_UNIMP,
"CXL 8 byte cache mem registers not implemented\n");
return;
- }
- mask = cregs->cache_mem_regs_write_mask[offset / sizeof(*cregs->cache_mem_regs_write_mask)];
- value &= mask;
- /* RO bits should remain constant. Done by reading existing value */
- value |= ~mask & cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)];
- if (cregs->special_ops && cregs->special_ops->write) {
- cregs->special_ops->write(cxl_cstate, offset, value, size);
- return;
- }
-
- if (offset >= A_CXL_HDM_DECODER_CAPABILITY &&
- offset <= A_CXL_HDM_DECODER3_TARGET_LIST_HI) {
- dumb_hdm_handler(cxl_cstate, offset, value);
- } else {
- cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)] = value;
+ default:
+ /*
+ * In line with specifiction limitaions on access sizes, this
+ * routine is not called with other sizes.
+ */
+ g_assert_not_reached();
}
}
--
2.39.2
^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH v3 3/4] hw/cxl: CXLDVSECPortExtensions renamed to CXLDVSECPortExt
2023-09-19 9:34 [PATCH v3 0/4] hw/cxl: Line length reduction and related Jonathan Cameron via
2023-09-19 9:34 ` [PATCH v3 1/4] hw/cxl: Use a switch to explicitly check size in caps_reg_read() Jonathan Cameron via
2023-09-19 9:34 ` [PATCH v3 2/4] hw/cxl: Use switch statements for read and write of cachemem registers Jonathan Cameron via
@ 2023-09-19 9:34 ` Jonathan Cameron via
2023-09-19 9:34 ` [PATCH v3 4/4] hw/cxl: Line length reductions Jonathan Cameron via
3 siblings, 0 replies; 10+ messages in thread
From: Jonathan Cameron via @ 2023-09-19 9:34 UTC (permalink / raw)
To: Michael Tokarev, qemu-devel, Michael Tsirkin, Fan Ni,
Philippe Mathieu-Daudé
Cc: linuxarm
Done to reduce line lengths where this is used.
Ext seems sufficiently obvious that it need not be spelt out
fully.
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Fan Ni <fan.ni@samsung.com>
---
include/hw/cxl/cxl_pci.h | 6 ++---
hw/cxl/cxl-component-utils.c | 49 ++++++++++++++++++++--------------
hw/pci-bridge/cxl_downstream.c | 2 +-
hw/pci-bridge/cxl_root_port.c | 2 +-
hw/pci-bridge/cxl_upstream.c | 2 +-
5 files changed, 35 insertions(+), 26 deletions(-)
diff --git a/include/hw/cxl/cxl_pci.h b/include/hw/cxl/cxl_pci.h
index 407be95b9e..ddf01a543b 100644
--- a/include/hw/cxl/cxl_pci.h
+++ b/include/hw/cxl/cxl_pci.h
@@ -86,7 +86,7 @@ typedef struct CXLDVSECDevice {
QEMU_BUILD_BUG_ON(sizeof(CXLDVSECDevice) != 0x38);
/* CXL 2.0 - 8.1.5 (ID 0003) */
-typedef struct CXLDVSECPortExtensions {
+typedef struct CXLDVSECPortExt {
DVSECHeader hdr;
uint16_t status;
uint16_t control;
@@ -100,8 +100,8 @@ typedef struct CXLDVSECPortExtensions {
uint32_t alt_prefetch_limit_high;
uint32_t rcrb_base;
uint32_t rcrb_base_high;
-} CXLDVSECPortExtensions;
-QEMU_BUILD_BUG_ON(sizeof(CXLDVSECPortExtensions) != 0x28);
+} CXLDVSECPortExt;
+QEMU_BUILD_BUG_ON(sizeof(CXLDVSECPortExt) != 0x28);
#define PORT_CONTROL_OFFSET 0xc
#define PORT_CONTROL_UNMASK_SBR 1
diff --git a/hw/cxl/cxl-component-utils.c b/hw/cxl/cxl-component-utils.c
index 7ef3ef2bd6..c8d632d540 100644
--- a/hw/cxl/cxl-component-utils.c
+++ b/hw/cxl/cxl-component-utils.c
@@ -392,26 +392,35 @@ void cxl_component_create_dvsec(CXLComponentState *cxl,
case NON_CXL_FUNCTION_MAP_DVSEC:
break; /* Not yet implemented */
case EXTENSIONS_PORT_DVSEC:
- wmask[offset + offsetof(CXLDVSECPortExtensions, control)] = 0x0F;
- wmask[offset + offsetof(CXLDVSECPortExtensions, control) + 1] = 0x40;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_bus_base)] = 0xFF;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_bus_limit)] = 0xFF;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_memory_base)] = 0xF0;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_memory_base) + 1] = 0xFF;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_memory_limit)] = 0xF0;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_memory_limit) + 1] = 0xFF;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base)] = 0xF0;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base) + 1] = 0xFF;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit)] = 0xF0;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit) + 1] = 0xFF;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base_high)] = 0xFF;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base_high) + 1] = 0xFF;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base_high) + 2] = 0xFF;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_base_high) + 3] = 0xFF;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit_high)] = 0xFF;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit_high) + 1] = 0xFF;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit_high) + 2] = 0xFF;
- wmask[offset + offsetof(CXLDVSECPortExtensions, alt_prefetch_limit_high) + 3] = 0xFF;
+ wmask[offset + offsetof(CXLDVSECPortExt, control)] = 0x0F;
