| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <20220128051600.GB5018@sunil-ThinkPad-T490>
Date: Fri, 28 Jan 2022 10:46:00 +0530
From: Sunil V L <sunilvl@...tanamicro.com>
To: Heinrich Schuchardt <xypron.glpk@....de>
Cc: Ard Biesheuvel <ardb@...nel.org>,
Paul Walmsley <paul.walmsley@...ive.com>,
Palmer Dabbelt <palmer@...belt.com>,
Albert Ou <aou@...s.berkeley.edu>, linux-efi@...r.kernel.org,
linux-kernel@...r.kernel.org, linux-riscv@...ts.infradead.org,
Anup Patel <apatel@...tanamicro.com>,
Atish Patra <atishp@...osinc.com>,
Abner Chang <abner.chang@....com>,
Jessica Clarke <jrtc27@...c27.com>
Subject: Re: [RFC PATCH 1/1] riscv/efi_stub: Add support for
RISCV_EFI_BOOT_PROTOCOL
On Thu, Jan 27, 2022 at 08:47:35AM +0100, Heinrich Schuchardt wrote:
> Am 26. Januar 2022 12:06:15 MEZ schrieb Sunil V L <sunilvl@...tanamicro.com>:
> >This patch adds the support for getting the boot hart ID in
> >Linux EFI stub using RISCV_EFI_BOOT_PROTOCOL.
>
> It would be helpful to add a link to the spec in the commit message and maybe a comment that this protocol is needed for the ACPI use case.
Sure. Will add.
>
> >
> >Signed-off-by: Sunil V L <sunilvl@...tanamicro.com>
> >---
> > drivers/firmware/efi/libstub/efistub.h | 15 ++++++++++++
> > drivers/firmware/efi/libstub/riscv-stub.c | 28 ++++++++++++++++++++---
> > include/linux/efi.h | 1 +
> > 3 files changed, 41 insertions(+), 3 deletions(-)
> >
> >diff --git a/drivers/firmware/efi/libstub/efistub.h b/drivers/firmware/efi/libstub/efistub.h
> >index edb77b0621ea..0428f8816942 100644
> >--- a/drivers/firmware/efi/libstub/efistub.h
> >+++ b/drivers/firmware/efi/libstub/efistub.h
> >@@ -720,6 +720,21 @@ union efi_tcg2_protocol {
> > } mixed_mode;
> > };
> >
> >+typedef union riscv_efi_boot_protocol riscv_efi_boot_protocol_t;
> >+
> >+union riscv_efi_boot_protocol {
> >+ struct {
> >+ u64 revision;
> >+ efi_status_t (__efiapi *get_boot_hartid)(
> >+ riscv_efi_boot_protocol_t *,
> >+ size_t *);
>
> I prefer to have parameter names for readability
Sure. Will add.
>
> According to the platform specification mhartid is MXLEN wide. UINTN (size_t) is SXLEN wide.
>
> Does this have any implications on how we define the protocol?
I don't think so. EFI and kernel will be at same privilige level. So, it
is not really an issue for this protocol.
But when MXLEN > SXLEN (ex: 64 vs 32), then implementation need to
ensure hartid value is 32 bit only so that when it is passed from M-mode
firmware to S-mode, it gets correct value. But again this is not an
issue from this EFI protocol perspective.
>
> >+ };
> >+ struct {
> >+ u32 revision;
> >+ u32 get_boot_hartid;>+ } mixed_mode;
> >+};
> >+
> > typedef union efi_load_file_protocol efi_load_file_protocol_t;
> > typedef union efi_load_file_protocol efi_load_file2_protocol_t;
> >
> >diff --git a/drivers/firmware/efi/libstub/riscv-stub.c b/drivers/firmware/efi/libstub/riscv-stub.c
> >index 380e4e251399..c7add4eb5453 100644
> >--- a/drivers/firmware/efi/libstub/riscv-stub.c
> >+++ b/drivers/firmware/efi/libstub/riscv-stub.c
> >@@ -46,12 +46,34 @@ static u32 get_boot_hartid_from_fdt(void)
> > return fdt32_to_cpu(*prop);
> > }
> >
> >+static u32 get_boot_hartid_from_efi(void)
> >+{
>
> The returned value must be UINTN /size_t like the protocol. This width must be carried to the legacy entry point of Linux.
