import qemu-kvm-6.2.0-28.module+el8.8.0+17721+63bacee8

This commit is contained in:
CentOS Sources 2023-01-31 01:26:15 +00:00 committed by root
parent 82e468bb05
commit 38978d44db
62 changed files with 8183 additions and 30 deletions

View File

@ -0,0 +1,171 @@
From 10fc28b61a6fba1e6dc44fd544cf31c7f313c622 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?C=C3=A9dric=20Le=20Goater?= <clg@redhat.com>
Date: Fri, 28 Oct 2022 17:48:00 +0100
Subject: [PATCH 05/42] Update linux headers to v6.0-rc4
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [5/41] ca55f497d1bf1e72179330f8f613781bf999d898
Based on upstream commit d525f73f9186a5bc641b8caf0b2c9bb94e5aa963
("Update linux headers to v6.0-rc4"), but this is focusing only on the
ZPCI and protected dump changes.
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
linux-headers/linux/kvm.h | 87 +++++++++++++++++++++++++++++++++
linux-headers/linux/vfio_zdev.h | 7 +++
2 files changed, 94 insertions(+)
diff --git a/linux-headers/linux/kvm.h b/linux-headers/linux/kvm.h
index 0d05d02ee4..c65930288c 100644
--- a/linux-headers/linux/kvm.h
+++ b/linux-headers/linux/kvm.h
@@ -1150,6 +1150,9 @@ struct kvm_ppc_resize_hpt {
#define KVM_CAP_DISABLE_QUIRKS2 213
/* #define KVM_CAP_VM_TSC_CONTROL 214 */
#define KVM_CAP_SYSTEM_EVENT_DATA 215
+#define KVM_CAP_S390_PROTECTED_DUMP 217
+#define KVM_CAP_S390_ZPCI_OP 221
+#define KVM_CAP_S390_CPU_TOPOLOGY 222
#ifdef KVM_CAP_IRQ_ROUTING
@@ -1651,6 +1654,55 @@ struct kvm_s390_pv_unp {
__u64 tweak;
};
+enum pv_cmd_dmp_id {
+ KVM_PV_DUMP_INIT,
+ KVM_PV_DUMP_CONFIG_STOR_STATE,
+ KVM_PV_DUMP_COMPLETE,
+ KVM_PV_DUMP_CPU,
+};
+
+struct kvm_s390_pv_dmp {
+ __u64 subcmd;
+ __u64 buff_addr;
+ __u64 buff_len;
+ __u64 gaddr; /* For dump storage state */
+ __u64 reserved[4];
+};
+
+enum pv_cmd_info_id {
+ KVM_PV_INFO_VM,
+ KVM_PV_INFO_DUMP,
+};
+
+struct kvm_s390_pv_info_dump {
+ __u64 dump_cpu_buffer_len;
+ __u64 dump_config_mem_buffer_per_1m;
+ __u64 dump_config_finalize_len;
+};
+
+struct kvm_s390_pv_info_vm {
+ __u64 inst_calls_list[4];
+ __u64 max_cpus;
+ __u64 max_guests;
+ __u64 max_guest_addr;
+ __u64 feature_indication;
+};
+
+struct kvm_s390_pv_info_header {
+ __u32 id;
+ __u32 len_max;
+ __u32 len_written;
+ __u32 reserved;
+};
+
+struct kvm_s390_pv_info {
+ struct kvm_s390_pv_info_header header;
+ union {
+ struct kvm_s390_pv_info_dump dump;
+ struct kvm_s390_pv_info_vm vm;
+ };
+};
+
enum pv_cmd_id {
KVM_PV_ENABLE,
KVM_PV_DISABLE,
@@ -1659,6 +1711,8 @@ enum pv_cmd_id {
KVM_PV_VERIFY,
KVM_PV_PREP_RESET,
KVM_PV_UNSHARE_ALL,
+ KVM_PV_INFO,
+ KVM_PV_DUMP,
};
struct kvm_pv_cmd {
@@ -2066,4 +2120,37 @@ struct kvm_stats_desc {
/* Available with KVM_CAP_XSAVE2 */
#define KVM_GET_XSAVE2 _IOR(KVMIO, 0xcf, struct kvm_xsave)
+/* Available with KVM_CAP_S390_PROTECTED_DUMP */
+#define KVM_S390_PV_CPU_COMMAND _IOWR(KVMIO, 0xd0, struct kvm_pv_cmd)
+
+/* Available with KVM_CAP_S390_ZPCI_OP */
+#define KVM_S390_ZPCI_OP _IOW(KVMIO, 0xd1, struct kvm_s390_zpci_op)
+
+struct kvm_s390_zpci_op {
+ /* in */
+ __u32 fh; /* target device */
+ __u8 op; /* operation to perform */
+ __u8 pad[3];
+ union {
+ /* for KVM_S390_ZPCIOP_REG_AEN */
+ struct {
+ __u64 ibv; /* Guest addr of interrupt bit vector */
+ __u64 sb; /* Guest addr of summary bit */
+ __u32 flags;
+ __u32 noi; /* Number of interrupts */
+ __u8 isc; /* Guest interrupt subclass */
+ __u8 sbo; /* Offset of guest summary bit vector */
+ __u16 pad;
+ } reg_aen;
+ __u64 reserved[8];
+ } u;
+};
+
+/* types for kvm_s390_zpci_op->op */
+#define KVM_S390_ZPCIOP_REG_AEN 0
+#define KVM_S390_ZPCIOP_DEREG_AEN 1
+
+/* flags for kvm_s390_zpci_op->u.reg_aen.flags */
+#define KVM_S390_ZPCIOP_REGAEN_HOST (1 << 0)
+
#endif /* __LINUX_KVM_H */
diff --git a/linux-headers/linux/vfio_zdev.h b/linux-headers/linux/vfio_zdev.h
index b4309397b6..77f2aff1f2 100644
--- a/linux-headers/linux/vfio_zdev.h
+++ b/linux-headers/linux/vfio_zdev.h
@@ -29,6 +29,9 @@ struct vfio_device_info_cap_zpci_base {
__u16 fmb_length; /* Measurement Block Length (in bytes) */
__u8 pft; /* PCI Function Type */
__u8 gid; /* PCI function group ID */
+ /* End of version 1 */
+ __u32 fh; /* PCI function handle */
+ /* End of version 2 */
};
/**
@@ -47,6 +50,10 @@ struct vfio_device_info_cap_zpci_group {
__u16 noi; /* Maximum number of MSIs */
__u16 maxstbl; /* Maximum Store Block Length */
__u8 version; /* Supported PCI Version */
+ /* End of version 1 */
+ __u8 reserved;
+ __u16 imaxstbl; /* Maximum Interpreted Store Block Length */
+ /* End of version 2 */
};
/**
--
2.37.3

View File

@ -0,0 +1,71 @@
From 60da56e3685969493ae483c3cc2c66af13d00baf Mon Sep 17 00:00:00 2001
From: Thomas Huth <thuth@redhat.com>
Date: Wed, 10 Aug 2022 14:57:18 +0200
Subject: [PATCH 1/3] backends/hostmem: Fix support of memory-backend-memfd in
qemu_maxrampagesize()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <None>
RH-MergeRequest: 221: backends/hostmem: Fix support of memory-backend-memfd in qemu_maxrampagesize()
RH-Bugzilla: 2117149
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: David Hildenbrand <david@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Commit: [1/1] b5a1047750af32c0a261b8385ea0e819eb16681a
It is currently not possible yet to use "memory-backend-memfd" on s390x
with hugepages enabled. This problem is caused by qemu_maxrampagesize()
not taking memory-backend-memfd objects into account yet, so the code
in s390_memory_init() fails to enable the huge page support there via
s390_set_max_pagesize(). Fix it by generalizing the code, so that it
looks at qemu_ram_pagesize(memdev->mr.ram_block) instead of re-trying
to get the information from the filesystem.
Suggested-by: David Hildenbrand <david@redhat.com>
Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=2116496
Message-Id: <20220810125720.3849835-2-thuth@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit 8be934b70e923104da883b990dee18f02552d40e)
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2117149
[clg: Resolved conflict on qemu_real_host_page_size() ]
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
backends/hostmem.c | 14 ++------------
1 file changed, 2 insertions(+), 12 deletions(-)
diff --git a/backends/hostmem.c b/backends/hostmem.c
index 4c05862ed5..0c4654ea85 100644
--- a/backends/hostmem.c
+++ b/backends/hostmem.c
@@ -305,22 +305,12 @@ bool host_memory_backend_is_mapped(HostMemoryBackend *backend)
return backend->is_mapped;
}
-#ifdef __linux__
size_t host_memory_backend_pagesize(HostMemoryBackend *memdev)
{
- Object *obj = OBJECT(memdev);
- char *path = object_property_get_str(obj, "mem-path", NULL);
- size_t pagesize = qemu_mempath_getpagesize(path);
-
- g_free(path);
+ size_t pagesize = qemu_ram_pagesize(memdev->mr.ram_block);
+ g_assert(pagesize >= qemu_real_host_page_size);
return pagesize;
}
-#else
-size_t host_memory_backend_pagesize(HostMemoryBackend *memdev)
-{
- return qemu_real_host_page_size;
-}
-#endif
static void
host_memory_backend_memory_complete(UserCreatable *uc, Error **errp)
--
2.35.3

View File

@ -0,0 +1,70 @@
From 407e23d7f0c9020404247afe7d4df98505222bbb Mon Sep 17 00:00:00 2001
From: Thomas Huth <thuth@redhat.com>
Date: Mon, 14 Nov 2022 14:25:02 +0100
Subject: [PATCH 1/3] docs/system/s390x: Document the "loadparm" machine
property
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Thomas Huth <thuth@redhat.com>
RH-MergeRequest: 233: s390x: Document the "loadparm" machine property
RH-Bugzilla: 2128225
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Cédric Le Goater <clg@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [1/2] e9589ea32d2a8f82971476b644e1063fa14cf822
The "loadparm" machine property is useful for selecting alternative
kernels on the disk of the guest, but so far we do not tell the users
yet how to use it. Add some documentation to fill this gap.
Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=2128235
Message-Id: <20221114132502.110213-1-thuth@redhat.com>
Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit be5df2edb5d69ff3107c5616aa035a9ba8d0422e)
---
docs/system/s390x/bootdevices.rst | 26 ++++++++++++++++++++++++++
1 file changed, 26 insertions(+)
diff --git a/docs/system/s390x/bootdevices.rst b/docs/system/s390x/bootdevices.rst
index 9e591cb9dc..d4bf3b9f0b 100644
--- a/docs/system/s390x/bootdevices.rst
+++ b/docs/system/s390x/bootdevices.rst
@@ -53,6 +53,32 @@ recommended to specify a CD-ROM device via ``-device scsi-cd`` (as mentioned
above) instead.
+Selecting kernels with the ``loadparm`` property
+------------------------------------------------
+
+The ``s390-ccw-virtio`` machine supports the so-called ``loadparm`` parameter
+which can be used to select the kernel on the disk of the guest that the
+s390-ccw bios should boot. When starting QEMU, it can be specified like this::
+
+ qemu-system-s390x -machine s390-ccw-virtio,loadparm=<string>
+
+The first way to use this parameter is to use the word ``PROMPT`` as the
+``<string>`` here. In that case the s390-ccw bios will show a list of
+installed kernels on the disk of the guest and ask the user to enter a number
+to chose which kernel should be booted -- similar to what can be achieved by
+specifying the ``-boot menu=on`` option when starting QEMU. Note that the menu
+list will only show the names of the installed kernels when using a DASD-like
+disk image with 4k byte sectors. On normal SCSI-style disks with 512-byte
+sectors, there is not enough space for the zipl loader on the disk to store
+the kernel names, so you only get a list without names here.
+
+The second way to use this parameter is to use a number in the range from 0
+to 31. The numbers that can be used here correspond to the numbers that are
+shown when using the ``PROMPT`` option, and the s390-ccw bios will then try
+to automatically boot the kernel that is associated with the given number.
+Note that ``0`` can be used to boot the default entry.
+
+
Booting from a network device
-----------------------------
--
2.37.3

View File

@ -0,0 +1,356 @@
From f2f3efff83dddd38a97699cd2701f46f61a732e3 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Mon, 17 Oct 2022 11:32:10 +0000
Subject: [PATCH 36/42] dump: Add architecture section and section string table
support
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [36/41] 83b98ff185e93e62703f686b65546d60c783d783
Add hooks which architectures can use to add arbitrary data to custom
sections.
Also add a section name string table in order to identify section
contents
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20221017113210.41674-1-frankja@linux.ibm.com>
(cherry picked from commit 9b72224f44612ddd5b434a1bccf79346946d11da)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 186 +++++++++++++++++++++++++++++++------
include/sysemu/dump-arch.h | 3 +
include/sysemu/dump.h | 3 +
3 files changed, 166 insertions(+), 26 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 7a42401790..4aa8fb64d2 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -104,6 +104,7 @@ static int dump_cleanup(DumpState *s)
memory_mapping_list_free(&s->list);
close(s->fd);
g_free(s->guest_note);
+ g_array_unref(s->string_table_buf);
s->guest_note = NULL;
if (s->resume) {
if (s->detached) {
@@ -153,11 +154,10 @@ static void prepare_elf64_header(DumpState *s, Elf64_Ehdr *elf_header)
elf_header->e_phoff = cpu_to_dump64(s, s->phdr_offset);
elf_header->e_phentsize = cpu_to_dump16(s, sizeof(Elf64_Phdr));
elf_header->e_phnum = cpu_to_dump16(s, phnum);
- if (s->shdr_num) {
- elf_header->e_shoff = cpu_to_dump64(s, s->shdr_offset);
- elf_header->e_shentsize = cpu_to_dump16(s, sizeof(Elf64_Shdr));
- elf_header->e_shnum = cpu_to_dump16(s, s->shdr_num);
- }
+ elf_header->e_shoff = cpu_to_dump64(s, s->shdr_offset);
+ elf_header->e_shentsize = cpu_to_dump16(s, sizeof(Elf64_Shdr));
+ elf_header->e_shnum = cpu_to_dump16(s, s->shdr_num);
+ elf_header->e_shstrndx = cpu_to_dump16(s, s->shdr_num - 1);
}
static void prepare_elf32_header(DumpState *s, Elf32_Ehdr *elf_header)
@@ -181,11 +181,10 @@ static void prepare_elf32_header(DumpState *s, Elf32_Ehdr *elf_header)
elf_header->e_phoff = cpu_to_dump32(s, s->phdr_offset);
elf_header->e_phentsize = cpu_to_dump16(s, sizeof(Elf32_Phdr));
elf_header->e_phnum = cpu_to_dump16(s, phnum);
- if (s->shdr_num) {
- elf_header->e_shoff = cpu_to_dump32(s, s->shdr_offset);
- elf_header->e_shentsize = cpu_to_dump16(s, sizeof(Elf32_Shdr));
- elf_header->e_shnum = cpu_to_dump16(s, s->shdr_num);
- }
+ elf_header->e_shoff = cpu_to_dump32(s, s->shdr_offset);
+ elf_header->e_shentsize = cpu_to_dump16(s, sizeof(Elf32_Shdr));
+ elf_header->e_shnum = cpu_to_dump16(s, s->shdr_num);
+ elf_header->e_shstrndx = cpu_to_dump16(s, s->shdr_num - 1);
}
static void write_elf_header(DumpState *s, Error **errp)
@@ -196,6 +195,8 @@ static void write_elf_header(DumpState *s, Error **errp)
void *header_ptr;
int ret;
+ /* The NULL header and the shstrtab are always defined */
+ assert(s->shdr_num >= 2);
if (dump_is_64bit(s)) {
prepare_elf64_header(s, &elf64_header);
header_size = sizeof(elf64_header);
@@ -394,17 +395,49 @@ static void prepare_elf_section_hdr_zero(DumpState *s)
}
}
-static void prepare_elf_section_hdrs(DumpState *s)
+static void prepare_elf_section_hdr_string(DumpState *s, void *buff)
+{
+ uint64_t index = s->string_table_buf->len;
+ const char strtab[] = ".shstrtab";
+ Elf32_Shdr shdr32 = {};
+ Elf64_Shdr shdr64 = {};
+ int shdr_size;
+ void *shdr;
+
+ g_array_append_vals(s->string_table_buf, strtab, sizeof(strtab));
+ if (dump_is_64bit(s)) {
+ shdr_size = sizeof(Elf64_Shdr);
+ shdr64.sh_type = SHT_STRTAB;
+ shdr64.sh_offset = s->section_offset + s->elf_section_data_size;
+ shdr64.sh_name = index;
+ shdr64.sh_size = s->string_table_buf->len;
+ shdr = &shdr64;
+ } else {
+ shdr_size = sizeof(Elf32_Shdr);
+ shdr32.sh_type = SHT_STRTAB;
+ shdr32.sh_offset = s->section_offset + s->elf_section_data_size;
+ shdr32.sh_name = index;
+ shdr32.sh_size = s->string_table_buf->len;
+ shdr = &shdr32;
+ }
+ memcpy(buff, shdr, shdr_size);
+}
+
+static bool prepare_elf_section_hdrs(DumpState *s, Error **errp)
{
size_t len, sizeof_shdr;
+ void *buff_hdr;
/*
* Section ordering:
* - HDR zero
+ * - Arch section hdrs
+ * - String table hdr
*/
sizeof_shdr = dump_is_64bit(s) ? sizeof(Elf64_Shdr) : sizeof(Elf32_Shdr);
len = sizeof_shdr * s->shdr_num;
s->elf_section_hdrs = g_malloc0(len);
+ buff_hdr = s->elf_section_hdrs;
/*
* The first section header is ALWAYS a special initial section
@@ -420,6 +453,26 @@ static void prepare_elf_section_hdrs(DumpState *s)
if (s->phdr_num >= PN_XNUM) {
prepare_elf_section_hdr_zero(s);
}
+ buff_hdr += sizeof_shdr;
+
+ /* Add architecture defined section headers */
+ if (s->dump_info.arch_sections_write_hdr_fn
+ && s->shdr_num > 2) {
+ buff_hdr += s->dump_info.arch_sections_write_hdr_fn(s, buff_hdr);
+
+ if (s->shdr_num >= SHN_LORESERVE) {
+ error_setg_errno(errp, EINVAL,
+ "dump: too many architecture defined sections");
+ return false;
+ }
+ }
+
+ /*
+ * String table is the last section since strings are added via
+ * arch_sections_write_hdr().
+ */
+ prepare_elf_section_hdr_string(s, buff_hdr);
+ return true;
}
static void write_elf_section_headers(DumpState *s, Error **errp)
@@ -427,7 +480,9 @@ static void write_elf_section_headers(DumpState *s, Error **errp)
size_t sizeof_shdr = dump_is_64bit(s) ? sizeof(Elf64_Shdr) : sizeof(Elf32_Shdr);
int ret;
- prepare_elf_section_hdrs(s);
+ if (!prepare_elf_section_hdrs(s, errp)) {
+ return;
+ }
ret = fd_write_vmcore(s->elf_section_hdrs, s->shdr_num * sizeof_shdr, s);
if (ret < 0) {
@@ -437,6 +492,29 @@ static void write_elf_section_headers(DumpState *s, Error **errp)
g_free(s->elf_section_hdrs);
}
+static void write_elf_sections(DumpState *s, Error **errp)
+{
+ int ret;
+
+ if (s->elf_section_data_size) {
+ /* Write architecture section data */
+ ret = fd_write_vmcore(s->elf_section_data,
+ s->elf_section_data_size, s);
+ if (ret < 0) {
+ error_setg_errno(errp, -ret,
+ "dump: failed to write architecture section data");
+ return;
+ }
+ }
+
+ /* Write string table */
+ ret = fd_write_vmcore(s->string_table_buf->data,
+ s->string_table_buf->len, s);
+ if (ret < 0) {
+ error_setg_errno(errp, -ret, "dump: failed to write string table data");
+ }
+}
+
static void write_data(DumpState *s, void *buf, int length, Error **errp)
{
int ret;
@@ -693,6 +771,31 @@ static void dump_iterate(DumpState *s, Error **errp)
}
}
+static void dump_end(DumpState *s, Error **errp)
+{
+ int rc;
+ ERRP_GUARD();
+
+ if (s->elf_section_data_size) {
+ s->elf_section_data = g_malloc0(s->elf_section_data_size);
+ }
+
+ /* Adds the architecture defined section data to s->elf_section_data */
+ if (s->dump_info.arch_sections_write_fn &&
+ s->elf_section_data_size) {
+ rc = s->dump_info.arch_sections_write_fn(s, s->elf_section_data);
+ if (rc) {
+ error_setg_errno(errp, rc,
+ "dump: failed to get arch section data");
+ g_free(s->elf_section_data);
+ return;
+ }
+ }
+
+ /* write sections to vmcore */
+ write_elf_sections(s, errp);
+}
+
static void create_vmcore(DumpState *s, Error **errp)
{
ERRP_GUARD();
@@ -702,7 +805,14 @@ static void create_vmcore(DumpState *s, Error **errp)
return;
}
+ /* Iterate over memory and dump it to file */
dump_iterate(s, errp);
+ if (*errp) {
+ return;
+ }
+
+ /* Write the section data */
+ dump_end(s, errp);
}
static int write_start_flat_header(int fd)
@@ -1720,6 +1830,14 @@ static void dump_init(DumpState *s, int fd, bool has_format,
s->filter_area_begin = begin;
s->filter_area_length = length;
+ /* First index is 0, it's the special null name */
+ s->string_table_buf = g_array_new(FALSE, TRUE, 1);
+ /*
+ * Allocate the null name, due to the clearing option set to true
+ * it will be 0.
+ */
+ g_array_set_size(s->string_table_buf, 1);
+
memory_mapping_list_init(&s->list);
guest_phys_blocks_init(&s->guest_phys_blocks);
@@ -1856,26 +1974,42 @@ static void dump_init(DumpState *s, int fd, bool has_format,
}
/*
- * calculate phdr_num
+ * The first section header is always a special one in which most
+ * fields are 0. The section header string table is also always
+ * set.
+ */
+ s->shdr_num = 2;
+
+ /*
+ * Adds the number of architecture sections to shdr_num and sets
+ * elf_section_data_size so we know the offsets and sizes of all
+ * parts.
+ */
+ if (s->dump_info.arch_sections_add_fn) {
+ s->dump_info.arch_sections_add_fn(s);
+ }
+
+ /*
+ * calculate shdr_num so we know the offsets and sizes of all
+ * parts.
+ * Calculate phdr_num
*
- * the type of ehdr->e_phnum is uint16_t, so we should avoid overflow
+ * The absolute maximum amount of phdrs is UINT32_MAX - 1 as
+ * sh_info is 32 bit. There's special handling once we go over
+ * UINT16_MAX - 1 but that is handled in the ehdr and section
+ * code.
*/
- s->phdr_num = 1; /* PT_NOTE */
- if (s->list.num < UINT16_MAX - 2) {
- s->shdr_num = 0;
+ s->phdr_num = 1; /* Reserve PT_NOTE */
+ if (s->list.num <= UINT32_MAX - 1) {
s->phdr_num += s->list.num;
} else {
- /* sh_info of section 0 holds the real number of phdrs */
- s->shdr_num = 1;
-
- /* the type of shdr->sh_info is uint32_t, so we should avoid overflow */
- if (s->list.num <= UINT32_MAX - 1) {
- s->phdr_num += s->list.num;
- } else {
- s->phdr_num = UINT32_MAX;
- }
+ s->phdr_num = UINT32_MAX;
}
+ /*
+ * Now that the number of section and program headers is known we
+ * can calculate the offsets of the headers and data.
+ */
if (dump_is_64bit(s)) {
s->shdr_offset = sizeof(Elf64_Ehdr);
s->phdr_offset = s->shdr_offset + sizeof(Elf64_Shdr) * s->shdr_num;
diff --git a/include/sysemu/dump-arch.h b/include/sysemu/dump-arch.h
index e25b02e990..59bbc9be38 100644
--- a/include/sysemu/dump-arch.h
+++ b/include/sysemu/dump-arch.h
@@ -21,6 +21,9 @@ typedef struct ArchDumpInfo {
uint32_t page_size; /* The target's page size. If it's variable and
* unknown, then this should be the maximum. */
uint64_t phys_base; /* The target's physmem base. */
+ void (*arch_sections_add_fn)(DumpState *s);
+ uint64_t (*arch_sections_write_hdr_fn)(DumpState *s, uint8_t *buff);
+ int (*arch_sections_write_fn)(DumpState *s, uint8_t *buff);
} ArchDumpInfo;
struct GuestPhysBlockList; /* memory_mapping.h */
diff --git a/include/sysemu/dump.h b/include/sysemu/dump.h
index 9ed811b313..38ccac7190 100644
--- a/include/sysemu/dump.h
+++ b/include/sysemu/dump.h
@@ -180,6 +180,9 @@ typedef struct DumpState {
hwaddr note_offset;
void *elf_section_hdrs; /* Pointer to section header buffer */
+ void *elf_section_data; /* Pointer to section data buffer */
+ uint64_t elf_section_data_size; /* Size of section data */
+ GArray *string_table_buf; /* String table data buffer */
uint8_t *note_buf; /* buffer for notes */
size_t note_buf_offset; /* the writing place in note_buf */
--
2.37.3

View File

@ -0,0 +1,138 @@
From bee31226b87d0b05faae84e88cce3af1b8dabbfd Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Wed, 30 Mar 2022 12:35:59 +0000
Subject: [PATCH 17/42] dump: Add more offset variables
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [17/41] fbe629e1476e8a0e039f989af6e1f4707075ba01
Offset calculations are easy enough to get wrong. Let's add a few
variables to make moving around elf headers and data sections easier.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20220330123603.107120-6-frankja@linux.ibm.com>
(cherry picked from commit e71d353360bb09a8e784e35d78370c691f6ea185)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 35 +++++++++++++++--------------------
include/sysemu/dump.h | 4 ++++
2 files changed, 19 insertions(+), 20 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 5cc2322325..85a402b38c 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -142,13 +142,11 @@ static void write_elf64_header(DumpState *s, Error **errp)
elf_header.e_machine = cpu_to_dump16(s, s->dump_info.d_machine);
elf_header.e_version = cpu_to_dump32(s, EV_CURRENT);
elf_header.e_ehsize = cpu_to_dump16(s, sizeof(elf_header));
- elf_header.e_phoff = cpu_to_dump64(s, sizeof(Elf64_Ehdr));
+ elf_header.e_phoff = cpu_to_dump64(s, s->phdr_offset);
elf_header.e_phentsize = cpu_to_dump16(s, sizeof(Elf64_Phdr));
elf_header.e_phnum = cpu_to_dump16(s, phnum);
if (s->shdr_num) {
- uint64_t shoff = sizeof(Elf64_Ehdr) + sizeof(Elf64_Phdr) * s->phdr_num;
-
- elf_header.e_shoff = cpu_to_dump64(s, shoff);
+ elf_header.e_shoff = cpu_to_dump64(s, s->shdr_offset);
elf_header.e_shentsize = cpu_to_dump16(s, sizeof(Elf64_Shdr));
elf_header.e_shnum = cpu_to_dump16(s, s->shdr_num);
}
@@ -179,13 +177,11 @@ static void write_elf32_header(DumpState *s, Error **errp)
elf_header.e_machine = cpu_to_dump16(s, s->dump_info.d_machine);
elf_header.e_version = cpu_to_dump32(s, EV_CURRENT);
elf_header.e_ehsize = cpu_to_dump16(s, sizeof(elf_header));
- elf_header.e_phoff = cpu_to_dump32(s, sizeof(Elf32_Ehdr));
+ elf_header.e_phoff = cpu_to_dump32(s, s->phdr_offset);
elf_header.e_phentsize = cpu_to_dump16(s, sizeof(Elf32_Phdr));
elf_header.e_phnum = cpu_to_dump16(s, phnum);
if (s->shdr_num) {
- uint32_t shoff = sizeof(Elf32_Ehdr) + sizeof(Elf32_Phdr) * s->phdr_num;
-
- elf_header.e_shoff = cpu_to_dump32(s, shoff);
+ elf_header.e_shoff = cpu_to_dump32(s, s->shdr_offset);
elf_header.e_shentsize = cpu_to_dump16(s, sizeof(Elf32_Shdr));
elf_header.e_shnum = cpu_to_dump16(s, s->shdr_num);
}
@@ -248,12 +244,11 @@ static void write_elf32_load(DumpState *s, MemoryMapping *memory_mapping,
static void write_elf64_note(DumpState *s, Error **errp)
{
Elf64_Phdr phdr;
- hwaddr begin = s->memory_offset - s->note_size;
int ret;
memset(&phdr, 0, sizeof(Elf64_Phdr));
phdr.p_type = cpu_to_dump32(s, PT_NOTE);
- phdr.p_offset = cpu_to_dump64(s, begin);
+ phdr.p_offset = cpu_to_dump64(s, s->note_offset);
phdr.p_paddr = 0;
phdr.p_filesz = cpu_to_dump64(s, s->note_size);
phdr.p_memsz = cpu_to_dump64(s, s->note_size);
@@ -313,13 +308,12 @@ static void write_elf64_notes(WriteCoreDumpFunction f, DumpState *s,
static void write_elf32_note(DumpState *s, Error **errp)
{
- hwaddr begin = s->memory_offset - s->note_size;
Elf32_Phdr phdr;
int ret;
memset(&phdr, 0, sizeof(Elf32_Phdr));
phdr.p_type = cpu_to_dump32(s, PT_NOTE);
- phdr.p_offset = cpu_to_dump32(s, begin);
+ phdr.p_offset = cpu_to_dump32(s, s->note_offset);
phdr.p_paddr = 0;
phdr.p_filesz = cpu_to_dump32(s, s->note_size);
phdr.p_memsz = cpu_to_dump32(s, s->note_size);
@@ -1826,15 +1820,16 @@ static void dump_init(DumpState *s, int fd, bool has_format,
}
if (s->dump_info.d_class == ELFCLASS64) {
- s->memory_offset = sizeof(Elf64_Ehdr) +
- sizeof(Elf64_Phdr) * s->phdr_num +
- sizeof(Elf64_Shdr) * s->shdr_num +
- s->note_size;
+ s->phdr_offset = sizeof(Elf64_Ehdr);
+ s->shdr_offset = s->phdr_offset + sizeof(Elf64_Phdr) * s->phdr_num;
+ s->note_offset = s->shdr_offset + sizeof(Elf64_Shdr) * s->shdr_num;
+ s->memory_offset = s->note_offset + s->note_size;
} else {
- s->memory_offset = sizeof(Elf32_Ehdr) +
- sizeof(Elf32_Phdr) * s->phdr_num +
- sizeof(Elf32_Shdr) * s->shdr_num +
- s->note_size;
+
+ s->phdr_offset = sizeof(Elf32_Ehdr);
+ s->shdr_offset = s->phdr_offset + sizeof(Elf32_Phdr) * s->phdr_num;
+ s->note_offset = s->shdr_offset + sizeof(Elf32_Shdr) * s->shdr_num;
+ s->memory_offset = s->note_offset + s->note_size;
}
return;
diff --git a/include/sysemu/dump.h b/include/sysemu/dump.h
index 19458bffbd..ffc2ea1072 100644
--- a/include/sysemu/dump.h
+++ b/include/sysemu/dump.h
@@ -159,6 +159,10 @@ typedef struct DumpState {
bool resume;
bool detached;
ssize_t note_size;
+ hwaddr shdr_offset;
+ hwaddr phdr_offset;
+ hwaddr section_offset;
+ hwaddr note_offset;
hwaddr memory_offset;
int fd;
--
2.37.3

View File

@ -0,0 +1,94 @@
From cbb653d73e32513ccd46b293a52384eed6a5f84f Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Wed, 30 Mar 2022 12:36:02 +0000
Subject: [PATCH 20/42] dump: Cleanup dump_begin write functions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [20/41] 18ea1457a3e54fd368e556d96c3be50c6ad0a6bd
There's no need to have a gigantic if in there let's move the elf
32/64 bit logic into the section, segment or note code.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220330123603.107120-9-frankja@linux.ibm.com>
(cherry picked from commit 5ff2e5a3e1e67930e523486e39549a33fcf97227)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 42 +++++++++++-------------------------------
1 file changed, 11 insertions(+), 31 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 823ca32883..88abde355a 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -565,46 +565,26 @@ static void dump_begin(DumpState *s, Error **errp)
return;
}
- if (dump_is_64bit(s)) {
- /* write all PT_LOAD to vmcore */
- write_elf_loads(s, errp);
+ /* write all PT_LOAD to vmcore */
+ write_elf_loads(s, errp);
+ if (*errp) {
+ return;
+ }
+
+ /* write section to vmcore */
+ if (s->shdr_num) {
+ write_elf_section(s, 1, errp);
if (*errp) {
return;
}
+ }
- /* write section to vmcore */
- if (s->shdr_num) {
- write_elf_section(s, 1, errp);
- if (*errp) {
- return;
- }
- }
-
+ if (dump_is_64bit(s)) {
/* write notes to vmcore */
write_elf64_notes(fd_write_vmcore, s, errp);
- if (*errp) {
- return;
- }
} else {
- /* write all PT_LOAD to vmcore */
- write_elf_loads(s, errp);
- if (*errp) {
- return;
- }
-
- /* write section to vmcore */
- if (s->shdr_num) {
- write_elf_section(s, 0, errp);
- if (*errp) {
- return;
- }
- }
-
/* write notes to vmcore */
write_elf32_notes(fd_write_vmcore, s, errp);
- if (*errp) {
- return;
- }
}
}
--
2.37.3

View File

@ -0,0 +1,67 @@
From 0547599cf507930f91943f22d5f917ebacf69484 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Wed, 30 Mar 2022 12:36:03 +0000
Subject: [PATCH 21/42] dump: Consolidate elf note function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [21/41] 52298c098c116aea75ad15894731ff412c2c4e73
Just like with the other write functions let's move the 32/64 bit elf
handling to a function to improve readability.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220330123603.107120-10-frankja@linux.ibm.com>
(cherry picked from commit c68124738bc29017e4254c898bc40be7be477af7)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 18 +++++++++++-------
1 file changed, 11 insertions(+), 7 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 88abde355a..a451abc590 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -520,6 +520,15 @@ static void write_elf_loads(DumpState *s, Error **errp)
}
}
+static void write_elf_notes(DumpState *s, Error **errp)
+{
+ if (dump_is_64bit(s)) {
+ write_elf64_notes(fd_write_vmcore, s, errp);
+ } else {
+ write_elf32_notes(fd_write_vmcore, s, errp);
+ }
+}
+
/* write elf header, PT_NOTE and elf note to vmcore. */
static void dump_begin(DumpState *s, Error **errp)
{
@@ -579,13 +588,8 @@ static void dump_begin(DumpState *s, Error **errp)
}
}
- if (dump_is_64bit(s)) {
- /* write notes to vmcore */
- write_elf64_notes(fd_write_vmcore, s, errp);
- } else {
- /* write notes to vmcore */
- write_elf32_notes(fd_write_vmcore, s, errp);
- }
+ /* write notes to vmcore */
+ write_elf_notes(s, errp);
}
static int get_next_block(DumpState *s, GuestPhysBlock *block)
--
2.37.3

View File

@ -0,0 +1,169 @@
From f87abe1ef14e80731249ebe9fe1bea569a68e9b4 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Wed, 30 Mar 2022 12:36:01 +0000
Subject: [PATCH 19/42] dump: Consolidate phdr note writes
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [19/41] 180c4c0ab4941a0bf366dc7f32ee035e03daa6c0
There's no need to have two write functions. Let's rather have two
functions that set the data for elf 32/64 and then write it in a
common function.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220330123603.107120-8-frankja@linux.ibm.com>
(cherry picked from commit bc7d558017e6700f9a05c61b0b638a8994945f0d)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 94 +++++++++++++++++++++++++++--------------------------
1 file changed, 48 insertions(+), 46 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 6394e94023..823ca32883 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -246,24 +246,15 @@ static void write_elf32_load(DumpState *s, MemoryMapping *memory_mapping,
}
}
-static void write_elf64_note(DumpState *s, Error **errp)
+static void write_elf64_phdr_note(DumpState *s, Elf64_Phdr *phdr)
{
- Elf64_Phdr phdr;
- int ret;
-
- memset(&phdr, 0, sizeof(Elf64_Phdr));
- phdr.p_type = cpu_to_dump32(s, PT_NOTE);
- phdr.p_offset = cpu_to_dump64(s, s->note_offset);
- phdr.p_paddr = 0;
- phdr.p_filesz = cpu_to_dump64(s, s->note_size);
- phdr.p_memsz = cpu_to_dump64(s, s->note_size);
- phdr.p_vaddr = 0;
-
- ret = fd_write_vmcore(&phdr, sizeof(Elf64_Phdr), s);
- if (ret < 0) {
- error_setg_errno(errp, -ret,
- "dump: failed to write program header table");
- }
+ memset(phdr, 0, sizeof(*phdr));
+ phdr->p_type = cpu_to_dump32(s, PT_NOTE);
+ phdr->p_offset = cpu_to_dump64(s, s->note_offset);
+ phdr->p_paddr = 0;
+ phdr->p_filesz = cpu_to_dump64(s, s->note_size);
+ phdr->p_memsz = cpu_to_dump64(s, s->note_size);
+ phdr->p_vaddr = 0;
}
static inline int cpu_index(CPUState *cpu)
@@ -311,24 +302,15 @@ static void write_elf64_notes(WriteCoreDumpFunction f, DumpState *s,
write_guest_note(f, s, errp);
}
-static void write_elf32_note(DumpState *s, Error **errp)
+static void write_elf32_phdr_note(DumpState *s, Elf32_Phdr *phdr)
{
- Elf32_Phdr phdr;
- int ret;
-
- memset(&phdr, 0, sizeof(Elf32_Phdr));
- phdr.p_type = cpu_to_dump32(s, PT_NOTE);
- phdr.p_offset = cpu_to_dump32(s, s->note_offset);
- phdr.p_paddr = 0;
- phdr.p_filesz = cpu_to_dump32(s, s->note_size);
- phdr.p_memsz = cpu_to_dump32(s, s->note_size);
- phdr.p_vaddr = 0;
-
- ret = fd_write_vmcore(&phdr, sizeof(Elf32_Phdr), s);
- if (ret < 0) {
- error_setg_errno(errp, -ret,
- "dump: failed to write program header table");
- }
+ memset(phdr, 0, sizeof(*phdr));
+ phdr->p_type = cpu_to_dump32(s, PT_NOTE);
+ phdr->p_offset = cpu_to_dump32(s, s->note_offset);
+ phdr->p_paddr = 0;
+ phdr->p_filesz = cpu_to_dump32(s, s->note_size);
+ phdr->p_memsz = cpu_to_dump32(s, s->note_size);
+ phdr->p_vaddr = 0;
}
static void write_elf32_notes(WriteCoreDumpFunction f, DumpState *s,
@@ -358,6 +340,32 @@ static void write_elf32_notes(WriteCoreDumpFunction f, DumpState *s,
write_guest_note(f, s, errp);
}
+static void write_elf_phdr_note(DumpState *s, Error **errp)
+{
+ ERRP_GUARD();
+ Elf32_Phdr phdr32;
+ Elf64_Phdr phdr64;
+ void *phdr;
+ size_t size;
+ int ret;
+
+ if (dump_is_64bit(s)) {
+ write_elf64_phdr_note(s, &phdr64);
+ size = sizeof(phdr64);
+ phdr = &phdr64;
+ } else {
+ write_elf32_phdr_note(s, &phdr32);
+ size = sizeof(phdr32);
+ phdr = &phdr32;
+ }
+
+ ret = fd_write_vmcore(phdr, size, s);
+ if (ret < 0) {
+ error_setg_errno(errp, -ret,
+ "dump: failed to write program header table");
+ }
+}
+
static void write_elf_section(DumpState *s, int type, Error **errp)
{
Elf32_Shdr shdr32;
@@ -551,13 +559,13 @@ static void dump_begin(DumpState *s, Error **errp)
return;
}
- if (dump_is_64bit(s)) {
- /* write PT_NOTE to vmcore */
- write_elf64_note(s, errp);
- if (*errp) {
- return;
- }
+ /* write PT_NOTE to vmcore */
+ write_elf_phdr_note(s, errp);
+ if (*errp) {
+ return;
+ }
+ if (dump_is_64bit(s)) {
/* write all PT_LOAD to vmcore */
write_elf_loads(s, errp);
if (*errp) {
@@ -578,12 +586,6 @@ static void dump_begin(DumpState *s, Error **errp)
return;
}
} else {
- /* write PT_NOTE to vmcore */
- write_elf32_note(s, errp);
- if (*errp) {
- return;
- }
-
/* write all PT_LOAD to vmcore */
write_elf_loads(s, errp);
if (*errp) {
--
2.37.3

