search for: stosb

...+ inc si ; Align to the real characters + mov cx,13 ; 13 characters per VFAT DIRENT +.vfat_cp_chr: + gs lodsw ; Unicode here!! + mov bp,ds + mov es,bp + call ucs2_to_cp ; Convert to local codepage + mov bp,fs + mov es,bp + jc .vfat_abort ;-; Use short name if character not on codepage + stosb ; CAN NOT OVERRIDE es + cmp al,0 + jz .vfat_find_next ; Null-terminated string; don't process more + cmp cx,3 + je .vfat_adj_add2 + cmp cx,9 + jne .vfat_adj_add0 +.vfat_adj_add3: inc si +.vfat_adj_add2: inc si +.vfat_adj_add1: inc si +.vfat_adj_add0: + loop .vfat_cp_chr + cmp dh,1 ; Is...

...e is my hack to change "ip=#:#:#:#" to "ip=#:#:#:#::eth0:" in pxelinux.asm (after "genipopt"): ... mov eax,[Netmask] call gendotquad ; Zero-terminates its output mov al,':' ; HACK stosb ; HACK mov al,':' ; HACK stosb ; HACK mov al,'e' ; HACK stosb ; HACK mov al,'t' ; HACK...

...,16 + xchg ah,al +.hexify_loop1: rol eax,4 + push eax + and al,0Fh + cmp al,10 + jae .high1 +.low1: add al,'0' + jmp short .char1 +.high1: add al,'A'-10 +.char1: stosb + pop eax + loop .hexify_loop1 + + mov cx,6 ; 6 nibbles in netcard part of MAC + mov eax,[MyMAC2] ; 3 byte netcard part of MAC xchg ah,al ; Convert to host byte order...

...; Store the next line we need + mov dx,ax ; Use DX to store the progress + mov bx,13 + mov ah,0 + mul bl ; Offset for DI + add di,ax ; Increment DI + inc si ; Align to the real characters + mov cx,13 ; 13 characters per VFAT DIRENT +.vfat_cp_chr: + gs lodsb + inc si ; Unicode here!! + stosb + cmp al,0 + jz .vfat_find_next ; Null-terminated string; don't process more + cmp cx,3 + je .vfat_adj_add2 + cmp cx,9 + jne .vfat_adj_add0 +.vfat_adj_add3: inc si +.vfat_adj_add2: inc si +.vfat_adj_add1: inc si +.vfat_adj_add0: + loop .vfat_cp_chr + cmp dh,1 ; Is this the first round?...

...dLen] + dec cx + and cx,cx + jne continue + + mov ax,2 + mov bx, msg_usage + int 22h + ret +continue: + mov di,pspCmdArg+1 + add di,cx + dec di + std + mov al,' ' + repe scasb + inc cx + cld + mov [pspCmdLen],cl + mov si,pspCmdArg+1 + mov di,si + add di,cx + xor al,al + stosb + mov si,pspCmdArg+1 + +; get config file name + mov ax,0eh + int 22h + +; open config file + mov si,bx ; es:bx config file name + mov ax,6 + int 22h + jc no_config_file + and eax,eax + jz no_config_file + jmp got_config_file +no_config_file: + push es + push bx + push cs + pop es +...

...bler should guess what the user is doing”. The patches you’re referring to are dealing with a very specific issue: there are multiple ways users specify prefixes for certain instructions. Sometimes the prefixes are integrated into the instruction. Sometimes they’re written separately, like “rep; stosb”. And sometimes users end up using raw “.byte” directives. So we add a rule: if we see something that looks like a prefix, followed by an instruction, we assume the combination might actually be a prefixed instruction, and don’t rewrite it. This is a heuristic in some sense: we can’t prove whethe...

...dLen] + dec cx + and cx,cx + jne continue + + mov ax,2 + mov bx, msg_usage + int 22h + ret +continue: + mov di,pspCmdArg+1 + add di,cx + dec di + std + mov al,' ' + repe scasb + inc cx + cld + mov [pspCmdLen],cl + mov si,pspCmdArg+1 + mov di,si + add di,cx + xor al,al + stosb + mov si,pspCmdArg+1 + +; get config file name + mov ax,0eh + int 22h + +; open config file + mov si,bx ; es:bx config file name + mov ax,6 + int 22h + jc no_config_file + and eax,eax + jz no_config_file + jmp got_config_file +no_config_file: + push es + push bx + push cs + pop es +...