+ wmask[offset + offsetof(CXLDVSECPortExt, control) + 1] = 0x40;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_bus_base)] = 0xFF;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_bus_limit)] = 0xFF;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_memory_base)] = 0xF0;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_memory_base) + 1] = 0xFF;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_memory_limit)] = 0xF0;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_memory_limit) + 1] = 0xFF;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_base)] = 0xF0;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_base) + 1] = 0xFF;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_limit)] = 0xF0;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_limit) + 1] =
+ 0xFF;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_base_high)] =
+ 0xFF;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_base_high) + 1] =
+ 0xFF;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_base_high) + 2] =
+ 0xFF;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_base_high) + 3] =
+ 0xFF;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_limit_high)] =
+ 0xFF;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_limit_high) + 1] =
+ 0xFF;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_limit_high) + 2] =
+ 0xFF;
+ wmask[offset + offsetof(CXLDVSECPortExt, alt_prefetch_limit_high) + 3] =
+ 0xFF;
break;
case GPF_PORT_DVSEC:
wmask[offset + offsetof(CXLDVSECPortGPF, phase1_ctrl)] = 0x0F;
diff --git a/hw/pci-bridge/cxl_downstream.c b/hw/pci-bridge/cxl_downstream.c
index 54f507318f..cdd79ea1c0 100644
--- a/hw/pci-bridge/cxl_downstream.c
+++ b/hw/pci-bridge/cxl_downstream.c
@@ -98,7 +98,7 @@ static void build_dvsecs(CXLComponentState *cxl)
{
uint8_t *dvsec;
- dvsec = (uint8_t *)&(CXLDVSECPortExtensions){ 0 };
+ dvsec = (uint8_t *)&(CXLDVSECPortExt){ 0 };
cxl_component_create_dvsec(cxl, CXL2_DOWNSTREAM_PORT,
EXTENSIONS_PORT_DVSEC_LENGTH,
EXTENSIONS_PORT_DVSEC,
diff --git a/hw/pci-bridge/cxl_root_port.c b/hw/pci-bridge/cxl_root_port.c
index 7dfd20aa67..8f97697631 100644
--- a/hw/pci-bridge/cxl_root_port.c
+++ b/hw/pci-bridge/cxl_root_port.c
@@ -107,7 +107,7 @@ static void build_dvsecs(CXLComponentState *cxl)
{
uint8_t *dvsec;
- dvsec = (uint8_t *)&(CXLDVSECPortExtensions){ 0 };
+ dvsec = (uint8_t *)&(CXLDVSECPortExt){ 0 };
cxl_component_create_dvsec(cxl, CXL2_ROOT_PORT,
EXTENSIONS_PORT_DVSEC_LENGTH,
EXTENSIONS_PORT_DVSEC,
diff --git a/hw/pci-bridge/cxl_upstream.c b/hw/pci-bridge/cxl_upstream.c
index a57806fb31..b81bb5fec9 100644
--- a/hw/pci-bridge/cxl_upstream.c
+++ b/hw/pci-bridge/cxl_upstream.c
@@ -116,7 +116,7 @@ static void build_dvsecs(CXLComponentState *cxl)
{
uint8_t *dvsec;
- dvsec = (uint8_t *)&(CXLDVSECPortExtensions){
+ dvsec = (uint8_t *)&(CXLDVSECPortExt){
.status = 0x1, /* Port Power Management Init Complete */
};
cxl_component_create_dvsec(cxl, CXL2_UPSTREAM_PORT,
--
2.39.2
^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH v3 4/4] hw/cxl: Line length reductions
2023-09-19 9:34 [PATCH v3 0/4] hw/cxl: Line length reduction and related Jonathan Cameron via
` (2 preceding siblings ...)
2023-09-19 9:34 ` [PATCH v3 3/4] hw/cxl: CXLDVSECPortExtensions renamed to CXLDVSECPortExt Jonathan Cameron via
@ 2023-09-19 9:34 ` Jonathan Cameron via
3 siblings, 0 replies; 10+ messages in thread
From: Jonathan Cameron via @ 2023-09-19 9:34 UTC (permalink / raw)
To: Michael Tokarev, qemu-devel, Michael Tsirkin, Fan Ni,
Philippe Mathieu-Daudé
Cc: linuxarm
Michael Tsirkin observed that there were some unnecessarily
long lines in the CXL code in a recent review.
This patch is intended to rectify that where it does not
hurt readability.
Reviewed-by: Michael Tokarev <mjt@tls.msk.ru>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
---
v3: Put removed space back in and rewrap some text (Fan Ni)
---
include/hw/cxl/cxl_component.h | 3 ++-
include/hw/cxl/cxl_device.h | 5 +++--
include/hw/cxl/cxl_events.h | 3 ++-
hw/cxl/cxl-cdat.c | 3 ++-
hw/cxl/cxl-component-utils.c | 14 ++++++++------
hw/cxl/cxl-events.c | 9 ++++++---
hw/cxl/cxl-mailbox-utils.c | 21 ++++++++++++++-------
hw/mem/cxl_type3.c | 31 +++++++++++++++++++------------
hw/mem/cxl_type3_stubs.c | 5 +++--
9 files changed, 59 insertions(+), 35 deletions(-)
diff --git a/include/hw/cxl/cxl_component.h b/include/hw/cxl/cxl_component.h
index 3c795a6278..e52dd8d2b9 100644
--- a/include/hw/cxl/cxl_component.h
+++ b/include/hw/cxl/cxl_component.h
@@ -175,7 +175,8 @@ HDM_DECODER_INIT(3);
(CXL_IDE_REGISTERS_OFFSET + CXL_IDE_REGISTERS_SIZE)
#define CXL_SNOOP_REGISTERS_SIZE 0x8
-QEMU_BUILD_BUG_MSG((CXL_SNOOP_REGISTERS_OFFSET + CXL_SNOOP_REGISTERS_SIZE) >= 0x1000,
+QEMU_BUILD_BUG_MSG((CXL_SNOOP_REGISTERS_OFFSET +
+ CXL_SNOOP_REGISTERS_SIZE) >= 0x1000,
"No space for registers");
typedef struct component_registers {
diff --git a/include/hw/cxl/cxl_device.h b/include/hw/cxl/cxl_device.h
index 51cd0d9ce3..007ddaf078 100644
--- a/include/hw/cxl/cxl_device.h
+++ b/include/hw/cxl/cxl_device.h
@@ -192,7 +192,7 @@ void cxl_device_register_init_common(CXLDeviceState *dev);
* Documented as a 128 bit register, but 64 bit accesses and the second
* 64 bits are currently reserved.
