[patchlet] One liner "fix" to mm/vmalloc.c

if (alloc_area_pmd(pmd, address, end - address, gfp_mask, prot)) break; -- Francis Galiegue, [EMAIL PROTECTED] - Normand et fier de l'être "Programming is a race between programmers, who try and make more and more idiot-proof software, and universe, which pr

Re: [patchlet] One liner "fix" to mm/vmalloc.c

On Fri, 16 Feb 2001, Burton Windle wrote: > Date: Fri, 16 Feb 2001 17:10:27 -0500 (EST) > From: Burton Windle <[EMAIL PROTECTED]> > To: Francis Galiegue <[EMAIL PROTECTED]> > Subject: Re: [patchlet] One liner "fix" to mm/vmalloc.c > > What kernel i

Re: SMP: bind process to cpu

d guess: as this is a bitmask, you must "or" the bitmask with (1 << cpu_number). 1 for CPU 0 only, 5 for CPU 0 and 2, etc, etc. -- Francis Galiegue, [EMAIL PROTECTED] - Normand et fier de l'être "Programming is a race between programmers, who try and make more and more

[PATCH] 2.4.1-ac16, arch/i386/kernel/irq.c: spurious #if CONFIG_SMP.. #endif

set_affinity(irq, new_value); -- Francis Galiegue, [EMAIL PROTECTED] - Normand et fier de l'être "Programming is a race between programmers, who try and make more and more idiot-proof software, and universe, which produces more and more remarkable idiots. Until now, universe leads t

Re: Linux stifles innovation...

tter how hard one will yell it around - vendors don't lose anything in releasing open source drivers and/or specs; on the opposite, they gain broader support. -- Francis Galiegue, [EMAIL PROTECTED] - Normand et fier de l'être "Programming is a race between programmers, who try an

[PATCH] 2.4.1-ac16: add CDROM_LOCKDOOR ioctl to IDE floppies

if (cmd == CDROMEJECT) { + idefloppy_create_start_stop_cmd (&pc, 2); + (void) idefloppy_queue_pc_tail (drive, &pc); + } return 0; case IDEFLOPPY_IOCTL_FORMAT_SUPPORTED: return (0); --

[PATCHES] 3 of them, resubmit, over 2.4.1-ac20

NOMEM whereas once would be enough. What's more, the loop in which this spurious assignment appears runs with the kernel lock held. This patch fixes this. Have fun, -- Francis Galiegue, [EMAIL PROTECTED] - Normand et fier de l'être "Programming is a race between programmers, who t

Re: [PATCHES] 3 of them, resubmit, over 2.4.1-ac20

On Wed, 21 Feb 2001, Alan Cox wrote: > > Care to include the actual patches tho.. > Argh! Brain damage... Sorry... -- Francis Galiegue, [EMAIL PROTECTED] - Normand et fier de l'être "Programming is a race between programmers, who try and make more and more idiot-proof sof

[PATCH] 2.4.2: cure the kapm-idled taking (100-epsilon)% CPU time

As attachment. Don't ask me why it works. Rather, if you see why it works, I'd like to know why :) Patch also applies cleanly over 2.4.2-ac10. -- Francis Galiegue, [EMAIL PROTECTED] - Normand et fier de l'être "Programming is a race between programmers, who try and make

Re: [PATCH] 2.4.2: cure the kapm-idled taking (100-epsilon)% CPUtime

On Sun, 4 Mar 2001, Francis Galiegue wrote: > > As attachment. Don't ask me why it works. Rather, if you see why it works, I'd > like to know why :) > BTW, in case this matters, this is with gcc 2.95.3 compiler... -- Francis Galiegue, [EMAIL PROTECTED] - Nor

Re: [PATCH] 2.4.2: cure the kapm-idled taking (100-epsilon)% CPUtime

saving > Well, from reading the source, I don't see how this can break APM... What am I missing? -- Francis Galiegue, [EMAIL PROTECTED] - Normand et fier de l'être "Programming is a race between programmers, who try and make more and more idiot-proof software, and universe, which

Re: [kernel] Re: [PATCH] 2.4.2: cure the kapm-idled taking (100-epsilon)%CPU time

: [...] APM_DO_SAVE_SEGS; { int cx, dx, si; [...] Aren't cx, dx and si really meant to be u32? -- Francis Galiegue, [EMAIL PROTECTED] - Normand et fier de l'être "Programming is a race between programmers, who try and make more and more idiot-proof software, and universe, whi

Re: Annoying CD-rom driver error messages

> without the kernel generating spurious WARNINGS/ERRORS? > The main problem is, in fact: none of floppy drives/IDE floppies/CDROMs/whatever can do asynchronous medium change notifications (at least not that I know of), so you'll need to poll anyway... And of course, the command

Re: UDMA/66

in specific VIA chipset support! > > shouldn't DMA be enabled upon bootup? > It is if and only if you compiled your kernel with the CONFIG_IDEDMA (at least this is the name of the option in 2.2.x). -- Francis Galiegue, [EMAIL PROTECTED] "Programming is a race between progra

Re: A question about memory fragmentation

: create your own slab cache for your objects. -- Francis Galiegue, [EMAIL PROTECTED] "Programming is a race between programmers, who try and make more and more idiot-proof software, and universe, which produces more and more remarkable idiots. Until now, universe leads the race" --

Re: Modprobe local root exploit

ue; > + if (*p == '_' || *p == '-') > + continue; > + return -EINVAL; > + } > Just in case... Some modules have uppercase letters too :) -- Francis Galiegue, [EMAIL PROTECTED] "Programming is a race between programmers, who try

Re: Calling current() from interrupt context

interrupt-level code that calls current() or is it a design > principle that it cannot be called? > Normally no, you can't - IRQ handlers are called from system context, not user context, and are in no way associated to any process. -- Francis Galiegue, [EMAIL PROTECTED] "Programmi