From patchwork Tue Sep 19 03:57:08 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Pu Lehui X-Patchwork-Id: 13390675 X-Patchwork-Delegate: bpf@iogearbox.net Received: from lindbergh.monkeyblade.net (lindbergh.monkeyblade.net [23.128.96.19]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 5907330F93; Tue, 19 Sep 2023 03:57:01 +0000 (UTC) Received: from dggsgout11.his.huawei.com (unknown [45.249.212.51]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 856F1111; Mon, 18 Sep 2023 20:56:58 -0700 (PDT) Received: from mail02.huawei.com (unknown [172.30.67.143]) by dggsgout11.his.huawei.com (SkyGuard) with ESMTP id 4RqSXM1dbpz4f3lft; Tue, 19 Sep 2023 11:56:51 +0800 (CST) Received: from localhost.localdomain (unknown [10.67.175.61]) by APP4 (Coremail) with SMTP id gCh0CgBH09IDHAlldnxXAw--.51651S3; Tue, 19 Sep 2023 11:56:52 +0800 (CST) From: Pu Lehui To: bpf@vger.kernel.org, linux-riscv@lists.infradead.org, netdev@vger.kernel.org Cc: =?utf-8?b?QmrDtnJuIFTDtnBlbA==?= , Alexei Starovoitov , Daniel Borkmann , Andrii Nakryiko , Martin KaFai Lau , Song Liu , Yonghong Song , John Fastabend , KP Singh , Stanislav Fomichev , Hao Luo , Jiri Olsa , Palmer Dabbelt , Conor Dooley , Luke Nelson , Pu Lehui , Pu Lehui Subject: [PATCH bpf-next 1/4] riscv, bpf: Remove redundant ctx->offset initialization Date: Tue, 19 Sep 2023 11:57:08 +0800 Message-Id: <20230919035711.3297256-2-pulehui@huaweicloud.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20230919035711.3297256-1-pulehui@huaweicloud.com> References: <20230919035711.3297256-1-pulehui@huaweicloud.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-CM-TRANSID: gCh0CgBH09IDHAlldnxXAw--.51651S3 X-Coremail-Antispam: 1UD129KBjvdXoW7XFW8Jr15KF1fuF45XrW5KFg_yoW3Arg_C3 4xta4fXrWfJa1qkryj9w4rWr1qkw1rtFyrur1fXry5Can0qw1vy393try7Gry7Zrs8ZrWx WFZ7J3yxZw4YgjkaLaAFLSUrUUUUbb8apTn2vfkv8UJUUUU8Yxn0WfASr-VFAUDa7-sFnT 9fnUUIcSsGvfJTRUUUbkxFF20E14v26rWj6s0DM7CY07I20VC2zVCF04k26cxKx2IYs7xG 6rWj6s0DM7CIcVAFz4kK6r1j6r18M28IrcIa0xkI8VA2jI8067AKxVWUGwA2048vs2IY02 0Ec7CjxVAFwI0_Gr0_Xr1l8cAvFVAK0II2c7xJM28CjxkF64kEwVA0rcxSw2x7M28EF7xv wVC0I7IYx2IY67AKxVWDJVCq3wA2z4x0Y4vE2Ix0cI8IcVCY1x0267AKxVW8Jr0_Cr1UM2 8EF7xvwVC2z280aVAFwI0_GcCE3s1l84ACjcxK6I8E87Iv6xkF7I0E14v26rxl6s0DM2AI xVAIcxkEcVAq07x20xvEncxIr21l5I8CrVACY4xI64kE6c02F40Ex7xfMcIj6xIIjxv20x vE14v26r1j6r18McIj6I8E87Iv67AKxVWUJVW8JwAm72CE4IkC6x0Yz7v_Jr0_Gr1lF7xv r2IYc2Ij64vIr41lF7I21c0EjII2zVCS5cI20VAGYxC7M4IIrI8v6xkF7I0E8cxan2IY04 v7MxkF7I0En4kS14v26r1q6r43MxAIw28IcxkI7VAKI48JMxC20s026xCaFVCjc4AY6r1j 6r4UMI8I3I0E5I8CrVAFwI0_Jr0_Jr4lx2IqxVCjr7xvwVAFwI0_JrI_JrWlx4CE17CEb7 AF67AKxVW8ZVWrXwCIc40Y0x0EwIxGrwCI42IY6xIIjxv20xvE14v26r1j6r1xMIIF0xvE 2Ix0cI8IcVCY1x0267AKxVW8JVWxJwCI42IY6xAIw20EY4v20xvaj40_Jr0_JF4lIxAIcV C2z280aVAFwI0_Gr0_Cr1lIxAIcVC2z280aVCY1x0267AKxVW8JVW8JrUvcSsGvfC2Kfnx nUUI43ZEXa7VUjrHUDUUUUU== X-CM-SenderInfo: psxovxtxl6x35dzhxuhorxvhhfrp/ X-CFilter-Loop: Reflected X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00,KHOP_HELO_FCRDNS, RCVD_IN_MSPIKE_H2,SPF_HELO_NONE,SPF_NONE autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on lindbergh.monkeyblade.net X-Patchwork-Delegate: bpf@iogearbox.net From: Pu Lehui There is no gain in initializing ctx->offset and prev_insns, and ctx->offset is already zero-initialized. Let's remove this code. Signed-off-by: Pu Lehui --- arch/riscv/net/bpf_jit_core.