View File

@ -0,0 +1,118 @@
From c851676d202b5b76962529f3b6d433936becbd8a Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Wed, 30 Mar 2022 12:36:00 +0000
Subject: [PATCH 18/42] dump: Introduce dump_is_64bit() helper function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [18/41] a0fd2d1985c61b8e50d4a7ca26bc0ee6fcaa6196
Checking d_class in dump_info leads to lengthy conditionals so let's
shorten things a bit by introducing a helper function.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220330123603.107120-7-frankja@linux.ibm.com>
(cherry picked from commit 05bbaa5040ccb3419e8b93af8040485430e2db42)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 25 +++++++++++++++----------
1 file changed, 15 insertions(+), 10 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 85a402b38c..6394e94023 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -55,6 +55,11 @@ static Error *dump_migration_blocker;
DIV_ROUND_UP((name_size), 4) + \
DIV_ROUND_UP((desc_size), 4)) * 4)
+static inline bool dump_is_64bit(DumpState *s)
+{
+ return s->dump_info.d_class == ELFCLASS64;
+}
+
uint16_t cpu_to_dump16(DumpState *s, uint16_t val)
{
if (s->dump_info.d_endian == ELFDATA2LSB) {
@@ -489,7 +494,7 @@ static void write_elf_loads(DumpState *s, Error **errp)
get_offset_range(memory_mapping->phys_addr,
memory_mapping->length,
s, &offset, &filesz);
- if (s->dump_info.d_class == ELFCLASS64) {
+ if (dump_is_64bit(s)) {
write_elf64_load(s, memory_mapping, phdr_index++, offset,
filesz, errp);
} else {
@@ -537,7 +542,7 @@ static void dump_begin(DumpState *s, Error **errp)
*/
/* write elf header to vmcore */
- if (s->dump_info.d_class == ELFCLASS64) {
+ if (dump_is_64bit(s)) {
write_elf64_header(s, errp);
} else {
write_elf32_header(s, errp);
@@ -546,7 +551,7 @@ static void dump_begin(DumpState *s, Error **errp)
return;
}
- if (s->dump_info.d_class == ELFCLASS64) {
+ if (dump_is_64bit(s)) {
/* write PT_NOTE to vmcore */
write_elf64_note(s, errp);
if (*errp) {
@@ -757,7 +762,7 @@ static void get_note_sizes(DumpState *s, const void *note,
uint64_t name_sz;
uint64_t desc_sz;
- if (s->dump_info.d_class == ELFCLASS64) {
+ if (dump_is_64bit(s)) {
const Elf64_Nhdr *hdr = note;
note_head_sz = sizeof(Elf64_Nhdr);
name_sz = tswap64(hdr->n_namesz);
@@ -1017,10 +1022,10 @@ out:
static void write_dump_header(DumpState *s, Error **errp)
{
- if (s->dump_info.d_class == ELFCLASS32) {
- create_header32(s, errp);
- } else {
+ if (dump_is_64bit(s)) {
create_header64(s, errp);
+ } else {
+ create_header32(s, errp);
}
}
@@ -1715,8 +1720,8 @@ static void dump_init(DumpState *s, int fd, bool has_format,
uint32_t size;
uint16_t format;
- note_head_size = s->dump_info.d_class == ELFCLASS32 ?
- sizeof(Elf32_Nhdr) : sizeof(Elf64_Nhdr);
+ note_head_size = dump_is_64bit(s) ?
+ sizeof(Elf64_Nhdr) : sizeof(Elf32_Nhdr);
format = le16_to_cpu(vmci->vmcoreinfo.guest_format);
size = le32_to_cpu(vmci->vmcoreinfo.size);
@@ -1819,7 +1824,7 @@ static void dump_init(DumpState *s, int fd, bool has_format,
}
}
- if (s->dump_info.d_class == ELFCLASS64) {
+ if (dump_is_64bit(s)) {
s->phdr_offset = sizeof(Elf64_Ehdr);
s->shdr_offset = s->phdr_offset + sizeof(Elf64_Phdr) * s->phdr_num;
s->note_offset = s->shdr_offset + sizeof(Elf64_Shdr) * s->shdr_num;
--
2.37.3

View File

@ -0,0 +1,136 @@
From 255722667a4fa4d522bb0b7e0825cbbe635abb8d Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Wed, 30 Mar 2022 12:35:57 +0000
Subject: [PATCH 15/42] dump: Introduce shdr_num to decrease complexity
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [15/41] b0215ea5d381ef7f6abfe3f3bafea51ce933da56
Let's move from a boolean to a int variable which will later enable us
to store the number of sections that are in the dump file.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220330123603.107120-4-frankja@linux.ibm.com>
(cherry picked from commit 862a395858e5a302ed5921487777acdc95a3a31b)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 24 ++++++++++++------------
include/sysemu/dump.h | 2 +-
2 files changed, 13 insertions(+), 13 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 7236b167cc..972e28b089 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -145,12 +145,12 @@ static void write_elf64_header(DumpState *s, Error **errp)
elf_header.e_phoff = cpu_to_dump64(s, sizeof(Elf64_Ehdr));
elf_header.e_phentsize = cpu_to_dump16(s, sizeof(Elf64_Phdr));
elf_header.e_phnum = cpu_to_dump16(s, phnum);
- if (s->have_section) {
+ if (s->shdr_num) {
uint64_t shoff = sizeof(Elf64_Ehdr) + sizeof(Elf64_Phdr) * s->phdr_num;
elf_header.e_shoff = cpu_to_dump64(s, shoff);
elf_header.e_shentsize = cpu_to_dump16(s, sizeof(Elf64_Shdr));
- elf_header.e_shnum = cpu_to_dump16(s, 1);
+ elf_header.e_shnum = cpu_to_dump16(s, s->shdr_num);
}
ret = fd_write_vmcore(&elf_header, sizeof(elf_header), s);
@@ -182,12 +182,12 @@ static void write_elf32_header(DumpState *s, Error **errp)
elf_header.e_phoff = cpu_to_dump32(s, sizeof(Elf32_Ehdr));
elf_header.e_phentsize = cpu_to_dump16(s, sizeof(Elf32_Phdr));
elf_header.e_phnum = cpu_to_dump16(s, phnum);
- if (s->have_section) {
+ if (s->shdr_num) {
uint32_t shoff = sizeof(Elf32_Ehdr) + sizeof(Elf32_Phdr) * s->phdr_num;
elf_header.e_shoff = cpu_to_dump32(s, shoff);
elf_header.e_shentsize = cpu_to_dump16(s, sizeof(Elf32_Shdr));
- elf_header.e_shnum = cpu_to_dump16(s, 1);
+ elf_header.e_shnum = cpu_to_dump16(s, s->shdr_num);
}
ret = fd_write_vmcore(&elf_header, sizeof(elf_header), s);
@@ -566,7 +566,7 @@ static void dump_begin(DumpState *s, Error **errp)
}
/* write section to vmcore */
- if (s->have_section) {
+ if (s->shdr_num) {
write_elf_section(s, 1, errp);
if (*errp) {
return;
@@ -592,7 +592,7 @@ static void dump_begin(DumpState *s, Error **errp)
}
/* write section to vmcore */
- if (s->have_section) {
+ if (s->shdr_num) {
write_elf_section(s, 0, errp);
if (*errp) {
return;
@@ -1811,11 +1811,11 @@ static void dump_init(DumpState *s, int fd, bool has_format,
*/
s->phdr_num = 1; /* PT_NOTE */
if (s->list.num < UINT16_MAX - 2) {
+ s->shdr_num = 0;
s->phdr_num += s->list.num;
- s->have_section = false;
} else {
/* sh_info of section 0 holds the real number of phdrs */
- s->have_section = true;
+ s->shdr_num = 1;
/* the type of shdr->sh_info is uint32_t, so we should avoid overflow */
if (s->list.num <= UINT32_MAX - 1) {
@@ -1826,19 +1826,19 @@ static void dump_init(DumpState *s, int fd, bool has_format,
}
if (s->dump_info.d_class == ELFCLASS64) {
- if (s->have_section) {
+ if (s->shdr_num) {
s->memory_offset = sizeof(Elf64_Ehdr) +
sizeof(Elf64_Phdr) * s->phdr_num +
- sizeof(Elf64_Shdr) + s->note_size;
+ sizeof(Elf64_Shdr) * s->shdr_num + s->note_size;
} else {
s->memory_offset = sizeof(Elf64_Ehdr) +
sizeof(Elf64_Phdr) * s->phdr_num + s->note_size;
}
} else {
- if (s->have_section) {
+ if (s->shdr_num) {
s->memory_offset = sizeof(Elf32_Ehdr) +
sizeof(Elf32_Phdr) * s->phdr_num +
- sizeof(Elf32_Shdr) + s->note_size;
+ sizeof(Elf32_Shdr) * s->shdr_num + s->note_size;
} else {
s->memory_offset = sizeof(Elf32_Ehdr) +
sizeof(Elf32_Phdr) * s->phdr_num + s->note_size;
diff --git a/include/sysemu/dump.h b/include/sysemu/dump.h
index b463fc9c02..19458bffbd 100644
--- a/include/sysemu/dump.h
+++ b/include/sysemu/dump.h
@@ -155,7 +155,7 @@ typedef struct DumpState {
ArchDumpInfo dump_info;
MemoryMappingList list;
uint32_t phdr_num;
- bool have_section;
+ uint32_t shdr_num;
bool resume;
bool detached;
ssize_t note_size;
--
2.37.3

View File

@ -0,0 +1,142 @@
From a18ba2fbaf132724e81be92da42b36d8f365e66c Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 11 Aug 2022 12:10:56 +0000
Subject: [PATCH 24/42] dump: Refactor dump_iterate and introduce
dump_filter_memblock_*()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [24/41] 74ef470f24d9d98093c4d63730a99474587033fd
The iteration over the memblocks in dump_iterate() is hard to
understand so it's about time to clean it up. Instead of manually
grabbing the next memblock we can use QTAILQ_FOREACH to iterate over
all memblocks.
Additionally we move the calculation of the offset and length out by
introducing and using the dump_filter_memblock_*() functions. These
functions will later be used to cleanup other parts of dump.c.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220811121111.9878-4-frankja@linux.ibm.com>
(cherry picked from commit 1e8113032f5b1efc5da66382470ce4809c76f8f2)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 74 ++++++++++++++++++++++++++++++-----------------------
1 file changed, 42 insertions(+), 32 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index fa787f379f..d981e843dd 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -592,31 +592,43 @@ static void dump_begin(DumpState *s, Error **errp)
write_elf_notes(s, errp);
}
-static int get_next_block(DumpState *s, GuestPhysBlock *block)
+static int64_t dump_filtered_memblock_size(GuestPhysBlock *block,
+ int64_t filter_area_start,
+ int64_t filter_area_length)
{
- while (1) {
- block = QTAILQ_NEXT(block, next);
- if (!block) {
- /* no more block */
- return 1;
- }
+ int64_t size, left, right;
- s->start = 0;
- s->next_block = block;
- if (s->has_filter) {
- if (block->target_start >= s->begin + s->length ||
- block->target_end <= s->begin) {
- /* This block is out of the range */
- continue;
- }
+ /* No filter, return full size */
+ if (!filter_area_length) {
+ return block->target_end - block->target_start;
+ }
- if (s->begin > block->target_start) {
- s->start = s->begin - block->target_start;
- }
+ /* calculate the overlapped region. */
+ left = MAX(filter_area_start, block->target_start);
+ right = MIN(filter_area_start + filter_area_length, block->target_end);
+ size = right - left;
+ size = size > 0 ? size : 0;
+
+ return size;
+}
+
+static int64_t dump_filtered_memblock_start(GuestPhysBlock *block,
+ int64_t filter_area_start,
+ int64_t filter_area_length)
+{
+ if (filter_area_length) {
+ /* return -1 if the block is not within filter area */
+ if (block->target_start >= filter_area_start + filter_area_length ||
+ block->target_end <= filter_area_start) {
+ return -1;
}
- return 0;
+ if (filter_area_start > block->target_start) {
+ return filter_area_start - block->target_start;
+ }
}
+
+ return 0;
}
/* write all memory to vmcore */
@@ -624,24 +636,22 @@ static void dump_iterate(DumpState *s, Error **errp)
{
ERRP_GUARD();
GuestPhysBlock *block;
- int64_t size;
-
- do {
- block = s->next_block;
+ int64_t memblock_size, memblock_start;
- size = block->target_end - block->target_start;
- if (s->has_filter) {
- size -= s->start;
- if (s->begin + s->length < block->target_end) {
- size -= block->target_end - (s->begin + s->length);
- }
+ QTAILQ_FOREACH(block, &s->guest_phys_blocks.head, next) {
+ memblock_start = dump_filtered_memblock_start(block, s->begin, s->length);
+ if (memblock_start == -1) {
+ continue;
}
- write_memory(s, block, s->start, size, errp);
+
+ memblock_size = dump_filtered_memblock_size(block, s->begin, s->length);
+
+ /* Write the memory to file */
+ write_memory(s, block, memblock_start, memblock_size, errp);
if (*errp) {
return;
}
-
- } while (!get_next_block(s, block));
+ }
}
static void create_vmcore(DumpState *s, Error **errp)
--
2.37.3

View File

@ -0,0 +1,45 @@
From 6932fe3afbec443bbf6acff5b707536254e1bc37 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Mon, 17 Oct 2022 08:38:16 +0000
Subject: [PATCH 35/42] dump: Reintroduce memory_offset and section_offset
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [35/41] e60c0d066aeeedb42e724712bc3aa7b7591c6c79
section_offset will later be used to store the offset to the section
data which will be stored last. For now memory_offset is only needed
to make section_offset look nicer.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20221017083822.43118-5-frankja@linux.ibm.com>
(cherry picked from commit 13fd417ddc81a1685c6a8f4e1c80bbfe7150f164)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/dump/dump.c b/dump/dump.c
index d17537d4e9..7a42401790 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -1885,6 +1885,8 @@ static void dump_init(DumpState *s, int fd, bool has_format,
s->phdr_offset = s->shdr_offset + sizeof(Elf32_Shdr) * s->shdr_num;
s->note_offset = s->phdr_offset + sizeof(Elf32_Phdr) * s->phdr_num;
}
+ s->memory_offset = s->note_offset + s->note_size;
+ s->section_offset = s->memory_offset + s->total_size;
return;
--
2.37.3

View File

@ -0,0 +1,70 @@
From a8eeab6936a2bd27b33b63aed7e2ef96034f7772 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Wed, 30 Mar 2022 12:35:58 +0000
Subject: [PATCH 16/42] dump: Remove the section if when calculating the memory
offset
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [16/41] ff214d2c23b9cb16fd49d22d976829267df43133
When s->shdr_num is 0 we'll add 0 bytes of section headers which is
equivalent to not adding section headers but with the multiplication
we can remove a if/else.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220330123603.107120-5-frankja@linux.ibm.com>
(cherry picked from commit 344107e07bd81546474a54ab83800158ca953059)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 24 ++++++++----------------
1 file changed, 8 insertions(+), 16 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 972e28b089..5cc2322325 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -1826,23 +1826,15 @@ static void dump_init(DumpState *s, int fd, bool has_format,
}
if (s->dump_info.d_class == ELFCLASS64) {
- if (s->shdr_num) {
- s->memory_offset = sizeof(Elf64_Ehdr) +
- sizeof(Elf64_Phdr) * s->phdr_num +
- sizeof(Elf64_Shdr) * s->shdr_num + s->note_size;
- } else {
- s->memory_offset = sizeof(Elf64_Ehdr) +
- sizeof(Elf64_Phdr) * s->phdr_num + s->note_size;
- }
+ s->memory_offset = sizeof(Elf64_Ehdr) +
+ sizeof(Elf64_Phdr) * s->phdr_num +
+ sizeof(Elf64_Shdr) * s->shdr_num +
+ s->note_size;
} else {
- if (s->shdr_num) {
- s->memory_offset = sizeof(Elf32_Ehdr) +
- sizeof(Elf32_Phdr) * s->phdr_num +
- sizeof(Elf32_Shdr) * s->shdr_num + s->note_size;
- } else {
- s->memory_offset = sizeof(Elf32_Ehdr) +
- sizeof(Elf32_Phdr) * s->phdr_num + s->note_size;
- }
+ s->memory_offset = sizeof(Elf32_Ehdr) +
+ sizeof(Elf32_Phdr) * s->phdr_num +
+ sizeof(Elf32_Shdr) * s->shdr_num +
+ s->note_size;
}
return;
--
2.37.3

View File

@ -0,0 +1,176 @@
From eb763bec53d6b9aea7a6b60b0cf8c5d8b5f1b35c Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 7 Apr 2022 09:48:24 +0000
Subject: [PATCH 14/42] dump: Remove the sh_info variable
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [14/41] 24af12b78c8f5a02cf85df2f6b1d64249f9499c9
There's no need to have phdr_num and sh_info at the same time. We can
make phdr_num 32 bit and set PN_XNUM when we write the header if
phdr_num >= PN_XNUM.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220407094824.5074-1-frankja@linux.ibm.com>
(cherry picked from commit 046bc4160bc780eaacc2d702a2589f1a7a01188d)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 44 +++++++++++++++++++++++--------------------
include/sysemu/dump.h | 3 +--
2 files changed, 25 insertions(+), 22 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 9876123f2e..7236b167cc 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -124,6 +124,12 @@ static int fd_write_vmcore(const void *buf, size_t size, void *opaque)
static void write_elf64_header(DumpState *s, Error **errp)
{
+ /*
+ * phnum in the elf header is 16 bit, if we have more segments we
+ * set phnum to PN_XNUM and write the real number of segments to a
+ * special section.
+ */
+ uint16_t phnum = MIN(s->phdr_num, PN_XNUM);
Elf64_Ehdr elf_header;
int ret;
@@ -138,9 +144,9 @@ static void write_elf64_header(DumpState *s, Error **errp)
elf_header.e_ehsize = cpu_to_dump16(s, sizeof(elf_header));
elf_header.e_phoff = cpu_to_dump64(s, sizeof(Elf64_Ehdr));
elf_header.e_phentsize = cpu_to_dump16(s, sizeof(Elf64_Phdr));
- elf_header.e_phnum = cpu_to_dump16(s, s->phdr_num);
+ elf_header.e_phnum = cpu_to_dump16(s, phnum);
if (s->have_section) {
- uint64_t shoff = sizeof(Elf64_Ehdr) + sizeof(Elf64_Phdr) * s->sh_info;
+ uint64_t shoff = sizeof(Elf64_Ehdr) + sizeof(Elf64_Phdr) * s->phdr_num;
elf_header.e_shoff = cpu_to_dump64(s, shoff);
elf_header.e_shentsize = cpu_to_dump16(s, sizeof(Elf64_Shdr));
@@ -155,6 +161,12 @@ static void write_elf64_header(DumpState *s, Error **errp)
static void write_elf32_header(DumpState *s, Error **errp)
{
+ /*
+ * phnum in the elf header is 16 bit, if we have more segments we
+ * set phnum to PN_XNUM and write the real number of segments to a
+ * special section.
+ */
+ uint16_t phnum = MIN(s->phdr_num, PN_XNUM);
Elf32_Ehdr elf_header;
int ret;
@@ -169,9 +181,9 @@ static void write_elf32_header(DumpState *s, Error **errp)
elf_header.e_ehsize = cpu_to_dump16(s, sizeof(elf_header));
elf_header.e_phoff = cpu_to_dump32(s, sizeof(Elf32_Ehdr));
elf_header.e_phentsize = cpu_to_dump16(s, sizeof(Elf32_Phdr));
- elf_header.e_phnum = cpu_to_dump16(s, s->phdr_num);
+ elf_header.e_phnum = cpu_to_dump16(s, phnum);
if (s->have_section) {
- uint32_t shoff = sizeof(Elf32_Ehdr) + sizeof(Elf32_Phdr) * s->sh_info;
+ uint32_t shoff = sizeof(Elf32_Ehdr) + sizeof(Elf32_Phdr) * s->phdr_num;
elf_header.e_shoff = cpu_to_dump32(s, shoff);
elf_header.e_shentsize = cpu_to_dump16(s, sizeof(Elf32_Shdr));
@@ -358,12 +370,12 @@ static void write_elf_section(DumpState *s, int type, Error **errp)
if (type == 0) {
shdr_size = sizeof(Elf32_Shdr);
memset(&shdr32, 0, shdr_size);
- shdr32.sh_info = cpu_to_dump32(s, s->sh_info);
+ shdr32.sh_info = cpu_to_dump32(s, s->phdr_num);
shdr = &shdr32;
} else {
shdr_size = sizeof(Elf64_Shdr);
memset(&shdr64, 0, shdr_size);
- shdr64.sh_info = cpu_to_dump32(s, s->sh_info);
+ shdr64.sh_info = cpu_to_dump32(s, s->phdr_num);
shdr = &shdr64;
}
@@ -478,13 +490,6 @@ static void write_elf_loads(DumpState *s, Error **errp)
hwaddr offset, filesz;
MemoryMapping *memory_mapping;
uint32_t phdr_index = 1;
- uint32_t max_index;
-
- if (s->have_section) {
- max_index = s->sh_info;
- } else {
- max_index = s->phdr_num;
- }
QTAILQ_FOREACH(memory_mapping, &s->list.head, next) {
get_offset_range(memory_mapping->phys_addr,
@@ -502,7 +507,7 @@ static void write_elf_loads(DumpState *s, Error **errp)
return;
}
- if (phdr_index >= max_index) {
+ if (phdr_index >= s->phdr_num) {
break;
}
}
@@ -1809,22 +1814,21 @@ static void dump_init(DumpState *s, int fd, bool has_format,
s->phdr_num += s->list.num;
s->have_section = false;
} else {
+ /* sh_info of section 0 holds the real number of phdrs */
s->have_section = true;
- s->phdr_num = PN_XNUM;
- s->sh_info = 1; /* PT_NOTE */
/* the type of shdr->sh_info is uint32_t, so we should avoid overflow */
if (s->list.num <= UINT32_MAX - 1) {
- s->sh_info += s->list.num;
+ s->phdr_num += s->list.num;
} else {
- s->sh_info = UINT32_MAX;
+ s->phdr_num = UINT32_MAX;
}
}
if (s->dump_info.d_class == ELFCLASS64) {
if (s->have_section) {
s->memory_offset = sizeof(Elf64_Ehdr) +
- sizeof(Elf64_Phdr) * s->sh_info +
+ sizeof(Elf64_Phdr) * s->phdr_num +
sizeof(Elf64_Shdr) + s->note_size;
} else {
s->memory_offset = sizeof(Elf64_Ehdr) +
@@ -1833,7 +1837,7 @@ static void dump_init(DumpState *s, int fd, bool has_format,
} else {
if (s->have_section) {
s->memory_offset = sizeof(Elf32_Ehdr) +
- sizeof(Elf32_Phdr) * s->sh_info +
+ sizeof(Elf32_Phdr) * s->phdr_num +
sizeof(Elf32_Shdr) + s->note_size;
} else {
s->memory_offset = sizeof(Elf32_Ehdr) +
diff --git a/include/sysemu/dump.h b/include/sysemu/dump.h
index 250143cb5a..b463fc9c02 100644
--- a/include/sysemu/dump.h
+++ b/include/sysemu/dump.h
@@ -154,8 +154,7 @@ typedef struct DumpState {
GuestPhysBlockList guest_phys_blocks;
ArchDumpInfo dump_info;
MemoryMappingList list;
- uint16_t phdr_num;
- uint32_t sh_info;
+ uint32_t phdr_num;
bool have_section;
bool resume;
bool detached;
--
2.37.3

View File

@ -0,0 +1,69 @@
From 18fef7f02801d51207d67b8f8ec5f0d828889c78 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 11 Aug 2022 12:11:01 +0000
Subject: [PATCH 29/42] dump: Rename write_elf*_phdr_note to
prepare_elf*_phdr_note
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [29/41] 876cea6f6e51be8df2763f56d0daef99d11fdd49
The functions in question do not actually write to the file descriptor
they set up a buffer which is later written to the fd.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220811121111.9878-9-frankja@linux.ibm.com>
(cherry picked from commit 2341a94d3a0a8a93a5a977e642da1807b8edaab8)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 8d5226f861..c2c1341ad7 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -261,7 +261,7 @@ static void write_elf32_load(DumpState *s, MemoryMapping *memory_mapping,
}
}
-static void write_elf64_phdr_note(DumpState *s, Elf64_Phdr *phdr)
+static void prepare_elf64_phdr_note(DumpState *s, Elf64_Phdr *phdr)
{
memset(phdr, 0, sizeof(*phdr));
phdr->p_type = cpu_to_dump32(s, PT_NOTE);
@@ -317,7 +317,7 @@ static void write_elf64_notes(WriteCoreDumpFunction f, DumpState *s,
write_guest_note(f, s, errp);
}
-static void write_elf32_phdr_note(DumpState *s, Elf32_Phdr *phdr)
+static void prepare_elf32_phdr_note(DumpState *s, Elf32_Phdr *phdr)
{
memset(phdr, 0, sizeof(*phdr));
phdr->p_type = cpu_to_dump32(s, PT_NOTE);
@@ -365,11 +365,11 @@ static void write_elf_phdr_note(DumpState *s, Error **errp)
int ret;
if (dump_is_64bit(s)) {
- write_elf64_phdr_note(s, &phdr64);
+ prepare_elf64_phdr_note(s, &phdr64);
size = sizeof(phdr64);
phdr = &phdr64;
} else {
- write_elf32_phdr_note(s, &phdr32);
+ prepare_elf32_phdr_note(s, &phdr32);
size = sizeof(phdr32);
phdr = &phdr32;
}
--
2.37.3

View File

@ -0,0 +1,57 @@
From 04d4947a22fe3192384ff486d0a979d799ded98e Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 11 Aug 2022 12:10:55 +0000
Subject: [PATCH 23/42] dump: Rename write_elf_loads to write_elf_phdr_loads
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [23/41] 18e3ef70b97c525b7c43cf12143204bdb1060e4f
Let's make it a bit clearer that we write the program headers of the
PT_LOAD type.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Reviewed-by: Steffen Eiden <seiden@ibm.linux.com>
Message-Id: <20220811121111.9878-3-frankja@linux.ibm.com>
(cherry picked from commit afae6056ea79e2d89fd90867de3a01732eae724f)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index a451abc590..fa787f379f 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -491,7 +491,7 @@ static void get_offset_range(hwaddr phys_addr,
}
}
-static void write_elf_loads(DumpState *s, Error **errp)
+static void write_elf_phdr_loads(DumpState *s, Error **errp)
{
ERRP_GUARD();
hwaddr offset, filesz;
@@ -574,8 +574,8 @@ static void dump_begin(DumpState *s, Error **errp)
return;
}
- /* write all PT_LOAD to vmcore */
- write_elf_loads(s, errp);
+ /* write all PT_LOADs to vmcore */
+ write_elf_phdr_loads(s, errp);
if (*errp) {
return;
}
--
2.37.3

View File

@ -0,0 +1,68 @@
From 7e8d6290099b33f88621b45e62652a97704c9573 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Mon, 17 Oct 2022 08:38:15 +0000
Subject: [PATCH 34/42] dump: Reorder struct DumpState
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [34/41] 8d44e5e8c86ea5b33644eba141046cd657d0071e
Let's move ELF related members into one block and guest memory related
ones into another to improve readability.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20221017083822.43118-4-frankja@linux.ibm.com>
(cherry picked from commit 8384b73c46fd474847d7e74d121318e344edc3c4)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
include/sysemu/dump.h | 16 +++++++++-------
1 file changed, 9 insertions(+), 7 deletions(-)
diff --git a/include/sysemu/dump.h b/include/sysemu/dump.h
index 9995f65dc8..9ed811b313 100644
--- a/include/sysemu/dump.h
+++ b/include/sysemu/dump.h
@@ -154,15 +154,8 @@ typedef struct DumpState {
GuestPhysBlockList guest_phys_blocks;
ArchDumpInfo dump_info;
MemoryMappingList list;
- uint32_t phdr_num;
- uint32_t shdr_num;
bool resume;
bool detached;
- ssize_t note_size;
- hwaddr shdr_offset;
- hwaddr phdr_offset;
- hwaddr section_offset;
- hwaddr note_offset;
hwaddr memory_offset;
int fd;
@@ -177,6 +170,15 @@ typedef struct DumpState {
int64_t filter_area_begin; /* Start address of partial guest memory area */
int64_t filter_area_length; /* Length of partial guest memory area */
+ /* Elf dump related data */
+ uint32_t phdr_num;
+ uint32_t shdr_num;
+ ssize_t note_size;
+ hwaddr shdr_offset;
+ hwaddr phdr_offset;
+ hwaddr section_offset;
+ hwaddr note_offset;
+
void *elf_section_hdrs; /* Pointer to section header buffer */
uint8_t *note_buf; /* buffer for notes */
--
2.37.3

View File

@ -0,0 +1,467 @@
From 8f674e0e12e4b88fc035948612a0b0949e0ad892 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 11 Aug 2022 12:10:54 +0000
Subject: [PATCH 22/42] dump: Replace opaque DumpState pointer with a typed one
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [22/41] 5f071d7ef441ae6f5da70eb56018c4657deee3d7
It's always better to convey the type of a pointer if at all
possible. So let's add the DumpState typedef to typedefs.h and move
the dump note functions from the opaque pointers to DumpState
pointers.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
CC: Peter Maydell <peter.maydell@linaro.org>
CC: Cédric Le Goater <clg@kaod.org>
CC: Daniel Henrique Barboza <danielhb413@gmail.com>
CC: David Gibson <david@gibson.dropbear.id.au>
CC: Greg Kurz <groug@kaod.org>
CC: Palmer Dabbelt <palmer@dabbelt.com>
CC: Alistair Francis <alistair.francis@wdc.com>
CC: Bin Meng <bin.meng@windriver.com>
CC: Cornelia Huck <cohuck@redhat.com>
CC: Thomas Huth <thuth@redhat.com>
CC: Richard Henderson <richard.henderson@linaro.org>
CC: David Hildenbrand <david@redhat.com>
Acked-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220811121111.9878-2-frankja@linux.ibm.com>
(cherry picked from commit 1af0006ab959864dfa2f59e9136c5fb93000b61f)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
include/hw/core/sysemu-cpu-ops.h | 8 ++++----
include/qemu/typedefs.h | 1 +
target/arm/arch_dump.c | 6 ++----
target/arm/cpu.h | 4 ++--
target/i386/arch_dump.c | 30 +++++++++++++++---------------
target/i386/cpu.h | 8 ++++----
target/ppc/arch_dump.c | 18 +++++++++---------
target/ppc/cpu.h | 4 ++--
target/riscv/arch_dump.c | 6 ++----
target/riscv/cpu.h | 4 ++--
target/s390x/arch_dump.c | 10 +++++-----
target/s390x/s390x-internal.h | 2 +-
12 files changed, 49 insertions(+), 52 deletions(-)
diff --git a/include/hw/core/sysemu-cpu-ops.h b/include/hw/core/sysemu-cpu-ops.h
index a9ba39e5f2..ee169b872c 100644
--- a/include/hw/core/sysemu-cpu-ops.h
+++ b/include/hw/core/sysemu-cpu-ops.h
@@ -53,25 +53,25 @@ typedef struct SysemuCPUOps {
* 32-bit VM coredump.
*/
int (*write_elf32_note)(WriteCoreDumpFunction f, CPUState *cpu,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
/**
* @write_elf64_note: Callback for writing a CPU-specific ELF note to a
* 64-bit VM coredump.
*/
int (*write_elf64_note)(WriteCoreDumpFunction f, CPUState *cpu,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
/**
* @write_elf32_qemunote: Callback for writing a CPU- and QEMU-specific ELF
* note to a 32-bit VM coredump.
*/
int (*write_elf32_qemunote)(WriteCoreDumpFunction f, CPUState *cpu,
- void *opaque);
+ DumpState *s);
/**
* @write_elf64_qemunote: Callback for writing a CPU- and QEMU-specific ELF
* note to a 64-bit VM coredump.
*/
int (*write_elf64_qemunote)(WriteCoreDumpFunction f, CPUState *cpu,
- void *opaque);
+ DumpState *s);
/**
* @virtio_is_big_endian: Callback to return %true if a CPU which supports
* runtime configurable endianness is currently big-endian.
diff --git a/include/qemu/typedefs.h b/include/qemu/typedefs.h
index ee60eb3de4..ac9d031be6 100644
--- a/include/qemu/typedefs.h
+++ b/include/qemu/typedefs.h
@@ -125,6 +125,7 @@ typedef struct VirtIODevice VirtIODevice;
typedef struct Visitor Visitor;
typedef struct VMChangeStateEntry VMChangeStateEntry;
typedef struct VMStateDescription VMStateDescription;
+typedef struct DumpState DumpState;
/*
* Pointer types
diff --git a/target/arm/arch_dump.c b/target/arm/arch_dump.c
index 0184845310..3a824e0aa6 100644
--- a/target/arm/arch_dump.c
+++ b/target/arm/arch_dump.c
@@ -232,12 +232,11 @@ static int aarch64_write_elf64_sve(WriteCoreDumpFunction f,
#endif
int arm_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
struct aarch64_note note;
ARMCPU *cpu = ARM_CPU(cs);
CPUARMState *env = &cpu->env;
- DumpState *s = opaque;
uint64_t pstate, sp;
int ret, i;
@@ -360,12 +359,11 @@ static int arm_write_elf32_vfp(WriteCoreDumpFunction f, CPUARMState *env,
}
int arm_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
struct arm_note note;
ARMCPU *cpu = ARM_CPU(cs);
CPUARMState *env = &cpu->env;
- DumpState *s = opaque;
int ret, i;
bool fpvalid = cpu_isar_feature(aa32_vfp_simd, cpu);
diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index e33f37b70a..8d2f496ef9 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -1065,9 +1065,9 @@ int arm_gen_dynamic_svereg_xml(CPUState *cpu, int base_reg);
const char *arm_gdb_get_dynamic_xml(CPUState *cpu, const char *xmlname);
int arm_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
int arm_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
#ifdef TARGET_AARCH64
int aarch64_cpu_gdb_read_register(CPUState *cpu, GByteArray *buf, int reg);
diff --git a/target/i386/arch_dump.c b/target/i386/arch_dump.c
index 004141fc04..c290910a04 100644
--- a/target/i386/arch_dump.c
+++ b/target/i386/arch_dump.c
@@ -42,7 +42,7 @@ typedef struct {
static int x86_64_write_elf64_note(WriteCoreDumpFunction f,
CPUX86State *env, int id,
- void *opaque)
+ DumpState *s)
{
x86_64_user_regs_struct regs;
Elf64_Nhdr *note;
@@ -94,7 +94,7 @@ static int x86_64_write_elf64_note(WriteCoreDumpFunction f,
buf += descsz - sizeof(x86_64_user_regs_struct)-sizeof(target_ulong);
memcpy(buf, &regs, sizeof(x86_64_user_regs_struct));
- ret = f(note, note_size, opaque);
+ ret = f(note, note_size, s);
g_free(note);
if (ret < 0) {
return -1;
@@ -148,7 +148,7 @@ static void x86_fill_elf_prstatus(x86_elf_prstatus *prstatus, CPUX86State *env,
}
static int x86_write_elf64_note(WriteCoreDumpFunction f, CPUX86State *env,
- int id, void *opaque)
+ int id, DumpState *s)
{
x86_elf_prstatus prstatus;
Elf64_Nhdr *note;
@@ -170,7 +170,7 @@ static int x86_write_elf64_note(WriteCoreDumpFunction f, CPUX86State *env,
buf += ROUND_UP(name_size, 4);
memcpy(buf, &prstatus, sizeof(prstatus));
- ret = f(note, note_size, opaque);
+ ret = f(note, note_size, s);
g_free(note);
if (ret < 0) {
return -1;
@@ -180,7 +180,7 @@ static int x86_write_elf64_note(WriteCoreDumpFunction f, CPUX86State *env,
}
int x86_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
X86CPU *cpu = X86_CPU(cs);
int ret;
@@ -189,10 +189,10 @@ int x86_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
bool lma = !!(first_x86_cpu->env.hflags & HF_LMA_MASK);
if (lma) {
- ret = x86_64_write_elf64_note(f, &cpu->env, cpuid, opaque);
+ ret = x86_64_write_elf64_note(f, &cpu->env, cpuid, s);
} else {
#endif
- ret = x86_write_elf64_note(f, &cpu->env, cpuid, opaque);
+ ret = x86_write_elf64_note(f, &cpu->env, cpuid, s);
#ifdef TARGET_X86_64
}
#endif
@@ -201,7 +201,7 @@ int x86_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
}
int x86_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
X86CPU *cpu = X86_CPU(cs);
x86_elf_prstatus prstatus;
@@ -224,7 +224,7 @@ int x86_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cs,
buf += ROUND_UP(name_size, 4);
memcpy(buf, &prstatus, sizeof(prstatus));
- ret = f(note, note_size, opaque);
+ ret = f(note, note_size, s);
g_free(note);
if (ret < 0) {
return -1;
@@ -329,7 +329,7 @@ static void qemu_get_cpustate(QEMUCPUState *s, CPUX86State *env)
static inline int cpu_write_qemu_note(WriteCoreDumpFunction f,
CPUX86State *env,
- void *opaque,
+ DumpState *s,
int type)
{
QEMUCPUState state;
@@ -369,7 +369,7 @@ static inline int cpu_write_qemu_note(WriteCoreDumpFunction f,
buf += ROUND_UP(name_size, 4);
memcpy(buf, &state, sizeof(state));
- ret = f(note, note_size, opaque);
+ ret = f(note, note_size, s);
g_free(note);
if (ret < 0) {
return -1;
@@ -379,19 +379,19 @@ static inline int cpu_write_qemu_note(WriteCoreDumpFunction f,
}
int x86_cpu_write_elf64_qemunote(WriteCoreDumpFunction f, CPUState *cs,
- void *opaque)
+ DumpState *s)
{
X86CPU *cpu = X86_CPU(cs);
- return cpu_write_qemu_note(f, &cpu->env, opaque, 1);
+ return cpu_write_qemu_note(f, &cpu->env, s, 1);
}
int x86_cpu_write_elf32_qemunote(WriteCoreDumpFunction f, CPUState *cs,
- void *opaque)
+ DumpState *s)
{
X86CPU *cpu = X86_CPU(cs);
- return cpu_write_qemu_note(f, &cpu->env, opaque, 0);
+ return cpu_write_qemu_note(f, &cpu->env, s, 0);
}
int cpu_get_dump_info(ArchDumpInfo *info,
diff --git a/target/i386/cpu.h b/target/i386/cpu.h
index 006b735fe4..5d2ddd81b9 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
@@ -1887,13 +1887,13 @@ extern const VMStateDescription vmstate_x86_cpu;
int x86_cpu_pending_interrupt(CPUState *cs, int interrupt_request);
int x86_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cpu,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
int x86_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cpu,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
int x86_cpu_write_elf64_qemunote(WriteCoreDumpFunction f, CPUState *cpu,
- void *opaque);
+ DumpState *s);
int x86_cpu_write_elf32_qemunote(WriteCoreDumpFunction f, CPUState *cpu,
- void *opaque);
+ DumpState *s);
void x86_cpu_get_memory_mapping(CPUState *cpu, MemoryMappingList *list,
Error **errp);
diff --git a/target/ppc/arch_dump.c b/target/ppc/arch_dump.c
index bb392f6d88..e9f512bcd4 100644
--- a/target/ppc/arch_dump.c
+++ b/target/ppc/arch_dump.c
@@ -270,23 +270,23 @@ ssize_t cpu_get_note_size(int class, int machine, int nr_cpus)
static int ppc_write_all_elf_notes(const char *note_name,
WriteCoreDumpFunction f,
PowerPCCPU *cpu, int id,
- void *opaque)
+ DumpState *s)
{
- NoteFuncArg arg = { .state = opaque };
+ NoteFuncArg arg = { .state = s };
int ret = -1;
int note_size;
const NoteFuncDesc *nf;
for (nf = note_func; nf->note_contents_func; nf++) {
- arg.note.hdr.n_namesz = cpu_to_dump32(opaque, sizeof(arg.note.name));
- arg.note.hdr.n_descsz = cpu_to_dump32(opaque, nf->contents_size);
+ arg.note.hdr.n_namesz = cpu_to_dump32(s, sizeof(arg.note.name));
+ arg.note.hdr.n_descsz = cpu_to_dump32(s, nf->contents_size);
strncpy(arg.note.name, note_name, sizeof(arg.note.name));
(*nf->note_contents_func)(&arg, cpu);
note_size =
sizeof(arg.note) - sizeof(arg.note.contents) + nf->contents_size;
- ret = f(&arg.note, note_size, opaque);
+ ret = f(&arg.note, note_size, s);
if (ret < 0) {
return -1;
}
@@ -295,15 +295,15 @@ static int ppc_write_all_elf_notes(const char *note_name,
}
int ppc64_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
PowerPCCPU *cpu = POWERPC_CPU(cs);
- return ppc_write_all_elf_notes("CORE", f, cpu, cpuid, opaque);
+ return ppc_write_all_elf_notes("CORE", f, cpu, cpuid, s);
}
int ppc32_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
PowerPCCPU *cpu = POWERPC_CPU(cs);
- return ppc_write_all_elf_notes("CORE", f, cpu, cpuid, opaque);
+ return ppc_write_all_elf_notes("CORE", f, cpu, cpuid, s);
}
diff --git a/target/ppc/cpu.h b/target/ppc/cpu.h
index 23e8b76c85..f5fb284706 100644
--- a/target/ppc/cpu.h
+++ b/target/ppc/cpu.h
@@ -1289,9 +1289,9 @@ void ppc_gdb_gen_spr_xml(PowerPCCPU *cpu);
const char *ppc_gdb_get_dynamic_xml(CPUState *cs, const char *xml_name);
#endif
int ppc64_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
int ppc32_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
#ifndef CONFIG_USER_ONLY
void ppc_cpu_do_interrupt(CPUState *cpu);
bool ppc_cpu_exec_interrupt(CPUState *cpu, int int_req);
diff --git a/target/riscv/arch_dump.c b/target/riscv/arch_dump.c
index 709f621d82..736a232956 100644
--- a/target/riscv/arch_dump.c
+++ b/target/riscv/arch_dump.c
@@ -64,12 +64,11 @@ static void riscv64_note_init(struct riscv64_note *note, DumpState *s,
}
int riscv_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
struct riscv64_note note;
RISCVCPU *cpu = RISCV_CPU(cs);
CPURISCVState *env = &cpu->env;
- DumpState *s = opaque;
int ret, i = 0;
const char name[] = "CORE";
@@ -134,12 +133,11 @@ static void riscv32_note_init(struct riscv32_note *note, DumpState *s,
}
int riscv_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
struct riscv32_note note;
RISCVCPU *cpu = RISCV_CPU(cs);
CPURISCVState *env = &cpu->env;
- DumpState *s = opaque;
int ret, i;
const char name[] = "CORE";
diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h
index 0760c0af93..4cce524b2c 100644
--- a/target/riscv/cpu.h
+++ b/target/riscv/cpu.h
@@ -344,9 +344,9 @@ extern const char * const riscv_fpr_regnames[];
const char *riscv_cpu_get_trap_name(target_ulong cause, bool async);
void riscv_cpu_do_interrupt(CPUState *cpu);
int riscv_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
int riscv_cpu_write_elf32_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
int riscv_cpu_gdb_read_register(CPUState *cpu, GByteArray *buf, int reg);
int riscv_cpu_gdb_write_register(CPUState *cpu, uint8_t *buf, int reg);
bool riscv_cpu_fp_enabled(CPURISCVState *env);
diff --git a/target/s390x/arch_dump.c b/target/s390x/arch_dump.c
index 08daf93ae1..f60a14920d 100644
--- a/target/s390x/arch_dump.c
+++ b/target/s390x/arch_dump.c
@@ -204,7 +204,7 @@ static const NoteFuncDesc note_linux[] = {
static int s390x_write_elf64_notes(const char *note_name,
WriteCoreDumpFunction f,
S390CPU *cpu, int id,
- void *opaque,
+ DumpState *s,
const NoteFuncDesc *funcs)
{
Note note;
@@ -222,7 +222,7 @@ static int s390x_write_elf64_notes(const char *note_name,
(*nf->note_contents_func)(&note, cpu, id);
note_size = sizeof(note) - sizeof(note.contents) + nf->contents_size;
- ret = f(&note, note_size, opaque);
+ ret = f(&note, note_size, s);
if (ret < 0) {
return -1;
@@ -235,16 +235,16 @@ static int s390x_write_elf64_notes(const char *note_name,
int s390_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque)
+ int cpuid, DumpState *s)
{
S390CPU *cpu = S390_CPU(cs);
int r;
- r = s390x_write_elf64_notes("CORE", f, cpu, cpuid, opaque, note_core);
+ r = s390x_write_elf64_notes("CORE", f, cpu, cpuid, s, note_core);
if (r) {
return r;
}
- return s390x_write_elf64_notes("LINUX", f, cpu, cpuid, opaque, note_linux);
+ return s390x_write_elf64_notes("LINUX", f, cpu, cpuid, s, note_linux);
}
int cpu_get_dump_info(ArchDumpInfo *info,
diff --git a/target/s390x/s390x-internal.h b/target/s390x/s390x-internal.h
index 1a178aed41..02cf6c3f43 100644
--- a/target/s390x/s390x-internal.h
+++ b/target/s390x/s390x-internal.h
@@ -228,7 +228,7 @@ static inline hwaddr decode_basedisp_s(CPUS390XState *env, uint32_t ipb,
/* arch_dump.c */
int s390_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
- int cpuid, void *opaque);
+ int cpuid, DumpState *s);
/* cc_helper.c */
--
2.37.3

