linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Li Zhijian <lizhijian@fujitsu.com>
To: Alison Schofield <alison.schofield@intel.com>, terry.bowman@amd.com
Cc: y-goto@fujitsu.com, nafonten@amd.com, david@redhat.com,
	nifan.cxl@gmail.com, linux-cxl@vger.kernel.org,
	dan.j.williams@intel.com, linux-mm@kvack.org, gourry@gourry.net
Subject: Re: [PATCH v2 1/4] kernel/resource: Introduce managed SOFT RESERVED resources
Date: Mon, 10 Mar 2025 13:52:34 +0800	[thread overview]
Message-ID: <20250310055234.3704571-1-lizhijian@fujitsu.com> (raw)
In-Reply-To: <Z8sjPkLpcmM8G93K@aschofie-mobl2.lan>

>On Fri, Mar 07, 2025 at 05:56:25AM +0000, Zhijian Li (Fujitsu) wrote:
>> Hello Fontenot,
>> 
>> I hope this email finds you well.
>> 
>> Thank you very much for this patch. We've encountered the same issue in our product,
>> and your patch works.
>> 
>> We do hope this issue will be resolved in the upstream kernel soon.
>> 
>
>snip
>
>> 
>> 
>> I noticed your earlier discussions about designing a new approach to solve this problem,
>> which I'm pretty excited about. Do you have any idea when you might post the updated
>> version? We'd love to help out with reviewing and testing..
>> 
>> If you run into any roadblocks and need a hand, just let us know. we'd be delighted to help.
>> 
>> As far as I know, this issue usually arises on Real CXL machines, but for ease of testing
>> and validation, we modified QEMU to simulate the intersection of 'Soft Reserved' and
>> the CXL region, which would aid in verification.
>
>Hi Zhijian,
>
>Wow - I want that! Can you share that QEMU branch?  With QEMU cmdline too.
>

Hi Terry and Alison

Well, it's a simple hack within the QEMU and seabios project. Their
modifications are as below:

Note: the QEMU's modification exposed the CFMWs region instead of the CXL
memory region(that the Real CXL machine did)

After the guest started, you could see the iomem tree:

a90000000-1a8fffffff : CXL Window 0
  a90000000-1a8fffffff : Soft Reserved


QEMU:
=====================
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
index f199a8c7ad19..484ad7e5e632 100644
--- a/hw/i386/pc.c
+++ b/hw/i386/pc.c
@@ -963,6 +963,9 @@ void pc_memory_init(PCMachineState *pcms,
                 memory_region_init_io(&fw->mr, OBJECT(machine), &cfmws_ops, fw,
                                       "cxl-fixed-memory-region", fw->size);
                 memory_region_add_subregion(system_memory, fw->base, &fw->mr);
+#define E820_TYPE_SOFT_RESERVED 0xefffffff
+                /* add special purpose memory */
+                e820_add_entry(fw->mr.addr, fw->mr.size , E820_TYPE_SOFT_RESERVED);
                 cxl_fmw_base += fw->size;
                 cxl_resv_end = cxl_fmw_base;
             }


seabios
=====================
diff --git a/src/e820map.c b/src/e820map.c
index c761e5e98a75..9440039541a6 100644
--- a/src/e820map.c
+++ b/src/e820map.c
@@ -54,6 +54,7 @@ e820_type_name(u32 type)
     case E820_ACPI:     return "ACPI";
     case E820_NVS:      return "NVS";
     case E820_UNUSABLE: return "UNUSABLE";
+    case E820_TYPE_SOFT_RESERVED: return "Soft Reserved";
     default:            return "UNKNOWN";
     }
 }
diff --git a/src/e820map.h b/src/e820map.h
index 07ce16ec213f..dd416d5ba3df 100644
--- a/src/e820map.h
+++ b/src/e820map.h
@@ -8,6 +8,7 @@
 #define E820_ACPI         3
 #define E820_NVS          4
 #define E820_UNUSABLE     5
+#define E820_TYPE_SOFT_RESERVED 0xefffffff
 