Even with -ffreestanding LLVM generates memcpy/memset? Does this mean some passes do not honor this flag? If you really wanted to prevent libcalls, you could technically translate those memcpy/memset to loops in lowering. Kevin On Sat, Oct 15, 2016 at 4:10 PM, Wolfgang McSneed via llvm-dev < llvm-dev at lists.llvm.org> wrote: > Hi Mehdi and Joerg, > > Thanks for your fast

...([32 x i8], [32 x i8]* @.str.75, i64 0, i64 0)) #58 to label %7 unwind label %10 ; <label>:7: ; preds = %6 unreachable ; <label>:8: ; preds = %2 %9 = tail call { i64, i8* } asm sideeffect "rep stosb", "={cx},={di},{ax},0,1,~{memory},~{dirflag},~{fpsr},~{flags}"(i32 0, i64 16, i8* nonnull %3) #53, !srcloc !1070 invoke void @_ZN8CryptoPP19UnalignedDeallocateEPv(i8* nonnull %3) to label %13 unwind label %10 ; <label>:10: ; pred...

I found that the version of isolinux.bin (and the debug version too) doesn't work on my Via C3 system. I get the SAY text, but never the boot: prompt. C-A-Del doesn't reboot either. The debug version does not print anything after announcing it's opening the config file. Only the single SAY text follows. The cursor sits on the next blank line, useless (no responses to key presses).

I've begun tinkering with syslinux v3 and have uncovered an anomoly with the DISPLAY file when called by pxelinux.0. If I have: DISPLAY boot.msg With an "^Xsplash.lss\n" in the file boot.msg (exactly: 18 73 70 6C 61 73 68 2E 6C 73 73 0A) file, it will fail to load the rle image. And produce some goofy results in the tftpd log: Jun 5 06:27:06 pxe in.tftpd[24649]: RRQ from

...k. diff -ru syslinux-1.66.old/pxelinux.asm syslinux-1.66/pxelinux.asm --- syslinux-1.66.old/pxelinux.asm Wed Jan 2 07:53:39 2002 +++ syslinux-1.66/pxelinux.asm Sat Jan 12 18:40:34 2002 @@ -1327,7 +1327,7 @@ .not_ctrl_f: cmp di,max_cmd_len+command_line ; Check there's space jnb get_char stosb ; Save it - call writechr ; Echo to screen +;Martin call writechr ; Echo to screen get_char_2: jmp short get_char not_ascii: mov byte [FuncFlag],0 cmp al,0Dh ; Enter @@ -1339,8 +1339,8 @@ backspace: cmp di,command_line ; Make sure there is anything je get_char ; to erase d...

...nux's variables when ES is still pointing to UNDI code segment. My fix is denoted by [KTT]. File : pxelinux.asm * Bug#1 ; Wipe the F-key area mov al,NULLFILE mov di,FKeyName mov cx,10*(1 << FILENAME_MAX_LG2) [KTT] push es [KTT] push ds [KTT] pop es rep stosb [KTT] pop es * Bug#2 mov di,Sockets mov cx,(MAX_SOCKETS*tftp_port_t_size)/4 xor eax,eax [KTT] push es [KTT] push ds [KTT] pop es rep stosd [KTT] pop es Can you submit these changes to Linux community? Thanks, ------------------------------------------...

...lue (0: no timeout) call far [gfx_bc_input] ret @@ -786,6 +785,42 @@ do_default: .noparm: ret +do_timeout: + call skipspace + jz .eof + jc .noparm + call ungetc + push es + push di + push cs + pop es + mov di,NumBuf +.getnum: + cmp di,NumBufEnd + jae .loaded + call getc + stosb + cmp al,'-' + jnb .getnum + call ungetc + dec di +.loaded: + mov byte [di],0 + pop di + pop es + push cs + pop ds + mov si,NumBuf + push ebx + call parseint + jc .err + mov [menu_timeout],ebx +.err: + pop ebx +.eof: +.noparm: + ret + skipline: cmp al,10 je .end @@ -8...