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/arch/riscv/net/bpf_jit_core.c b/arch/riscv/net/bpf_jit_core.c index 7a26a3e1c..dd00103a2 100644 --- a/arch/riscv/net/bpf_jit_core.c +++ b/arch/riscv/net/bpf_jit_core.c @@ -89,11 +89,6 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog) goto out_offset; } - for (i = 0; i < prog->len; i++) { - prev_ninsns += 32; - ctx->offset[i] = prev_ninsns; - } - for (i = 0; i < NR_JIT_ITERATIONS; i++) { pass++; ctx->ninsns = 0; From patchwork Tue Sep 19 03:57:09 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Pu Lehui X-Patchwork-Id: 13390676 X-Patchwork-Delegate: bpf@iogearbox.net Received: from lindbergh.monkeyblade.net (lindbergh.monkeyblade.net [23.128.96.19]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id B9DF531A9C; Tue, 19 Sep 2023 03:57:01 +0000 (UTC) Received: from dggsgout11.his.huawei.com (unknown [45.249.212.51]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 4A00C10F; Mon, 18 Sep 2023 20:56:58 -0700 (PDT) Received: from mail02.huawei.com (unknown [172.30.67.143]) by dggsgout11.his.huawei.com (SkyGuard) with ESMTP id 4RqSXN31W5z4f3kkT; Tue, 19 Sep 2023 11:56:52 +0800 (CST) Received: from localhost.localdomain (unknown [10.67.175.61]) by APP4 (Coremail) with SMTP id gCh0CgBH09IDHAlldnxXAw--.51651S4; Tue, 19 Sep 2023 11:56:53 +0800 (CST) From: Pu Lehui To: bpf@vger.kernel.org, linux-riscv@lists.infradead.org, netdev@vger.kernel.org Cc: =?utf-8?b?QmrDtnJuIFTDtnBlbA==?= , Alexei Starovoitov , Daniel Borkmann , Andrii Nakryiko , Martin KaFai Lau , Song Liu , Yonghong Song , John Fastabend , KP Singh , Stanislav Fomichev , Hao Luo , Jiri Olsa , Palmer Dabbelt , Conor Dooley , Luke Nelson , Pu Lehui , Pu Lehui Subject: [PATCH bpf-next 2/4] riscv, bpf: Using kvcalloc to allocate cache buffer Date: Tue, 19 Sep 2023 11:57:09 +0800 Message-Id: <20230919035711.3297256-3-pulehui@huaweicloud.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20230919035711.3297256-1-pulehui@huaweicloud.com> References: <20230919035711.3297256-1-pulehui@huaweicloud.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-CM-TRANSID: gCh0CgBH09IDHAlldnxXAw--.51651S4 X-Coremail-Antispam: 1UD129KBjvJXoW7WFW3Cw13GrWfGw4Utr4fXwb_yoW8Cw1kpF 4DKrnxA398Zr1kXF40qr4kXFy5J3Wqg3W7GFWUuFWrWF90qrW8Xan5C345urWrCrWFyrWF v3yY9rnxu34kZwUanT9S1TB71UUUUUDqnTZGkaVYY2UrUUUUjbIjqfuFe4nvWSU5nxnvy2 9KBjDU0xBIdaVrnRJUUUmF14x267AKxVWrJVCq3wAFc2x0x2IEx4CE42xK8VAvwI8IcIk0 rVWrJVCq3wAFIxvE14AKwVWUJVWUGwA2048vs2IY020E87I2jVAFwI0_Jryl82xGYIkIc2 x26xkF7I0E14v26ryj6s0DM28lY4IEw2IIxxk0rwA2F7IY1VAKz4vEj48ve4kI8wA2z4x0 Y4vE2Ix0cI8IcVAFwI0_tr0E3s1l84ACjcxK6xIIjxv20xvEc7CjxVAFwI0_Gr1j6F4UJw A2z4x0Y4vEx4A2jsIE14v26rxl6s0DM28EF7xvwVC2z280aVCY1x0267AKxVW0oVCq3wAS 0I0E0xvYzxvE52x082IY62kv0487Mc02F40EFcxC0VAKzVAqx4xG6I80ewAv7VC0I7IYx2 IY67AKxVWUJVWUGwAv7VC2z280aVAFwI0_Jr0_Gr1lOx8S6xCaFVCjc4AY6r1j6r4UM4x0 Y48IcxkI7VAKI48JM4x0x7Aq67IIx4CEVc8vx2IErcIFxwACI402YVCY1x02628vn2kIc2 xKxwCY1x0262kKe7AKxVWUtVW8ZwCF04k20xvY0x0EwIxGrwCFx2IqxVCFs4IE7xkEbVWU