| Message ID | 20211227193120.1084176-2-imammedo@redhat.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | Fix -acpitable regression | expand |
On 12/27/21 20:31, Igor Mammedov wrote: > if QEMU is started with used provided SLIC table blob, > > -acpitable sig=SLIC,oem_id='CRASH ',oem_table_id="ME",oem_rev=00002210,asl_compiler_id="",asl_compiler_rev=00000000,data=/dev/null > it will assert with: > > hw/acpi/aml-build.c:61:build_append_padded_str: assertion failed: (len <= maxlen) > > and following backtrace: > > ... > build_append_padded_str (array=0x555556afe320, str=0x555556afdb2e "CRASH ME", maxlen=0x6, pad=0x20) at hw/acpi/aml-build.c:61 > acpi_table_begin (desc=0x7fffffffd1b0, array=0x555556afe320) at hw/acpi/aml-build.c:1727 > build_fadt (tbl=0x555556afe320, linker=0x555557ca3830, f=0x7fffffffd318, oem_id=0x555556afdb2e "CRASH ME", oem_table_id=0x555556afdb34 "ME") at hw/acpi/aml-build.c:2064 > ... > > which happens due to acpi_table_begin() expecting NULL terminated > oem_id and oem_table_id strings, which is normally the case, but > in case of user provided SLIC table, oem_id points to table's blob > directly and as result oem_id became longer than expected. > > Fix issue by handling oem_id consistently and make acpi_get_slic_oem() > return NULL terminated strings. > > PS: > After [1] refactoring, oem_id semantics became inconsistent, where > NULL terminated string was coming from machine and old way pointer > into byte array coming from -acpitable option. That used to work > since build_header() wasn't expecting NULL terminated string and > blindly copied the 1st 6 bytes only. > > However commit [2] broke that by replacing build_header() with > acpi_table_begin(), which was expecting NULL terminated string > and was checking oem_id size. > > 1) 602b45820 ("acpi: Permit OEM ID and OEM table ID fields to be changed") > 2) > Fixes: 4b56e1e4eb08 ("acpi: build_fadt: use acpi_table_begin()/acpi_table_end() instead of build_header()") > Resolves: https://gitlab.com/qemu-project/qemu/-/issues/786 > Signed-off-by: Igor Mammedov <imammedo@redhat.com> > --- > hw/acpi/core.c | 4 ++-- > hw/i386/acpi-build.c | 2 ++ > 2 files changed, 4 insertions(+), 2 deletions(-) Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
On Monday, 27 December 2021 22:31:17 MSK Igor Mammedov wrote: > if QEMU is started with used provided SLIC table blob, > > -acpitable sig=SLIC,oem_id='CRASH > ',oem_table_id="ME",oem_rev=00002210,asl_compiler_id="",asl_compiler_rev=00 > 000000,data=/dev/null it will assert with: > > hw/acpi/aml-build.c:61:build_append_padded_str: assertion failed: (len <= > maxlen) > > and following backtrace: > > ... > build_append_padded_str (array=0x555556afe320, str=0x555556afdb2e "CRASH > ME", maxlen=0x6, pad=0x20) at hw/acpi/aml-build.c:61 acpi_table_begin > (desc=0x7fffffffd1b0, array=0x555556afe320) at hw/acpi/aml-build.c:1727 > build_fadt (tbl=0x555556afe320, linker=0x555557ca3830, f=0x7fffffffd318, > oem_id=0x555556afdb2e "CRASH ME", oem_table_id=0x555556afdb34 "ME") at > hw/acpi/aml-build.c:2064 ... > > which happens due to acpi_table_begin() expecting NULL terminated > oem_id and oem_table_id strings, which is normally the case, but > in case of user provided SLIC table, oem_id points to table's blob > directly and as result oem_id became longer than expected. > > Fix issue by handling oem_id consistently and make acpi_get_slic_oem() > return NULL terminated strings. > > PS: > After [1] refactoring, oem_id semantics became inconsistent, where > NULL terminated string was coming from machine and old