*/
-REG64(CXL_DEV_CAP_ARRAY, 0) /* Documented as 128 bit register but 64 byte accesses */
+REG64(CXL_DEV_CAP_ARRAY, 0)
FIELD(CXL_DEV_CAP_ARRAY, CAP_ID, 0, 16)
FIELD(CXL_DEV_CAP_ARRAY, CAP_VERSION, 16, 8)
FIELD(CXL_DEV_CAP_ARRAY, CAP_COUNT, 32, 16)
@@ -361,7 +361,8 @@ struct CXLType3Class {
uint64_t offset);
void (*set_lsa)(CXLType3Dev *ct3d, const void *buf, uint64_t size,
uint64_t offset);
- bool (*set_cacheline)(CXLType3Dev *ct3d, uint64_t dpa_offset, uint8_t *data);
+ bool (*set_cacheline)(CXLType3Dev *ct3d, uint64_t dpa_offset,
+ uint8_t *data);
};
MemTxResult cxl_type3_read(PCIDevice *d, hwaddr host_addr, uint64_t *data,
diff --git a/include/hw/cxl/cxl_events.h b/include/hw/cxl/cxl_events.h
index 089ba2091f..d778487b7e 100644
--- a/include/hw/cxl/cxl_events.h
+++ b/include/hw/cxl/cxl_events.h
@@ -92,7 +92,8 @@ typedef enum CXLEventIntMode {
CXL_INT_RES = 0x03,
} CXLEventIntMode;
#define CXL_EVENT_INT_MODE_MASK 0x3
-#define CXL_EVENT_INT_SETTING(vector) ((((uint8_t)vector & 0xf) << 4) | CXL_INT_MSI_MSIX)
+#define CXL_EVENT_INT_SETTING(vector) \
+ ((((uint8_t)vector & 0xf) << 4) | CXL_INT_MSI_MSIX)
typedef struct CXLEventInterruptPolicy {
uint8_t info_settings;
uint8_t warn_settings;
diff --git a/hw/cxl/cxl-cdat.c b/hw/cxl/cxl-cdat.c
index d246d6885b..639a2db3e1 100644
--- a/hw/cxl/cxl-cdat.c
+++ b/hw/cxl/cxl-cdat.c
@@ -60,7 +60,8 @@ static void ct3_build_cdat(CDATObject *cdat, Error **errp)
return;
}
- cdat->built_buf_len = cdat->build_cdat_table(&cdat->built_buf, cdat->private);
+ cdat->built_buf_len = cdat->build_cdat_table(&cdat->built_buf,
+ cdat->private);
if (!cdat->built_buf_len) {
/* Build later as not all data available yet */
diff --git a/hw/cxl/cxl-component-utils.c b/hw/cxl/cxl-component-utils.c
index c8d632d540..486cd5200a 100644
--- a/hw/cxl/cxl-component-utils.c
+++ b/hw/cxl/cxl-component-utils.c
@@ -240,7 +240,8 @@ static void hdm_init_common(uint32_t *reg_state, uint32_t *write_msk,
ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_CAPABILITY, TARGET_COUNT, 1);
ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_CAPABILITY, INTERLEAVE_256B, 1);
ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_CAPABILITY, INTERLEAVE_4K, 1);
- ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_CAPABILITY, POISON_ON_ERR_CAP, 0);
+ ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_CAPABILITY,
+ POISON_ON_ERR_CAP, 0);
ARRAY_FIELD_DP32(reg_state, CXL_HDM_DECODER_GLOBAL_CONTROL,
HDM_DECODER_ENABLE, 0);
write_msk[R_CXL_HDM_DECODER_GLOBAL_CONTROL] = 0x3;
@@ -263,15 +264,16 @@ static void hdm_init_common(uint32_t *reg_state, uint32_t *write_msk,
}
}
-void cxl_component_register_init_common(uint32_t *reg_state, uint32_t *write_msk,
+void cxl_component_register_init_common(uint32_t *reg_state,
+ uint32_t *write_msk,
enum reg_type type)
{
int caps = 0;
/*
- * In CXL 2.0 the capabilities required for each CXL component are such that,
- * with the ordering chosen here, a single number can be used to define
- * which capabilities should be provided.
+ * In CXL 2.0 the capabilities required for each CXL component are such
+ * that, with the ordering chosen here, a single number can be used to
+ * define which capabilities should be provided.
*/
switch (type) {
case CXL2_DOWNSTREAM_PORT:
@@ -448,7 +450,7 @@ void cxl_component_create_dvsec(CXLComponentState *cxl,
default: /* Registers are RO for other component types */
break;
}
- /* There are rw1cs bits in the status register but never set currently */
+ /* There are rw1cs bits in the status register but never set */
break;
}
diff --git a/hw/cxl/cxl-events.c b/hw/cxl/cxl-events.c
index d161d57456..10a4531904 100644
--- a/hw/cxl/cxl-events.c
+++ b/hw/cxl/cxl-events.c
@@ -170,8 +170,10 @@ CXLRetCode cxl_event_get_records(CXLDeviceState *cxlds, CXLGetEventPayload *pl,
if (log->overflow_err_count) {
pl->flags |= CXL_GET_EVENT_FLAG_OVERFLOW;
pl->overflow_err_count = cpu_to_le16(log->overflow_err_count);
- pl->first_overflow_timestamp = cpu_to_le64(log->first_overflow_timestamp);
- pl->last_overflow_timestamp = cpu_to_le64(log->last_overflow_timestamp);
+ pl->first_overflow_timestamp =
+ cpu_to_le64(log->first_overflow_timestamp);
+ pl->last_overflow_timestamp =
+ cpu_to_le64(log->last_overflow_timestamp);
}
pl->record_count = cpu_to_le16(nr);
@@ -180,7 +182,8 @@ CXLRetCode cxl_event_get_records(CXLDeviceState *cxlds, CXLGetEventPayload *pl,
return CXL_MBOX_SUCCESS;
}
-CXLRetCode cxl_event_clear_records(CXLDeviceState *cxlds, CXLClearEventPayload *pl)
+CXLRetCode cxl_event_clear_records(CXLDeviceState *cxlds,
+ CXLClearEventPayload *pl)
{
CXLEventLog *log;
uint8_t log_type;
diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c
index 02f9b5a870..3977af23f1 100644
--- a/hw/cxl/cxl-mailbox-utils.c
+++ b/hw/cxl/cxl-mailbox-utils.c
@@ -366,9 +366,12 @@ static CXLRetCode cmd_identify_memory_device(struct cxl_cmd *cmd,
snprintf(id->fw_revision, 0x10, "BWFW VERSION %02d", 0);
- stq_le_p(&id->total_capacity, cxl_dstate->mem_size / CXL_CAPACITY_MULTIPLIER);
- stq_le_p(&id->persistent_capacity, cxl_dstate->pmem_size / CXL_CAPACITY_MULTIPLIER);
- stq_le_p(&id->volatile_capacity, cxl_dstate->vmem_size / CXL_CAPACITY_MULTIPLIER);
+ stq_le_p(&id->total_capacity,
+ cxl_dstate->mem_size / CXL_CAPACITY_MULTIPLIER);
+ stq_le_p(&id->persistent_capacity,
+ cxl_dstate->pmem_size / CXL_CAPACITY_MULTIPLIER);
+ stq_le_p(&id->volatile_capacity,
+ cxl_dstate->vmem_size / CXL_CAPACITY_MULTIPLIER);
stl_le_p(&id->lsa_size, cvc->get_lsa_size(ct3d));
/* 256 poison records */
st24_le_p(id->poison_list_max_mer, 256);
@@ -396,13 +399,15 @@ static CXLRetCode cmd_ccls_get_partition_info(struct cxl_cmd *cmd,
return CXL_MBOX_INTERNAL_ERROR;
}
- stq_le_p(&part_info->active_vmem, cxl_dstate->vmem_size / CXL_CAPACITY_MULTIPLIER);
+ stq_le_p(&part_info->active_vmem,
+ cxl_dstate->vmem_size / CXL_CAPACITY_MULTIPLIER);
/*
* When both next_vmem and next_pmem are 0, there is no pending change to
* partitioning.