JVW8JwC20s026c02F40E14v26r1j6r18MI8I3I0E7480Y4vE14v26r106r1rMI8E67AF67 kF1VAFwI0_GFv_WrylIxkGc2Ij64vIr41lIxAIcVC0I7IYx2IY67AKxVWUJVWUCwCI42IY 6xIIjxv20xvEc7CjxVAFwI0_Cr0_Gr1UMIIF0xvE42xK8VAvwI8IcIk0rVWUJVWUCwCI42 IY6I8E87Iv67AKxVW8JVWxJwCI42IY6I8E87Iv6xkF7I0E14v26r4UJVWxJrUvcSsGvfC2 KfnxnUUI43ZEXa7VUbH5lUUUUUU== X-CM-SenderInfo: psxovxtxl6x35dzhxuhorxvhhfrp/ X-CFilter-Loop: Reflected X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00,KHOP_HELO_FCRDNS, RCVD_IN_MSPIKE_H2,SPF_HELO_NONE,SPF_NONE autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on lindbergh.monkeyblade.net X-Patchwork-Delegate: bpf@iogearbox.net From: Pu Lehui It is unnecessary to allocate continuous physical memory for cache buffer, and when ebpf program is too large, it may cause memory allocation failure. Signed-off-by: Pu Lehui --- arch/riscv/net/bpf_jit_comp64.c | 4 ++-- arch/riscv/net/bpf_jit_core.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/riscv/net/bpf_jit_comp64.c b/arch/riscv/net/bpf_jit_comp64.c index 8423f4ddf..57535cfe1 100644 --- a/arch/riscv/net/bpf_jit_comp64.c +++ b/arch/riscv/net/bpf_jit_comp64.c @@ -906,7 +906,7 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, } if (fmod_ret->nr_links) { - branches_off = kcalloc(fmod_ret->nr_links, sizeof(int), GFP_KERNEL); + branches_off = kvcalloc(fmod_ret->nr_links, sizeof(int), GFP_KERNEL); if (!branches_off) return -ENOMEM; @@ -993,7 +993,7 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, ret = ctx->ninsns; out: - kfree(branches_off); + kvfree(branches_off); return ret; } diff --git a/arch/riscv/net/bpf_jit_core.c b/arch/riscv/net/bpf_jit_core.c index dd00103a2..d9c82537d 100644 --- a/arch/riscv/net/bpf_jit_core.c +++ b/arch/riscv/net/bpf_jit_core.c @@ -78,7 +78,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog) } ctx->prog = prog; - ctx->offset = kcalloc(prog->len, sizeof(int), GFP_KERNEL); + ctx->offset = kvcalloc(prog->len, sizeof(int), GFP_KERNEL); if (!ctx->offset) { prog = orig_prog; goto out_offset; @@ -170,7 +170,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog) ctx->offset[i] = ninsns_rvoff(ctx->offset[i]); bpf_prog_fill_jited_linfo(prog, ctx->offset); out_offset: - kfree(ctx->offset); + kvfree(ctx->offset); kfree(jit_data); prog->aux->jit_data = NULL; } From patchwork Tue Sep 19 03:57:10 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Pu Lehui X-Patchwork-Id: 13390673 X-Patchwork-Delegate: bpf@iogearbox.net Received: from lindbergh.monkeyblade.net (lindbergh.monkeyblade.net [23.128.96.19]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 6AFB78489; Tue, 19 Sep 2023 03:57:00 +0000 (UTC) Received: from dggsgout12.his.huawei.com (unknown [45.249.212.56]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 8BABB102; Mon, 18 Sep 2023 20:56:56 -0700 (PDT) Received: from mail02.huawei.com (unknown [172.30.67.143]) by dggsgout12.his.huawei.com (SkyGuard) with ESMTP id 4RqSXK0DLXz4f3khm; Tue, 19 Sep 2023 11:56:49 +0800 (CST) Received: from localhost.localdomain (unknown [10.67.175.61]) by APP4 (Coremail) with SMTP id gCh0CgBH09IDHAlldnxXAw--.51651S5; Tue, 19 Sep 2023 11:56:53 +0800 (CST) From: Pu Lehui To: bpf@vger.kernel.org, linux-riscv@lists.infradead.org, netdev@vger.kernel.org Cc: =?utf-8?b?QmrDtnJuIFTDtnBlbA==?