View File

@ -0,0 +1,73 @@
From 1f7cb73592a1922b3a981eb3232098281e07679f Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 11 Aug 2022 12:10:59 +0000
Subject: [PATCH 27/42] dump: Rework dump_calculate_size function
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [27/41] eaa05c39109b57a119752ad3df66f4c2ace2cbe4
dump_calculate_size() sums up all the sizes of the guest memory
blocks. Since we already have a function that calculates the size of a
single memory block (dump_get_memblock_size()) we can simply iterate
over the blocks and use the function instead of calculating the size
ourselves.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Message-Id: <20220811121111.9878-7-frankja@linux.ibm.com>
(cherry picked from commit c370d5300f9ac1f90f8158082d22262b904fe30e)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 22 ++++++++--------------
1 file changed, 8 insertions(+), 14 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index f6fe13e258..902a85ef8e 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -1557,25 +1557,19 @@ bool dump_in_progress(void)
return (qatomic_read(&state->status) == DUMP_STATUS_ACTIVE);
}
-/* calculate total size of memory to be dumped (taking filter into
- * acoount.) */
+/*
+ * calculate total size of memory to be dumped (taking filter into
+ * account.)
+ */
static int64_t dump_calculate_size(DumpState *s)
{
GuestPhysBlock *block;
- int64_t size = 0, total = 0, left = 0, right = 0;
+ int64_t total = 0;
QTAILQ_FOREACH(block, &s->guest_phys_blocks.head, next) {
- if (dump_has_filter(s)) {
- /* calculate the overlapped region. */
- left = MAX(s->filter_area_begin, block->target_start);
- right = MIN(s->filter_area_begin + s->filter_area_length, block->target_end);
- size = right - left;
- size = size > 0 ? size : 0;
- } else {
- /* count the whole region in */
- size = (block->target_end - block->target_start);
- }
- total += size;
+ total += dump_filtered_memblock_size(block,
+ s->filter_area_begin,
+ s->filter_area_length);
}
return total;
--
2.37.3

View File

@ -0,0 +1,187 @@
From 411f5354b809f6b783946e58d7655135814fb809 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 11 Aug 2022 12:10:58 +0000
Subject: [PATCH 26/42] dump: Rework filter area variables
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [26/41] f10a5523dfd2724f7a8637fca3ed68ba6df659a5
While the DumpState begin and length variables directly mirror the API
variable names they are not very descriptive. So let's add a
"filter_area_" prefix and make has_filter a function checking length > 0.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220811121111.9878-6-frankja@linux.ibm.com>
(cherry picked from commit dddf725f70bfe7f5adb41fa31dbd06e767271bda)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 53 +++++++++++++++++++++++++------------------
include/sysemu/dump.h | 13 ++++++++---
2 files changed, 41 insertions(+), 25 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index e6aa037f59..f6fe13e258 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -60,6 +60,11 @@ static inline bool dump_is_64bit(DumpState *s)
return s->dump_info.d_class == ELFCLASS64;
}
+static inline bool dump_has_filter(DumpState *s)
+{
+ return s->filter_area_length > 0;
+}
+
uint16_t cpu_to_dump16(DumpState *s, uint16_t val)
{
if (s->dump_info.d_endian == ELFDATA2LSB) {
@@ -444,29 +449,30 @@ static void get_offset_range(hwaddr phys_addr,
*p_offset = -1;
*p_filesz = 0;
- if (s->has_filter) {
- if (phys_addr < s->begin || phys_addr >= s->begin + s->length) {
+ if (dump_has_filter(s)) {
+ if (phys_addr < s->filter_area_begin ||
+ phys_addr >= s->filter_area_begin + s->filter_area_length) {
return;
}
}
QTAILQ_FOREACH(block, &s->guest_phys_blocks.head, next) {
- if (s->has_filter) {
- if (block->target_start >= s->begin + s->length ||
- block->target_end <= s->begin) {
+ if (dump_has_filter(s)) {
+ if (block->target_start >= s->filter_area_begin + s->filter_area_length ||
+ block->target_end <= s->filter_area_begin) {
/* This block is out of the range */
continue;
}
- if (s->begin <= block->target_start) {
+ if (s->filter_area_begin <= block->target_start) {
start = block->target_start;
} else {
- start = s->begin;
+ start = s->filter_area_begin;
}
size_in_block = block->target_end - start;
- if (s->begin + s->length < block->target_end) {
- size_in_block -= block->target_end - (s->begin + s->length);
+ if (s->filter_area_begin + s->filter_area_length < block->target_end) {
+ size_in_block -= block->target_end - (s->filter_area_begin + s->filter_area_length);
}
} else {
start = block->target_start;
@@ -639,12 +645,12 @@ static void dump_iterate(DumpState *s, Error **errp)
int64_t memblock_size, memblock_start;
QTAILQ_FOREACH(block, &s->guest_phys_blocks.head, next) {
- memblock_start = dump_filtered_memblock_start(block, s->begin, s->length);
+ memblock_start = dump_filtered_memblock_start(block, s->filter_area_begin, s->filter_area_length);
if (memblock_start == -1) {
continue;
}
- memblock_size = dump_filtered_memblock_size(block, s->begin, s->length);
+ memblock_size = dump_filtered_memblock_size(block, s->filter_area_begin, s->filter_area_length);
/* Write the memory to file */
write_memory(s, block, memblock_start, memblock_size, errp);
@@ -1513,14 +1519,14 @@ static int validate_start_block(DumpState *s)
{
GuestPhysBlock *block;
- if (!s->has_filter) {
+ if (!dump_has_filter(s)) {
return 0;
}
QTAILQ_FOREACH(block, &s->guest_phys_blocks.head, next) {
/* This block is out of the range */
- if (block->target_start >= s->begin + s->length ||
- block->target_end <= s->begin) {
+ if (block->target_start >= s->filter_area_begin + s->filter_area_length ||
+ block->target_end <= s->filter_area_begin) {
continue;
}
return 0;
@@ -1559,10 +1565,10 @@ static int64_t dump_calculate_size(DumpState *s)
int64_t size = 0, total = 0, left = 0, right = 0;
QTAILQ_FOREACH(block, &s->guest_phys_blocks.head, next) {
- if (s->has_filter) {
+ if (dump_has_filter(s)) {
/* calculate the overlapped region. */
- left = MAX(s->begin, block->target_start);
- right = MIN(s->begin + s->length, block->target_end);
+ left = MAX(s->filter_area_begin, block->target_start);
+ right = MIN(s->filter_area_begin + s->filter_area_length, block->target_end);
size = right - left;
size = size > 0 ? size : 0;
} else {
@@ -1652,9 +1658,12 @@ static void dump_init(DumpState *s, int fd, bool has_format,
}
s->fd = fd;
- s->has_filter = has_filter;
- s->begin = begin;
- s->length = length;
+ if (has_filter && !length) {
+ error_setg(errp, QERR_INVALID_PARAMETER, "length");
+ goto cleanup;
+ }
+ s->filter_area_begin = begin;
+ s->filter_area_length = length;
memory_mapping_list_init(&s->list);
@@ -1787,8 +1796,8 @@ static void dump_init(DumpState *s, int fd, bool has_format,
return;
}
- if (s->has_filter) {
- memory_mapping_filter(&s->list, s->begin, s->length);
+ if (dump_has_filter(s)) {
+ memory_mapping_filter(&s->list, s->filter_area_begin, s->filter_area_length);
}
/*
diff --git a/include/sysemu/dump.h b/include/sysemu/dump.h
index 7fce1d4af6..b62513d87d 100644
--- a/include/sysemu/dump.h
+++ b/include/sysemu/dump.h
@@ -166,9 +166,16 @@ typedef struct DumpState {
hwaddr memory_offset;
int fd;
- bool has_filter;
- int64_t begin;
- int64_t length;
+ /*
+ * Dump filter area variables
+ *
+ * A filtered dump only contains the guest memory designated by
+ * the start address and length variables defined below.
+ *
+ * If length is 0, no filtering is applied.
+ */
+ int64_t filter_area_begin; /* Start address of partial guest memory area */
+ int64_t filter_area_length; /* Length of partial guest memory area */
uint8_t *note_buf; /* buffer for notes */
size_t note_buf_offset; /* the writing place in note_buf */
--
2.37.3

View File

@ -0,0 +1,102 @@
From b56c362132baef40cc25d910c1e0d217d83cfe44 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 11 Aug 2022 12:10:57 +0000
Subject: [PATCH 25/42] dump: Rework get_start_block
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [25/41] c93842a1aaeadcc11e91c194452fcd05d163b3ca
get_start_block() returns the start address of the first memory block
or -1.
With the GuestPhysBlock iterator conversion we don't need to set the
start address and can therefore remove that code and the "start"
DumpState struct member. The only functionality left is the validation
of the start block so it only makes sense to re-name the function to
validate_start_block()
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Message-Id: <20220811121111.9878-5-frankja@linux.ibm.com>
(cherry picked from commit 0c2994ac9009577b967529ce18e269da5b280351)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 20 ++++++--------------
include/sysemu/dump.h | 2 --
2 files changed, 6 insertions(+), 16 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index d981e843dd..e6aa037f59 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -1509,30 +1509,22 @@ static void create_kdump_vmcore(DumpState *s, Error **errp)
}
}
-static ram_addr_t get_start_block(DumpState *s)
+static int validate_start_block(DumpState *s)
{
GuestPhysBlock *block;
if (!s->has_filter) {
- s->next_block = QTAILQ_FIRST(&s->guest_phys_blocks.head);
return 0;
}
QTAILQ_FOREACH(block, &s->guest_phys_blocks.head, next) {
+ /* This block is out of the range */
if (block->target_start >= s->begin + s->length ||
block->target_end <= s->begin) {
- /* This block is out of the range */
continue;
}
-
- s->next_block = block;
- if (s->begin > block->target_start) {
- s->start = s->begin - block->target_start;
- } else {
- s->start = 0;
- }
- return s->start;
- }
+ return 0;
+ }
return -1;
}
@@ -1679,8 +1671,8 @@ static void dump_init(DumpState *s, int fd, bool has_format,
goto cleanup;
}
- s->start = get_start_block(s);
- if (s->start == -1) {
+ /* Is the filter filtering everything? */
+ if (validate_start_block(s) == -1) {
error_setg(errp, QERR_INVALID_PARAMETER, "begin");
goto cleanup;
}
diff --git a/include/sysemu/dump.h b/include/sysemu/dump.h
index ffc2ea1072..7fce1d4af6 100644
--- a/include/sysemu/dump.h
+++ b/include/sysemu/dump.h
@@ -166,8 +166,6 @@ typedef struct DumpState {
hwaddr memory_offset;
int fd;
- GuestPhysBlock *next_block;
- ram_addr_t start;
bool has_filter;
int64_t begin;
int64_t length;
--
2.37.3

View File

@ -0,0 +1,173 @@
From d1e147a3133d4d31d4b0c02c05916366fadd9c30 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Thu, 11 Aug 2022 12:11:00 +0000
Subject: [PATCH 28/42] dump: Split elf header functions into prepare and write
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [28/41] f70a13ad443835e7f46b7c5e176e372d370ac797
Let's split the write from the modification of the elf header so we
can consolidate the write of the data in one function.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220811121111.9878-8-frankja@linux.ibm.com>
(cherry picked from commit 670e76998a61ca171200fcded3865b294a2d1243)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 100 ++++++++++++++++++++++++++++------------------------
1 file changed, 53 insertions(+), 47 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 902a85ef8e..8d5226f861 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -132,7 +132,7 @@ static int fd_write_vmcore(const void *buf, size_t size, void *opaque)
return 0;
}
-static void write_elf64_header(DumpState *s, Error **errp)
+static void prepare_elf64_header(DumpState *s, Elf64_Ehdr *elf_header)
{
/*
* phnum in the elf header is 16 bit, if we have more segments we
@@ -140,34 +140,27 @@ static void write_elf64_header(DumpState *s, Error **errp)
* special section.
*/
uint16_t phnum = MIN(s->phdr_num, PN_XNUM);
- Elf64_Ehdr elf_header;
- int ret;
- memset(&elf_header, 0, sizeof(Elf64_Ehdr));
- memcpy(&elf_header, ELFMAG, SELFMAG);
- elf_header.e_ident[EI_CLASS] = ELFCLASS64;
- elf_header.e_ident[EI_DATA] = s->dump_info.d_endian;
- elf_header.e_ident[EI_VERSION] = EV_CURRENT;
- elf_header.e_type = cpu_to_dump16(s, ET_CORE);
- elf_header.e_machine = cpu_to_dump16(s, s->dump_info.d_machine);
- elf_header.e_version = cpu_to_dump32(s, EV_CURRENT);
- elf_header.e_ehsize = cpu_to_dump16(s, sizeof(elf_header));
- elf_header.e_phoff = cpu_to_dump64(s, s->phdr_offset);
- elf_header.e_phentsize = cpu_to_dump16(s, sizeof(Elf64_Phdr));
- elf_header.e_phnum = cpu_to_dump16(s, phnum);
+ memset(elf_header, 0, sizeof(Elf64_Ehdr));
+ memcpy(elf_header, ELFMAG, SELFMAG);
+ elf_header->e_ident[EI_CLASS] = ELFCLASS64;
+ elf_header->e_ident[EI_DATA] = s->dump_info.d_endian;
+ elf_header->e_ident[EI_VERSION] = EV_CURRENT;
+ elf_header->e_type = cpu_to_dump16(s, ET_CORE);
+ elf_header->e_machine = cpu_to_dump16(s, s->dump_info.d_machine);
+ elf_header->e_version = cpu_to_dump32(s, EV_CURRENT);
+ elf_header->e_ehsize = cpu_to_dump16(s, sizeof(elf_header));
+ elf_header->e_phoff = cpu_to_dump64(s, s->phdr_offset);
+ elf_header->e_phentsize = cpu_to_dump16(s, sizeof(Elf64_Phdr));
+ elf_header->e_phnum = cpu_to_dump16(s, phnum);
if (s->shdr_num) {
- elf_header.e_shoff = cpu_to_dump64(s, s->shdr_offset);
- elf_header.e_shentsize = cpu_to_dump16(s, sizeof(Elf64_Shdr));
- elf_header.e_shnum = cpu_to_dump16(s, s->shdr_num);
- }
-
- ret = fd_write_vmcore(&elf_header, sizeof(elf_header), s);
- if (ret < 0) {
- error_setg_errno(errp, -ret, "dump: failed to write elf header");
+ elf_header->e_shoff = cpu_to_dump64(s, s->shdr_offset);
+ elf_header->e_shentsize = cpu_to_dump16(s, sizeof(Elf64_Shdr));
+ elf_header->e_shnum = cpu_to_dump16(s, s->shdr_num);
}
}
-static void write_elf32_header(DumpState *s, Error **errp)
+static void prepare_elf32_header(DumpState *s, Elf32_Ehdr *elf_header)
{
/*
* phnum in the elf header is 16 bit, if we have more segments we
@@ -175,28 +168,45 @@ static void write_elf32_header(DumpState *s, Error **errp)
* special section.
*/
uint16_t phnum = MIN(s->phdr_num, PN_XNUM);
- Elf32_Ehdr elf_header;
- int ret;
- memset(&elf_header, 0, sizeof(Elf32_Ehdr));
- memcpy(&elf_header, ELFMAG, SELFMAG);
- elf_header.e_ident[EI_CLASS] = ELFCLASS32;
- elf_header.e_ident[EI_DATA] = s->dump_info.d_endian;
- elf_header.e_ident[EI_VERSION] = EV_CURRENT;
- elf_header.e_type = cpu_to_dump16(s, ET_CORE);
- elf_header.e_machine = cpu_to_dump16(s, s->dump_info.d_machine);
- elf_header.e_version = cpu_to_dump32(s, EV_CURRENT);
- elf_header.e_ehsize = cpu_to_dump16(s, sizeof(elf_header));
- elf_header.e_phoff = cpu_to_dump32(s, s->phdr_offset);
- elf_header.e_phentsize = cpu_to_dump16(s, sizeof(Elf32_Phdr));
- elf_header.e_phnum = cpu_to_dump16(s, phnum);
+ memset(elf_header, 0, sizeof(Elf32_Ehdr));
+ memcpy(elf_header, ELFMAG, SELFMAG);
+ elf_header->e_ident[EI_CLASS] = ELFCLASS32;
+ elf_header->e_ident[EI_DATA] = s->dump_info.d_endian;
+ elf_header->e_ident[EI_VERSION] = EV_CURRENT;
+ elf_header->e_type = cpu_to_dump16(s, ET_CORE);
+ elf_header->e_machine = cpu_to_dump16(s, s->dump_info.d_machine);
+ elf_header->e_version = cpu_to_dump32(s, EV_CURRENT);
+ elf_header->e_ehsize = cpu_to_dump16(s, sizeof(elf_header));
+ elf_header->e_phoff = cpu_to_dump32(s, s->phdr_offset);
+ elf_header->e_phentsize = cpu_to_dump16(s, sizeof(Elf32_Phdr));
+ elf_header->e_phnum = cpu_to_dump16(s, phnum);
if (s->shdr_num) {
- elf_header.e_shoff = cpu_to_dump32(s, s->shdr_offset);
- elf_header.e_shentsize = cpu_to_dump16(s, sizeof(Elf32_Shdr));
- elf_header.e_shnum = cpu_to_dump16(s, s->shdr_num);
+ elf_header->e_shoff = cpu_to_dump32(s, s->shdr_offset);
+ elf_header->e_shentsize = cpu_to_dump16(s, sizeof(Elf32_Shdr));
+ elf_header->e_shnum = cpu_to_dump16(s, s->shdr_num);
}
+}
- ret = fd_write_vmcore(&elf_header, sizeof(elf_header), s);
+static void write_elf_header(DumpState *s, Error **errp)
+{
+ Elf32_Ehdr elf32_header;
+ Elf64_Ehdr elf64_header;
+ size_t header_size;
+ void *header_ptr;
+ int ret;
+
+ if (dump_is_64bit(s)) {
+ prepare_elf64_header(s, &elf64_header);
+ header_size = sizeof(elf64_header);
+ header_ptr = &elf64_header;
+ } else {
+ prepare_elf32_header(s, &elf32_header);
+ header_size = sizeof(elf32_header);
+ header_ptr = &elf32_header;
+ }
+
+ ret = fd_write_vmcore(header_ptr, header_size, s);
if (ret < 0) {
error_setg_errno(errp, -ret, "dump: failed to write elf header");
}
@@ -565,11 +575,7 @@ static void dump_begin(DumpState *s, Error **errp)
*/
/* write elf header to vmcore */
- if (dump_is_64bit(s)) {
- write_elf64_header(s, errp);
- } else {
- write_elf32_header(s, errp);
- }
+ write_elf_header(s, errp);
if (*errp) {
return;
}
--
2.37.3

View File

@ -0,0 +1,420 @@
From 4ca61efe246d62d420eb332655c0c8ead4cc762b Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Wed, 30 Mar 2022 12:35:55 +0000
Subject: [PATCH 13/42] dump: Use ERRP_GUARD()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [13/41] f735cd1dab0230000cfadd878765fdf4647b239c
Let's move to the new way of handling errors before changing the dump
code. This patch has mostly been generated by the coccinelle script
scripts/coccinelle/errp-guard.cocci.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220330123603.107120-2-frankja@linux.ibm.com>
(cherry picked from commit 86a518bba4f4d7c9016fc5b104fe1e58b00ad756)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 144 ++++++++++++++++++++++------------------------------
1 file changed, 61 insertions(+), 83 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 662d0a62cd..9876123f2e 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -390,23 +390,21 @@ static void write_data(DumpState *s, void *buf, int length, Error **errp)
static void write_memory(DumpState *s, GuestPhysBlock *block, ram_addr_t start,
int64_t size, Error **errp)
{
+ ERRP_GUARD();
int64_t i;
- Error *local_err = NULL;
for (i = 0; i < size / s->dump_info.page_size; i++) {
write_data(s, block->host_addr + start + i * s->dump_info.page_size,
- s->dump_info.page_size, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ s->dump_info.page_size, errp);
+ if (*errp) {
return;
}
}
if ((size % s->dump_info.page_size) != 0) {
write_data(s, block->host_addr + start + i * s->dump_info.page_size,
- size % s->dump_info.page_size, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ size % s->dump_info.page_size, errp);
+ if (*errp) {
return;
}
}
@@ -476,11 +474,11 @@ static void get_offset_range(hwaddr phys_addr,
static void write_elf_loads(DumpState *s, Error **errp)
{
+ ERRP_GUARD();
hwaddr offset, filesz;
MemoryMapping *memory_mapping;
uint32_t phdr_index = 1;
uint32_t max_index;
- Error *local_err = NULL;
if (s->have_section) {
max_index = s->sh_info;
@@ -494,14 +492,13 @@ static void write_elf_loads(DumpState *s, Error **errp)
s, &offset, &filesz);
if (s->dump_info.d_class == ELFCLASS64) {
write_elf64_load(s, memory_mapping, phdr_index++, offset,
- filesz, &local_err);
+ filesz, errp);
} else {
write_elf32_load(s, memory_mapping, phdr_index++, offset,
- filesz, &local_err);
+ filesz, errp);
}
- if (local_err) {
- error_propagate(errp, local_err);
+ if (*errp) {
return;
}
@@ -514,7 +511,7 @@ static void write_elf_loads(DumpState *s, Error **errp)
/* write elf header, PT_NOTE and elf note to vmcore. */
static void dump_begin(DumpState *s, Error **errp)
{
- Error *local_err = NULL;
+ ERRP_GUARD();
/*
* the vmcore's format is:
@@ -542,73 +539,64 @@ static void dump_begin(DumpState *s, Error **errp)
/* write elf header to vmcore */
if (s->dump_info.d_class == ELFCLASS64) {
- write_elf64_header(s, &local_err);
+ write_elf64_header(s, errp);
} else {
- write_elf32_header(s, &local_err);
+ write_elf32_header(s, errp);
}
- if (local_err) {
- error_propagate(errp, local_err);
+ if (*errp) {
return;
}
if (s->dump_info.d_class == ELFCLASS64) {
/* write PT_NOTE to vmcore */
- write_elf64_note(s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf64_note(s, errp);
+ if (*errp) {
return;
}
/* write all PT_LOAD to vmcore */
- write_elf_loads(s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf_loads(s, errp);
+ if (*errp) {
return;
}
/* write section to vmcore */
if (s->have_section) {
- write_elf_section(s, 1, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf_section(s, 1, errp);
+ if (*errp) {
return;
}
}
/* write notes to vmcore */
- write_elf64_notes(fd_write_vmcore, s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf64_notes(fd_write_vmcore, s, errp);
+ if (*errp) {
return;
}
} else {
/* write PT_NOTE to vmcore */
- write_elf32_note(s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf32_note(s, errp);
+ if (*errp) {
return;
}
/* write all PT_LOAD to vmcore */
- write_elf_loads(s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf_loads(s, errp);
+ if (*errp) {
return;
}
/* write section to vmcore */
if (s->have_section) {
- write_elf_section(s, 0, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf_section(s, 0, errp);
+ if (*errp) {
return;
}
}
/* write notes to vmcore */
- write_elf32_notes(fd_write_vmcore, s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf32_notes(fd_write_vmcore, s, errp);
+ if (*errp) {
return;
}
}
@@ -644,9 +632,9 @@ static int get_next_block(DumpState *s, GuestPhysBlock *block)
/* write all memory to vmcore */
static void dump_iterate(DumpState *s, Error **errp)
{
+ ERRP_GUARD();
GuestPhysBlock *block;
int64_t size;
- Error *local_err = NULL;
do {
block = s->next_block;
@@ -658,9 +646,8 @@ static void dump_iterate(DumpState *s, Error **errp)
size -= block->target_end - (s->begin + s->length);
}
}
- write_memory(s, block, s->start, size, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_memory(s, block, s->start, size, errp);
+ if (*errp) {
return;
}
@@ -669,11 +656,10 @@ static void dump_iterate(DumpState *s, Error **errp)
static void create_vmcore(DumpState *s, Error **errp)
{
- Error *local_err = NULL;
+ ERRP_GUARD();
- dump_begin(s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ dump_begin(s, errp);
+ if (*errp) {
return;
}
@@ -810,6 +796,7 @@ static bool note_name_equal(DumpState *s,
/* write common header, sub header and elf note to vmcore */
static void create_header32(DumpState *s, Error **errp)
{
+ ERRP_GUARD();
DiskDumpHeader32 *dh = NULL;
KdumpSubHeader32 *kh = NULL;
size_t size;
@@ -818,7 +805,6 @@ static void create_header32(DumpState *s, Error **errp)
uint32_t bitmap_blocks;
uint32_t status = 0;
uint64_t offset_note;
- Error *local_err = NULL;
/* write common header, the version of kdump-compressed format is 6th */
size = sizeof(DiskDumpHeader32);
@@ -894,9 +880,8 @@ static void create_header32(DumpState *s, Error **errp)
s->note_buf_offset = 0;
/* use s->note_buf to store notes temporarily */
- write_elf32_notes(buf_write_note, s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf32_notes(buf_write_note, s, errp);
+ if (*errp) {
goto out;
}
if (write_buffer(s->fd, offset_note, s->note_buf,
@@ -922,6 +907,7 @@ out:
/* write common header, sub header and elf note to vmcore */
static void create_header64(DumpState *s, Error **errp)
{
+ ERRP_GUARD();
DiskDumpHeader64 *dh = NULL;
KdumpSubHeader64 *kh = NULL;
size_t size;
@@ -930,7 +916,6 @@ static void create_header64(DumpState *s, Error **errp)
uint32_t bitmap_blocks;
uint32_t status = 0;
uint64_t offset_note;
- Error *local_err = NULL;
/* write common header, the version of kdump-compressed format is 6th */
size = sizeof(DiskDumpHeader64);
@@ -1006,9 +991,8 @@ static void create_header64(DumpState *s, Error **errp)
s->note_buf_offset = 0;
/* use s->note_buf to store notes temporarily */
- write_elf64_notes(buf_write_note, s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_elf64_notes(buf_write_note, s, errp);
+ if (*errp) {
goto out;
}
@@ -1472,8 +1456,8 @@ out:
static void create_kdump_vmcore(DumpState *s, Error **errp)
{
+ ERRP_GUARD();
int ret;
- Error *local_err = NULL;
/*
* the kdump-compressed format is:
@@ -1503,21 +1487,18 @@ static void create_kdump_vmcore(DumpState *s, Error **errp)
return;
}
- write_dump_header(s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_dump_header(s, errp);
+ if (*errp) {
return;
}
- write_dump_bitmap(s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_dump_bitmap(s, errp);
+ if (*errp) {
return;
}
- write_dump_pages(s, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ write_dump_pages(s, errp);
+ if (*errp) {
return;
}
@@ -1647,10 +1628,10 @@ static void dump_init(DumpState *s, int fd, bool has_format,
DumpGuestMemoryFormat format, bool paging, bool has_filter,
int64_t begin, int64_t length, Error **errp)
{
+ ERRP_GUARD();
VMCoreInfoState *vmci = vmcoreinfo_find();
CPUState *cpu;
int nr_cpus;
- Error *err = NULL;
int ret;
s->has_format = has_format;
@@ -1769,9 +1750,8 @@ static void dump_init(DumpState *s, int fd, bool has_format,
/* get memory mapping */
if (paging) {
- qemu_get_guest_memory_mapping(&s->list, &s->guest_phys_blocks, &err);
- if (err != NULL) {
- error_propagate(errp, err);
+ qemu_get_guest_memory_mapping(&s->list, &s->guest_phys_blocks, errp);
+ if (*errp) {
goto cleanup;
}
} else {
@@ -1870,33 +1850,32 @@ cleanup:
/* this operation might be time consuming. */
static void dump_process(DumpState *s, Error **errp)
{
- Error *local_err = NULL;
+ ERRP_GUARD();
DumpQueryResult *result = NULL;
if (s->has_format && s->format == DUMP_GUEST_MEMORY_FORMAT_WIN_DMP) {
#ifdef TARGET_X86_64
- create_win_dump(s, &local_err);
+ create_win_dump(s, errp);
#endif
} else if (s->has_format && s->format != DUMP_GUEST_MEMORY_FORMAT_ELF) {
- create_kdump_vmcore(s, &local_err);
+ create_kdump_vmcore(s, errp);
} else {
- create_vmcore(s, &local_err);
+ create_vmcore(s, errp);
}
/* make sure status is written after written_size updates */
smp_wmb();
qatomic_set(&s->status,
- (local_err ? DUMP_STATUS_FAILED : DUMP_STATUS_COMPLETED));
+ (*errp ? DUMP_STATUS_FAILED : DUMP_STATUS_COMPLETED));
/* send DUMP_COMPLETED message (unconditionally) */
result = qmp_query_dump(NULL);
/* should never fail */
assert(result);
- qapi_event_send_dump_completed(result, !!local_err, (local_err ?
- error_get_pretty(local_err) : NULL));
+ qapi_event_send_dump_completed(result, !!*errp, (*errp ?
+ error_get_pretty(*errp) : NULL));
qapi_free_DumpQueryResult(result);
- error_propagate(errp, local_err);
dump_cleanup(s);
}
@@ -1925,10 +1904,10 @@ void qmp_dump_guest_memory(bool paging, const char *file,
int64_t length, bool has_format,
DumpGuestMemoryFormat format, Error **errp)
{
+ ERRP_GUARD();
const char *p;
int fd = -1;
DumpState *s;
- Error *local_err = NULL;
bool detach_p = false;
if (runstate_check(RUN_STATE_INMIGRATE)) {
@@ -2028,9 +2007,8 @@ void qmp_dump_guest_memory(bool paging, const char *file,
dump_state_prepare(s);
dump_init(s, fd, has_format, format, paging, has_begin,
- begin, length, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
+ begin, length, errp);
+ if (*errp) {
qatomic_set(&s->status, DUMP_STATUS_FAILED);
return;
}
--
2.37.3

View File

@ -0,0 +1,150 @@
From a918c7305ec7c68e8bc37b449f71e75d84124cd0 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Mon, 17 Oct 2022 08:38:13 +0000
Subject: [PATCH 32/42] dump: Use a buffer for ELF section data and headers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [32/41] e1a03e202e67764581e486f37e13e479200e5846
Currently we're writing the NULL section header if we overflow the
physical header number in the ELF header. But in the future we'll add
custom section headers AND section data.
To facilitate this we need to rearange section handling a bit. As with
the other ELF headers we split the code into a prepare and a write
step.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20221017083822.43118-2-frankja@linux.ibm.com>
(cherry picked from commit e41ed29bcee5cb16715317bcf290f6b5c196eb0a)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 75 +++++++++++++++++++++++++++++--------------
include/sysemu/dump.h | 2 ++
2 files changed, 53 insertions(+), 24 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 88177fa886..4142b4cc0c 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -381,31 +381,60 @@ static void write_elf_phdr_note(DumpState *s, Error **errp)
}
}
-static void write_elf_section(DumpState *s, int type, Error **errp)
+static void prepare_elf_section_hdr_zero(DumpState *s)
{
- Elf32_Shdr shdr32;
- Elf64_Shdr shdr64;
- int shdr_size;
- void *shdr;
- int ret;
+ if (dump_is_64bit(s)) {
+ Elf64_Shdr *shdr64 = s->elf_section_hdrs;
- if (type == 0) {
- shdr_size = sizeof(Elf32_Shdr);
- memset(&shdr32, 0, shdr_size);
- shdr32.sh_info = cpu_to_dump32(s, s->phdr_num);
- shdr = &shdr32;
+ shdr64->sh_info = cpu_to_dump32(s, s->phdr_num);
} else {
- shdr_size = sizeof(Elf64_Shdr);
- memset(&shdr64, 0, shdr_size);
- shdr64.sh_info = cpu_to_dump32(s, s->phdr_num);
- shdr = &shdr64;
+ Elf32_Shdr *shdr32 = s->elf_section_hdrs;
+
+ shdr32->sh_info = cpu_to_dump32(s, s->phdr_num);
+ }
+}
+
+static void prepare_elf_section_hdrs(DumpState *s)
+{
+ size_t len, sizeof_shdr;
+
+ /*
+ * Section ordering:
+ * - HDR zero
+ */
+ sizeof_shdr = dump_is_64bit(s) ? sizeof(Elf64_Shdr) : sizeof(Elf32_Shdr);
+ len = sizeof_shdr * s->shdr_num;
+ s->elf_section_hdrs = g_malloc0(len);
+
+ /*
+ * The first section header is ALWAYS a special initial section
+ * header.
+ *
+ * The header should be 0 with one exception being that if
+ * phdr_num is PN_XNUM then the sh_info field contains the real
+ * number of segment entries.
+ *
+ * As we zero allocate the buffer we will only need to modify
+ * sh_info for the PN_XNUM case.
+ */
+ if (s->phdr_num >= PN_XNUM) {
+ prepare_elf_section_hdr_zero(s);
}
+}
- ret = fd_write_vmcore(shdr, shdr_size, s);
+static void write_elf_section_headers(DumpState *s, Error **errp)
+{
+ size_t sizeof_shdr = dump_is_64bit(s) ? sizeof(Elf64_Shdr) : sizeof(Elf32_Shdr);
+ int ret;
+
+ prepare_elf_section_hdrs(s);
+
+ ret = fd_write_vmcore(s->elf_section_hdrs, s->shdr_num * sizeof_shdr, s);
if (ret < 0) {
- error_setg_errno(errp, -ret,
- "dump: failed to write section header table");
+ error_setg_errno(errp, -ret, "dump: failed to write section headers");
}
+
+ g_free(s->elf_section_hdrs);
}
static void write_data(DumpState *s, void *buf, int length, Error **errp)
@@ -592,12 +621,10 @@ static void dump_begin(DumpState *s, Error **errp)
return;
}
- /* write section to vmcore */
- if (s->shdr_num) {
- write_elf_section(s, 1, errp);
- if (*errp) {
- return;
- }
+ /* write section headers to vmcore */
+ write_elf_section_headers(s, errp);
+ if (*errp) {
+ return;
}
/* write notes to vmcore */
diff --git a/include/sysemu/dump.h b/include/sysemu/dump.h
index b62513d87d..9995f65dc8 100644
--- a/include/sysemu/dump.h
+++ b/include/sysemu/dump.h
@@ -177,6 +177,8 @@ typedef struct DumpState {
int64_t filter_area_begin; /* Start address of partial guest memory area */
int64_t filter_area_length; /* Length of partial guest memory area */
+ void *elf_section_hdrs; /* Pointer to section header buffer */
+
uint8_t *note_buf; /* buffer for notes */
size_t note_buf_offset; /* the writing place in note_buf */
uint32_t nr_cpus; /* number of guest's cpu */
--
2.37.3