*/
stq_le_p(&part_info->next_vmem, 0);
- stq_le_p(&part_info->active_pmem, cxl_dstate->pmem_size / CXL_CAPACITY_MULTIPLIER);
+ stq_le_p(&part_info->active_pmem,
+ cxl_dstate->pmem_size / CXL_CAPACITY_MULTIPLIER);
stq_le_p(&part_info->next_pmem, 0);
*len = sizeof(*part_info);
@@ -681,8 +686,10 @@ static struct cxl_cmd cxl_cmd_set[256][256] = {
[FIRMWARE_UPDATE][GET_INFO] = { "FIRMWARE_UPDATE_GET_INFO",
cmd_firmware_update_get_info, 0, 0 },
[TIMESTAMP][GET] = { "TIMESTAMP_GET", cmd_timestamp_get, 0, 0 },
- [TIMESTAMP][SET] = { "TIMESTAMP_SET", cmd_timestamp_set, 8, IMMEDIATE_POLICY_CHANGE },
- [LOGS][GET_SUPPORTED] = { "LOGS_GET_SUPPORTED", cmd_logs_get_supported, 0, 0 },
+ [TIMESTAMP][SET] = { "TIMESTAMP_SET", cmd_timestamp_set,
+ 8, IMMEDIATE_POLICY_CHANGE },
+ [LOGS][GET_SUPPORTED] = { "LOGS_GET_SUPPORTED", cmd_logs_get_supported,
+ 0, 0 },
[LOGS][GET_LOG] = { "LOGS_GET_LOG", cmd_logs_get_log, 0x18, 0 },
[IDENTIFY][MEMORY_DEVICE] = { "IDENTIFY_MEMORY_DEVICE",
cmd_identify_memory_device, 0, 0 },
diff --git a/hw/mem/cxl_type3.c b/hw/mem/cxl_type3.c
index 1658e0cc59..c5855d4e7d 100644
--- a/hw/mem/cxl_type3.c
+++ b/hw/mem/cxl_type3.c
@@ -197,10 +197,9 @@ static int ct3_build_cdat_table(CDATSubHeader ***cdat_table, void *priv)
}
if (nonvolatile_mr) {
+ uint64_t base = volatile_mr ? memory_region_size(volatile_mr) : 0;
rc = ct3_build_cdat_entries_for_mr(&(table[cur_ent]), dsmad_handle++,
- nonvolatile_mr, true,
- (volatile_mr ?
- memory_region_size(volatile_mr) : 0));
+ nonvolatile_mr, true, base);
if (rc < 0) {
goto error_cleanup;
}
@@ -503,7 +502,8 @@ static void ct3d_reg_write(void *opaque, hwaddr offset, uint64_t value,
case A_CXL_RAS_UNC_ERR_STATUS:
{
uint32_t capctrl = ldl_le_p(cache_mem + R_CXL_RAS_ERR_CAP_CTRL);
- uint32_t fe = FIELD_EX32(capctrl, CXL_RAS_ERR_CAP_CTRL, FIRST_ERROR_POINTER);
+ uint32_t fe = FIELD_EX32(capctrl, CXL_RAS_ERR_CAP_CTRL,
+ FIRST_ERROR_POINTER);
CXLError *cxl_err;
uint32_t unc_err;
@@ -522,7 +522,8 @@ static void ct3d_reg_write(void *opaque, hwaddr offset, uint64_t value,
* closest to behavior of hardware not capable of multiple
* header recording.