= , Alexei Starovoitov , Daniel Borkmann , Andrii Nakryiko , Martin KaFai Lau , Song Liu , Yonghong Song , John Fastabend , KP Singh , Stanislav Fomichev , Hao Luo , Jiri Olsa , Palmer Dabbelt , Conor Dooley , Luke Nelson , Pu Lehui , Pu Lehui Subject: [PATCH bpf-next 3/4] riscv, bpf: Add RV_TAILCALL_OFFSET macro to format tailcall offset Date: Tue, 19 Sep 2023 11:57:10 +0800 Message-Id: <20230919035711.3297256-4-pulehui@huaweicloud.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20230919035711.3297256-1-pulehui@huaweicloud.com> References: <20230919035711.3297256-1-pulehui@huaweicloud.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-CM-TRANSID: gCh0CgBH09IDHAlldnxXAw--.51651S5 X-Coremail-Antispam: 1UD129KBjvJXoW7WrWrAFyxJr4rGw1UtF1rCrg_yoW8Ww4xpF 4UCwn3CF9Yqr18CryxAF4DW3Wakrn0qFsrGrWakrW3G3ZF9rykGa1DK3yYqry5uFy8Jry8 Zryqvw13ua4kZrDanT9S1TB71UUUUUDqnTZGkaVYY2UrUUUUjbIjqfuFe4nvWSU5nxnvy2 9KBjDU0xBIdaVrnRJUUUmF14x267AKxVWrJVCq3wAFc2x0x2IEx4CE42xK8VAvwI8IcIk0 rVWrJVCq3wAFIxvE14AKwVWUJVWUGwA2048vs2IY020E87I2jVAFwI0_JrWl82xGYIkIc2 x26xkF7I0E14v26ryj6s0DM28lY4IEw2IIxxk0rwA2F7IY1VAKz4vEj48ve4kI8wA2z4x0 Y4vE2Ix0cI8IcVAFwI0_tr0E3s1l84ACjcxK6xIIjxv20xvEc7CjxVAFwI0_Gr1j6F4UJw A2z4x0Y4vEx4A2jsIE14v26rxl6s0DM28EF7xvwVC2z280aVCY1x0267AKxVW0oVCq3wAS 0I0E0xvYzxvE52x082IY62kv0487Mc02F40EFcxC0VAKzVAqx4xG6I80ewAv7VC0I7IYx2 IY67AKxVWUJVWUGwAv7VC2z280aVAFwI0_Jr0_Gr1lOx8S6xCaFVCjc4AY6r1j6r4UM4x0 Y48IcxkI7VAKI48JM4x0x7Aq67IIx4CEVc8vx2IErcIFxwACI402YVCY1x02628vn2kIc2 xKxwCY1x0262kKe7AKxVWUtVW8ZwCF04k20xvY0x0EwIxGrwCFx2IqxVCFs4IE7xkEbVWU JVW8JwC20s026c02F40E14v26r1j6r18MI8I3I0E7480Y4vE14v26r106r1rMI8E67AF67 kF1VAFwI0_GFv_WrylIxkGc2Ij64vIr41lIxAIcVC0I7IYx2IY67AKxVWUJVWUCwCI42IY 6xIIjxv20xvEc7CjxVAFwI0_Cr0_Gr1UMIIF0xvE42xK8VAvwI8IcIk0rVWUJVWUCwCI42 IY6I8E87Iv67AKxVW8JVWxJwCI42IY6I8E87Iv6xkF7I0E14v26r4UJVWxJrUvcSsGvfC2 KfnxnUUI43ZEXa7VUbpwZ7UUUUU== X-CM-SenderInfo: psxovxtxl6x35dzhxuhorxvhhfrp/ X-CFilter-Loop: Reflected X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00,KHOP_HELO_FCRDNS, RCVD_IN_MSPIKE_H2,SPF_HELO_NONE,SPF_NONE autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on lindbergh.monkeyblade.net X-Patchwork-Delegate: bpf@iogearbox.net From: Pu Lehui Add RV_TAILCALL_OFFSET macro to format tailcall offset, and correct the relevant comments. Signed-off-by: Pu Lehui --- arch/riscv/net/bpf_jit_comp64.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/arch/riscv/net/bpf_jit_comp64.c b/arch/riscv/net/bpf_jit_comp64.c index 57535cfe1..f2ded1151 100644 --- a/arch/riscv/net/bpf_jit_comp64.c +++ b/arch/riscv/net/bpf_jit_comp64.c @@ -13,7 +13,9 @@ #include #include "bpf_jit.h" -#define RV_FENTRY_NINSNS 2 +#define RV_FENTRY_NINSNS 2 +/* fentry and TCC init insns will be skipped on tailcall */ +#define RV_TAILCALL_OFFSET ((RV_FENTRY_NINSNS + 1) * 4) #define RV_REG_TCC RV_REG_A6 #define RV_REG_TCC_SAVED RV_REG_S6 /* Store A6 in S6 if program do calls */ @@ -243,8 +245,7 @@ static void __build_epilogue(bool is_tail_call, struct rv_jit_context *ctx) if (!is_tail_call) emit_mv(RV_REG_A0, RV_REG_A5, ctx); emit_jalr(RV_REG_ZERO, is_tail_call ? RV_REG_T3 : RV_REG_RA, - is_tail_call ? (RV_FENTRY_NINSNS + 1) * 4 : 0, /* skip reserved nops and TCC init */ - ctx); + is_tail_call ? RV_TAILCALL_OFFSET : 0, ctx); } static void emit_bcc(u8 cond, u8 rd, u8 rs, int rvoff, @@ -371,7 +372,7 @@ static int emit_bpf_tail_call(int insn, struct rv_jit_context *ctx) off = ninsns_rvoff(tc_ninsn - (ctx->ninsns - start_insn)); emit_branch(BPF_JEQ, RV_REG_T2, RV_REG_ZERO, off, ctx); - /* goto *(prog->bpf_func + 4); */ + /* goto *(prog->bpf_func + RV_TAILCALL_OFFSET); */ off = offsetof(struct bpf_prog, bpf_func); if (is_12b_check(off, insn)) return -1; From patchwork Tue Sep 19 03:57:11 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Pu Lehui X-Patchwork-Id: 13390677 X-Patchwork-Delegate: bpf@iogearbox.net Received: from lindbergh.monkeyblade.net (lindbergh.monkeyblade.net [23.128.96.19]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 9985C341B7; Tue, 19 Sep 2023 03:57:02 +0000 (UTC) Received: from dggsgout11.his.huawei.com (unknown [45.249.212.51]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 21CAC10E; Mon, 18 Sep 2023 20:56:58 -0700 (PDT) Received: from mail02.huawei.com (unknown [172.30.67.143]) by dggsgout11.his.huawei.com (SkyGuard) with ESMTP id 4RqSXN4qqCz4f3kq4; Tue, 19 Sep 2023 11:56:52 +0800 (CST) Received: from localhost.localdomain (unknown [10.67.175.61]) by APP4 (Coremail) with SMTP id gCh0CgBH09IDHAlldnxXAw--.51651S6; Tue, 19 Sep 2023 11:56:53 +0800 (CST) From: Pu Lehui To: bpf@vger.kernel.org, linux-riscv@lists.infradead.org, netdev@vger.kernel.org Cc: =?utf-8?b?QmrDtnJuIFTDtnBlbA==?= , Alexei Starovoitov , Daniel Borkmann , Andrii Nakryiko , Martin KaFai Lau , Song Liu , Yonghong Song , John Fastabend , KP Singh , Stanislav Fomichev , Hao Luo , Jiri Olsa , Palmer Dabbelt , Conor Dooley , Luke Nelson , Pu Lehui , Pu Lehui Subject: [PATCH bpf-next 4/4] riscv, bpf: Mixing bpf2bpf and tailcalls Date: Tue, 19 Sep 2023 11:57:11 +0800 Message-Id: <20230919035711.3297256-5-pulehui@huaweicloud.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20230919035711.3297256-1-pulehui@huaweicloud.com> References: <20230919035711.3297256-1-pulehui@huaweicloud.com> Precedence: bulk X-Mailing-List: bpf@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-CM-TRANSID: gCh0CgBH09IDHAlldnxXAw--.51651S6 X-Coremail-Antispam: 1UD129KBjvJXoWxtF4DWr1kZFy8Zr4ruF15Jwb_yoWfKF17pa 4kKw4fCFW0qa15JFZrGF1DXw1akw4vv3ZIkry3Kw4Ska1qqrykG3WxKa1YyFy5Ar95Zw1x Xr4Dt3ZFga17JrJanT9S1TB71UUUUUDqnTZGkaVYY2UrUUUUjbIjqfuFe4nvWSU5nxnvy2 9KBjDU0xBIdaVrnRJUUUma14x267AKxVWrJVCq3wAFc2x0x2IEx4CE42xK8VAvwI8IcIk0 rVWrJVCq3wAFIxvE14AKwVWUJVWUGwA2048vs2IY020E87I2jVAFwI0_JF0E3s1l82xGYI kIc2x26xkF7I0E14v26ryj6s0DM28lY4IEw2IIxxk0rwA2F7IY1VAKz4vEj48ve4kI8wA2 z4x0Y4vE2Ix0cI8IcVAFwI0_tr0E3s1l84ACjcxK6xIIjxv20xvEc7CjxVAFwI0_Gr1j6F 4UJwA2z4x0Y4vEx4A2jsIE14v26rxl6s0DM28EF7xvwVC2z280aVCY1x0267AKxVW0oVCq 3wAS0I0E0xvYzxvE52x082IY62kv0487Mc02F40EFcxC0VAKzVAqx4xG6I80ewAv7VC0I7 IYx2IY67AKxVWUJVWUGwAv7VC2z280aVAFwI0_Jr0_Gr1lOx8S6xCaFVCjc4AY6r1j6r4U M4x0Y48IcxkI7VAKI48JM4x0x7Aq67IIx4CEVc8vx2IErcIFxwACI402YVCY1x02628vn2 kIc2xKxwCY1x0262kKe7AKxVWUtVW8ZwCF04k20xvY0x0EwIxGrwCFx2IqxVCFs4IE7xkE bVWUJVW8JwC20s026c02F40E14v26r1j6r18MI8I3I0E7480Y4vE14v26r106r1rMI8E67 