View File

@ -0,0 +1,104 @@
From 987ede93fa4e3d058acddc19874e467faa116ede Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Mon, 17 Oct 2022 08:38:14 +0000
Subject: [PATCH 33/42] dump: Write ELF section headers right after ELF header
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [33/41] e956040753533ac376e9763145192de1e216027d
Let's start bundling the writes of the headers and of the data so we
have a clear ordering between them. Since the ELF header uses offsets
to the headers we can freely order them.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20221017083822.43118-3-frankja@linux.ibm.com>
(cherry picked from commit cb415fd61e48d52f81dcf38956e3f913651cff1c)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 31 ++++++++++++++-----------------
1 file changed, 14 insertions(+), 17 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 4142b4cc0c..d17537d4e9 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -584,6 +584,8 @@ static void dump_begin(DumpState *s, Error **errp)
* --------------
* | elf header |
* --------------
+ * | sctn_hdr |
+ * --------------
* | PT_NOTE |
* --------------
* | PT_LOAD |
@@ -592,8 +594,6 @@ static void dump_begin(DumpState *s, Error **errp)
* --------------
* | PT_LOAD |
* --------------
- * | sec_hdr |
- * --------------
* | elf note |
* --------------
* | memory |
@@ -609,20 +609,20 @@ static void dump_begin(DumpState *s, Error **errp)
return;
}
- /* write PT_NOTE to vmcore */
- write_elf_phdr_note(s, errp);
+ /* write section headers to vmcore */
+ write_elf_section_headers(s, errp);
if (*errp) {
return;
}
- /* write all PT_LOADs to vmcore */
- write_elf_phdr_loads(s, errp);
+ /* write PT_NOTE to vmcore */
+ write_elf_phdr_note(s, errp);
if (*errp) {
return;
}
- /* write section headers to vmcore */
- write_elf_section_headers(s, errp);
+ /* write all PT_LOADs to vmcore */
+ write_elf_phdr_loads(s, errp);
if (*errp) {
return;
}
@@ -1877,16 +1877,13 @@ static void dump_init(DumpState *s, int fd, bool has_format,
}
if (dump_is_64bit(s)) {
- s->phdr_offset = sizeof(Elf64_Ehdr);
- s->shdr_offset = s->phdr_offset + sizeof(Elf64_Phdr) * s->phdr_num;
- s->note_offset = s->shdr_offset + sizeof(Elf64_Shdr) * s->shdr_num;
- s->memory_offset = s->note_offset + s->note_size;
+ s->shdr_offset = sizeof(Elf64_Ehdr);
+ s->phdr_offset = s->shdr_offset + sizeof(Elf64_Shdr) * s->shdr_num;
+ s->note_offset = s->phdr_offset + sizeof(Elf64_Phdr) * s->phdr_num;
} else {
-
- s->phdr_offset = sizeof(Elf32_Ehdr);
- s->shdr_offset = s->phdr_offset + sizeof(Elf32_Phdr) * s->phdr_num;
- s->note_offset = s->shdr_offset + sizeof(Elf32_Shdr) * s->shdr_num;
- s->memory_offset = s->note_offset + s->note_size;
+ s->shdr_offset = sizeof(Elf32_Ehdr);
+ s->phdr_offset = s->shdr_offset + sizeof(Elf32_Shdr) * s->shdr_num;
+ s->note_offset = s->phdr_offset + sizeof(Elf32_Phdr) * s->phdr_num;
}
return;
--
2.37.3

View File

@ -0,0 +1,173 @@
From deaf4e0f5e90d227b7b9f3e5d1dff7fd0bc0206a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Marc-Andr=C3=A9=20Lureau?= <marcandre.lureau@redhat.com>
Date: Mon, 5 Sep 2022 16:06:21 +0400
Subject: [PATCH 31/42] dump: fix kdump to work over non-aligned blocks
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [31/41] b307bdce4a4791fc30160fa2a1678bd238f2432e
Rewrite get_next_page() to work over non-aligned blocks. When it
encounters non aligned addresses, it will try to fill a page provided by
the caller.
This solves a kdump crash with "tpm-crb-cmd" RAM memory region,
qemu-kvm: ../dump/dump.c:1162: _Bool get_next_page(GuestPhysBlock **,
uint64_t *, uint8_t **, DumpState *): Assertion `(block->target_start &
~target_page_mask) == 0' failed.
because:
guest_phys_block_add_section: target_start=00000000fed40080 target_end=00000000fed41000: added (count: 4)
Fixes:
https://bugzilla.redhat.com/show_bug.cgi?id=2120480
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Acked-by: David Hildenbrand <david@redhat.com>
(cherry picked from commit 94d788408d2d5a6474c99b2c9cf06913b9db7c58)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 79 +++++++++++++++++++++++++++++++++++++----------------
1 file changed, 56 insertions(+), 23 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 1c49232390..88177fa886 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -1117,50 +1117,81 @@ static uint64_t dump_pfn_to_paddr(DumpState *s, uint64_t pfn)
}
/*
- * exam every page and return the page frame number and the address of the page.
- * bufptr can be NULL. note: the blocks here is supposed to reflect guest-phys
- * blocks, so block->target_start and block->target_end should be interal
- * multiples of the target page size.
+ * Return the page frame number and the page content in *bufptr. bufptr can be
+ * NULL. If not NULL, *bufptr must contains a target page size of pre-allocated
+ * memory. This is not necessarily the memory returned.
*/
static bool get_next_page(GuestPhysBlock **blockptr, uint64_t *pfnptr,
uint8_t **bufptr, DumpState *s)
{
GuestPhysBlock *block = *blockptr;
- hwaddr addr, target_page_mask = ~((hwaddr)s->dump_info.page_size - 1);
- uint8_t *buf;
+ uint32_t page_size = s->dump_info.page_size;
+ uint8_t *buf = NULL, *hbuf;
+ hwaddr addr;
/* block == NULL means the start of the iteration */
if (!block) {
block = QTAILQ_FIRST(&s->guest_phys_blocks.head);
*blockptr = block;
addr = block->target_start;
+ *pfnptr = dump_paddr_to_pfn(s, addr);
} else {
- addr = dump_pfn_to_paddr(s, *pfnptr + 1);
+ *pfnptr += 1;
+ addr = dump_pfn_to_paddr(s, *pfnptr);
}
assert(block != NULL);
- if ((addr >= block->target_start) &&
- (addr + s->dump_info.page_size <= block->target_end)) {
- buf = block->host_addr + (addr - block->target_start);
- } else {
- /* the next page is in the next block */
- block = QTAILQ_NEXT(block, next);
- *blockptr = block;
- if (!block) {
- return false;
+ while (1) {
+ if (addr >= block->target_start && addr < block->target_end) {
+ size_t n = MIN(block->target_end - addr, page_size - addr % page_size);
+ hbuf = block->host_addr + (addr - block->target_start);
+ if (!buf) {
+ if (n == page_size) {
+ /* this is a whole target page, go for it */
+ assert(addr % page_size == 0);
+ buf = hbuf;
+ break;
+ } else if (bufptr) {
+ assert(*bufptr);
+ buf = *bufptr;
+ memset(buf, 0, page_size);
+ } else {
+ return true;
+ }
+ }
+
+ memcpy(buf + addr % page_size, hbuf, n);
+ addr += n;
+ if (addr % page_size == 0) {
+ /* we filled up the page */
+ break;
+ }
+ } else {
+ /* the next page is in the next block */
+ *blockptr = block = QTAILQ_NEXT(block, next);
+ if (!block) {
+ break;
+ }
+
+ addr = block->target_start;
+ /* are we still in the same page? */
+ if (dump_paddr_to_pfn(s, addr) != *pfnptr) {
+ if (buf) {
+ /* no, but we already filled something earlier, return it */
+ break;
+ } else {
+ /* else continue from there */
+ *pfnptr = dump_paddr_to_pfn(s, addr);
+ }
+ }
}
- addr = block->target_start;
- buf = block->host_addr;
}
- assert((block->target_start & ~target_page_mask) == 0);
- assert((block->target_end & ~target_page_mask) == 0);
- *pfnptr = dump_paddr_to_pfn(s, addr);
if (bufptr) {
*bufptr = buf;
}
- return true;
+ return buf != NULL;
}
static void write_dump_bitmap(DumpState *s, Error **errp)
@@ -1306,6 +1337,7 @@ static void write_dump_pages(DumpState *s, Error **errp)
uint8_t *buf;
GuestPhysBlock *block_iter = NULL;
uint64_t pfn_iter;
+ g_autofree uint8_t *page = NULL;
/* get offset of page_desc and page_data in dump file */
offset_desc = s->offset_page;
@@ -1341,12 +1373,13 @@ static void write_dump_pages(DumpState *s, Error **errp)
}
offset_data += s->dump_info.page_size;
+ page = g_malloc(s->dump_info.page_size);
/*
* dump memory to vmcore page by page. zero page will all be resided in the
* first page of page section
*/
- while (get_next_page(&block_iter, &pfn_iter, &buf, s)) {
+ for (buf = page; get_next_page(&block_iter, &pfn_iter, &buf, s); buf = page) {
/* check zero page */
if (is_zero_page(buf, s->dump_info.page_size)) {
ret = write_cache(&page_desc, &pd_zero, sizeof(PageDescriptor),
--
2.37.3

View File

@ -0,0 +1,75 @@
From bb55fde4d8ca587e2ef52ce58a0c22e4d66a08dc Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Marc-Andr=C3=A9=20Lureau?= <marcandre.lureau@redhat.com>
Date: Thu, 25 Aug 2022 12:40:12 +0400
Subject: [PATCH 30/42] dump: simplify a bit kdump get_next_page()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [30/41] 417ac19fa96036e0242f40121ac6e87a9f3f70ba
This should be functionally equivalent, but slightly easier to read,
with simplified paths and checks at the end of the function.
The following patch is a major rewrite to get rid of the assert().
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
(cherry picked from commit 08df343874fcddd260021a04ce3c5a34f2c48164)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 21 ++++++++-------------
1 file changed, 8 insertions(+), 13 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index c2c1341ad7..1c49232390 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -1133,17 +1133,11 @@ static bool get_next_page(GuestPhysBlock **blockptr, uint64_t *pfnptr,
if (!block) {
block = QTAILQ_FIRST(&s->guest_phys_blocks.head);
*blockptr = block;
- assert((block->target_start & ~target_page_mask) == 0);
- assert((block->target_end & ~target_page_mask) == 0);
- *pfnptr = dump_paddr_to_pfn(s, block->target_start);
- if (bufptr) {
- *bufptr = block->host_addr;
- }
- return true;
+ addr = block->target_start;
+ } else {
+ addr = dump_pfn_to_paddr(s, *pfnptr + 1);
}
-
- *pfnptr = *pfnptr + 1;
- addr = dump_pfn_to_paddr(s, *pfnptr);
+ assert(block != NULL);
if ((addr >= block->target_start) &&
(addr + s->dump_info.page_size <= block->target_end)) {
@@ -1155,12 +1149,13 @@ static bool get_next_page(GuestPhysBlock **blockptr, uint64_t *pfnptr,
if (!block) {
return false;
}
- assert((block->target_start & ~target_page_mask) == 0);
- assert((block->target_end & ~target_page_mask) == 0);
- *pfnptr = dump_paddr_to_pfn(s, block->target_start);
+ addr = block->target_start;
buf = block->host_addr;
}
+ assert((block->target_start & ~target_page_mask) == 0);
+ assert((block->target_end & ~target_page_mask) == 0);
+ *pfnptr = dump_paddr_to_pfn(s, addr);
if (bufptr) {
*bufptr = buf;
}
--
2.37.3

View File

@ -0,0 +1,81 @@
From edead46187b1e55ad5e238332780aef19f1bc214 Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@redhat.com>
Date: Wed, 9 Nov 2022 18:41:18 -0500
Subject: [PATCH 1/2] hw/acpi: Add ospm_status hook implementation for acpi-ged
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 228: qemu-kvm: backport some aarch64 fixes
RH-Bugzilla: 2132609
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Igor Mammedov <imammedo@redhat.com>
RH-Acked-by: Eric Auger <eric.auger@redhat.com>
RH-Acked-by: Gavin Shan <gshan@redhat.com>
RH-Commit: [1/2] 99730b1a27666ca745dc28d90751c938d43f1682 (jmaloy/qemu-kvm)
BZ: https://bugzilla.redhat.com/show_bug.cgi?id=2132609
Upstream: Merged
commit d4424bebceaa8ffbc23060ce45e52a9bb817e3c9
Author: Keqian Zhu <zhukeqian1@huawei.com>
Date: Tue Aug 16 17:49:57 2022 +0800
hw/acpi: Add ospm_status hook implementation for acpi-ged
Setup an ARM virtual machine of machine virt and execute qmp "query-acpi-ospm-status"
causes segmentation fault with following dumpstack:
#1 0x0000aaaaab64235c in qmp_query_acpi_ospm_status (errp=errp@entry=0xfffffffff030) at ../monitor/qmp-cmds.c:312
#2 0x0000aaaaabfc4e20 in qmp_marshal_query_acpi_ospm_status (args=<optimized out>, ret=0xffffea4ffe90, errp=0xffffea4ffe88) at qapi/qapi-commands-acpi.c:63
#3 0x0000aaaaabff8ba0 in do_qmp_dispatch_bh (opaque=0xffffea4ffe98) at ../qapi/qmp-dispatch.c:128
#4 0x0000aaaaac02e594 in aio_bh_call (bh=0xffffe0004d80) at ../util/async.c:150
#5 aio_bh_poll (ctx=ctx@entry=0xaaaaad0f6040) at ../util/async.c:178
#6 0x0000aaaaac00bd40 in aio_dispatch (ctx=ctx@entry=0xaaaaad0f6040) at ../util/aio-posix.c:421
#7 0x0000aaaaac02e010 in aio_ctx_dispatch (source=0xaaaaad0f6040, callback=<optimized out>, user_data=<optimized out>) at ../util/async.c:320
#8 0x0000fffff76f6884 in g_main_context_dispatch () at /usr/lib64/libglib-2.0.so.0
#9 0x0000aaaaac0452d4 in glib_pollfds_poll () at ../util/main-loop.c:297
#10 os_host_main_loop_wait (timeout=0) at ../util/main-loop.c:320
#11 main_loop_wait (nonblocking=nonblocking@entry=0) at ../util/main-loop.c:596
#12 0x0000aaaaab5c9e50 in qemu_main_loop () at ../softmmu/runstate.c:734
#13 0x0000aaaaab185370 in qemu_main (argc=argc@entry=47, argv=argv@entry=0xfffffffff518, envp=envp@entry=0x0) at ../softmmu/main.c:38
#14 0x0000aaaaab16f99c in main (argc=47, argv=0xfffffffff518) at ../softmmu/main.c:47
Fixes: ebb62075021a ("hw/acpi: Add ACPI Generic Event Device Support")
Signed-off-by: Keqian Zhu <zhukeqian1@huawei.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Message-id: 20220816094957.31700-1-zhukeqian1@huawei.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
(cherry picked from commit d4424bebceaa8ffbc23060ce45e52a9bb817e3c9)
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
hw/acpi/generic_event_device.c | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/hw/acpi/generic_event_device.c b/hw/acpi/generic_event_device.c
index e28457a7d1..a3d31631fe 100644
--- a/hw/acpi/generic_event_device.c
+++ b/hw/acpi/generic_event_device.c
@@ -267,6 +267,13 @@ static void acpi_ged_unplug_cb(HotplugHandler *hotplug_dev,
}
}
+static void acpi_ged_ospm_status(AcpiDeviceIf *adev, ACPIOSTInfoList ***list)
+{
+ AcpiGedState *s = ACPI_GED(adev);
+
+ acpi_memory_ospm_status(&s->memhp_state, list);
+}
+
static void acpi_ged_send_event(AcpiDeviceIf *adev, AcpiEventStatusBits ev)
{
AcpiGedState *s = ACPI_GED(adev);
@@ -409,6 +416,7 @@ static void acpi_ged_class_init(ObjectClass *class, void *data)
hc->unplug_request = acpi_ged_unplug_request_cb;
hc->unplug = acpi_ged_unplug_cb;
+ adevc->ospm_status = acpi_ged_ospm_status;
adevc->send_event = acpi_ged_send_event;
}
--
2.37.3

View File

@ -0,0 +1,52 @@
From 100f33ff8a1d55986e43b99ba8726abc29ee8d26 Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@redhat.com>
Date: Mon, 5 Dec 2022 15:32:55 -0500
Subject: [PATCH 5/5] hw/display/qxl: Assert memory slot fits in preallocated
MemoryRegion
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 240: hw/display/qxl: Have qxl_log_command Return early if no log_cmd handler
RH-Bugzilla: 2148545
RH-Acked-by: Gerd Hoffmann <kraxel@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Marc-André Lureau <marcandre.lureau@redhat.com>
RH-Commit: [5/5] f809ce48e7989dd6547b7c8bf1a5efc3fdcacbac (jmaloy/jons-qemu-kvm)
BZ: https://bugzilla.redhat.com/show_bug.cgi?id=2148545
CVE: CVE-2022-4144
Upstream: Merged
commit 86fdb0582c653a9824183679403a85f588260d62
Author: Philippe Mathieu-Daudé <philmd@linaro.org>
Date: Mon Nov 28 21:27:41 2022 +0100
hw/display/qxl: Assert memory slot fits in preallocated MemoryRegion
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20221128202741.4945-6-philmd@linaro.org>
(cherry picked from commit 86fdb0582c653a9824183679403a85f588260d62)
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
hw/display/qxl.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/hw/display/qxl.c b/hw/display/qxl.c
index 2a4b2d4158..bcd9e8716a 100644
--- a/hw/display/qxl.c
+++ b/hw/display/qxl.c
@@ -1372,6 +1372,7 @@ static int qxl_add_memslot(PCIQXLDevice *d, uint32_t slot_id, uint64_t delta,
qxl_set_guest_bug(d, "%s: pci_region = %d", __func__, pci_region);
return 1;
}
+ assert(guest_end - pci_start <= memory_region_size(mr));
virt_start = (intptr_t)memory_region_get_ram_ptr(mr);
memslot.slot_id = slot_id;
--
2.37.3

View File

@ -0,0 +1,130 @@
From 4e1bfbe3a0a113fe3cf39336a9d7da4e8c2a21ea Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@redhat.com>
Date: Mon, 5 Dec 2022 15:32:55 -0500
Subject: [PATCH 4/5] hw/display/qxl: Avoid buffer overrun in qxl_phys2virt
(CVE-2022-4144)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 240: hw/display/qxl: Have qxl_log_command Return early if no log_cmd handler
RH-Bugzilla: 2148545
RH-Acked-by: Gerd Hoffmann <kraxel@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Marc-André Lureau <marcandre.lureau@redhat.com>
RH-Commit: [4/5] afe53f8d9b31c6fd8211fe172173151f3255e67c (jmaloy/jons-qemu-kvm)
BZ: https://bugzilla.redhat.com/show_bug.cgi?id=2148545
CVE: CVE-2022-4144
Upstream: Merged
commit 6dbbf055148c6f1b7d8a3251a65bd6f3d1e1f622
Author: Philippe Mathieu-Daudé <philmd@linaro.org>
Date: Mon Nov 28 21:27:40 2022 +0100
hw/display/qxl: Avoid buffer overrun in qxl_phys2virt (CVE-2022-4144)
Have qxl_get_check_slot_offset() return false if the requested
buffer size does not fit within the slot memory region.
Similarly qxl_phys2virt() now returns NULL in such case, and
qxl_dirty_one_surface() aborts.
This avoids buffer overrun in the host pointer returned by
memory_region_get_ram_ptr().
Fixes: CVE-2022-4144 (out-of-bounds read)
Reported-by: Wenxu Yin (@awxylitol)
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1336
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20221128202741.4945-5-philmd@linaro.org>
(cherry picked from commit 6dbbf055148c6f1b7d8a3251a65bd6f3d1e1f622)
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
hw/display/qxl.c | 27 +++++++++++++++++++++++----
hw/display/qxl.h | 2 +-
2 files changed, 24 insertions(+), 5 deletions(-)
diff --git a/hw/display/qxl.c b/hw/display/qxl.c
index aa9065183e..2a4b2d4158 100644
--- a/hw/display/qxl.c
+++ b/hw/display/qxl.c
@@ -1412,11 +1412,13 @@ static void qxl_reset_surfaces(PCIQXLDevice *d)
/* can be also called from spice server thread context */
static bool qxl_get_check_slot_offset(PCIQXLDevice *qxl, QXLPHYSICAL pqxl,
- uint32_t *s, uint64_t *o)
+ uint32_t *s, uint64_t *o,
+ size_t size_requested)
{
uint64_t phys = le64_to_cpu(pqxl);
uint32_t slot = (phys >> (64 - 8)) & 0xff;
uint64_t offset = phys & 0xffffffffffff;
+ uint64_t size_available;
if (slot >= NUM_MEMSLOTS) {
qxl_set_guest_bug(qxl, "slot too large %d >= %d", slot,
@@ -1440,6 +1442,23 @@ static bool qxl_get_check_slot_offset(PCIQXLDevice *qxl, QXLPHYSICAL pqxl,
slot, offset, qxl->guest_slots[slot].size);
return false;
}
+ size_available = memory_region_size(qxl->guest_slots[slot].mr);
+ if (qxl->guest_slots[slot].offset + offset >= size_available) {
+ qxl_set_guest_bug(qxl,
+ "slot %d offset %"PRIu64" > region size %"PRIu64"\n",
+ slot, qxl->guest_slots[slot].offset + offset,
+ size_available);
+ return false;
+ }
+ size_available -= qxl->guest_slots[slot].offset + offset;
+ if (size_requested > size_available) {
+ qxl_set_guest_bug(qxl,
+ "slot %d offset %"PRIu64" size %zu: "
+ "overrun by %"PRIu64" bytes\n",
+ slot, offset, size_requested,
+ size_requested - size_available);
+ return false;
+ }
*s = slot;
*o = offset;
@@ -1459,7 +1478,7 @@ void *qxl_phys2virt(PCIQXLDevice *qxl, QXLPHYSICAL pqxl, int group_id,
offset = le64_to_cpu(pqxl) & 0xffffffffffff;
return (void *)(intptr_t)offset;
case MEMSLOT_GROUP_GUEST:
- if (!qxl_get_check_slot_offset(qxl, pqxl, &slot, &offset)) {
+ if (!qxl_get_check_slot_offset(qxl, pqxl, &slot, &offset, size)) {
return NULL;
}
ptr = memory_region_get_ram_ptr(qxl->guest_slots[slot].mr);
@@ -1925,9 +1944,9 @@ static void qxl_dirty_one_surface(PCIQXLDevice *qxl, QXLPHYSICAL pqxl,
uint32_t slot;
bool rc;
- rc = qxl_get_check_slot_offset(qxl, pqxl, &slot, &offset);
- assert(rc == true);
size = (uint64_t)height * abs(stride);
+ rc = qxl_get_check_slot_offset(qxl, pqxl, &slot, &offset, size);
+ assert(rc == true);
trace_qxl_surfaces_dirty(qxl->id, offset, size);
qxl_set_dirty(qxl->guest_slots[slot].mr,
qxl->guest_slots[slot].offset + offset,
diff --git a/hw/display/qxl.h b/hw/display/qxl.h
index c784315daa..89ca832cf9 100644
--- a/hw/display/qxl.h
+++ b/hw/display/qxl.h
@@ -157,7 +157,7 @@ OBJECT_DECLARE_SIMPLE_TYPE(PCIQXLDevice, PCI_QXL)
*
* Returns a host pointer to a buffer placed at offset @phys within the
* active slot @group_id of the PCI VGA RAM memory region associated with
- * the @qxl device. If the slot is inactive, or the offset is out
+ * the @qxl device. If the slot is inactive, or the offset + size are out
* of the memory region, returns NULL.
*
* Use with care; by the time this function returns, the returned pointer is
--
2.37.3

View File

@ -0,0 +1,70 @@
From 068c531fb968ec04509b85f524d0745e6acf5449 Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@redhat.com>
Date: Mon, 5 Dec 2022 15:32:55 -0500
Subject: [PATCH 2/5] hw/display/qxl: Document qxl_phys2virt()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 240: hw/display/qxl: Have qxl_log_command Return early if no log_cmd handler
RH-Bugzilla: 2148545
RH-Acked-by: Gerd Hoffmann <kraxel@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Marc-André Lureau <marcandre.lureau@redhat.com>
RH-Commit: [2/5] f84c0b379022c527fc2508a242443d86454944c0 (jmaloy/jons-qemu-kvm)
BZ: https://bugzilla.redhat.com/show_bug.cgi?id=2148545
CVE: CVE-2022-4144
Upstream: Merged
commit b1901de83a9456cde26fc755f71ca2b7b3ef50fc
Author: Philippe Mathieu-Daudé <philmd@linaro.org>
Date: Mon Nov 28 21:27:38 2022 +0100
hw/display/qxl: Document qxl_phys2virt()
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20221128202741.4945-3-philmd@linaro.org>
(cherry picked from commit b1901de83a9456cde26fc755f71ca2b7b3ef50fc)
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
hw/display/qxl.h | 19 +++++++++++++++++++
1 file changed, 19 insertions(+)
diff --git a/hw/display/qxl.h b/hw/display/qxl.h
index 30d21f4d0b..c938f88a2f 100644
--- a/hw/display/qxl.h
+++ b/hw/display/qxl.h
@@ -147,6 +147,25 @@ OBJECT_DECLARE_SIMPLE_TYPE(PCIQXLDevice, PCI_QXL)
#define QXL_DEFAULT_REVISION (QXL_REVISION_STABLE_V12 + 1)
/* qxl.c */
+/**
+ * qxl_phys2virt: Get a pointer within a PCI VRAM memory region.
+ *
+ * @qxl: QXL device
+ * @phys: physical offset of buffer within the VRAM
+ * @group_id: memory slot group
+ *
+ * Returns a host pointer to a buffer placed at offset @phys within the
+ * active slot @group_id of the PCI VGA RAM memory region associated with
+ * the @qxl device. If the slot is inactive, or the offset is out
+ * of the memory region, returns NULL.
+ *
+ * Use with care; by the time this function returns, the returned pointer is
+ * not protected by RCU anymore. If the caller is not within an RCU critical
+ * section and does not hold the iothread lock, it must have other means of
+ * protecting the pointer, such as a reference to the region that includes
+ * the incoming ram_addr_t.
+ *
+ */
void *qxl_phys2virt(PCIQXLDevice *qxl, QXLPHYSICAL phys, int group_id);
void qxl_set_guest_bug(PCIQXLDevice *qxl, const char *msg, ...)
GCC_FMT_ATTR(2, 3);
--
2.37.3

View File

@ -0,0 +1,74 @@
From 5ec8d909d40fa04ef2c3572e01509a1866786070 Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@redhat.com>
Date: Mon, 5 Dec 2022 15:32:55 -0500
Subject: [PATCH 1/5] hw/display/qxl: Have qxl_log_command Return early if no
log_cmd handler
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 240: hw/display/qxl: Have qxl_log_command Return early if no log_cmd handler
RH-Bugzilla: 2148545
RH-Acked-by: Gerd Hoffmann <kraxel@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Marc-André Lureau <marcandre.lureau@redhat.com>
RH-Commit: [1/5] 33d94f40c46cccbc32d108d1035365917bf90356 (jmaloy/jons-qemu-kvm)
BZ: https://bugzilla.redhat.com/show_bug.cgi?id=2148545
CVE: CVE-2022-4144
Upstream: Merged
commit 61c34fc194b776ecadc39fb26b061331107e5599
Author: Philippe Mathieu-Daudé <philmd@linaro.org>
Date: Mon Nov 28 21:27:37 2022 +0100
hw/display/qxl: Have qxl_log_command Return early if no log_cmd handler
Only 3 command types are logged: no need to call qxl_phys2virt()
for the other types. Using different cases will help to pass
different structure sizes to qxl_phys2virt() in a pair of commits.
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20221128202741.4945-2-philmd@linaro.org>
(cherry picked from commit 61c34fc194b776ecadc39fb26b061331107e5599)
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
hw/display/qxl-logger.c | 11 +++++++++++
1 file changed, 11 insertions(+)
diff --git a/hw/display/qxl-logger.c b/hw/display/qxl-logger.c
index 68bfa47568..1bcf803db6 100644
--- a/hw/display/qxl-logger.c
+++ b/hw/display/qxl-logger.c
@@ -247,6 +247,16 @@ int qxl_log_command(PCIQXLDevice *qxl, const char *ring, QXLCommandExt *ext)
qxl_name(qxl_type, ext->cmd.type),
compat ? "(compat)" : "");
+ switch (ext->cmd.type) {
+ case QXL_CMD_DRAW:
+ break;
+ case QXL_CMD_SURFACE:
+ break;
+ case QXL_CMD_CURSOR:
+ break;
+ default:
+ goto out;
+ }
data = qxl_phys2virt(qxl, ext->cmd.data, ext->group_id);
if (!data) {
return 1;
@@ -269,6 +279,7 @@ int qxl_log_command(PCIQXLDevice *qxl, const char *ring, QXLCommandExt *ext)
qxl_log_cmd_cursor(qxl, data, ext->group_id);
break;
}
+out:
fprintf(stderr, "\n");
return 0;
}
--
2.37.3

View File

@ -0,0 +1,234 @@
From 0e6bd3911c4971f575aac7e9cd726467b52fe544 Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@redhat.com>
Date: Mon, 5 Dec 2022 15:32:55 -0500
Subject: [PATCH 3/5] hw/display/qxl: Pass requested buffer size to
qxl_phys2virt()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 240: hw/display/qxl: Have qxl_log_command Return early if no log_cmd handler
RH-Bugzilla: 2148545
RH-Acked-by: Gerd Hoffmann <kraxel@redhat.com>
RH-Acked-by: Stefan Hajnoczi <stefanha@redhat.com>
RH-Acked-by: Marc-André Lureau <marcandre.lureau@redhat.com>
RH-Commit: [3/5] 8e362d67fe7fef9eb457cfb15d75b298fed725c3 (jmaloy/jons-qemu-kvm)
BZ: https://bugzilla.redhat.com/show_bug.cgi?id=2148545
CVE: CVE-2022-4144
Upstream: Merged
commit 8efec0ef8bbc1e75a7ebf6e325a35806ece9b39f
Author: Philippe Mathieu-Daudé <philmd@linaro.org>
Date: Mon Nov 28 21:27:39 2022 +0100
hw/display/qxl: Pass requested buffer size to qxl_phys2virt()
Currently qxl_phys2virt() doesn't check for buffer overrun.
In order to do so in the next commit, pass the buffer size
as argument.
For QXLCursor in qxl_render_cursor() -> qxl_cursor() we
verify the size of the chunked data ahead, checking we can
access 'sizeof(QXLCursor) + chunk->data_size' bytes.
Since in the SPICE_CURSOR_TYPE_MONO case the cursor is
assumed to fit in one chunk, no change are required.
In SPICE_CURSOR_TYPE_ALPHA the ahead read is handled in
qxl_unpack_chunks().
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20221128202741.4945-4-philmd@linaro.org>
(cherry picked from commit 8efec0ef8bbc1e75a7ebf6e325a35806ece9b39f)
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
hw/display/qxl-logger.c | 11 ++++++++---
hw/display/qxl-render.c | 20 ++++++++++++++++----
hw/display/qxl.c | 14 +++++++++-----
hw/display/qxl.h | 4 +++-
4 files changed, 36 insertions(+), 13 deletions(-)
diff --git a/hw/display/qxl-logger.c b/hw/display/qxl-logger.c
index 1bcf803db6..35c38f6252 100644
--- a/hw/display/qxl-logger.c
+++ b/hw/display/qxl-logger.c
@@ -106,7 +106,7 @@ static int qxl_log_image(PCIQXLDevice *qxl, QXLPHYSICAL addr, int group_id)
QXLImage *image;
QXLImageDescriptor *desc;
- image = qxl_phys2virt(qxl, addr, group_id);
+ image = qxl_phys2virt(qxl, addr, group_id, sizeof(QXLImage));
if (!image) {
return 1;
}
@@ -214,7 +214,8 @@ int qxl_log_cmd_cursor(PCIQXLDevice *qxl, QXLCursorCmd *cmd, int group_id)
cmd->u.set.position.y,
cmd->u.set.visible ? "yes" : "no",
cmd->u.set.shape);
- cursor = qxl_phys2virt(qxl, cmd->u.set.shape, group_id);
+ cursor = qxl_phys2virt(qxl, cmd->u.set.shape, group_id,
+ sizeof(QXLCursor));
if (!cursor) {
return 1;
}
@@ -236,6 +237,7 @@ int qxl_log_command(PCIQXLDevice *qxl, const char *ring, QXLCommandExt *ext)
{
bool compat = ext->flags & QXL_COMMAND_FLAG_COMPAT;
void *data;
+ size_t datasz;
int ret;
if (!qxl->cmdlog) {
@@ -249,15 +251,18 @@ int qxl_log_command(PCIQXLDevice *qxl, const char *ring, QXLCommandExt *ext)
switch (ext->cmd.type) {
case QXL_CMD_DRAW:
+ datasz = compat ? sizeof(QXLCompatDrawable) : sizeof(QXLDrawable);
break;
case QXL_CMD_SURFACE:
+ datasz = sizeof(QXLSurfaceCmd);
break;
case QXL_CMD_CURSOR:
+ datasz = sizeof(QXLCursorCmd);
break;
default:
goto out;
}
- data = qxl_phys2virt(qxl, ext->cmd.data, ext->group_id);
+ data = qxl_phys2virt(qxl, ext->cmd.data, ext->group_id, datasz);
if (!data) {
return 1;
}
diff --git a/hw/display/qxl-render.c b/hw/display/qxl-render.c
index ca217004bf..fcfd40c3ac 100644
--- a/hw/display/qxl-render.c
+++ b/hw/display/qxl-render.c
@@ -107,7 +107,9 @@ static void qxl_render_update_area_unlocked(PCIQXLDevice *qxl)
qxl->guest_primary.resized = 0;
qxl->guest_primary.data = qxl_phys2virt(qxl,
qxl->guest_primary.surface.mem,
- MEMSLOT_GROUP_GUEST);
+ MEMSLOT_GROUP_GUEST,
+ qxl->guest_primary.abs_stride
+ * height);
if (!qxl->guest_primary.data) {
goto end;
}
@@ -228,7 +230,8 @@ static void qxl_unpack_chunks(void *dest, size_t size, PCIQXLDevice *qxl,
if (offset == size) {
return;
}
- chunk = qxl_phys2virt(qxl, chunk->next_chunk, group_id);
+ chunk = qxl_phys2virt(qxl, chunk->next_chunk, group_id,
+ sizeof(QXLDataChunk) + chunk->data_size);
if (!chunk) {
return;
}
@@ -295,7 +298,8 @@ fail:
/* called from spice server thread context only */
int qxl_render_cursor(PCIQXLDevice *qxl, QXLCommandExt *ext)
{
- QXLCursorCmd *cmd = qxl_phys2virt(qxl, ext->cmd.data, ext->group_id);
+ QXLCursorCmd *cmd = qxl_phys2virt(qxl, ext->cmd.data, ext->group_id,
+ sizeof(QXLCursorCmd));
QXLCursor *cursor;
QEMUCursor *c;
@@ -314,7 +318,15 @@ int qxl_render_cursor(PCIQXLDevice *qxl, QXLCommandExt *ext)
}
switch (cmd->type) {
case QXL_CURSOR_SET:
- cursor = qxl_phys2virt(qxl, cmd->u.set.shape, ext->group_id);
+ /* First read the QXLCursor to get QXLDataChunk::data_size ... */
+ cursor = qxl_phys2virt(qxl, cmd->u.set.shape, ext->group_id,
+ sizeof(QXLCursor));
+ if (!cursor) {
+ return 1;
+ }
+ /* Then read including the chunked data following QXLCursor. */
+ cursor = qxl_phys2virt(qxl, cmd->u.set.shape, ext->group_id,
+ sizeof(QXLCursor) + cursor->chunk.data_size);
if (!cursor) {
return 1;
}
diff --git a/hw/display/qxl.c b/hw/display/qxl.c
index 29c80b4289..aa9065183e 100644
--- a/hw/display/qxl.c
+++ b/hw/display/qxl.c
@@ -274,7 +274,8 @@ static void qxl_spice_monitors_config_async(PCIQXLDevice *qxl, int replay)
QXL_IO_MONITORS_CONFIG_ASYNC));
}
- cfg = qxl_phys2virt(qxl, qxl->guest_monitors_config, MEMSLOT_GROUP_GUEST);
+ cfg = qxl_phys2virt(qxl, qxl->guest_monitors_config, MEMSLOT_GROUP_GUEST,
+ sizeof(QXLMonitorsConfig));
if (cfg != NULL && cfg->count == 1) {
qxl->guest_primary.resized = 1;
qxl->guest_head0_width = cfg->heads[0].width;
@@ -459,7 +460,8 @@ static int qxl_track_command(PCIQXLDevice *qxl, struct QXLCommandExt *ext)
switch (le32_to_cpu(ext->cmd.type)) {
case QXL_CMD_SURFACE:
{
- QXLSurfaceCmd *cmd = qxl_phys2virt(qxl, ext->cmd.data, ext->group_id);
+ QXLSurfaceCmd *cmd = qxl_phys2virt(qxl, ext->cmd.data, ext->group_id,
+ sizeof(QXLSurfaceCmd));
if (!cmd) {
return 1;
@@ -494,7 +496,8 @@ static int qxl_track_command(PCIQXLDevice *qxl, struct QXLCommandExt *ext)
}
case QXL_CMD_CURSOR:
{
- QXLCursorCmd *cmd = qxl_phys2virt(qxl, ext->cmd.data, ext->group_id);
+ QXLCursorCmd *cmd = qxl_phys2virt(qxl, ext->cmd.data, ext->group_id,
+ sizeof(QXLCursorCmd));
if (!cmd) {
return 1;
@@ -1444,7 +1447,8 @@ static bool qxl_get_check_slot_offset(PCIQXLDevice *qxl, QXLPHYSICAL pqxl,
}
/* can be also called from spice server thread context */
-void *qxl_phys2virt(PCIQXLDevice *qxl, QXLPHYSICAL pqxl, int group_id)
+void *qxl_phys2virt(PCIQXLDevice *qxl, QXLPHYSICAL pqxl, int group_id,
+ size_t size)
{
uint64_t offset;
uint32_t slot;
@@ -1952,7 +1956,7 @@ static void qxl_dirty_surfaces(PCIQXLDevice *qxl)
}
cmd = qxl_phys2virt(qxl, qxl->guest_surfaces.cmds[i],
- MEMSLOT_GROUP_GUEST);
+ MEMSLOT_GROUP_GUEST, sizeof(QXLSurfaceCmd));
assert(cmd);
assert(cmd->type == QXL_SURFACE_CMD_CREATE);
qxl_dirty_one_surface(qxl, cmd->u.surface_create.data,
diff --git a/hw/display/qxl.h b/hw/display/qxl.h
index c938f88a2f..c784315daa 100644
--- a/hw/display/qxl.h
+++ b/hw/display/qxl.h
@@ -153,6 +153,7 @@ OBJECT_DECLARE_SIMPLE_TYPE(PCIQXLDevice, PCI_QXL)
* @qxl: QXL device
* @phys: physical offset of buffer within the VRAM
* @group_id: memory slot group
+ * @size: size of the buffer
*
* Returns a host pointer to a buffer placed at offset @phys within the
* active slot @group_id of the PCI VGA RAM memory region associated with
@@ -166,7 +167,8 @@ OBJECT_DECLARE_SIMPLE_TYPE(PCIQXLDevice, PCI_QXL)
* the incoming ram_addr_t.
*
*/
-void *qxl_phys2virt(PCIQXLDevice *qxl, QXLPHYSICAL phys, int group_id);
+void *qxl_phys2virt(PCIQXLDevice *qxl, QXLPHYSICAL phys, int group_id,
+ size_t size);
void qxl_set_guest_bug(PCIQXLDevice *qxl, const char *msg, ...)
GCC_FMT_ATTR(2, 3);
--
2.37.3

