diff mbox series

[v4,5/6] hw/cxl: Add poison injection via the mailbox.

Message ID 20230303150908.27889-6-Jonathan.Cameron@huawei.com
State Superseded
Headers show
Series hw/cxl: Poison get, inject, clear | expand

Commit Message

Jonathan Cameron March 3, 2023, 3:09 p.m. UTC
Very simple implementation to allow testing of corresponding
kernel code. Note that for now we track each 64 byte section
independently.  Whilst a valid implementation choice, it may
make sense to fuse entries so as to prove out more complex
corners of the kernel code.

Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
---
v4: No change
---
 hw/cxl/cxl-mailbox-utils.c | 41 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 41 insertions(+)

Comments

nifan@outlook.com March 14, 2023, 5:22 a.m. UTC | #1
The 03/03/2023 15:09, Jonathan Cameron wrote:
> Very simple implementation to allow testing of corresponding
> kernel code. Note that for now we track each 64 byte section
> independently.  Whilst a valid implementation choice, it may
> make sense to fuse entries so as to prove out more complex
> corners of the kernel code.
> 
> Reviewed-by: Ira Weiny <ira.weiny@intel.com>
> Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> ---

Reviewed-by: Fan Ni <fan.ni@samsung.com>

> v4: No change
> ---
>  hw/cxl/cxl-mailbox-utils.c | 41 ++++++++++++++++++++++++++++++++++++++
>  1 file changed, 41 insertions(+)
> 
> diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c
> index 25933cf62c..64a3f3c1bf 100644
> --- a/hw/cxl/cxl-mailbox-utils.c
> +++ b/hw/cxl/cxl-mailbox-utils.c
> @@ -64,6 +64,7 @@ enum {
>          #define SET_LSA       0x3
>      MEDIA_AND_POISON = 0x43,
>          #define GET_POISON_LIST        0x0
> +        #define INJECT_POISON          0x1
>  };
>  
>  /* 8.2.8.4.5.1 Command Return Codes */
> @@ -472,6 +473,44 @@ static CXLRetCode cmd_media_get_poison_list(struct cxl_cmd *cmd,
>      return CXL_MBOX_SUCCESS;
>  }
>  
> +static CXLRetCode cmd_media_inject_poison(struct cxl_cmd *cmd,
> +                                          CXLDeviceState *cxl_dstate,
> +                                          uint16_t *len)
> +{
> +    CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate);
> +    CXLPoisonList *poison_list = &ct3d->poison_list;
> +    CXLPoison *ent;
> +    struct inject_poison_pl {
> +        uint64_t dpa;
> +    };
> +    struct inject_poison_pl *in = (void *)cmd->payload;
> +    uint64_t dpa = ldq_le_p(&in->dpa);
> +    CXLPoison *p;
> +
> +    QLIST_FOREACH(ent, poison_list, node) {
> +        if (dpa >= ent->start && dpa + 64 <= ent->start + ent->length) {
> +            return CXL_MBOX_SUCCESS;
> +        }
> +    }
> +
> +    if (ct3d->poison_list_cnt == CXL_POISON_LIST_LIMIT) {
> +        return CXL_MBOX_INJECT_POISON_LIMIT;
> +    }
> +    p = g_new0(CXLPoison, 1);
> +
> +    p->length = 64;
> +    p->start = dpa;
> +    p->type = CXL_POISON_TYPE_INJECTED;
> +
> +    /*
> +     * Possible todo: Merge with existing entry if next to it and if same type
> +     */
> +    QLIST_INSERT_HEAD(poison_list, p, node);
> +    ct3d->poison_list_cnt++;
> +
> +    return CXL_MBOX_SUCCESS;
> +}
> +
>  #define IMMEDIATE_CONFIG_CHANGE (1 << 1)
>  #define IMMEDIATE_DATA_CHANGE (1 << 2)
>  #define IMMEDIATE_POLICY_CHANGE (1 << 3)
> @@ -501,6 +540,8 @@ static struct cxl_cmd cxl_cmd_set[256][256] = {
>          ~0, IMMEDIATE_CONFIG_CHANGE | IMMEDIATE_DATA_CHANGE },
>      [MEDIA_AND_POISON][GET_POISON_LIST] = { "MEDIA_AND_POISON_GET_POISON_LIST",
>          cmd_media_get_poison_list, 16, 0 },
> +    [MEDIA_AND_POISON][INJECT_POISON] = { "MEDIA_AND_POISON_INJECT_POISON",
> +        cmd_media_inject_poison, 8, 0 },
>  };
>  
>  void cxl_process_mailbox(CXLDeviceState *cxl_dstate)
> -- 
> 2.37.2
>
Philippe Mathieu-Daudé March 14, 2023, 6:27 a.m. UTC | #2
On 3/3/23 16:09, Jonathan Cameron wrote:
> Very simple implementation to allow testing of corresponding
> kernel code. Note that for now we track each 64 byte section
> independently.  Whilst a valid implementation choice, it may
> make sense to fuse entries so as to prove out more complex
> corners of the kernel code.
> 
> Reviewed-by: Ira Weiny <ira.weiny@intel.com>
> Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> ---
> v4: No change
> ---
>   hw/cxl/cxl-mailbox-utils.c | 41 ++++++++++++++++++++++++++++++++++++++
>   1 file changed, 41 insertions(+)


> +static CXLRetCode cmd_media_inject_poison(struct cxl_cmd *cmd,
> +                                          CXLDeviceState *cxl_dstate,
> +                                          uint16_t *len)
> +{
> +    CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate);

This makes me wonder why CXLDeviceState isn't QDev based.