AF67kF1VAFwI0_GFv_WrylIxkGc2Ij64vIr41lIxAIcVC0I7IYx2IY67AKxVWUJVWUCwCI 42IY6xIIjxv20xvEc7CjxVAFwI0_Cr0_Gr1UMIIF0xvE42xK8VAvwI8IcIk0rVWUJVWUCw CI42IY6I8E87Iv67AKxVW8JVWxJwCI42IY6I8E87Iv6xkF7I0E14v26r4UJVWxJrUvcSsG vfC2KfnxnUUI43ZEXa7VUbPC7UUUUUU== X-CM-SenderInfo: psxovxtxl6x35dzhxuhorxvhhfrp/ X-CFilter-Loop: Reflected X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00,KHOP_HELO_FCRDNS, RCVD_IN_MSPIKE_H2,SPF_HELO_NONE,SPF_NONE autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on lindbergh.monkeyblade.net X-Patchwork-Delegate: bpf@iogearbox.net From: Pu Lehui In the current RV64 JIT, if we just don't initialize the TCC in subprog, the TCC can be propagated from the parent process to the subprocess, but the TCC of the parent process cannot be restored when the subprocess exits. Since the RV64 TCC is initialized before saving the callee saved registers into the stack, we cannot use the callee saved register to pass the TCC, otherwise the original value of the callee saved register will be destroyed. So we implemented mixing bpf2bpf and tailcalls similar to x86_64, i.e. using a non-callee saved register to transfer the TCC between functions, and saving that register to the stack to protect the TCC value. At the same time, we also consider the scenario of mixing trampoline. Tests test_bpf.ko and test_verifier have passed, as well as the relative testcases of test_progs*. Signed-off-by: Pu Lehui --- arch/riscv/net/bpf_jit.h | 1 + arch/riscv/net/bpf_jit_comp64.c | 91 ++++++++++++++------------------- 2 files changed, 39 insertions(+), 53 deletions(-) diff --git a/arch/riscv/net/bpf_jit.h b/arch/riscv/net/bpf_jit.h index d21c6c92a..ca518846c 100644 --- a/arch/riscv/net/bpf_jit.h +++ b/arch/riscv/net/bpf_jit.h @@ -75,6 +75,7 @@ struct rv_jit_context { int nexentries; unsigned long flags; int stack_size; + int tcc_offset; }; /* Convert from ninsns to bytes. */ diff --git a/arch/riscv/net/bpf_jit_comp64.c b/arch/riscv/net/bpf_jit_comp64.c index f2ded1151..f37be4911 100644 --- a/arch/riscv/net/bpf_jit_comp64.c +++ b/arch/riscv/net/bpf_jit_comp64.c @@ -13,13 +13,11 @@ #include #include "bpf_jit.h" +#define RV_REG_TCC RV_REG_A6 #define RV_FENTRY_NINSNS 2 /* fentry and TCC init insns will be skipped on tailcall */ #define RV_TAILCALL_OFFSET ((RV_FENTRY_NINSNS + 1) * 4) -#define RV_REG_TCC RV_REG_A6 -#define RV_REG_TCC_SAVED RV_REG_S6 /* Store A6 in S6 if program do calls */ - static const int regmap[] = { [BPF_REG_0] = RV_REG_A5, [BPF_REG_1] = RV_REG_A0, @@ -51,14 +49,12 @@ static const int pt_regmap[] = { }; enum { - RV_CTX_F_SEEN_TAIL_CALL = 0, RV_CTX_F_SEEN_CALL = RV_REG_RA, RV_CTX_F_SEEN_S1 = RV_REG_S1, RV_CTX_F_SEEN_S2 = RV_REG_S2, RV_CTX_F_SEEN_S3 = RV_REG_S3, RV_CTX_F_SEEN_S4 = RV_REG_S4, RV_CTX_F_SEEN_S5 = RV_REG_S5, - RV_CTX_F_SEEN_S6 = RV_REG_S6, }; static u8 bpf_to_rv_reg(int bpf_reg, struct rv_jit_context *ctx) @@ -71,7 +67,6 @@ static u8 bpf_to_rv_reg(int bpf_reg, struct rv_jit_context *ctx) case RV_CTX_F_SEEN_S3: case RV_CTX_F_SEEN_S4: case RV_CTX_F_SEEN_S5: - case RV_CTX_F_SEEN_S6: __set_bit(reg, &ctx->flags); } return reg; @@ -86,7 +81,6 @@ static bool seen_reg(int reg, struct