View File

@ -1,16 +1,15 @@
From 1bd939d374ec2e994ff47c84e16fa3bc1323a0fd Mon Sep 17 00:00:00 2001 From f96220d64a31a4a52b2d132a503048579946f982 Mon Sep 17 00:00:00 2001
From: Vitaly Kuznetsov <vkuznets@redhat.com> From: Vitaly Kuznetsov <vkuznets@redhat.com>
Date: Thu, 18 Aug 2022 17:01:13 +0200 Date: Thu, 18 Aug 2022 17:01:13 +0200
Subject: [PATCH 2/2] i386: do kvm_put_msr_feature_control() first thing when Subject: [PATCH 3/3] i386: do kvm_put_msr_feature_control() first thing when
vCPU is reset vCPU is reset
RH-Author: Vitaly Kuznetsov <vkuznets@redhat.com> RH-Author: Miroslav Rezanina <mrezanin@redhat.com>
RH-MergeRequest: 216: i386: fix 'system_reset' when the VM is in VMX root operation RH-MergeRequest: 219: Synchronize qemu-6.2.0-20.el8.1 build from RHEL 8.7 to RHEL 8.8
RH-Bugzilla: 2116743 RH-Bugzilla: 2125271
RH-Acked-by: Emanuele Giuseppe Esposito <eesposit@redhat.com> RH-Acked-by: Vitaly Kuznetsov <vkuznets@redhat.com>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com> RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com> RH-Commit: [2/2] 08e1e67db96801e4a35aa6b60a93b2c2f1641220
RH-Commit: [2/2] f838a57f74487eb394794de00006d5d2b9e84344
kvm_put_sregs2() fails to reset 'locked' CR4/CR0 bits upon vCPU reset when kvm_put_sregs2() fails to reset 'locked' CR4/CR0 bits upon vCPU reset when
it is in VMX root operation. Do kvm_put_msr_feature_control() before it is in VMX root operation. Do kvm_put_msr_feature_control() before
@ -64,5 +63,5 @@ index 81d729dc40..a06221d3e5 100644
if (level == KVM_PUT_FULL_STATE) { if (level == KVM_PUT_FULL_STATE) {
-- --
2.31.1 2.35.3

View File

@ -1,15 +1,14 @@
From 4ad00e318f8afbee0e455cfbb6bc693c808d87f3 Mon Sep 17 00:00:00 2001 From 46e54544c3480658111d6f111d6c265dcea2e19b Mon Sep 17 00:00:00 2001
From: Vitaly Kuznetsov <vkuznets@redhat.com> From: Vitaly Kuznetsov <vkuznets@redhat.com>
Date: Thu, 18 Aug 2022 17:01:12 +0200 Date: Thu, 18 Aug 2022 17:01:12 +0200
Subject: [PATCH 1/2] i386: reset KVM nested state upon CPU reset Subject: [PATCH 2/3] i386: reset KVM nested state upon CPU reset
RH-Author: Vitaly Kuznetsov <vkuznets@redhat.com> RH-Author: Miroslav Rezanina <mrezanin@redhat.com>
RH-MergeRequest: 216: i386: fix 'system_reset' when the VM is in VMX root operation RH-MergeRequest: 219: Synchronize qemu-6.2.0-20.el8.1 build from RHEL 8.7 to RHEL 8.8
RH-Bugzilla: 2116743 RH-Bugzilla: 2125271
RH-Acked-by: Emanuele Giuseppe Esposito <eesposit@redhat.com> RH-Acked-by: Vitaly Kuznetsov <vkuznets@redhat.com>
RH-Acked-by: Miroslav Rezanina <mrezanin@redhat.com> RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Acked-by: Peter Xu <peterx@redhat.com> RH-Commit: [1/2] de4db7bceb6baaf69aec8b0ae9aa8887aa869e15
RH-Commit: [1/2] 20d2dabeda74b8cd5135228980a2414e66dc64f3
Make sure env->nested_state is cleaned up when a vCPU is reset, it may Make sure env->nested_state is cleaned up when a vCPU is reset, it may
be stale after an incoming migration, kvm_arch_put_registers() may be stale after an incoming migration, kvm_arch_put_registers() may
@ -91,5 +90,5 @@ index bd439e56ad..81d729dc40 100644
} }
-- --
2.31.1 2.35.3

View File

@ -0,0 +1,43 @@
From 3fceb3b60a60c5008eecf99e45e269b757042b5a Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Mon, 17 Oct 2022 08:38:20 +0000
Subject: [PATCH 39/42] include/elf.h: add s390x note types
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [39/41] ebf0873744905abbe9cfc423a56c6d1b4f2ae936
Adding two s390x note types
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20221017083822.43118-9-frankja@linux.ibm.com>
(cherry picked from commit 5433669c7a1884cc0394c360148965edf7519884)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
include/elf.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/include/elf.h b/include/elf.h
index 811bf4a1cb..4edab8e5a2 100644
--- a/include/elf.h
+++ b/include/elf.h
@@ -1647,6 +1647,8 @@ typedef struct elf64_shdr {
#define NT_TASKSTRUCT 4
#define NT_AUXV 6
#define NT_PRXFPREG 0x46e62b7f /* copied from gdb5.1/include/elf/common.h */
+#define NT_S390_PV_CPU_DATA 0x30e /* s390 protvirt cpu dump data */
+#define NT_S390_RI_CB 0x30d /* s390 runtime instrumentation */
#define NT_S390_GS_CB 0x30b /* s390 guarded storage registers */
#define NT_S390_VXRS_HIGH 0x30a /* s390 vector registers 16-31 */
#define NT_S390_VXRS_LOW 0x309 /* s390 vector registers 0-15 (lower half) */
--
2.37.3

View File

@ -0,0 +1,111 @@
From 21b19213328826327eba18199b790425659af7d8 Mon Sep 17 00:00:00 2001
From: Thomas Huth <thuth@redhat.com>
Date: Fri, 29 Jul 2022 16:55:34 +0200
Subject: [PATCH 1/3] redhat: Update linux-headers/linux/kvm.h to v5.18-rc6
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <None>
RH-MergeRequest: 220: s390x: Fix skey test in kvm_unit_test
RH-Bugzilla: 2124757
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: David Hildenbrand <david@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Commit: [1/2] e514a00305cb0caab9d3acc0efb325853daa6d51
Upstream Status: RHEL-only
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2124757
Based on upstream commit e4082063e47e9731dbeb1c26174c17f6038f577f
("linux-headers: Update to v5.18-rc6"), but this is focusing on
the file linux-headers/linux/kvm.h only (since the other changes
related to the VFIO renaming might break some stuff).
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit 71516db15469a02600932a5c1f0d4a9626a91193)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
linux-headers/linux/kvm.h | 27 +++++++++++++++++++++------
1 file changed, 21 insertions(+), 6 deletions(-)
diff --git a/linux-headers/linux/kvm.h b/linux-headers/linux/kvm.h
index d232feaae9..0d05d02ee4 100644
--- a/linux-headers/linux/kvm.h
+++ b/linux-headers/linux/kvm.h
@@ -445,7 +445,11 @@ struct kvm_run {
#define KVM_SYSTEM_EVENT_RESET 2
#define KVM_SYSTEM_EVENT_CRASH 3
__u32 type;
- __u64 flags;
+ __u32 ndata;
+ union {
+ __u64 flags;
+ __u64 data[16];
+ };
} system_event;
/* KVM_EXIT_S390_STSI */
struct {
@@ -562,9 +566,12 @@ struct kvm_s390_mem_op {
__u32 op; /* type of operation */
__u64 buf; /* buffer in userspace */
union {
- __u8 ar; /* the access register number */
+ struct {
+ __u8 ar; /* the access register number */
+ __u8 key; /* access key, ignored if flag unset */
+ };
__u32 sida_offset; /* offset into the sida */
- __u8 reserved[32]; /* should be set to 0 */
+ __u8 reserved[32]; /* ignored */
};
};
/* types for kvm_s390_mem_op->op */
@@ -572,9 +579,12 @@ struct kvm_s390_mem_op {
#define KVM_S390_MEMOP_LOGICAL_WRITE 1
#define KVM_S390_MEMOP_SIDA_READ 2
#define KVM_S390_MEMOP_SIDA_WRITE 3
+#define KVM_S390_MEMOP_ABSOLUTE_READ 4
+#define KVM_S390_MEMOP_ABSOLUTE_WRITE 5
/* flags for kvm_s390_mem_op->flags */
#define KVM_S390_MEMOP_F_CHECK_ONLY (1ULL << 0)
#define KVM_S390_MEMOP_F_INJECT_EXCEPTION (1ULL << 1)
+#define KVM_S390_MEMOP_F_SKEY_PROTECTION (1ULL << 2)
/* for KVM_INTERRUPT */
struct kvm_interrupt {
@@ -1134,6 +1144,12 @@ struct kvm_ppc_resize_hpt {
#define KVM_CAP_VM_GPA_BITS 207
#define KVM_CAP_XSAVE2 208
#define KVM_CAP_SYS_ATTRIBUTES 209
+#define KVM_CAP_PPC_AIL_MODE_3 210
+#define KVM_CAP_S390_MEM_OP_EXTENSION 211
+#define KVM_CAP_PMU_CAPABILITY 212
+#define KVM_CAP_DISABLE_QUIRKS2 213
+/* #define KVM_CAP_VM_TSC_CONTROL 214 */
+#define KVM_CAP_SYSTEM_EVENT_DATA 215
#ifdef KVM_CAP_IRQ_ROUTING
@@ -1624,9 +1640,6 @@ struct kvm_enc_region {
#define KVM_S390_NORMAL_RESET _IO(KVMIO, 0xc3)
#define KVM_S390_CLEAR_RESET _IO(KVMIO, 0xc4)
-/* Available with KVM_CAP_XSAVE2 */
-#define KVM_GET_XSAVE2 _IOR(KVMIO, 0xcf, struct kvm_xsave)
-
struct kvm_s390_pv_sec_parm {
__u64 origin;
__u64 length;
@@ -1973,6 +1986,8 @@ struct kvm_dirty_gfn {
#define KVM_BUS_LOCK_DETECTION_OFF (1 << 0)
#define KVM_BUS_LOCK_DETECTION_EXIT (1 << 1)
+#define KVM_PMU_CAP_DISABLE (1 << 0)
+
/**
* struct kvm_stats_header - Header of per vm/vcpu binary statistics data.
* @flags: Some extra information for header, always 0 for now.
--
2.35.3

View File

@ -0,0 +1,124 @@
From 95c229506a6e7261fce184488e880a94f9ba0789 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Mon, 17 Oct 2022 08:38:21 +0000
Subject: [PATCH 40/42] s390x: Add KVM PV dump interface
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [40/41] 5df512a63b2ed17991489565b70f89f4efc0b639
Let's add a few bits of code which hide the new KVM PV dump API from
us via new functions.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
[ Marc-André: fix up for compilation issue ]
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20221017083822.43118-10-frankja@linux.ibm.com>
(cherry picked from commit 753ca06f4706cd6e57750a606afb08c5c5299643)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/s390x/pv.c | 51 +++++++++++++++++++++++++++++++++++++++++++
include/hw/s390x/pv.h | 9 ++++++++
2 files changed, 60 insertions(+)
diff --git a/hw/s390x/pv.c b/hw/s390x/pv.c
index 4c012f2eeb..728ba24547 100644
--- a/hw/s390x/pv.c
+++ b/hw/s390x/pv.c
@@ -175,6 +175,57 @@ bool kvm_s390_pv_info_basic_valid(void)
return info_valid;
}
+static int s390_pv_dump_cmd(uint64_t subcmd, uint64_t uaddr, uint64_t gaddr,
+ uint64_t len)
+{
+ struct kvm_s390_pv_dmp dmp = {
+ .subcmd = subcmd,
+ .buff_addr = uaddr,
+ .buff_len = len,
+ .gaddr = gaddr,
+ };
+ int ret;
+
+ ret = s390_pv_cmd(KVM_PV_DUMP, (void *)&dmp);
+ if (ret) {
+ error_report("KVM DUMP command %ld failed", subcmd);
+ }
+ return ret;
+}
+
+int kvm_s390_dump_cpu(S390CPU *cpu, void *buff)
+{
+ struct kvm_s390_pv_dmp dmp = {
+ .subcmd = KVM_PV_DUMP_CPU,
+ .buff_addr = (uint64_t)buff,
+ .gaddr = 0,
+ .buff_len = info_dump.dump_cpu_buffer_len,
+ };
+ struct kvm_pv_cmd pv = {
+ .cmd = KVM_PV_DUMP,
+ .data = (uint64_t)&dmp,
+ };
+
+ return kvm_vcpu_ioctl(CPU(cpu), KVM_S390_PV_CPU_COMMAND, &pv);
+}
+
+int kvm_s390_dump_init(void)
+{
+ return s390_pv_dump_cmd(KVM_PV_DUMP_INIT, 0, 0, 0);
+}
+
+int kvm_s390_dump_mem_state(uint64_t gaddr, size_t len, void *dest)
+{
+ return s390_pv_dump_cmd(KVM_PV_DUMP_CONFIG_STOR_STATE, (uint64_t)dest,
+ gaddr, len);
+}
+
+int kvm_s390_dump_completion_data(void *buff)
+{
+ return s390_pv_dump_cmd(KVM_PV_DUMP_COMPLETE, (uint64_t)buff, 0,
+ info_dump.dump_config_finalize_len);
+}
+
#define TYPE_S390_PV_GUEST "s390-pv-guest"
OBJECT_DECLARE_SIMPLE_TYPE(S390PVGuest, S390_PV_GUEST)
diff --git a/include/hw/s390x/pv.h b/include/hw/s390x/pv.h
index e5ea0eca16..9360aa1091 100644
--- a/include/hw/s390x/pv.h
+++ b/include/hw/s390x/pv.h
@@ -51,6 +51,10 @@ uint64_t kvm_s390_pv_dmp_get_size_cpu(void);
uint64_t kvm_s390_pv_dmp_get_size_mem_state(void);
uint64_t kvm_s390_pv_dmp_get_size_completion_data(void);
bool kvm_s390_pv_info_basic_valid(void);
+int kvm_s390_dump_init(void);
+int kvm_s390_dump_cpu(S390CPU *cpu, void *buff);
+int kvm_s390_dump_mem_state(uint64_t addr, size_t len, void *dest);
+int kvm_s390_dump_completion_data(void *buff);
#else /* CONFIG_KVM */
static inline bool s390_is_pv(void) { return false; }
static inline int s390_pv_query_info(void) { return 0; }
@@ -66,6 +70,11 @@ static inline uint64_t kvm_s390_pv_dmp_get_size_cpu(void) { return 0; }
static inline uint64_t kvm_s390_pv_dmp_get_size_mem_state(void) { return 0; }
static inline uint64_t kvm_s390_pv_dmp_get_size_completion_data(void) { return 0; }
static inline bool kvm_s390_pv_info_basic_valid(void) { return false; }
+static inline int kvm_s390_dump_init(void) { return 0; }
+static inline int kvm_s390_dump_cpu(S390CPU *cpu, void *buff) { return 0; }
+static inline int kvm_s390_dump_mem_state(uint64_t addr, size_t len,
+ void *dest) { return 0; }
+static inline int kvm_s390_dump_completion_data(void *buff) { return 0; }
#endif /* CONFIG_KVM */
int s390_pv_kvm_init(ConfidentialGuestSupport *cgs, Error **errp);
--
2.37.3

View File

@ -0,0 +1,113 @@
From 7634eed5aea61dc94f9a828c62ef3da9aeaa62ae Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Mon, 17 Oct 2022 08:38:18 +0000
Subject: [PATCH 37/42] s390x: Add protected dump cap
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [37/41] 52e1e7bf1a00ce3a220d3db2f733a65548bfec6d
Add a protected dump capability for later feature checking.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Message-Id: <20221017083822.43118-7-frankja@linux.ibm.com>
[ Marc-André - Add missing stubs when !kvm ]
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
(cherry picked from commit ad3b2e693daac6ed92db7361236028851d37c77c)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
target/s390x/kvm/kvm.c | 7 +++++++
target/s390x/kvm/kvm_s390x.h | 1 +
target/s390x/kvm/meson.build | 2 ++
target/s390x/kvm/stubs.c | 12 ++++++++++++
4 files changed, 22 insertions(+)
create mode 100644 target/s390x/kvm/stubs.c
diff --git a/target/s390x/kvm/kvm.c b/target/s390x/kvm/kvm.c
index 30712487d4..d36b44f32a 100644
--- a/target/s390x/kvm/kvm.c
+++ b/target/s390x/kvm/kvm.c
@@ -159,6 +159,7 @@ static int cap_hpage_1m;
static int cap_vcpu_resets;
static int cap_protected;
static int cap_zpci_op;
+static int cap_protected_dump;
static bool mem_op_storage_key_support;
@@ -365,6 +366,7 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
cap_vcpu_resets = kvm_check_extension(s, KVM_CAP_S390_VCPU_RESETS);
cap_protected = kvm_check_extension(s, KVM_CAP_S390_PROTECTED);
cap_zpci_op = kvm_check_extension(s, KVM_CAP_S390_ZPCI_OP);
+ cap_protected_dump = kvm_check_extension(s, KVM_CAP_S390_PROTECTED_DUMP);
kvm_vm_enable_cap(s, KVM_CAP_S390_USER_SIGP, 0);
kvm_vm_enable_cap(s, KVM_CAP_S390_VECTOR_REGISTERS, 0);
@@ -2042,6 +2044,11 @@ int kvm_s390_assign_subch_ioeventfd(EventNotifier *notifier, uint32_t sch,
return kvm_vm_ioctl(kvm_state, KVM_IOEVENTFD, &kick);
}
+int kvm_s390_get_protected_dump(void)
+{
+ return cap_protected_dump;
+}
+
int kvm_s390_get_ri(void)
{
return cap_ri;
diff --git a/target/s390x/kvm/kvm_s390x.h b/target/s390x/kvm/kvm_s390x.h
index aaae8570de..f9785564d0 100644
--- a/target/s390x/kvm/kvm_s390x.h
+++ b/target/s390x/kvm/kvm_s390x.h
@@ -26,6 +26,7 @@ int kvm_s390_set_cpu_state(S390CPU *cpu, uint8_t cpu_state);
void kvm_s390_vcpu_interrupt_pre_save(S390CPU *cpu);
int kvm_s390_vcpu_interrupt_post_load(S390CPU *cpu);
int kvm_s390_get_hpage_1m(void);
+int kvm_s390_get_protected_dump(void);
int kvm_s390_get_ri(void);
int kvm_s390_get_zpci_op(void);
int kvm_s390_get_clock(uint8_t *tod_high, uint64_t *tod_clock);
diff --git a/target/s390x/kvm/meson.build b/target/s390x/kvm/meson.build
index d1356356b1..aef52b6686 100644
--- a/target/s390x/kvm/meson.build
+++ b/target/s390x/kvm/meson.build
@@ -1,6 +1,8 @@
s390x_ss.add(when: 'CONFIG_KVM', if_true: files(
'kvm.c'
+), if_false: files(
+ 'stubs.c'
))
# Newer kernels on s390 check for an S390_PGSTE program header and
diff --git a/target/s390x/kvm/stubs.c b/target/s390x/kvm/stubs.c
new file mode 100644
index 0000000000..5fd63b9a7e
--- /dev/null
+++ b/target/s390x/kvm/stubs.c
@@ -0,0 +1,12 @@
+/*
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#include "qemu/osdep.h"
+
+#include "kvm_s390x.h"
+
+int kvm_s390_get_protected_dump(void)
+{
+ return false;
+}
--
2.37.3

View File

@ -0,0 +1,174 @@
From 760236b3633a8f532631256a899cab969e772196 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Mon, 17 Oct 2022 08:38:19 +0000
Subject: [PATCH 38/42] s390x: Introduce PV query interface
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [38/41] 3090615d81ec6b9e4c306f7fc3709e1935ff5a79
Introduce an interface over which we can get information about UV data.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Reviewed-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Acked-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20221017083822.43118-8-frankja@linux.ibm.com>
(cherry picked from commit 03d83ecfae46bf5e0074cb5808043b30df34064b)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/s390x/pv.c | 61 ++++++++++++++++++++++++++++++++++++++
hw/s390x/s390-virtio-ccw.c | 6 ++++
include/hw/s390x/pv.h | 10 +++++++
3 files changed, 77 insertions(+)
diff --git a/hw/s390x/pv.c b/hw/s390x/pv.c
index 401b63d6cb..4c012f2eeb 100644
--- a/hw/s390x/pv.c
+++ b/hw/s390x/pv.c
@@ -20,6 +20,11 @@
#include "exec/confidential-guest-support.h"
#include "hw/s390x/ipl.h"
#include "hw/s390x/pv.h"
+#include "target/s390x/kvm/kvm_s390x.h"
+
+static bool info_valid;
+static struct kvm_s390_pv_info_vm info_vm;
+static struct kvm_s390_pv_info_dump info_dump;
static int __s390_pv_cmd(uint32_t cmd, const char *cmdname, void *data)
{
@@ -56,6 +61,42 @@ static int __s390_pv_cmd(uint32_t cmd, const char *cmdname, void *data)
} \
}
+int s390_pv_query_info(void)
+{
+ struct kvm_s390_pv_info info = {
+ .header.id = KVM_PV_INFO_VM,
+ .header.len_max = sizeof(info.header) + sizeof(info.vm),
+ };
+ int rc;
+
+ /* Info API's first user is dump so they are bundled */
+ if (!kvm_s390_get_protected_dump()) {
+ return 0;
+ }
+
+ rc = s390_pv_cmd(KVM_PV_INFO, &info);
+ if (rc) {
+ error_report("KVM PV INFO cmd %x failed: %s",
+ info.header.id, strerror(-rc));
+ return rc;
+ }
+ memcpy(&info_vm, &info.vm, sizeof(info.vm));
+
+ info.header.id = KVM_PV_INFO_DUMP;
+ info.header.len_max = sizeof(info.header) + sizeof(info.dump);
+ rc = s390_pv_cmd(KVM_PV_INFO, &info);
+ if (rc) {
+ error_report("KVM PV INFO cmd %x failed: %s",
+ info.header.id, strerror(-rc));
+ return rc;
+ }
+
+ memcpy(&info_dump, &info.dump, sizeof(info.dump));
+ info_valid = true;
+
+ return rc;
+}
+
int s390_pv_vm_enable(void)
{
return s390_pv_cmd(KVM_PV_ENABLE, NULL);
@@ -114,6 +155,26 @@ void s390_pv_inject_reset_error(CPUState *cs)
env->regs[r1 + 1] = DIAG_308_RC_INVAL_FOR_PV;
}
+uint64_t kvm_s390_pv_dmp_get_size_cpu(void)
+{
+ return info_dump.dump_cpu_buffer_len;
+}
+
+uint64_t kvm_s390_pv_dmp_get_size_completion_data(void)
+{
+ return info_dump.dump_config_finalize_len;
+}
+
+uint64_t kvm_s390_pv_dmp_get_size_mem_state(void)
+{
+ return info_dump.dump_config_mem_buffer_per_1m;
+}
+
+bool kvm_s390_pv_info_basic_valid(void)
+{
+ return info_valid;
+}
+
#define TYPE_S390_PV_GUEST "s390-pv-guest"
OBJECT_DECLARE_SIMPLE_TYPE(S390PVGuest, S390_PV_GUEST)
diff --git a/hw/s390x/s390-virtio-ccw.c b/hw/s390x/s390-virtio-ccw.c
index bd80e72cf8..a9617ab79f 100644
--- a/hw/s390x/s390-virtio-ccw.c
+++ b/hw/s390x/s390-virtio-ccw.c
@@ -365,6 +365,12 @@ static int s390_machine_protect(S390CcwMachineState *ms)
ms->pv = true;
+ /* Will return 0 if API is not available since it's not vital */
+ rc = s390_pv_query_info();
+ if (rc) {
+ goto out_err;
+ }
+
/* Set SE header and unpack */
rc = s390_ipl_prepare_pv_header();
if (rc) {
diff --git a/include/hw/s390x/pv.h b/include/hw/s390x/pv.h
index 1f1f545bfc..e5ea0eca16 100644
--- a/include/hw/s390x/pv.h
+++ b/include/hw/s390x/pv.h
@@ -38,6 +38,7 @@ static inline bool s390_is_pv(void)
return ccw->pv;
}
+int s390_pv_query_info(void);
int s390_pv_vm_enable(void);
void s390_pv_vm_disable(void);
int s390_pv_set_sec_parms(uint64_t origin, uint64_t length);
@@ -46,8 +47,13 @@ void s390_pv_prep_reset(void);
int s390_pv_verify(void);
void s390_pv_unshare(void);
void s390_pv_inject_reset_error(CPUState *cs);
+uint64_t kvm_s390_pv_dmp_get_size_cpu(void);
+uint64_t kvm_s390_pv_dmp_get_size_mem_state(void);
+uint64_t kvm_s390_pv_dmp_get_size_completion_data(void);
+bool kvm_s390_pv_info_basic_valid(void);
#else /* CONFIG_KVM */
static inline bool s390_is_pv(void) { return false; }
+static inline int s390_pv_query_info(void) { return 0; }
static inline int s390_pv_vm_enable(void) { return 0; }
static inline void s390_pv_vm_disable(void) {}
static inline int s390_pv_set_sec_parms(uint64_t origin, uint64_t length) { return 0; }
@@ -56,6 +62,10 @@ static inline void s390_pv_prep_reset(void) {}
static inline int s390_pv_verify(void) { return 0; }
static inline void s390_pv_unshare(void) {}
static inline void s390_pv_inject_reset_error(CPUState *cs) {};
+static inline uint64_t kvm_s390_pv_dmp_get_size_cpu(void) { return 0; }
+static inline uint64_t kvm_s390_pv_dmp_get_size_mem_state(void) { return 0; }
+static inline uint64_t kvm_s390_pv_dmp_get_size_completion_data(void) { return 0; }
+static inline bool kvm_s390_pv_info_basic_valid(void) { return false; }
#endif /* CONFIG_KVM */
int s390_pv_kvm_init(ConfidentialGuestSupport *cgs, Error **errp);
--
2.37.3

View File

@ -0,0 +1,209 @@
From 429c4cc750affe82b89867668ff2515a8a66732e Mon Sep 17 00:00:00 2001
From: Thomas Huth <thuth@redhat.com>
Date: Fri, 18 Nov 2022 15:23:19 +0100
Subject: [PATCH 2/3] s390x: Register TYPE_S390_CCW_MACHINE properties as class
properties
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Thomas Huth <thuth@redhat.com>
RH-MergeRequest: 233: s390x: Document the "loadparm" machine property
RH-Bugzilla: 2128225
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Cédric Le Goater <clg@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [2/2] 28a0086cb0e8be2535deafdd9115cadd7ff033f3
Currently, when running 'qemu-system-s390x -M s390-ccw-virtio,help'
the s390x-specific properties are not listed anymore. This happens
because since commit d8fb7d0969 ("vl: switch -M parsing to keyval")
the properties have to be defined at the class level and not at the
instance level anymore. Fix it on s390x now, too, by moving the
registration of the properties to the class level"
Fixes: d8fb7d0969 ("vl: switch -M parsing to keyval")
Signed-off-by: Pierre Morel <pmorel@linux.ibm.com>
Message-Id: <20221103170150.20789-2-pmorel@linux.ibm.com>
[thuth: Add patch description]
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit 1fd396e32288bbf536483c74b68cb3ee86005a9f)
Conflicts:
hw/s390x/s390-virtio-ccw.c
(dropped the "zpcii-disable" property code - it's not used in downstream)
Signed-off-by: Thomas Huth <thuth@redhat.com>
---
hw/s390x/s390-virtio-ccw.c | 117 +++++++++++++++++++++----------------
1 file changed, 67 insertions(+), 50 deletions(-)
diff --git a/hw/s390x/s390-virtio-ccw.c b/hw/s390x/s390-virtio-ccw.c
index a9617ab79f..4a7cd21cac 100644
--- a/hw/s390x/s390-virtio-ccw.c
+++ b/hw/s390x/s390-virtio-ccw.c
@@ -42,6 +42,7 @@
#include "sysemu/sysemu.h"
#include "hw/s390x/pv.h"
#include "migration/blocker.h"
+#include "qapi/visitor.h"
static Error *pv_mig_blocker;
@@ -588,38 +589,6 @@ static ram_addr_t s390_fixup_ram_size(ram_addr_t sz)
return newsz;
}
-static void ccw_machine_class_init(ObjectClass *oc, void *data)
-{
- MachineClass *mc = MACHINE_CLASS(oc);
- NMIClass *nc = NMI_CLASS(oc);
- HotplugHandlerClass *hc = HOTPLUG_HANDLER_CLASS(oc);
- S390CcwMachineClass *s390mc = S390_CCW_MACHINE_CLASS(mc);
-
- s390mc->ri_allowed = true;
- s390mc->cpu_model_allowed = true;
- s390mc->css_migration_enabled = true;
- s390mc->hpage_1m_allowed = true;
- mc->init = ccw_init;
- mc->reset = s390_machine_reset;
- mc->block_default_type = IF_VIRTIO;
- mc->no_cdrom = 1;
- mc->no_floppy = 1;
- mc->no_parallel = 1;
- mc->no_sdcard = 1;
- mc->max_cpus = S390_MAX_CPUS;
- mc->has_hotpluggable_cpus = true;
- assert(!mc->get_hotplug_handler);
- mc->get_hotplug_handler = s390_get_hotplug_handler;
- mc->cpu_index_to_instance_props = s390_cpu_index_to_props;
- mc->possible_cpu_arch_ids = s390_possible_cpu_arch_ids;
- /* it is overridden with 'host' cpu *in kvm_arch_init* */
- mc->default_cpu_type = S390_CPU_TYPE_NAME("qemu");
- hc->plug = s390_machine_device_plug;
- hc->unplug_request = s390_machine_device_unplug_request;
- nc->nmi_monitor_handler = s390_nmi;
- mc->default_ram_id = "s390.ram";
-}
-
static inline bool machine_get_aes_key_wrap(Object *obj, Error **errp)
{
S390CcwMachineState *ms = S390_CCW_MACHINE(obj);
@@ -694,19 +663,29 @@ bool hpage_1m_allowed(void)
return get_machine_class()->hpage_1m_allowed;
}
-static char *machine_get_loadparm(Object *obj, Error **errp)
+static void machine_get_loadparm(Object *obj, Visitor *v,
+ const char *name, void *opaque,
+ Error **errp)
{
S390CcwMachineState *ms = S390_CCW_MACHINE(obj);
+ char *str = g_strndup((char *) ms->loadparm, sizeof(ms->loadparm));
- /* make a NUL-terminated string */
- return g_strndup((char *) ms->loadparm, sizeof(ms->loadparm));
+ visit_type_str(v, name, &str, errp);
+ g_free(str);
}
-static void machine_set_loadparm(Object *obj, const char *val, Error **errp)
+static void machine_set_loadparm(Object *obj, Visitor *v,
+ const char *name, void *opaque,
+ Error **errp)
{
S390CcwMachineState *ms = S390_CCW_MACHINE(obj);
+ char *val;
int i;
+ if (!visit_type_str(v, name, &val, errp)) {
+ return;
+ }
+
for (i = 0; i < sizeof(ms->loadparm) && val[i]; i++) {
uint8_t c = qemu_toupper(val[i]); /* mimic HMC */
@@ -724,29 +703,67 @@ static void machine_set_loadparm(Object *obj, const char *val, Error **errp)
ms->loadparm[i] = ' '; /* pad right with spaces */
}
}
-static inline void s390_machine_initfn(Object *obj)
+
+static void ccw_machine_class_init(ObjectClass *oc, void *data)
{
- object_property_add_bool(obj, "aes-key-wrap",
- machine_get_aes_key_wrap,
- machine_set_aes_key_wrap);
- object_property_set_description(obj, "aes-key-wrap",
+ MachineClass *mc = MACHINE_CLASS(oc);
+ NMIClass *nc = NMI_CLASS(oc);
+ HotplugHandlerClass *hc = HOTPLUG_HANDLER_CLASS(oc);
+ S390CcwMachineClass *s390mc = S390_CCW_MACHINE_CLASS(mc);
+
+ s390mc->ri_allowed = true;
+ s390mc->cpu_model_allowed = true;
+ s390mc->css_migration_enabled = true;
+ s390mc->hpage_1m_allowed = true;
+ mc->init = ccw_init;
+ mc->reset = s390_machine_reset;
+ mc->block_default_type = IF_VIRTIO;
+ mc->no_cdrom = 1;
+ mc->no_floppy = 1;
+ mc->no_parallel = 1;
+ mc->no_sdcard = 1;
+ mc->max_cpus = S390_MAX_CPUS;
+ mc->has_hotpluggable_cpus = true;
+ assert(!mc->get_hotplug_handler);
+ mc->get_hotplug_handler = s390_get_hotplug_handler;
+ mc->cpu_index_to_instance_props = s390_cpu_index_to_props;
+ mc->possible_cpu_arch_ids = s390_possible_cpu_arch_ids;
+ /* it is overridden with 'host' cpu *in kvm_arch_init* */
+ mc->default_cpu_type = S390_CPU_TYPE_NAME("qemu");
+ hc->plug = s390_machine_device_plug;
+ hc->unplug_request = s390_machine_device_unplug_request;
+ nc->nmi_monitor_handler = s390_nmi;
+ mc->default_ram_id = "s390.ram";
+
+ object_class_property_add_bool(oc, "aes-key-wrap",
+ machine_get_aes_key_wrap,
+ machine_set_aes_key_wrap);
+ object_class_property_set_description(oc, "aes-key-wrap",
"enable/disable AES key wrapping using the CPACF wrapping key");
- object_property_set_bool(obj, "aes-key-wrap", true, NULL);
- object_property_add_bool(obj, "dea-key-wrap",
- machine_get_dea_key_wrap,
- machine_set_dea_key_wrap);
- object_property_set_description(obj, "dea-key-wrap",
+ object_class_property_add_bool(oc, "dea-key-wrap",
+ machine_get_dea_key_wrap,
+ machine_set_dea_key_wrap);
+ object_class_property_set_description(oc, "dea-key-wrap",
"enable/disable DEA key wrapping using the CPACF wrapping key");
- object_property_set_bool(obj, "dea-key-wrap", true, NULL);
- object_property_add_str(obj, "loadparm",
- machine_get_loadparm, machine_set_loadparm);
- object_property_set_description(obj, "loadparm",
+
+ object_class_property_add(oc, "loadparm", "loadparm",
+ machine_get_loadparm, machine_set_loadparm,
+ NULL, NULL);
+ object_class_property_set_description(oc, "loadparm",
"Up to 8 chars in set of [A-Za-z0-9. ] (lower case chars converted"
" to upper case) to pass to machine loader, boot manager,"
" and guest kernel");
}
+static inline void s390_machine_initfn(Object *obj)
+{
+ S390CcwMachineState *ms = S390_CCW_MACHINE(obj);
+
+ ms->aes_key_wrap = true;
+ ms->dea_key_wrap = true;
+}
+
static const TypeInfo ccw_machine_info = {
.name = TYPE_S390_CCW_MACHINE,
.parent = TYPE_MACHINE,
--
2.37.3

View File

@ -0,0 +1,178 @@
From 8020177f1c40da2a9ca09fa20dc90eda65739671 Mon Sep 17 00:00:00 2001
From: Matthew Rosato <mjrosato@linux.ibm.com>
Date: Fri, 2 Sep 2022 13:27:31 -0400
Subject: [PATCH 06/42] s390x/pci: add routine to get host function handle from
CLP info
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [6/41] 8ab652cf4095e61f5f55726d41111de227d452e7
In order to interface with the underlying host zPCI device, we need
to know its function handle. Add a routine to grab this from the
vfio CLP capabilities chain.
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Pierre Morel <pmorel@linux.ibm.com>
Message-Id: <20220902172737.170349-3-mjrosato@linux.ibm.com>
[thuth: Replace free(info) with g_free(info)]
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit 21fa15298d88db2050a713cdf79c10cb0e09146f)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/s390x/s390-pci-vfio.c | 83 ++++++++++++++++++++++++++------
include/hw/s390x/s390-pci-vfio.h | 5 ++
2 files changed, 72 insertions(+), 16 deletions(-)
diff --git a/hw/s390x/s390-pci-vfio.c b/hw/s390x/s390-pci-vfio.c
index 6f80a47e29..08bcc55e85 100644
--- a/hw/s390x/s390-pci-vfio.c
+++ b/hw/s390x/s390-pci-vfio.c
@@ -124,6 +124,27 @@ static void s390_pci_read_base(S390PCIBusDevice *pbdev,
pbdev->zpci_fn.pft = 0;
}
+static bool get_host_fh(S390PCIBusDevice *pbdev, struct vfio_device_info *info,
+ uint32_t *fh)
+{
+ struct vfio_info_cap_header *hdr;
+ struct vfio_device_info_cap_zpci_base *cap;
+ VFIOPCIDevice *vpci = container_of(pbdev->pdev, VFIOPCIDevice, pdev);
+
+ hdr = vfio_get_device_info_cap(info, VFIO_DEVICE_INFO_CAP_ZPCI_BASE);
+
+ /* Can only get the host fh with version 2 or greater */
+ if (hdr == NULL || hdr->version < 2) {
+ trace_s390_pci_clp_cap(vpci->vbasedev.name,
+ VFIO_DEVICE_INFO_CAP_ZPCI_BASE);
+ return false;
+ }
+ cap = (void *) hdr;
+
+ *fh = cap->fh;
+ return true;
+}
+
static void s390_pci_read_group(S390PCIBusDevice *pbdev,
struct vfio_device_info *info)
{
@@ -217,25 +238,13 @@ static void s390_pci_read_pfip(S390PCIBusDevice *pbdev,
memcpy(pbdev->zpci_fn.pfip, cap->pfip, CLP_PFIP_NR_SEGMENTS);
}
-/*
- * This function will issue the VFIO_DEVICE_GET_INFO ioctl and look for
- * capabilities that contain information about CLP features provided by the
- * underlying host.
- * On entry, defaults have already been placed into the guest CLP response
- * buffers. On exit, defaults will have been overwritten for any CLP features
- * found in the capability chain; defaults will remain for any CLP features not
- * found in the chain.
- */
-void s390_pci_get_clp_info(S390PCIBusDevice *pbdev)
+static struct vfio_device_info *get_device_info(S390PCIBusDevice *pbdev,
+ uint32_t argsz)
{
- g_autofree struct vfio_device_info *info = NULL;
+ struct vfio_device_info *info = g_malloc0(argsz);
VFIOPCIDevice *vfio_pci;
- uint32_t argsz;
int fd;
- argsz = sizeof(*info);
- info = g_malloc0(argsz);
-
vfio_pci = container_of(pbdev->pdev, VFIOPCIDevice, pdev);
fd = vfio_pci->vbasedev.fd;
@@ -250,7 +259,8 @@ retry:
if (ioctl(fd, VFIO_DEVICE_GET_INFO, info)) {
trace_s390_pci_clp_dev_info(vfio_pci->vbasedev.name);
- return;
+ g_free(info);
+ return NULL;
}
if (info->argsz > argsz) {
@@ -259,6 +269,47 @@ retry:
goto retry;
}
+ return info;
+}
+
+/*
+ * Get the host function handle from the vfio CLP capabilities chain. Returns
+ * true if a fh value was placed into the provided buffer. Returns false
+ * if a fh could not be obtained (ioctl failed or capabilitiy version does
+ * not include the fh)
+ */
+bool s390_pci_get_host_fh(S390PCIBusDevice *pbdev, uint32_t *fh)
+{
+ g_autofree struct vfio_device_info *info = NULL;
+
+ assert(fh);
+
+ info = get_device_info(pbdev, sizeof(*info));
+ if (!info) {
+ return false;
+ }
+
+ return get_host_fh(pbdev, info, fh);
+}
+
+/*
+ * This function will issue the VFIO_DEVICE_GET_INFO ioctl and look for
+ * capabilities that contain information about CLP features provided by the
+ * underlying host.
+ * On entry, defaults have already been placed into the guest CLP response
+ * buffers. On exit, defaults will have been overwritten for any CLP features
+ * found in the capability chain; defaults will remain for any CLP features not
+ * found in the chain.
+ */
+void s390_pci_get_clp_info(S390PCIBusDevice *pbdev)
+{
+ g_autofree struct vfio_device_info *info = NULL;
+
+ info = get_device_info(pbdev, sizeof(*info));
+ if (!info) {
+ return;
+ }
+
/*
* Find the CLP features provided and fill in the guest CLP responses.
* Always call s390_pci_read_base first as information from this could
diff --git a/include/hw/s390x/s390-pci-vfio.h b/include/hw/s390x/s390-pci-vfio.h
index ff708aef50..ae1b126ff7 100644
--- a/include/hw/s390x/s390-pci-vfio.h
+++ b/include/hw/s390x/s390-pci-vfio.h
@@ -20,6 +20,7 @@ bool s390_pci_update_dma_avail(int fd, unsigned int *avail);
S390PCIDMACount *s390_pci_start_dma_count(S390pciState *s,
S390PCIBusDevice *pbdev);
void s390_pci_end_dma_count(S390pciState *s, S390PCIDMACount *cnt);
+bool s390_pci_get_host_fh(S390PCIBusDevice *pbdev, uint32_t *fh);
void s390_pci_get_clp_info(S390PCIBusDevice *pbdev);
#else
static inline bool s390_pci_update_dma_avail(int fd, unsigned int *avail)
@@ -33,6 +34,10 @@ static inline S390PCIDMACount *s390_pci_start_dma_count(S390pciState *s,
}
static inline void s390_pci_end_dma_count(S390pciState *s,
S390PCIDMACount *cnt) { }
+static inline bool s390_pci_get_host_fh(S390PCIBusDevice *pbdev, uint32_t *fh)
+{
+ return false;
+}
static inline void s390_pci_get_clp_info(S390PCIBusDevice *pbdev) { }
#endif
--
2.37.3