*/
- QTAILQ_FOREACH_SAFE(cxl_err, &ct3d->error_list, node, cxl_next) {
+ QTAILQ_FOREACH_SAFE(cxl_err, &ct3d->error_list, node,
+ cxl_next) {
if ((1 << cxl_err->type) & value) {
QTAILQ_REMOVE(&ct3d->error_list, cxl_err, node);
g_free(cxl_err);
@@ -1061,7 +1062,8 @@ void qmp_cxl_inject_poison(const char *path, uint64_t start, uint64_t length,
if (((start >= p->start) && (start < p->start + p->length)) ||
((start + length > p->start) &&
(start + length <= p->start + p->length))) {
- error_setg(errp, "Overlap with existing poisoned region not supported");
+ error_setg(errp,
+ "Overlap with existing poisoned region not supported");
return;
}
}
@@ -1074,7 +1076,8 @@ void qmp_cxl_inject_poison(const char *path, uint64_t start, uint64_t length,
p = g_new0(CXLPoison, 1);
p->length = length;
p->start = start;
- p->type = CXL_POISON_TYPE_INTERNAL; /* Different from injected via the mbox */
+ /* Different from injected via the mbox */
+ p->type = CXL_POISON_TYPE_INTERNAL;
QLIST_INSERT_HEAD(&ct3d->poison_list, p, node);
ct3d->poison_list_cnt++;
@@ -1211,7 +1214,8 @@ void qmp_cxl_inject_correctable_error(const char *path, CxlCorErrorType type,
return;
}
/* If the error is masked, nothting to do here */
- if (!((1 << cxl_err_type) & ~ldl_le_p(reg_state + R_CXL_RAS_COR_ERR_MASK))) {
+ if (!((1 << cxl_err_type) &
+ ~ldl_le_p(reg_state + R_CXL_RAS_COR_ERR_MASK))) {
return;
}
@@ -1361,7 +1365,8 @@ void qmp_cxl_inject_dram_event(const char *path, CxlEventLog log, uint8_t flags,
bool has_bank, uint8_t bank,
bool has_row, uint32_t row,
bool has_column, uint16_t column,
- bool has_correction_mask, uint64List *correction_mask,
+ bool has_correction_mask,
+ uint64List *correction_mask,
Error **errp)
{
Object *obj = object_resolve_path(path, NULL);
@@ -1462,7 +1467,7 @@ void qmp_cxl_inject_memory_module_event(const char *path, CxlEventLog log,
int16_t temperature,
uint32_t dirty_shutdown_count,
uint32_t corrected_volatile_error_count,
- uint32_t corrected_persistent_error_count,
+ uint32_t corrected_persist_error_count,
Error **errp)
{
Object *obj = object_resolve_path(path, NULL);
@@ -1502,8 +1507,10 @@ void qmp_cxl_inject_memory_module_event(const char *path, CxlEventLog log,
module.life_used = life_used;
stw_le_p(&module.temperature, temperature);
stl_le_p(&module.dirty_shutdown_count, dirty_shutdown_count);
- stl_le_p(&module.corrected_volatile_error_count, corrected_volatile_error_count);
- stl_le_p(&module.corrected_persistent_error_count, corrected_persistent_error_count);
+ stl_le_p(&module.corrected_volatile_error_count,
+ corrected_volatile_error_count);
+ stl_le_p(&module.corrected_persistent_error_count,
+ corrected_persist_error_count);
if (cxl_event_insert(cxlds, enc_log, (CXLEventRecordRaw *)&module)) {
cxl_event_irq_assert(ct3d);
diff --git a/hw/mem/cxl_type3_stubs.c b/hw/mem/cxl_type3_stubs.c
index f3e4a9fa72..56d60b4fd8 100644
--- a/hw/mem/cxl_type3_stubs.c
+++ b/hw/mem/cxl_type3_stubs.c
@@ -23,7 +23,8 @@ void qmp_cxl_inject_dram_event(const char *path, CxlEventLog log, uint8_t flags,
bool has_bank, uint8_t bank,
bool has_row, uint32_t row,
bool has_column, uint16_t column,
- bool has_correction_mask, uint64List *correction_mask,
+ bool has_correction_mask,
+ uint64List *correction_mask,
Error **errp) {}
void qmp_cxl_inject_memory_module_event(const char *path, CxlEventLog log,
@@ -35,7 +36,7 @@ void qmp_cxl_inject_memory_module_event(const char *path, CxlEventLog log,
int16_t temperature,
uint32_t dirty_shutdown_count,
uint32_t corrected_volatile_error_count,
- uint32_t corrected_persistent_error_count,
+ uint32_t corrected_persist_error_count,
Error **errp) {}
void qmp_cxl_inject_poison(const char *path, uint64_t start, uint64_t length,
--
2.39.2
^ permalink raw reply related [flat|nested] 10+ messages in thread
* Re: [PATCH v3 2/4] hw/cxl: Use switch statements for read and write of cachemem registers
2023-09-19 9:34 ` [PATCH v3 2/4] hw/cxl: Use switch statements for read and write of cachemem registers Jonathan Cameron via
@ 2023-09-20 5:08 ` Michael Tokarev
2023-09-20 11:40 ` Jonathan Cameron via
2023-10-03 20:42 ` Michael S. Tsirkin
1 sibling, 1 reply; 10+ messages in thread
From: Michael Tokarev @ 2023-09-20 5:08 UTC (permalink / raw)
To: Jonathan Cameron, qemu-devel, Michael Tsirkin, Fan Ni,
Philippe Mathieu-Daudé
Cc: linuxarm
19.09.2023 12:34, Jonathan Cameron via wrote:
> Establishing that only register accesses of size 4 and 8 can occur
> using these functions requires looking at their callers. Make it
> easier to see that by using switch statements.
> Assertions are used to enforce that the register storage is of the
> matching size, allowing fixed values to be used for divisors of
> the array indices.
>
> Suggested-by: Michael Tokarev <mjt@tls.msk.ru>
> Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> Reviewed-by: Fan Ni <fan.ni@samsung.com>
> @@ -117,25 +125,36 @@ static void cxl_cache_mem_write_reg(void *opaque, hwaddr offset, uint64_t value,
> ComponentRegisters *cregs = &cxl_cstate->crb;
> uint32_t mask;
..
This hunk does not apply to qemu/master. Is it based on some other
change missing in this area?
I thought about collecting all this and pushing trivial-patches but
stumbled upon this one.
Thanks,
/mjt
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH v3 2/4] hw/cxl: Use switch statements for read and write of cachemem registers
2023-09-20 5:08 ` Michael Tokarev
@ 2023-09-20 11:40 ` Jonathan Cameron via
0 siblings, 0 replies; 10+ messages in thread
From: Jonathan Cameron via @ 2023-09-20 11:40 UTC (permalink / raw)
To: Michael Tokarev
Cc: qemu-devel, Michael Tsirkin, Fan Ni, Philippe Mathieu-Daudé,
linuxarm
On Wed, 20 Sep 2023 08:08:39 +0300
Michael Tokarev <mjt@tls.msk.ru> wrote:
> 19.09.2023 12:34, Jonathan Cameron via wrote:
> > Establishing that only register accesses of size 4 and 8 can occur
> > using these functions requires looking at their callers. Make it
> > easier to see that by using switch statements.
> > Assertions are used to enforce that the register storage is of the
> > matching size, allowing fixed values to be used for divisors of
> > the array indices.
> >
> > Suggested-by: Michael Tokarev <mjt@tls.msk.ru>
> > Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> > Reviewed-by: Fan Ni <fan.ni@samsung.com>
>
> > @@ -117,25 +125,36 @@ static void cxl_cache_mem_write_reg(void *opaque, hwaddr offset, uint64_t value,
> > ComponentRegisters *cregs = &cxl_cstate->crb;
> > uint32_t mask;
> ..