...X86_FEATURE_SSE4_2 20 /* Streaming SIMD Extensions 4.2 */ @@ -132,5 +133,6 @@ #define X86_FEATURE_SMEP 7 /* Supervisor Mode Execution Protection */ #define X86_FEATURE_BMI2 8 /* 2nd group bit manipulation extensions */ #define X86_FEATURE_ERMS 9 /* Enhanced REP MOVSB/STOSB */ +#define X86_FEATURE_INVPCID 10 /* Invalidate Process Context ID */ #endif /* __LIBXC_CPUFEATURE_H */ diff -r 0b15aa9541dc tools/libxc/xc_cpuid_x86.c --- a/tools/libxc/xc_cpuid_x86.c Thu Nov 17 18:41:59 2011 +0800 +++ b/tools/libxc/xc_cpuid_x86.c Thu Nov 17 23:09:45 2011 +0800 @@ -481,6 +...

...e_no_enter + cmp al,' ' ; ASCII? + jb vmware_no_enter ; don't paste no non-ASCII + ja vmware_enter + cmp di,command_line ; Space must not be first + je short vmware_no_enter +vmware_enter: cmp di,max_cmd_len+command_line ; Check there's space + jnb short vmware_no_enter + stosb ; Save it + call writechr ; Echo to screen +vmware_no_enter: + shr eax,8 + dec ebx + je vmware_done + dec cx + jne vmware_paste_char + jmp vmware_get_data + +vmware_clear_buffer: + sub ebx,ecx + jle vmware_done + shr ebx,2 +vmware_clear_loop: + mov eax,VMwareMagic + mov cx,00007h...

...ll Rights Reserved @@ -345,9 +345,12 @@ mov [fs:kern_cmd_offset],di ; Store pointer mov cx,[CmdLineLen] - add cx,byte 3 - shr cx,2 ; Convert to dwords - fs rep movsd + cmp cx,255 + jna .len_ok + mov cx,255 ; Protocol < 0x202 has 255 as hard limit +.len_ok: + fs rep movsb + fs stosb ; Final null, note AL == 0 here push fs pop es

.../* movsw/movsl */ @@ -299,16 +328,28 @@ } else { thread_inst->op_size = LONG; } - strcpy((char *)thread_inst->i_name, "movs"); - return DECODE_success; - + case 0xaa: + /* stosb */ + thread_inst->op_size = BYTE; + strcpy((char *)thread_inst->i_name, "stosb"); + return DECODE_success; + case 0xab: + /* stosw/stosl */ + if (thread_inst->op_size == WORD) { + strcpy((char *)thread_in...

.../* Hardware Lock Elision */ #define X86_FEATURE_AVX2 (7*32+ 5) /* AVX2 instructions */ #define X86_FEATURE_SMEP (7*32+ 7) /* Supervisor Mode Execution Protection */ #define X86_FEATURE_BMI2 (7*32+ 8) /* 2nd bit manipulation extensions */ #define X86_FEATURE_ERMS (7*32+ 9) /* Enhanced REP MOVSB/STOSB */ #define X86_FEATURE_INVPCID (7*32+10) /* Invalidate Process Context ID */ +#define X86_FEATURE_RTM (7*32+11) /* Restricted Transactional Memory */ #define cpu_has(c, bit) test_bit(bit, (c)->x86_capability) #define boot_cpu_has(bit) test_bit(bit, boot_cpu_data.x86_capability) _________...

The first patch really isn''t as much of a follow-up than what triggered the security issues to be noticed in the first place. 1: x86: preserve FPU selectors for 32-bit guest code 2: x86: fix XCR0 handling 3: x86/xsave: adjust state management 4: x86/fxsave: bring in line with recent xsave adjustments The first two I would see as candidates for 4.3 (as well as subsequent backporting,