(Also, why include/hw/cxl/cxl_device.h is under GPL-2.0-only license?)
Jonathan Cameron April 21, 2023, 5:33 p.m. UTC | #3
On Tue, 14 Mar 2023 07:27:52 +0100
Philippe Mathieu-Daudé <philmd@linaro.org> wrote:

> On 3/3/23 16:09, Jonathan Cameron wrote:
> > Very simple implementation to allow testing of corresponding
> > kernel code. Note that for now we track each 64 byte section
> > independently.  Whilst a valid implementation choice, it may
> > make sense to fuse entries so as to prove out more complex
> > corners of the kernel code.
> > 
> > Reviewed-by: Ira Weiny <ira.weiny@intel.com>
> > Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> > ---
> > v4: No change
> > ---
> >   hw/cxl/cxl-mailbox-utils.c | 41 ++++++++++++++++++++++++++++++++++++++
> >   1 file changed, 41 insertions(+)  
> 
> 
> > +static CXLRetCode cmd_media_inject_poison(struct cxl_cmd *cmd,
> > +                                          CXLDeviceState *cxl_dstate,
> > +                                          uint16_t *len)
> > +{
> > +    CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate);  
> 
> This makes me wonder why CXLDeviceState isn't QDev based.

Interesting question that I'll look into, but I hope you don't mind if
I separate that question from this series.

Logically it's a one of a couple of different subsets of functionality and
different CXL components have a different mix of those. I'm not sure
that will map to a QDev based approach. I'll need to take more time to look into
this.

> 
> (Also, why include/hw/cxl/cxl_device.h is under GPL-2.0-only license?)

Not a clue.   Ben, any comment?
diff mbox series

Patch

diff --git a/hw/cxl/cxl-mailbox-utils.c b/hw/cxl/cxl-mailbox-utils.c
index 25933cf62c..64a3f3c1bf 100644
--- a/hw/cxl/cxl-mailbox-utils.c
+++ b/hw/cxl/cxl-mailbox-utils.c
@@ -64,6 +64,7 @@  enum {
         #define SET_LSA       0x3
     MEDIA_AND_POISON = 0x43,
         #define GET_POISON_LIST        0x0
+        #define INJECT_POISON          0x1
 };
 
 /* 8.2.8.4.5.1 Command Return Codes */
@@ -472,6 +473,44 @@  static CXLRetCode cmd_media_get_poison_list(struct cxl_cmd *cmd,
     return CXL_MBOX_SUCCESS;
 }
 
+static CXLRetCode cmd_media_inject_poison(struct cxl_cmd *cmd,
+                                          CXLDeviceState *cxl_dstate,
+                                          uint16_t *len)
+{
+    CXLType3Dev *ct3d = container_of(cxl_dstate, CXLType3Dev, cxl_dstate);
+    CXLPoisonList *poison_list = &ct3d->poison_list;
+    CXLPoison *ent;
+    struct inject_poison_pl {
+        uint64_t dpa;
+    };
+    struct inject_poison_pl *in = (void *)cmd->payload;
+    uint64_t dpa = ldq_le_p(&in->dpa);
+    CXLPoison *p;
+
+    QLIST_FOREACH(ent, poison_list, node) {
+        if (dpa >= ent->start && dpa + 64 <= ent->start + ent->length) {
+            return CXL_MBOX_SUCCESS;
+        }
+    }
+
+    if (ct3d->poison_list_cnt == CXL_POISON_LIST_LIMIT) {
+        return CXL_MBOX_INJECT_POISON_LIMIT;
+    }
+    p = g_new0(CXLPoison, 1);
+
+    p->length = 64;
+    p->start = dpa;
+    p->type = CXL_POISON_TYPE_INJECTED;
+
+    /*
+     * Possible todo: Merge with existing entry if next to it and if same type
+     */
+    QLIST_INSERT_HEAD(poison_list, p, node);
+    ct3d->poison_list_cnt++;
+
+    return CXL_MBOX_SUCCESS;
+}
+
 #define IMMEDIATE_CONFIG_CHANGE (1 << 1)
 #define IMMEDIATE_DATA_CHANGE (1 << 2)
 #define IMMEDIATE_POLICY_CHANGE (1 << 3)
@@ -501,6 +540,8 @@  static struct cxl_cmd cxl_cmd_set[256][256] = {
         ~0, IMMEDIATE_CONFIG_CHANGE | IMMEDIATE_DATA_CHANGE },
     [MEDIA_AND_POISON][GET_POISON_LIST] = { "MEDIA_AND_POISON_GET_POISON_LIST",
         cmd_media_get_poison_list, 16, 0 },
+    [MEDIA_AND_POISON][INJECT_POISON] = { "MEDIA_AND_POISON_INJECT_POISON",
+        cmd_media_inject_poison, 8, 0 },
 };
 
 void cxl_process_mailbox(CXLDeviceState *cxl_dstate)