> This hunk does not apply to qemu/master. Is it based on some other
> change missing in this area?
>
> I thought about collecting all this and pushing trivial-patches but
> stumbled upon this one.
See the dependencies listed in the cover letter for this set.
The HDM decoder series in particular affects the same code.
There are several series that are more urgent than this one as I'm keen
to get some features upstream as well as cleanup this cycle.
I could reorder the tree, but that would have knock on impacts on those
series. Hopefully nothing that would require re-review, but still noisy.
Thanks for considering / trying these though and for picking up
the ones that were dependency free.
Jonathan
>
> Thanks,
>
> /mjt
>
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH v3 2/4] hw/cxl: Use switch statements for read and write of cachemem registers
2023-09-19 9:34 ` [PATCH v3 2/4] hw/cxl: Use switch statements for read and write of cachemem registers Jonathan Cameron via
2023-09-20 5:08 ` Michael Tokarev
@ 2023-10-03 20:42 ` Michael S. Tsirkin
2023-10-12 12:09 ` Jonathan Cameron via
1 sibling, 1 reply; 10+ messages in thread
From: Michael S. Tsirkin @ 2023-10-03 20:42 UTC (permalink / raw)
To: Jonathan Cameron
Cc: Michael Tokarev, qemu-devel, Fan Ni, Philippe Mathieu-Daudé,
linuxarm
On Tue, Sep 19, 2023 at 10:34:32AM +0100, Jonathan Cameron wrote:
> Establishing that only register accesses of size 4 and 8 can occur
> using these functions requires looking at their callers. Make it
> easier to see that by using switch statements.
> Assertions are used to enforce that the register storage is of the
> matching size, allowing fixed values to be used for divisors of
> the array indices.
>
> Suggested-by: Michael Tokarev <mjt@tls.msk.ru>
> Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> Reviewed-by: Fan Ni <fan.ni@samsung.com>
> ---
> hw/cxl/cxl-component-utils.c | 65 +++++++++++++++++++++++-------------
> 1 file changed, 42 insertions(+), 23 deletions(-)
>
> diff --git a/hw/cxl/cxl-component-utils.c b/hw/cxl/cxl-component-utils.c
> index f3bbf0fd13..7ef3ef2bd6 100644
> --- a/hw/cxl/cxl-component-utils.c
> +++ b/hw/cxl/cxl-component-utils.c
> @@ -67,16 +67,24 @@ static uint64_t cxl_cache_mem_read_reg(void *opaque, hwaddr offset,
> CXLComponentState *cxl_cstate = opaque;
> ComponentRegisters *cregs = &cxl_cstate->crb;
>
> - if (size == 8) {
> + switch (size) {
> + case 4:
> + if (cregs->special_ops && cregs->special_ops->read) {
> + return cregs->special_ops->read(cxl_cstate, offset, 4);
> + } else {
> + static_assert(sizeof(*cregs->cache_mem_registers) == 4);
In C currently static_assert needs a message.
In fact clang warns:
../hw/cxl/cxl-component-utils.c:75:67: error: '_Static_assert' with no message is a C2x extension [-Werror,-Wc2x-extensions]
static_assert(sizeof(*cregs->cache_mem_registers) == 4);
> + return cregs->cache_mem_registers[offset / 4];
> + }
> + case 8:
> qemu_log_mask(LOG_UNIMP,
> "CXL 8 byte cache mem registers not implemented\n");
> return 0;
> - }
> -
> - if (cregs->special_ops && cregs->special_ops->read) {
> - return cregs->special_ops->read(cxl_cstate, offset, size);
> - } else {
> - return cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)];
> + default:
> + /*
> + * In line with specifiction limitaions on access sizes, this
> + * routine is not called with other sizes.
> + */
> + g_assert_not_reached();
> }
> }
>
> @@ -117,25 +125,36 @@ static void cxl_cache_mem_write_reg(void *opaque, hwaddr offset, uint64_t value,
> ComponentRegisters *cregs = &cxl_cstate->crb;
> uint32_t mask;
>
> - if (size == 8) {
> + switch (size) {
> + case 4:
> + static_assert(sizeof(*cregs->cache_mem_regs_write_mask) == 4);
> + static_assert(sizeof(*cregs->cache_mem_registers) == 4);
> + mask = cregs->cache_mem_regs_write_mask[offset / 4];
> + value &= mask;
> + /* RO bits should remain constant. Done by reading existing value */
> + value |= ~mask & cregs->cache_mem_registers[offset / 4];
> + if (cregs->special_ops && cregs->special_ops->write) {
> + cregs->special_ops->write(cxl_cstate, offset, value, size);
> + return;
> + }
> +
> + if (offset >= A_CXL_HDM_DECODER_CAPABILITY &&
> + offset <= A_CXL_HDM_DECODER3_TARGET_LIST_HI) {
> + dumb_hdm_handler(cxl_cstate, offset, value);
> + } else {
> + cregs->cache_mem_registers[offset / 4] = value;
> + }
> + return;
> + case 8:
> qemu_log_mask(LOG_UNIMP,
> "CXL 8 byte cache mem registers not implemented\n");
> return;
> - }
> - mask = cregs->cache_mem_regs_write_mask[offset / sizeof(*cregs->cache_mem_regs_write_mask)];
> - value &= mask;
> - /* RO bits should remain constant. Done by reading existing value */
> - value |= ~mask & cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)];
> - if (cregs->special_ops && cregs->special_ops->write) {
> - cregs->special_ops->write(cxl_cstate, offset, value, size);
> - return;
> - }
> -
> - if (offset >= A_CXL_HDM_DECODER_CAPABILITY &&
> - offset <= A_CXL_HDM_DECODER3_TARGET_LIST_HI) {
> - dumb_hdm_handler(cxl_cstate, offset, value);
> - } else {
> - cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)] = value;
> + default:
> + /*
> + * In line with specifiction limitaions on access sizes, this
> + * routine is not called with other sizes.