View File

@ -0,0 +1,99 @@
From de6319fe0ce09297beae5ff4636c03217abe6f26 Mon Sep 17 00:00:00 2001
From: Matthew Rosato <mjrosato@linux.ibm.com>
Date: Fri, 3 Dec 2021 09:27:06 -0500
Subject: [PATCH 04/42] s390x/pci: add supported DT information to clp response
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [4/41] 275668f6d38fbc1dfa2f1aa8f58b2c319de2657d
The DTSM is a mask that specifies which I/O Address Translation designation
types are supported. Today QEMU only supports DT=1.
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Eric Farman <farman@linux.ibm.com>
Reviewed-by: Pierre Morel <pmorel@linux.ibm.com>
Message-Id: <20211203142706.427279-5-mjrosato@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit ac6aa30ac47b2abaf142f76de46374da2a98f6e7)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/s390x/s390-pci-bus.c | 1 +
hw/s390x/s390-pci-inst.c | 1 +
hw/s390x/s390-pci-vfio.c | 1 +
include/hw/s390x/s390-pci-bus.h | 1 +
include/hw/s390x/s390-pci-clp.h | 3 ++-
5 files changed, 6 insertions(+), 1 deletion(-)
diff --git a/hw/s390x/s390-pci-bus.c b/hw/s390x/s390-pci-bus.c
index 1b51a72838..01b58ebc70 100644
--- a/hw/s390x/s390-pci-bus.c
+++ b/hw/s390x/s390-pci-bus.c
@@ -782,6 +782,7 @@ static void s390_pci_init_default_group(void)
resgrp->i = 128;
resgrp->maxstbl = 128;
resgrp->version = 0;
+ resgrp->dtsm = ZPCI_DTSM;
}
static void set_pbdev_info(S390PCIBusDevice *pbdev)
diff --git a/hw/s390x/s390-pci-inst.c b/hw/s390x/s390-pci-inst.c
index 07bab85ce5..6d400d4147 100644
--- a/hw/s390x/s390-pci-inst.c
+++ b/hw/s390x/s390-pci-inst.c
@@ -329,6 +329,7 @@ int clp_service_call(S390CPU *cpu, uint8_t r2, uintptr_t ra)
stw_p(&resgrp->i, group->zpci_group.i);
stw_p(&resgrp->maxstbl, group->zpci_group.maxstbl);
resgrp->version = group->zpci_group.version;
+ resgrp->dtsm = group->zpci_group.dtsm;
stw_p(&resgrp->hdr.rsp, CLP_RC_OK);
break;
}
diff --git a/hw/s390x/s390-pci-vfio.c b/hw/s390x/s390-pci-vfio.c
index 2a153fa8c9..6f80a47e29 100644
--- a/hw/s390x/s390-pci-vfio.c
+++ b/hw/s390x/s390-pci-vfio.c
@@ -160,6 +160,7 @@ static void s390_pci_read_group(S390PCIBusDevice *pbdev,
resgrp->i = cap->noi;
resgrp->maxstbl = cap->maxstbl;
resgrp->version = cap->version;
+ resgrp->dtsm = ZPCI_DTSM;
}
}
diff --git a/include/hw/s390x/s390-pci-bus.h b/include/hw/s390x/s390-pci-bus.h
index 2727e7bdef..da3cde2bb4 100644
--- a/include/hw/s390x/s390-pci-bus.h
+++ b/include/hw/s390x/s390-pci-bus.h
@@ -37,6 +37,7 @@
#define ZPCI_MAX_UID 0xffff
#define UID_UNDEFINED 0
#define UID_CHECKING_ENABLED 0x01
+#define ZPCI_DTSM 0x40
OBJECT_DECLARE_SIMPLE_TYPE(S390pciState, S390_PCI_HOST_BRIDGE)
OBJECT_DECLARE_SIMPLE_TYPE(S390PCIBus, S390_PCI_BUS)
diff --git a/include/hw/s390x/s390-pci-clp.h b/include/hw/s390x/s390-pci-clp.h
index 96b8e3f133..cc8c8662b8 100644
--- a/include/hw/s390x/s390-pci-clp.h
+++ b/include/hw/s390x/s390-pci-clp.h
@@ -163,7 +163,8 @@ typedef struct ClpRspQueryPciGrp {
uint8_t fr;
uint16_t maxstbl;
uint16_t mui;
- uint64_t reserved3;
+ uint8_t dtsm;
+ uint8_t reserved3[7];
uint64_t dasm; /* dma address space mask */
uint64_t msia; /* MSI address */
uint64_t reserved4;
--
2.37.3

View File

@ -0,0 +1,60 @@
From 5bd57d8ac3a4e75337eae81a3623b4dc2b417e2f Mon Sep 17 00:00:00 2001
From: Matthew Rosato <mjrosato@linux.ibm.com>
Date: Fri, 2 Sep 2022 13:27:33 -0400
Subject: [PATCH 08/42] s390x/pci: don't fence interpreted devices without
MSI-X
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [8/41] 52bad4368e9494c43133338b386dc0cc159aeedc
Lack of MSI-X support is not an issue for interpreted passthrough
devices, so let's let these in. This will allow, for example, ISM
devices to be passed through -- but only when interpretation is
available and being used.
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Pierre Morel <pmorel@linux.ibm.com>
Message-Id: <20220902172737.170349-5-mjrosato@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit 15d0e7942d3b31ff71d8e0e8cec3a8203214f19b)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/s390x/s390-pci-bus.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/hw/s390x/s390-pci-bus.c b/hw/s390x/s390-pci-bus.c
index 18bfae0465..07c7c155e3 100644
--- a/hw/s390x/s390-pci-bus.c
+++ b/hw/s390x/s390-pci-bus.c
@@ -881,6 +881,10 @@ static int s390_pci_msix_init(S390PCIBusDevice *pbdev)
static void s390_pci_msix_free(S390PCIBusDevice *pbdev)
{
+ if (pbdev->msix.entries == 0) {
+ return;
+ }
+
memory_region_del_subregion(&pbdev->iommu->mr, &pbdev->msix_notify_mr);
object_unparent(OBJECT(&pbdev->msix_notify_mr));
}
@@ -1093,7 +1097,7 @@ static void s390_pcihost_plug(HotplugHandler *hotplug_dev, DeviceState *dev,
pbdev->interp = false;
}
- if (s390_pci_msix_init(pbdev)) {
+ if (s390_pci_msix_init(pbdev) && !pbdev->interp) {
error_setg(errp, "MSI-X support is mandatory "
"in the S390 architecture");
return;
--
2.37.3

View File

@ -0,0 +1,77 @@
From 67ebb71d56e95adf185ab4971939e31c4c899863 Mon Sep 17 00:00:00 2001
From: Matthew Rosato <mjrosato@linux.ibm.com>
Date: Fri, 3 Dec 2021 09:27:04 -0500
Subject: [PATCH 02/42] s390x/pci: don't use hard-coded dma range in reg_ioat
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [2/41] c7897321f9848ef8f115130832774bbcd6724f03
Instead use the values from clp info, they will either be the hard-coded
values or what came from the host driver via vfio.
Fixes: 9670ee752727 ("s390x/pci: use a PCI Function structure")
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Eric Farman <farman@linux.ibm.com>
Reviewed-by: Pierre Morel <pmorel@linux.ibm.com>
Message-Id: <20211203142706.427279-3-mjrosato@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit df7ce0a94d9283f0656b4bc0f21566973ff649a3)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/s390x/s390-pci-inst.c | 9 +++++----
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/hw/s390x/s390-pci-inst.c b/hw/s390x/s390-pci-inst.c
index 1c8ad91175..11b7f6bfa1 100644
--- a/hw/s390x/s390-pci-inst.c
+++ b/hw/s390x/s390-pci-inst.c
@@ -916,9 +916,10 @@ int pci_dereg_irqs(S390PCIBusDevice *pbdev)
return 0;
}
-static int reg_ioat(CPUS390XState *env, S390PCIIOMMU *iommu, ZpciFib fib,
+static int reg_ioat(CPUS390XState *env, S390PCIBusDevice *pbdev, ZpciFib fib,
uintptr_t ra)
{
+ S390PCIIOMMU *iommu = pbdev->iommu;
uint64_t pba = ldq_p(&fib.pba);
uint64_t pal = ldq_p(&fib.pal);
uint64_t g_iota = ldq_p(&fib.iota);
@@ -927,7 +928,7 @@ static int reg_ioat(CPUS390XState *env, S390PCIIOMMU *iommu, ZpciFib fib,
pba &= ~0xfff;
pal |= 0xfff;
- if (pba > pal || pba < ZPCI_SDMA_ADDR || pal > ZPCI_EDMA_ADDR) {
+ if (pba > pal || pba < pbdev->zpci_fn.sdma || pal > pbdev->zpci_fn.edma) {
s390_program_interrupt(env, PGM_OPERAND, ra);
return -EINVAL;
}
@@ -1125,7 +1126,7 @@ int mpcifc_service_call(S390CPU *cpu, uint8_t r1, uint64_t fiba, uint8_t ar,
} else if (pbdev->iommu->enabled) {
cc = ZPCI_PCI_LS_ERR;
s390_set_status_code(env, r1, ZPCI_MOD_ST_SEQUENCE);
- } else if (reg_ioat(env, pbdev->iommu, fib, ra)) {
+ } else if (reg_ioat(env, pbdev, fib, ra)) {
cc = ZPCI_PCI_LS_ERR;
s390_set_status_code(env, r1, ZPCI_MOD_ST_INSUF_RES);
}
@@ -1150,7 +1151,7 @@ int mpcifc_service_call(S390CPU *cpu, uint8_t r1, uint64_t fiba, uint8_t ar,
s390_set_status_code(env, r1, ZPCI_MOD_ST_SEQUENCE);
} else {
pci_dereg_ioat(pbdev->iommu);
- if (reg_ioat(env, pbdev->iommu, fib, ra)) {
+ if (reg_ioat(env, pbdev, fib, ra)) {
cc = ZPCI_PCI_LS_ERR;
s390_set_status_code(env, r1, ZPCI_MOD_ST_INSUF_RES);
}
--
2.37.3

View File

@ -0,0 +1,265 @@
From 362fae654bbae03741003e565fb95d73d8c0025f Mon Sep 17 00:00:00 2001
From: Matthew Rosato <mjrosato@linux.ibm.com>
Date: Fri, 2 Sep 2022 13:27:34 -0400
Subject: [PATCH 09/42] s390x/pci: enable adapter event notification for
interpreted devices
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [9/41] 771975c436c7cb608e0e9e40edd732ac310beb69
Use the associated kvm ioctl operation to enable adapter event notification
and forwarding for devices when requested. This feature will be set up
with or without firmware assist based upon the 'forwarding_assist' setting.
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Message-Id: <20220902172737.170349-6-mjrosato@linux.ibm.com>
[thuth: Rename "forwarding_assist" property to "forwarding-assist"]
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit d0bc7091c2013ad2fa164100cf7b17962370e8ab)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/s390x/s390-pci-bus.c | 20 ++++++++++++++---
hw/s390x/s390-pci-inst.c | 40 +++++++++++++++++++++++++++++++--
hw/s390x/s390-pci-kvm.c | 30 +++++++++++++++++++++++++
include/hw/s390x/s390-pci-bus.h | 1 +
include/hw/s390x/s390-pci-kvm.h | 14 ++++++++++++
5 files changed, 100 insertions(+), 5 deletions(-)
diff --git a/hw/s390x/s390-pci-bus.c b/hw/s390x/s390-pci-bus.c
index 07c7c155e3..cd152ce711 100644
--- a/hw/s390x/s390-pci-bus.c
+++ b/hw/s390x/s390-pci-bus.c
@@ -190,7 +190,10 @@ void s390_pci_sclp_deconfigure(SCCB *sccb)
rc = SCLP_RC_NO_ACTION_REQUIRED;
break;
default:
- if (pbdev->summary_ind) {
+ if (pbdev->interp && (pbdev->fh & FH_MASK_ENABLE)) {
+ /* Interpreted devices were using interrupt forwarding */
+ s390_pci_kvm_aif_disable(pbdev);
+ } else if (pbdev->summary_ind) {
pci_dereg_irqs(pbdev);
}
if (pbdev->iommu->enabled) {
@@ -1082,6 +1085,7 @@ static void s390_pcihost_plug(HotplugHandler *hotplug_dev, DeviceState *dev,
} else {
DPRINTF("zPCI interpretation facilities missing.\n");
pbdev->interp = false;
+ pbdev->forwarding_assist = false;
}
}
pbdev->iommu->dma_limit = s390_pci_start_dma_count(s, pbdev);
@@ -1090,11 +1094,13 @@ static void s390_pcihost_plug(HotplugHandler *hotplug_dev, DeviceState *dev,
if (!pbdev->interp) {
/* Do vfio passthrough but intercept for I/O */
pbdev->fh |= FH_SHM_VFIO;
+ pbdev->forwarding_assist = false;
}
} else {
pbdev->fh |= FH_SHM_EMUL;
/* Always intercept emulated devices */
pbdev->interp = false;
+ pbdev->forwarding_assist = false;
}
if (s390_pci_msix_init(pbdev) && !pbdev->interp) {
@@ -1244,7 +1250,10 @@ static void s390_pcihost_reset(DeviceState *dev)
/* Process all pending unplug requests */
QTAILQ_FOREACH_SAFE(pbdev, &s->zpci_devs, link, next) {
if (pbdev->unplug_requested) {
- if (pbdev->summary_ind) {
+ if (pbdev->interp && (pbdev->fh & FH_MASK_ENABLE)) {
+ /* Interpreted devices were using interrupt forwarding */
+ s390_pci_kvm_aif_disable(pbdev);
+ } else if (pbdev->summary_ind) {
pci_dereg_irqs(pbdev);
}
if (pbdev->iommu->enabled) {
@@ -1382,7 +1391,10 @@ static void s390_pci_device_reset(DeviceState *dev)
break;
}
- if (pbdev->summary_ind) {
+ if (pbdev->interp && (pbdev->fh & FH_MASK_ENABLE)) {
+ /* Interpreted devices were using interrupt forwarding */
+ s390_pci_kvm_aif_disable(pbdev);
+ } else if (pbdev->summary_ind) {
pci_dereg_irqs(pbdev);
}
if (pbdev->iommu->enabled) {
@@ -1428,6 +1440,8 @@ static Property s390_pci_device_properties[] = {
DEFINE_PROP_S390_PCI_FID("fid", S390PCIBusDevice, fid),
DEFINE_PROP_STRING("target", S390PCIBusDevice, target),
DEFINE_PROP_BOOL("interpret", S390PCIBusDevice, interp, true),
+ DEFINE_PROP_BOOL("forwarding-assist", S390PCIBusDevice, forwarding_assist,
+ true),
DEFINE_PROP_END_OF_LIST(),
};
diff --git a/hw/s390x/s390-pci-inst.c b/hw/s390x/s390-pci-inst.c
index 651ec38635..20a9bcc7af 100644
--- a/hw/s390x/s390-pci-inst.c
+++ b/hw/s390x/s390-pci-inst.c
@@ -1066,6 +1066,32 @@ static void fmb_update(void *opaque)
timer_mod(pbdev->fmb_timer, t + pbdev->pci_group->zpci_group.mui);
}
+static int mpcifc_reg_int_interp(S390PCIBusDevice *pbdev, ZpciFib *fib)
+{
+ int rc;
+
+ rc = s390_pci_kvm_aif_enable(pbdev, fib, pbdev->forwarding_assist);
+ if (rc) {
+ DPRINTF("Failed to enable interrupt forwarding\n");
+ return rc;
+ }
+
+ return 0;
+}
+
+static int mpcifc_dereg_int_interp(S390PCIBusDevice *pbdev, ZpciFib *fib)
+{
+ int rc;
+
+ rc = s390_pci_kvm_aif_disable(pbdev);
+ if (rc) {
+ DPRINTF("Failed to disable interrupt forwarding\n");
+ return rc;
+ }
+
+ return 0;
+}
+
int mpcifc_service_call(S390CPU *cpu, uint8_t r1, uint64_t fiba, uint8_t ar,
uintptr_t ra)
{
@@ -1120,7 +1146,12 @@ int mpcifc_service_call(S390CPU *cpu, uint8_t r1, uint64_t fiba, uint8_t ar,
switch (oc) {
case ZPCI_MOD_FC_REG_INT:
- if (pbdev->summary_ind) {
+ if (pbdev->interp) {
+ if (mpcifc_reg_int_interp(pbdev, &fib)) {
+ cc = ZPCI_PCI_LS_ERR;
+ s390_set_status_code(env, r1, ZPCI_MOD_ST_SEQUENCE);
+ }
+ } else if (pbdev->summary_ind) {
cc = ZPCI_PCI_LS_ERR;
s390_set_status_code(env, r1, ZPCI_MOD_ST_SEQUENCE);
} else if (reg_irqs(env, pbdev, fib)) {
@@ -1129,7 +1160,12 @@ int mpcifc_service_call(S390CPU *cpu, uint8_t r1, uint64_t fiba, uint8_t ar,
}
break;
case ZPCI_MOD_FC_DEREG_INT:
- if (!pbdev->summary_ind) {
+ if (pbdev->interp) {
+ if (mpcifc_dereg_int_interp(pbdev, &fib)) {
+ cc = ZPCI_PCI_LS_ERR;
+ s390_set_status_code(env, r1, ZPCI_MOD_ST_SEQUENCE);
+ }
+ } else if (!pbdev->summary_ind) {
cc = ZPCI_PCI_LS_ERR;
s390_set_status_code(env, r1, ZPCI_MOD_ST_SEQUENCE);
} else {
diff --git a/hw/s390x/s390-pci-kvm.c b/hw/s390x/s390-pci-kvm.c
index 0f16104a74..9134fe185f 100644
--- a/hw/s390x/s390-pci-kvm.c
+++ b/hw/s390x/s390-pci-kvm.c
@@ -11,12 +11,42 @@
#include "qemu/osdep.h"
+#include <linux/kvm.h>
+
#include "kvm/kvm_s390x.h"
#include "hw/s390x/pv.h"
+#include "hw/s390x/s390-pci-bus.h"
#include "hw/s390x/s390-pci-kvm.h"
+#include "hw/s390x/s390-pci-inst.h"
#include "cpu_models.h"
bool s390_pci_kvm_interp_allowed(void)
{
return kvm_s390_get_zpci_op() && !s390_is_pv();
}
+
+int s390_pci_kvm_aif_enable(S390PCIBusDevice *pbdev, ZpciFib *fib, bool assist)
+{
+ struct kvm_s390_zpci_op args = {
+ .fh = pbdev->fh,
+ .op = KVM_S390_ZPCIOP_REG_AEN,
+ .u.reg_aen.ibv = fib->aibv,
+ .u.reg_aen.sb = fib->aisb,
+ .u.reg_aen.noi = FIB_DATA_NOI(fib->data),
+ .u.reg_aen.isc = FIB_DATA_ISC(fib->data),
+ .u.reg_aen.sbo = FIB_DATA_AISBO(fib->data),
+ .u.reg_aen.flags = (assist) ? 0 : KVM_S390_ZPCIOP_REGAEN_HOST
+ };
+
+ return kvm_vm_ioctl(kvm_state, KVM_S390_ZPCI_OP, &args);
+}
+
+int s390_pci_kvm_aif_disable(S390PCIBusDevice *pbdev)
+{
+ struct kvm_s390_zpci_op args = {
+ .fh = pbdev->fh,
+ .op = KVM_S390_ZPCIOP_DEREG_AEN
+ };
+
+ return kvm_vm_ioctl(kvm_state, KVM_S390_ZPCI_OP, &args);
+}
diff --git a/include/hw/s390x/s390-pci-bus.h b/include/hw/s390x/s390-pci-bus.h
index a9843dfe97..5b09f0cf2f 100644
--- a/include/hw/s390x/s390-pci-bus.h
+++ b/include/hw/s390x/s390-pci-bus.h
@@ -351,6 +351,7 @@ struct S390PCIBusDevice {
bool pci_unplug_request_processed;
bool unplug_requested;
bool interp;
+ bool forwarding_assist;
QTAILQ_ENTRY(S390PCIBusDevice) link;
};
diff --git a/include/hw/s390x/s390-pci-kvm.h b/include/hw/s390x/s390-pci-kvm.h
index 80a2e7d0ca..933814a402 100644
--- a/include/hw/s390x/s390-pci-kvm.h
+++ b/include/hw/s390x/s390-pci-kvm.h
@@ -12,13 +12,27 @@
#ifndef HW_S390_PCI_KVM_H
#define HW_S390_PCI_KVM_H
+#include "hw/s390x/s390-pci-bus.h"
+#include "hw/s390x/s390-pci-inst.h"
+
#ifdef CONFIG_KVM
bool s390_pci_kvm_interp_allowed(void);
+int s390_pci_kvm_aif_enable(S390PCIBusDevice *pbdev, ZpciFib *fib, bool assist);
+int s390_pci_kvm_aif_disable(S390PCIBusDevice *pbdev);
#else
static inline bool s390_pci_kvm_interp_allowed(void)
{
return false;
}
+static inline int s390_pci_kvm_aif_enable(S390PCIBusDevice *pbdev, ZpciFib *fib,
+ bool assist)
+{
+ return -EINVAL;
+}
+static inline int s390_pci_kvm_aif_disable(S390PCIBusDevice *pbdev)
+{
+ return -EINVAL;
+}
#endif
#endif
--
2.37.3

View File

@ -0,0 +1,319 @@
From 62fbb66d18f598d0896164383aab465e093fb0c1 Mon Sep 17 00:00:00 2001
From: Matthew Rosato <mjrosato@linux.ibm.com>
Date: Fri, 2 Sep 2022 13:27:32 -0400
Subject: [PATCH 07/42] s390x/pci: enable for load/store interpretation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [7/41] 3a96e901e295bb9e0c530638c45b5da5d60c00bd
If the ZPCI_OP ioctl reports that is is available and usable, then the
underlying KVM host will enable load/store intepretation for any guest
device without a SHM bit in the guest function handle. For a device that
will be using interpretation support, ensure the guest function handle
matches the host function handle; this value is re-checked every time the
guest issues a SET PCI FN to enable the guest device as it is the only
opportunity to reflect function handle changes.
By default, unless interpret=off is specified, interpretation support will
always be assumed and exploited if the necessary ioctl and features are
available on the host kernel. When these are unavailable, we will silently
revert to the interception model; this allows existing guest configurations
to work unmodified on hosts with and without zPCI interpretation support,
allowing QEMU to choose the best support model available.
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Acked-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20220902172737.170349-4-mjrosato@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit dd1d5fd9684beeb0c14c39f497ef2aa9ac683aa7)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/s390x/meson.build | 1 +
hw/s390x/s390-pci-bus.c | 66 ++++++++++++++++++++++++++++++++-
hw/s390x/s390-pci-inst.c | 16 ++++++++
hw/s390x/s390-pci-kvm.c | 22 +++++++++++
include/hw/s390x/s390-pci-bus.h | 1 +
include/hw/s390x/s390-pci-kvm.h | 24 ++++++++++++
target/s390x/kvm/kvm.c | 7 ++++
target/s390x/kvm/kvm_s390x.h | 1 +
8 files changed, 137 insertions(+), 1 deletion(-)
create mode 100644 hw/s390x/s390-pci-kvm.c
create mode 100644 include/hw/s390x/s390-pci-kvm.h
diff --git a/hw/s390x/meson.build b/hw/s390x/meson.build
index 28484256ec..6e6e47fcda 100644
--- a/hw/s390x/meson.build
+++ b/hw/s390x/meson.build
@@ -23,6 +23,7 @@ s390x_ss.add(when: 'CONFIG_KVM', if_true: files(
's390-skeys-kvm.c',
's390-stattrib-kvm.c',
'pv.c',
+ 's390-pci-kvm.c',
))
s390x_ss.add(when: 'CONFIG_TCG', if_true: files(
'tod-tcg.c',
diff --git a/hw/s390x/s390-pci-bus.c b/hw/s390x/s390-pci-bus.c
index 01b58ebc70..18bfae0465 100644
--- a/hw/s390x/s390-pci-bus.c
+++ b/hw/s390x/s390-pci-bus.c
@@ -16,6 +16,7 @@
#include "qapi/visitor.h"
#include "hw/s390x/s390-pci-bus.h"
#include "hw/s390x/s390-pci-inst.h"
+#include "hw/s390x/s390-pci-kvm.h"
#include "hw/s390x/s390-pci-vfio.h"
#include "hw/pci/pci_bus.h"
#include "hw/qdev-properties.h"
@@ -971,12 +972,51 @@ static void s390_pci_update_subordinate(PCIDevice *dev, uint32_t nr)
}
}
+static int s390_pci_interp_plug(S390pciState *s, S390PCIBusDevice *pbdev)
+{
+ uint32_t idx, fh;
+
+ if (!s390_pci_get_host_fh(pbdev, &fh)) {
+ return -EPERM;
+ }
+
+ /*
+ * The host device is already in an enabled state, but we always present
+ * the initial device state to the guest as disabled (ZPCI_FS_DISABLED).
+ * Therefore, mask off the enable bit from the passthrough handle until
+ * the guest issues a CLP SET PCI FN later to enable the device.
+ */
+ pbdev->fh = fh & ~FH_MASK_ENABLE;
+
+ /* Next, see if the idx is already in-use */
+ idx = pbdev->fh & FH_MASK_INDEX;
+ if (pbdev->idx != idx) {
+ if (s390_pci_find_dev_by_idx(s, idx)) {
+ return -EINVAL;
+ }
+ /*
+ * Update the idx entry with the passed through idx
+ * If the relinquished idx is lower than next_idx, use it
+ * to replace next_idx
+ */
+ g_hash_table_remove(s->zpci_table, &pbdev->idx);
+ if (idx < s->next_idx) {
+ s->next_idx = idx;
+ }
+ pbdev->idx = idx;
+ g_hash_table_insert(s->zpci_table, &pbdev->idx, pbdev);
+ }
+
+ return 0;
+}
+
static void s390_pcihost_plug(HotplugHandler *hotplug_dev, DeviceState *dev,
Error **errp)
{
S390pciState *s = S390_PCI_HOST_BRIDGE(hotplug_dev);
PCIDevice *pdev = NULL;
S390PCIBusDevice *pbdev = NULL;
+ int rc;
if (object_dynamic_cast(OBJECT(dev), TYPE_PCI_BRIDGE)) {
PCIBridge *pb = PCI_BRIDGE(dev);
@@ -1022,12 +1062,35 @@ static void s390_pcihost_plug(HotplugHandler *hotplug_dev, DeviceState *dev,
set_pbdev_info(pbdev);
if (object_dynamic_cast(OBJECT(dev), "vfio-pci")) {
- pbdev->fh |= FH_SHM_VFIO;
+ /*
+ * By default, interpretation is always requested; if the available
+ * facilities indicate it is not available, fallback to the
+ * interception model.
+ */
+ if (pbdev->interp) {
+ if (s390_pci_kvm_interp_allowed()) {
+ rc = s390_pci_interp_plug(s, pbdev);
+ if (rc) {
+ error_setg(errp, "Plug failed for zPCI device in "
+ "interpretation mode: %d", rc);
+ return;
+ }
+ } else {
+ DPRINTF("zPCI interpretation facilities missing.\n");
+ pbdev->interp = false;
+ }
+ }
pbdev->iommu->dma_limit = s390_pci_start_dma_count(s, pbdev);
/* Fill in CLP information passed via the vfio region */
s390_pci_get_clp_info(pbdev);
+ if (!pbdev->interp) {
+ /* Do vfio passthrough but intercept for I/O */
+ pbdev->fh |= FH_SHM_VFIO;
+ }
} else {
pbdev->fh |= FH_SHM_EMUL;
+ /* Always intercept emulated devices */
+ pbdev->interp = false;
}
if (s390_pci_msix_init(pbdev)) {
@@ -1360,6 +1423,7 @@ static Property s390_pci_device_properties[] = {
DEFINE_PROP_UINT16("uid", S390PCIBusDevice, uid, UID_UNDEFINED),
DEFINE_PROP_S390_PCI_FID("fid", S390PCIBusDevice, fid),
DEFINE_PROP_STRING("target", S390PCIBusDevice, target),
+ DEFINE_PROP_BOOL("interpret", S390PCIBusDevice, interp, true),
DEFINE_PROP_END_OF_LIST(),
};
diff --git a/hw/s390x/s390-pci-inst.c b/hw/s390x/s390-pci-inst.c
index 6d400d4147..651ec38635 100644
--- a/hw/s390x/s390-pci-inst.c
+++ b/hw/s390x/s390-pci-inst.c
@@ -18,6 +18,8 @@
#include "sysemu/hw_accel.h"
#include "hw/s390x/s390-pci-inst.h"
#include "hw/s390x/s390-pci-bus.h"
+#include "hw/s390x/s390-pci-kvm.h"
+#include "hw/s390x/s390-pci-vfio.h"
#include "hw/s390x/tod.h"
#ifndef DEBUG_S390PCI_INST
@@ -246,6 +248,20 @@ int clp_service_call(S390CPU *cpu, uint8_t r2, uintptr_t ra)
goto out;
}
+ /*
+ * Take this opportunity to make sure we still have an accurate
+ * host fh. It's possible part of the handle changed while the
+ * device was disabled to the guest (e.g. vfio hot reset for
+ * ISM during plug)
+ */
+ if (pbdev->interp) {
+ /* Take this opportunity to make sure we are sync'd with host */
+ if (!s390_pci_get_host_fh(pbdev, &pbdev->fh) ||
+ !(pbdev->fh & FH_MASK_ENABLE)) {
+ stw_p(&ressetpci->hdr.rsp, CLP_RC_SETPCIFN_FH);
+ goto out;
+ }
+ }
pbdev->fh |= FH_MASK_ENABLE;
pbdev->state = ZPCI_FS_ENABLED;
stl_p(&ressetpci->fh, pbdev->fh);
diff --git a/hw/s390x/s390-pci-kvm.c b/hw/s390x/s390-pci-kvm.c
new file mode 100644
index 0000000000..0f16104a74
--- /dev/null
+++ b/hw/s390x/s390-pci-kvm.c
@@ -0,0 +1,22 @@
+/*
+ * s390 zPCI KVM interfaces
+ *
+ * Copyright 2022 IBM Corp.
+ * Author(s): Matthew Rosato <mjrosato@linux.ibm.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or (at
+ * your option) any later version. See the COPYING file in the top-level
+ * directory.
+ */
+
+#include "qemu/osdep.h"
+
+#include "kvm/kvm_s390x.h"
+#include "hw/s390x/pv.h"
+#include "hw/s390x/s390-pci-kvm.h"
+#include "cpu_models.h"
+
+bool s390_pci_kvm_interp_allowed(void)
+{
+ return kvm_s390_get_zpci_op() && !s390_is_pv();
+}
diff --git a/include/hw/s390x/s390-pci-bus.h b/include/hw/s390x/s390-pci-bus.h
index da3cde2bb4..a9843dfe97 100644
--- a/include/hw/s390x/s390-pci-bus.h
+++ b/include/hw/s390x/s390-pci-bus.h
@@ -350,6 +350,7 @@ struct S390PCIBusDevice {
IndAddr *indicator;
bool pci_unplug_request_processed;
bool unplug_requested;
+ bool interp;
QTAILQ_ENTRY(S390PCIBusDevice) link;
};
diff --git a/include/hw/s390x/s390-pci-kvm.h b/include/hw/s390x/s390-pci-kvm.h
new file mode 100644
index 0000000000..80a2e7d0ca
--- /dev/null
+++ b/include/hw/s390x/s390-pci-kvm.h
@@ -0,0 +1,24 @@
+/*
+ * s390 PCI KVM interfaces
+ *
+ * Copyright 2022 IBM Corp.
+ * Author(s): Matthew Rosato <mjrosato@linux.ibm.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2 or (at
+ * your option) any later version. See the COPYING file in the top-level
+ * directory.
+ */
+
+#ifndef HW_S390_PCI_KVM_H
+#define HW_S390_PCI_KVM_H
+
+#ifdef CONFIG_KVM
+bool s390_pci_kvm_interp_allowed(void);
+#else
+static inline bool s390_pci_kvm_interp_allowed(void)
+{
+ return false;
+}
+#endif
+
+#endif
diff --git a/target/s390x/kvm/kvm.c b/target/s390x/kvm/kvm.c
index ba04997da1..30712487d4 100644
--- a/target/s390x/kvm/kvm.c
+++ b/target/s390x/kvm/kvm.c
@@ -158,6 +158,7 @@ static int cap_ri;
static int cap_hpage_1m;
static int cap_vcpu_resets;
static int cap_protected;
+static int cap_zpci_op;
static bool mem_op_storage_key_support;
@@ -363,6 +364,7 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
cap_s390_irq = kvm_check_extension(s, KVM_CAP_S390_INJECT_IRQ);
cap_vcpu_resets = kvm_check_extension(s, KVM_CAP_S390_VCPU_RESETS);
cap_protected = kvm_check_extension(s, KVM_CAP_S390_PROTECTED);
+ cap_zpci_op = kvm_check_extension(s, KVM_CAP_S390_ZPCI_OP);
kvm_vm_enable_cap(s, KVM_CAP_S390_USER_SIGP, 0);
kvm_vm_enable_cap(s, KVM_CAP_S390_VECTOR_REGISTERS, 0);
@@ -2579,3 +2581,8 @@ bool kvm_arch_cpu_check_are_resettable(void)
{
return true;
}
+
+int kvm_s390_get_zpci_op(void)
+{
+ return cap_zpci_op;
+}
diff --git a/target/s390x/kvm/kvm_s390x.h b/target/s390x/kvm/kvm_s390x.h
index 05a5e1e6f4..aaae8570de 100644
--- a/target/s390x/kvm/kvm_s390x.h
+++ b/target/s390x/kvm/kvm_s390x.h
@@ -27,6 +27,7 @@ void kvm_s390_vcpu_interrupt_pre_save(S390CPU *cpu);
int kvm_s390_vcpu_interrupt_post_load(S390CPU *cpu);
int kvm_s390_get_hpage_1m(void);
int kvm_s390_get_ri(void);
+int kvm_s390_get_zpci_op(void);
int kvm_s390_get_clock(uint8_t *tod_high, uint64_t *tod_clock);
int kvm_s390_get_clock_ext(uint8_t *tod_high, uint64_t *tod_clock);
int kvm_s390_set_clock(uint8_t tod_high, uint64_t tod_clock);
--
2.37.3