 struct e820entry {
     u64 start;
diff --git a/src/fw/paravirt.c b/src/fw/paravirt.c
index e5d4eca0cb5a..38a9bfed04df 100644
--- a/src/fw/paravirt.c
+++ b/src/fw/paravirt.c
@@ -781,6 +781,12 @@ static int qemu_early_e820(void)
                 if (RamSizeOver4G < table.address + table.length - 0x100000000LL)
                     RamSizeOver4G = table.address + table.length - 0x100000000LL;
             }
+            break;
+        case E820_TYPE_SOFT_RESERVED:
+            e820_add(table.address, table.length, table.type);
+            dprintf(1, "qemu/e820: addr 0x%016llx len 0x%016llx [Sort reserved]\n",
+                 table.address, table.length);
+            break;
         }
     }


====================
QEMU comand line(nothing special to QEMU expect specifying your own bios):

/path/to/qemu <...args...> \
-machine type=q35,cxl=on \
-bios /path/to/seabios/out/bios.bin \
-nographic \
-object memory-backend-ram,id=vmem0,share=on,size=4G \
-device pxb-cxl,bus_nr=12,bus=pcie.0,id=cxl.1 \
-device cxl-rp,port=0,bus=cxl.1,id=root_port13,chassis=0,slot=2 \
-device cxl-type3,bus=root_port13,volatile-memdev=vmem0,id=cxl-vmem0 \
-M cxl-fmw.0.targets.0=cxl.1,cxl-fmw.0.size=64G




  reply	other threads:[~2025-03-10  5:52 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-01-16 17:42 [PATCH v2 0/4] Add managed SOFT RESERVE resource handling Nathan Fontenot
2025-01-16 17:42 ` [PATCH v2 1/4] kernel/resource: Introduce managed SOFT RESERVED resources Nathan Fontenot
2025-01-21  8:19   ` David Hildenbrand
2025-01-21 18:57     ` Fontenot, Nathan
2025-01-22  6:03       ` Fan Ni
2025-01-23 15:49         ` Fontenot, Nathan
2025-01-27 14:40           ` David Hildenbrand
2025-01-27 18:46             ` Fontenot, Nathan
2025-03-07  5:56               ` Zhijian Li (Fujitsu)
2025-03-07 16:47                 ` Alison Schofield
2025-03-10  5:52                   ` Li Zhijian [this message]
2025-03-07 23:05                 ` Bowman, Terry
2025-03-10  6:00                   ` Zhijian Li (Fujitsu)
2025-03-23  8:24                   ` Zhijian Li (Fujitsu)
2025-03-23  8:33                     ` Zhijian Li (Fujitsu)
2025-01-22  5:52   ` Fan Ni
2025-01-23 15:55     ` Fontenot, Nathan
2025-01-16 17:42 ` [PATCH v2 2/4] cxl: Update Soft Reserve resources upon region creation Nathan Fontenot
2025-01-16 17:42 ` [PATCH v2 3/4] dax: Update hmem resource/device registration Nathan Fontenot
2025-01-16 22:28   ` Ira Weiny
2025-01-21 18:49     ` Fontenot, Nathan
2025-01-21 23:14       ` Ira Weiny
2025-01-23 16:01         ` Fontenot, Nathan
2025-01-27 18:44           ` Fontenot, Nathan
2025-01-16 17:42 ` [PATCH v2 4/4] Add SOFT RESERVE resource notification chain Nathan Fontenot

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=20250310055234.3704571-1-lizhijian@fujitsu.com \
    --to=lizhijian@fujitsu.com \
    --cc=alison.schofield@intel.com \
    --cc=dan.j.williams@intel.com \
    --cc=david@redhat.com \
    --cc=gourry@gourry.net \
    --cc=linux-cxl@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=nafonten@amd.com \
    --cc=nifan.cxl@gmail.com \
    --cc=terry.bowman@amd.com \
    --cc=y-goto@fujitsu.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