> + */
> + g_assert_not_reached();
> }
> }
>
> --
> 2.39.2
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH v3 2/4] hw/cxl: Use switch statements for read and write of cachemem registers
2023-10-03 20:42 ` Michael S. Tsirkin
@ 2023-10-12 12:09 ` Jonathan Cameron via
2023-10-12 12:09 ` Jonathan Cameron
0 siblings, 1 reply; 10+ messages in thread
From: Jonathan Cameron via @ 2023-10-12 12:09 UTC (permalink / raw)
To: Michael S. Tsirkin
Cc: Michael Tokarev, qemu-devel, Fan Ni, Philippe Mathieu-Daudé,
linuxarm
On Tue, 3 Oct 2023 16:42:37 -0400
"Michael S. Tsirkin" <mst@redhat.com> wrote:
> On Tue, Sep 19, 2023 at 10:34:32AM +0100, Jonathan Cameron wrote:
> > Establishing that only register accesses of size 4 and 8 can occur
> > using these functions requires looking at their callers. Make it
> > easier to see that by using switch statements.
> > Assertions are used to enforce that the register storage is of the
> > matching size, allowing fixed values to be used for divisors of
> > the array indices.
> >
> > Suggested-by: Michael Tokarev <mjt@tls.msk.ru>
> > Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> > Reviewed-by: Fan Ni <fan.ni@samsung.com>
> > ---
> > hw/cxl/cxl-component-utils.c | 65 +++++++++++++++++++++++-------------
> > 1 file changed, 42 insertions(+), 23 deletions(-)
> >
> > diff --git a/hw/cxl/cxl-component-utils.c b/hw/cxl/cxl-component-utils.c
> > index f3bbf0fd13..7ef3ef2bd6 100644
> > --- a/hw/cxl/cxl-component-utils.c
> > +++ b/hw/cxl/cxl-component-utils.c
> > @@ -67,16 +67,24 @@ static uint64_t cxl_cache_mem_read_reg(void *opaque, hwaddr offset,
> > CXLComponentState *cxl_cstate = opaque;
> > ComponentRegisters *cregs = &cxl_cstate->crb;
> >
> > - if (size == 8) {
> > + switch (size) {
> > + case 4:
> > + if (cregs->special_ops && cregs->special_ops->read) {
> > + return cregs->special_ops->read(cxl_cstate, offset, 4);
> > + } else {
> > + static_assert(sizeof(*cregs->cache_mem_registers) == 4);
>
> In C currently static_assert needs a message.
I switched to QEMU_BUILD_BUG_ON() which seems better
+ doesn't need a additional message.
>
> In fact clang warns:
> ../hw/cxl/cxl-component-utils.c:75:67: error: '_Static_assert' with no message is a C2x extension [-Werror,-Wc2x-extensions]
> static_assert(sizeof(*cregs->cache_mem_registers) == 4);
>
Thanks,
Jonathan
>
> > + return cregs->cache_mem_registers[offset / 4];
> > + }
> > + case 8:
> > qemu_log_mask(LOG_UNIMP,
> > "CXL 8 byte cache mem registers not implemented\n");
> > return 0;
> > - }
> > -
> > - if (cregs->special_ops && cregs->special_ops->read) {
> > - return cregs->special_ops->read(cxl_cstate, offset, size);
> > - } else {
> > - return cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)];
> > + default:
> > + /*
> > + * In line with specifiction limitaions on access sizes, this
> > + * routine is not called with other sizes.
> > + */
> > + g_assert_not_reached();
> > }
> > }
> >
> > @@ -117,25 +125,36 @@ static void cxl_cache_mem_write_reg(void *opaque, hwaddr offset, uint64_t value,
> > ComponentRegisters *cregs = &cxl_cstate->crb;
> > uint32_t mask;
> >
> > - if (size == 8) {
> > + switch (size) {
> > + case 4:
> > + static_assert(sizeof(*cregs->cache_mem_regs_write_mask) == 4);
> > + static_assert(sizeof(*cregs->cache_mem_registers) == 4);
> > + mask = cregs->cache_mem_regs_write_mask[offset / 4];
> > + value &= mask;
> > + /* RO bits should remain constant. Done by reading existing value */
> > + value |= ~mask & cregs->cache_mem_registers[offset / 4];
> > + if (cregs->special_ops && cregs->special_ops->write) {
> > + cregs->special_ops->write(cxl_cstate, offset, value, size);
> > + return;
> > + }
> > +
> > + if (offset >= A_CXL_HDM_DECODER_CAPABILITY &&
> > + offset <= A_CXL_HDM_DECODER3_TARGET_LIST_HI) {
> > + dumb_hdm_handler(cxl_cstate, offset, value);
> > + } else {
> > + cregs->cache_mem_registers[offset / 4] = value;
> > + }
> > + return;
> > + case 8:
> > qemu_log_mask(LOG_UNIMP,
> > "CXL 8 byte cache mem registers not implemented\n");
> > return;
> > - }
> > - mask = cregs->cache_mem_regs_write_mask[offset / sizeof(*cregs->cache_mem_regs_write_mask)];
> > - value &= mask;
> > - /* RO bits should remain constant. Done by reading existing value */
> > - value |= ~mask & cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)];
> > - if (cregs->special_ops && cregs->special_ops->write) {
> > - cregs->special_ops->write(cxl_cstate, offset, value, size);
> > - return;
> > - }
> > -
> > - if (offset >= A_CXL_HDM_DECODER_CAPABILITY &&
> > - offset <= A_CXL_HDM_DECODER3_TARGET_LIST_HI) {
> > - dumb_hdm_handler(cxl_cstate, offset, value);
> > - } else {
> > - cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)] = value;
> > + default:
> > + /*
> > + * In line with specifiction limitaions on access sizes, this
> > + * routine is not called with other sizes.
> > + */
> > + g_assert_not_reached();
> > }
> > }
> >
> > --
> > 2.39.2
>
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH v3 2/4] hw/cxl: Use switch statements for read and write of cachemem registers
2023-10-12 12:09 ` Jonathan Cameron via
@ 2023-10-12 12:09 ` Jonathan Cameron
0 siblings, 0 replies; 10+ messages in thread
From: Jonathan Cameron @ 2023-10-12 12:09 UTC (permalink / raw)
To: Michael S. Tsirkin
Cc: Michael Tokarev, qemu-devel, Fan Ni, Philippe Mathieu-Daudé,
linuxarm
On Tue, 3 Oct 2023 16:42:37 -0400
"Michael S. Tsirkin" <mst@redhat.com> wrote:
> On Tue, Sep 19, 2023 at 10:34:32AM +0100, Jonathan Cameron wrote:
> > Establishing that only register accesses of size 4 and 8 can occur
> > using these functions requires looking at their callers. Make it
> > easier to see that by using switch statements.