View File

@ -0,0 +1,192 @@
From b98a5bc4c21284dd0a8a1c86b91af81fcb75f060 Mon Sep 17 00:00:00 2001
From: Matthew Rosato <mjrosato@linux.ibm.com>
Date: Fri, 2 Sep 2022 13:27:35 -0400
Subject: [PATCH 10/42] s390x/pci: let intercept devices have separate PCI
groups
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [10/41] 1545bdcd2e21386afa9869f0414e96eecb62647d
Let's use the reserved pool of simulated PCI groups to allow intercept
devices to have separate groups from interpreted devices as some group
values may be different. If we run out of simulated PCI groups, subsequent
intercept devices just get the default group.
Furthermore, if we encounter any PCI groups from hostdevs that are marked
as simulated, let's just assign them to the default group to avoid
conflicts between host simulated groups and our own simulated groups.
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Pierre Morel <pmorel@linux.ibm.com>
Message-Id: <20220902172737.170349-7-mjrosato@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit 30dcf4f7fd23bef7d72a2454c60881710fd4c785)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/s390x/s390-pci-bus.c | 19 ++++++++++++++--
hw/s390x/s390-pci-vfio.c | 40 ++++++++++++++++++++++++++++++---
include/hw/s390x/s390-pci-bus.h | 6 ++++-
3 files changed, 59 insertions(+), 6 deletions(-)
diff --git a/hw/s390x/s390-pci-bus.c b/hw/s390x/s390-pci-bus.c
index cd152ce711..d8b1e44a02 100644
--- a/hw/s390x/s390-pci-bus.c
+++ b/hw/s390x/s390-pci-bus.c
@@ -748,13 +748,14 @@ static void s390_pci_iommu_free(S390pciState *s, PCIBus *bus, int32_t devfn)
object_unref(OBJECT(iommu));
}
-S390PCIGroup *s390_group_create(int id)
+S390PCIGroup *s390_group_create(int id, int host_id)
{
S390PCIGroup *group;
S390pciState *s = s390_get_phb();
group = g_new0(S390PCIGroup, 1);
group->id = id;
+ group->host_id = host_id;
QTAILQ_INSERT_TAIL(&s->zpci_groups, group, link);
return group;
}
@@ -772,12 +773,25 @@ S390PCIGroup *s390_group_find(int id)
return NULL;
}
+S390PCIGroup *s390_group_find_host_sim(int host_id)
+{
+ S390PCIGroup *group;
+ S390pciState *s = s390_get_phb();
+
+ QTAILQ_FOREACH(group, &s->zpci_groups, link) {
+ if (group->id >= ZPCI_SIM_GRP_START && group->host_id == host_id) {
+ return group;
+ }
+ }
+ return NULL;
+}
+
static void s390_pci_init_default_group(void)
{
S390PCIGroup *group;
ClpRspQueryPciGrp *resgrp;
- group = s390_group_create(ZPCI_DEFAULT_FN_GRP);
+ group = s390_group_create(ZPCI_DEFAULT_FN_GRP, ZPCI_DEFAULT_FN_GRP);
resgrp = &group->zpci_group;
resgrp->fr = 1;
resgrp->dasm = 0;
@@ -825,6 +839,7 @@ static void s390_pcihost_realize(DeviceState *dev, Error **errp)
NULL, g_free);
s->zpci_table = g_hash_table_new_full(g_int_hash, g_int_equal, NULL, NULL);
s->bus_no = 0;
+ s->next_sim_grp = ZPCI_SIM_GRP_START;
QTAILQ_INIT(&s->pending_sei);
QTAILQ_INIT(&s->zpci_devs);
QTAILQ_INIT(&s->zpci_dma_limit);
diff --git a/hw/s390x/s390-pci-vfio.c b/hw/s390x/s390-pci-vfio.c
index 08bcc55e85..338f436e87 100644
--- a/hw/s390x/s390-pci-vfio.c
+++ b/hw/s390x/s390-pci-vfio.c
@@ -150,13 +150,18 @@ static void s390_pci_read_group(S390PCIBusDevice *pbdev,
{
struct vfio_info_cap_header *hdr;
struct vfio_device_info_cap_zpci_group *cap;
+ S390pciState *s = s390_get_phb();
ClpRspQueryPciGrp *resgrp;
VFIOPCIDevice *vpci = container_of(pbdev->pdev, VFIOPCIDevice, pdev);
+ uint8_t start_gid = pbdev->zpci_fn.pfgid;
hdr = vfio_get_device_info_cap(info, VFIO_DEVICE_INFO_CAP_ZPCI_GROUP);
- /* If capability not provided, just use the default group */
- if (hdr == NULL) {
+ /*
+ * If capability not provided or the underlying hostdev is simulated, just
+ * use the default group.
+ */
+ if (hdr == NULL || pbdev->zpci_fn.pfgid >= ZPCI_SIM_GRP_START) {
trace_s390_pci_clp_cap(vpci->vbasedev.name,
VFIO_DEVICE_INFO_CAP_ZPCI_GROUP);
pbdev->zpci_fn.pfgid = ZPCI_DEFAULT_FN_GRP;
@@ -165,11 +170,40 @@ static void s390_pci_read_group(S390PCIBusDevice *pbdev,
}
cap = (void *) hdr;
+ /*
+ * For an intercept device, let's use an existing simulated group if one
+ * one was already created for other intercept devices in this group.
+ * If not, create a new simulated group if any are still available.
+ * If all else fails, just fall back on the default group.
+ */
+ if (!pbdev->interp) {
+ pbdev->pci_group = s390_group_find_host_sim(pbdev->zpci_fn.pfgid);
+ if (pbdev->pci_group) {
+ /* Use existing simulated group */
+ pbdev->zpci_fn.pfgid = pbdev->pci_group->id;
+ return;
+ } else {
+ if (s->next_sim_grp == ZPCI_DEFAULT_FN_GRP) {
+ /* All out of simulated groups, use default */
+ trace_s390_pci_clp_cap(vpci->vbasedev.name,
+ VFIO_DEVICE_INFO_CAP_ZPCI_GROUP);
+ pbdev->zpci_fn.pfgid = ZPCI_DEFAULT_FN_GRP;
+ pbdev->pci_group = s390_group_find(ZPCI_DEFAULT_FN_GRP);
+ return;
+ } else {
+ /* We can assign a new simulated group */
+ pbdev->zpci_fn.pfgid = s->next_sim_grp;
+ s->next_sim_grp++;
+ /* Fall through to create the new sim group using CLP info */
+ }
+ }
+ }
+
/* See if the PCI group is already defined, create if not */
pbdev->pci_group = s390_group_find(pbdev->zpci_fn.pfgid);
if (!pbdev->pci_group) {
- pbdev->pci_group = s390_group_create(pbdev->zpci_fn.pfgid);
+ pbdev->pci_group = s390_group_create(pbdev->zpci_fn.pfgid, start_gid);
resgrp = &pbdev->pci_group->zpci_group;
if (cap->flags & VFIO_DEVICE_INFO_ZPCI_FLAG_REFRESH) {
diff --git a/include/hw/s390x/s390-pci-bus.h b/include/hw/s390x/s390-pci-bus.h
index 5b09f0cf2f..0605fcea24 100644
--- a/include/hw/s390x/s390-pci-bus.h
+++ b/include/hw/s390x/s390-pci-bus.h
@@ -315,13 +315,16 @@ typedef struct ZpciFmb {
QEMU_BUILD_BUG_MSG(offsetof(ZpciFmb, fmt0) != 48, "padding in ZpciFmb");
#define ZPCI_DEFAULT_FN_GRP 0xFF
+#define ZPCI_SIM_GRP_START 0xF0
typedef struct S390PCIGroup {
ClpRspQueryPciGrp zpci_group;
int id;
+ int host_id;
QTAILQ_ENTRY(S390PCIGroup) link;
} S390PCIGroup;
-S390PCIGroup *s390_group_create(int id);
+S390PCIGroup *s390_group_create(int id, int host_id);
S390PCIGroup *s390_group_find(int id);
+S390PCIGroup *s390_group_find_host_sim(int host_id);
struct S390PCIBusDevice {
DeviceState qdev;
@@ -370,6 +373,7 @@ struct S390pciState {
QTAILQ_HEAD(, S390PCIBusDevice) zpci_devs;
QTAILQ_HEAD(, S390PCIDMACount) zpci_dma_limit;
QTAILQ_HEAD(, S390PCIGroup) zpci_groups;
+ uint8_t next_sim_grp;
};
S390pciState *s390_get_phb(void);
--
2.37.3

View File

@ -0,0 +1,52 @@
From 65f90bfccf7500978879c15104a79de58173a06b Mon Sep 17 00:00:00 2001
From: Matthew Rosato <mjrosato@linux.ibm.com>
Date: Fri, 2 Sep 2022 13:27:36 -0400
Subject: [PATCH 11/42] s390x/pci: reflect proper maxstbl for groups of
interpreted devices
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [11/41] 9ac2f5dedef3d743ef621525eef222a3e09d63b3
The maximum supported store block length might be different depending
on whether the instruction is interpretively executed (firmware-reported
maximum) or handled via userspace intercept (host kernel API maximum).
Choose the best available value during group creation.
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Pierre Morel <pmorel@linux.ibm.com>
Message-Id: <20220902172737.170349-8-mjrosato@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit 9ee8f7e46a7d42ede69a4780200129bf1acb0d01)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/s390x/s390-pci-vfio.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/hw/s390x/s390-pci-vfio.c b/hw/s390x/s390-pci-vfio.c
index 338f436e87..2aefa508a0 100644
--- a/hw/s390x/s390-pci-vfio.c
+++ b/hw/s390x/s390-pci-vfio.c
@@ -213,7 +213,11 @@ static void s390_pci_read_group(S390PCIBusDevice *pbdev,
resgrp->msia = cap->msi_addr;
resgrp->mui = cap->mui;
resgrp->i = cap->noi;
- resgrp->maxstbl = cap->maxstbl;
+ if (pbdev->interp && hdr->version >= 2) {
+ resgrp->maxstbl = cap->imaxstbl;
+ } else {
+ resgrp->maxstbl = cap->maxstbl;
+ }
resgrp->version = cap->version;
resgrp->dtsm = ZPCI_DTSM;
}
--
2.37.3

View File

@ -0,0 +1,49 @@
From 55294fc4a955491f1fd947e4d98bd6df832c88ba Mon Sep 17 00:00:00 2001
From: Matthew Rosato <mjrosato@linux.ibm.com>
Date: Fri, 3 Dec 2021 09:27:03 -0500
Subject: [PATCH 01/42] s390x/pci: use a reserved ID for the default PCI group
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [1/41] ad3ed38dec95acf0da04d7669fe772d798d039fc
The current default PCI group being used can technically collide with a
real group ID passed from a hostdev. Let's instead use a group ID that
comes from a special pool (0xF0-0xFF) that is architected to be reserved
for simulated devices.
Fixes: 28dc86a072 ("s390x/pci: use a PCI Group structure")
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Eric Farman <farman@linux.ibm.com>
Reviewed-by: Pierre Morel <pmorel@linux.ibm.com>
Message-Id: <20211203142706.427279-2-mjrosato@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit b2892a2b9d45d25b909108ca633d19f9d8d673f5)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
include/hw/s390x/s390-pci-bus.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/include/hw/s390x/s390-pci-bus.h b/include/hw/s390x/s390-pci-bus.h
index aa891c178d..2727e7bdef 100644
--- a/include/hw/s390x/s390-pci-bus.h
+++ b/include/hw/s390x/s390-pci-bus.h
@@ -313,7 +313,7 @@ typedef struct ZpciFmb {
} ZpciFmb;
QEMU_BUILD_BUG_MSG(offsetof(ZpciFmb, fmt0) != 48, "padding in ZpciFmb");
-#define ZPCI_DEFAULT_FN_GRP 0x20
+#define ZPCI_DEFAULT_FN_GRP 0xFF
typedef struct S390PCIGroup {
ClpRspQueryPciGrp zpci_group;
int id;
--
2.37.3

View File

@ -0,0 +1,59 @@
From c251a7a16e776072b9c7bbc4e61cfa4f52599b0a Mon Sep 17 00:00:00 2001
From: Matthew Rosato <mjrosato@linux.ibm.com>
Date: Fri, 3 Dec 2021 09:27:05 -0500
Subject: [PATCH 03/42] s390x/pci: use the passthrough measurement update
interval
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [3/41] bc31ea731fe64e51522f1202e65528311397b919
We may have gotten a measurement update interval from the underlying host
via vfio -- Use it to set the interval via which we update the function
measurement block.
Fixes: 28dc86a072 ("s390x/pci: use a PCI Group structure")
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Eric Farman <farman@linux.ibm.com>
Reviewed-by: Pierre Morel <pmorel@linux.ibm.com>
Message-Id: <20211203142706.427279-4-mjrosato@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit cb6d6a3e6aa1226b67fd218953dcb3866c3a6845)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/s390x/s390-pci-inst.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/hw/s390x/s390-pci-inst.c b/hw/s390x/s390-pci-inst.c
index 11b7f6bfa1..07bab85ce5 100644
--- a/hw/s390x/s390-pci-inst.c
+++ b/hw/s390x/s390-pci-inst.c
@@ -1046,7 +1046,7 @@ static void fmb_update(void *opaque)
sizeof(pbdev->fmb.last_update))) {
return;
}
- timer_mod(pbdev->fmb_timer, t + DEFAULT_MUI);
+ timer_mod(pbdev->fmb_timer, t + pbdev->pci_group->zpci_group.mui);
}
int mpcifc_service_call(S390CPU *cpu, uint8_t r1, uint64_t fiba, uint8_t ar,
@@ -1204,7 +1204,8 @@ int mpcifc_service_call(S390CPU *cpu, uint8_t r1, uint64_t fiba, uint8_t ar,
}
pbdev->fmb_addr = fmb_addr;
timer_mod(pbdev->fmb_timer,
- qemu_clock_get_ms(QEMU_CLOCK_VIRTUAL) + DEFAULT_MUI);
+ qemu_clock_get_ms(QEMU_CLOCK_VIRTUAL) +
+ pbdev->pci_group->zpci_group.mui);
break;
}
default:
--
2.37.3

View File

@ -0,0 +1,445 @@
From 86aeb4fd7ff9395afba574e422d83f990ce1f047 Mon Sep 17 00:00:00 2001
From: Janosch Frank <frankja@linux.ibm.com>
Date: Mon, 17 Oct 2022 08:38:22 +0000
Subject: [PATCH 41/42] s390x: pv: Add dump support
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [41/41] 2731c2329276e76013e3b3df21e9743bc74edd2b
Sometimes dumping a guest from the outside is the only way to get the
data that is needed. This can be the case if a dumping mechanism like
KDUMP hasn't been configured or data needs to be fetched at a specific
point. Dumping a protected guest from the outside without help from
fw/hw doesn't yield sufficient data to be useful. Hence we now
introduce PV dump support.
The PV dump support works by integrating the firmware into the dump
process. New Ultravisor calls are used to initiate the dump process,
dump cpu data, dump memory state and lastly complete the dump process.
The UV calls are exposed by KVM via the new KVM_PV_DUMP command and
its subcommands. The guest's data is fully encrypted and can only be
decrypted by the entity that owns the customer communication key for
the dumped guest. Also dumping needs to be allowed via a flag in the
SE header.
On the QEMU side of things we store the PV dump data in the newly
introduced architecture ELF sections (storage state and completion
data) and the cpu notes (for cpu dump data).
Users can use the zgetdump tool to convert the encrypted QEMU dump to an
unencrypted one.
Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: Steffen Eiden <seiden@linux.ibm.com>
Message-Id: <20221017083822.43118-11-frankja@linux.ibm.com>
(cherry picked from commit 113d8f4e95cf0450bea421263de6ec016c779ad0)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
dump/dump.c | 12 +-
include/sysemu/dump.h | 5 +
target/s390x/arch_dump.c | 262 +++++++++++++++++++++++++++++++++++----
3 files changed, 246 insertions(+), 33 deletions(-)
diff --git a/dump/dump.c b/dump/dump.c
index 4aa8fb64d2..5dee060b73 100644
--- a/dump/dump.c
+++ b/dump/dump.c
@@ -709,9 +709,9 @@ static void dump_begin(DumpState *s, Error **errp)
write_elf_notes(s, errp);
}
-static int64_t dump_filtered_memblock_size(GuestPhysBlock *block,
- int64_t filter_area_start,
- int64_t filter_area_length)
+int64_t dump_filtered_memblock_size(GuestPhysBlock *block,
+ int64_t filter_area_start,
+ int64_t filter_area_length)
{
int64_t size, left, right;
@@ -729,9 +729,9 @@ static int64_t dump_filtered_memblock_size(GuestPhysBlock *block,
return size;
}
-static int64_t dump_filtered_memblock_start(GuestPhysBlock *block,
- int64_t filter_area_start,
- int64_t filter_area_length)
+int64_t dump_filtered_memblock_start(GuestPhysBlock *block,
+ int64_t filter_area_start,
+ int64_t filter_area_length)
{
if (filter_area_length) {
/* return -1 if the block is not within filter area */
diff --git a/include/sysemu/dump.h b/include/sysemu/dump.h
index 38ccac7190..4ffed0b659 100644
--- a/include/sysemu/dump.h
+++ b/include/sysemu/dump.h
@@ -215,4 +215,9 @@ typedef struct DumpState {
uint16_t cpu_to_dump16(DumpState *s, uint16_t val);
uint32_t cpu_to_dump32(DumpState *s, uint32_t val);
uint64_t cpu_to_dump64(DumpState *s, uint64_t val);
+
+int64_t dump_filtered_memblock_size(GuestPhysBlock *block, int64_t filter_area_start,
+ int64_t filter_area_length);
+int64_t dump_filtered_memblock_start(GuestPhysBlock *block, int64_t filter_area_start,
+ int64_t filter_area_length);
#endif
diff --git a/target/s390x/arch_dump.c b/target/s390x/arch_dump.c
index f60a14920d..a2329141e8 100644
--- a/target/s390x/arch_dump.c
+++ b/target/s390x/arch_dump.c
@@ -12,11 +12,13 @@
*/
#include "qemu/osdep.h"
+#include "qemu/units.h"
#include "cpu.h"
#include "s390x-internal.h"
#include "elf.h"
#include "sysemu/dump.h"
-
+#include "hw/s390x/pv.h"
+#include "kvm/kvm_s390x.h"
struct S390xUserRegsStruct {
uint64_t psw[2];
@@ -76,9 +78,16 @@ typedef struct noteStruct {
uint64_t todcmp;
uint32_t todpreg;
uint64_t ctrs[16];
+ uint8_t dynamic[1]; /*
+ * Would be a flexible array member, if
+ * that was legal inside a union. Real
+ * size comes from PV info interface.
+ */
} contents;
} QEMU_PACKED Note;
+static bool pv_dump_initialized;
+
static void s390x_write_elf64_prstatus(Note *note, S390CPU *cpu, int id)
{
int i;
@@ -177,28 +186,39 @@ static void s390x_write_elf64_prefix(Note *note, S390CPU *cpu, int id)
note->contents.prefix = cpu_to_be32((uint32_t)(cpu->env.psa));
}
+static void s390x_write_elf64_pv(Note *note, S390CPU *cpu, int id)
+{
+ note->hdr.n_type = cpu_to_be32(NT_S390_PV_CPU_DATA);
+ if (!pv_dump_initialized) {
+ return;
+ }
+ kvm_s390_dump_cpu(cpu, &note->contents.dynamic);
+}
typedef struct NoteFuncDescStruct {
int contents_size;
+ uint64_t (*note_size_func)(void); /* NULL for non-dynamic sized contents */
void (*note_contents_func)(Note *note, S390CPU *cpu, int id);
+ bool pvonly;
} NoteFuncDesc;
static const NoteFuncDesc note_core[] = {
- {sizeof_field(Note, contents.prstatus), s390x_write_elf64_prstatus},
- {sizeof_field(Note, contents.fpregset), s390x_write_elf64_fpregset},
- { 0, NULL}
+ {sizeof_field(Note, contents.prstatus), NULL, s390x_write_elf64_prstatus, false},
+ {sizeof_field(Note, contents.fpregset), NULL, s390x_write_elf64_fpregset, false},
+ { 0, NULL, NULL, false}
};
static const NoteFuncDesc note_linux[] = {
- {sizeof_field(Note, contents.prefix), s390x_write_elf64_prefix},
- {sizeof_field(Note, contents.ctrs), s390x_write_elf64_ctrs},
- {sizeof_field(Note, contents.timer), s390x_write_elf64_timer},
- {sizeof_field(Note, contents.todcmp), s390x_write_elf64_todcmp},
- {sizeof_field(Note, contents.todpreg), s390x_write_elf64_todpreg},
- {sizeof_field(Note, contents.vregslo), s390x_write_elf64_vregslo},
- {sizeof_field(Note, contents.vregshi), s390x_write_elf64_vregshi},
- {sizeof_field(Note, contents.gscb), s390x_write_elf64_gscb},
- { 0, NULL}
+ {sizeof_field(Note, contents.prefix), NULL, s390x_write_elf64_prefix, false},
+ {sizeof_field(Note, contents.ctrs), NULL, s390x_write_elf64_ctrs, false},
+ {sizeof_field(Note, contents.timer), NULL, s390x_write_elf64_timer, false},
+ {sizeof_field(Note, contents.todcmp), NULL, s390x_write_elf64_todcmp, false},
+ {sizeof_field(Note, contents.todpreg), NULL, s390x_write_elf64_todpreg, false},
+ {sizeof_field(Note, contents.vregslo), NULL, s390x_write_elf64_vregslo, false},
+ {sizeof_field(Note, contents.vregshi), NULL, s390x_write_elf64_vregshi, false},
+ {sizeof_field(Note, contents.gscb), NULL, s390x_write_elf64_gscb, false},
+ {0, kvm_s390_pv_dmp_get_size_cpu, s390x_write_elf64_pv, true},
+ { 0, NULL, NULL, false}
};
static int s390x_write_elf64_notes(const char *note_name,
@@ -207,22 +227,41 @@ static int s390x_write_elf64_notes(const char *note_name,
DumpState *s,
const NoteFuncDesc *funcs)
{
- Note note;
+ Note note, *notep;
const NoteFuncDesc *nf;
- int note_size;
+ int note_size, content_size;
int ret = -1;
assert(strlen(note_name) < sizeof(note.name));
for (nf = funcs; nf->note_contents_func; nf++) {
- memset(&note, 0, sizeof(note));
- note.hdr.n_namesz = cpu_to_be32(strlen(note_name) + 1);
- note.hdr.n_descsz = cpu_to_be32(nf->contents_size);
- g_strlcpy(note.name, note_name, sizeof(note.name));
- (*nf->note_contents_func)(&note, cpu, id);
+ notep = &note;
+ if (nf->pvonly && !s390_is_pv()) {
+ continue;
+ }
+
+ content_size = nf->note_size_func ? nf->note_size_func() : nf->contents_size;
+ note_size = sizeof(note) - sizeof(notep->contents) + content_size;
+
+ /* Notes with dynamic sizes need to allocate a note */
+ if (nf->note_size_func) {
+ notep = g_malloc(note_size);
+ }
+
+ memset(notep, 0, sizeof(note));
- note_size = sizeof(note) - sizeof(note.contents) + nf->contents_size;
- ret = f(&note, note_size, s);
+ /* Setup note header data */
+ notep->hdr.n_descsz = cpu_to_be32(content_size);
+ notep->hdr.n_namesz = cpu_to_be32(strlen(note_name) + 1);
+ g_strlcpy(notep->name, note_name, sizeof(notep->name));
+
+ /* Get contents and write them out */
+ (*nf->note_contents_func)(notep, cpu, id);
+ ret = f(notep, note_size, s);
+
+ if (nf->note_size_func) {
+ g_free(notep);
+ }
if (ret < 0) {
return -1;
@@ -247,13 +286,179 @@ int s390_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
return s390x_write_elf64_notes("LINUX", f, cpu, cpuid, s, note_linux);
}
+/* PV dump section size functions */
+static uint64_t get_mem_state_size_from_len(uint64_t len)
+{
+ return (len / (MiB)) * kvm_s390_pv_dmp_get_size_mem_state();
+}
+
+static uint64_t get_size_mem_state(DumpState *s)
+{
+ return get_mem_state_size_from_len(s->total_size);
+}
+
+static uint64_t get_size_completion_data(DumpState *s)
+{
+ return kvm_s390_pv_dmp_get_size_completion_data();
+}
+
+/* PV dump section data functions*/
+static int get_data_completion(DumpState *s, uint8_t *buff)
+{
+ int rc;
+
+ if (!pv_dump_initialized) {
+ return 0;
+ }
+ rc = kvm_s390_dump_completion_data(buff);
+ if (!rc) {
+ pv_dump_initialized = false;
+ }
+ return rc;
+}
+
+static int get_mem_state(DumpState *s, uint8_t *buff)
+{
+ int64_t memblock_size, memblock_start;
+ GuestPhysBlock *block;
+ uint64_t off;
+ int rc;
+
+ QTAILQ_FOREACH(block, &s->guest_phys_blocks.head, next) {
+ memblock_start = dump_filtered_memblock_start(block, s->filter_area_begin,
+ s->filter_area_length);
+ if (memblock_start == -1) {
+ continue;
+ }
+
+ memblock_size = dump_filtered_memblock_size(block, s->filter_area_begin,
+ s->filter_area_length);
+
+ off = get_mem_state_size_from_len(block->target_start);
+
+ rc = kvm_s390_dump_mem_state(block->target_start,
+ get_mem_state_size_from_len(memblock_size),
+ buff + off);
+ if (rc) {
+ return rc;
+ }
+ }
+
+ return 0;
+}
+
+static struct sections {
+ uint64_t (*sections_size_func)(DumpState *s);
+ int (*sections_contents_func)(DumpState *s, uint8_t *buff);
+ char sctn_str[12];
+} sections[] = {
+ { get_size_mem_state, get_mem_state, "pv_mem_meta"},
+ { get_size_completion_data, get_data_completion, "pv_compl"},
+ {NULL , NULL, ""}
+};
+
+static uint64_t arch_sections_write_hdr(DumpState *s, uint8_t *buff)
+{
+ Elf64_Shdr *shdr = (void *)buff;
+ struct sections *sctn = sections;
+ uint64_t off = s->section_offset;
+
+ if (!pv_dump_initialized) {
+ return 0;
+ }
+
+ for (; sctn->sections_size_func; off += shdr->sh_size, sctn++, shdr++) {
+ memset(shdr, 0, sizeof(*shdr));
+ shdr->sh_type = SHT_PROGBITS;
+ shdr->sh_offset = off;
+ shdr->sh_size = sctn->sections_size_func(s);
+ shdr->sh_name = s->string_table_buf->len;
+ g_array_append_vals(s->string_table_buf, sctn->sctn_str, sizeof(sctn->sctn_str));
+ }
+
+ return (uintptr_t)shdr - (uintptr_t)buff;
+}
+
+
+/* Add arch specific number of sections and their respective sizes */
+static void arch_sections_add(DumpState *s)
+{
+ struct sections *sctn = sections;
+
+ /*
+ * We only do a PV dump if we are running a PV guest, KVM supports
+ * the dump API and we got valid dump length information.
+ */
+ if (!s390_is_pv() || !kvm_s390_get_protected_dump() ||
+ !kvm_s390_pv_info_basic_valid()) {
+ return;
+ }
+
+ /*
+ * Start the UV dump process by doing the initialize dump call via
+ * KVM as the proxy.
+ */
+ if (!kvm_s390_dump_init()) {
+ pv_dump_initialized = true;
+ } else {
+ /*
+ * Dump init failed, maybe the guest owner disabled dumping.
+ * We'll continue the non-PV dump process since this is no
+ * reason to crash qemu.
+ */
+ return;
+ }
+
+ for (; sctn->sections_size_func; sctn++) {
+ s->shdr_num += 1;
+ s->elf_section_data_size += sctn->sections_size_func(s);
+ }
+}
+
+/*
+ * After the PV dump has been initialized, the CPU data has been
+ * fetched and memory has been dumped, we need to grab the tweak data
+ * and the completion data.
+ */
+static int arch_sections_write(DumpState *s, uint8_t *buff)
+{
+ struct sections *sctn = sections;
+ int rc;
+
+ if (!pv_dump_initialized) {
+ return -EINVAL;
+ }
+
+ for (; sctn->sections_size_func; sctn++) {
+ rc = sctn->sections_contents_func(s, buff);
+ buff += sctn->sections_size_func(s);
+ if (rc) {
+ return rc;
+ }
+ }
+ return 0;
+}
+
int cpu_get_dump_info(ArchDumpInfo *info,
const struct GuestPhysBlockList *guest_phys_blocks)
{
info->d_machine = EM_S390;
info->d_endian = ELFDATA2MSB;
info->d_class = ELFCLASS64;
-
+ /*
+ * This is evaluated for each dump so we can freely switch
+ * between PV and non-PV.
+ */
+ if (s390_is_pv() && kvm_s390_get_protected_dump() &&
+ kvm_s390_pv_info_basic_valid()) {
+ info->arch_sections_add_fn = *arch_sections_add;
+ info->arch_sections_write_hdr_fn = *arch_sections_write_hdr;
+ info->arch_sections_write_fn = *arch_sections_write;
+ } else {
+ info->arch_sections_add_fn = NULL;
+ info->arch_sections_write_hdr_fn = NULL;
+ info->arch_sections_write_fn = NULL;
+ }
return 0;
}
@@ -261,7 +466,7 @@ ssize_t cpu_get_note_size(int class, int machine, int nr_cpus)
{
int name_size = 8; /* "LINUX" or "CORE" + pad */
size_t elf_note_size = 0;
- int note_head_size;
+ int note_head_size, content_size;
const NoteFuncDesc *nf;
assert(class == ELFCLASS64);
@@ -270,12 +475,15 @@ ssize_t cpu_get_note_size(int class, int machine, int nr_cpus)
note_head_size = sizeof(Elf64_Nhdr);
for (nf = note_core; nf->note_contents_func; nf++) {
- elf_note_size = elf_note_size + note_head_size + name_size +
- nf->contents_size;
+ elf_note_size = elf_note_size + note_head_size + name_size + nf->contents_size;
}
for (nf = note_linux; nf->note_contents_func; nf++) {
+ if (nf->pvonly && !s390_is_pv()) {
+ continue;
+ }
+ content_size = nf->contents_size ? nf->contents_size : nf->note_size_func();
elf_note_size = elf_note_size + note_head_size + name_size +
- nf->contents_size;
+ content_size;
}
return (elf_note_size) * nr_cpus;
--
2.37.3

View File

@ -0,0 +1,55 @@
From c7b14d3af7c73a3bf0c00911b85f202840d9b414 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?C=C3=A9dric=20Le=20Goater?= <clg@redhat.com>
Date: Mon, 7 Nov 2022 17:13:49 +0100
Subject: [PATCH 12/42] s390x/s390-virtio-ccw: Switch off zPCI enhancements on
older machines
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 226: s390: Enhanced Interpretation for PCI Functions and Secure Execution guest dump
RH-Bugzilla: 1664378 2043909
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Jon Maloy <jmaloy@redhat.com>
RH-Commit: [12/41] 61e32bab6d68ee9abd6a0751944af82e002b05b4
zPCI enhancement features (interpretation and forward assist) were
recently introduced to improve performance on PCI passthrough devices.
To maintain the same behaviour on older Z machines, deactivate the
features with the associated properties.
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Message-Id: <20221107161349.1032730-3-clg@kaod.org>
Reviewed-by: Matthew Rosato <mjrosato@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit d3d1a406127f7da482eafbdc871c120c2770bb91)
[ clg: applied zPCI restrictions to rhel8.5.0 machine and below ]
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/s390x/s390-virtio-ccw.c | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/hw/s390x/s390-virtio-ccw.c b/hw/s390x/s390-virtio-ccw.c
index bec270598b..bd80e72cf8 100644
--- a/hw/s390x/s390-virtio-ccw.c
+++ b/hw/s390x/s390-virtio-ccw.c
@@ -1130,8 +1130,14 @@ static void ccw_machine_rhel850_instance_options(MachineState *machine)
static void ccw_machine_rhel850_class_options(MachineClass *mc)
{
+ static GlobalProperty compat[] = {
+ { TYPE_S390_PCI_DEVICE, "interpret", "off", },
+ { TYPE_S390_PCI_DEVICE, "forwarding-assist", "off", },
+ };
+
ccw_machine_rhel860_class_options(mc);
compat_props_add(mc->compat_props, hw_compat_rhel_8_5, hw_compat_rhel_8_5_len);
+ compat_props_add(mc->compat_props, compat, G_N_ELEMENTS(compat));
mc->smp_props.prefer_sockets = true;
}
DEFINE_CCW_MACHINE(rhel850, "rhel8.5.0", false);
--
2.37.3

View File

@ -0,0 +1,70 @@
From 90fcfe4b07afc5299c5ee69fa663ca46b597fd4b Mon Sep 17 00:00:00 2001
From: Nico Boehr <nrb@linux.ibm.com>
Date: Wed, 12 Oct 2022 14:32:29 +0200
Subject: [PATCH] s390x/tod-kvm: don't save/restore the TOD in PV guests
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <clg@redhat.com>
RH-MergeRequest: 244: s390x/tod-kvm: don't save/restore the TOD in PV guests
RH-Bugzilla: 2155448
RH-Acked-by: David Hildenbrand <david@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Commit: [1/1] 3cb3154dd7c1549c54cf8c0483b5f23b235f6db3
Under PV, the guest's TOD clock is under control of the ultravisor and the
hypervisor cannot change it.
With upcoming kernel changes[1], the Linux kernel will reject QEMU's
request to adjust the guest's clock in this case, so don't attempt to set
the clock.
This avoids the following warning message on save/restore of a PV guest:
warning: Unable to set KVM guest TOD clock: Operation not supported
[1] https://lore.kernel.org/all/20221011160712.928239-2-nrb@linux.ibm.com/
Fixes: c3347ed0d2ee ("s390x: protvirt: Support unpack facility")
Signed-off-by: Nico Boehr <nrb@linux.ibm.com>
Message-Id: <20221012123229.1196007-1-nrb@linux.ibm.com>
[thuth: Add curly braces]
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit 38621181ae3cbec62e3490fbc14f6ac01642d07a)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
hw/s390x/tod-kvm.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/hw/s390x/tod-kvm.c b/hw/s390x/tod-kvm.c
index ec855811ae..c804c979b5 100644
--- a/hw/s390x/tod-kvm.c
+++ b/hw/s390x/tod-kvm.c
@@ -13,6 +13,7 @@
#include "qemu/module.h"
#include "sysemu/runstate.h"
#include "hw/s390x/tod.h"
+#include "hw/s390x/pv.h"
#include "kvm/kvm_s390x.h"
static void kvm_s390_get_tod_raw(S390TOD *tod, Error **errp)
@@ -84,6 +85,14 @@ static void kvm_s390_tod_vm_state_change(void *opaque, bool running,
S390TODState *td = opaque;
Error *local_err = NULL;
+ /*
+ * Under PV, the clock is under ultravisor control, hence we cannot restore
+ * it on resume.
+ */
+ if (s390_is_pv()) {
+ return;
+ }
+
if (running && td->stopped) {
/* Set the old TOD when running the VM - start the TOD clock. */
kvm_s390_set_tod_raw(&td->base, &local_err);
--
2.37.3

View File

@ -0,0 +1,65 @@
From 3014c7c11b6e64433fe9f3c463bd91e318ac96b6 Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@redhat.com>
Date: Wed, 9 Nov 2022 18:41:18 -0500
Subject: [PATCH 2/2] target/arm/kvm: Retry KVM_CREATE_VM call if it fails
EINTR
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 228: qemu-kvm: backport some aarch64 fixes
RH-Bugzilla: 2132609
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Igor Mammedov <imammedo@redhat.com>
RH-Acked-by: Eric Auger <eric.auger@redhat.com>
RH-Acked-by: Gavin Shan <gshan@redhat.com>
RH-Commit: [2/2] 8494bbfb3fcd8693f56312f984d2964d1ca275c2 (jmaloy/qemu-kvm)
BZ: https://bugzilla.redhat.com/show_bug.cgi?id=2132609
Upstream: Merged
commit bbde13cd14ad4eec18529ce0bf5876058464e124
Author: Peter Maydell <peter.maydell@linaro.org>
Date: Fri Sep 30 12:38:24 2022 +0100
target/arm/kvm: Retry KVM_CREATE_VM call if it fails EINTR
Occasionally the KVM_CREATE_VM ioctl can return EINTR, even though
there is no pending signal to be taken. In commit 94ccff13382055
we added a retry-on-EINTR loop to the KVM_CREATE_VM call in the
generic KVM code. Adopt the same approach for the use of the
ioctl in the Arm-specific KVM code (where we use it to create a
scratch VM for probing for various things).
For more information, see the mailing list thread:
https://lore.kernel.org/qemu-devel/8735e0s1zw.wl-maz@kernel.org/
Reported-by: Vitaly Chikunov <vt@altlinux.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Vitaly Chikunov <vt@altlinux.org>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Acked-by: Marc Zyngier <maz@kernel.org>
Message-id: 20220930113824.1933293-1-peter.maydell@linaro.org
(cherry picked from commit bbde13cd14ad4eec18529ce0bf5876058464e124)
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
target/arm/kvm.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/target/arm/kvm.c b/target/arm/kvm.c
index bbf1ce7ba3..1ae4e51055 100644
--- a/target/arm/kvm.c
+++ b/target/arm/kvm.c
@@ -80,7 +80,9 @@ bool kvm_arm_create_scratch_host_vcpu(const uint32_t *cpus_to_try,
if (max_vm_pa_size < 0) {
max_vm_pa_size = 0;
}
- vmfd = ioctl(kvmfd, KVM_CREATE_VM, max_vm_pa_size);
+ do {
+ vmfd = ioctl(kvmfd, KVM_CREATE_VM, max_vm_pa_size);
+ } while (vmfd == -1 && errno == EINTR);
if (vmfd < 0) {
goto err;
}
--
2.37.3

View File

@ -0,0 +1,106 @@
From 8d1a60069cddcc69ef1a6f50f2b55343de348b57 Mon Sep 17 00:00:00 2001
From: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Date: Fri, 6 May 2022 17:39:56 +0200
Subject: [PATCH 2/3] target/s390x: kvm: Honor storage keys during emulation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Cédric Le Goater <None>
RH-MergeRequest: 220: s390x: Fix skey test in kvm_unit_test
RH-Bugzilla: 2124757
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: David Hildenbrand <david@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Commit: [2/2] 980dbb4eba8d2f1da7cf4113230d0a6483cffc4f
Storage key controlled protection is currently not honored when
emulating instructions.
If available, enable key protection for the MEM_OP ioctl, thereby
enabling it for the s390_cpu_virt_mem_* functions, when using kvm.
As a result, the emulation of the following instructions honors storage
keys:
* CLP
The Synch I/O CLP command would need special handling in order
to support storage keys, but is currently not supported.
* CHSC
Performing commands asynchronously would require special
handling, but commands are currently always synchronous.
* STSI
* TSCH
Must (and does) not change channel if terminated due to
protection.
* MSCH
Suppressed on protection, works because fetching instruction.
* SSCH
Suppressed on protection, works because fetching instruction.
* STSCH
* STCRW
Suppressed on protection, this works because no partial store is
possible, because the operand cannot span multiple pages.
* PCISTB
* MPCIFC
* STPCIFC
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2124757
Signed-off-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com>
Message-Id: <20220506153956.2217601-3-scgl@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
(cherry picked from commit 54354861d21b69ec0781f43e67b8d4f6edad7e3f)
Signed-off-by: Cédric Le Goater <clg@redhat.com>
---
target/s390x/kvm/kvm.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/target/s390x/kvm/kvm.c b/target/s390x/kvm/kvm.c
index c52434985b..ba04997da1 100644
--- a/target/s390x/kvm/kvm.c
+++ b/target/s390x/kvm/kvm.c
@@ -152,12 +152,15 @@ const KVMCapabilityInfo kvm_arch_required_capabilities[] = {
static int cap_sync_regs;
static int cap_async_pf;
static int cap_mem_op;
+static int cap_mem_op_extension;
static int cap_s390_irq;
static int cap_ri;
static int cap_hpage_1m;
static int cap_vcpu_resets;
static int cap_protected;
+static bool mem_op_storage_key_support;
+
static int active_cmma;
static int kvm_s390_query_mem_limit(uint64_t *memory_limit)
@@ -355,6 +358,8 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
cap_sync_regs = kvm_check_extension(s, KVM_CAP_SYNC_REGS);
cap_async_pf = kvm_check_extension(s, KVM_CAP_ASYNC_PF);
cap_mem_op = kvm_check_extension(s, KVM_CAP_S390_MEM_OP);
+ cap_mem_op_extension = kvm_check_extension(s, KVM_CAP_S390_MEM_OP_EXTENSION);
+ mem_op_storage_key_support = cap_mem_op_extension > 0;
cap_s390_irq = kvm_check_extension(s, KVM_CAP_S390_INJECT_IRQ);
cap_vcpu_resets = kvm_check_extension(s, KVM_CAP_S390_VCPU_RESETS);
cap_protected = kvm_check_extension(s, KVM_CAP_S390_PROTECTED);
@@ -843,6 +848,7 @@ int kvm_s390_mem_op(S390CPU *cpu, vaddr addr, uint8_t ar, void *hostbuf,
: KVM_S390_MEMOP_LOGICAL_READ,
.buf = (uint64_t)hostbuf,
.ar = ar,
+ .key = (cpu->env.psw.mask & PSW_MASK_KEY) >> PSW_SHIFT_KEY,
};
int ret;
@@ -852,6 +858,9 @@ int kvm_s390_mem_op(S390CPU *cpu, vaddr addr, uint8_t ar, void *hostbuf,
if (!hostbuf) {
mem_op.flags |= KVM_S390_MEMOP_F_CHECK_ONLY;
}
+ if (mem_op_storage_key_support) {
+ mem_op.flags |= KVM_S390_MEMOP_F_SKEY_PROTECTION;
+ }
ret = kvm_vcpu_ioctl(CPU(cpu), KVM_S390_MEM_OP, &mem_op);
if (ret < 0) {
--
2.35.3