rv_jit_context *ctx) case RV_CTX_F_SEEN_S3: case RV_CTX_F_SEEN_S4: case RV_CTX_F_SEEN_S5: - case RV_CTX_F_SEEN_S6: return test_bit(reg, &ctx->flags); } return false; @@ -102,32 +96,6 @@ static void mark_call(struct rv_jit_context *ctx) __set_bit(RV_CTX_F_SEEN_CALL, &ctx->flags); } -static bool seen_call(struct rv_jit_context *ctx) -{ - return test_bit(RV_CTX_F_SEEN_CALL, &ctx->flags); -} - -static void mark_tail_call(struct rv_jit_context *ctx) -{ - __set_bit(RV_CTX_F_SEEN_TAIL_CALL, &ctx->flags); -} - -static bool seen_tail_call(struct rv_jit_context *ctx) -{ - return test_bit(RV_CTX_F_SEEN_TAIL_CALL, &ctx->flags); -} - -static u8 rv_tail_call_reg(struct rv_jit_context *ctx) -{ - mark_tail_call(ctx); - - if (seen_call(ctx)) { - __set_bit(RV_CTX_F_SEEN_S6, &ctx->flags); - return RV_REG_S6; - } - return RV_REG_A6; -} - static bool is_32b_int(s64 val) { return -(1L << 31) <= val && val < (1L << 31); @@ -235,10 +203,7 @@ static void __build_epilogue(bool is_tail_call, struct rv_jit_context *ctx) emit_ld(RV_REG_S5, store_offset, RV_REG_SP, ctx); store_offset -= 8; } - if (seen_reg(RV_REG_S6, ctx)) { - emit_ld(RV_REG_S6, store_offset, RV_REG_SP, ctx); - store_offset -= 8; - } + emit_ld(RV_REG_TCC, store_offset, RV_REG_SP, ctx); emit_addi(RV_REG_SP, RV_REG_SP, stack_adjust, ctx); /* Set return value. */ @@ -332,7 +297,6 @@ static void emit_zext_32(u8 reg, struct rv_jit_context *ctx) static int emit_bpf_tail_call(int insn, struct rv_jit_context *ctx) { int tc_ninsn, off, start_insn = ctx->ninsns; - u8 tcc = rv_tail_call_reg(ctx); /* a0: &ctx * a1: &array @@ -355,9 +319,11 @@ static int emit_bpf_tail_call(int insn, struct rv_jit_context *ctx) /* if (--TCC < 0) * goto out; */ - emit_addi(RV_REG_TCC, tcc, -1, ctx); + emit_ld(RV_REG_TCC, ctx->tcc_offset, RV_REG_SP, ctx); + emit_addi(RV_REG_TCC, RV_REG_TCC, -1, ctx); off = ninsns_rvoff(tc_ninsn - (ctx->ninsns - start_insn)); emit_branch(BPF_JSLT, RV_REG_TCC, RV_REG_ZERO, off, ctx); + emit_sd(RV_REG_SP, ctx->tcc_offset, RV_REG_TCC, ctx); /* prog = array->ptrs[index]; * if (!prog) @@ -763,7 +729,7 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, int i, ret, offset; int *branches_off = NULL; int stack_size = 0, nregs = m->nr_args; - int retval_off, args_off, nregs_off, ip_off, run_ctx_off, sreg_off; + int retval_off, args_off, nregs_off, ip_off, run_ctx_off, sreg_off, tcc_off; struct bpf_tramp_links *fentry = &tlinks[BPF_TRAMP_FENTRY]; struct bpf_tramp_links *fexit = &tlinks[BPF_TRAMP_FEXIT]; struct bpf_tramp_links *fmod_ret = &tlinks[BPF_TRAMP_MODIFY_RETURN]; @@ -807,6 +773,8 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, * * FP - sreg_off [ callee saved reg ] * + * FP - tcc_off [ tail call count ] BPF_TRAMP_F_TAIL_CALL_CTX + * * [ pads ] pads for 16 bytes alignment */ @@ -848,6 +816,11 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, stack_size += 8; sreg_off = stack_size; + if (flags & BPF_TRAMP_F_TAIL_CALL_CTX) { + stack_size += 8; + tcc_off = stack_size; + } + stack_size = round_up(stack_size, 16); if (func_addr) { @@ -874,6 +847,10 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, emit_addi(RV_REG_FP, RV_REG_SP, stack_size, ctx); } + /* store tail