> > Assertions are used to enforce that the register storage is of the
> > matching size, allowing fixed values to be used for divisors of
> > the array indices.
> >
> > Suggested-by: Michael Tokarev <mjt@tls.msk.ru>
> > Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> > Reviewed-by: Fan Ni <fan.ni@samsung.com>
> > ---
> > hw/cxl/cxl-component-utils.c | 65 +++++++++++++++++++++++-------------
> > 1 file changed, 42 insertions(+), 23 deletions(-)
> >
> > diff --git a/hw/cxl/cxl-component-utils.c b/hw/cxl/cxl-component-utils.c
> > index f3bbf0fd13..7ef3ef2bd6 100644
> > --- a/hw/cxl/cxl-component-utils.c
> > +++ b/hw/cxl/cxl-component-utils.c
> > @@ -67,16 +67,24 @@ static uint64_t cxl_cache_mem_read_reg(void *opaque, hwaddr offset,
> > CXLComponentState *cxl_cstate = opaque;
> > ComponentRegisters *cregs = &cxl_cstate->crb;
> >
> > - if (size == 8) {
> > + switch (size) {
> > + case 4:
> > + if (cregs->special_ops && cregs->special_ops->read) {
> > + return cregs->special_ops->read(cxl_cstate, offset, 4);
> > + } else {
> > + static_assert(sizeof(*cregs->cache_mem_registers) == 4);
>
> In C currently static_assert needs a message.
I switched to QEMU_BUILD_BUG_ON() which seems better
+ doesn't need a additional message.
>
> In fact clang warns:
> ../hw/cxl/cxl-component-utils.c:75:67: error: '_Static_assert' with no message is a C2x extension [-Werror,-Wc2x-extensions]
> static_assert(sizeof(*cregs->cache_mem_registers) == 4);
>
Thanks,
Jonathan
>
> > + return cregs->cache_mem_registers[offset / 4];
> > + }
> > + case 8:
> > qemu_log_mask(LOG_UNIMP,
> > "CXL 8 byte cache mem registers not implemented\n");
> > return 0;
> > - }
> > -
> > - if (cregs->special_ops && cregs->special_ops->read) {
> > - return cregs->special_ops->read(cxl_cstate, offset, size);
> > - } else {
> > - return cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)];
> > + default:
> > + /*
> > + * In line with specifiction limitaions on access sizes, this
> > + * routine is not called with other sizes.
> > + */
> > + g_assert_not_reached();
> > }
> > }
> >
> > @@ -117,25 +125,36 @@ static void cxl_cache_mem_write_reg(void *opaque, hwaddr offset, uint64_t value,
> > ComponentRegisters *cregs = &cxl_cstate->crb;
> > uint32_t mask;
> >
> > - if (size == 8) {
> > + switch (size) {
> > + case 4:
> > + static_assert(sizeof(*cregs->cache_mem_regs_write_mask) == 4);
> > + static_assert(sizeof(*cregs->cache_mem_registers) == 4);
> > + mask = cregs->cache_mem_regs_write_mask[offset / 4];
> > + value &= mask;
> > + /* RO bits should remain constant. Done by reading existing value */
> > + value |= ~mask & cregs->cache_mem_registers[offset / 4];
> > + if (cregs->special_ops && cregs->special_ops->write) {
> > + cregs->special_ops->write(cxl_cstate, offset, value, size);
> > + return;
> > + }
> > +
> > + if (offset >= A_CXL_HDM_DECODER_CAPABILITY &&
> > + offset <= A_CXL_HDM_DECODER3_TARGET_LIST_HI) {
> > + dumb_hdm_handler(cxl_cstate, offset, value);
> > + } else {
> > + cregs->cache_mem_registers[offset / 4] = value;
> > + }
> > + return;
> > + case 8:
> > qemu_log_mask(LOG_UNIMP,
> > "CXL 8 byte cache mem registers not implemented\n");
> > return;
> > - }
> > - mask = cregs->cache_mem_regs_write_mask[offset / sizeof(*cregs->cache_mem_regs_write_mask)];
> > - value &= mask;
> > - /* RO bits should remain constant. Done by reading existing value */
> > - value |= ~mask & cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)];
> > - if (cregs->special_ops && cregs->special_ops->write) {
> > - cregs->special_ops->write(cxl_cstate, offset, value, size);
> > - return;
> > - }
> > -
> > - if (offset >= A_CXL_HDM_DECODER_CAPABILITY &&
> > - offset <= A_CXL_HDM_DECODER3_TARGET_LIST_HI) {
> > - dumb_hdm_handler(cxl_cstate, offset, value);
> > - } else {
> > - cregs->cache_mem_registers[offset / sizeof(*cregs->cache_mem_registers)] = value;
> > + default:
> > + /*
> > + * In line with specifiction limitaions on access sizes, this
> > + * routine is not called with other sizes.
> > + */
> > + g_assert_not_reached();
> > }
> > }
> >
> > --
> > 2.39.2
>
^ permalink raw reply [flat|nested] 10+ messages in thread
end of thread, other threads:[~2023-10-12 12:10 UTC | newest]
Thread overview: 10+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2023-09-19 9:34 [PATCH v3 0/4] hw/cxl: Line length reduction and related Jonathan Cameron via
2023-09-19 9:34 ` [PATCH v3 1/4] hw/cxl: Use a switch to explicitly check size in caps_reg_read() Jonathan Cameron via
2023-09-19 9:34 ` [PATCH v3 2/4] hw/cxl: Use switch statements for read and write of cachemem registers Jonathan Cameron via
2023-09-20 5:08 ` Michael Tokarev
2023-09-20 11:40 ` Jonathan Cameron via
2023-10-03 20:42 ` Michael S. Tsirkin
2023-10-12 12:09 ` Jonathan Cameron via
2023-10-12 12:09 ` Jonathan Cameron
2023-09-19 9:34 ` [PATCH v3 3/4] hw/cxl: CXLDVSECPortExtensions renamed to CXLDVSECPortExt Jonathan Cameron via
2023-09-19 9:34 ` [PATCH v3 4/4] hw/cxl: Line length reductions Jonathan Cameron via
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).