way pointer > into byte array coming from -acpitable option. That used to work > since build_header() wasn't expecting NULL terminated string and > blindly copied the 1st 6 bytes only. > > However commit [2] broke that by replacing build_header() with > acpi_table_begin(), which was expecting NULL terminated string > and was checking oem_id size. > > 1) 602b45820 ("acpi: Permit OEM ID and OEM table ID fields to be changed") > 2) > Fixes: 4b56e1e4eb08 ("acpi: build_fadt: use > acpi_table_begin()/acpi_table_end() instead of build_header()") Resolves: > https://gitlab.com/qemu-project/qemu/-/issues/786 > Signed-off-by: Igor Mammedov <imammedo@redhat.com> > --- > hw/acpi/core.c | 4 ++-- > hw/i386/acpi-build.c | 2 ++ > 2 files changed, 4 insertions(+), 2 deletions(-) > > diff --git a/hw/acpi/core.c b/hw/acpi/core.c > index 1e004d0078..3e811bf03c 100644 > --- a/hw/acpi/core.c > +++ b/hw/acpi/core.c > @@ -345,8 +345,8 @@ int acpi_get_slic_oem(AcpiSlicOem *oem) > struct acpi_table_header *hdr = (void *)(u - sizeof(hdr->_length)); > > if (memcmp(hdr->sig, "SLIC", 4) == 0) { > - oem->id = hdr->oem_id; > - oem->table_id = hdr->oem_table_id; > + oem->id = g_strndup(hdr->oem_id, 6); > + oem->table_id = g_strndup(hdr->oem_table_id, 8); > return 0; > } > } > diff --git a/hw/i386/acpi-build.c b/hw/i386/acpi-build.c > index 8383b83ee3..0234fe7588 100644 > --- a/hw/i386/acpi-build.c > +++ b/hw/i386/acpi-build.c > @@ -2723,6 +2723,8 @@ void acpi_build(AcpiBuildTables *tables, MachineState > *machine) > > /* Cleanup memory that's no longer used. */ > g_array_free(table_offsets, true); > + g_free(slic_oem.id); > + g_free(slic_oem.table_id); > } > > static void acpi_ram_update(MemoryRegion *mr, GArray *data) Tested-by: Denis Lisov <dennis.lissov@gmail.com>
В Пн, 27/12/2021 в 14:31 -0500, Igor Mammedov пишет: > if QEMU is started with used provided SLIC table blob, > > -acpitable sig=SLIC,oem_id='CRASH > ',oem_table_id="ME",oem_rev=00002210,asl_compiler_id="",asl_compiler_re > v=00000000,data=/dev/null > it will assert with: > > hw/acpi/aml-build.c:61:build_append_padded_str: assertion failed: > (len <= maxlen) > > and following backtrace: > > ... > build_append_padded_str (array=0x555556afe320, str=0x555556afdb2e > "CRASH ME", maxlen=0x6, pad=0x20) at hw/acpi/aml-build.c:61 > acpi_table_begin (desc=0x7fffffffd1b0, array=0x555556afe320) at > hw/acpi/aml-build.c:1727 > build_fadt (tbl=0x555556afe320, linker=0x555557ca3830, > f=0x7fffffffd318, oem_id=0x555556afdb2e "CRASH ME", > oem_table_id=0x555556afdb34 "ME") at hw/acpi/aml-build.c:2064 > ... > > which happens due to acpi_table_begin() expecting NULL terminated > oem_id and oem_table_id strings, which is normally the case, but > in case of user provided SLIC table, oem_id points to table's blob > directly and as result oem_id became longer than expected. > > Fix issue by handling oem_id consistently and make acpi_get_slic_oem() > return NULL terminated strings. > > PS: > After [1] refactoring, oem_id semantics became inconsistent, where > NULL terminated string was coming from machine and old way pointer > into byte array coming from -acpitable option. That used to work > since build_header() wasn't expecting NULL terminated string and > blindly copied the 1st 6 bytes only. > > However commit [2] broke that by replacing build_header() with > acpi_table_begin(), which was expecting NULL terminated string > and was checking oem_id size. > > 1) 602b45820 ("acpi: Permit OEM ID and OEM table ID fields to be > changed") > 2) > Fixes: 4b56e1e4eb08 ("acpi: build_fadt: use > acpi_table_begin()/acpi_table_end() instead of build_header()") > Resolves: https://gitlab.com/qemu-project/qemu/-/issues/786 > Signed-off-by: Igor Mammedov <imammedo@redhat.com> > --- > hw/acpi/core.c | 4 ++-- > hw/i386/acpi-build.c | 2 ++ > 2 files changed, 4 insertions(+), 2 deletions(-) > > diff --git a/hw/acpi/core.c b/hw/acpi/core.c > index 1e004d0078..3e811bf03c 100644 > --- a/hw/acpi/core.c > +++ b/hw/acpi/core.c > @@ -345,8 +345,8 @@ int acpi_get_slic_oem(AcpiSlicOem *oem) > struct acpi_table_header *hdr = (void *)(u - sizeof(hdr- > >_length)); > > if (memcmp(hdr->sig, "SLIC", 4) == 0) { > - oem->id = hdr->oem_id; > - oem->table_id = hdr->oem_table_id; > + oem->id = g_strndup(hdr->oem_id, 6); > + oem->table_id = g_strndup(hdr->oem_table_id, 8); > return 0; > } > } > diff --git a/hw/i386/acpi-build.c b/hw/i386/acpi-build.c > index 8383b83ee3..0234fe7588 100644 > --- a/hw/i386/acpi-build.c > +++ b/hw/i386/acpi-build.c > @@ -2723,6 +2723,8 @@ void acpi_build(AcpiBuildTables *tables, > MachineState *machine) > > /* Cleanup memory that's no longer used. */ > g_array_free(table_offsets, true); > + g_free(slic_oem.id); > + g_free(slic_oem.table_id); > } > > static void acpi_ram_update(MemoryRegion *mr, GArray *data) Tested-by: Alexander Tsoy <alexander@tsoy.me>
CCing, qemu-stable@ On Mon, 27 Dec 2021 14:31:17 -0500 Igor Mammedov <imammedo@redhat.com> wrote: > if QEMU is started with used provided SLIC table blob, > > -acpitable sig=SLIC,oem_id='CRASH ',oem_table_id="ME",oem_rev=00002210,asl_compiler_id="",asl_compiler_rev=00000000,data=/dev/null > it will assert with: > > hw/acpi/aml-build.c:61:build_append_padded_str: assertion failed: (len <= maxlen) > > and following backtrace: > > ... > build_append_padded_str (array=0x555556afe320, str=0x555556afdb2e "CRASH ME", maxlen=0x6, pad=0x20) at hw/acpi/aml-build.c:61 > acpi_table_begin (desc=0x7fffffffd1b0, array=0x555556afe320) at hw/acpi/aml-build.c:1727 > build_fadt (tbl=0x555556afe320, linker=0x555557ca3830, f=0x7fffffffd318, oem_id=0x555556afdb2e "CRASH ME", oem_table_id=0x555556afdb34 "ME") at hw/acpi/aml-build.c:2064 > ... > > which happens due to acpi_table_begin() expecting NULL terminated > oem_id and oem_table_id strings, which is normally the case, but > in case of user provided SLIC table, oem_id points to table's blob > directly and as result oem_id became longer than expected. > > Fix issue by handling oem_id consistently and make acpi_get_slic_oem() > return NULL terminated strings. > > PS: > After [1] refactoring, oem_id semantics became inconsistent, where > NULL terminated string was coming from machine and old way pointer > into byte array coming from -acpitable option. That used to work > since build_header() wasn't expecting NULL terminated string and > blindly copied the 1st 6 bytes only. > > However commit [2] broke that by replacing build_header() with > acpi_table_begin(), which was expecting NULL terminated string > and was checking oem_id size. > > 1) 602b45820 ("acpi: Permit OEM ID and OEM table ID fields to be changed") > 2) > Fixes: 4b56e1e4eb08 ("acpi: build_fadt: use acpi_table_begin()/acpi_table_end() instead of build_header()") > Resolves: https://gitlab.com/qemu-project/qemu/-/issues/786 > Signed-off-by: Igor Mammedov <imammedo@redhat.com> > --- > hw/acpi/core.c | 4 ++-- > hw/i386/acpi-build.c | 2 ++ > 2 files changed, 4 insertions(+), 2 deletions(-) > > diff --git a/hw/acpi/core.c b/hw/acpi/core.c > index 1e004d0078..3e811bf03c 100644 > --- a/hw/acpi/core.c > +++ b/hw/acpi/core.c > @@ -345,8 +345,8 @@ int acpi_get_slic_oem(AcpiSlicOem *oem) > struct acpi_table_header *hdr = (void *)(u - sizeof(hdr->_length)); > > if (memcmp(hdr->sig, "SLIC", 4) == 0) { > - oem->id = hdr->oem_id; > - oem->table_id = hdr->oem_table_id; > + oem->id = g_strndup(hdr->oem_id, 6); > + oem->table_id = g_strndup(hdr->oem_table_id, 8); > return 0; > } > } > diff --git a/hw/i386/acpi-build.c b/hw/i386/acpi-build.c > index 8383b83ee3..0234fe7588 100644 > --- a/hw/i386/acpi-build.c > +++ b/hw/i386/acpi-build.c > @@ -2723,6 +2723,8 @@ void acpi_build(AcpiBuildTables *tables, MachineState *machine) > > /* Cleanup memory that's no longer used. */ > g_array_free(table_offsets, true); > + g_free(slic_oem.id); > + g_free(slic_oem.table_id); > } > > static void acpi_ram_update(MemoryRegion *mr, GArray *data)
diff --git a/hw/acpi/core.c b/hw/acpi/core.c index 1e004d0078..3e811bf03c 100644 --- a/hw/acpi/core.c +++ b/hw/acpi/core.c @@ -345,8 +345,8 @@ int acpi_get_slic_oem(AcpiSlicOem *oem) struct acpi_table_header *hdr = (void *)(u - sizeof(hdr->_length)); if (memcmp(hdr->sig, "SLIC", 4) == 0) { - oem->id = hdr->oem_id; - oem->table_id = hdr->oem_table_id; + oem->id = g_strndup(hdr->oem_id, 6); + oem->table_id = g_strndup(hdr->oem_table_id, 8); return 0; } } diff --git a/hw/i386/acpi-build.c b/hw/i386/acpi-build.c index 8383b83ee3..0234fe7588 100644 --- a/hw/i386/acpi-build.c +++ b/hw/i386/acpi-build.c @@ -2723,6 +2723,8 @@ void acpi_build(AcpiBuildTables *tables, MachineState *machine) /* Cleanup memory that's no longer used. */ g_array_free(table_offsets, true); + g_free(slic_oem.id); + g_free(slic_oem.table_id); } static void acpi_ram_update(MemoryRegion *mr, GArray *data)
if QEMU is started with used provided SLIC table blob, -acpitable sig=SLIC,oem_id='CRASH ',oem_table_id="ME",oem_rev=00002210,asl_compiler_id="",asl_compiler_rev=00000000,data=/dev/null it will assert with: hw/acpi/aml-build.c:61:build_append_padded_str: assertion failed: (len <= maxlen) and following backtrace: ... build_append_padded_str (array=0x555556afe320, str=0x555556afdb2e "CRASH ME", maxlen=0x6, pad=0x20) at hw/acpi/aml-build.c:61 acpi_table_begin (desc=0x7fffffffd1b0, array=0x555556afe320) at hw/acpi/aml-build.c:1727 build_fadt (tbl=0x555556afe320, linker=0x555557ca3830, f=0x7fffffffd318, oem_id=0x555556afdb2e "CRASH ME", oem_table_id=0x555556afdb34 "ME") at hw/acpi/aml-build.c:2064 ... which happens due to acpi_table_begin() expecting NULL terminated oem_id and oem_table_id strings, which is normally the case, but in case of user provided SLIC table, oem_id points to table's blob directly and as result oem_id became longer than expected. Fix issue by handling oem_id consistently and make acpi_get_slic_oem() return NULL terminated strings. PS: After [1] refactoring, oem_id semantics became inconsistent, where NULL terminated string was coming from machine and old way pointer into byte array coming from -acpitable option. That used to work since build_header() wasn't expecting NULL terminated string and blindly copied the 1st 6 bytes only. However commit [2] broke that by replacing build_header() with acpi_table_begin(), which was expecting NULL terminated string and was checking oem_id size. 1) 602b45820 ("acpi: Permit OEM ID and OEM table ID fields to be changed") 2) Fixes: 4b56e1e4eb08 ("acpi: build_fadt: use acpi_table_begin()/acpi_table_end() instead of build_header()") Resolves: https://gitlab.com/qemu-project/qemu/-/issues/786 Signed-off-by: Igor Mammedov <imammedo@redhat.com> --- hw/acpi/core.c | 4 ++-- hw/i386/acpi-build.c | 2 ++ 2 files changed, 4 insertions(+), 2 deletions(-)