call count */ + if (flags & BPF_TRAMP_F_TAIL_CALL_CTX) + emit_sd(RV_REG_FP, -tcc_off, RV_REG_TCC, ctx); + /* callee saved register S1 to pass start time */ emit_sd(RV_REG_FP, -sreg_off, RV_REG_S1, ctx); @@ -927,6 +904,9 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, if (flags & BPF_TRAMP_F_CALL_ORIG) { restore_args(nregs, args_off, ctx); + /* restore TCC to RV_REG_TCC before calling the original function */ + if (flags & BPF_TRAMP_F_TAIL_CALL_CTX) + emit_ld(RV_REG_TCC, -tcc_off, RV_REG_FP, ctx); ret = emit_call((const u64)orig_call, true, ctx); if (ret) goto out; @@ -967,6 +947,10 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, emit_ld(RV_REG_S1, -sreg_off, RV_REG_FP, ctx); + /* restore TCC to RV_REG_TCC before calling the original function */ + if (flags & BPF_TRAMP_F_TAIL_CALL_CTX) + emit_ld(RV_REG_TCC, -tcc_off, RV_REG_FP, ctx); + if (func_addr) { /* trampoline called from function entry */ emit_ld(RV_REG_T0, stack_size - 8, RV_REG_SP, ctx); @@ -1476,6 +1460,9 @@ int bpf_jit_emit_insn(const struct bpf_insn *insn, struct rv_jit_context *ctx, if (ret < 0) return ret; + /* restore TCC from stack to RV_REG_TCC */ + emit_ld(RV_REG_TCC, ctx->tcc_offset, RV_REG_SP, ctx); + ret = emit_call(addr, fixed_addr, ctx); if (ret) return ret; @@ -1735,6 +1722,7 @@ int bpf_jit_emit_insn(const struct bpf_insn *insn, struct rv_jit_context *ctx, void bpf_jit_build_prologue(struct rv_jit_context *ctx) { int i, stack_adjust = 0, store_offset, bpf_stack_adjust; + bool is_main = ctx->prog->aux->func_idx == 0; bpf_stack_adjust = round_up(ctx->prog->aux->stack_depth, 16); if (bpf_stack_adjust) @@ -1753,8 +1741,7 @@ void bpf_jit_build_prologue(struct rv_jit_context *ctx) stack_adjust += 8; if (seen_reg(RV_REG_S5, ctx)) stack_adjust += 8; - if (seen_reg(RV_REG_S6, ctx)) - stack_adjust += 8; + stack_adjust += 8; /* RV_REG_TCC */ stack_adjust = round_up(stack_adjust, 16); stack_adjust += bpf_stack_adjust; @@ -1769,7 +1756,8 @@ void bpf_jit_build_prologue(struct rv_jit_context *ctx) * (TCC) register. This instruction is skipped for tail calls. * Force using a 4-byte (non-compressed) instruction. */ - emit(rv_addi(RV_REG_TCC, RV_REG_ZERO, MAX_TAIL_CALL_CNT), ctx); + if (is_main) + emit(rv_addi(RV_REG_TCC, RV_REG_ZERO, MAX_TAIL_CALL_CNT), ctx); emit_addi(RV_REG_SP, RV_REG_SP, -stack_adjust, ctx); @@ -1799,22 +1787,14 @@ void bpf_jit_build_prologue(struct rv_jit_context *ctx) emit_sd(RV_REG_SP, store_offset, RV_REG_S5, ctx); store_offset -= 8; } - if (seen_reg(RV_REG_S6, ctx)) { - emit_sd(RV_REG_SP, store_offset, RV_REG_S6, ctx); - store_offset -= 8; - } + emit_sd(RV_REG_SP, store_offset, RV_REG_TCC, ctx); + ctx->tcc_offset = store_offset; emit_addi(RV_REG_FP, RV_REG_SP, stack_adjust, ctx); if (bpf_stack_adjust) emit_addi(RV_REG_S5, RV_REG_SP, bpf_stack_adjust, ctx); - /* Program contains calls and tail calls, so RV_REG_TCC need - * to be saved across calls. - */ - if (seen_tail_call(ctx) && seen_call(ctx)) - emit_mv(RV_REG_TCC_SAVED, RV_REG_TCC, ctx); - ctx->stack_size = stack_adjust; } @@ -1827,3 +1807,8 @@ bool bpf_jit_supports_kfunc_call(void) { return true; } + +bool bpf_jit_supports_subprog_tailcalls(void) +{ + return true; +}