>
> >+ efi_guid_t boot_protocol_guid = RISCV_EFI_BOOT_PROTOCOL_GUID;
> >+ efi_status_t status;
> >+ riscv_efi_boot_protocol_t *boot_protocol;
> >+ size_t boot_hart_id;
> >+
> >+ status = efi_bs_call(locate_protocol, &boot_protocol_guid, NULL,
> >+ (void **)&boot_protocol);
> >+ if (status == EFI_SUCCESS) {
> >+ status = efi_call_proto(boot_protocol,
> >+ get_boot_hartid, &boot_hart_id);
> >+ if (status == EFI_SUCCESS) {
> >+ return (u32)boot_hart_id;
> >+ }
> >+ }
> >+ return U32_MAX;
>
> U32_MAX is a legal value for the hart id.
Yeah. This is an existing issue in get_boot_hartid_from_fdt() for which
I have sent a fix patch. Once that gets accepted, I will fix
get_boot_hartid_from_efi() and send RFC V2 patch.
>
> >+}
> >+
> > efi_status_t check_platform_features(void)
> > {
> >- hartid = get_boot_hartid_from_fdt();
> >+ hartid = get_boot_hartid_from_efi();
> > if (hartid == U32_MAX) {
> >- efi_err("/chosen/boot-hartid missing or invalid!\n");
> >- return EFI_UNSUPPORTED;
> >+ hartid = get_boot_hartid_from_fdt();
> >+ if (hartid == U32_MAX) {
>
> U32_MAX is a legal value for the hart id. Please, separate status and value.
Will fix it.
Thank you very much for the feedback.
Sunil
>
> Best regards
>
> Heinrich
>
> >+ efi_err("/chosen/boot-hartid missing or invalid!\n");
> >+ return EFI_UNSUPPORTED;
> >+ }
> > }
> > return EFI_SUCCESS;
> > }
> >diff --git a/include/linux/efi.h b/include/linux/efi.h
> >index ccd4d3f91c98..9822c730207c 100644
> >--- a/include/linux/efi.h
> >+++ b/include/linux/efi.h
> >@@ -380,6 +380,7 @@ void efi_native_runtime_setup(void);
> > #define EFI_CONSOLE_OUT_DEVICE_GUID EFI_GUID(0xd3b36f2c, 0xd551, 0x11d4, 0x9a, 0x46, 0x00, 0x90, 0x27, 0x3f, 0xc1, 0x4d)
> > #define APPLE_PROPERTIES_PROTOCOL_GUID EFI_GUID(0x91bd12fe, 0xf6c3, 0x44fb, 0xa5, 0xb7, 0x51, 0x22, 0xab, 0x30, 0x3a, 0xe0)
> > #define EFI_TCG2_PROTOCOL_GUID EFI_GUID(0x607f766c, 0x7455, 0x42be, 0x93, 0x0b, 0xe4, 0xd7, 0x6d, 0xb2, 0x72, 0x0f)
> >+#define RISCV_EFI_BOOT_PROTOCOL_GUID EFI_GUID(0xccd15fec, 0x6f73, 0x4eec, 0x83, 0x95, 0x3e, 0x69, 0xe4, 0xb9, 0x40, 0xbf)
> > #define EFI_LOAD_FILE_PROTOCOL_GUID EFI_GUID(0x56ec3091, 0x954c, 0x11d2, 0x8e, 0x3f, 0x00, 0xa0, 0xc9, 0x69, 0x72, 0x3b)
> > #define EFI_LOAD_FILE2_PROTOCOL_GUID EFI_GUID(0x4006c0c1, 0xfcb3, 0x403e, 0x99, 0x6d, 0x4a, 0x6c, 0x87, 0x24, 0xe0, 0x6d)
> > #define EFI_RT_PROPERTIES_TABLE_GUID EFI_GUID(0xeb66918a, 0x7eef, 0x402a, 0x84, 0x2e, 0x93, 0x1d, 0x21, 0xc3, 0x8a, 0xe9)
>
Powered by blists - more mailing lists