View File

@ -0,0 +1,80 @@
From d3602e5afa1e90c5e33625fc528db7f96195bada Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@redhat.com>
Date: Mon, 7 Nov 2022 19:59:46 -0500
Subject: [PATCH 42/42] ui/vnc-clipboard: fix integer underflow in
vnc_client_cut_text_ext
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 227: ui/vnc-clipboard: fix integer underflow in vnc_client_cut_text_ext
RH-Bugzilla: 2129760
RH-Acked-by: Mauro Matteo Cascella <None>
RH-Acked-by: Marc-André Lureau <marcandre.lureau@redhat.com>
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Gerd Hoffmann <kraxel@redhat.com>
RH-Commit: [1/1] ac19a6c0777e308061bcb6d1de5cc9beaa105a3a (jmaloy/qemu-kvm)
BZ: https://bugzilla.redhat.com/show_bug.cgi?id=2129760
CVE: CVE-2022-3165
Upstream: Merged
commit d307040b18bfcb1393b910f1bae753d5c12a4dc7
Author: Mauro Matteo Cascella <mcascell@redhat.com>
Date: Sun Sep 25 22:45:11 2022 +0200
ui/vnc-clipboard: fix integer underflow in vnc_client_cut_text_ext
Extended ClientCutText messages start with a 4-byte header. If len < 4,
an integer underflow occurs in vnc_client_cut_text_ext. The result is
used to decompress data in a while loop in inflate_buffer, leading to
CPU consumption and denial of service. Prevent this by checking dlen in
protocol_client_msg.
Fixes: CVE-2022-3165
Fixes: 0bf41cab93e5 ("ui/vnc: clipboard support")
Reported-by: TangPeng <tangpeng@qianxin.com>
Signed-off-by: Mauro Matteo Cascella <mcascell@redhat.com>
Message-Id: <20220925204511.1103214-1-mcascell@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit d307040b18bfcb1393b910f1bae753d5c12a4dc7)
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
ui/vnc.c | 11 ++++++++---
1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/ui/vnc.c b/ui/vnc.c
index af02522e84..a14b6861be 100644
--- a/ui/vnc.c
+++ b/ui/vnc.c
@@ -2442,8 +2442,8 @@ static int protocol_client_msg(VncState *vs, uint8_t *data, size_t len)
if (len == 1) {
return 8;
}
+ uint32_t dlen = abs(read_s32(data, 4));
if (len == 8) {
- uint32_t dlen = abs(read_s32(data, 4));
if (dlen > (1 << 20)) {
error_report("vnc: client_cut_text msg payload has %u bytes"
" which exceeds our limit of 1MB.", dlen);
@@ -2456,8 +2456,13 @@ static int protocol_client_msg(VncState *vs, uint8_t *data, size_t len)
}
if (read_s32(data, 4) < 0) {
- vnc_client_cut_text_ext(vs, abs(read_s32(data, 4)),
- read_u32(data, 8), data + 12);
+ if (dlen < 4) {
+ error_report("vnc: malformed payload (header less than 4 bytes)"
+ " in extended clipboard pseudo-encoding.");
+ vnc_client_error(vs);
+ break;
+ }
+ vnc_client_cut_text_ext(vs, dlen, read_u32(data, 8), data + 12);
break;
}
vnc_client_cut_text(vs, read_u32(data, 4), data + 8);
--
2.37.3

View File

@ -0,0 +1,90 @@
From ffdf44cb8b4c743e7ab0ce46d62687d7178f9a49 Mon Sep 17 00:00:00 2001
From: Jon Maloy <jmaloy@redhat.com>
Date: Fri, 18 Nov 2022 20:26:54 -0500
Subject: [PATCH 3/3] ui/vnc.c: Fixed a deadlock bug.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Jon Maloy <jmaloy@redhat.com>
RH-MergeRequest: 234: ui/vnc.c: Fixed a deadlock bug.
RH-Bugzilla: 2141896
RH-Acked-by: Gerd Hoffmann <kraxel@redhat.com>
RH-Acked-by: Marc-André Lureau <marcandre.lureau@redhat.com>
RH-Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-Commit: [1/1] d3d1d28d7b621a8ae8a593a5bd5303fa7951c17c (jmaloy/qemu-kvm)
BZ: https://bugzilla.redhat.com/show_bug.cgi?id=2141896
Upstream: Merged
commit 1dbbe6f172810026c51dc84ed927a3cc23017949
Author: Rao Lei <lei.rao@intel.com>
Date: Wed Jan 5 10:08:08 2022 +0800
ui/vnc.c: Fixed a deadlock bug.
The GDB statck is as follows:
(gdb) bt
0 __lll_lock_wait (futex=futex@entry=0x56211df20360, private=0) at lowlevellock.c:52
1 0x00007f263caf20a3 in __GI___pthread_mutex_lock (mutex=0x56211df20360) at ../nptl/pthread_mutex_lock.c:80
2 0x000056211a757364 in qemu_mutex_lock_impl (mutex=0x56211df20360, file=0x56211a804857 "../ui/vnc-jobs.h", line=60)
at ../util/qemu-thread-posix.c:80
3 0x000056211a0ef8c7 in vnc_lock_output (vs=0x56211df14200) at ../ui/vnc-jobs.h:60
4 0x000056211a0efcb7 in vnc_clipboard_send (vs=0x56211df14200, count=1, dwords=0x7ffdf1701338) at ../ui/vnc-clipboard.c:138
5 0x000056211a0f0129 in vnc_clipboard_notify (notifier=0x56211df244c8, data=0x56211dd1bbf0) at ../ui/vnc-clipboard.c:209
6 0x000056211a75dde8 in notifier_list_notify (list=0x56211afa17d0 <clipboard_notifiers>, data=0x56211dd1bbf0) at ../util/notify.c:39
7 0x000056211a0bf0e6 in qemu_clipboard_update (info=0x56211dd1bbf0) at ../ui/clipboard.c:50
8 0x000056211a0bf05d in qemu_clipboard_peer_release (peer=0x56211df244c0, selection=QEMU_CLIPBOARD_SELECTION_CLIPBOARD)
at ../ui/clipboard.c:41
9 0x000056211a0bef9b in qemu_clipboard_peer_unregister (peer=0x56211df244c0) at ../ui/clipboard.c:19
10 0x000056211a0d45f3 in vnc_disconnect_finish (vs=0x56211df14200) at ../ui/vnc.c:1358
11 0x000056211a0d4c9d in vnc_client_read (vs=0x56211df14200) at ../ui/vnc.c:1611
12 0x000056211a0d4df8 in vnc_client_io (ioc=0x56211ce70690, condition=G_IO_IN, opaque=0x56211df14200) at ../ui/vnc.c:1649
13 0x000056211a5b976c in qio_channel_fd_source_dispatch
(source=0x56211ce50a00, callback=0x56211a0d4d71 <vnc_client_io>, user_data=0x56211df14200) at ../io/channel-watch.c:84
14 0x00007f263ccede8e in g_main_context_dispatch () at /lib/x86_64-linux-gnu/libglib-2.0.so.0
15 0x000056211a77d4a1 in glib_pollfds_poll () at ../util/main-loop.c:232
16 0x000056211a77d51f in os_host_main_loop_wait (timeout=958545) at ../util/main-loop.c:255
17 0x000056211a77d630 in main_loop_wait (nonblocking=0) at ../util/main-loop.c:531
18 0x000056211a45bc8e in qemu_main_loop () at ../softmmu/runstate.c:726
19 0x000056211a0b45fa in main (argc=69, argv=0x7ffdf1701778, envp=0x7ffdf17019a8) at ../softmmu/main.c:50
From the call trace, we can see it is a deadlock bug.
vnc_disconnect_finish will acquire the output_mutex.
But, the output_mutex will be acquired again in vnc_clipboard_send.
Repeated locking will cause deadlock. So, I move
qemu_clipboard_peer_unregister() behind vnc_unlock_output();
Fixes: 0bf41cab93e ("ui/vnc: clipboard support")
Signed-off-by: Lei Rao <lei.rao@intel.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20220105020808.597325-1-lei.rao@intel.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
(cherry picked from commit 1dbbe6f172810026c51dc84ed927a3cc23017949)
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
ui/vnc.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/ui/vnc.c b/ui/vnc.c
index a14b6861be..76372ca1de 100644
--- a/ui/vnc.c
+++ b/ui/vnc.c
@@ -1354,12 +1354,12 @@ void vnc_disconnect_finish(VncState *vs)
/* last client gone */
vnc_update_server_surface(vs->vd);
}
+ vnc_unlock_output(vs);
+
if (vs->cbpeer.update.notify) {
qemu_clipboard_peer_unregister(&vs->cbpeer);
}
- vnc_unlock_output(vs);
-
qemu_mutex_destroy(&vs->output_mutex);
if (vs->bh != NULL) {
qemu_bh_delete(vs->bh);
--
2.37.3

View File

@ -0,0 +1,102 @@
From 56e2aef97e750ffdc572dcecbfc31314728d37a9 Mon Sep 17 00:00:00 2001
From: Halil Pasic <pasic@linux.ibm.com>
Date: Mon, 7 Mar 2022 12:29:39 +0100
Subject: [PATCH 2/2] virtio: fix feature negotiation for ACCESS_PLATFORM
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-MergeRequest: 224: virtiofs on s390 secure execution
RH-Bugzilla: 2116302
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Cédric Le Goater <None>
RH-Commit: [2/2] 264d3bdbbde985f16ed6f5a1786547c25fb8cc04
Unlike most virtio features ACCESS_PLATFORM is considered mandatory by
QEMU, i.e. the driver must accept it if offered by the device. The
virtio specification says that the driver SHOULD accept the
ACCESS_PLATFORM feature if offered, and that the device MAY fail to
operate if ACCESS_PLATFORM was offered but not negotiated.
While a SHOULD ain't exactly a MUST, we are certainly allowed to fail
the device when the driver fences ACCESS_PLATFORM. With commit
2943b53f68 ("virtio: force VIRTIO_F_IOMMU_PLATFORM") we already made the
decision to do so whenever the get_dma_as() callback is implemented (by
the bus), which in practice means for the entirety of virtio-pci.
That means, if the device needs to translate I/O addresses, then
ACCESS_PLATFORM is mandatory. The aforementioned commit tells us in the
commit message that this is for security reasons. More precisely if we
were to allow a less then trusted driver (e.g. an user-space driver, or
a nested guest) to make the device bypass the IOMMU by not negotiating
ACCESS_PLATFORM, then the guest kernel would have no ability to
control/police (by programming the IOMMU) what pieces of guest memory
the driver may manipulate using the device. Which would break security
assumptions within the guest.
If ACCESS_PLATFORM is offered not because we want the device to utilize
an IOMMU and do address translation, but because the device does not
have access to the entire guest RAM, and needs the driver to grant
access to the bits it needs access to (e.g. confidential guest support),
we still require the guest to have the corresponding logic and to accept
ACCESS_PLATFORM. If the driver does not accept ACCESS_PLATFORM, then
things are bound to go wrong, and we may see failures much less graceful
than failing the device because the driver didn't negotiate
ACCESS_PLATFORM.
So let us make ACCESS_PLATFORM mandatory for the driver regardless
of whether the get_dma_as() callback is implemented or not.
Signed-off-by: Halil Pasic <pasic@linux.ibm.com>
Fixes: 2943b53f68 ("virtio: force VIRTIO_F_IOMMU_PLATFORM")
Message-Id: <20220307112939.2780117-1-pasic@linux.ibm.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
(cherry picked from commit 06134e2bc35dc21543d4cbcf31f858c03d383442)
---
hw/virtio/virtio-bus.c | 22 ++++++++++++++--------
1 file changed, 14 insertions(+), 8 deletions(-)
diff --git a/hw/virtio/virtio-bus.c b/hw/virtio/virtio-bus.c
index 0f69d1c742..d7ec023adf 100644
--- a/hw/virtio/virtio-bus.c
+++ b/hw/virtio/virtio-bus.c
@@ -78,17 +78,23 @@ void virtio_bus_device_plugged(VirtIODevice *vdev, Error **errp)
return;
}
- vdev_has_iommu = virtio_host_has_feature(vdev, VIRTIO_F_IOMMU_PLATFORM);
- if (klass->get_dma_as != NULL && has_iommu) {
+ vdev->dma_as = &address_space_memory;
+ if (has_iommu) {
+ vdev_has_iommu = virtio_host_has_feature(vdev, VIRTIO_F_IOMMU_PLATFORM);
+ /*
+ * Present IOMMU_PLATFORM to the driver iff iommu_plattform=on and
+ * device operational. If the driver does not accept IOMMU_PLATFORM
+ * we fail the device.
+ */
virtio_add_feature(&vdev->host_features, VIRTIO_F_IOMMU_PLATFORM);
- vdev->dma_as = klass->get_dma_as(qbus->parent);
- if (!vdev_has_iommu && vdev->dma_as != &address_space_memory) {
- error_setg(errp,
+ if (klass->get_dma_as) {
+ vdev->dma_as = klass->get_dma_as(qbus->parent);
+ if (!vdev_has_iommu && vdev->dma_as != &address_space_memory) {
+ error_setg(errp,
"iommu_platform=true is not supported by the device");
- return;
+ return;
+ }
}
- } else {
- vdev->dma_as = &address_space_memory;
}
}
--
2.37.3

View File

@ -0,0 +1,115 @@
From c731ffdf9faee74e9522dff06e61cda817902088 Mon Sep 17 00:00:00 2001
From: Halil Pasic <pasic@linux.ibm.com>
Date: Mon, 7 Feb 2022 12:28:57 +0100
Subject: [PATCH 1/2] virtio: fix the condition for iommu_platform not
supported
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-MergeRequest: 224: virtiofs on s390 secure execution
RH-Bugzilla: 2116302
RH-Acked-by: Thomas Huth <thuth@redhat.com>
RH-Acked-by: Cornelia Huck <cohuck@redhat.com>
RH-Acked-by: Cédric Le Goater <None>
RH-Commit: [1/2] d7edc7e3905a04644c9ff44b0d36122c72068e08
The commit 04ceb61a40 ("virtio: Fail if iommu_platform is requested, but
unsupported") claims to fail the device hotplug when iommu_platform
is requested, but not supported by the (vhost) device. On the first
glance the condition for detecting that situation looks perfect, but
because a certain peculiarity of virtio_platform it ain't.
In fact the aforementioned commit introduces a regression. It breaks
virtio-fs support for Secure Execution, and most likely also for AMD SEV
or any other confidential guest scenario that relies encrypted guest
memory. The same also applies to any other vhost device that does not
support _F_ACCESS_PLATFORM.
The peculiarity is that iommu_platform and _F_ACCESS_PLATFORM collates
"device can not access all of the guest RAM" and "iova != gpa, thus
device needs to translate iova".
Confidential guest technologies currently rely on the device/hypervisor
offering _F_ACCESS_PLATFORM, so that, after the feature has been
negotiated, the guest grants access to the portions of memory the
device needs to see. So in for confidential guests, generally,
_F_ACCESS_PLATFORM is about the restricted access to memory, but not
about the addresses used being something else than guest physical
addresses.
This is the very reason for which commit f7ef7e6e3b ("vhost: correctly
turn on VIRTIO_F_IOMMU_PLATFORM") fences _F_ACCESS_PLATFORM from the
vhost device that does not need it, because on the vhost interface it
only means "I/O address translation is needed".
This patch takes inspiration from f7ef7e6e3b ("vhost: correctly turn on
VIRTIO_F_IOMMU_PLATFORM"), and uses the same condition for detecting the
situation when _F_ACCESS_PLATFORM is requested, but no I/O translation
by the device, and thus no device capability is needed. In this
situation claiming that the device does not support iommu_plattform=on
is counter-productive. So let us stop doing that!
Signed-off-by: Halil Pasic <pasic@linux.ibm.com>
Reported-by: Jakob Naucke <Jakob.Naucke@ibm.com>
Fixes: 04ceb61a40 ("virtio: Fail if iommu_platform is requested, but
unsupported")
Acked-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Tested-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Cc: Kevin Wolf <kwolf@redhat.com>
Cc: qemu-stable@nongnu.org
Message-Id: <20220207112857.607829-1-pasic@linux.ibm.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
(cherry picked from commit e65902a913bf31ba79a83a3bd3621108b85cf645)
---
hw/virtio/virtio-bus.c | 12 +++++++-----
1 file changed, 7 insertions(+), 5 deletions(-)
diff --git a/hw/virtio/virtio-bus.c b/hw/virtio/virtio-bus.c
index d23db98c56..0f69d1c742 100644
--- a/hw/virtio/virtio-bus.c
+++ b/hw/virtio/virtio-bus.c
@@ -48,6 +48,7 @@ void virtio_bus_device_plugged(VirtIODevice *vdev, Error **errp)
VirtioBusClass *klass = VIRTIO_BUS_GET_CLASS(bus);
VirtioDeviceClass *vdc = VIRTIO_DEVICE_GET_CLASS(vdev);
bool has_iommu = virtio_host_has_feature(vdev, VIRTIO_F_IOMMU_PLATFORM);
+ bool vdev_has_iommu;
Error *local_err = NULL;
DPRINTF("%s: plug device.\n", qbus->name);
@@ -69,11 +70,6 @@ void virtio_bus_device_plugged(VirtIODevice *vdev, Error **errp)
return;
}
- if (has_iommu && !virtio_host_has_feature(vdev, VIRTIO_F_IOMMU_PLATFORM)) {
- error_setg(errp, "iommu_platform=true is not supported by the device");
- return;
- }
-
if (klass->device_plugged != NULL) {
klass->device_plugged(qbus->parent, &local_err);
}
@@ -82,9 +78,15 @@ void virtio_bus_device_plugged(VirtIODevice *vdev, Error **errp)
return;
}
+ vdev_has_iommu = virtio_host_has_feature(vdev, VIRTIO_F_IOMMU_PLATFORM);
if (klass->get_dma_as != NULL && has_iommu) {
virtio_add_feature(&vdev->host_features, VIRTIO_F_IOMMU_PLATFORM);
vdev->dma_as = klass->get_dma_as(qbus->parent);
+ if (!vdev_has_iommu && vdev->dma_as != &address_space_memory) {
+ error_setg(errp,
+ "iommu_platform=true is not supported by the device");
+ return;
+ }
} else {
vdev->dma_as = &address_space_memory;
}
--
2.37.3

View File

@ -0,0 +1,65 @@
From ebf6be5ba316ffda354af5eb1f1241ad6543b3cd Mon Sep 17 00:00:00 2001
From: Yusuke Okada <okada.yusuke@jp.fujitsu.com>
Date: Thu, 18 Aug 2022 14:46:19 -0400
Subject: [PATCH 3/3] virtiofsd: use g_date_time_get_microsecond to get
subsecond
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
RH-Author: Dr. David Alan Gilbert <dgilbert@redhat.com>
RH-MergeRequest: 222: virtiofsd: use g_date_time_get_microsecond to get subsecond
RH-Bugzilla: 2018885
RH-Acked-by: Vivek Goyal <None>
RH-Acked-by: Daniel P. Berrangé <berrange@redhat.com>
RH-Acked-by: Sergio Lopez <slp@redhat.com>
RH-Commit: [1/1] da8795576acc7029044a801ef42676d66471a577
The "%f" specifier in g_date_time_format() is only available in glib
2.65.2 or later. If combined with older glib, the function returns null
and the timestamp displayed as "(null)".
For backward compatibility, g_date_time_get_microsecond should be used
to retrieve subsecond.
In this patch the g_date_time_format() leaves subsecond field as "%06d"
and let next snprintf to format with g_date_time_get_microsecond.
Signed-off-by: Yusuke Okada <okada.yusuke@jp.fujitsu.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-id: 20220818184618.2205172-1-yokada.996@gmail.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
(cherry picked from commit f16d15c9276bd8f501f861c39cbd4adc812d0c1d)
---
tools/virtiofsd/passthrough_ll.c | 7 +++++--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/tools/virtiofsd/passthrough_ll.c b/tools/virtiofsd/passthrough_ll.c
index b3d0674f6d..523d8fbe1e 100644
--- a/tools/virtiofsd/passthrough_ll.c
+++ b/tools/virtiofsd/passthrough_ll.c
@@ -3791,6 +3791,7 @@ static void setup_nofile_rlimit(unsigned long rlimit_nofile)
static void log_func(enum fuse_log_level level, const char *fmt, va_list ap)
{
g_autofree char *localfmt = NULL;
+ char buf[64];
if (current_log_level < level) {
return;
@@ -3803,9 +3804,11 @@ static void log_func(enum fuse_log_level level, const char *fmt, va_list ap)
fmt);
} else {
g_autoptr(GDateTime) now = g_date_time_new_now_utc();
- g_autofree char *nowstr = g_date_time_format(now, "%Y-%m-%d %H:%M:%S.%f%z");
+ g_autofree char *nowstr = g_date_time_format(now,
+ "%Y-%m-%d %H:%M:%S.%%06d%z");
+ snprintf(buf, 64, nowstr, g_date_time_get_microsecond(now));
localfmt = g_strdup_printf("[%s] [ID: %08ld] %s",
- nowstr, syscall(__NR_gettid), fmt);
+ buf, syscall(__NR_gettid), fmt);
}
fmt = localfmt;
}
--
2.35.3

View File

@ -83,7 +83,7 @@ Obsoletes: %1-rhev <= %{epoch}:%{version}-%{release}
Summary: QEMU is a machine emulator and virtualizer Summary: QEMU is a machine emulator and virtualizer
Name: qemu-kvm Name: qemu-kvm
Version: 6.2.0 Version: 6.2.0
Release: 20%{?rcrel}%{?dist}.1 Release: 28%{?rcrel}%{?dist}
# Epoch because we pushed a qemu-1.0 package. AIUI this can't ever be dropped # Epoch because we pushed a qemu-1.0 package. AIUI this can't ever be dropped
Epoch: 15 Epoch: 15
License: GPLv2 and GPLv2+ and CC-BY License: GPLv2 and GPLv2+ and CC-BY
@ -427,10 +427,169 @@ Patch163: kvm-QIOChannelSocket-Add-support-for-MSG_ZEROCOPY-IPV6.patch
Patch164: kvm-pc-bios-s390-ccw-Fix-booting-with-logical-block-size.patch Patch164: kvm-pc-bios-s390-ccw-Fix-booting-with-logical-block-size.patch
# For bz#2120279 - Wrong max_sectors_kb and Maximum transfer length on the pass-through device [rhel-8.7] # For bz#2120279 - Wrong max_sectors_kb and Maximum transfer length on the pass-through device [rhel-8.7]
Patch165: kvm-scsi-generic-Fix-emulated-block-limits-VPD-page.patch Patch165: kvm-scsi-generic-Fix-emulated-block-limits-VPD-page.patch
# For bz#2116743 - [RHEL8.7] Guests in VMX root operation fail to reboot with QEMU's 'system_reset' command # For bz#2117149 - Can't run when memory backing with hugepages and backend type memfd
Patch166: kvm-i386-reset-KVM-nested-state-upon-CPU-reset.patch Patch166: kvm-backends-hostmem-Fix-support-of-memory-backend-memfd.patch
# For bz#2116743 - [RHEL8.7] Guests in VMX root operation fail to reboot with QEMU's 'system_reset' command # For bz#2125271 - [RHEL8.7] Guests in VMX root operation fail to reboot with QEMU's 'system_reset' command [rhel-8.8.0]
Patch167: kvm-i386-do-kvm_put_msr_feature_control-first-thing-when.patch Patch167: kvm-i386-reset-KVM-nested-state-upon-CPU-reset.patch
# For bz#2125271 - [RHEL8.7] Guests in VMX root operation fail to reboot with QEMU's 'system_reset' command [rhel-8.8.0]
Patch168: kvm-i386-do-kvm_put_msr_feature_control-first-thing-when.patch
# For bz#2124757 - RHEL8: skey test in kvm_unit_test got failed
Patch169: kvm-redhat-Update-linux-headers-linux-kvm.h-to-v5.18-rc6.patch
# For bz#2124757 - RHEL8: skey test in kvm_unit_test got failed
Patch170: kvm-target-s390x-kvm-Honor-storage-keys-during-emulation.patch
# For bz#2018885 - [virtiofs] virtiofsd debug log's timestamp is NULL
Patch171: kvm-virtiofsd-use-g_date_time_get_microsecond-to-get-sub.patch
# For bz#2116302 - RHEL8.6 - virtiofs will not mount fs on secure execution guest
Patch172: kvm-virtio-fix-the-condition-for-iommu_platform-not-supp.patch
# For bz#2116302 - RHEL8.6 - virtiofs will not mount fs on secure execution guest
Patch173: kvm-virtio-fix-feature-negotiation-for-ACCESS_PLATFORM.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch174: kvm-s390x-pci-use-a-reserved-ID-for-the-default-PCI-grou.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch175: kvm-s390x-pci-don-t-use-hard-coded-dma-range-in-reg_ioat.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch176: kvm-s390x-pci-use-the-passthrough-measurement-update-int.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch177: kvm-s390x-pci-add-supported-DT-information-to-clp-respon.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch178: kvm-Update-linux-headers-to-v6.0-rc4.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch179: kvm-s390x-pci-add-routine-to-get-host-function-handle-fr.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch180: kvm-s390x-pci-enable-for-load-store-interpretation.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch181: kvm-s390x-pci-don-t-fence-interpreted-devices-without-MS.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch182: kvm-s390x-pci-enable-adapter-event-notification-for-inte.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch183: kvm-s390x-pci-let-intercept-devices-have-separate-PCI-gr.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch184: kvm-s390x-pci-reflect-proper-maxstbl-for-groups-of-inter.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch185: kvm-s390x-s390-virtio-ccw-Switch-off-zPCI-enhancements-o.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch186: kvm-dump-Use-ERRP_GUARD.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch187: kvm-dump-Remove-the-sh_info-variable.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch188: kvm-dump-Introduce-shdr_num-to-decrease-complexity.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch189: kvm-dump-Remove-the-section-if-when-calculating-the-memo.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch190: kvm-dump-Add-more-offset-variables.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch191: kvm-dump-Introduce-dump_is_64bit-helper-function.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch192: kvm-dump-Consolidate-phdr-note-writes.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch193: kvm-dump-Cleanup-dump_begin-write-functions.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch194: kvm-dump-Consolidate-elf-note-function.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch195: kvm-dump-Replace-opaque-DumpState-pointer-with-a-typed-o.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch196: kvm-dump-Rename-write_elf_loads-to-write_elf_phdr_loads.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch197: kvm-dump-Refactor-dump_iterate-and-introduce-dump_filter.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch198: kvm-dump-Rework-get_start_block.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch199: kvm-dump-Rework-filter-area-variables.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch200: kvm-dump-Rework-dump_calculate_size-function.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch201: kvm-dump-Split-elf-header-functions-into-prepare-and-wri.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch202: kvm-dump-Rename-write_elf-_phdr_note-to-prepare_elf-_phd.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch203: kvm-dump-simplify-a-bit-kdump-get_next_page.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch204: kvm-dump-fix-kdump-to-work-over-non-aligned-blocks.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch205: kvm-dump-Use-a-buffer-for-ELF-section-data-and-headers.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch206: kvm-dump-Write-ELF-section-headers-right-after-ELF-heade.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch207: kvm-dump-Reorder-struct-DumpState.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch208: kvm-dump-Reintroduce-memory_offset-and-section_offset.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch209: kvm-dump-Add-architecture-section-and-section-string-tab.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch210: kvm-s390x-Add-protected-dump-cap.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch211: kvm-s390x-Introduce-PV-query-interface.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch212: kvm-include-elf.h-add-s390x-note-types.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch213: kvm-s390x-Add-KVM-PV-dump-interface.patch
# For bz#1664378 - [IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part
# For bz#2043909 - [IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part
Patch214: kvm-s390x-pv-Add-dump-support.patch
# For bz#2129760 - CVE-2022-3165 virt:rhel/qemu-kvm: QEMU: VNC: integer underflow in vnc_client_cut_text_ext leads to CPU exhaustion [rhel-8]
Patch215: kvm-ui-vnc-clipboard-fix-integer-underflow-in-vnc_client.patch
# For bz#2132609 - qemu-kvm: backport some aarch64 fixes
Patch216: kvm-hw-acpi-Add-ospm_status-hook-implementation-for-acpi.patch
# For bz#2132609 - qemu-kvm: backport some aarch64 fixes
Patch217: kvm-target-arm-kvm-Retry-KVM_CREATE_VM-call-if-it-fails-.patch
# For bz#2128225 - [s390x] [RHEL8][s390x-ccw bios] lacking document about parameter loadparm in qemu
Patch218: kvm-docs-system-s390x-Document-the-loadparm-machine-prop.patch
# For bz#2128225 - [s390x] [RHEL8][s390x-ccw bios] lacking document about parameter loadparm in qemu
Patch219: kvm-s390x-Register-TYPE_S390_CCW_MACHINE-properties-as-c.patch
# For bz#2141896 - VMs hung on vnc_clipboard_send
Patch220: kvm-ui-vnc.c-Fixed-a-deadlock-bug.patch
# For bz#2148545 - CVE-2022-4144 virt:rhel/qemu-kvm: QEMU: QXL: qxl_phys2virt unsafe address translation can lead to out-of-bounds read [rhel-8]
Patch221: kvm-hw-display-qxl-Have-qxl_log_command-Return-early-if-.patch
# For bz#2148545 - CVE-2022-4144 virt:rhel/qemu-kvm: QEMU: QXL: qxl_phys2virt unsafe address translation can lead to out-of-bounds read [rhel-8]
Patch222: kvm-hw-display-qxl-Document-qxl_phys2virt.patch
# For bz#2148545 - CVE-2022-4144 virt:rhel/qemu-kvm: QEMU: QXL: qxl_phys2virt unsafe address translation can lead to out-of-bounds read [rhel-8]
Patch223: kvm-hw-display-qxl-Pass-requested-buffer-size-to-qxl_phy.patch
# For bz#2148545 - CVE-2022-4144 virt:rhel/qemu-kvm: QEMU: QXL: qxl_phys2virt unsafe address translation can lead to out-of-bounds read [rhel-8]
Patch224: kvm-hw-display-qxl-Avoid-buffer-overrun-in-qxl_phys2virt.patch
# For bz#2148545 - CVE-2022-4144 virt:rhel/qemu-kvm: QEMU: QXL: qxl_phys2virt unsafe address translation can lead to out-of-bounds read [rhel-8]
Patch225: kvm-hw-display-qxl-Assert-memory-slot-fits-in-preallocat.patch
# For bz#2155448 - RHEL8.8 - KVM: s390: pv: don't allow userspace to set the clock under PV - QEMU part
Patch226: kvm-s390x-tod-kvm-don-t-save-restore-the-TOD-in-PV-guest.patch
BuildRequires: wget BuildRequires: wget
BuildRequires: rpm-build BuildRequires: rpm-build
@ -1600,11 +1759,108 @@ sh %{_sysconfdir}/sysconfig/modules/kvm.modules &> /dev/null || :
%changelog %changelog
* Wed Sep 14 2022 Miroslav Rezanina <mrezanin@redhat.com> - 6.2.0-20.el8.1 * Wed Jan 04 2023 Jon Maloy <jmaloy@redhat.com> - 6.2.0-28
- kvm-i386-reset-KVM-nested-state-upon-CPU-reset.patch [bz#2116743] - kvm-s390x-tod-kvm-don-t-save-restore-the-TOD-in-PV-guest.patch [bz#2155448]
- kvm-i386-do-kvm_put_msr_feature_control-first-thing-when.patch [bz#2116743] - Resolves: bz#2155448
- Resolves: bz#2116743 (RHEL8.8 - KVM: s390: pv: don't allow userspace to set the clock under PV - QEMU part)
([RHEL8.7] Guests in VMX root operation fail to reboot with QEMU's 'system_reset' command)
* Thu Dec 08 2022 Jon Maloy <jmaloy@redhat.com> - 6.2.0-27
- kvm-hw-display-qxl-Have-qxl_log_command-Return-early-if-.patch [bz#2148545]
- kvm-hw-display-qxl-Document-qxl_phys2virt.patch [bz#2148545]
- kvm-hw-display-qxl-Pass-requested-buffer-size-to-qxl_phy.patch [bz#2148545]
- kvm-hw-display-qxl-Avoid-buffer-overrun-in-qxl_phys2virt.patch [bz#2148545]
- kvm-hw-display-qxl-Assert-memory-slot-fits-in-preallocat.patch [bz#2148545]
- Resolves: bz#2148545
(CVE-2022-4144 virt:rhel/qemu-kvm: QEMU: QXL: qxl_phys2virt unsafe address translation can lead to out-of-bounds read [rhel-8])
* Wed Nov 23 2022 Jon Maloy <jmaloy@redhat.com> - 6.2.0-26
- kvm-docs-system-s390x-Document-the-loadparm-machine-prop.patch [bz#2128225]
- kvm-s390x-Register-TYPE_S390_CCW_MACHINE-properties-as-c.patch [bz#2128225]
- kvm-ui-vnc.c-Fixed-a-deadlock-bug.patch [bz#2141896]
- Resolves: bz#2128225
([s390x] [RHEL8][s390x-ccw bios] lacking document about parameter loadparm in qemu)
- Resolves: bz#2141896
(VMs hung on vnc_clipboard_send)
* Wed Nov 16 2022 Jon Maloy <jmaloy@redhat.com> - 6.2.0-25
- kvm-hw-acpi-Add-ospm_status-hook-implementation-for-acpi.patch [bz#2132609]
- kvm-target-arm-kvm-Retry-KVM_CREATE_VM-call-if-it-fails-.patch [bz#2132609]
- Resolves: bz#2132609
(qemu-kvm: backport some aarch64 fixes)
* Thu Nov 10 2022 Jon Maloy <jmaloy@redhat.com> - 6.2.0-24
- kvm-s390x-pci-use-a-reserved-ID-for-the-default-PCI-grou.patch [bz#1664378 bz#2043909]
- kvm-s390x-pci-don-t-use-hard-coded-dma-range-in-reg_ioat.patch [bz#1664378 bz#2043909]
- kvm-s390x-pci-use-the-passthrough-measurement-update-int.patch [bz#1664378 bz#2043909]
- kvm-s390x-pci-add-supported-DT-information-to-clp-respon.patch [bz#1664378 bz#2043909]
- kvm-Update-linux-headers-to-v6.0-rc4.patch [bz#1664378 bz#2043909]
- kvm-s390x-pci-add-routine-to-get-host-function-handle-fr.patch [bz#1664378 bz#2043909]
- kvm-s390x-pci-enable-for-load-store-interpretation.patch [bz#1664378 bz#2043909]
- kvm-s390x-pci-don-t-fence-interpreted-devices-without-MS.patch [bz#1664378 bz#2043909]
- kvm-s390x-pci-enable-adapter-event-notification-for-inte.patch [bz#1664378 bz#2043909]
- kvm-s390x-pci-let-intercept-devices-have-separate-PCI-gr.patch [bz#1664378 bz#2043909]
- kvm-s390x-pci-reflect-proper-maxstbl-for-groups-of-inter.patch [bz#1664378 bz#2043909]
- kvm-s390x-s390-virtio-ccw-Switch-off-zPCI-enhancements-o.patch [bz#1664378 bz#2043909]
- kvm-dump-Use-ERRP_GUARD.patch [bz#1664378 bz#2043909]
- kvm-dump-Remove-the-sh_info-variable.patch [bz#1664378 bz#2043909]
- kvm-dump-Introduce-shdr_num-to-decrease-complexity.patch [bz#1664378 bz#2043909]
- kvm-dump-Remove-the-section-if-when-calculating-the-memo.patch [bz#1664378 bz#2043909]
- kvm-dump-Add-more-offset-variables.patch [bz#1664378 bz#2043909]
- kvm-dump-Introduce-dump_is_64bit-helper-function.patch [bz#1664378 bz#2043909]
- kvm-dump-Consolidate-phdr-note-writes.patch [bz#1664378 bz#2043909]
- kvm-dump-Cleanup-dump_begin-write-functions.patch [bz#1664378 bz#2043909]
- kvm-dump-Consolidate-elf-note-function.patch [bz#1664378 bz#2043909]
- kvm-dump-Replace-opaque-DumpState-pointer-with-a-typed-o.patch [bz#1664378 bz#2043909]
- kvm-dump-Rename-write_elf_loads-to-write_elf_phdr_loads.patch [bz#1664378 bz#2043909]
- kvm-dump-Refactor-dump_iterate-and-introduce-dump_filter.patch [bz#1664378 bz#2043909]
- kvm-dump-Rework-get_start_block.patch [bz#1664378 bz#2043909]
- kvm-dump-Rework-filter-area-variables.patch [bz#1664378 bz#2043909]
- kvm-dump-Rework-dump_calculate_size-function.patch [bz#1664378 bz#2043909]
- kvm-dump-Split-elf-header-functions-into-prepare-and-wri.patch [bz#1664378 bz#2043909]
- kvm-dump-Rename-write_elf-_phdr_note-to-prepare_elf-_phd.patch [bz#1664378 bz#2043909]
- kvm-dump-simplify-a-bit-kdump-get_next_page.patch [bz#1664378 bz#2043909]
- kvm-dump-fix-kdump-to-work-over-non-aligned-blocks.patch [bz#1664378 bz#2043909]
- kvm-dump-Use-a-buffer-for-ELF-section-data-and-headers.patch [bz#1664378 bz#2043909]
- kvm-dump-Write-ELF-section-headers-right-after-ELF-heade.patch [bz#1664378 bz#2043909]
- kvm-dump-Reorder-struct-DumpState.patch [bz#1664378 bz#2043909]
- kvm-dump-Reintroduce-memory_offset-and-section_offset.patch [bz#1664378 bz#2043909]
- kvm-dump-Add-architecture-section-and-section-string-tab.patch [bz#1664378 bz#2043909]
- kvm-s390x-Add-protected-dump-cap.patch [bz#1664378 bz#2043909]
- kvm-s390x-Introduce-PV-query-interface.patch [bz#1664378 bz#2043909]
- kvm-include-elf.h-add-s390x-note-types.patch [bz#1664378 bz#2043909]
- kvm-s390x-Add-KVM-PV-dump-interface.patch [bz#1664378 bz#2043909]
- kvm-s390x-pv-Add-dump-support.patch [bz#1664378 bz#2043909]
- kvm-ui-vnc-clipboard-fix-integer-underflow-in-vnc_client.patch [bz#2129760]
- Resolves: bz#1664378
([IBM 8.8 FEAT] Enhanced Interpretation for PCI Functions (kvm) - qemu part)
- Resolves: bz#2043909
([IBM 8.8 FEAT] KVM: Secure Execution guest dump encryption with customer keys - qemu part)
- Resolves: bz#2129760
(CVE-2022-3165 virt:rhel/qemu-kvm: QEMU: VNC: integer underflow in vnc_client_cut_text_ext leads to CPU exhaustion [rhel-8])
* Wed Oct 26 2022 Jon Maloy <jmaloy@redhat.com> - 6.2.0-23
- kvm-virtio-fix-the-condition-for-iommu_platform-not-supp.patch [bz#2116302]
- kvm-virtio-fix-feature-negotiation-for-ACCESS_PLATFORM.patch [bz#2116302]
- Resolves: bz#2116302
(RHEL8.6 - virtiofs will not mount fs on secure execution guest)
* Wed Oct 05 2022 Jon Maloy <jmaloy@redhat.com> - 6.2.0-22
- kvm-redhat-Update-linux-headers-linux-kvm.h-to-v5.18-rc6.patch [bz#2124757]
- kvm-target-s390x-kvm-Honor-storage-keys-during-emulation.patch [bz#2124757]
- kvm-virtiofsd-use-g_date_time_get_microsecond-to-get-sub.patch [bz#2018885]
- Resolves: bz#2124757
(RHEL8: skey test in kvm_unit_test got failed)
- Resolves: bz#2018885
([virtiofs] virtiofsd debug log's timestamp is NULL)
* Thu Sep 29 2022 Jon Maloy <jmaloy@redhat.com> - 6.2.0-21
- kvm-backends-hostmem-Fix-support-of-memory-backend-memfd.patch [bz#2117149]
- kvm-i386-reset-KVM-nested-state-upon-CPU-reset.patch [bz#2125271]
- kvm-i386-do-kvm_put_msr_feature_control-first-thing-when.patch [bz#2125271]
- Resolves: bz#2117149
(Can't run when memory backing with hugepages and backend type memfd)
- Resolves: bz#2125271
([RHEL8.7] Guests in VMX root operation fail to reboot with QEMU's 'system_reset' command [rhel-8.8.0])
* Fri Aug 26 2022 Jon Maloy <jmaloy@redhat.com> - 6.2.0-20 * Fri Aug 26 2022 Jon Maloy <jmaloy@redhat.com> - 6.2.0-20
- kvm-scsi-generic-Fix-emulated-block-limits-VPD-page.patch [bz#2120279] - kvm-scsi-generic-Fix-emulated-block-limits-VPD-page.patch [bz#2120279]